This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project to make the world's books discoverable online. It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the publisher to a library and finally to you. Usage guidelines Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to prevent abuse by commercial parties, including placing technical restrictions on automated querying. We also ask that you: + Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for personal, non-commercial purposes. + Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the use of public domain materials for these purposes and may be able to help. + Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find additional materials through Google Book Search. Please do not remove it. + Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner anywhere in the world. Copyright infringement liability can be quite severe. About Google Book Search Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web at|http : //books . google . com/ HARVARD UNIVERSITY Library of the Museum of Comparative Zoology Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC }} ^v PROCEEDINGS OF THK g0sl0n Somlg of |latural ^istorji* VOL. XXIX. WITH FORTY-ONE PLATES. "^BOSTON: PRINTED FOR THE SOCIETY. 1901. Digitized by VjOOQ IC v^^ PrHLlSlIING COMMITTEE. Charles 8. Minuj, Roland Thaxter, Alpiiei's Hyatt, Jay B. Woodworth, Glover M. Allen. Digitized by VjOOQ IC CONTENTS OF VOL. XXIX. Page No. 1. — Proceedings of the Annual Meeting, May 3, 1899. Report of the Curator, Alphbus Hyatt 1 Report of the Secretary and Librarian, Samtel Henshaw ... 14 Report of the Treasurer, £dwari> T. Boivfe 22 Officers for 1899-1900 24 List of members 26 By-laws 38 June, 1899. No. 2. — Variation and Sexual Selection in Man. By Edwin Tennky Brewster. July, 1899 46 No. 3. — Notes on the Reptiles and Amphibians of Intervale, New Hamp- shire. By Glover M. Allen. July, 1899 63 No. 4. — Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United States. By Garev i>e N. Hoigh. July, 1899. ... 77 No. 6. — Contributions from the Gray Herbarium of Harvard University. New Series. — No. 17. By B. L. Robinson and J. M. Grbenman. August, 1899 87 No. 6. — The Development of Penilia schmackeri Richai*d. By Mervin T. Sri)LER. (3 Plates.) October, 1899 109 No. 7. — List of Marine Mollusca of Coldspring Harbor, Long Island, with descriptions of one new Genus and two new Species of Nudibranchs. By Francis Noves Balch. (1 Plate.) October, 1899. ... 133 No. 8. — The B1(mk1 Vessels of the Heart in Carcharias, Raja, and Amia. Bv G. H. Parker and Frei>erica K. Davis. (3 Plates.) October, 1899. 103 No. 9. — The Oceurrt»nce of Fossils in the Roxburj' Conglomerate. By Henry T. Birr and R^uieht E. Birke. (1 Plate.) April, 1900. 179 No. 10. — On a hitherto unrecognized fonn of bUxMl circulation without capillaries in the organs of the Veitebrata. By Charles SEiMiwuK Minot, LL. D. April, HKH) 186 No. 11. — A Revision of the Systematic Names employed by Writers on the Morpholo^' of the Acmaeidae. By M. A. Wii.lcox, Ph. D. April, liHK) 217 No. 12. — Proceedings of the Annual Meeting, May 2, 1900. Re|X)rt of the Curator, A LPiiEi-s Hyatt 223 Report of the Secretary- and Librarian, Charles F. Batcheldbr 234 Report of the Treasurer, Ei»ward T. Borv6 239 May, 1900. No. 13. — The Embryonic History of Imaginal Discs in Melophagus ovinus L., together witli an Account of the Earlier Stages in the Develop- ment of the Insect, By H. S. Pratt, Ph. D. (7 Plates.) June, 19(>f) 241 No. 14. — Glacial Erosion in France, Switzerland and Norway. By William Morris Davis. (3 Plates.) July, 1900. . . .273 No. 16. — Echinoderms from Puget Soiuid : Ol)servations made on the Echinoderins collected by the parties from Colimibia University, in Puget Sound in 1890 and 1897. By Hubert Lyman Clark. (4 Plates.) May, 1901 323 No. 16. — Bermudan Echinoderms. A Report on Observations and Collec- tions made in 18m). By Hubert Lyman Clark. May, 1901. . 339 No. 17. — Proceedings of the Annual Meeting, May 1, 1901. Report of the Curator, Alpheis Hyatt 347 Report of the Secretary and Librarian, Charles F. Batchelder . 358 Report of the Treasurer, £dwari> T. Bouvfe 304 Officers for 1901-1902 300 List of Members 307 June, 1901. No. 18. — The Polychaeta of the Puget Souna Region. By Herbert Farlin Johnson. (19 Plates.) August, 1901 381 Digitized by VjOOQ IC Digitized by VjOOQ IC JfO Prc»ca«dliie» ol the Boston Soolet^ of Hatur&i Hftntorr. :bSb Vol. t% No. I, l'lt*iCKKnrNr}H OK TIIK ANNTAL MKKTINO. MAY :i, IHOO BOSTON PHINTEI> FflR TITE SOOIETT June, 1899- Digitized by Google Digitized by VjOOQ IC tu/nt ibh i^^j No. 1. — Proceedings of the Annual Meeting^ May 3, 1899. Report of the Curator, Alpiieus Hyatt. The death of Mr. John Cummings of Woburn removes a member to whom this Society is indebted for valuable time freely given for many years in its Council and for much actual work done in its Museum, especially in the botanical department. While his pecun- iary means were ample, he was as generous in giving aid with them as with his brain and hands. For a long time he carried on the botanical department, working in it himself and paying the salary of an assistant; he also maintained another assistant in the 3Iuseum, and at the same time supported the Teachers' school of science. His services in the botanical department were described in detail in my report for 1898 on the occasion of his retirement from that department which he had sustained from 1873 to 1898 and which he had succeeded in placing in excellent. condition. Another monument to Mr. Cummings's generosity is the collection of European fossils filling Room H. This is the Eser collection, which is entirely his gift, and is one of the most famous of the older and smaller European collections. This collection is a great prize for any museum since it possesses very rare and valuable specimens and is especially suitable for the purposes of our educational series. The Teachers' school of science owes its foundation to Mr. Cummings and arose in the Council in conseqlience of his offer to support a series of lectures for teachers. The Curator immediately accepted this offer, and the school began in the following autumn. These are only his principal claims to our remembrance and gratitude, since it is impossible to take notice of the thousand and one ser- vices to the cause of science which he gave so generously and with- out expectation of credit or reward. In his last annual report the Curator called attention to the necessity of doing something to perpetuate the office of Guide to the Museum. This has been held by Mr. Grabau and has become really a free lectureship, that has not only made our collections more instructive to the public but also interested a number of persons in the study of natural history and led to the giving of Digitized by VjOOQ IC 2 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. many lessons on the sea- shore and in our laboratory or lecture room. This has greatly increased our usefulness to the community, and some means should be found to continue it. An appeal was also made for aid in the botanical department. This is still cared for by Miss Carter but largely as a voluntar}* service, since only somewhat more than is sufficient to meet her expenses is now paid her. Another part of the same report set forth the claims of the New England collection in our Museum, and the great need that exists of separating this from the educational and general systematic collections. The New England collections are scattered in the different departments and should be brought together in order to give a complete exposition of the natural products of our own neighborhood. The need of exhibiting in connected succession the mineralogy, geology, botany, and zoology of New England was fully demonstrated during the past summer when the American association for the advancement of science met in our building and in that of the Institute of technology and many of its members visited our Museum. If we had been able to throw open to them a series of collections showing such a connected history of the natural products of New England and of the geology of this region, it would have been a revelation of the teaching j)ower of collections and might have had a far-reaching influence upon other museums in this country. The Curator has always held that a New England museum should be brought together in this way within our building, and the whole be placed in order before any General guide could be effectively written. This year, however, the need for some guide to explain the apparent confusion to scientific visitors became urgent, and he wrote a small pamphlet of forty-seven pages, accompanied by a diagram of the building. This appeared as a second edition of a General guide written and published on the celebration of the fiftieth anniversary of this Society, but it is in reality new. Teaching ix the Museum. The operations of this department have ceased for want of funds, as has been noted above. Digitized by VjOOQ IC HYATT: REPORT OF THE CURATOR. Dynamical Zoology. Considerable work on this collection has been done. The differ- ent series have been rearranged and general descriptive labels prepared by the Curator assisted by Miss Bryant. The large relief map of Oahu shows with the aid of colored pins the migrations of the species of three of the principal genera of Achatinellidae. The genera are represented by pins with heads of different colors, and the species are represented by different numbers and labels on the pins. In this way the migrations of the species may be followed along the mountain ridges of the island, and the correlation of these movements with topography is brought out more clearly than by the use of the shells themselves. A valuable and interesting series of lamprey eels has been given to this collection by Prof. S. H. Gage of Cornell. Mineralogy and Geology. During the summer Professor Crosby gave considerable time to the geological department of the Museum, putting the specimens on exhibition in more perfect order, labeling new specimens, etc., and he also labeled and arranged, as well as the limited space now available would permit, the illustrative specimens and maps pre- pared for Parts 1 and 2 of the Boston Basin work, and for Part 3 so far as that was then completed. The same assistant has also completed a much needed revision of the general collection of min- erals and incorporated with the mounted s})ecimens on exhibition all of the materials that have been accumulating in the past ten years. Professor Crosby has also personally paid for the necessary clerical assistance employed in this work. During the meeting of the American association for the advance- ment of science he presented before Section E an outline of the history of the Blue Hills complex, and abstracts of this paper appeared in the American geologist, in Science, and in the Pro- ceedings of the American association. He also conducted seven parties of geologists to points of local interest, including the Blue Hills, and had the satisfaction of learning that his views concerning the geology of this exceedingly difficult region were endorsed by competent authorities. Digitized by VjOOQ IC 4 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Another gratifying result of going over the ground with these geologists was the offer of further cooperation. Professor Crosby's work is necessarily mainly areal and structural, and although he has himself paid for a number of chemical analyses and has received additional assistance from several of his students, it was very de- sirable that this should be supplemented, especially for the igneous rocks, by microscopic and chemical work. Dr. White of Columbia has very kindly done considerable microscopical work on the Blue Hills rocks, and Dr. A. S. Eakle of Harvard has kindly supple- mented this by investigations of the plutonic rocks, especially on chemical lines. The intricate nature of the problem of the geology of this region has been the cause of the successive delays in com- pleting the work, and it would still have remained unfinished if Professor Crosby had not this year given an extraordinary amount of time to this purpose. The manuscript of Part 3, *' The Blue Hills complex," is now in the hands of the Secretary, 375 pages, with 8 plates and 26 figures. Two chapters which Mr. Grabau is writing, one on the fossils and one on Lake Bouv6, an extinct glacial lake, noticed in the report for 1895-96, are in the hands of the author, but are stated to be practically finished. An important accession of new fossils lately received, including the Sears collections from Nahant and Mr. "W. W. Dodge's collections from Braintree, has prevented the com- pletion of the descriptions of the fossils in time for this report, but it is gratifying to notice that these accessions have about tripled the materials for investigation. The work on the Neponset Valley, Part 4, of the Boston Basin work, has been actively pressed. Outside the areal and structural work done by Professor Crosby, Dr. Florence Bascom, of Bryn Mawr college, accepted an invitation to contribute by studying the volcanic rocks, and Mr. F. C. Ohm of the U. S. geological survey has also kindly taken the work on the thin sections, while Professor Crosby has arranged for a sufiicient number of chemical anal3\ses to supplement Miss Bascom 's investigations. Professor Crosby ends his report to the Curator as follows : '* The willingness of other geologists to cooperate with me is a great encouragement and leads me to hope that the ideal end of this work may be realized ; viz : an illustrate PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. thing on Pelecypoda. A considerable number of fossil Gasteropoda have also been selected from the general collections, mounted, and placed on exhibition. Mrs. Sheldon has also described about fifty- eight species of Crustacea and selected and mounted a limited num- ber of specimens. Miss Martin has been as usual more or less occupied in making colored drawings for this collection, especially during the summer. The labeling of the tablets has been done by Miss Bryant and also the preparation of some of the fossils. Botany. Less work than in previous years has been done u})on the herba- rium, owing in part to the unfortunate sickness of the assistant. Miss Carter. She nevertheless continued through a large part of the past official year the work u])on the labeling of the Lowell col- lection and systematic arrangement of the duplicates and the j)oi- soning of the plants. During the summer she also looked over and cleaned the specimens on exhibition and placed in position a series of shelf labels indicating the different divisions of plants, and Miss Bryant a.ssisted her by making these labels. Thirty species of North American lichens were added by purchase from Cummings, Seymour, and Earle. Nineteen persons have been permitted to consult and study in the herbarium. Palaeontology. Miss Bryant has this year worked upon the renovation of the tablets and specimens in the European room, especially in Oolite, and has rewritten about four hundred labels. She has also worked over all of the American Jurassic and Cretaceous fossils and has cleaned from the matrix and mounted a number of Silurian fossils of the American collection from Anticosti, and a collection of fos- sils from the St. John's group has also been cleaned and mounted by the same assistant. A number of Devonian corals purchased from Mr. Greene have been identified and labeled, and the Vertebrata from the Cretaceous freshly labeled. The same assistant revised the work previously done by other assistants upon the Trenton and Niagara fossils and relabeled them. She also finished the Carboni- ferous fossils including the Brachiopoda, Lamellibranchiata, Gastero- poda, and Cephalopoda. Digitized by VjOOQ IC HYATT : REPORT OF THE CURATOR. MOLLUSCA. The large collection purchased of Rev. J. T. Gulick last year and described in the report of 189H has been completely catalogued and every lot labeled, each shell having been also numbered. The col- lections of Achatinellidae previously mentioned have been still further increased by the loan of eight thousand shells belonging to Mr. C. M. Cooke, Jr., of Honolulu, making in all about thirty thousand shells of this family at present available for study. The Curator has found it essential to trace as far as practicable the migrations of the Achatinellae from island to island, and has made good progress during the past year, having reviewed most of the genera that are found in islands outside of Oahu. A small but valuable collection of Achatinellidae from the island of Molokai that filled a number of blanks in our own collection has been given us by Dr. W. P. Wesselhoeft. Miss Martin completed the work of labeling the duplicate Gasteropoda having no locality labels, and these were in part given away to the public schools and in part exchanged for invertebrates collected in the West Indies by Mr. C. J. Maynard. Miss Martin has made a complete series of colored drawings of the animals of Pteropoda and arranged these in com- pany with a series of the shells purchased from Mr. Sowerby. The death of Mr. Edward W. Roper, a member who had taken a great interest in this collection and was practically an assistant in this department, was a severe loss. He had, in the short time that he was connected with us before his health obliged him to seek refuge in southern California, done much efficient work, and had also planned the revision of all our land shells. His entire con- chological collection was bequeathed to us and is now in our pos- session, with the exception of the Cyrenidae. These are in the hands of Mrs. Roper, who will deliver them to us as soon as prac- ticable after her return to the east. A detailed re])ort on this collection will be made after these have been received and the whole collection has been catalogued. This work has been begun by Miss Martin. Ckustacea. The work of restoring the faded labels of alcoholic preparations, begun by Miss Martin last year, was renewed this year, and will be Digitized by VjOOQ IC 8 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. continued during the coining year. Professor Kingsley of Tufts college, to whom our collection of Amphipoda was loaned for investigation, reports that about one fourth of the specimens have been named and that the whole collection is in good condition. Birds and Mammals. The New England collection of birds was removed from the cases during the early part of this official year and the new backs put into the cases with its new shelving brings the specimens close to the glass where they can be readily seen and examined. The specimens were removed and returned to their places by Miss Martin, and the arrangement was subsequently revised by Mr. Batchelder. The Mammalia were also removed and the cases in the main gallery immediately adjoining the entrance to Room N were deepened. This enabled Miss Martin to place the collection of New England mammals outside of Room N and thus make suffi- cient space for the extension of the collection of birds within that room. The arrangement of the mammals also was revised by Mr. Batchelder. A railing case of improved pattern was built early in the summer along the east side of the upper gallery for the accommodation of the larger specimens of birds' nests and eggs promised by the Nuttall ornithological club. Mr. Batchelder removed from their cases all of the old collection of birds' nests and eggs, which had for some years been in bad condition, and cleaned and rearranged them, weeding out undesirable material during the process. A part of the donation of the Nuttall club has been received and placed in the cases by Mr. Batchelder. This makes a fine appearance, and the sincere thanks of the 'Society are due to the members of the Club for a donation which has so greatly improved this part of our New England collections, and to Mr. Batchelder for the expense and trouble incurred in appropriately mounting these handsome specimens. Unfortunately Mr. 15atchelder met with a serious accident a few weeks since, and this has stopped for the present the active work he was doing in the collections under his charge, but, as this report shows, his efforts this year had already laid the Society under obligations which should be gratefully acknowledged. Digitized by VjOOQ IC HYATT : REPORT OF THE CURATOR. 9 During the summer the Curator succeeded in obtaining a rare species of whale, Meaoplodon hidens^ noticed at the meeting of November 2, 1898. The bones of this specimen are now being prepared for the collection. Laboratory. The room in our basement has been this year used as in previous years by the classes of the Boston univei-sity and Teachers* school of science. A number of new diagrams have been made for use in the laboratory and the specimens have been looked after by the Curator and Miss Martin. Remarks. A much larger amount of work upon the collections not reported upon above has been done this year than usual, especial efforts having been made to bring into order and work up our miscella- neous alcoholic and dry materials. It should also be noted here that these annual reports take no notice whatever of this sort of work nor of similar work upon the various collections, such as the general inspection and repair of dried specimens, the poisoning of the same twice in each year, the inspection of alcoholic sj)ecimen8 and the refilling of bottles once in each year, the reception and preparation of specimens, and other items of daily routine. The Museum has been visited during private days by 340 pupils and teachers, representing 12 schools, all of whom have as heretofore been admitted without charge. It should also be noted here that this year as in previous years several artists have been allowed to draught specimens or study them in the Museum on closed days free of charge. Teachers' School of Science. Mr. Grabau has continued his excursions and lectures upon zool- ogy. The Satm-day field courses have been regularly carried on. Four excursions were made in May and June, 1898, in addition to those reported upon in the last annual report. These took in the more important localities on our coast and the fresh-water ponds of West Cambridge. A three days' excursion to Wood's IIoU was Digitized by VjOOQ IC 10 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. also made and notwithstanding the length of time consumed and the expense of such a prolonged trip, this was taken by ten persons. The days were spent in the general study of the fauna of the shores and the evenings in surface work. Reports and theses on the speci- mens found and preserved were subsequently prepared, each student selecting a distinct group of animals. Another longer excursion of ten days was made to Oouldsbo rough on Frenchman's Bay, Maine. Several boats together with dredging apparatus were placed at the command of this party. The habits of marine animals were the principal subject for observation on this trip. As on previous trips an effort was made to collect extensively, and lists of the species are in preparation as part of the theses of the members of the clans. Several members of this class made an indei)endent excursion of several weeks' duration to the island of Monhegan and collected a large number of marine invertebrates, some of which have not been heretofore reported from this part of the coast. The autumn course of field lessons in zoology also given by Mr. Grabau began Septem- ber 10, and ten lessons were given ; but owing to the exceptional inclemency of the weather, eight of these were necessarily trans- formed into laboratory exercises. The average attendance was twenty. The Hale house natural history club, founded by members of these classes, still continues to work in cooperation. All remunera- tion ceased with the close of the spring courses, and since then Mr. Grabau has been conducting the instruction without pay, excepting in so far as the members of the classes have defrayed his traveling expenses and the cost of circulars and correspondence. Unless some substantial support can be obtained for this work, it must cease, and the fruit of years of preparation as well as several years of direct work will be lost. A few hundred dollars per year would enable us to carry on these classes which must eventually exercise a large influence upon our efforts to investigate and excite a general interest in the natural history of New England. Our work upon the geology of this region is in full swing with two able men, one constantly investigating and publishing, the other at the head of a large, enthusiastic, and increasing class of teachers in the public and private schools. The zoology has also been successful and will do quite as much for the biological side of our work, if it meet with equal patronage. The spring work in zoology has been begun with one regular Digitized by VjOOQ IC m'ATT: REPORT OF THE CURATOR. 11 excursion to Nahant and a number of informal trips, and will be continued for the present. Professor Barton had arranged to give ten field lessons in geology to the pupils of the Boston normal school but the exceptionally bad weather in the autumn of 1898 reduced these to four. This class consisted of twenty young ladies and was conducted without remuneration. The indifference of the authorities to the continu- ance of this course, in spite of the exertions of the teacher in charge and of the head of the school to obtain an appropriation for this purpose, exhibits quite j)lainly the estimate in which science is held by the government of the public schools. Professor Barton states that these pupils, all of whom are to be teachers in the })ublic schools, come to him exceedingly well prepared for the work by their previous training and are very enthusiastic and successful, but as he has now been carrying these courses for several years without remuneration he will be unable to continue after the series of lessons now being given is finished. The spring course has begun and will be reported upon next year. The field lessons in geology in the autumn of 1898, altliougli no longer supported by the Lowell fund, were conducted voluntarily by Professor Barton in order that the teachers who had counted upon having them should not be disappointed and in the hope that means would be found to keep this important part of the work of the school from being given up altogether. Luckily our appeals for help were in this case answered by a generous friend of the Society, who has promised a sum sufficient to cany on these courses through the spring and autumn of 1899. The autumn course began September 17, 1898, and ended November 19, ten lessons having been given. The whole number of applications was 110 and the average attendance notwithstand- ing the b«id weather was twenty-seven. There was but one fair day out of the ten Saturdays upon which lessons were held. Most of these lessons occupy one half day but some, like those to Marble- head, Rockport, Haverhill, Fitchburg, require a whole day and those to Iloosac Tunnel and Mt. Ilolyoke took three days each. The spring course of 1899 has begun and will be noticed fully in the next annual report. A party of seventeen under Professor Barton's direction visited Nova Scotia during the summer of 1898, spending about three weeks in making a study of the geology, mineralog}% mining, and Digitized by VjOOQ IC 12 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. natural scenery of that country. Among the prominent points visited were the coal mines at the Joggins, the iron mines at Londonderry, the gold mines at Montagu, the gypsum quames at Windsor, the famous mineral localities at Partridge Island, Wason's Bluff, and Cape Blomidon, and the beautiful scenery around Hali- fax and the Annapolis Valley. Many courtesies were shown the party by the managers of the different mines and by others. The members of the class were charged with the teacher's expenses and a fee of five dollars each. A similar excursion into western Massachusetts and westwards as far as Niagara Falls has been planned for the coming summer. Lowell Fbee Courses. The regular spiing course of ten field lessons in geology car- ried on by Professor Barton began April 23, 1898, and ended June 25. Bad weather interfered with the work and reduced the attendance to less than it ever has been before. While in 1897 for the same course it was 34, this spring the average was only 10. The Curator regrets to announce that the Trustee of the Lowell institute having decided to discontinue the out-of-door work of the Lowell free course in the Teachers' school of science, will contrib- ute in future only to the support of lectures during the winter. To this determination he has been led by considerations of general policy which in no way reflect upon the value or success of the field courses. It would be difficult to overstate the obligations of this Society to Mr. Augustus Lowell. This gentleman as Trustee of the Lowell institute has carried on for many years past most of the les- sons given by the Teachers' school of science and preceding this time gave freely towards the maintenance of courses of free public evening lectures. He assumed the support of the field lessons in geology in 1890, and they have since then been a part of the work of the Lowell institute and have thus been able to build up results that ought to secure for them some permanent foundation. A new four years' course (1 20 hours) in geology under Profes- sor Barton began during the past winter. This course like that which preceded it will embrace at least fifteen lessons of two hours each, thirty hours for each year of the course. The subjects are to Digitized by VjOOQ IC HYATT: REPORT OF THE CURATOR. 13 be as follows : mineralogy, lithology and dynamic geology, struc- tural geology, and historical geology. The first year's course on mineralogy began on December 5, 1898, and ended on April 15, 1899. This included one extra lecture which was given by special request at the end of the regular lessons and the final examination, making seventeen exercises, in all thirty-four hours of instruction instead of the required thirty hours. The entire number of applications received for this course was 145, and special provision was made to accommodate 130 of these. The unusual prevalence of sickness during the winter caused the loss of several from the class and a few also withdrew as they found the work more difticult than they could attend to in connection with the necessary work in their schools. The average attendance for the course was 106. Of these 72 took all the exam- inations, including the final. The instruction included one intro- ductory lecture upon chemistry, four upon crystallography, and ten upon mineralogy proper. They were given by means of lectures, supplemented by a complete series of minerals illustrating the com- moner species, about 130 in number. Each two members of the class had one tray between them, containing all the species dis- cussed at any single lesson, and this enabled each member of the class to study with the specimens directly in hand. A complete set of notes was also supplied each member of the class, for which a cost price was charged. At each lesson except the first and the last the first half hour was devoted to an examination covering all the ground previously passed over, and at the end of the course a final examination of three hours was given which was so arranged as to present a concise r4sum6 of the whole subject. Dr. R. W. Greenleaf gave a course of fifteen lessons of two hours each upon the principles of the classification of flowering plants. One hundred species of native plants were collected, dried, mounted, and divided into sets by Miss E. B. Bryant, and each student was provided with one of these. Fresh material was also purchased when required for the use of the class. The class was especially mdebted to Miss Helen Sharp, formerly an assistant in this course, for the use of her collection of water-color drawings of 901 sheets, 774 of which represent American plants, the actual number of American species being 649. These drawings will shortly be assembled and exhibited in our laboratory, where they can be seen as a whole and properly inspected. This free public exhibition will Digitized by VjOOQ IC 14 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. take place between the 9th and 13th of May, and will be open daily from 12 to 4.30. There were forty persons in the class and the average attendance was twenty-seven. Twenty-one persons took the examination, out of which number nine passed with honor, seven with credit, four simply passed, and one failed. The Curator gave the fourth year of a five years' coui-se in zoology, consisting this year of sixteen lessons of two hours each. The subjects were Myriopoda, Arachnozoa, and Insecta through Coleoptera, leaving the Lepidoptera and other so-called higher orders to be treated next year. The number of tickets issued was forty-five and the average attendance twenty-five. The excessively bad weather had a depressing effect upon this course as upon most others. Thirteen only took the final examination and all of these passed. REPORT OF THE SECRETARY AND LIBRARIAN, SAMUEL HENSHAW. Membekship. During the past year eighty-four Corporate members have been elected by the Council. Nine Corporate members have become Life members. One Honorary member, James Hall, has died. Three Corresponding members, George Baur, Oliver Marcy, and O. C. Marsh, have died. From the list of Corporate members we have lost by death five names, John Cummings, Edraond E. Fernald, James I. Peck, Henry P. Quincy, and Edward W. Roper. Two Corporate members have resigned, and the names of four- teen have been stricken from the list for non-])ayment of dues. The membership of the Society, corrected to May 3, 1899, con- sists of 10 Honorary, 139 Corresponding, and 420 Corporate mem- bers, a total of 569. There are 17 Patrons. The numl>er of Corporate members reported last year was 363 ; twenty-three is the greatest numl>er elected in any of the antecedent seven years, and for the essential increase this year the Society is indebted mainly to the abundant faith and efticient work of its President, Professor Minot. Digitized by VjOOQ IC HENSHAW: REPORT OF SECRETARY AND LIBRARIAN. 15 The Dames of the Corporate members elected and the dates of their election are as follows : — Arthur Amory, Feb. 15, 1899. S. Reed Anthony, Feb. 15, 1899. Francis Blake, Feb. 15, 1899. Mrs. T. M. Brewer, Feb. 15, 1899. Shepherd Brooks, Feb. 15, 1899. Edward I. Browne, Feb. 15, 1899. Arthur T. Cabot, Mar. 29, 1S99. William B. Cabot, Apr. 26, 1899. Alvin Carl, Mar. 29, 1899. Mrs. J. B. Case, Mar. 29, 1899. John S. Clark, Mar. 29, 1899. Collier Cobb, Feb. 15, 1899. Miss Helen Collamore, Feb. 15, 1899. T. Jefferson Coolidge, Jr., Feb. 15, 1899. Charles U. Cotting, Feb. 15, 1899. Miss S. H. Crocker, Feb. 15, 1899. Miss Ada Dana, Oct. 19, 189><. Andrew McF. Davis, P\»b. 15, 1899. James C. Davis, Mar. 29, 1899. Gordon Dexter, Feb. 15, 1899. Thomas Doliber, Mar. 29, 1899. Richard S. Dow, Feb. 15, 1899. Raymond B. Earle, Feb. 15, 1899. G. B. Eisenhard, Oct. 19, 1898. Mrs. J. W. Elliot, Mar. 29, 1899. George W. Fitz, Feb. 15, 1899. Charles F. Folsom, Feb. 15, 1899. Eugene N. Foss, Mar. 29, 1899. Charles Fry, Mar. 29, 1899. George M. Garland, Feb. 15, 1899. George W. Gay, Feb. 15, 1899. Harold B. Goodrich, Mar. 29, 1899. Mrs. Mary T. Gorham, Feb. 15, 1899. Francis C. Gray, Feb. 15, 1899. Mrs. Caroline W. Greenough, Mar. 29, 1899. Charles P. Greenough, Feb. 15, 1899. Elisha H. Gregory, Jr., Oct. 19, 1898. Miss Minna B. Hall, Mar. 29, 1899. Digitized by VjOOQ IC 16 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Francis R. Hart, Feb. 15, 1899. Charles E. Hellier, Feb. 15, 1899. Mrs. Augustus Hemenway, Feb. 15, 1899. Joseph P. B. Henshaw, Feb. 15, 1899. Hibbert \V. Hill, Mar. 29, 1899. Robert C. Hooper, Feb. 15, 1899. Theodore Hough, Mar. 29, 1899. John E. Hudson, Feb. 15, 1899. Willard P. Hunnewell, Mar. 29, 1899. Charles E. Inches, Feb. 15. 1899. Charles F. Jenney, Oct. 19, 1898. Miss Marian H. Judd, Feb. 15, 1899. William Lawrence, Feb. 15, 1899. George W. Lee, Dec. 21, 1898. William C. Loring, Feb. 15, 1899. Charles Lowell, Apr. 26, 1899. Charles P. Lyman, Apr. 26, 1899. Vernon F. Marsters, Mar. 29, 1899. Asa E. Mattice, Mar. 29, 1899. Laurence Minot, Feb. 15, 1899. William Minot, Feb. 15, 1899. William J. Moenkhaus, Dec. 21, 1898. Miss Margaret W. Morley, Feb. 15, 1899. Elisha W. Morse, Dec. 21, 1898. Mrs. Edith J. Nichols, Mar. 29, 1899. Edgar W. OUve, Oct. 19, 1898. Edward C. Perkins, Feb. 15, 1899. John C. Phillips, Oct. 19, 1898. Dudley L. Pickman, Feb. 15, 1899. David Pingree, Feb. 15, 1899. William H. Ruddick, Dec. 21, 1898. Mrs. T. E. Ruggles, Oct. 19, 1898. Dudley A. Sargent, Mar. 29, 1899. Alfred L. T. Schaper, Feb. 15, 1899. Frederic F. Smith, Feb. 15, 1899. John E. Thayer, Feb. 15, 1899. Augustus L. Thorndike, Mar. 29, 1899. Miss Mary M. Webster, Feb. 15, 1899. Clarence M. Weed, Oct. 19, 1898. William P. Wesselhoeft, Feb. 15, 1899. Digitized by VjOOQ IC HEXSHAW; REPORT OF SECRETARY AND LIBRARIAN. 17 Henry M. Whitney, Feb. 15, 1899. George Wigglesworth, Feb. 15, 1899. Guy M. Winslow, Feb. 15, 1899. Edward S. Wood, Mar. 29, 1899. Miss Elvira Wood, Dec. 21, 1898. Frederick A. Woods, Dec. 21, 1898. Meetings. Fourteen regular meetings including the Annual meeting have been held during the year; one Special meeting has been held. The attendance shows an average of 55 plua^ the largest at any one meeting being 144, the smallest 24 ; the largest last year was 352, the smallest 24. Twenty-six communications have been made by twenty-two persons ; of the twenty-two persons seven have not previously spoken at our meetings. Twelve papers have been presented by title. The substitution of electricity for gas in the lecture room and the availableness of the lantern for even a modicum of illustration are features that should add to the success of future meetings. The Society and especially the Secretary are indebted to Dr. R. T. Jackson for his kindness in serving as recorder at meetings in January and February. The meetings, attendance, and communications have been as follows : — May 4, 1898. Annual meeting. Forty-two persons present. Reports 'of the Curator, Secretary, Librarian, Treasurer, and Trustees. Mr. J. Edmund Woodman. Geological history of the gold- bearing slates of Nova Scotia. Mr. M. L. Fernald. The genus Antennaria in New England. (By title.) Prof. C. S. Minot. On the veins of the Wolffian bodies in the pig. (By title.) Dr. P. P. Calvert. The odonate genus Macrothemis and its allies. (By title.) Mr. T. D. A. Cockerell. The North American bees of the genus Prosapis. (By title.) Digitized by VjOOQ IC 18 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. May 18, 1898. General meeting. Thirty-two persons present. Dr. T. A. Jaggar, Jr. The Absaroka range of the Rocky Mountains. November 2, 1898. General meeting. Thirty-seven persons present. Prof. Alpheus Hyatt. A rare whale from the Massachusetts coast. Dr. R. T. Jackson. Localized stages in growth in plants and animals. Dr. G. N. Calkins. Some hydroids from Puget Sound. (By title.) Mr. Reginald Heber Howe, Jr. North American wood frogs. (By title.) November 16, 1898. General meeting. Eighty-five persons present. Prof. W. Z. Ripley. Racial characteristics of the Jews — a study in physical anthropology. December 7, 1898. General meeting. Forty-five persons present. Dr. G. H. Parker. On the coronary circulation in some fishes. Prof. M. A. Willcox. On the occipital region of the head in the European trout. Prof. J. S. Kingsley. Ear-bones of mammals. Prof. C. S. Minot. Prof. O. van der Stricht's demonstration of the human ovum. Prof. C. S. Minot. Some classical embryological monographs. December 21, 1898. General meeting. Forty-three persons pres- ent. Prof. Alpheus Hyatt. Evolution and migrations of the land shells of the Hawaiian Islands. January 4, 1899. General meeting. Thirty-six persons present. Mr. J. B. Wood worth. The geology and geography of the Richmond area in Virginia. January 18, 1899. General meeting. Forty- four persons present. Dr. Frank Russell. An account of the Apache festival of San Antonio. February 1, 1899. General meeting. Seventy-six persons present. Dr. R. A. Daly. A geological tour in Russia, Finland, the Volga, the Caucasus, and the Crimea. February 15, 1899. General meeting. Thirty-two persons present. Mr. Myron L. Fuller. Rapidity of sand-plain growth. Prof. W. O. Crosby. Geology of the main dam and tunnel of the Metropolitan water works, near Clinton, Mass. Digitized by VjOOQ IC HENSHAW: REPORT OF SECRETARY AND LIBRARIAN. 19 February 17, 1899. Special meeting. Eighty-six persons present. Prof. H. F. Osbom. Evolution of the Mammalia in North America. March 1, 1899. General meeting. Thirty-four persons present. Dr. T. A. Jaggar, Jr. The intrusive rocks of the Black Hills. March 15, 1899. General meeting. Twenty-eight persons present. Mr. E. C. Jeffrey. Development and affinities of the genus Equisetum. Dr. C. R. Eastman. Some new North American fossil fishes. April 5, 1899. General meeting. Sixty-three persons present. Report of the Nominating committee. Prof. C. E. Fay. The Alpine features of the Canadian Rockies. April 26, 1899. General meeting. One hundred and forty-four persons present. Dr. W. McM. Wood worth. Samoa and the Samoans. Mr. G. M. Allen. Notes on the reptiles and amphibians of Intervale, N. H. (By title.) Dr. B. L. Robinson and Mr. J. M. Greenman. Contributions from the Gray herbarium of Harvard university, no. 17. (By title.) Dr. Gary de N. Hough. Studies in Diptera Cyclorhapha. (By title.) Mr. F. N. Balch. List of marine MoUusca of Cold Spring Harbor, Long Island, with descriptions of one new genus and two new species of nudibranchs. (By title.) Prof. G. E. Stone. The flora of Lake Quinsigamond. (By title.) PuBLirATIONS. The following publications have been issued during the year : — Localized stages in development in plants and animals, by Robert T. Jackson. Memoirs, vol. 5, no. 4, 65 pp., 10 plates, 5 cuts. The development, structure, and aftinities of the genus Equisetum, by Edward C. Jeffrey. Memoirs, vol. 5, no. 5, 36 pp., 5 plates. The genus Antennaria in New England, by Merrit L. Fernald. Proceedings, vol. 28, no. 8, 13 pp. Notes on a Carboniferous boulder train in eastern Massachusetts, by Myron L. Fuller. Proceedings, vol. 28, no. 9, 14 pp., 1 cut. Digitized by VjOOQ IC 20 PROCEEDDfGS : BOSTON SOCIETY NATURAL HISTORY. On the veins of the Wolffian bodies in the pig, by Charles S. Minot. Proceedings, vol. 28, no. 10, 10 pp., 1 plate, 1 cut. Proceedings of the annual meeting. May 4, 1898. Proceedings, •vol. 28, no. 11,26 pp. The odonate genus Macro themis and its allies, by Philip P. Calvert. Proceedings, vol. 28, no. 12, 32 pp., 2 plates. Some hydroids from Puget Sound, by Gary N. Calkins. Pro- ceedings, vol. 28, no. 13, 35 pp., 6 plates. North American wood frogs, by Reginald Heber Howe, Jr. Pro- ceedings, vol. 28, no. 14, 6 pp. Studies in the gold-bearing slates of Nova Scotia, by J. Edmund Woodman. Proceedings, vol. 28, no. 15, 33 pp., 3 plates, 1 cut. Moniloporidae, a new family of Palaeozoic corals, by Amadeus W. Grabau. Proceedings, vol. 28, no. 16, 16 pp., 4 plates. Library. The additions to the library have been as follows: — 8vo. ■tto. Folio. Total. Volumes 278 31 309 Parts 2,208 373 1 2,582 Pamphlets 651 26 25 702 Maps 45 45 Total 3,137 430 71 3,638 The library contains 24,879 volumes, 1,378 incomplete (including current) volumes, and 12,812 pamphlets. By exchange, gift, or purchase we have added twenty-eight serials: — Agricultural experiment stations of California, Con- necticut, Delaware, Florida, Georgia, Illinois, Maine, Maryland, Minnesota, New Mexico, Rhode Island, and South Carolina ; Bird lore ; Blue Hill observatory bulletin ; Bulletin Cooper ornithologi- cal club, Santa Clara ; Kansas university geological survey ; Lloyd mycological museum, Cincinnati ; Mineral collector. New York ; The museum, Albion, N. Y. ; Natural history association of Mira- michi, Chatham, N. "B. ; Nature study in school. West Newton ; Naturwissenschaftliche gesellschaft in Winterthur, Winterthur ; Ottawa literary and scientific society, Ottawa ; Pasadena academy of sciences, Pasadena; Queensland agricultural journal, Brisbane; Digitized by VjOOQ IC HENSHAW: REPORT OF SECRETARY AND LIBRARIAN. 21 Rhodora, Boston ; Washington academy of science, Washington ; Yorkshire naturalists' union, Leeds. The Society now exchanges its publications with 432 scientific institutions and periodicals. One thousand and thirty-seven books have been borrowed by 115 persons ; 480 volumes have been borrowed for use in the building, and the library has been consulted 450 times. Three hundred and sixty-eight volumes have been bound in 824 covers ; 362 pamphlets have been bound. Twenty-five volumes of the Verhandlungen zoologisch-botanische gesellschaft, Wien, have been indexed ; current volumes of serials previously indexed are indexed as received. Taking advantage of the aid afforded by a generous though anonymous patron, we have bound all the unbound books and pamphlets on several alcoves in the back library; the alcoves selected being those nearest the windows and thus the most exposed to light and dust. Walker Prizes. The subjects selected by the Walker Prize Committee for 1899 were : — 1. Is there fundamental difference between 'equation division' and * reduction division ' in the division of cells ? 2. The phenomena and laws of hybridization. The only essay received, one on the relations between the hybrid and parent forms of echinoid larvae, having been published in Philosophical transactions of the royal society of London, could not be considered by the Committee, as in all cases the memoirs are to be based upon original and unpublished work. It is greatly to be regretted that the annual awards of Walker prizes are so frequently omitted, and it is suggested that the selec- tion of subjects of broader scope would secure more general competition. The subjects for the award in May, 1900, are : — 1. Stratigraphy and correlation of the sedimentary formations of any part of New England. 2. A study in palaeozoic stratigraphy and correlation. Digitized by VjOOQ IC 22 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. 9* *l il li. > o Q Pi ft W PS" CO O o H PS s H < h O o H CO O 00 00 S o .SS 1- :c S|s|||s:S^: I S3S2S§?;§$|8882S?li S ^ CO 1-H "i ^ o k iS §k <2k aS ? = S.2 is M 9 si S^ C' 18 V « ^•s nn 8 S8 i?5 81? 1 "§ i gh- Ol 1 ri •» \^ •» "i I CO r i I • !•« i ! J I : : : : :i : :^: i >- •t: •ill .=5 i :1 i e2 H £ c « p «* .^ . . ;- • £" 2 s •J 2?-- =11 1 1-3 1. .:-.;.. 1 is g m 5 d 1 ll?^slcS5 » E % < i ? 233*"53*-35 2 3 2 * 3 Si W Ci/ a> •) u W e8 cc.ti . a 8- I «>2 Digitized by VjOOQ IC ANNUAL MEETING. 23 The reports of the Trustees and of the Auditing Committee were read and it was voted to accept and adopt the several reports. The Society then proceeded to ballot for officers for 1899- 1900. Messrs. R. P. Bigelow and R. Hayward were appointed to collect and count the votes. They reported the election of PRESIDENT, CHARLES SEDGWICK MINOT. VICE-PRESIDENTS, CHARLES p. BOWDITCH. HENRY W. HAYNES. WILLIAM G. FARLOW. CURATOR, ALPHEUS HYATT. SECRETARY, samlt:l henshaw. TREASURER, EDWARD T. BOUVE. LIBRARIAN, SAMUEL HENSHAW. councillor for three tears, Miss Cora H. Clarke. George H. Parker. Robert T. Jackson. A. Lawrence Rotch. J. Arnold Lowell. William F. Whitney. Edward 8. Morse. J. B. Woodworth. councillor for one tear, H. C. BuMPUs. Digitized by VjOOQ IC 24 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. OFFICERS FOR 1899-1900. PRESIDENT, CHARLES SEDGWICK MINOT. VICE-PRESIDENTS, CHARLES p. BOWDITCH. HENRY W. HAYNES. WILLIAM G. FARLOW. CURATOR, ' ALPHEUS HYATT. SECRETARY, SAMUEL HENSHAW. TREASURER, EDWARD T. BOUVfe. LIBRARIAN, SAMUEL HENSHAW. COUNCILLORS FOR THREE YEARS, Miss Cora H. Clarke. Robert T. Jackson. J. Arnold Lowell. Edward S. Morse. George H. Parker. A. Lawrence Rotch. William F. Whitney, j. b. woodworth. councillors for TWO YEARS, Charles B. Davenport. James H. Emerton. William A. Jeffries. George G. Kennedy. Augustus Lowell. Miss Susannah Minns. Thomas A. Watson. Samuel Wells. S. L. Abbot. William S. Bryant. H. C. Bumpus. William M. Davis. COUNCILLORS FOR ONE YEAR, Miss Catharine I. Ireland. Benjamin Joy Jeffries. N. T. Kidder. Willla^m H. Niles. George L. Goodale. COUNCILLORS ez-officUs^ F. W. Putnam. Samuel H. Sc udder. Digitized by VjOOQ IC LIST OF MEMBERS. 25 LIST OF MEMBERS. HONORARY MEMBERS. Alexander Agassiz, Cambridge Adolph Bastian, Berlin. John William Dawson, Montreal. Michael Foster, Cambridge. Ernst Haeckel, Jena. Joseph D. Hooker, London. Albert v. KoUiker, Wtlrzburg. Henri Lacaze-Duthiers, Paris. Edward B. Tylor, London. Kudolph Vlrchow, Berlin. WiUiam Allen, James Anderson, Francis Archer, Francesco Ardissone, Loring W. Bailey, A. S. Baldwin, Mariano Barcena, A. Constantino Barry, Charles E. Beecher, Hermann Behr, £. Tan Beneden, William G. Binney, Nathaniel H. Bishop, William P. Blake, George A. Boardman, William H. Brewer, William K. Brooks, John Brown, Giovanni Capellini, Antonio del Castillo, Ferdinand Cohn, Guido Cora, CORRESPONDING MEMBERS. Boston. Liverpool, Eng. Liverpool, Eng. Milan, Italy. Fredericton, N. B. Jacksonville, Fla. Mexico, Mex. Prairie du Sac, Wis. New Haven, Ct. San Francisco, Cal. Li^ge, Belgium. Burlington, N. J. Lake George, N. Y. New Haven, Ct. Calais, Me. New Haven, Ct. Baltimore, Md. New York, N. Y. Bologna, Italy. Mexico, Mex. Breslau, Germany. Rome, Italy. Digitized by VjOOQ IC 26 PKOCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Elliott Coues, John M. Coulter, Hermann Credner, EzraT. Cresson, Josiah Curtis, Henry Davis, William Boyd Dawkins, William Dean, Anton Dohm, Sanford B. Dole, Henry E. Dresser, Paul B. Du Chaillu, W. T. T. Dyer, Arthur M. Edwards, William H. Edwards, D. G. Elliot, Sigmund Exner, Roswell Field, William H. Flower, F. Fouqu6, M. Ganin, J. T. Gardner, Albert Gaudry, Archibald Geikie, James Geikie, Hans B. Geinitz, Theodore N. Gill, Augustus R. Grote, Albert C. L. G. GUnther, John T. Gulick, L. H. Gulick, Edwin Harrison, James Hector, Angelo Heilprin, George Henslow, Henry Y. Hind, Charles H. Hitchcock, John Hjaltalin, W. J. Hoffman, W. J. Holland, Bernard A. Hoopes, A. W. Howitt, Oliver P. Hubbard, Samuel Hubbard, Wadiington, D. C. Chicago, 111. Leipsic, Germany. Philadelphia, Pa. Washmgton, D. C. McGregor, Iowa. Manchester, Eng. Bangkok, Siam. Naples, Italy? Honolulu, H. I. Kent, Eng. New York, N. Y. London, Eng. Newark, N. J. Coalburg, W. Va. Chicago, 111. Vienna, Austria. GUI. London, Eng. Paris, France. Nice, France. Washington, D. C. Paris, France. Edinburgh, Scotland. Edinburgh, Scotland. Dresden, Germany. Washington, D. C. Hildesheim, Germany. London, Eng. Osaka, Japan. Honolulu, Oahu. St. Louis, Mo. Wellington, N. Z. Philadelphia, Pa. London, Eng. Windsor, N. S. Hanover, N. H. Rejkyavik, Iceland. Reading, Pa. Alleghany, Pa. Philadelphia, Pa. Melbourne, Vict. Hanover, N. H. San Francisco, Cal. Digitized by VjOOQ IC LIST OF MEMBERS. 27 Christopher Johnson, David S. Jordan, Clarence King, John King, Cornelius Kollock, A. Kowalewsky, Carl Kupffer, Arnold Lang, E. Ray Lankester, Joseph Leconte, R. von Lendenfeld, J. Peter Lesley, A. M. L6vy, F. W. Lewis, Franz Leydig, Christian F. Ltitken, Richard Lydekker, Robert McLachlan, E. J. Marey, Paul Mayer, Joseph B. Meader, C. Hart Merriam, Charles L. Metz, Alphonse Milne-Edwards, S. Wier Mitchell, John Murray, Francis P. Nash, Alfred Newton, Henry F. Osbom, C. R. von Osten Sacken, Emile Oustalet, TTiomas F. Perley, F6lix Plateau, Edward B. Poulton, John W. Powell, Raphael Pumpelly, Richard Rathbun, Ferd von Richthofen, Robert Ridgway, Heinrich Rosenbtosch, Henri de Saussure, C. M. Scammon, Baltimore, Md. Lelstfid Stanford, Cal. Washington, D. C. Boone, Iowa. Cheraw, S. C. St. Petersburg, Russia. Munich, Germany. Zurich, Switzerland. London, Eng. Berkeley, Cal. Czernowitz, Austria. Milton. Paris, France. PhUadelphia, Pa. Bonn, Germany. Copenhagen, Denmark. Harpenden, Eng. London, Eng. Paris, France. Naples, Italy. Stoneham. Washington, D. C. Madison ville, Ohio. Paris; France. Philadelphia, Pa. Edinburgh, Scotland. Geneva, N. Y. Cambridge, Eng. New York, N. Y. Heidelberg, Germany. Paris, France. Portland, Me. Li6ge, Belgium. Oxford, Eng. Washington, D. C. Newport, R. I. Washington, D. C. Bonn, Germany. Washington, D. C. Heidelberg, Germany. Geneva, Switzerland. Washington, D. C. Digitized by VjOOQ IC 28 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Philip L. Sclater, A. R. C. Selwyn, William Sharswobd, Hamilton L. Smith, Hermami Snellen, Armand Thieleus, Tamerlan Thorell, William Trelease, Gustav Tschermak, Philip R. Uhler, Addison E. Verrill, Sydney H. Vines, W. Waagen, Henry A. Ward, R. A. Ward, Carl Wedt, August Weismann, George M. Wheeler, WUliam T. White, R. P. Whitfield, Robert E. E. Wiedersheim, Burt G. Wilder, C. S. Wilkinson, Edmund B. Wilson, Henry Woodward, J. J. Woodward, Ferdinand Zirkel, Carl A. Zittel, London, Eng. Ottawa, Can. Philadelphia, Pa. Geneva, N. Y. Utrecht, Holland. Tirlemont, Belgium. Montpellier, France. St. Louis, Mo. Vienna, Austria. Baltimore, Md. New Haven, Ct. Oxford, Eng. Vienna, Austria. Rochester, N. Y. Troy, N. Y. Vienna, Austria. Freiburg, Germany. Washington, D. C. New York, N. Y. New York, N. Y. Freiburg, Germany. Ithaca, N. Y. Sydney, N. S. W. New York, N. Y. London, Eng. Washington, D. C. Leipsic, Germany. Munich, Germany. CORPORATE MEMBERS. Samuel L. Abbot, M. D., John E. Alden, Jane Alexander, Henry F. Allen, Joel A. Allen, Edward P. Allis, Jr., Arthur Amory, Robert Amory, M. D., S. Reed Anthony, Nathan Appleton, William S. Appleton, Anuetta F. Armes, Edward P. Austin, 00 Mt. Vernon St. Newton. 91 Mt. Vernon St. Jamaica Plain. Absent. Milwaukee, Wis. 133 Marlborough St. 279 Beacon St. 63 State St. Somerset Club. 462 Beacon St. Dorchester. Absent. Digitized by VjOOQ IC LIST OF MEMBERS. 29 Lucas Baker, Francis N. Balch, Edward E. Bancroft, M. D., Edward A. Bangs, Outram Bangs, James M. Barnard, Walter B. Barrows, George H. Barton, Charles F. Batchelder, George W. Beaman, Mrs. George W. Beaman, Henry B. Bigelow, Joseph S. Bigelow, Jr., Robert P. Bigelow, William S. Bigelow, M. D., aarence J. Blake, M. D., Francis Blake, James H. Blake, Joseph W. Blankinship, Albert N. Blodgett, M. D., >Ir8. Alice L. Boardman, Elizabeth D. Boardman, Edward T. Bouv6, Charles P. Bowditch, Frederic C. Bowditch, Henry P. Bowditch, M. D., Mrs. Ella F. Boyd, Arthur C. Boy den, Francis W. Brewer, Mrs. Thomas M. Brewer, Willard S. Brewer, Edwin T. Brewster, William Brewster, Albert P. Brigham, Henry Brooks, Shepherd Brooks, J. Frank Brown, Edward I. Browne, Elizabeth B. Bryant, John Bryant, M. D., William S. Bryant, M. D., Charles Bullard, Williani N. Bullard, M. D., Hermon C. Biunpus, Absent. Jamaica Plain. Wellesley. 240 Beacon St. 240 Beacon St. Milton. Absent. Mass. Inst, of Technology. Cambridge. Absent. Cambridge. 261 Commonwealth Ave. 251 Commonwealth Ave. Mass.. Inst, of Technology. 60 Beacon St. 226 Marlborough St. Auburndale. Cambridge. Bozeman, Mont. 61 Massachusetts Ave. Absent. 416 Marlborough St. 436 Marlborough St. 28 State St. Brookline. Harvard Medical School. Hyde Park. Bridgewater. Hiugham. 233 Beacon St. Hingham. Andover. Cambridge. Absent. Lincoln. 02 Beacon St. 89 State St. 63 State St. AUston. Cohasset. 63 State St. Cambridge. 89 Marlborough St. Providence, R. I. Arthur T. Cabot, M. D., Edward C. Cabot, Louis Cabot, 1 Marlborough St. Brookline. Brookline. Digitized by VjOOQ IC 30 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. William B. Cabot, Mrs. Gar>' N. Calkins, Alvin Carl, Charles T. Carruth, Mrs. J. B. Case, Arthur P. Chadbourne, M. D., Montage Chamberlain, Walter G. Chase, Francis S. Child, Henr>' L. Clapp, Mabel D. Clapp, John S. Clark, T. W. B. Clark, Cora H. Clarke, Collier Cobb, Edward W. Ctxlman, Helen Collamore, Frank S. Collins, Grace E. Cooley, Algernon Coolidge, M. D., T. Jefferson Coolidge, Jr., Mrs. Elizabeth R. Cormier, Charles B. Cor>', Charles U. Cotting, Sarah H. Crocker, William O. Crosby, Charles R. Cross, Clara E. Cmnmiugs, Chestnut Hill. New York, N. Y. Jamaica Plain. 84 India St. 468 Beacon St. 226 Marlborough St. Cambridge. Brookline. Absent. Roxbury. 9 Mas.sachusetts Ave. 046 Washington St. 326 Atlantic Ave. 01 Mt. Vernon St. Chapel Hill, N.C. 63 Marlborough St. 317 Commonwealth Ave. Maiden. Wellesley. 81 Marlborough St. 184 Beacon St. Charlestown. 100 Boylston St. 249 Commonwealth Ave. 319 Commonwealth Ave. Mass. Inst, of Technology. Mass. Inst, of Technology. Wellesley. Ulric Dahlgren, William H. Dall, Reginald A. Daly, Ada Daiia, John Dane, F. Graef Darlington, Charles B. Davenport, Andrew McF. Davis, James C. Davis, Simon Davis, William M. Davis, Henry G. Denny, Franklin Dexter, M. D., F. Gordon Dexter, Gordon Dexter, Joseph S. Diller, George Dimmock, Richard E. Dodge, Charles C. Doe, Princeton, N. J. Absent. Absent. Newton. 29 Marlborough St. Absent. Cambridge. Cambridge. 70 Kilby St. 277 Beacon St. Cambridge. 08 Devonshire St. 148 Marlborough St. 65 Beacon St. 66 Beacon St. Absent. Spriugtield. Absent. South Newbur>', Vt. Digitized by VjOOQ IC LIST OF MEMBERS. 31 Thomas Doliber, Jonathan Dorr, George A. Dorsey, Richard S. Dow, Sara A. Downs, Thomas Dwight, M. D., Harrison G. Dyar, Raymond B Earle, Charles R. Eastman, Charles W. Eliot, Mrs. J. W. Elliot. Mary L. Ells, James H. Emerton, William Endicott, Harold C. Ernst, M. D., James F. Estes, William G. Farlow, M. D., Mrs. Eva D. Farquhar, Charles E. Faxon, Heiirj' H. Fay, Joseph S. Fay. Charles S. Fellows, Cliarles H. Fernald, J. Walter Fewkes, William L. W. Field, Frank S. Fiske, George W. Fitz, M. D., Augustus Flagg, Charles F. Folsom, M. D., Justus W. Folsom, Eugene N. Foss, John Foster, William Foster, Harriet .E. Freeman, Nathaniel S. French, Sophia W. French, Charles Fr>', Myron L. Fuller, Sarah S. Fuller, Thomas Gaffield, Charles W. Galloupe, T. W. Galloway, William F. Ganong, Edward G. Gardiner, John L. Gardner, Jr., Brookline. 27 School St. Chicago, 111. 27 State St. 68 Berkeley St. 236 Beacon St. Washington, D. C. Newton. Cambridge. Cambridge. 124 Beacim St. Cambridge. Clarendon St. 33 Summer St. Hanard Medical School. Absent. Cambridge. Roxbury. Jamaica Plain. 416 Beacon St. Id9 Commonwealth Ave. Fairbanks, Fla. Amherst. Absent. Milton. U. S. Court House. Cambridge. 274 Clarendon St. 16 Marlborough St. Cambridge. Jamaica Plain. 113 Broad St. Absent. 384 Commonwealth Ave. West R«>xbury. Wollaston. 21 Commonwealth Ave. Mass. Inst, of Technology. Absent. 64 Allen St. 46 Broad St. Marshall, Mo. Absent. 131 Mt. Vernon St. 61 Commonwealth Ave. Digitized by VjOOQ IC 32 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. George M. Garland, M. D., George W. Gay, M. D., B. W. Gilbert, George L. Goodale, M. D., Harold B. Groodrich, ^Irs. Mary T. Gorhain, Amadeus W. Grabau, Francis C. Gray, John C. Gray, Robert W. Greenleaf, M.D., Mrs. C. W. Greenough, Charles P. Greenough, H. S. Greenough, William D. Grier, Leon S. Griswold, 227 Newbury St. 665 Boylston St. 202 Purchase St. Cambridge. 72 Westland Ave. 108 Marlborough St. Cambridge. 7 Mt. Vernon Place. 176 Beacon St. 561 B»)ylston St. Jamaica Plain. 39 Court St. Absent. 20 Kilby St. Dorchester. Minna B. Hall, Robert W. Hall, Susan M. Hallowell, Mrs. M. L. Hammatt, Ida S. Hammerle, Edward D. Harris, T. W. Harris, Francis R. Hart, Franklin Haven, Gustavus Hay, M.D., Ellen Hayes, Henry W. Haynes, Roland Hayward, Charles E. Hellier, Augustus Hemenway, Mrs. Augustus Hemenway, Joseph P. B. Henshaw, Samuel Henshaw, Warren W. Herman, Francis H. Herrick, Ella J. Hill, Harriet A. Hill, Hibbert W. Hill, M. D., Mary H. Hinckley, John Hobbs, Walter E. Hobbs, John Hogg, Frederick S. HoUis, John Homans, M. D., Robert C. Hooper, Mrs. S. E. Hooper, Samuel A. Hopkins, '.M. D., Brookline. Cambridge. Wellesley. Hyde Park. Roxbury. Absent. Absent. Milton. 07 Mt. Vernon St. 388 Marlborough St. Wellesley. 289 Beacon St. 346 Marlboroui?h;St. 57 Etiuitable Bldg. 273 Clarendon St. 273 Clarendon St. 77 Newbury St. Cambridge. P. O. Box 1848. Cleveland, Ohio. 223 Newbury St. Belmont. 72 Pinckney St. Mattapau. 99 St. Botolph St. Stony brook. 280 Commonwealth Ave. Newton Highlands. 164 Beacon St. 448 Beacon St. Cambridge. 235 Marlborough St. Digitized by VjOOQ IC LIST OF MEMBERS. 33 Gany de N. Hough, M.D., Theodore Hou^, Helen Hubbard, John G. Hubbard, L. L. Hubbard, John E. Hudson, Henry S. Himnewett, VVillard P. Huimewell, Alpheus Hyatt, Charles E. Inches, M. D., Catherine I. Ireland, John G. Jack, John C. Jackson, Robert T. Jackson, William D. Jackson, Thomas A. Ja^jgar, Jr., B. Joy Jeffries, M. D., William A. Jeffries, Charles W. Jenks, Charles F. Jenney, Isabel L. Johnson, Samuel Johnson, Marian H. Judd, Charles S. Kendall, Mrs. Caroline A. Kennard, Getjrge G. Kennedy, M. D., Harris Kennedy, M. I)., Nathaniel T. Kidder, John S. Kingsley, New Bedford. Mass. Inst, of Technology. (^harlestown, N. H. Brookline. Absent. 126 Milk St. 9 Park St. 261 Commonwealth Ave. Cambridge. 380 Beacon St. Cambridge, Jamaica Plain. Absent. 3.S Gloucester St. Bridgewater. Cambridge. 16 Chestnut St. 126 Beaccm St. Bedford. Hyde Park. 467 Massachusetts Ave. 7 Commonwealth Ave. 186 Conmionwealth Ave. 91 Federal St. Brookline. Readville. Readville. Milton. Tufts College. F. D. Lambert, Alfred C. Lane, Amory A. Lawrence, William Lawrence, Geora^e W. Ixie, David F. Lincoln, M. D., James L. Little, William R. Livennore, William C. Loring, Augustus Ivowell, Charles Lowell, James Arnold Lowell, Mrs. Louisa F. Lowery, Arthur T. Lyman, Charles P. Lyman, M. D., Auburn, Me. Houghton, Mich, 69 (\)nunonwealth Ave. 122 (Commonwealth Ave. Brookline. 73 Pinckney St. Brookline. P. O. Box 168. 2 Gloucester St. 63 State St. 149 Beacon St. 297 Beacon St. AbstMit. P. O. Box 1717. 60 Village St. Digitized by VjOOQ IC 34 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. George H. Mackay, W. Duncan MoKim, J. Playfair McMurrich, W. D. McPherson, B. Pickman Mann, Warren H. Manning, Edward L. Mark, Mrs. W. H. K. Marrs, Vernon F. Marsters, Asa E. Mattice, F. W. G. May, Alfred G. Mayer, Charles J. Maynard, James Means, James C. Merrill, M. D., Selah Merrill, Gerrit S. Miller, Jr., Susannah Minns, Charles S. Minot, Laurence Minot, William Minot, George Mixter, William J. Moenkhaus, Henry L. Moody, Alexander Moore, Margaret W. Morley, Albro D. Morrill, Albert P. Morse, Edward S. Morse, Elisha W. Morse, John Murdoch, Albert L. Murdock, Nathaniel C. Nash, Herbert V. Neal, Frederick H. Newell, Mrs. Edith J. Nichols, Sereno D. Nickerson, William H. NUes, Grenville H. Norcross, Edward E. Norton, William E. Norton, John Ome, Jr. Alpheus S. Packard, M. D., George H. Parker, Edith A- Parkhurst, George L. Parmelee, 218 Commonwealth Ave. Portsmouth, N. H. Absent. South Framingham. Absent Brookline. Cambridge. Absent. Bloomington, Ind. Concord, Mich. Dorchester. Cambridge. West Newton. 106 Beacon St. Absent, Absent. Washington, D. C. 14 Louisburg Sq. Harvard Medical School. 24 Marlborough St. 39 Court St. 219 Beacon St. Williamstown. Absent. 3 School St 28 St James Ave. Clinton, N. Y. South Natick. Salem. Jamaica Plain. Roxbury.. 15 Causeway St. Cambridge. Absent. Absent 294 Marlborough St. Masonic Temple, Cambridge. 9 Commonwealth Ave. 419 Washington St Absent. Cambridge. Providence, R. L Cambridge. Somerville. Absent. Digitized by VjOOQ IC LIST OF MEMBERS. 35 William Patten, Francis H. Peabody, James £. Peabody, Edward C. Perkins, William H. Phelps, John C. Phillips, Dudley L. Pickman, David Pingree. Julia B. Piatt, William G. Preston, Frances C. Prince, Loring W. Puffer, Charles P. Putnam, M. D., Frederick W. Putnam, James J. Putnam, M. D., Hanover, N. H. 113 Devonshire St. Absent. 706 Sears Bldg. Absent. 299 Berkeley St. 98 Beacon St. Salem. Absent. 186 Devonshire St. 17 Joy St. Brockton. 63 Marlborough St. Cambridge. 106 Marlborough. Motte A. Read, Mrs. William H. Reed, John P. Reynolds, M. D., Stephen H. Rhodes, Mrs. Ellen H. Richards, George H. Richards, Harriet E. Richards, Robert H. Richards, William L. Richardson, M. D., Everett W. Ricker, William Z. Ripley, Thomas P. Ritchie, Benjamin L. Robinson, Alfred P. Rockwell, Mrs. William B. Rogers, A. Lawrence Rotch, WUUam H. Ruddick, M. D., Mrs. T. E. Ruggles, John D. Runkle, Frederick W. Russell, M. D., « William E. Safford, Lilian V. Sampson, Charles S. Sargent, Dudley A. Sargent, M. D., Frederick Le R. Sargent, Mrs. Marian E. Y. Saville, Marshall H. Saville, Henry Sayles, Alfred L. T. Schaper, M. D., Barthold Schlesinger, Samuel H. Scudder, Absent. 37 Commonwealth Ave. 416 Marlborough St. 641 Commonwealth Ave. Jamaica Plain. 14 Chestnut St. Boston. Jamaica Plain. 225 Commonwealth Ave. City Hall. Mass. Inst, of Technology. Newton Highlands, Cambridge. 281 Beacon St. 117 Marlborough St. 63 State St. 602 E. Broadway. Milton. Mass. Inst, of Technology. Winchendon. Absent. Germantown, Pa. Brookline. Cambridge. Absent. Waban. Absent. Somerset Club. Harvard Medical School. 131 Devonshire St. Cambridge. Digitized by VjOOQ IC 36 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. John H. Sears, Salem. Mrs. Mary L. Seavey, Brookline. William T. Sedgwick, Mass. Inst, of Tt*chnology Nathaniel S. Shaler, Cambridge. J. C. Sharp, Jr., 64 Commonwealth Ave. Stephen P. Sharpies, 13 Broad St. Mrs. J. M. Sheldon, 108 Mt. Vernon St. Augustine Shurtleff, M. D., Brookline. A. D. Sinclair, M. D., 35 Newbury St. Charles C. Smith, 286 Marlborough St. Frederic F. Smitli, Spriiigfield. Caroline G. Soule, Brookline. Edmund D. Spear, M. D., 20 Mt. Vernon SU A. W. Spencer, 31 State St. Charles J. Sprague, 380 Marlborough St. Frank F. Stanley, 108 Summer St. George E. Stt)ne, Amherst. Charles S. Street, Absent. Mrs. Charles P. Strong, Cambridge. William C. Sturgis, New Haven, Ccmn. John 0. Sumner, Ab.M'nt. Charles W. Swan, M. D., Brookline. J. Brooks Taft, 28 Peniberton S(i. Ralph S. Tarr, Absent. Levi L. Thaxter, 13 Tremont St. Roland Thaxter, Cainbrldm'. John E. Thayer, Lancaster. Mary F. Thompson, 413 Shawmut Ave. Augustus L. Thorndikc, 722 Treinont Bldg. Townsend W. Thorndikc, 22 Newbury St. G. Francis Topliff, 48 Congi-ess St. Mrs. Helen M. Tower, Cainbridgeport. Samuel F. Tower, English High School. William L. Tower, Cambridge. W. Porter Truesdell, 12 South St. Frederick Tuckerman, M. D., Andierst. William Tudor, Absent. Warren Upham, Absent. J. F. Urie, M. D., Absent. Balfour H. Van Vleck, Boston Soc. of Nat. Hist. T. Wayland Vaughan, Absent. M. Edward Wadsworth, Absent. Oliver F. Wadsworth, M. D., 520 Beacon St. Robert Wainwright, Absent. Digitized by VjOOQ IC LIST OF MEMBERS. 37 Frederick C. Waite, Mary L. Ware, Joseph W. Warren, M. D., Mrs. Elizabeth S. Wat«on, Thomas A. Watson, Mary M. Webster, Clarence M. Weed, Andrew G. Weeks, Andrew G. Weeks, Jr., Charles G. Weld, M. D., Samuel M. Weld, Samuel Wells, William P. Wesselhoeft, Arthur W. Weysse, Mrs. Katharine K. Wheeler, Charles T. White, James C. White, M. D., Charles O. Whitman, Henry M. Whitney, Solon F. Whitney, William F. Whitney, M. D., W. H. Whittemore, George Wigglesworth, Thomas Wigglesworth, Mary A. Willcox, Emile F. Williams, Henry V. Wilson, WiUiam F. WUson, ClifU>n E. Wing, M. D., Guy M. Winslow, Roger Wolcott, John E. Wolff, Edward S. Womi, M. D., Elvira Wood, J. Edmund Woodman, Frederick A. Woods, M. D., Jay B. Woodworth, William McM. Woodworth, G. Frederick Wright, New York, N. Y. 41 Brimmer St. Absent. Weymouth. Weymouth. 232 Newbury St. Durham, N. H. 400 Beac(m St. 360 Washington St. 6 Commonwealth Ave. North Chathaju. 46 Commonwealth Ave. M. D., 176 Commonwealth Ave. Mass. Inst, of Tech noli ►gy. Roxbury. 213 Connnonwealth Ave. 250 Marlborough St. Chicago, III. 107 (^)mmonwealth Ave. Wjit^rtown. 228 Mariborough St. Milfnrd, N. H. 53 State St. 36 Hawhy St. WelU'Kh.y. 352 MiiKsachusctts Ave. Chapt'l Hill, N. C. Abst^nt. Hoxbuiy. Auburndale. 173 Commonwealth Ave. Cambridge. Harvanl Medical S<*hool. Mjuss. Inst, of Technology, (^and^ridgf. Brookline. Cambridge. C'ambridge. ObeHin, Ohio. E. Bentley Young, Francis Zirngiebel, 104 Apploton St. Roxbury. Digitized by VjOOQ IC 38 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. PATRONS. Loring W. Bailey. Nathan Matthews. W. W. BaUey. John J. May. James M. Barnard. Samuel May. Miss Emellne Binney. Mrs. Mary G. P. Binney. Henry Sayles. Mrs. G. H. vShaw. George R. Carter. Charles J. Sprague. Henry Cross. W. W. Stone. Thomas Wigglesworth. C. W. Galloupe. Edward Wyman. BY-LAWS. Section 1. Membebs. 1. The Society shall consist of Corporate Members, Corre- sponding Members, Honorary Members, and Patrons. 2. Members shall be elected by the Council only upon the recommendation of the Committee on Membership. Nominations must be in writing and endorsed by two Members. Corporate Members and Patrons only shall be entitled to vote, to transact business, or to hold office. 8. Corresponding and Honorary Members may be chosen from among persons who have contributed to the advancement of science on whom the Society may wish to confer a mark of respect. 4. Members may withdraw from the Society by giving written notice of such intention and paying all arrears. Members who have neglected to pay their regular assessments for two successive years, and have received due notification thereof from the Treas- urer, shall become liable to forfeit their membership at any time when the Council shall so order. 5. A Member may be expelled from the Society upon the recommendation of the Council, by a vote of three fourths of the Digitized by VjOOQ IC BY-LAWS. 39 Members present at a meeting of the Society after notice of such proposed action has been sent to all Corporate Members at least four days before such meeting. 6. Any person who contributes at one time, to the funds of the Society, a sum not less than five hundred dollars shall become a Patron. Section 2. Admission Fee and Assessments. 1. Every person elected a Corporate Member, before obtaining^ the privileges of membership, must pay an admission fee of five dollars. 2. Every Corporate Member shall be subject to an annual assessment of five dollars payable on the first day of October in each year, but for one hundred dollars life membership may be granted free from annual assessment. 3. Members who have given notice that they will be unable to avail themselves of the privileges of membership for an entire year dating from October 1 shall be exempt from the payment of the assessment for that year. 4. The President and Treasurer shall be empowered to return (sub silentio) to a Corporate Member the admission fee and the annual assessment, if they deem it for the interest of the Society to do so. 5. Corresponding and Honorary Members shall not be required to pay an admission fee or any assessment. Section 3. Nominating Committee. 1. At the General Meeting held on the third Wednesday of March a Nominating Committee of five persons shall be appointed who shall report at the next meeting of the Society a list of candidates for the oflUces to be filled. Additional nominations may be made by any Corporate Member. 2. Unless otherwise ordered the Nominating Committee shall be appointed by the Chair. Digitized by VjOOQ IC 4() PROC EEDINGS : BOSTON SOCIETY NATURAL HISTORY. 3. No person shall be eligible for election to any office except after nomination at a preceding meeting. Section 4. Officees. 1. The officers of the Society shall be a President, three Vice- Presidents, a Curator, a Secretary, a Librarian, a Treasurer, and twenty-four Councillors, who together shall constitute a board for the management of the affairs of the Society, and shall be called the Council, and shall hold office until their successors have quali- fied in their places. All Presidents and Vice-Presidents prior to January 7, 1891, are ex-officio members of the Council. All officers shall l>e chosen by ballot. Of the twenty-four Councillors eight shall be chosen at each Annual Meeting to serve for three years. 2. The President shall preside at the meetings of the Society and of the Council. 3. The Vice-Presidents shall perform the duties of President in his absence in the order of seniority in office. 4. The Curator shall be a person of acknowledged scientific attainments. IFnder direction of the Council he shall have general charge of the museum and its contents, shall be responsible for the proper arrangement of the collections, and shall perform such other duties as may be prescribed by the Council. 5. The Secretary shall record the proceedings of the Society, of its Sections, and of the Council, in books to be kej)t for that purpose; shall have charge of all records belonging to the Society, and of its publications; shall conduct the correspondence of the Society, and keep a record thereof; shall inform Members of their election, and Committees of their appointment ; and shall give notice of all meetings, and inform officers of all matters which occur at any meeting requiring their action. 6. The Librarian shall have charge of the library and shall observe and enforce such regulations as the Council may make for its use. Digitized by VjOOQ IC BY-LAWS. 41 7. The Treasurer shall receive all moneys due or payable to the Society ; shall pay all accounts against it when the same have been approved by the Executive Committee ; and shall keep a correct account of all his receipts and expenditures in books belonging to the Society. lie shall receive officially all moneys given or bequeathed to the Society, transmit the same to the Trustees, and report his action to the Council. When requested by the Council he shall make a detailed report of all his doings. Section 5. Council. 1. The Council shall have full power to act for the interests of the Society, shall control all expenditures, and shall make rules for the use of the library and of the museum, and for the direction of the Curator, the Secretary, and the Treasurer. 2. The Annual Meeting of the Council shall be held on the second Wednesday in May, at which meeting a Board of three Trustees shall be elected to hold all the funded property of the Society in trust with power to sell and to reinvest according to their judgment. The following Standing Committees shall also be elected ; viz. : an Executive Committee of five, in whom the affairs of the Council shall be vested so far as it shall determine by its votes and rules; a Membership Committee of five to recommend to the Council persons for election as Members; a Library and Publishing Committee of five to superintend the library and the issue and exchange of the publications of the Society; a Walker Prize Committee of three to have charge of matters relating to that foundation ; and an Auditing Committee of two to audit the accounts of the Treasurer. These Committees and Boards need not be chosen from the Council, and they shall hold office until their successors have qualified. Special committees also may be appointed if needed. 3. The Council may fill the vacancies occurring among the twenty- four Councillors until the next Annual Meeting of the Society. Section 6. Meetings. 1. The Annual Meeting for the election of officers and for other general purposes shall be held on the first Wednesdajr in Digitized by VjOOQ IC 42 PROCEEDINGS : BOSTON SO(^IETY NATURAL JIISTORY. May. At this meeting the following reports sliall be presented : bj the Curator uj)on tlie condition and progress of tlie museum, the lectures which he has superintended, and any other matters of general interest; by the Secretary upon the publications, meetings, and the lectures which he has superintended ; by the Librarian upon the library; by the Treasurer upon the receipts and expenditures for the year; and by the Trustees upon the tinancial condition of the Society. 2. General meetings of the Society shall be held on the first and third Wednesdays of every month from November to May inclusive ; and at such other times as the Council may determine. When any meeting provided for by these By-Laws falls upon a legal holiday, said meeting shall be held on the Wednesday follow- ing such holiday. 3. At any meeting business not provided for by these By-Laws may be transacted, provided notice of such ])roposed business has been mailed postpaid at least four days before the meeting to the address furnished by each Corporate ^Member. 4. Eleven Corporate Members shall form a quorum for business. 5. The order of proceeding at the (General Meetings shall be 1. Approving the record of preceding meeting. 2. Business. 3. Scientific communications and their discussion. 4. Adjournment. Section 7. Libkaky. 1. Members of the Society shall have access to and may take books from the library under the direction of the Library and Publishing Committee, which may by special vote extend the use of the books to others than Members, specifying the conditions under which books may be taken, or may restrict their use even by Members, when, in their opinion, the interests of the Society demand it. 2. The rules and regulations for the use of the library shall be posted in the library room, and a digest of them aftixed to each volume. Digitized by VjOOQ IC BY-LAWS. 43 Section 8. Museu3i. 1. The Executive Committee shall determine the conditions of admission to the museum. 2. Specimens may be removed from the museum only by leave of the Curator who shall take a receipt of the same and be respon- sible for their return in good order. Section 9. Sections. 1. The Council may authorize the formation of Sections of the Society on the written ai)plication of ten Members. 2. Meetings may be held at the pleasure of the Section. 3. The requirements of membership in a Section shall be : — Membership in the Society. Written nomination by two Members of the Section at one of its regular meetings. Election by a three fourths vote of the Members present at a subsequent meeting of the Section. 4. The records of meetings of Sections shall be entered in chronological order upon the book containing the records of the ordinary meetings of the Society. 6. Such report of each meeting as may be judged by the Library and Publishing Committee suitable for publication in the Proceed- ings or Memoirs of the Society, shall be announced by the Secretary at the next regular meeting of the Society. 6. Each Section shall have the right to make additional regula- tions concerning its organization, subject to the approval of the Council. Section 10. Change of By-Laws. 1. The By-Laws of the Society may be altered or amended by a two thirds vote of the Corporate Members present at any meeting, provided a notice of such intended change shall have been given at a preceding meeting and shall have been mailed postpaid to the address furnished by each Corporate Member, at least four days before final action. Printed, June, 1899. Digitized by VjOOQIC Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. MBMOIR8. 4U). The development, structure, and affinities of the genus Equisetum. By E ^ 8 ea g H •< H *-• 1^ 1! 1 1 1 1 I o n n I 0» ?1l i1 1 s »0 lO CO Mean of 15 trans- verse di- mensions. •♦ OS CO CO* r» Mean of 5 dorso- ventral dimensions. CO eo ©< CO o lit: n n J »o '4, IP + 3 W 1-1 "<< «5 ^ s O »0 oo Mean of 8 transverse dimen- sions. iC^CO^CO'^-^*'^^ r« Mean of 6 dorso- ventral dimen. sions. CD niinnii kO Mean of 3 dimen- t sions of nose. cdcDkCUdiOkOodoi'^ "<< Mean of 10 dimen- sions of face. ^•^COCO"^"^"^^"^ CO ©^ Mean of 13 dimen- sions of face and nose. ?3§§S§S2SJ2 1-H Mean of 5 dimen. sions of head. Javanese S Javanese ? Tahitians ? Sudanese ? Roumanians b ss 00 ^ CO CO ^ 00 C>« CO CO kO " iCU9'^^C0'^'^-^»6 M0i^O^©iPC0»0 00tO r^ $ ^ QQ i i 3. 1 •^ s s 1 S "d •^ 0 n § S •S ® CO a 1 2 S «j o d '^ 1 s ^ CO « § &< 1 CO o s • 3 S 3 -s CO eo c ^ Of Digitized by VjOOQ IC BREWSTER : VARIATION AND SEXUAL SELECTION IN MAN.^57 s o aa o O 4.72 4.70 If CO "^ S5g 1-1 CO •^ CO If S 2 CO »6 II ^ 2 ji 8 S i if 3 S CO ©1 ft CO U3 CO w' CO CO 5 S 5 CO ©1 oo o CO 0^ Digitized by VjOOQ IC 58 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. o < K S 2 S &^ < > 11 8 S S § >5 « t* (N CO ** «j . *a 55 rH »: §-| Oi o CO Oi t^ CO QO CO 3 i.! '^cc ©ico^oor^ , -^ , © © »o® r^co«® o&cD ooao9iMU5 1-4 piA«!«4©oeOkS^f^aD uicoi^ujco 'co 'cO'^co©^'^c»eJi'^"«>e4 cipSci© I© |®c^coo3r-, t^aqS'* QDQD^Or-i^ 'co 'c4COCO9i00kd.94<<1«<^CO e4©^t^Od9900»OOkO»O^CO©Qp©QOi-4 i-(i5cOCOt^'^u5f^r^(M0O»CO^©©QO© lOCO^'^949Jc6-^r4^&io4cO«li'^OkOc6 cooooo«5»5»oaOi^^©i(Nr^»oo^^^© ^©'^'^"i>:'*j;»o»-Icc©ic6»oco'54^*t-^c6©t-^'^ 1—1 5| S 5o .^t_^ . . a a = ^ 5 II -^ ^ br 2P =r ^ ^ S? a> o *j Digitized by VjOOQ IC BREWSTER: VARIATION AND SEXUAL SELECTION IN MAN. 61 •-3 s d 3 s s u H o o ^d If o to o r^ ei 3^ 04 XI all 3 g 'N Ol Vl^ d' 5; s 04 05 Max head eigbt. to c4 a It § ?? sg ** Ol 1^3 ^il S 1^ ^P 04 04 S5 a 1^1 00 s f^ 00 Ol 14 oo oi 5 oi ©■a *»ti . Bo£ CO CO 5fc3 © p IP CO CO « 2*- • III s J^ • > p l9^ CO CO K IIIK UNITKIl STATES. llr ffAKiiv nr, N Ittitioii BOSTON: PHINTFhD for the 80CIKTV ) Digitized by Google Digitized by Google No. 4. — Studies in Diptera Cyclorhapha, 1. The Pipunculidae of the United States, By Garry de N. Hough, New Bedford, Mass. This is a small family of mostly small flies, from three to five millimeters in length. A better idea of their appearance can be obtained from the figures on p. 473 of Professor Comstock's Man- ual for the study of insects than by any description that can be written. The family contains but four genera which may be thus dis- tinguished : — 1. Occiput scooped out, closely applied to the convex cephalic surface of the thorax. Body, long, hairy ... 2 Occiput not scooped out, not so applied to the thorax. Body naked or very slightly hairy 3 2. Discal cell present .... Prothecus Rond. Discal cell absent .... Cualarus Walk. 3. Abdomen elongate, thorax with well-developed bristles Nephrocerus Zett. Abdomen not particularly elongate, thorax without well- developed bristles .... Pipunculus Latr. For an elaborate characterization of these genera see Becker's monograph of the Eui:opean species, Berl. ent. zeits., 1897, vol. 42, p. 25-100. Of the four genera of this family all except Nephrocerus are now known to occur in this country. Chalarus. In my collection are two specimens, apparently of different species. One was collected here by myself, the other in Colorado by C. F. Baker. Neither is in sufficiently good condition to use as the basis of a description. Prothecus. Pipunculus lateralis Walk. (Dipt. Saund., p. 216) is referred to Prothecus by Mr. Coquillett (Proc. acad. nat. sci. Phil., 1895, p. 331). Pipunculus opacus Will. (Trans. Amer. ent. soc, 1886, vol. 13, p. 295) also belongs here. The femora of P. lateralis are " serrated beneath for half their length with very small black teeth " ; the femora of P. opacus are not so serrated. Pipunculus. By the kindness of Mr. Samuel Henshaw I have Digitized by VjOOQ IC 78 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. recently been permitted to examine the types of Loew's species and thus to make sure of the correctness of the determinations of the species in my collection. As far as I am aware, none of the species described and tabulated by Professor Williston in Biologia Cen- trali- Americana, vol. 3, have yet been found in the United States. It is not unlikely that some of them will be found in our most south- em states. I strongly suspect that P, reipublicae Walk, is the same as P, fuscita Lw. and that P. tranalatua Walk, is P. subvi- reacens Lw. Professor Williston's description of P, aridus (Dipt, of Death Valley expedition, p. 255-250) applies exactly to P. aubvirescens Lw., with which it is therefore probably synonymous. I shall arrange the species in accordance with the tables in Becker's monograph. Division 1. Stigma wholly or partly colored; abdomen wholly opaque ; third antennal joint usually long acuminate. Division 2. Stigma wholly or partly colored; abdomen wholly or partly shining ; third antennal joint usually obtuse. Division 3. Stigma not colored at all. Division 1. 1. Fourth longitudinal vein without an appendix, 2. Legs not wholly black, at least the knees yellow. 3. Abdomen naked, at most with a few scattered, erect, fine hairs. 5. Hypopygium (sixth abdominal segment) of varying size but not larger than two abdominal segments together. 7. Stigma (subcostal cell) not colored its whole length. 8. Males 9 Females 10 9. Fourth longitudinal vein with a very distinct angle at its junction with the hind cross vein. a. Hypopygium without a cleft. Third antennal joint obtuse . . fasciatus Lw. Third antennal joint prolonged into a long white process ..... subopacus Lw. b. Hypopygium with a cleft. Hypopygium small, shiny, only thinly poUinose; cleft to the right of the median line . 7iigripes Lw. Hypopygium large, opaque, thickly gray pollinose except a spot near the cephalic border ; cleft to left of median line . . . atlafHicvs sp. nov. P. fasciatus Lw. In the Loew collection there is a single male. Digitized by VjOOQ IC HOUGH: PrPUNCULIDAE OF THE UNITED STATES. 79 Opaque, black ; halteres and antennae black ; third antennal joint short and obtuse. Dorsum of thorax opaque with brownish gray pollen. Base of each abdominal segment black and opaque; the remainder of the first segment densely white pollinose; remainder of the other segments with grayish pollen which becomes whiter, thinner, and less opaque toward the sides. Sixth segment of mod- erate size, black. Legs black ; apices of femora and bases of tibiae yellow or reddish yellow. Wings gray, toward base more purely hyaline; stigma brown; veins black; small cross vein near the junction of the basal and middle thirds of the discal cell; third segment of costa shorter than the fourth. P. subo2)acu8 Lw. In the Loew collection a single female. Brownish black. Halteres yellow. Antennae black ; third joint prolonged into a long white process. Dorsum of thorax lightly sprinkled with brownish gray pollen. Abdomen brown black almost shining; first segment except the cephalic margin, other segments toward the sides and caudolateral angles whitish polli- nose. Sixth segment of male small and without cleft. Femora black, thickly gray pollinose, except mesal surface of posterior femora, which is shining black. Apex of femur sometimes yellow- ish. Tibiae vary from yellowish with a black ring to black (some- what gray pollinose) with yellowish base. The tarsi are black or brown with more or less yellow on the first one or two joints. Wings grayish hyaline; stigma brown; veins black; small cross vein at about the junction of the basal and middle thirds of the discal cell ; third segment of costa about equal to the fourth. P. nigripeB Lw. In the Loew collection there is a single male without a head. Brownish black. Antennae black; third joint short acuminate. Dorsum of thorax and scutellum brown polli- nose, subopaque. Abdomen wholly opaque. First segment, except at base, white pollinose. Second, third, and fourth segments vel- vety black, each with a caudal, transverse, dark gray pollinose fascia. Fifth segment brown pollinose, subshining, whitish polli- nose on the sides. Sixth segment black, rather shiny, thinly gray pollinose; its cleft considerably to the right of the median line. Legs black. Femora grayish pollinose to a varying extent except the mesal surface of the posterior femora, which is shining black ; the extreme tips of the femora are yellow. Flexor surface of anterior and middle femora with rows of tiny black spines. Tibiae yellow at base to a varying extent. First one or two joints of the Digitized by VjOOQ IC 80 PROCEEDINGS : BOSTON SOCIETY NATUHAL HISTORY. tarsi yellow or brownish yellow to a varying extent. Wings gray- ish ; stigma brown ; veins black ; small cross vein at a point apicad the junction of the basal and middle thirds of the discal cell, but before the middle of that cell ; third costal segment about twice as long as the fourth. Ilalteres black except base of peduncles yellowish. P, atlanticua sp. no v. Two males and three females. New Bedford, Mass., Hough; Montgomery Co., Pa., C. W. Johnson. Length 3 mm., length of wing 5 mm. Brown black, mostly gray pollinose, with wholly yellow legs and very distinctly banded abdo- men. Antennae brown with very fine whitish pubescence, third joint short, oval, obtuse. Dorsum of thorax and scutellum black brown pollinose, toward the sides and on the humeri more grayish. Pleurae, metanotum, and coxae gray pollinose. Legs wholly yellow, appearing more or less silvery gray pollinose according to the inci- dence of the light, except the tips of the tarsi, which are black. The four anterior femora on the flexor surfaces and all the tibiae have rows of extremely minute black spines. Abdomen wholly opaque. First segment wholly gray pollinose except a narrow black cephalic border. Second segment wholly gray pollinose, the pollen thinner on the middle of the segment, so that in some speci- mens there seems to be a faint median, transverse, black brown fascia. Third, fourth, and fifth segments wholly gray pollinose, except a cephalic, transverse, blackish brown fascia which gradually fades out toward the sides and a median, caudal, blackish brown spot which is connected with the cephalic fascia. On the dorsum of the abdo- men the relative widths (measured cephalo-caudad) of the blackish brown fascia and the gray pollinose portion vary in different speci- mens, sometimes the one and sometimes the other being the wider; the sides of the segments are, however, wholly gray pollinose. The sixth segment is wholly gray pollinose except a small brownish black spot cephalad; this segment is rather large and is divided into three very unequal portions by two dorso-ventral clefts, both of which are to the left of the median line. The one nearest the median line is about halfway to the lateral border, while the one furthest laterad cuts off not over one sixth of the segment. Remainder of genital apparatus yellow. Halteres with yellowish brown peduncle and black knobs. Wings grayish hyaline, veins black; small cross vein at about the junction of the basal and middle thirds of the discal cell ; third costal segment about equal to the fourth. Digitized by VjOOQ IC HOUGH: PIPUNCULIDAE OF THE UNITED STATES. 81 10. Ovipositor straight. 11. Sixth abdominal segment not furrowed. a. The lanceolate terminal piece of the ovipositor not longer than the fifth abdominal segment atlanticus sp. nov. b. The awl-shaped terminal piece of the ovipositor as long as the second, third, fourth, and fifth abdominal segments together ...... subopacus Lw. These females are like the males except for the sexual differences. As far as I know, the females of P. fasciatus and P. 7i%gripes are unknown. Division 2. Dorsum of thorax with delicate but distinct hairs . . 1 1. Abdomen shining black or bronze with opaque black fasciae or spots on the cephalic borders of the segments ; males . 2 Abdomen shining black without opaque black fasciae but with gray side spots ; mostly females 8 2. Fourth longitudinal vein not interrupted ... 3 3. Dorsum of thorax poUinose on the cephalic half without any clearly defined boundary line between the poUinose portion and the shining caudal portion 4 4. Abdomen with gray side spots ..... 5 5. Abdominal segments with moderately broad opaque black fasciae at their cephalic borders. a. Tibiae largely brownish black . . chifpilatus Lw. b. Tibiae yellow with scarcely a suggestion of brown ; hind tibiae rather strongly curved . . . fuscus Lw. 8. Dorsum of thorax more or less poUinose . . . 11 11. Tibiae between a brown and a yellow, seeming now one, now the other, according to the incidence of the light. Awl-shaped terminal piece of ovipositor straight . nitidii'entris Lw. P./uscus and P. nitidiventris belong to the difficult group of P. cUer Meig. It is not unlikely that they will prove to be the two sexes of one species. P. cingulatus Lw. In the Loew collection a single male. ** Grayish black. Antennae wholly black ; third joint short, rather obtuse. Dorsum of thorax opaque with brownish cinereous pollen ; pleurae sprinkled with whitish cinereous pollen. Scutellum black, moderately shining. Metanotum white ])olIinose. First abdomi- nal segment white pollinose with a median black spot ; the rest of Digitized by VjOOQ IC 82 PROCEEDINGS: BOSTON SOCIETY NATUBAL mSTORT. the abdomen black, shining, with white pollinose sides, each seg- ment with a slender, basal, dark cinereous pollinose fascia. Fe- mora black except the yellowish apices," hind femora shining black on mesal surface, the others not at all shining; "tibiae brown black, basal third of all and apex of the anterior ones reddish testa- ceous ; tarsi reddish testaceous, towards the apex brown, last joint black or black brown. Wings slightly brownish cinereous, stigma brown." P, fuscuH Lw. In the Loew collection three males. Black brown. Antennae black, third joint short, rather obtuse. Dorsum of thorax brown pollinose, opaque, towards the caudal margin sub- shining. Scutellum black, shining. Metanotum and first abdominal segment, except its base which is black, cinereous pollinose ; second, third, fourth, and fifth segments velvety black on their cephalic halves and metallic, shining bronze-colored on their caudal halves (a favorable incidence of light is necessary clearly to make out the extent of the velvety black portion); the black of the second seg- ment is to a certain extent cinereous pollinose, and each segment has on the sides an indistinct grayish pollinose spot; the fifth segment is much longer than the preceding ones ; the sixth is much larger still, not symmetrical, and wholly shining bronzy except for a widely interrupted transverse fascia on its cephalic border. Tibiae yellow with scarcely a suggestion of brown ; hind tibiae rather strongly cur\'ed. Femora black except base and apex which are yellow ; lightly white pollinose except the mesal surface of the hind ones, which is shining black ; all of them with a certain amount of white pile, the pile on the posterior surface of the middle femora much longer and denser than elsewhere ; flexor surfaces of all with rows of very minute black spines. Wings cinereous; stigma brown ; small cross vein at junction of basal and middle thirds of discal cell; veins black; third costal segment longer than the fourth. The relative lengths of the third and fourth costal seg- ments in this and the following species are very difficult to deter- mine on account of the extremely acute angle at which the first vein meets the costa, the brown color of the stigma, and the rather large size of the veins, so that the exact point of junction of the first vein and costa is hard to see. P. nitidwentrls Lw. In the Loew collection a single female. Blackish cinereous. Antennae black brown, thii-d joint with the apex short acuminate. Dorsum of thorax opaque with whitish cine- Digitized by VjOOQ IC HOUGH: PIPUNCULIDAB OF THE UNITED STATES. 83 reous pollen, in the middle brownish cinereous. Scntellum black, moderately shining. Metanotum whitish pollinose. Abdomen wholly black and shining; first segment except at base white pilose ; second very slightly sprinkled with cinereous pollen so that it is less shiny than the following segments, its sides and caudal margin almost wholly white pollinose ; third segment with a cine- reous pollinose stripe, dilated cephalad, obsolete caudad ; the sides, the caudo-lateral angles, and the lateral portions of the caudal margins of segments 3, 4, and 5 are white pollinose ; the sixth seg- ment has its lateral margins and its caudo-lateral angles white pol- linose. The tibiae are between a brown and a yellow color, seem- ing now one and now the other according to the incidence of the light. Femora black except the extreme bases and apices, which are yellow ; they are wholly pollinose except the mesal surface of the posterior pair, which is shining ; I can see no spines or pile on either femur. The awl-shaped terminal piece of the ovipositor is straight. Wings hyaline ; veins black ; stigma brown ; small cross vein at junction of basal and middle thirds of discal cell; third costal segment shorter than fourth. Division 3. 1. Stigma not colored 2 2. Small cross vein at or apicad the middle of discal cell . 3 Small cross vein at basal fourth or tifth of discal cell . 10 Small cross vein before the middle but beyond the first third of discal cell 22 3. Eyes of the male in contact near the middle of front for a greater or less distance ....... 4 4. Abdomen, especially the fifth segment, with short, erect black bristles 5 Abdomen with delicate whitish hairs or almost naked . 6 5. Humeri yellow. Ilypopygium asymmetrical ; looked at from the left it is as long as the fifth sogment, from the right it is shorter than the fifth ; on its caudal end is an oblique oval depression which is usually very distinct. Legs black, grayish pollinose ; apices of femora, basal third of tibiae, and tarsi except last joint yellow; hind femora shining black on the mesal surface. Wholly greenish black, lightly gray pollinose, pollen thickest on first and fifth abdominal segnuMits. Anten- nae black, very delicately whitish pubescent: third joint with a rather long drawn out white point. Wings hyaline ; veins Digitized by VjOOQ IC 84 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. black; third costal segment one half the fourth. Length 3.5 to 4 mm similis sp. nov. This species is very like P. sylvaticus Meig., from which its pol- linosity and the form of its hypopygium clearly separate it. The terminal segment of the ovipositor is straight, awl-shaped, and about as long as the hind tibiae. I have eleven males, collected by G. R. Pilate at Tifton, Ga., in Oct. and Nov., and two females, collected by C. F. Baker in Alabama. 6. Tarsi yellow, only the last joint black .... 8 8. Humeri yellow; front of male very small, silvery pollinose, without depression. Thorax and scutellum shining greenish black, dusted with brown pollen in the middle and towards the cephalic border. Hypopygium small, hardly half as long as the fifth segment; its rima oblique and to the right of the median line. Terminal piece of ovipositor straight and about as long as the hind tibiae. Wings hyaline ; veins black ; fourth costal segment hardly twice the third. Abdomen greenish black, shining. Legs black ; the very apex of the femora and base of the tibiae yellowish ; tarsi brownish yellow, toward the tip black. Halteres with yellow knob and brownish ped- uncle aubvtrescens Lw. In the Loew collection a single rubbed male. I have numerous specimens, of both sexes, from New Bedford, Mass., Tifton, Ga., and Opelousas, La. 10. Males 11 11. Eyes not in contact, front not swollen. Shining black, lightly white pollinose. Abdomen with a large yellow spot on each side; fifth and sixth segments large. Legs and coxae yellow except the basal two thirds of anterior coxae black. Length 4 mm. Face and ventral half of front silvery pollinose. Antennae black, third joint yellow and very finely white pubescent, it-s apex prolonged in a white point. Thorax shining black, very lightly white pollinose especially near the humeri. Scutellum shining black. Metanotum and ])leurae somewhat whitish gray pollinose. First abdominal segment black with a transverse yellow fascia at the cephalic border ; second segment black except the caudo- lateral angles, which are yellow; incisure between second and third segments yellow ; third and fourth segments black on the middle of the dorsum, their sides almost wholly yellow ; incisures Digitized by VjOOQ IC HOUGH: PIPUNCULIDAE OF THE UNITED STATES. 85 between third and fourth and between fourth and fifth segments yellow ; fifth segment wholly shining black, much longer and wider than fourth segment, beset with a few delicate, scattered, white hairs; sixth segment wholly black, somewhat grayish pollinose, quite as large as the fifth segment, not symmetrical, being larger on the left side, no visible cleft or depression on its apex. Wing grayish hyaline ; veins black, yellow towards the base ; small cross vein at basal fourth or fifth of discal cell ; fourth costal segment much longer than the third. One male ; Horse Neck Beach, Mass., Aug. 6 flavomacidata sp. nov. 12. Brown black, poUinose. Abdomen with segments one and five densely gray pollinose. Femora black, gray poUinose. Tibiae brownish yellow. Tarsi yellow except last joint black. Length 3.5 mm. Eyes in contact. Face and the small, smooth frontal triangle silvery white. Antennae black, third joint silvery with a rather short sharp point. Dorsum of thorax and scutellum brown polli- nose tending to gray pollinose at the sides and near the scutellum. Humeri yellow. Pleurae gray pollinose. Metanotum densely gray pollinose. First abdominal segment black on cephalic half, densely gray pollinose on caudal half; segments two, three, and four are blackish brown, very lightly gray pollinose on the dorsum, the pollen becoming gradually much thicker towards and on the sides ; fifth segment very densely gray pollinose ; on it there is visible in some lights a suggestion of a median, cephalo-c^udal brown stripe ; sixth segment shining black, only very slightly pollinose save at sides and on venter, about as large as the fifth segment, very slightly asymmetrical ; no rima or depression can be seen ; segments two, three, and four are of equal length, fiye is distinctly longer. Femora black, their extreme apices yellow, wholly gray pollinose except the mesal surface of the posterior, which is shining black. Tibiae appear to vary somewhat in color, according to the strength and incidence of the light, between a yellow and a brown; the anterior tibiae usually look brown except the extreme base which is yellow; the other tibiae look yellowish brown. Tarsi: anterior brown with yellow base ; the others yellow with black tips. Wings grayish hyaline; veins black; small cross vein a little apicad the junction of the basal and middle thirds of the discal cell; third costal segment one half as long as the fourth. Halteres blackish brown, the middle of the peduncles yellowish cUbo/asciata sp. nov. Digitized by VjOOQ IC 86 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. One male, Opelousas, La. Collected by Mi-. G. R. Pilate in May, 1897. In the last two species, as in P.fuacuSy the relative lengths of the third and fourth costal segments are very difficult to determine ; moreover I cannot be quite sure that there is not a very little brown color in the stigma at its extreme apical angle. There is, however, so little of this color, if any, that there can be no impropriety in placing the species in this third division. Only about ninety species of this family are known, of which fifty- eight are European. Undoubtedly there are many more species to be discovered. Printed, July, 1800, Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by Google Boston Society of Natural History. HECENT PUBLICATIONS. MEMotfiA, Uf\ IHJ pp., JO pliit#'8. riJUl Kiikr& oil Uif rvpiUe* and itinphilunns f T'^Uaeozoio ciimls, By Aiiuvirufr W, Gl,l^Ml , Ifl pi>., 4 pi M ten, 25 cL^. BtuUitisin tbe gttld-b**!iri%' wlntt^w uf Nov» Scutm. Ry J. Edmund WixwiuiAir .1:1 pp,, » pluiei*, Ml eiR, Norib AmLTii^iin wood tfo*^. By Hegiimld lieber Ijowt!, Jr. *( pp. 10 cis Soiue llydroida fToiir I'uget Soiuid. By Gury N Catkitia, 35 pp.* n plaie<^ The Odr>naie gemiit MmMHitliemla and it-s nHiea. By l*liiHp P. nahort, *Ti pp. 2 plntes, 60 cts. I hi the* veiiiBof the Wolffian bodivs in the pig, By Cl»ark*s Sedgwiek MUuh to pp., 1 piEiie. 25 et». Noie^ ijtj a Carboiilferouit boukler irrvm in vtmU^rn MasAnelRi^i'tt^ By M}fi>ri L, Fuiler. 14 pp. lo et«. Tlvb genua Ant«iuuw*itt in Nt'W Erigkad. By Merritt L, Kernald. lit pp. Ih cl* The land itiamnml^ tif |iiMiiiiBu!iir FloHdi^ and The t^nnst region tif Ucargia B> Ontram Biini?H. TPpp, 75 cU. A contribution to thi* p^trograpLy of the Hunmti Ranin, By Thendcire i} Wliitc. 40 i>p„ 5 piaLt»s, 05 ets. riymene pttKluutasp nf»v, By Margaret Lewis & pp , 2 plulea. I5c3l». The Harvard geographical mndeli. By W. M, Dnvia. 2(i pp., 4 plaieft, 2^5 ctfc. The rak; of wtvter in growth. Ry (•- 11 Dnvenpurt. \2 pp. 15 ctji. Digitized by Google /y^ AU... Ptoca^dine* of tile Bostou BoQietf of IVaturai Hislorj. Vol, 2% No, 5, riiXTJnrU'1'IONS KKHM THK <»UAV IIKRBARII'M UF JlARVArd) Rt B L Rotii\sijs imo *[. M. (iiff;f,Mwt% BOHTON: PRINTED FOR THE SOCIETT. AutirsT, 1890. Digitized by VjOOQIC Digitized by Google No. 5. — Contributions from the Gray Jlerbarium of Harvard University. Nein Series. — No. 17. By B. L. Robinson and J. M. Greexman, Camiiridge, Mass. 1. REVISION OF THE GENUS GYMNOLOMIA. The genus Gymnolomia, so far as known, is exclusively Ameri- can and extends from the northwestern parts of the Unit€»d States to Brazil. More than half its species, however, are confined to Mexico and Central America. The genus in its present circum- scription has never been monograj)hed. Kunth when founding it in 1S20 described 4 species. De Candolle in the Prodromus, 5, 5(51 (1^86) changes rather arbitrarily Gymnolomia to Gymnopsis under which 10 species are described of which several have since been referred to other genera. In 1^78, Bentham and Hooker f. ascribe to the genus Gymnolomia 16 species. Baillon in his His- toire des plantes regards the genus Gymnolomia as of doubtful value. Hoffmann in Engler and Prantl's Nat. Pflanzenfamilien maintains the genus and ascribes to it 20 species. Aside from the brief citations of the Index Kewensis by far the best bibliographic enumeration of the Gymnolomias is that of Hemsley (Biol. Cent.- Amer. Bot., 2, 161) who cites literature and stations for 15 species, found in Mexico and Central America. The present revision, including 87 recognized species and several varieties, has been chiefly based upon the specimens in the Gray herbarium, now largely supplemented by the incorporation of the Klatt herbarium. But through the kindness of Mr. Coville and Dr. Rose, the representation of the genus in the U. S. national museum has also been examined with profit. The revision of the genus has presented three chief difficulties, viz.: — 1. The distinction of our own western and southwestern forms, which have, notwithstanding great diversity of foliage, pu- bescence, size of heads, and duration, been loosely grouped under G, niidtiflora. 2. The exact interpretation of the Central Ameri- can G. suhflexuosa^ Benth., upon which must depend the varietal and specific nomenclature of some of the nearly related 8])ecies. 3. The explanation and proper treatment of several large-headed Digitized by VjOOQ IC 88 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY Mexican species which although they differ conspicuously among themselves, closely simulate a parallel series of Viguieras. As examples of this parallelism may be mentioned G. ensifolia with the habit of V. blepharolepU. G. meyacephala^ var. sitnulajis with the habit of V. exctlsa. G. decurnbe7i8 with the habit of V, (jhiesbrtfjhtii. G, ghiesbreghtii with the habit of V. buddleiaefomiis. This simulation has various degrees of closeness. In G. yfiies- breghtii and V. buddlti(teforniis^ although the similarity is striking, the heads are always perceptibly smaller in the Viguiera. G. me- gacephala var. simulans and G. deciunbefitiy however, so closely resemble V. excelsa and F! yhitHhreghtii res[)ectively, that no satisfactory external characters have as yet been found by which to separate them. The examination of the achenes, however, at once reveals striking differences. The achenes are in the Gymno- lomias more compressed, quite glabrous, and completely destitute of pappus. In the Viguieras on the contrary they are thickish, obtusely 4-angled, upwar. Smith, Cnrtiss, no. 1434, Small ; fl. August to October. t- t- Outer involucral bracts (rarely elongated) usually appressed-pubescent or even hoary : ligides 10 to 14, linear or oblong : western U. S. and Mexico. 7. G. MULTiPLORA, Bcuth. and Hook.f. Perennial : stems usually several (1.5 to) 9 or 12 dm. high, from a thickish lignescent stock, finely striate and covered with a very short-appressed grayish pubescence: leaves lanceolate or linear-lanceolate and acutish to lance-oblong or oblong and obtuse, 4 to 7 cm. long, 4 to 20 mm. Digitized by VjOOQ IC 92 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. broad, attenuate at the base, finely pubescent, becoming scabrous especially above : beads (exclusive of rays) 1.2 cm. in diameter; involucral scales canescent with appressed hairs. — Benth. and Hook, f., ace. to Rothr. in Wheeler Rep., G, IGO, as to synon. ; Gray, Syn. ft., 1, pt. 2, 269 only in part. Jleliomeria multiflora, Xutt., Journ. acad. Phila., ser. 2, 1, 171 ; Wats., Bot. King exp., 170, at least in great part. — Rocky mountains and plains of Idaho, Palmer^ no. 422 ; Wyoming, Hay den, no. 34, 7\ceedy^ Btirglehaus^ A. Nelson^ nos. 39, 1064, 2663; Colorado, Jiothrock\ no. 551, Parry ^ no. 420, Greene, no. 194, Jones^ (\ S. Sheldon, nos. 152, 468, Cowen (low and many-stemmed), Miss EasHnood (coll. July, 1H89); Utah, Jones ^ nos. 5S20, 599(> ; Nevada, in the Uintas, Watson, no. 606 ; New Mexico, Wooton, no. 4S4 ; California, Ganibel, Co- ville and Eunsto/i, no. 806; and in the Rocky Mountains, without locality, by Gordon, Burke, and by Frhnoyit, no. 121. Toward the southwest this species passes into somewhat narrower- leaved forms (Utah, Ward^ nos. 647, 652, Jones^ no. 5996 r; Arizona, Palmer, coll. of 1869 without numbqr and no. 241 coll. of 1877, Knoxrl- ton, no. 6 ; S. E. California, Purpus, no. 5025), which, however, retain their perennial character. 8. Q*. longifolia. Erect paniculately branched annual : stem single, 6 to 12 dm. high, finely pubescent with appressed hairs: leaves lance-linear or oblong-linear, narrowed at both ends, those of the stem 8 to 11 cm. long, 5 to 10 mm. broad, entire or obscurely and remotely crenate-serrate, finely pubescent on both surfaces, becoming very scabrous and tuberculate-hispid above, often ciliate toward the base ; peduncle and involucral bracts covered with fine appressed pubescence; heads numerous, in size and floral characters essentially like those of G, mvltiflora. — G. multiflora, Hemsl., 1. c, 162, as to Mex. pi. ; Gray in Wats., Proc. Amer. acad., 21, 432 ; Syn. fl., 1, pt. 2, 269, in part; not Benth. and Hook. f. Ileliomeris multiflora, Gray, PI. Wright., 1, 107; 2, 87, in part, not Nutt. — W. Texas, Wright^ no. 328, Jlamird, Nealley, no. 432 (depaupe- rate), Dr. Smart, no. 422 ; New Mexico, ** on pine hills between the copper mines and the mimbres,*' Wright, no. 1221 ; Burro Mts., Pushy, no. 172; Arizona, Rucker's Valley, Lemmon^ nos. 345, 2765, and 383 (a form with double flowers); Nagle's Ranch, Jones, no. 6054 p. ; Mexico, between San Luis Potosi and Tampico, Palm- er, no. 1102; hills and plains near the town of Chihuahua, Pringle, no. 615; S. W. Chihuahua, Palmer, no. 392; Strawberry Valley, Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 93 Chihuahua, Ilartman^ no. 777 ; vicinity of Durango, Pahner^ no. 754 ; Chiapas, Ghiesbreyht^ no. 68. Types in herb. Gray. 9. Q*. annua. Annual, slender, often branched from the base, 3 to 5 dm. high, copiously branched above : leaves linear or nearly so, finely appressed-pubescent and even somewhat canescent, slightly ciliated at the base or not at all, the cauline 4 to 7 cm. long, 2 to 3 mm. broad, the rameal smaller : pubescence on the peduncles and on the involucral bracts fine and apj)re8sed ; heads rather small, usually 7 to 10 mm. in diameter (exclusive of rays): rays 10 to 14, yellow, 4 to VI mm. long : otherwise like G. niiilti' flora. — G. inultiflora. Gray, Syn. fl., 1, pt. 2, 2(59, in part, not Benth. and Hook. f. G. multiflora,, var. annua^ Jones, Proc. Calif, acad. sci., ser. 2, 5, t)9S. Jleliomeris multiflora, Gray, PI. Fendl., H4, as to pi. of Wislizenus ; PI. Wright., 2, S7, in part, not Nutt. — W. Texas, Wright, no. 834, Pope ; New Mexico, on sides of moun- tains near the copper mines, Wright ; Arizona, on sandy river bot- toms, near Ft. Whipple, Coues and Palmer^ no. 559, at Patano, Pringlt ; near Defiance, Mursh, no. 229 ; Arizona, Knotrlton^ no. 2S8, Wilcox, no. 455, Mothrock^ no. 779 ; Mexico, at Llanos, Wislizenus \ in Sonora, Wright, r\o. 1220, F. E. Lloyd, nos. 410, 411. Distinguished from G. muUifloni by its annual root, much narrower leaves, somewhat smaller heads, shorter rays and dis- tinctly Sonoran range ; from G, hispidn^ var. ciliaUt less satisfac- torily by the different pubescence of its leaves and involucral scales. 10. O. hispida. Probably annual, 6 to S dm. high, densely hispid and hoary throughout with long stiff white hairs, those of the stem and peduncles widely spreading : leaves alternate, narrow, linear, 5 to 7 cm. long, 3 or 4 mm. broad, thickish, channeled above: involucral bracts considerably exceeding the disk, hirsute and hoary: flowers as in G, multiflora. — Heliomeris niultifloruy var. hisjnda, Gray, PI. Wright., 2, H7, as to ty})ical canescent form. — Low damp soil, near Sta. Cruz and San Bernardino, Sonora, Wright, no. 1220 in part. In the almost shaggy pubescence of the leaves and copious spreading pubescence of the peduncles this form differs rather markedly from the others here enumerated, but it certainly passes to var. ciliata. Slender annual, commonly branched from the base, 3 to 6 dm. high : stems with sparse spreading pubescence or nearly glabrous below: leaves linear, green, conspicuously ciliate Digitized by VjOOQ IC 94 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. for the greater part of their length : pubescence of the peduncles sparing, mostly appressed ; involucral scales ciliate but scarcely pubescent otherwise. — Ileliomeris mxdtijiora^ var. hispida^ Gray, 1. c, in part, i. e., as to greener forms. — S. Utah at Beaver, Palmer, no. 245 ; New Mexico, Zuni Mts., Sitgreaves -£>/>., San Antonita, Bigdow in part ; S. California at Sta. Monica, Hasse ; Sonora, Mexico, Wright, no. 1222 in part. * * Heads small or mediuui-sized ; involucral scales linear and acute or the inner narrowly oblon^^ and obtusish : leaves ovate or ovate-lanceolate : perennials so fai- as the duration is known. ■•- Heads small and numerous, corymbose at the ends of the branches : leaves petioled (shortly so in var. abbreviata). ■^ Leaves distinctly serrate. = Pappus none : Mexico to Venezuela. 11. G. PATENS, Gray. Smoothish : stems flexuous, reclining, becoming 5 m. in length, at length soft-woody at the base, striate- angidate, sparingly and minutely appressed-pubescent ; leaves ovate, acuminate, serrate, thin, obliquely truncate at the base, finely pubes- cent, more or less scabrous above; petioles slender, 1 to 3.2 cm. long; peduncles slender, finely appressed-pubescent; involucre cylindric, 5 mm. in diameter; outer scales lanceolate acute or attenuate, the inner longer, oblong, striate, obtusish : ligules 6 to 10, 1 cm. long, yellow turning whitish during or after drying; disk- fiowers, 80 to 35. — Proc. Amer. acad., 5, 182; Hemsl., Biol. Cent.- Amer. Bot., 2, 103. ? Wedclia cordata. Hook, and Arn., Bot. Beech., 435. 3Iicroceph(dum ehrtnhergiamnn, Sch. Bip., and Gymnolomia ehrenhergiiuia, Klatt, Leopoldina, 23, 90. Montanoa thamasiiy Klatt, Abh. naturf. gesell. Halle, 15, 328. — State of San Luis Potosi, Palmer, no. 1099, Pringle, no. 3937; Orizaba, Schaff)ter, Botteri, no. 497, 810, Thomas\ Wartenberg in Hua- steca, Eri^endherg, no. 95 ; Oaxaca, El Paridnetla, Comatii and Ganzdlez, no. 906; Guatemala, von Tnerckheim, no. 287. var. ABBREVIATA, Robiuson and Greenman. Stems 1 to 2 m. long, sparingly appressed-pubescent: petioles very short, 4 to 6 mm. in length ; leaves somewhat narrower and thicker than in the type : pappus none. — Proc. Amer. acad., 29, 387. — Jalisco at Tequila, Pr ingle, no. 4595 ; Guerrero, between Chilapa and Tixtla, altitude 1,600 to 2,100 m., E, W. Nelson, no. 2164. var. gnatemalensis. With copious spreading pubescence on the stem : pedicels 1 to 2.5 cm. long: pappus none. — Guatemala, Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 95 San Miguel Uspantdn, Dept. Quiche, Heyde and Lux, no. 3870 of Mr. J. Donnell Smith's sets. var. brachypoda. Pubescence, at least on the upper part of the stem, spreading : pedicels very short (2 to 5 mm. in length) : pappus none. — Guatemala, Alta Verapaz, altitude 1,300 m., von Tuerckheim, no. 852 of Mr. J. Donnell Smith's sets ; also in Colima, S. W. Mexico, Palmer, no. 1223 (coll. of 1891). ▼ar. macrophylla. Pubescence of the stem copious, spread- ing: leaves (1.5 dm. long, 8 to 10 cm. broad) much larger than in the other forms of the species, subcrenately serrate : pappus none. — Venezuela near To var, at 1,000 m. altitude, Fendler^ no. 686. 12. G. cosTARicENsis, Bcuth. Habit, foliage, and pubescence as in G. patens : heads larger (disk-flowers about 50) ; involucre campanulate : pubescence fine, appressed or subappressed : rays about 10: achenes of the disk-flowers destitute of pappus but sometimes crowned with an obscure ring or rim. — Benth. ace. to Klatt, Bull. soc. hot. Belg., 81, 199, at least as to pi. of Oersted (Pitticr's no. 2614, referred to this species by Klatt, is, in herb. Klatt, jBaltimora scolosperum, Steetz!). G. rudbeckioidts, Hemsl., Biol. Cent.-Amer. Bot., 2, 163, in part, probably not HBK. Gym- nopsisf costaricensis, Benth. in Oerst., Vidensk. meddel., 1852, p. 90. G. vulcanica^ Steetz in Seem. Bot. herald, 157. Aspiiia costaricensis^ Klatt, I.e., 201, as to synon. and Pittier's no. 8276 in part. — Costa Rica, Aguacate, and Ujaras, Oersted ; Rodeo de Pacaca, Pittier^ no. 3276 in part, San Francisco de Guadalupe, Tonduz^ nos. 7251, 7288; Rio Maria near S. Jos6, Tonduz, no. 7271 ; Panama Boquete, Veraguas, Seemaiin, no. 158S. = = Pappus none : Peruvian. 13. G. RUDBECKioiDEs, HBK. Of habit and foliage of the last two species but with short petioles (4 to 6 mm. long) , spread- ing pubescence, and (ex icon.) more axillary or irregular (less definitely corymbose) inflorescence. — Nov. gen. et spec, 4, 219, t. 374. — Temperate regions, Agavaca, Peru, Humboldt and Bon- plaTul. Not seen by the writers, who have been unable to match satisfactorily the original description and figure of this species with any Mexican or Central American plant. = = = A delinite pappus present, consisting of several fimbriate scales. 14. ? G. suBPLExuosA, Benth. and Hook. f. Habit, foliage, pu- bescence, and inflorescence much as in G. patens^ from which we are Digitized by VjOOQ IC 96 PROCEEDINGS : BOS'iX)N SOCIETY NATURAL HISTORY. able to separate it solely by the pappus. — Gen., 2, 364 ; Hemsl.,1. c, 163. Wedelia subjfexuosa, Hook, and Arn., Bot. Beech., 435. — Realejo, Nicaragua, Sinclair; Orizaba, Botteri, no. 435 (ace. to Hemsl.). To this species of which we have seen but a single achene we refer with doubt 8j)ecimens from the following localities : Vera Cruz, Valley of Cordova, Bouryeau^ no. 1961; Oaxaca, Jaya- catlan, Z. C, Smith, no. 896, Monte Alban, Pringle, no. 4859, C. Z. Smith, no. 244, Valley of Etla, Z. C. Smith, no. 849, Alvarez^ no. 723 ; Costa Kica, Prov. of San Jos^, */. Doimell Smith, no. 4871. A better knowledge of the type of this species will doubtless justify at least a varietal separation of some of the specimens here mentioned. •^ ^ Leaves entire or nearly so, canesceut -sericeous beneath. 15. G. f'AXEscENs, Robinson. Leaves oj)posite, ovate, subtrun- cate or subcordate at the base, acute or often obtuse at the apex, finely gray pubescent above, silky and silvery beneath ; petioles 2 cm. long: branches of the naked inflorescence long, bearing few clustered heads at their ends. — Proc. Amer. acad., 27, 174. — San Luis Potosi, brackish marsh, Las Tablas, Pringle, no. 3611, alkaline plains. Hacienda de Angosture, Primjle, no. 3763. •^ -^ Heads somewhat larger (of medium size) rather few, mostly solitary or borne by 3's at the ends of the branches. •»- Outer bracts of the involucre shorter than or little exceeding the inner : species of Mexico, Cent. America, and Andean South America. = Leaves sessile or petioles short (2 to 8 mm. long). a. Pappus none : species of Mexico. 16. G. ovATA, Gray. Stem 4 to 8 dm. long or more, covered with a short spreading pubescence : leaves ovate, subsessile, 3 to 7 cm. long, 2 to 4 cm. broad, serrate, mostly acute: heads 3 to 8, long, unequally long-peduncled, 1.5 to 1.8 cm. in diameter (exclu- sive of rays), many- flowered ; the bracts of the involucre narrow, loose, very numerous. — Proc. Amer. acad., 19, 4. — Chiapas, Ghies- breght, no. 554 ; Oaxaca, Sierra de San Filipe, altitude 1,800 to 2,500 m., Pringle^ no. 5674, Conzatti and Gonzdlez, no. 551. 17. G. LiEBMANXii, Klatt. Smaller: leaves sessile, cordate, green on both sides, 2 cm. long, nearly as broad : heads few, 1 or 2 or perhaps 3 at the ends of the branches ; involucres 6 mm. in diameter ; bracts 2-seriate, rather few, lanceolate ; rays about 9, con- spicuously 5-nerved. — Leopoldina, 23, 90. Microcephalum lieb- Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 97 mannii, Sch. Bip^ ace. to Klatt, 1. c. — S. Andres, Mecatlan, Mex- ico, Liebmann, no. 588. Type in the herbarium of the Botanic garden of Copenhagen ; a tracing and fragment in herb. Gray. 6. A rudimentary pappus generally present a« a short-fringed cup : South American species. 18. G. TEXELLA, HBK. Prostrate, much-branched: stems slen- der, hard : leaves opposite, ovate, subcordate, obtusish, thick, rugose, very scabrous above, short- j)etioled, serrate or seiTulate, 2 to 5 cm. long : heads solitary, terminal, on filiform peduncles ; involucre 8 to 10 mm. in diameter; the bracts oblong-lanceolate, canescent- pubescent; disk achenes obovate, apj^ressed-pubescent, somewhat contracted above into the at length obscure pappus-cup. — Nov. gen. et spec, 4, 21H, t. 878; Klatt in Engl. Jahrb., 8, 42. —New Granada near Ibague and Contreras, lhnnholdt\ in bushy under- growths and on savannas of the Kio Dagua, Cauca, altitude 1,000 to 1,S00 m., Lthmann, nos. 19r)(), 2X65, 2994. 19. G. iiONDENsis, HBK., I. c, 21S. From description close to if not identical with the preceding: leaves said to be subacumi- nate at the apex and rounded at the base. — Torrid regions of New Grenada near Honda, Humboldt, Sprengel (Syst., 3, 610) reduces this species to G. tenella^ where it may well belong. 20. G. TRiPLiNERviA, HBK. Branches terete, smooth, glabrate, the younger with appressed pilosity: leaves opposite, ovate, nar- rowly subacuminate, acute at the base, crenate-dentate, reticulate- veiny, 3-nerved, hispidulous with appressed hairs on both surfaces, about 6 cm. long, half as broad; petioles 6 to 8 mm. long: heads solitary on long appressed-pubescent peduncles; involucral bracts numerous, appressed-pubescent, the outer slightly exceeding the inner: disk-flowers and achenes unknown. — Nov. gen. et spec, 4, 219. — On high plains of Bogota, U. S. Colombia, Ilumholdt. Not seen by the writers, the descr. compiled. = = Petioles longer (8 to 20 nun. in leninh): ligules rather showy, 1 to 2 cm. long. a Rays about 12, narrow, strongly carinate or conduplicate. 21. G. CRUCIATA, Klatt. Tall erect shrub, 3 m. high : branches tetragonal, rough : leaves ovate- oblong, shortly cuneate or one- sided at the base, 7 cm. in length, half as broad ; the petioles 1 cm. long : rays about 12. — Bull. Herb. Boiss., 4, 480. — Near Castanal, U. S. Colombia, Sonntag^ no. 8, June, 1888. Type in herb. Univ. of Zurich ; a head and sketch in herb. Gray. Digitized by VjOOQ IC 98 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. 6. Rays 8 to 10, not conspicuously cariuate or couduplicate : South Americau. 22. G. QUiTENsis, Bentb. and Hook. f. Shrub witb rather copious pubescence : leaves all opposite, acuminate, from subcordate to shortly cuneate at the base, 4 to 9 cm. long, half as broad heads cymose, the peduncles rather short, 2 to 2.5 cm. long much exceeded by the leaves. — Gen., 2, 364. Andrieuxia qxdten sis, Benth., PI. Hartw., 206. — Andes of Ecuador, Hartxceg, no, 1142, Hall, Couthouy, ■^ •*^ Outer bracts of the involucre elongated, the inner short : species of Cen- tral Brazil. 23. G. KUNTHiANA, Baker. Erect perennial pubescent herb: leaves petiolate, ovate, serrate, the upper alternate : involucre 1 to 1.2 cm. in diameter; outer bracts oblong, large, unequal, acute, 1.2 to 2.4 cm. long: ligules 10 to 15. — Baker in Mart. Fl. Bras., 6, pt. 3, 172. Gyynnapsis kwithiana^ Gardner in Hook. Lond. journ. bot., 7, 292. — Dry mountainous regions near Concei^Ao Goyaz, Brazil, Gardner, no. 3846. * * * Heads and involucres as in the last group : leaves (all opposite) narrowly lanceolate : pappus none : South American. •.- Pedicels bracteolate. 24. G. GOEBELii, Klatt. Finely and cinereously appressed- pubescent : leaves rather small : branches terminated by 3-headed cymes; pedicels long and slender, bearing above the middle two subulate bractlets; heads 1.2 to 1.5 cm. broad exclusive of the 16 to 20 narrow linear rays. — Klatt in Goebel, Pflanzenbiologische schilderungen, 2, 49. — Dry slopes near the small town of Muru- chies, Venezuela, K. GoeheL ■»- -.- Pedicels naked. 25. G. HiRsuTA, Klatt. Branching shrub, scarcely or not at all canescent : leaves long-lanceolate, caudate-attenuate, obtusish at the base, 7 cm. long, 1.5 cm. broad, bullate-rugose above, veiny and pubescent beneath : ligules about 12, oblong, 5 mm. broad. — Klatt in Engl. Jahrb., 8, 42 (18^<6). — Among shrubs, borders of woods, Tungu-ragua, Ecuador, altitude 2,200 m., Lehmann, October, 1879, no. 360 a. **** Ht^ads laree, but liuules short; involucral scales oblong, the tips slightly squarrose ; peduncles long : leaves oblong or elliptic, sessile, harsh in texture and hirsute : perennial herb. 26. G. PARKiNsoxii, Hemsl. Erect, coarsely pubescent and Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 1K> scabrous : leaves alternate or scattered (the middle ones occasion- ally verticillate) , 6 to 13 cm. long, 1/2 to 3.4 cm. broad, promi- nently 3-nerved : beads on long stout ascending alternate peduncles, subglobose, 2 to 2.5 cm. in diameter. — Biol. Cent.-Amer. Bot., 2, 163. G. rudisy Gray, Proc. Amer. acad., 22, 424. G. rudisy var. minor^ Rob. and Greenm., ibid, 29, 3S7 (small form) . — The type (in herb. Kew), collected in Mexic/O, without locality, by Park- inson, has brownish-purple flowers. This form is represented also by Palmer's no. 531 from the Rio Blanco, Jalisco, Pringle's no. 4584 from rocky hills near Tequila (small form) , and Pose's no. 3031, coll. between Bolanos and Guadalajara, and no. 3700 coll. at Bolanos. This typical form is connected by variegated transi- tions with a forma flaviflora in which both disk- and ray-flowers are golden yellow. The latter form is rej)resented by Pringle's DO. 2460, from hillsides near Guadalajara, Pabner's no. 533 from the Rio Blanco, and Pose's no. 2827, coll. between Colotlan and Bolanos, Jalisco. No morphological differences between the indi- viduals with brown-purple flowers and those with yellow flowers have been detected. * • • • « Heads lar;^ (1.6 to 1.8 cm. broad excl. of rays), terminal, solitary- ; involueral bractn narrow, lance-linear, attenuate: rays long: leaves lance- oblong. 27. O. serrata. Perennial : stems several, 5 dm. high, un- branched, striate, terete, puberulent, leafy, 1 (-2)- headed: leaves lance-oblong, opposite or alternate, attenuate, conspicuously ser- rate, shortly petiolate, scabrous- puberulent above, scarcely paler, pubescent, and reticulate veiny beneath, 6 to 12 cm. long, 1 to 2 cm. broad : involueral bracts herbaceous, lance-linear, attenuate, appressed-canescent-pubescent especially on the mid-nerve, 1 to 1.2 cm. long: rays deep yellow, oblong, 1.5 to 2.5 cm. long, 5 to 8 mm. broad : disk-corollas yellow, essentially glabrous, i\ mm. long : immature achenes 4 mm. long, glabrous, destitute of pappus. — Collected in S. W. Chihuahua on Mt. Mohinora, 1 September, 189s, E. W. Nelson, no. 4891. Types in herb. U. S. nat. museum and herb. Gray. • •«««« Heads of medium size, most nf them sessile or shortly pecfuncled in few many-headed leafy corj-mbs : involueral bracts ovate, obtuse, con- spicuously squarrose : leaves ovate, sessile. 28. G. SQUARROSA, Bcuth. and Hook. f. Erect perennial herb, simple below, coarsely and copiously pubescent: leaves serrate Digitized by VjOOQ IC 100 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. scabrous above, paler and somewhat canescent-pubescent beneath, 5 to 7 cm. long : ligules (about 13) 1 to 1.3 cm. long. — Gen., 2, 3G2 ; Hemsl., 1. c. ; Gray in Wats., Proc. Amer. acad., 22, 424. Zaliiza- nia squarrosa^ Sch. Bip., Flora, 1S64, p. 217. — Ravines near Guada- lajara, Jalisco, Oliva (lHb4) , Palmer, nos. 486, 741 (1886), Pm?^/c, nos. 2194, 2474 (1888-89) . Flowering from September to Novem- ber. «««*««« Heads small or iiR'diuni -sized, s lender-ped uncled : involucral bracts (all but a few (»f the outermost) broad, obovate or oblon«r, mostly obtuse or rounded at the apex : leaves ovate or lanceolate to linear-oblon*;. +- Scales of the involucre not canescent. 29. G. LATiBRACTEATA, Hcmsl. Stems terete, striate, purple, alternately branched above, sparingly scabrous-pubescent: leaves alternate (so far as known), lanceolate, attenuate at each end, short-petioled, obsoletely serrate, 6 to S ("to 18") cm. long, 1 to 3 cm. broad : peduncles o to 7 cm. long, 1-headed; heads (excl. of ligules), 1.5 cm. in diameter; involucre campanulate: rays few, 1.2 to 1.4 cm. long. — Hemsl., 1. c, 162. — Cerro Pinar, North Mexico, Stemann, no. 1485. Collected but once. -•- -4- Scales of the involucre canescent. 30. G. EXSiFOLiA, Benth. and Hook. f. Stems slender, terete, strigillose, and scabrous: leaves opposite, sword-shaped, 1 to 1.2 dm. long, 1 to 1.5 cm. broad, gradually narrowed from near the sessile base to the attenuate apex, very scabrous above, pale beneath : heads cymose ; involucral scales all rounded at the apex, the outer much shorter. — Gen., 2, 364; Hemsl., 1. c, 161. Montac/)iea ensi- folia, Sch. Bip. in Seem. Bot. herald, 304. 2kiluzama eitsifolia, Sch. Bip., Flora, 1864, p. 216. — Sierra Madre, Northern Mexico, Seemann, no. 2007 in part; Tepic, Palmer (coll. of 1892). 31. G. SERICEA, Klatt. " Suffrutescent " : leaves chiefly oppo- site, ovate, shortly connate-petiolate, entire, 3-nerved, acute, acu- minate, 3 to 4 cm. long, half as broad, above green and scabrous, beneath silky canescent: heads 1.2 cm. in diameter; involucral bracts biseriate, oblong, subequal : rays about 20. narrowly oblong. — L^opoldina, 23, 90. Mia*ocephalum sericetim^ Sch. Bip. ace. to Klatt, I.e. — Mexico, without locality, JJebma7i7i, no. 609. Type in herb. Botanic garden of Copenhagen; excellent drawing and some fragments in herb. Gray. Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 101 ««»««••« Heads large (1.7 to 7 cm. broad excl. of rays) ; involucral bracts broad, the inner large, spatulate with rounded apex: perennial herbs: leaves lanceolate to oblong or ovate ; petioles sliort winged, or almost none. 1- Leaves oblong-lanceolate, pale beneath. 32. G. GHiESBREGHTii, Hemsl. Erect, simple or branched from the base, 6 to 15 dm. high: leaves all alternate or some opposite, serrate or serrulate, mostly acute at each end, 5 to 10 cm. long, 1 to 4 cm. broad: heads few, 2 to 2.5 cm. in diameter; outer scales short, squarrose: ligules about 20, golden yellow, 2 cm. long. — Hemsl., 1. c, 162. — South 3Iexico, hills of Oaxaca, Ghiesbreght, nos. 385 (ace. to Hemsl.), 3S2, Printjle^ no. 4903, E. W, Kelson y no3. 1371, 1464, L, t\ Smith, no. 799, Coiizatti and Gonzdlez^ no. 553 ; Michoacan on wooded hills near Patzcuaro, Fringle, no. 3339. 1- ■»- leaves elliptical, green beneath. 33. G. DEcuMBExs, Robiuson. Stems several from a lignescent stock, decumbent, 4 to 5 dm. long: leaves chiefly opposite, 2.5 to 4 cm. long, half as broad, scabrous upon both surfaces: heads sohtary or borne by threes at the summit of each stem ; outer scales of the involucre ovate-lanceolate, acute, not squarrose, dark colored, covered with appressed white hairs. — Proc. Amer. acad., 26, 165. — Rocky hills, Tultenango, State of Mexico, Pi-inyle^ no. 3263 in part; Zacatecas between Bolanos and (Guadalajara, Bose^ no. 3044 (leaves somewhat longer). The achenes of this species are perfectly glabrous, obovate- oblong, strongly compressed, striate, and completely destitute of pappus. It is now found that another plant of astonishing similarity was also collected and distributed under the type number (Pringle's no. 3263). While in all observed external characters it cannot be distinguished from the former, it has thick obtusely 4-angled mot- tled and pubescent achenes which are crowned with 2 (to 4) palea- ceous awns and intermediate scales. The latter plant seems to be identical with Viguiera ghiesbreghtiij Gray. -!-■»-••- Leaves ovate, large and broad. •M. Petioles not auricled. = Outer involucral bracts large, squarrosely spreading or reflexed, about as long as the inner ones, acute or acuminate. 34. G. FLAVA, Hemsl. Stems strigillose becoming hirsute with coarse spreading hairs toward the summit and upon the peduncles : Digitized by VjOOQ IC 102 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. leaves 7 to 10 cm. long, 2 to 3 cm. broad, above scabrous, beneath hirsute upon the veins and finely canescent-pubescent upon the surface: heads (excl. the rays) '2.5 to 3.5 cm. in diameter: ligules about 15. — Hemsl., 1. c, 161. — Oaxaca, Ghiesbreght, no. 216 (type in herb. Kew). We follow Dr. Gray in referring to this species GhiesbreghVs no. 3^<8, from which the above characteriza- tion is drawn. 35. G*. megacephala. Stems simple, striate, scarcely at all sca- brous, covered with minute appressed hairs, also lanate in lines decurrent each from either side of the short winged petioles : leaves ovate-lanceolate with short slender caudate acumination, serrate, scabrous above, hirsutulous upon the veins, but green and glabrous upon the surface beneath, 8 to 12 cm. .long, 2 to 4.5 cm. broad: heads terminal, solitary, about 7 cm. broad : no rays observed. — Collected between Tlapa and Ayusinapa, Guerrero, Mexico, alti- tude 1,380 to 1,700 m., 13 December, 1894, E. W. Nelsoji, no. 2105, distributed as G.flava. var. simulans. Pubescence of the stem coarser, spreading : bases of the hairs enlarged, white, tuberculate : leaves somewhat larger, less pubescent, merely cuspidate at the obtusish tip: rays about 13, oblong, bright yellow, 3 cm. long, 7 to 10 mm. broad. — Collected on the Sierra de los Morones near Flateado, Zacatecas, altitude 2,300 m. 1 September, 1897, Dr. J, N, Rose, no. 2740. This vari- ety closely simulates Viguiera excelsa, Benth. and Hook.f., which, however, differs technically and, as it appears, with constancy in its pubescent achenes which bear at the summit two conspicuous awns and some intermediate scales. = = Outer involucral bracts much shorter than the large inner ones. 36. G. PLATYLEPis, Gray. Hirsute shrub, 1 to 3 m. high : leaves chiefly opposite, acuminate or attenuate at each end, very sca- brous above, paler and densely soft-pubescent beneath, 1 dm. long, 3-nerved from considerably above the base : peduncles thickened upwards; heads (excl. of rays) 2 to 3 cm. in diameter; ligules 15 to 20 in number, 2 to 3 cm. long. — Proc. Amer. acad., 19, 5; Klatt, Bull. soc. bot. Belg., 31, 199. G, decurrens^ Klatt, Leopol- dina, 23, 90. Peri7nenio2ysi8 jyerfoliata^ Sch. Bip., ace. to Klatt, 1. c. Tithonia scaherrima, Benth. in Oerst., Vidensk. meddel., 1852, p. 91. T, platylepis, Sch. Bip., ace. to Benth. and Hook.f., 1. c, 368 (name only). JlirasoHa scaberrima, Benth. and Hook.f. Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 103 ace. to Ilemsl., I. c, lOS. — South Mexico, Mirador, JAehintfmi^ no. 251 ; Chiapas, Ghieahreght^ no. 572, Caec. and Ed. Seler, no. 2187 ; Valley of Cordova, Bonrf/edu, no. 1, no. 539. Types in herb. Gray and herb. U. S. nat. museum. Doubtful species. G. coNNATA, Spreng., Syst. 3, 610 [conudtion). — Branches hispid : leaves oblong, attenuate at either end, connate, hispidulous : ]>eduncles terminal, short: foliaceous involucre subsimple. — 13razil, tSeUo, A species not mentioned in Martins, Fl. Hras., and probably not of this genus. The description is a translation of the original characterization . G. MiCROCEPHALA, Lcss. Herb with 3-nerved leaves, small cylindrical discoid short-pedicelled heads: leaves petiolate, opposite, paler beneath, oblong-ovate, long-acuminate, very obtuse at the base, sharply serrate, hirsute-hispid, scabrous, 3.1 to 5 cm. long, 1 to 3 cm. broad : scales of the involucre passing from Hat semilanceo- late foliaceous bracts, hirsute on the midnerve, to longer (about 4 mm. long) and broader dry linear shortly acuminate glabrous bracts, fimbriate-dentate at the a})ex, scarious on the margin, and traversed by green nerves : branches opposite, diverging at the apex into 3 short filiform branchlets bearing several hea0) with Kunth's description and plate (IIBK., Nov. ^aw, et spec, 4, 295, t. 407) we can find no significant difference. C. DENsiFLORA, Klatt, Lcopoldiua, 20, 9(), is Af/eratum coinj- zouies, L. C. PELLuriDiyERViA, Klatt, Bull. soc. hot. Belg., 31, 207, is not to be satisfactorily distinguished from C (wilfaris, DC. C. AXILLARIS, DC. To this species should also be referred C. prunifolla, Klatt, 1. c, 208. in part (as to Pittier's nos. 3219, 3095, 4520, 4938, and 7023), not IIBK. The distinctions between C. iisrillftris and its variety urticnefoUa seem to be scarcely worthy of recognition. C. pittierl sp. n. Shrub with slender flexuous terete puberu- lent or toward the end tomentulose branchis : loaves ovate to elliptic, thickish, crenate- serrate, acutish, 4.5 to 7.5 cm. long, 2.5 Digitized by VjOOQ IC 10<) PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. to 8.7 cm. broad, rugose and very scabrous above, veiny grayish- tomentose and resinous-dotted beneath, 5-nerved from near the obtuse or subcordate base ; i)etioles 0 to 9 nira. long, covered with a fine fuscous pubescence : heads numerous, rather small, discoid ; pedicels filiform, 1 to 1.2 cm. long, some fascicled In the upper axils but most of them in regular hemispherical umbelliform leafy-bracted clusters (4 to 5 cm. in diameter) at the ends of the branches ; invo- lucre cylindric-ovoid, somewhat turbinate at the base, the outer scales very short, linear-oblong to narrowly ovate, herbaceous, tomentulose, the intermediate oblong, obtuse, often erose and squar- rose at the tip, the inner lance-oblong, acute, bright yellow, peta- loid : achenes narrowly fusiform, angled, 2.7 mm. long, callose at the base, covered with short spreading gray hairs ; pappus-awns about 20, narrow, attenuate, 4.() mm. long. — C, pruinfolia, Klatt, Bull. soc. bot. Belg., :il, 2()S, in part, not HBK. — Costa Rica, banks of the Kio Ceibo near Buenos Ayres, altitude 200 m., Janu- ary, 1892, Prof. If. Plttler^ no. 4918, Chennn de la Caldera between San Mateo and San Kamon, P. Biolleyy 25 January, 1892, no. 7015; also by Pt-of. Pittier in woods at Boruca, altitude 450 m., February, 1S91. C. prunifoUn, HBK., to which this plant was referred by Dr. Klatt, has larger leaves nearly or quite smooth and somewhat coriaceous. It also differs in the calyculate bractlets of the involucre, which are suborbicular and of larger size. C. GRAYii, Klatt, Leopoldina, 20, 90, overlooked in our revision, should replace C. to)nentosa^ Gray, not Gardner. C. TKKNiFoLiA, Oliver, Trans. Linn. soc. ser. 2, 2, 277, t. 48, figs. 9-10 (188()). For this species the name of which is antedated by n. ternifoU'i, IIBK., Nov. gen. et spec, 4, 294, we would propose C. oliverii. TiUDAX (iALEOTTii, Klatt, 1. c, at least as to the plant with lobed pilose leaves. T. tnhtrosa, Rob. and Greenm., Proc. Amer. acad., 82,4. In the Klatt herbarium T. fftth ottii is represented by two sheets of well executed drawings. On one is an excellent repre- sentation of the j)lant we have called T. tuberosd.^ with lobed leaves, cuneate at the base and distinctly hirsute. On the other sheet two ])lants are represented, one being identical with that just mentioned, the other having the more oval serrate unlobed leaves of T. brachy- Ifpf's^ llemsl. Dr. Klatt 's description clearly shows that he had in mind the plant with lobed leaves. The description, however, is quite in error as to the involucral scales being 2-seriate. They are clearly Digitized by VjOOQ IC ROBINSON & GREENMAN : CONTRIB. GRAY HERBARIUM. 107 3— 4-8eriate in both his drawings and in a single autlientic head pre- served in a pooket on one of the sheets of drawings. In the arrangement of species T. galeottii should, accordingly, be inserted where T, tuber osa has been placed. T. iMBRicATA, Sch. Bip. in Klatt, Flora, 1885, p. 202. T. petro- phitot Rob. and Greenm. 1. c, 5. The authentic specimen of T. inibricata in the Klatt herbarium is very poor, but sufficient to show with a fair degree of certainty the identity of T. 2>etrophila. The involucre is here also 8-4-8eriate, not biseriate as we had supposed. T. vEETiciLLATA, Klatt, Lcopoldiua, 25, 107 (1889). This South American species, omitted from our revision, is the only member of the genus with verticillate leaves. In the arrangement of the species it may well be placed in subgenus Eutridax after T. angtistifolia, T. BHBBNBEBGii, Sch. Bip. in Klatt, 1. c. Klatt's drawing of this species shows a plant with long internodes and habit of T, dubia^ but the achenes preserved in the Klatt herbarium have the pappus rather of a Galea than a Tridax. The species must remain doubtful until more complete material can be seen. MiKANiA OLIVACKA, Klatt, Bull. soc. l)ot. Holg., 31, 11)5. — An examination of the type in the Klatt herbarium shows it to be Pittier's no. 4938, not 4433, as cited in the original publication of the species. M. PUNCTATA, Klatt, 1. c, originally described as having the heads 7-flowered, proves, on careful dissections of portions of the type material, to have uniformly 4 flowers in each head, fully agreeing in this regard with the other sj)ecie8 of the genus. M. FENDLERi, Klatt, Abh. naturf. gesellsch. Halle, 15, 324, although reduced in our revision to a synonym of 3f. cordlfoUa^ Willd., is probably worthy of specific rank. M. gonzalezii, sp. n. Vigorous smoothish twiner : branches obsoletely 6-angled, finely striate, glabrous : leaves large, thin, gla- brous, ovate, entire, acute, broadly and shal lowly cordate or sul>- cordate, 5-nerved from the base, 4 to 12 cm. l')ng, 3 to 10 cm. broad; petioles 1.5 to 5 cm. long: heads in an elongated open panicle with short opposite corymbiferous branches (5 to 15 cm. long); bracts of the inflorescence oblong-lanceolate, ])etiolate ; involucres 7 mm. long, subtended by shorter sessile oblong-ellip- tic or oval bractlets ; pedicels puberulent under a lens; involucral Digitized by VjOOQ IC 108 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. scales oblong, essentially glabrous except at the obtusish or barely acute tip : corolla 5 to 6 mm. long, about equalling the sordid pap- pus : achenes 4 mm. long, nearly glabrous. — Collected by Prof. C. Cofizatti at Colonia Melchor Ocampo, Vera Cruz, Mexico, alti- tude 1,200 m., 8 December, 1S95, no. 18, and by Prof. C. Comatti, and V. Gonzdlez in Canton de Cordoba, Vera Cruz, 27 December, 1897, no. 637 in part. This noteworthy plant is clearly distinct from any Mexican species known to us, nor have we been success- ful in referring it to any South -American species. In the genus it should, probably, be placed near M. cordifolut. from which, how- ever, it differs in the size, form, texture, and glabrous character of the leaves, which are more or less nigrescent in dr^nng. We take pleasure in dedicating the species to Mr. V. Gonzdlez, who as the able assistant of Professor Conzatti, has done much to further the knowledge of Mexican plants. Printed, Awjud, ISOO. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by Google Boston Society of Natural History. RECENT PUBLICATIONS. M^MOfRtj. 4riK The rfevel n\mw n t» *?tTiicl u rt% an( 1 nM n HJes o! ih e ^^n us Ki | uist'tiun . By TA wmjiI ^fiie» on lUi' ivjitileA nmj umphibmiis nf IfitervnU*, N IL By Ulowr M Vimaiioij aiiit [«?xuitl aelf?€t.icm in iiii*it By KdwUi Teiiiit?y Bi'ewsUt MtJuiloprtriiUiA', H Uf'W family of PalatHJEnIc cornlvs. By AmaiUnis W. Giatmu. 113 pp., 4 plate?;;. 25 rtH. Studies in lilt! ii^uUI-lwarui^ wIuk^k of Ndva Sj'nitJt. By J. Eiimtimi WrxwInjAn S3 pp., H plaivH. 50 L't^. NoPtU Auierlcaii wood frogs* By R«gUiakl llebcr Howe, Jr. tf pp, 10 cu Sotue Hytlriiids from Puget Sound, By Gary N. Ciilkmi» i4& Jip . 0 p1af4« m) cts. The Ortouaie gr^nrns MaeroUieiiiis and Tt,H ?«lli^s By Philip I\ CHlven, 32 p(i 2 platen i &f* crtH. thi the veitm of the Wolffian bodies in tin? pig. By ClmrleB ^edgwiek Miiinj, 10 pp,, 1 plau?. 25 cls» Notes im a Carboniferoua boulder train in eaetern Mai^acbuspiis, By Mymn I.. Fuller, 14 pp. 15 eta, The genus Ant^uuaria in New EnglanU. By M err hi h. t'eruaUi i;i pp. 15 cts. Tho land aiauitnalfj of peuinsuiar Florida and ih eta. A couiribuiiun Tj> the petrography o( thi.^ Bost^^n Basin- By ThtJfHlort^ 0 White. 40 pp., 5 platos. WA ut^. Clymene prtxlucta sp. nov. By Margaret Ix'wie. ^ pp., 2 plates. 1 5 els, 'rhe Harvard geographical models. By W. M. Davis, 2« pp., 4 plates. 2& ct*. Tiie roll* of water In growth* By C, B. DaT4^npori. 12 pj), 15 els. Digitized by Google Proo#«diitg» oi the BostDti Socittty ol Ifatixred History. Vou 29, No. f>, TJIE DEVEUJPMEJrr OF PEMl-IA St IIMACKERI lUrHAHD. By Mkkviw T, ^itiJLKii* Wrrii fHict;u rr*\TK!*. BOSTON: PKINTED FOR THE SOCIETY. Digitized by Google Digitized by Google No. 6. — Ths Development of Penilia schmackeri Richard. By Mervin T. Sudler, Baltimore, Md. With three plates. During June, 1896, immense swarms of small crustaceans sud- denly appeared in the water of the harbor of Beaufort, N. C. Their 'numbers were so great that the animals usually found near the surface of the water were completely obscured and, in towing, the meshes of the net after being drawn through the water a few yards were clogged with these crustaceans. This abundance lasted but a few days when the creatures disappeared completely and as suddenly as they had appeared. Dr. C. P. Sigerfoos, then of the marine laboratory of the Johns Hopkins university, pre- served a number for future study. Three methods were used to preserve and fix these animals'; viz. : — 1. A saturated solution of corrosive sublimate used cold for a few minutes. 2. A corrosive sublimate and acetic acid reagent, composed of Corrosive sublimate (saturated solution) . . 90 parts Glacial acetic acid 10 " 3. 80% alcoholic picro-sulphuric acid. The animals were washed in water after treatment with each of these solutions, run up through the graded series of alcohol until they were jn 80%, and kept until used. The material so preserved was placed in my hands and was studied by means of sections of the adults containing young and the separate embryos after they had been removed and by means of adults and embryos mounted whole. Kleinenberg*s haematoxylin as a staining agent gave the best results in the first case, and where they were mounted whole a few moments staining in Czokor's alum cochineal proved the most satisfactory. Xylol was used as a clearing agent and Canada balsam as a mounting medium. Systematic Position. The crustaceans were found to be Cladocera (Daphnoidea) and belong to Penilia, of which two species have been described by Digitized by VjOOQ IC 110 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Dana. All of the members of the Sididae thus far known are marine and especially distinguished from all other families of the Cladocera by the presence of six pairs of thoracic appendages and by the fact that one or both of the branches of the second antennae have less than three joints, the characteristic number for most of the members of this group. The species has been determined by Prof. E. A. Birge of Madison, Wis., who has compared the form with the type specimens of M. Richard. The embryology of no member of the Sididae has been studied, and most of the work on the entire group has been of a systematic or anatomical character. Only three papers of importance dealing with the embryology of the Cladocera are known to me ; these are by Grobben (79), Lebedinsky ('91), and Samassa ('98). Other articles, such as those of Dohrn ('(39) and of Zaddach ('41), are old and give but few details of development. Anatomy. Penilia when alive is transparent and has no pigment of any kind except the very small amount in its single median eye. It measures from 0.8 to 1.0 mm. in length, not including the biramose swim- ming antennae, which with their setae are about 0.8 mm. The long pair of abdominal setae are 0.4 mm. in length, making the entire animal from 2.0 to 2.2 mm. long. In general shape (Fig. 1) it somewhat resembles Daphnia similis (Claus). At one stage of the life of Penilia eleven pairs of appendages appear, but in the adult ten only arc present. The small anterior, or first, antennae are situated on tlie posterior side of the two prominent horns which project downward from either side of the head. They contain nerve celln, and one long sensory seta is attached to the anterior side of their distal end and to tlie ]>ost(*ri()r side a pencil of small sensory liairs. Tlie large biranio.se swimming antenna is composed of one large basal joint from which extend an exopodite and endo- j)odite, eaoh composed of two segments, the first about three times as long as the seconut the size of the qqq^ corresj)onds very nearly to the size of them all, and is therefore very much larger than any one cell. In a few (three) specimens I have seen what I have taken to be the act of fusion. Nuclei in partially broken states occurred in protoplasm which seemed separated in places and slightly broken and irregular. This fusion of the four cells must occur quite rapidly as transition stages are so rare. In some cases, in eggs which had not seg- mented, small fragments of material which stained as nuclear material appeared. These were embedded in the substance of the egg, and I interpreted them as fragments of the nuclei of the three cells that lose their identity in the formation of the egg, I found Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 113 nothing that I could call a polar-body. This structure is present in Daphnia similis until the eight-celled stage according to Lebedinsky and the sixteen-cell stage in Moina according to Grobben. These obserrers did not see the formation of the polar-body, but state that it is formed while the egg is in the ovary. From these descrip- tions it seems to me that there is a possibility of its not being a polar-body but a disintegrating nucleus of one of the ovarian cells which has entered into the composition of the egg. The Position of the Egg in the Brood-Ch amber. The eggs when laid lie in the brood-chamber with their long axes parallel to that of the mother where not more than two are laid on either side. In large individuals the orientation may be the same with a larger number of eggs ; but in most cases where the number on a side is more than two, the eggs overlap so that the pos- terior end of each egg is outside the anterior end of the egg directly behind it. In some cases this position is not assumed until cleavage has advanced to the four- or eight-cell stage, and in a few cases it may not be assumed until the appearance of the appendages. Later on the change in position is controlled by different factors, such as the change in the shape of the embryo and the change in the shape of the enlarging brood-chamber. These will be described later. The embryo is nourished by means of the blood-plasma or albu- men with which it is surrounded. Its increase in size is due entirely to material obtained in this way. In the preserved speci- mens this nourishing substance has coagulated and causes the embryos to adhere quite firmly to the mother. In sections showing a segmentation cavity we find a coagulum slightly stained which probably comes from the same source. Segmentation. The first cleavage plane is transverse to the long axis of the ogg and divides it into two separate and distinct cells, the anterior of which is slightly larger than the posterior. (Fig. 7.) The next plane of division is through the long axis of the egg at right angles to the first and is almost parallel with a plane pass- ing through the middle line of the adult. A four-celled embryo is thus formed with the two anterior cells slightly larger than the two posterior ones. (Fig. 8.) Digitized by VjOOQ IC 114 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Sections of eggs throughout their development show through the protoplasm a sharply defined and delicate reticulum with coarse meshes. In the young stages and with high magnification, small enlargements can be distinguished where the threads meet to form the meshes. The third plane of division is longitudinal but perpendicular to the second, forming an eight-celled embryo in which the cells of the anterior end are still slightly larger than those of the posterior end. (Fig. 9.) In the fourth stage of cleavage the eight cells have divided transversely making sixteen cells. (Fig. 10.) Two planes of division were nt'cessary to cause tliis. The eight anterior cells can still be seen to be a very little larger tlian the eight posterior ones. At this stage (Trobben has already distinguished a cell as the mother-cell of the genital organs, another as the primitive ento- derm cell and still others surrounding the "genitalzelle" which are thought to give rise to the mesoderm. Thus at this early stage these few cells foreshadow all of the germ layers as well as the sexual cells of the adult. These cells are distinguished 1st by their peculiar staining properties and 2d by difTerences in their action ; t. e. time of division in reference to the other cells of the egg. Samassa in his studies on the same animal finds the so-called " Grobben'sche zelle" ("genitalzelle " of Grobben) and the entoderm present but not distinguishable to the same extent as Grobben claims. He is unable to distinguish the '' Grobben'sche zelle " after the original one has divided into eight. He fails to find that it develops into any particular organ of the adult. In Penilia I can find no cell marked from its fellows by any peculiarity at this stage. The cells vary somewhat as to size but apparently not in such a definite manner as on that account to allow any particular one to be chosen and its history followed. In the fifth or thirty-two-celled stage regularity of cleavage planes is hard to observe and probably does not occur with exact- ness. From the arrangement of the cells, however, the planes must be longitudinal again in their genera] direction. It is no longer possible to distinguish any difference in the size of the cells lying at the anterior end from those at the posterior. A cross- section of this stage shows segmentation to be complete with a large segmentation cavity containing no yolk. (Fig. 11.) The sixth or sixty-four-celled stage still shows no sign of differen- Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 115 tiation. All of the planes of division cease to be distinguishable as such from this time. (Fig. 12.) The seventh stage probably has about one hundred and twenty- eight cells and shows the beginning of gastrulation. (Fig. 13.) Gastrulation. This process always t«akes place in a very definite part of the egg with reference to its position in the mother. No matter on which side of the intestine of the adult the ogg lies, gastrulation always occurs at its outer posterior comer. An exception to this rule was not observed in the scores of specimens examined at this stage. A median horizontal longitudinal section (Fig. 12) at this time shows projecting into the segmentation cavity from the outer and posterior angle of the outer wall of the egg an indefinite amount of protoplasm containing one or two nuclei. At the particular point to which this is attached to the outer wall one or rarely two cells have sunk below the surface and are in the process of migrating into the interior. (Fig. 13.) The nuclei and protoplasm shown in the interior have reached this position by a similar process. In all probability regularity in the planes of division has disappeared even before this time, biit it certainly has now. The cells, however, continue to divide at nearly the same time. An older embryo (Fig. 14) shows that both the amount of protoplasm and number of nuclei in it in the seg- mentation cavity have increased and a larger num])er of cells are concerned in the inwandering and formation of the mouth of the gastrula. In Moina, gastrulation begins later than in Penilia, and apparently the process is more rapid since more cells are concerned from the beginning. In an older Penilia larva this indefinite protoplasmic mass has become quite a definite rod running through the segmenta- tion cavity and contains numerous nuclei. The gastrula mouth is much deeper and enlarged. Xo differentiation except that noted can be seen. Cell walls are now made out with great difticulty, if at all, although the individual cells stand out as such very plainly. Boundaries between them are marked off by lighter staining areas of protoplasm. (Fig. 15.) At a corresponding or even earlier stage, Moina, according to Grobben, already shows the beginning Digitized by VjOOQ IC 116 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. •of the nervous system. Samassa states that the embryo is more tadvanced before such a differentiation can.be distinguished. Size of Embhyos. So far in Penilia the embryos have grown but little, the elements composing them simply becoming smaller with each succeeding stage ; but from this time on, increase in size of the whole is marked and rapid. In this respect it agrees with the development of Moina. The embryos and eggs from the same mother are all of the same size and stage, as far as can be detected, throughout their develop- ment. Embryos of like stages but from different mothers vary somewhat in size; a younger stage is larger than a more advanced one from another animal. Grobben found the same difference in size present in the embryos of Moina and thought the cause was traceable to variation in the size of the eggs when first laid by the different mothers. This furnishes a partial explanation for the phenomena seen in Penilia as the early segmentation stages of this animal show the same differences in size, but never to the same extent that the older embryos do. The amount and rapidity with which the nutritive blood-plasma is supplied to the embryos also vary in all probability, and this may be the principal cause in the greater variation in size seen in the older embryos. The Germ Layers. The protoplasmic rod occupying the segmentation cavity is com- posed almost entirely of the entoderm. The mesodenn originates on either side of the mid-ventral line at the angles of the gastrula, which at these particular points cannot be separated definitely into entoderm and mesoderm. Certain cells, instead of growing up into the center of the segmentation cavity, project out into these angles and toward the ectoderm. As they multiply they increase in length and width, but are only one-layered in thickness. They leave the entoderm completely and cling to the inside of the ectoderm. (Fig. 16.) These and their descendants form the mesoderm of the adult animal. Samassa finds on the ventral side of Moina a mass of tissue somewhat semilunar in cross-section. The central-lying cells form themselves into a solid cylindrical row. These are the ento- Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 117 derm, and the rest to either side of them are meso.derm. As has been shown, the separation in Penilia is much earlier. The facts observed in Penilia correspond to those given by Lebedinsky for Daphnia similis, where the mesoderm spreads forward as two sym- metrical rows from the gastrula invagination. At a later stage in Penilia (Figs. 17-23) the mesoderm has grown slowly forward and together at the mid-ventral line, making the outer wall two-layered with the exception of the dorsal side where only ectoderm exists. The gastrula mouth gradually becomes smaller, and by the time the second antennae show it has disappeared almost completely, leaving its position marked only by the entodermal rod, which soon breaks away, making it impossible to locate the spot where the anus breaks through. Grobben concludes that the mouth of the adult Moina arises directly from the gastrula mouth of the larva. This derivation of the mouth is impossible in Penilia, as will l)e seen later. In fact, it is the rule throughout the Crustacea for the anus and not the mouth to be formed from or near the position occupied by the gastrula mouth. The derivation of the anus in Penilia is open to question. The anus appears in the middle line arid in the same or immediate vicinity as that occupied before by the gastrula month. The anus may be formed de novo slightly above or below this point. The gastrula mouth completely disappears, and so it is impossible to say whether it ever reopens. The Appendages and Change in the Position of the Embryos. About this time a change in the outward form of the embryos makes itself apparent. The gastrula mouth still lies at the posterior and outer angle of each egg in reference to the mother and in the middle line in reference to the embryo. The anterior end flattens itself very slightly dorso-ventrally and becomes broader than the posterior. Two pairs of prominences appear anteriorly ; the ante- rior of which are very small and weakly developed, frequently noth- ing more than the smallest angles on this end of the embryo. The more posterior and larger are more sharply pronounced. They project backwards on either side from the anterior flattened area and make the flattening effect much more pronounced. Both pairs of these prominences lie in a plane perpendicular to one cutting the Digitized by VjOOQ IC 118 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. gastrula month and making the embryo bilaterally symmetrical. The first pair of these prominences become the first or sensory antennae, while the second will be the chief locomotor organ of the adult, the second or swimming antennae. Up to this time the position of the embryos has been practically the same and only a very slight change has occurred from that first noticed in the segmenting eggs. The embryos have elongated, and that is all. As the appendages appear the change in shape of the embryos and of the enlarging brood-chamber causes a shifting in position of the embryo with reference to the mother. Upon looking at the cross-section of an adult with a brood- ehaniber as previously described it is seen to be heart-shaped with the apex upwards at the dorsal wall and the intestine occupying the depression in the blunt end of the heart. Before any change in the outward shape of the embryos occurs they lie for the most part in the concavity on either side of the intestine with the posterior ends of the ones in front to the outside and somewhat beneath the anterior ends of the embryos directly behind. When the appendages push out the flattening at the anterior end occurs, and a resistance is encountered on either side by growing against the outer walls of the brood-chamber externally and against the wall of the intestine internally. The outer an- tennae strike the wall at an angle so that an increase in width causes the edge of the embryo to rise and revolve slightly on its other antenna, as an axis, in such a way as to cause the region once occupied by the gastrula mouth to look dorsally and outward instead of outward with a slight ventral tendency. An exception to this sometimes occurs in the case of the most anterior embryos which do not have the other factor ; i. e. the posterior end of another embryo under the outer side of its anterior end and chang- ing its position. These embryos in some cases may assume such a position that the angle formed by the antenna and the outer wall of the animal is in the opposite direction from the rest. This causes a rotation in the other direction so the position once occu- j)ied by the gastrula mouth is downward and in, instead of upward and out, as in the rest of the embryos. This opposite rotation seems to affect only tlie most anterior embryo and even in thia case it is the exception and not the rule. The normal or usual position assumed by these anterior embryos is not exactly like that taken by the rest in the brood-chamber but one in which the ven- Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 119 tral surface is farther away from the mid-line and more toward the side. Thus it is possible to distinguish almost exactly the com- parative importance of the two mechanical factors most concerned in the revolution. The embryos now increase in length so that the posterior end of each embryo lies still further back and to the outside of the anterior end of the embryo just behind it. The height of the brood-chamber increases, and still further growth and flattening of the embryo caused by the enlargement of the second antennae tend to raise the anterior of each embryo above the intestine and allow the smaller rounded end to drop so that it now lies to the outside and beneath the anterior end of the embryo immediately behind it. This constantly increasing factor adds to the rotation until it may cause the side which was originally next to the outer wall to lie on top of the intestine. With the increase in size in all directions the embryos become very much crowded in comparison with their condi- tion at first. The brood-chamber also enlarges, and the floor is raised until it is on a level with the intestine. A cross-section of it now shows it to be hemispherical in shape rather than heart-shaped. The embryos now lie almost wholly above the intestine and have lost to some extent the slant in the dorso ventral plane. The crowding occasioned upon the maturity of the embryos causes many changes of an irregular nature in their positions, so that they depart from the positions described, especially in regard to the ventral side being uppermost or directly dorsal. In nearly mature embryos, however, their long axes are more nearly parallel with that of the mother than before. The position of the embryos of Penilia is thus seen to be a changing one in the different stages of their develop- ment. To a certain extent, the degree to which the embryos are shifted depends upon their age, but throughout their development the anterior end of the embryo lies toward the corresponding end of the mother. In older stages, the ventral side is uppermost and looks toward the side, except as noted, where it is just the opposite, looking downward and toward the intestine of the mother. While the axis of the embryo is never exactly parallel to that of the mother, in a general way it always lies in the same direction, sufliciently so to see with perfect clearness that the egg is oriented and that as it develops its })osition in regard to the mother is changed by a series of mechanical factors resulting from the growth and the enlarging brood-chamber to accommodate its increasing contents. Digitized by VjOOQ IC 120 PROCEEDINGS: BOSTON SOCIETY NATUBAL HISTORY. The Appendages. The order of the appearance pf the appendages of Penilia departs from the usual rule for Crustacea. The first or sensory antennae do not appear first but only after the second antennae are clearly distinguishable, and in some cases where the retardation is excep- tional, they may not appear sharply differentiated until the second antennae, mandibles, and maxillary region are plainly formed. A maxillary region is differentiated before the thoracic appendages appear and the appendages themselves do not develop until rudi- ments of every other appendage of the entire animal are developed. The thoracic appendages appear in the usual order, as the most anterior is the oldest and those following show a perfect gradation in their development. The appendages will be taken up in the order of their position and not in reference to the time of their appearance. First Antennae. The first or sensory antennae, as has been just said, vary consider- ably as to the time of their appearancfe. They begin as angles on either side of the rounded anterior end of the larvae. They may appear in this way when the second antennae are just beginning to be plainly visible or they may not appear until the manSibles and maxillary region are marked off. This angle increases in prominence until it gradually forms a rounded lump projecting anteriorly, its connection with the body of the embryo being the smallest part of its diameter. (Fig. 28.) Its relation to the entire animal is shifted as the bending of the head and the outgrowth of the upper lip occur. From its extreme anterior position in front of the second antennae it changes to the ventral side of the embryo, and finally in the older embryos and adult its position is ventral and posterior to that occupied by the second antennae. The first antennae project on either side from outgrowths or horn-like processes developed from a part of the upper lip; They remain small throughout life and in the adult consist of but a single joint with nerve-cells and sensory fibers or setae. Digitized by VjOOQ IC 8UDLER : DEVELOPMENT OF PENILIA. 121 Thb Second Antennae. One of the first changes noticed in the outward shape of the embryo is a widening of the anterior third and the appearance of two rounded prominences projecting posteriorly on either side of this area which will form the second antennae of the adult. These two projections grow backward rapidly and, at a stage when the maxillary region and two thoracic segments are present, bifurcate. Their position in regard to the entire animal shifts, approaching more and more the anterior end of the embryo as the head bends and in the adult coming to occupy the extreme anterior dorsal corner of the animal. From their first appearance the second an- tennae are the most prominent appendages. They grow rapidly and appear first and thus foreshadow their great importance in the adult. The swimming hairs do not begin to show until just before the embryos are mature. The second antennae are folded back close to the side of the embryo as long as they are in the brood- chamber of the mother. The Mandibles. The mandibles are the next appendages to appear and the time when they are first recognizable varies but little. At a stage represented by Fig. 21, when the second antennae are well defined as two processes projecting posteriorly, the mandibles show as paired outgrowths just posterior and ventral to the second antennae. These eminences do not project either posteriorly or anteriorly but directly outward from the middle line. As the embryo grows older they assume an upright position (project ven- trally) and approach more and more the middle line. In the adult they meet in the middle line. They are stout, serrated on their distal borders, and moved by well-developed muscles. (Fig. 1.) Except for the change in position in the direction in which the appendages point, there is no shifting such as one sees in the case of the two previous appendages described. Their position in the adult is practically that occupied by them throughout the embryonic development. Digitized by VjOOQ IC 122 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. The Maxillae. A maxillary region is differentiated immediately after the man- dibles are well formed, and upon this the appendages appear later. The first pair of maxillae appear first and the second pair follow shortly. The first are the largest and show themselves as two small rounded eminences projecting at first away from the middle line but closer to it than the mandibles. Their position as devel- opment proceeds changes but little, except that the appendages of the adult point more toward the middle line. They are small even in the adult and consist of but a single joint bearing setae on its terminal extremity. The second maxillae develop at first as small thickenings which increase until the appendages are small projec- tions, reaching their maximum development in a stage represented by Fig. 27, when the thoracic appendages are just beginning to show the first trace of bifurcation. These appendages disappear as rapidly as they come, and by the time the bifurcation of the thoracic append- ages is well marked they have become reduced to mere darker stain- ing masses which soon cease to be recognizable as appendages. The second maxillae are formed sooner in Penilia than in Moina, where they appear after the thoracic appendages have already bifurcated. They persist, however, much longer in Moina, disappearing only just before the larvae are set free. They are much more transient in Penilia and never so well marked. The TnoRA( ic Api»KNi)A(iEs. The thoracic appendages follow the usual rule and appear in a regular sequence of order making the anterior the oldest and the posterior the youngest. Tlie thoracic region becomes divided into six segments in the same order. Furrows running from the ventral side up to the dorsal si«le ai)j)ear. These cause the prominences between them to become more i)roiiounced as development proceeds. This condition causes the peculiar lluted appearance presented by the embryo in side view as seen in Fig. 2(3. These i)rominences ap})roach much closer to the middle line on the dorsal than on the ventral side. A view of the embryo at this stage shows the prominences projecting slightly beyond the ventral surface and squarish in outline. This Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 123 portion represents the beginning of the thoracic appendages. Near the center of these squarish limb-rudiments which represent the basopodite a portion becomes more prominent, projecting directly downward. This eminence is the beginning of the endopodite and first shows itself in the most anterior thoracic appendage, although the difference in the time of their development is not so marked as that of the first appearance of the thoracic segmentation. The endo- podite grows ventral wards until we have it standing nearly at right angles with the exopodite. The free ends of these bend more toward the mid-line as they grow older and the shell grows back covering them. All of the thoracic appendages are bifurcated in the adult. The third of the series of thoracic apj)endages becomes the largest in the adult and they grade off from it smaller in either direction. The sixth or last pair is the smallest but preserves the structure typical of them all. The presence of six pairs of thoracic appendages is the chief dis- tinction of the genus Penilia, and Claus states in reference to the presence of six thoracic appendages in a larva of Leptodora hya- Una that we must consider this to be the original number for the Cladocera. The Shell. The shell of Penilia appears as a lobed fold of ectoderm on either side just above the maxillary region at a stage intermediate be- tween those shown in Figs. 1, 5, 2S. Tlie integument here rises up into a fold projecting j)osteriorly. This fold is lobed, as seen from a dorsal view, with the apex of tlie riiiL?ular area dividing the two shell-rudiments directed toward tlie head of the embryo. This is shown by the most anterior embryo of Fig. 1. On either side this fold runs well down to the vential surface. Seen from the ventral side when young, it resenil>K*s an aj)]»endau:e to a considerable extent, since it is about the same width and com- posed of two layers of cells and lies in the same general position as that assumed by a thoracic a[)pendage. These saddle-folds grow back gradually on either side, covering the body and apj>en- dages as we have it in the adult. The rate of growth and shapes can be seen in Figs. 1, 5, 6, 2><, and 29. Digitized by VjOOQ IC 124 PROCEEDINGS: BOSTON SOCIETY NATURAL mSTORY. The Nervous System. The nervous system of Peniliit is composed of a supraejBophageal ganglion and a series of ventral ganglia, but the supraesophageal and its commissures contain most of the nervous . material of the adult. As the most important part of the nervous system, it is the first to develop. At a stage when the second antennae, are plainly seen, represented by Fig. 20, the ectoderm is found to possess several layers on either side of the mid-line and at the anterior extremity of the embryo. These paired enlargements or accumulations of cells are the first beginning of the supraesophageal ganglion. The cells that compose them have no distinct walls, but in some sections appear almost circular, except where pressed against one another. The cytoplasm is liglit staining and the nuclei are very distinct. According to Samassa, the supraesophageal ganglion of Moina arises from paired rudiments in the same position as described for PeniUa, but earlier in its development. Grobben found in Moina that at first the rudiment of the supraesophageal ganglion was single, but afterwards became paired. He found this unpaired rudiment to appear even earlier than Samassa, and the latter says in regard to this that Grobben was mistaken in the stage of his embryos in which he first found it. The ventral chain of ganglia also rise from paired thickenings of ectoderm. These appear externally as two slight ridges on either side of a groove, running down the mid-line on the ventral side of the embryo just anterior to the stomodaeum to the abdominal part of the embryo. Grobben calls this groove the " primitivfurche " in Moina. At a stage represented by Fig. 29, the cells composing these two folds have separated themselves into nine groups, which can be seen in longitudinal sections. The third one from the anterior end of the group is the smallest and least well marked. This is the ganglion of the second maxillar segment, and probably fuses with the one just anterior to it, as Grobben describes for Moina. This condition corresponds almost exactly with the figure given by Grobben for Moina, except in that case only eight ganglia are present, which is what we should expect, Moina having one pair less of thoracic appendages. The supraesophageal commissure is seen in sections as a more or less definite string of cells running up to one side of the digestive tract. In the adult a cross-section Digitized by VjOOQ IC 8UDLER: DEVELOPMENT OF PENILIA. 125 of the veDtral nerve cord reveals a section shaped somewhat like a dumb-bell, showing that while the ganglia unite they do not entirely lose their double character. The Eye. The eye of the adult Penilia is quite small in comparison to that of other Cladocera. Its development and separation from the brain rudiment do not take place until quite late. The rudiments of the eye of Penilia do not arise separately or around distinct centers, but they are plainly seen to be lobed and paired to that extent. The Digestive Tract. The beginning of the mid-gut of Penilia is seen when the inwan- dering of cells occurs at gastrulation. The rod of tissue thus formed breaks away or very nearly so from the ectoderm, but retains its general position in the long axis of the embryo. The rod is solid, but is not regular in outline or cylindrical at first. See Figs. 15, 16, 17, 30, 31, and 32. Later this indefinite rod assumes definite- ness, and at a stage shown in Fig. 26 the mid-gut shows as a clearly defined cylindrical rod arranged radially. Later when the embryo enlarges and begins to assume the adult form, as at a stage repre- sented by Fig. 28, the lumen appears. The mouth and pharynx are formed from an inpitting which occui-s just under the rapidly elongating upper lip. The origin of the anus is also from an inpit- ting of the ectoderm at the opposite end of the embryo. Whether this inpitting occurs at the same spot at which gastrulation occurred or whether it is entirely new, I am unable to state definitely, but, judging from the position of the gastrula mouth when the embryo is revolved, I think the anus is formed dorsal to it. The anus forms in Penilia directly in the center of the posterior end and in the mid-line, while the position of the gastrula month when the embryo is revolved is in the mid-line but considerably ventral to the center of the posterior end. Grobben finds the anus of Moina arising at the posterior dorsal angle of the embryo. He also finds that the mouth and anus appear earlier in Moina than in Penilia. According to Samassa we do not have an entoderraal rod as dis- Digitized by VjOOQ IC 12(> PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. tinct in Moina as in Penilia but more or less mixed with the meso- derm. Only later the central cells of this mes-entoderm separate themselves from the main mass to form a cylindrical and solid rod of entoderm which later obtains a lumen as described for Penilia. The cells lying outside of this are the mesodeim. The Shell-Gland. The shell-gland of the adult Penilia is a marked departure from that organ as seen in other members of the group. Instead of being long, tubular, and coiled, it is almost circular in outline. It is composed of from six to ten large cells described elsewhere. The shell-gland has its origin in the mesoderm in Penilia. Certain cells in the mesoderm on either side of the digestive tract just above the second maxillary segment begin to enlarge. These cells have larger, darker staining nuclei than the surrounding mesoderm cells, and they soon show a spherical arrangement. These cells con- tinue to enlarge Until they form a solid sphere almost as large as the organ of the mature animal but with no lumen present. Later, but before they have assumed the definite histological struc- ture of the adult, they appear around a lumen with an opening to the exterior through the ectoderm. Grobben also confirms the mesodermal origin of this gland for Moina, but the different shape and histological character of the glands of the adults make their rudiments in the respective embryos somewhat different. Kingsley finds the homologous gland of Craiigon to be of mesodermal origin and that later in the devel()[)nient it acquires an ectodermal oi)ening to the exterior. Keichenbaeh in his work upon Astacus and Ishikawa in his work on Atyephira both find this gland of ectodermal origin and the mesoderm plan's no part in any stage of its development. The Cervical Gland. Througliout the arthropods one or more glands are formed in the neck and head region at various stages in their development. Different investigators have given the same structure different names such as cervical gland, neck-organ, dorsal gland, and endusium. In Penilia a cervical gland is present here in the Digitized by VjOOQ IC SUDLER : DEVELOPMENT OF PENILIA. 127 embryo but disappears before the animal is born leaving no trace of its existence in the adult. In a Pen ilia embryo at a stage rep- resented by Fig. 28 it shows its maximum development which is quite weak in comparison with other members of the Cladocera. At the stage referred to, it is composed of from fifteen to twenty cells, each of which is longer than broad and larger at the distal end than at the surface end. The whole organ is shaped somewhat like a truncated cone with a small depression at the smallest end where it comes to the surface. (Fig. 33.) It is situated in the middle line at the anterior dorsal angle of the embryo. Its posi- tion in Penilia is more anterior than in Moina. It appears about the same time in embryos of both animals but disappears much sooner in Penilia than in Moina, in which it is much stronger developed than in the former. The Ovary. The ovary of Penilia is of mesodermal origin. I have failed completely to find a genital cell from which the future genital organs originate. I have been unable also to distinguish any particular cell corresponding to the "Grobben'sche zelle " of Samassa. Lebedinsky and Samassa both agree with me in being nnable to find any trace of a differentiation of any kind at such an early stage to form the genital organs. In Penilia certain meso- dermal cells lying to either side of the intestine become larger and their cytoplasm stains clearer than those surrounding them. These form a row in a position occupied by the ovary in the adult. Samassa finds the ovaries of Moina derived from four mesodermal cells, two on either side. These cells multiply and grow backw^aid to form the ovary. In Penilia the origin of the ovary corresponds more closely to the origin of the same organ in Branchipus. According to Claus it originates here as paired rudiments lying in rows on either side of the digestive tract. The Heart. The heart of the adult Penilia is an oval-shaped body composed of about thirty semilunar 8hai)ed cells. It possesses a single pair of ostia situated toward the posterior end. The heart arises in the Digitized by VjOOQ IC 128 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. mesoderm in a position in the embryo corresponding to that occupied by it in the adult. Two lots of cells, one on either side of the middle line and above the intestine, become longer and tend to meet in a mid-dorsal position. These gradually become longer and more curved. These ends meet in the middle line and the structure of the heart is practically complete as seen in the adult (Figs. 1 and 34). Conclusion. In reviewing the general features of the embryology of Penilia certain characteristics present themselves prominently throughout its development. The orientation of the egg is one of these. An embryo always arises whose anterior end corresponds to the same end of the egg and whose long axis also agrees with that of the egg. Whether this is brought about by inherent differentiation of the egg protoplasm or whether by external influences is impos- sible of demonstration. The egg of Penilia divides transversely to the long axis of the egg into two cells, and judging from the position of these two cells, the descendants of the anterior one must have most to do with the formation of the anterior end of the embryo and the posterior one most to do with the formation of the posteiior end of the embryo. The first two cells are certainly not right and left in that their descendants form those halves of the body of the adult, as Roux found for the frog's egg or Watase for cephalopod eggs, but rather as Wilson found for Nereis, where the first cleavage is also transverse to the future animal. Grobben oriented the egg of Moina by means of the polar body which lies in the substance of the egg. This body marks the ani- mal pole, and the cell containing it in the two-celled embryo by its development gives rise to the left side of the animal. • The segmentation of Penilia is total and remains so throughout, in marked contrast to most of the Crustacea. Lucifer and Euphau- sia are the known exceptions in the higher forms. Much less yolk is present in the egg of Penilia than in most Crustacea even in nearly related fonns, and the segmentation cavity never at any time possesses a yolk plug as is found in Moina. One of the facts most frequently quoted from Grobben 's article is the very early differentiation of one cell destined to give rise to the reproductive organs. Samassa also finds in Moina the same cell, Digitized by VjOOQ IC " SUDLER: DEVELOPMENT OF PENILIA. 129 but he is unable to distinguish it from its fellows after it has sub- divided into eight cells. He cannot trace it as the mother-cell of the genital organs, and from ray work on Penilia and Lebedinsky's on Daphnia it is evident that Grobben's results are not confirmed. Certain facts in the development of Penilia lead me to think that it is a highly specialized rather than a primitive cladoceran. The second maxillae appear as late and disappear much sooner than they do in Moina. The dorsal gland is weakly developed in com- parison with other members of the group and lingeous a shorter time. The entoderm is nearly distinct as such from its origin and is never at any time intermingled with the mesoderm as it is in Moina. McMurrich thinks the origin of the entoderm and mesoderm together is the rule for the entire crustacean group. In the method of forming the eggs it resembles what is found in other Cladocera, since the egg is here apparently the survivor of four ovarian cells which unite to give rise to but one with little yolk. All of these facts indicate a specialized type. Penilia also presents other characteristics considered primitive for the Cladocera. Clans finds six pairs of thoracic appendages present in the metanauplius hatched from the winter egg of Leptodora hyalinay which he thinks must be considered the original or primi- tive number for the Cladocera. As we have seen, the presence of six pairs of thoracic appendages is the chief distinguishing feature of Penilia. If Clauses view be the correct one, then Penilia must have branched off from the cladoceran stem quite early but yet late enough to have inherited the peculiar yet characteristic method of forming its eggs each from four equal ovarian cells. I desire to express my thanks to Dr. C. P. Sigerfoos for the material upon which this work was done. Also my indebtedness to Prof. W. K. Brooks who suggested the work and throughout its progress gave advice, and to whose kind interest is due much of whatever value my results may possess ; and I also acknowledge my indebtedness to Dr. H. McE. Knower for his kindly criticisms. Digitized by VjOOQ IC 130 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. LITERATURE. Brooks, W. K. *82. Lucifer: a study in morphology. Philos. trans, roy. soc. London, vol. 173, p. 67-137, pi. 1-11. Claus, Carl. '73. Zur kenntniss des baues und der entwickluiig von Branchipus stagnalis und Apus cancriformis. Abhandl. kon. gesell. wissensch. Gottingen, bd. 18, p. 93-140, taf. 1-8. Claus, C. '70. Zur kenntniss der organisation und des feinen baues der daphniden und verwandter cladoceren. Zeitsch. f. wiss. zool., bd. 27, p. 302-402, taf. 26-28. Dohrn, Anton. '09. Untersuchungen tiber bau und entwicklung der arthropoden. 3. Die schalendrtise und die embryonale entwicklung der daplinien. Jena, zeitsch., bd. 6, p. 277-292, taf. 10. Gerstaecker, Adolph. '00-'79. Crustacea. Bronn's Thier-reich. Grobben, Carl. '79. Die entwickelungsgeschichte der Moina rectirostris. Zugleich ein beitrag zur kenntniss der anatomie der phyllopoden. Arb. zool. Inst. Wien, torn. 2, 00 pp., 7 taf. Grobben, C. '80. Die antennendrtise der crustaceen. Arb. zool. inst. Wien, torn. 3. 18 pp., 1 taf. Ishikawa, Chiyomatsu. '85. On the development of a freshwater macrurous crustacean, Atyephira compressa, De Haan. Quart, journ. micr. sci., vol. 26, p. 391-428, pis. 25-28. Kingsley, J. S. '87, '89. The development of Crangon vulgaris. Bull. Essex inst., vol. 18, p. 99-163, pis. 1-2 ; vol. 21, p. 1-42, pis. 1-3. Lang, Arnold. '91. Text-book of comparative anatomy. Translated by Herbert M. and Matilda Bernard. London. Lebedinsky, J. *91. Die entwicklung der Daphnia aus dem sommereie. Zool. anz., jahrg. 14, p. 149-162. McMurrich, J. P. '96. Embryology of the isopod Crustacea. Journ. morph., vol. 11, p. 03- 154, pis. 6-9. Digitized by VjOOQ IC SUDLER: DEVELOPMENT OF PENILIA. 131 Reichenbach, Heinrich. *86. Studien zur entwicklungsgeschichte des flusskrebses. Abhandl. senc- kenb. nat. gesell. Frankfurt a. M., bd. 14, 137 pp., 14 taf. Roux, Wilhelm. '88. Beitrftg© zur entwickelungsmechanik des embryo. Ueber die kUnstliche hervorbringung halber embryonen durch zerstorung einer der beiden ersten furchungskogeln sowie Uber die nachentwickelung (postregenera- tion) der fehlenden korperhalfte. Arch. path, anat., bd. 114, p. 113-154, 246-291, taf. 2-5. Samassa, Paul. '93. Die keimblfttterblldung bei den cladoceren. 1. Moina rectirostris, Baird. Archiv f. mikr. anat., bd. 41, p. 339-3^56, taf. 20-22. Watase, S. *91. Studies on cephalopoda. 1. Cleavage of the ovum. Joum. morph., vol. 4. p. 247-302, pis. 9-12. Weismann, August. 74. Ueber bau und lebenserscheinungen von Leptodora hyalina, Lilljeborg. Zeitsch. f. wiM. zool., bd. 24, p. 349-418, taf. 32-38. Weismann, A. '76. Zur naturgeschichte der daphniden. Zeitsch. f. wiss. zool., bd. 27, p. 61-112, taf. 6-7. Weismann, A., und Ischikawa, C. *89. Ueber die paracopulation im daphnidenei, sowie tiber reifung und befruchtung desselben. Zool. jahrb. Anat., bd. 4, p. 156-196, taf. 7-13. Wilson, E. B. *92. The cell-lineage of Nereis. A contribution to the cytogeny of the annelid body. Joum. morph., vol. 6, p. 361-480, pis. 13-20. Zaddach, E. G. *41. De Apodis cancriformis Schaeff. Anatonie et historia evolutionis. Bonnae, 1841, 72 pp., 4 pis. Pnnted, October, 1800. Digitized by VjOOQ IC SuoLEB. — Peailia. EXPLANATION OF PLATES. Abbreviations used. Ant. Anterior. Artt.l. 1st antenna. Ant. 2. 2d. antenna. B. C. Blood corpuscle. Bl Blastopore. Br. Supra-esophageal ganglion. Br. Ch. Brood-chamber. C. Supra-esophageal commissure. Cer. GL Cervical gland. Ed. Ectoderm. End. Endopodite. Ent. Entoderm. Ey. Eye. Ex. Exopodite. G. Fold between the two rudiments of the nervous system. Ht. Heart. Int. Intestine. L. Upper lip. Ms. Mesoderm. Mx.1. 1st Maxilla. Mz.2. 2d Maxilla. Mx.r. Maxillary region. Post Posterior. Sh. Shell. Sh. GL Shell-gland. Th. Seg, ' Thoracic segments. in. iSeg, ' inoracic segments. 1, 2j 5, 4, 5, and 6. 1st, 2d, 3d, 4th, 6th, and 6th thoracic segments. Digitized by VjOOQ IC Digitized by VjOOQ IC SuDLEB. — Penilia. PLATE 1. Fig. 1. Adult seen from the side ; brood-chamber contains three young, the most anterior of which is rotated in the opposite direction from the one usually seen. Fig. 2. Adult seen from the dorsal side. The most anterior embryo on the left side has turned its dorsal side uppermost. The remainder have assumed the typical position. Fig. 3. Brood-chamber of adult containing eggs in the four-cell stage. Seen diagonally ( dorso-laterally ) . Fig. 4. Brood-chamber of adult containing four eggs in the thirty -two cell stage. Seen from the dorsal surface. Fig. 6. Side view of a brood-chamber showing embryos with three thoracic segments. The most anterior embryo has its ventral surface turned inward and down. Fig. 6. Brood-chamber of an adult Penilia seen from the dorsal side. It contains four embryos in an advanced stage of development ; all of which present their ventral surface to view. Digitized by VjOOQ IC SrnLER. l^ENIUA. Pi. AT>: I . pRor. Bust. S(x*.^AT. HiHT. Vol.. *JI>. TMt MtLIOTVPC PRINTIMG CO. BOSTON. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC SuDLEB. — Penllia. PLATE 2. Fig. 7. Two-cell stage of the segmenting egg. Fig. 8. Four-cell stage. Fig. 9. Eight-cell stage. Fig. 10. Sixteen-cell stage. Fig. 11. Thirty-two cell stage. ^ Fig. 12. Sixty -four cell stage (section). Fig. 13. Section of egg having about one hundred and twenty -eight cells and showing the beginning of gastrulation. Fig. 14. A more advanced stage of the same process. Fig. 16. A still older stage showing the gastrula mouth at its greatest degree of development. Fig. 16. A section of a larva having all three germ layers differentiated. Fig. 17. A horizontal section of a larva which has the first rudiments of the two most anterior appendages. Fig. 18. Ventral view of an embryo whose outward form is beginning to change. Fig. 19. An older stage than the preceding seen from the ventral surface and showing the first traces of the first antennae. Fig. 20. An older stage seen from the ventral side and showing the broaden- ing at the anterior end and elongation of the embryo. Fig. 21. An embryo seen from the ventral side showing the first and second antennae and rudiments of the mandibles. Fig. 22. An embryo with the first three pairs of appendages clearly defined and a maxillary region marked off seen from the ventral side. Fig. 23. A transverse section just behind the second antennae of a stage cor- responding with Fig. 21. Fig. 24. A transverse section of the same embryo nearer the head. Digitized by VjOOQ IC SUDLER . — PENIUA . JO. ' .-."< 13. 13. 11 f . Plate 2. *t Vi •h't'f & * * ** • 17 0-0 yA /M v***-^ 21. 5 I^it pRoc. BosT. Soc.Nat.Hiht. Vol. 29. TMt NCUOTVM PffUmMC CO.. BOSTON. Digitized by VjOOQ IC Digitized by VjOOQ IC n Digitized by VjOOQ IC SuDLEB. — Penilia. PLATE 3. Fig. 26. An eiubryo seen from the ventral side and which has five pairs of thoracic ai>iHMHla,!matophora, 2 PpJecypoda^ I^-otobranchiata, 3. Filil>rancliiata, 6. PscMulolanielli- brauchiata, 2. Eulamollibranchi- ata, 23. Total, ()3 genera, S3 species 9. 34. Digitized by VjOOQ IC 136 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Among these, apart from the nudibranchs, are no noveltieB or even great rarities, but the following are not in Professor Verrill's (78) Report on the invertebrates of Vineyard Sound : Littorina littorea, Assiminea modesta^ Embletonia fuscctta, four species of Turbonilla, Polycerdla duvenportiij and Corambelia depresaa^ while the range of Aatarte undata^ Tergipea despecttcs^ Hemiaea cniciata, and Cratena gymnota is extended west of the localities then known. Sanderson Smith and Temple Prime (70) published a report on the MoUuaca of Long Island which represented eleven years' col- lecting and dredging and contained a great number of nominal species. It gave to the north side of Long Island inside of Mon- tauk Point the equivalents of about 111 modern species, of which 71, or 64% are in my list which also contains 17 species not in 5Smith and Prime's list. The additions are Littoriiia Httorea^ Alexia mgosotiSy four species of Turbonilla, several nudibranchs, and Macoma sabulosa^ which last, however, is represented only by one worn valve whose occurrence may be accidental. The gaps in the list are perhaps more interesting than the occurrences. An eastern port where neither Purpura lapiUus nor lAttorina irrorata occurs is fairly well defined, geographically. Ten years ago it might have been possible to define the spot within 60 miles by saying it was a place where P, lapiUua was not, and X. littorea was, found, but now the wave of the conquering European species has spread far down toward Virginia and at Coldspring the native competitor, Nassa obsokta, begins to yield room. The Caecidae, Bela, Lacuna, Skenea, Missoa aculexia^ Odostoniia impressa and 0, seminuda^ Turbonilla interrupta and T. elega^iSy the Elysiidae, Corbula, Saxicava, Siliqua, Donax, Tellina teneUa^ Pholas, all are conspicuous for absence and another summer might possibly account for some of them. In the list of the Gasteropods as it stands 18 % are ** northern " (i. e. forms characteristic of waters north of Cape Cod and usually not extending farther south than New Jersey unless in deep water), 41 % are " southern *' (i, e. forms extending south to Hatteras or Florida and found north of Cape Cod only in Massachusetts Bay or those scattered colonies of southern forms which still dot the coast up to Anticosti), while 42% are so general, so local, or so little known in distribution as to be unassignable to either category. Of the pelecypods 21% are '* northern." 52% are "southern,** and 27% are unassignable. Or, for the whole, 19-f-% are "northern," Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 137 45 -f- % are '* southern," and 36 -f % unassignable. I think the similarity of the figures in gasteropods and pelecypods consid- ered separately shows that these percentages stand for a real fact of distribution. Two faunas overlap, the more southerly contribut- ing rather more than twice as many species as the northerly. The fojlowing species which do not occur at Coldspring Harbor (so far as known) are given by Perkins ('69) as occurring at New Haven on the opposite shore of Long Island Sound about forty miles to the eastward : Amy da dissimilis Stimps. (= Astyris zonalis Linsley), three specimens found among a lot of A. zonata; Scalaria inuUistriata Say {^=,JScala m. Say), very rare; Odostoniia deaUmta Stimps., very rare; O, impreaaa Stimps., not common; O, seminuda Gould (= O. seminuda C. B. Adams) , very rare ; Pleurotoma brunnea Perkins {=: J^efa plicata Adsuns) , one speci- men thrown up by the waves ; Simnia xniiplicaUda Adams {^=.8, uniplicata Sowb.) , one specimen probably brought up on southern oysters; Jiiasoa aculens Stimps. (= B, aadeus Gould) , common under stones and on algae near low water; Rissoellaf ebumea Stimps. (= Hiasoa ehurneaf Stimps.) one beachworn specimen taken by Dr. Perkins which Vemll (73, p. 655) remarks ** may have been some other species .... I have seen no undoubted shells of this species south of Cape Cod " ; Skeiiea planorbis F. & H. (=z S, planorbis Fabr.), taken with Alexia myosotis, not com- mon; Lacuna vincta Gould (= X. vincta Turton) , not very com- mon ; IAttori7ia irrorata DeKay (= X. irrorata Say) , not at all common ; Tornatella puncto- striata C. B. Adams (= Actaeon p. C. B. Adams), dead on the beach, not common; Cylichna oryza Stimps. (=: (Jylichnella o, Totten) , dead on the beach, very rare ; Cyrtopleura truncata Tryon (= Pholas [Barnea) t. Say), clay and peat at high water, not rare ; Martesia cimeiforniiH Tryon (= M. Ciineifonnis Say) , one specimen in a pile of beach shells ; Saxicava arctica Linn., sand near low water, not common ; Corbula contra eta Say, sand at low water, not common ; Siliqua costata Con. (= A\ costata Say), very rare and small; Angulns polita? Tryon (= Tdlina p. Say) , a few specimens doubtfully so referred ; KeUia planidata Stimps., in accumulations of small shells on shore, not common; Brachydontes hamatiis Say (= Jfytilus h. Say), abun- dant on southern oysters, doubtful if it is naturalized. The following species occurring at Coldspring Harbor do not appear in Perkins's list: Chaetopleura apicidata Say, abundant; Digitized by VjOOQ IC 138 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Littorina littorea Linn., abundant; Assiniinea niodesta Lea, locally common ; two species of Turbonilla, rare ; Ilerrnaea cruciata A. Ag., rare ; Cratena gyrnnota'i Couth., rare ; C. pilata Gould, com- mon ; Terglpes despectus Johnst., rather rare ; EmhUtonia fuscata Gould, rare ; Polycerella davenportii nobis, locally not rare ; Co- ramhella dtpressa nobis, locally not uncommon ; Eriphyla^ lunulata Con., rare ; ? ? Lucinafilosa Stimps., a drift shell ; Macoma sabidosa Spengler, a drift shell ; GcinTtia inanhattensis Prime, rather rare. Doubtless a certain allowance must be made for doubtful identi- fications and not much importance can be attached to finds of one or two dead specimens on the beach, but it appears that there were resident at New Haven at least one Astyris, three Odostomias, one Rissoa, Skenea, Lacuna, JAttorina irrorata^ Pholas, Saxicava, Corbula, and Siliqua, which either do not live at Coldspring Harbor or are so excessively rare or so excessively local or capricious as to have escaped search. Any of them may yet be found there, how- ever, and some, such as JAttorina irrorata^ probably will be. One very abundant Chiton, a highly local Assirainea, two species of Turbonilla, Eriphyla, and a southern variety of Gemma appear to be legitimate inhabitants of Coldspring Harbor, but not of New Haven. A comj)arison of the relative abundance of various spe- cies emphasizes the difference. In Perkins's list, omitting the non-acclimatized forms imported on southern oysters (to which there is nothing corresponding at Colds]>ring Harbor), of the gasteropods ^Ih — ^Jf^, are '* northern," 48% are " southern, '* while 27-f-% are unassignable. Kather curi- ously the pelecypods give the same figures so that the total is the same. In the Coldspring Harbor list, omitting the nudibranchs (which apparently were not collected by Dr. I'erkins), of the gas- teropods 14-1-'^^, are "northern," 49— '^^ are *' southern '* while Z^-\-f/r are unassignable; of the pelecyi)0(ls 23-1- <}{ are " nortli- ern," 52-f-^/^ are "southern," while '24-f'^v are unassignable; or for the total 18 -f- 9;^ are " northern " 50+ f^f, are " southern " and 81 -h ^^ are unassignable. The omission of the nudibranchs emphasizes the southern aspect of the fauna on the face of the record, but there may well be doubt whether the attribution of many of the s|)ecies to northern waters only is not a mere effect of our lack of knowledge of this group. Doubtless figures like these give a false impression of accuracy, since opinion may differ as to whether to class a species as " south- Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 139 em '* or unassignable, but the sti'iking correspondence between the figures for gasteropods and pelecypods and between the results for Coldspring and New Haven (results obtained at first by independ- ent classings of the species which were only afterwards compared) shows, I repeat, that they represent real facts. The comparison between Coldspring and New Haven at least is just. The upshot is, as before, that two pretty distinct faunas in this region overlap, the more southern one contributing a quota rather more than twice that of the more northern one ; and further, the increase in the preponderance of southern forms can be detected in a range of forty miles. In the list which follows the nomenclature adopted is that of Dall ('86, '89, '89) wherever possible, while in one instance I have followed Apgar ('91). The arrangement of the gasteropods is that of Fischer ('87), excepting the nudibranchs where I have followed Bergh ('92), and that of the pelecypods is that of Pel- seneer ('94, '97). The only synonymy attempted in most cases is Verrill's names in his report ('78) and Smith and Prime's names in their report ('70), given for the sake of convenience in comparison. I have adopted the scale of "very abundant" "abundant" " very common " *' common " '* tolerably common " " rather uncom- mon " *' uncommon " " rare " and " very rare " as the best available way of describing the present condition of the mollusean population. Too vague to serve as an absolute measure it is yet to be hoped that it wilf prove accurate enough in relative terms to enable some future student to detennine what changes in distribution and adjustment of equilibrium a given number of years may have made in the so-called '* permanent residents" of a given locality. If it could be supplemented by a series of quantitative determinations of the actually prevailing " mode " in a series of the local forms it might prove a useful reference-point for future comparison. A capital N or 8 following the synonymy will indicate that the corresponding species was counted as " Northern " or "Southern." CEPHALOPODA. LOLIGIXJDAE. Ijoligo pealli Les. Smith and Prime ('70), p. 405. Verrill (73), p. 685. N. One was taken in August, 1899, and specimens from the harbor Digitized by VjOOQ IC 140 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. are preserred at the State fish commission, which has a station here. This species certainly occurs occasionally, if rarely. OASTEROPODA. AMPHINEURA. POLYPLACOPHORA. Chaetopleura apiculata (Say). Verrill (73), p. 661. Chiton a. Smith and Prime (70), p. 892. S. Abundant and unusually large on, and especially in, old shells, etc., 1-3 fath. PROSOBRANCHIATA. DOCOGLOSSA. Acmaea testudinalis (Muller;. Verrill (73), p. 661. Tectura L Smith and Prime (70), p. 392. N. One shell, empty but fresh. Doubtless occurs very rarely on the more exposed points where beach boulders offer some poor foothold. Ptenoglossa. Scala lineata (Say). Scalaria L Smith and Prime (70), p. 394. Verrill (73), p. 660. S. One fine specimen, empty but fresh, buried in clean, sandy mud at low water. Taenioglossa. Naticidae. ^everita duplicata (Say). Verrill (73), p. 646. Natica d. Smith and Prime (70), p. 396. S. Common on the firmer bottoms at and below low water. Lunatia heros (Say). Verrill (73), p. 646. Natica h. Smith and Prime (70) , p. 395. Common but less so than the preceding, with which it would seem to compete. The var. triseriata (Say) Verrill (73), p. 646. Natica t Smith Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 141 and Prime (70), p. 395, is abundant in deeper water where it out numbers either of the preceding. Capulidae. Crepidida convexa Say. Smith and Prime (70), p. 392. Ver- rill (73), p. 650. S. Common, but less so than the two succeeding species, and appar- ently seeks deeper water. £gg8 still in the two-cell stage up to August 20. Crepidvla fomicata (Lam.). Smith and Prime (70), p. 392. Verrill (73), p. 649. S. Abundant at and below low water on Limulus, Fulgur, etc. Eggs abundant in July and some still in 2-cell stage on August 23. Crepidula plana Say. Verrill (73), p. 650. C. ungxdformis Lam. Smith and Prime (70), p. 392. S. Abundant on and in other shells, especially Fulgur car tea and F, canaliculata, LiTTORINIDAE. Littoritia littorea (Linn^). N. Still a recent arrival (having reached New Haven only in 1880), and does not appear as yet seriously to threaten Xassa obsoleta, the native competitor for the mud flats. Far less abundant than X, paUiata or X. rudis^ from which it differs considerably in station, it is yet common on the edges of marsh and mud flat, and every- where scattered sparsely among the A\ obaoleta. It is not the clean dark shell of Massachusetts Bay, but dingy gray and green with vegetable growths like iV. obsoleta^ and appears to average larger than north of Cape Cod ; the body-whorl seems niore inflated, the suture less well marked. Bumpus's ('98) studies on the ven- tricosity of this species seem to confirm such a tendency in southern localities. Littorina paUiata (Say). Verrill (73), p. 652. X. littoralis var. jt>eco/iica Smith. Smith and Prime (70), p. 393. N. Abundant 6n sea walls, piles, and wherever Fucus will grow. As variable here as elsewhere. lAttorina rudis (Maton). Smith and Prime (70), p. 392. Verrill (73), p. 651. N. Digitized by VjOOQ IC 142 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Abundant with the preceding. Relatively more abundant in less brackish water. Extremely variable here as elsewhere. RiSSOIDAE. lihsoa mimita (Totten). Smith and Prime (*70), p. 393. Littorinella m. Verrill (73), p. G53. N. Very common on Ulva and in black mud, upper harbor, in water almost fresh. liiasoa laevis (DeKay). Smith and Prime ('70), p. 393. Litto- riaella L Verrill (73), p. 653. ILjdrohia't I. Ven-ill ('82), p. 523. ? Rissoa stiinpsoni Smith. Smith and Prime (70), p. 393. This and the preceding and two succeeding species are in such confusion as scarcely to rei)ay any attempt to straighten out the nomenclature without further knowledge. Stimpson ('05) on the strength of the dentition and anatomy of JL ininuta retained the genus Littorinella in his subfamily Hydrobiinae (= Amni- colinae of (4ill) distinguished from Rissoinae on dentition alone, but suggested a new genus — P^crobia — which Fischer ('ST) adopts. The shell has also been referred to Paludinella, Cingula, and Ilydrobia. As to II, laevis the case is much worse, for not only does it share the above changes but sometimes one appears in one genus or family and the other in another, while in truth the specific difference of the two is at least doubtful. As the dentition and anatomy of 11. laevis have never been examined (so far as I know), any attempt to i\n\\i^.)^:=i Bucrinnni irheatleyi DeKay, is tolerably common. Xassidae. Xassa obsoleta (Say). Smith and Prime (70), p. 897. llyan- assa o. Verrill (73), p. 641. Abundant everywhere, especially on mud flats where it is dom- inant. Kassa vibex Say. Smith and Prime (70), p. 397. Verrill (78), p. 640. S. Uncommon. Capriciously distributed among X. obsolete, from which it is not always readily distinguishable. A fairly well- marked variety, yerha\}H=^ X. /rete/isis Perkins ('69), occurs, occa- sionally in some numbers, on eel-grass. Nftssa trivittatfi Say. Smith and Prime (70), p. 397. Tritia i. Verrill (73), p. 641. Abundant in proportion as the situation becomes too exposed for X. obsoleta^ and in deeper water. TURBIXKLLIDAE. Fuhjvr carica {\Am\6) . Verrill (78), p. 640. Pi/Dfla c. Smith and Prime (70), p. 898. S. Common below tides, but less common than the following. Fulfjur canaUnddtd (Linnc). Pyrt/ht r. Smith and Prime (70), p. 398. /Si/cottjptis c. Verrill (78), p. 640. S. Very common, but not perhaps abundant, from '2 fathoms out. Digitized by VjOOQ IC 148 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. TOXOGLOSSA. CONIDAE. Ma7igilia cerina (Kurtz and Stimps.). Venill ('73), p. 637. Pleurotema c. Smith and Prime (70), p. 398. S. One specimen from sandy mud in 7 fathoms. OPISTHOBRANCHIATA. TECTIBRANCniATA. TORXATINIDAE. Tornatina canaliculata (Say). JinUci c. Smith and Prime (70), p. 399. Utricidtis c. Verrill (73), p. 663. S. Not uncommon below tide marks, but usually dead. The speci- mens fall into two well-separated groups, the first consisting of larger shells with yellow epidermis and tapered spire, the second of much smaller shells usually greatly eroded, the remains of the spire much thicker and flatter, the columella tooth less conspicuous. Some of the latter answer better to descriptions of lietusa pertenuis Mighels than they do to those of T, canaliculata. Why the old shells should have the epidermis usually intact while the young shells usually lack it, is a puzzle. The same thing is noticeable in Cylichna alba Brown, from more northern waters. BULLIDAE. Jlconinea solitaria (Say). Bulla 8. Smith and Prime (70), p. 399. Verrill (73), p. 662. S. Dead shells common in dredging. Rather uncommon alive. Occurs on marsh grass, top of sea wall, etc. ASCOGLOSSA. Hekmaeidae. Jltnuaea critciata A. Agassiz MSS. Gould (70), p. 253. Verrill (73), p. 667. Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSC A. 149 Two specimens from Ceramiura and other red sea-weeds on stony bottom off Lloyd's neck. Only once before reported, so far as I know, viz., by Agassiz at Naushon in 1808; but Professor Verrill informs me that he has taken it several times. Probably it is not so mnch rare as capricious in occurrence, like so many nudi- branchs. NlTDIBUANTHIxVTA. Kladoiiepatica. cuatenidak. ? Cratena yymnota (Couth.). Hergh ('92), p. 81. Coryphella (J. Verrill ('78), p. G()7. These specimens, found depositing eggs on eel-grass, Aug. 26, were unfortunately not preserved for complete identification. They differed markedly from var. gouldii Ven-ill (Bergh ('9*2), p. 81, := Mo)ita(/tui g. Verrill ('78), p. 067), of which one specimen was taken from compound tunicates. On the authority of Profes- sor Verrill this is made a variety of C. gymnota, (Jrntena pilata (Gould). Hergh ('92), p. 81. Montngna p. Ven-ill ('78), p. 000. 31, vemufera Verrill ('78), p. 000. Aeolis vermiferus Smith. Smith and Prime ('70), p. 891. Common but of small size on hydroids from lobster-pot lines a fathom or two below the surface. Also in dredgings of eel-grass. On the authority of Professor Verrill Aeolis vermifents Smith is reduced to synonymy. Specimens answering both descriptions were taken. Tkrgipedidak. Tergipea despectas (Johnst.). Verrill ('78), p. 007. Bergh ('92), p. 81. N. Rather uncommon on hydroids from lobster-pot lines. Embletoniafuacata Gould ('70), p. 251. Bergh ('92), p. 84. N. liather uacoraraon on hydroids from lobster-pot lines. This is the first reported occurrence south of Cape Cod so far as I know. Digitized by VjOOQ IC 150 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. HOLOHEPATICA. Phaxerobranchiata. polyc'eridae. PolycereUa davenportii sp. nov. PI. 1, figs. 2-7. Body limaciform, slightly constricted behind the rhinophores. Length 2- 4 mm. Breadth (when creeping) about i of length. Color dirty green, speckled with black and splashed with sulphur-yellow around and on rhino- phores. Mantle distinct, without marginal flap, not covering the foot which is pale yellowish and long in extension posteriorly where it tapers evenly to a notched point. Anterior angles of foot moderately prolonged as oral palps, anterior margin usually salient. Rhinophores long (in extension ^ length of body), heavy, clavate, simple, contractile, not foliate, laminate, or retractile, without sheaths. Gills three, small, rudimentary in appearance, dorsal, me- dian, each consisting of a recurved stem bearing three posteriorly directed branches between which is stretched, in perfect specimens, a delicate web. The middle gill is set on the very conspicuous pulsating cardiac prominence and shows scarcely any branching, the web being thicker than in the lateral members. Occasionally a single rudimentar>' fourth branch anteriorly directed appears. Dt)r8al papillae not on edge of mantle, usually one pair in front of, one at level of, and one behind the gills. Two smaller pairs form a posterior rosette. Papillae small (J length of contnicted rhinophores), inconspicuous. Mouth anterior, funnel-shaped, dorsally exposed in extension, aiuied with thin mandibular lamellae. Anus, a transveree silt, median, dorsal, just iR^hind and under gills. Radula almost as in P. emertonii Verrill ('80-81, p. 387 ; '82. p. 548), rhachidlan tooth wanting; pleurae strongly hooked with accessory point.s, large ; uncinl two, sickle-.shaped. Formula 2-1-0-1-2. This odd little sea-slug is most nearly related to P. e)nertonii Verrill ('80-81, p. 387; Bergh, '83, pi. 9, figs. 1-0 and pi. 8, figs. 9-19), which it resembles in general organization, color, size, and dentition and from which it differs in dentition (slightly), in the fewness and smallness of the papillae and the relatively much larger rhinophores, in the fact that the webbed gills are simply and singly phinate instead of simply but doubly (ahernately) pinnate, in the shape of the gills and number of their branches (which is much greater in P. emert07ii'i)^ and finally in the character of the foot which in P. emertonii is covered by the mantle. Professor VerrilTs great Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 151 kindness in showing me his unpublished sketches of P, emertonii has allowed me to satisfy myself of the very different appearance of the two. Were it not for the great similarity of the dentition F. davenportii would seem entitled to generic distinction on the ground of the very different gills and the uncovered foot. No armature of the penis could be made out in sections, but this was perhaps owing to poor preservation. Specimens of P. davenportii were first taken on August 16 from- hydroids on lobster-pot lines, and again about two weeks later appeared in jars of stones, weeds, hydroids, etc., which had been, dredged in about 3 fath. CORAMBIDAK. Corambella gen. no v. PI. 1, figs. 12-15. Form Corainbe-like (c/. Bergh, '71 and '92, and H. Fischer, '91), but more convex and proportionately longer. Notaeum as in Corambe, but without the anal notch. Rhiuophores not foliate laminate or branched ; tapered, retractile, with sheaths. Anus and gills as in Corambe, viz. : anus median, posterior, between the foot and the gill plates, which lie posteriorly on either side between foot and notaeum, completely hidden by foot in life, (ienital papilla anterior, left side, hidden in life. Jaws and pharyngeal bulb without plates or other armatui-e. RadiUa large, no rhachidian tooth ; pleurae large, twisted, the median ends bent up and back in a heavy hook ; uncini live, stout, claw-shaped. Formula 5-1-0-1-5. This genus is erected to contain a form closely allied to Cor- ambe sargasaicola Bergh and C. testudinaria H. Fischer, but not to be included within that genus because of the lack of the anal notch and the different dentition. The anal notch is by Bergh made a family character, yet no doubt the present form must fall within his Corambidae. Corambella depressa sp. no v. Form Doris-like, much flattened in life, broadly rounded in front and bluntly tapering posteriorly. Length 5 mm., broadth 3 mm. Color sometimes dull brown with gi-ay irregular reticulation, but occasionally rather bright and conspicuous, when a translucent dull blue gi*ound-col(»r is hidistinctly, coarsely, and irregularly blotched with dirty green or greenish brown and irregulaily scattered with small black spots, the i)attern darker and closer toward the center, producing a radial effect. Between the blotches in every direction run very conspicuous opaque yellowish white lines like the borings of larvae under Digitized by VjOOQ IC 152 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY bark, beoomiiif? coarser toward the margin and heightening the radial effect of the whole. These lines serve to pick out smartly the rhinophore oiH-ninp*. In forniol specimens the color becomes black, and in specimens cleared and viewed by transmitted light a coarse reticulatirm around the edge of the nota^mm becomes very conspicuous especially posteriorly. Notaeum l(M)8e and ami)ly covering all parts, often up-tiu-ned at the margin in life ; dorsal aspect, which in life is smooth, has a shiigieen-like surface in preserved specimens from the calcareous particles of the integument. Foot rather long, bluntly tai)ereil. anteriorly strongly cordate, posteriorly roundly pt)inted in life (in i)reser\'ed material usually emai'ginate). Oral disc only as broad as foot, strongly con- vex anteriorly ; oral pali)8 very thick and .short. Rhinophores long (i maxi- mum brea^Uh of specimen), evenly tapered, covered for about half the length by a delicate white sheath. Entirely retnu'tile together with sheath. Gills, which are most of the time hidden in living animals, lie posteriorly in the s])iu*e between foot and notaeum, and consist of a set of simple overlapi)ing l)lates on each side of the cardiat^ vessel which unites them. The.se, together with the anal papilla which lies ventral and anterior to the heart in the median line, are only exposed in a few even of the pre.st^rved specimens. On the left fiide, just bfliind the neck, lies the genital (»pening from which the genitalia are usually everted in preserved material even wIumi the animals were first .stupelied with magnesium sulphate. In some .sections through this .**exual extrovert ilure appeared stnmgly-staining plumose structures at first sugire.st- ing gills and possibly representing a complicated armature of the penis but not positively identified by me. No signs of mandibular lamellae, t)r of the horny "Balken" described from the mouth cavity of Corambe, api)ear. The radula is de.seribed HUj)ra. l^iifortunately this inconspicuous little nudibranch, which occurred during August in fair numbers under stones on whicli Fucus was growing on the edge of swift water, and in dredgings of stones and weed from *2 fathoms, was at first taken for DorideUa obscura Ver- rill (78, pp. 400 and 0()4) and ('Hl-'8!> p. 547) and therefore only casually studied and sketched in life. Consequently I am uncertain what is the exact structure of the rhinophore and sheath, but probably, as in Corambe, the sheath is longitudinally slit on its posterior (superior) face and the rhinophore itself is, for the pro- jecting portion, a sheet with the edges rolled inward posteriorly, as tlie comparison of my sections with Fischer's (*91) very com- plete account of the anatomy of Corambe testiid'marin makes it clear thai almost all the structures bear some general resemblance to that species. As the animal, in which periods of considerable activity ap[)ear to alternate with periods of obstinate passivity, creeps on the under surface of the water the rhinophores are com- monly turned back in a graceful curve on each side. The gills are Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSC A. IT);) usually hidden, but rarely the progress of the animal is stopped, the foot, which before extended beyond the mantle border, retracted, and the extremely delicate and glassy gill-plates rapidly thrust in and out first on one side then on the other, at such times closely resembling the tips of some small beetles' wings thrust from under the elytra. The form of the notaeum varies constantly in life, often appearing broadest posteriorly and sometimes slightly emarginate there but never so broad as long. The relation of this form to VerrilTs DorUhUa ohsctint, which it somewhat resembles superficially, is interesting but rather uncertain. Both Fischer and 15ergh j)lace), all these of one genus, and two in Long Island Sound occupying the same station and belonging to different genera. PULMONATA. Hassomatopiiora. AuniClMDAE. Alexin myosotin (Drap.). Verrill ('7v{), p. iMVl, X. On stones at and above high-water mark. Usually near brackish water. Sometimes associated with Assinnnea modesta. Uncom- mon. 3fel(unpu8 lineatus Say. M. bid^ntatus Verrill (78), p. (>()2. M, corneuH Smith and Prime (70), p. 899. S. Common on edges of salt marsh, on grass stems, etc. The banded variety occurs sparingly. Digitized by VjOOQ IC 154 PllOCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. PELECYPODA. PROTOBRANCIIIATA. NUCL'LIDAE. N'ucula proxima Say. Smith and Prime (70), p. 3S5. Ver- rill (73), p. 691. Abundant everywhere on muddy or shelly bottoms below I fath. Ledidae. Yoldla Ihaatnla (Say). Verrill (73), p. GiS9. J^edd Umatuht Smith and Prime (70), p. 385. Locally common in black mud, 3-6 fath., and attaining a length of IJ inches. The number of teeth is variable, ranging from 16 anterior and 13 posterior to 26 anterior and 24 posterior, so that the numbers usually given on the authority of Gould (22-18) are mis- leading. Probably, as Professor Verrill believes, they increase with age, but it is easy to find small specimens with many and large specimens with few teeth. In shape the specimens are all nearly typical limattdas, the form sei)arated as I" sajjotida not occurring. SOLEXOMYIDAE. ^oleiiomya veliua Say. Smith and Prime (70), p. 389. Ver- rill (73), p. 6S8. Locally not uncommon at and below low water; occurs a foot or so below the surface of the tine clean sand and mud in which it lives. FILIBP.VXCIIIATA. AXUMIIDAE. Anoiiiia shnph\c d'Orbigny. A. epluppium Smith and Prime C70\ p. 384. A. (jlahva Verrill (73), p. 690. S. Extremely abundant everywhere. In many places the dredge comes up tilled with the dead shells to the exclusion of almost everything else. Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 155 Arcidae. Area transversa Say. Smith and Prime (70), p. 8H4. ^Sca- pharca t, Verrill (78), p. G91. S. Common on shelly bottom, 2 fath. and below. Area pexata Say. Smith and Prime (70), p. 8S5. Argina p. Verrill (73), p. 092. S. Occurs with the last species but not one fifth as common. Per- kins ('09) notes the opposite condition at New Haven. Neither of these " bloody-clams " appears to liave the deep-burrowing habit which Lankester's explanation (78) of the presence of haemoglobin in the blood calls for. Mytilidae. Mytilas edulis Liun^\ Smith and Prime (70), p. 880. Verrill (78), p. 092. N. Abundant everywhere, but the great "mussel-ridges" so charac- teristic of some waters do not occur ; Gould's var. jtellucidHS is not uncommon. Modiola modiohts (Linn(»). Verrill (78), p. 098. Mi/tilus 7ii, Smith and Prime (70), p. 8S0. N. No live specimens were taken, but doubtless the species occurs in the harbor as very fresh valves were tolerably common in the dredge at certain spots. Modiola jylicatula Lam. Verrill (78), p. ()98. Mytilus p. Smith and Prime (70), p. 8S0. S. Abundant everywhere, especially about salt marshes. Specimens from a certain marsh near Lloyd's Harbor show curi- ous distortion and erosion effects so constant as to resemble a real variety; but doubtless the erosion, which is sharply confined to the beaks which are buried in the soil, is due to the presence of humous or other acids and probably the distortion is due to the same cause. In the same marsh the clam shells were very curiously distorted ; cf, infra. Digitized by VjOOQ IC ir)(> PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. rSEUDOLAMELLIBRANCIIIATA. OSTUEIDAK. Ostrea virgintca Gmelin. O. virr/iniana Smith and Prime ('70), p. 884. Verrill (78), p. ()97. O, horcalis Smith and Prime (70), p. 8H4. O. costata Smith and Prime (70), p. 884. S. Abundant, but largely artificially planted. Owing to the low specific gravity of the water and the abundant food this is a famous fattening ground and the industry is the principal one of the ])lace. In 1898 the natural "set" was very small (as it had been for sev- eral years), owing, as the fishermen believed, to severe thunder- storms during the period when the " fry " were " swimming '' which they variously estimate at from July 25 to August 7. In 1899 the ''set" was heavy. The method of cultivation here is simply to clear the ground of '' wrack " and '* sludge " and plant " native " oysters (mostly from Bridgeport, Conn., which has oysters in abun- dance but no such fatten ing-ground as this) on grounds in the outer waters sj)ecially prei)ared by ''cleaning u]> " and then spread- ing gravel of such a size that each oyster may attach itself to a separate bjvse histead of growing in bunches. As the young approach marketable size they are moved farther up the harbor to fatten. Pkctinidae. Pecten irradkins Lam. Smith and Prime (70), p. 884. Verrill (78), p. 095. S. Abundant on eel-grass beds in outer harbor. The spat covered the eel-grass on Aug. 4th. Scallops are to some extent cultivated like oysters here, but the industry does not attain to the importance it does in Buzzard's Bay and on some parts of Caj)e Cod. EULAMELLIBUANCIIIATA. ASTAUTIDAE. Astarte unduta Gould. Verrill (78), p. 684. A.sidcata Flem. Smith and Prime (70), p. 887. N. One live specimen from gravel and shelly bottom in 4 fath. This was one of the surprises of dredging, as the species is here at Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. lo7 its extreme southern limit, I believe, as a shoal-water form, and 4 fath. would be above its usual habitat even in much more northern waters. But the animal was alive and admitted of no doubt in identification. Smith and Prime admit the species only doubtfully on the authority of I)e Kay as " rare " and assign it a range from Stonington north, so this is really an addition to their list. Verrill mentions no locality west of New Haven, where Dr. Perkins found var. liUea (=zAstarte httea Perkins). Astarte castdjiea (Say). Smith and Prime (*7()), p. 887. Ver- rill (73), p. OSf). N. A few live specimens in black mud, 3-4 fath., and a few old valves. Rare. *n Astarte qttitdntns (4ould. Verrill {'78), j). 0^5. It is doubtful whether the shell so identified was correctly placed, even supposing this to be a good species. It is at all events a marked variant from the preceding. CllASSATKLLIDAK. Eriphyla Innnlata (Conrad). Astarte /. Smith and Prime (70), p. 387. Goahiia mactracea Gould, Verrill (78) , p. 085. S. One dead specimen, fairly fresh, and several much worn valves from hard bottom in 4 fath. LrciNinAE. ?? J^uci/iafilosa Stimj)s. Verrill (78), p. OSii. A very small shell, about 5 mm. high, extremely compressed and with very conspicuous concentric lamellae, is assigned to this species with the greatest uncertainty. It was fit first taken for the young of Vemcs tnerccnaria^ which it much resembles, but there is no pallial sinus, and the teeth, though so undeveloped as to be equivocal, point to the Lucinidae. Being eroded and immature, it is beyond positive identification and may have come from a distance, very possibly carried by some fish. Tellinidae. Tellina tenera Say. Smith and Prime (70), p. 889. A?tf/f(lus tener Verrill (73), p. 677. S. Tolerably common in black mud, 3-4 fath. Digitized by VjOOQ IC 158 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Macoma tenia (Say). Smith and Prime (70), p. 88&. Verrill (78), p. 678. S. Locally abundant in black mud below tide marks. Macoma balthica (Linn^). M, fragilis Adams. Verrill (73), p. 676. Tellina fusca Philippi. Smith and Prime (70), p. 389. Dead but perfectly fresh and unusually large specimens locally common 1-3 feet deep in black mud at low- water mark in Upper Harbor. The animals doubtless live in numbers close by, prob- ably deeper in the soft mud, although in more northern waters and on harder ground it may be found alive free upon the surface. The rather surprising combination of the heavy shell and epidermis characteristic of muddy localities with the brilliant rose color usually seen only in small shells from clean sand occurs. Macoma sahulosa (Spengler). Verrill (73), p. 677. N. One valve found by Dr. J. I. Ilamaker was so identified by him. Semelidae. Cumingia teUinoides (Conrad.). Smith and Prime (70), p. 388. Verrill (73), p. 679. S. A few valves on hard bottoms in 4 fath. Rare. Mactkidae. Macira solidhsima Dillwyn. Smith and Prime (70), p. 388. Verrill (73), p. 680. Relatively not very common or very large. Lives in mud and sand in more exposed situations. Mactra lateralis Say. Smith and Prime (70), p. 388. Verrill (73), p. 680. S. Abundant in black mud below tide marks, slightly buried. Vexeiudae. Ventis intrcenaria Linn)elow tide marks. A considerable clam-fishery exists. Cytherea conveo'a Say. Smith and Prime (70), p. 388. Cal- lista c, Verrill (73), p. 681. N. Digitized by VjOOQ IC BALCH: LIST OF MARINE MOLLUSCA. 159 Rather uncommon on soft bottom, 3-4 fath. Gemma purpurea H. C. Lea. Venus gemma Totten. Smith and Prime ('70), p. 388. Tottenia gernma Verrill ('73), p. 682. Locally abundant, lying on the surface of black mud and fine gravel on the flats. var. 7nanhattensis Prime. Venus m. Smith and Prime ('70), p. 38S. Tottenia m. Verrill (73), p. 082. S. A few specimens so identified occurred capriciously among the G. ptfrpurea. Rare. Petiuoolidak. Petrlcohi pholadiformis Lam. Smith and Prime ('70), p. 390. Verrill (73), p. 080. S. Common, boring in the salt marsh among roots of grass, etc., at top of beach. Cardiidae. Cardium pinnidatam Conrad. Smith and Prime ('70), p. 387. Verrill ('73), p. 083. N. One live and several dead specimens on hard bottom in 6 fath^ Rare. Liocardhim 7nortoni (Conrad). (Jardium m. Smith and Prime (70), p. 387. JLaevlcardium m. Verrill (73), p. 083. S. Abundant in soft mud at and below low water. Myidae. Mga arenaria Linn6. Smith and Prime (70), p. 390. Verrill (73), p. 072. Abundant but not large. In the marsh before mentioned (see under Jfodlola pllcatula) occurred a distorted variety, heavy, trun- cated, and gaping, which resembled the circurapolar Jfga truncata Linn, almost exactly, even to the tough and persistent epidermis. I think it would scarcely be distinguished from specimens of M. trim' cata lacking the epidermal tube. Perhaps adverse circumstances in both cases have produced parallel results. Doubtless the thick epidermis is a protection from the acids of the marsh. Digitized by VjOOQ IC l()0 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. SOLENIDAE. J^/isis americtma (Gould). Ensdtella a. Verrill (73), p. 674. ^Solen e?isis Linn^. Smith and Prime (70), p. 389. Tolerably common in sand or clean mud at and below extreme low water. Pandouidae. Cli(liop/i07'a (/ouldiana DM, ('. trilineata Verrill (73), p. 673. PiUiilora t. Smith and Prime (70), p. 390. Locally common in black mud, 3-() fath., with YohVut luiKitnld, Lyoxsidae. Lyonsiit hiiaUmt (Conrad). Smith and Prime (70j, p. 390. Verrill (73), p. 672. Rather uncommon. Locally on hard bottom in 3-6 fath. Tekedidae. Ttredo navalis Linn^. Verrill (73), p. 6(59. S. Not uncommon. The submerged timber is honey-combed by ship worms, but in only a few instances was the species determined. Of these all but one were of this species. Xylotnjafmbriata^Q^YGy^, Verrill (73), p. 670. S. Found in an old hull. Besides the above, fourteen species of land and fresh-water shells occurree Hat- teras. Journ. N. J. nat. hist, soc., vol. 2^ p. 76-162, pi. 2-4. Bergh, R. '71. Beitrftge zur kenntniss der mollusken des Sargassomeeres. Verhandl. zool.-bot. gesell. Wien., bd. 21, Abh., p. 1278-1308, taf. 11-13. Bergh, R. '83. Beitrttge zu einer monographie der polyceraden, 3. Verhandl. zool.- bot. gesell. Wien., bd. 33, Abh., p. 136-180, taf. 0-10. Bergh. R. '92. System der nudibranchiaten gasteropoden. Wiesbaden, 1892. Bumpiis, H. C. '98. The variations and mutations of the introduced Littorina. Zool. bull., vol. 1, p. 248-259, 14 pi. Cooke, A. H. '96. Molluscs. Cambridge natural history, vol. 3, p. 1-459. Dall, W. II. '86. Report on the Mollusca [of the '•Blake"], part 1. Brachiopoda and Pelecypoda. Bull. mus. comp. zool., vol. 12, p. 171-318, 9 pi. Dall, W. H. '89*. Report on the Mollusca [of the "Blake"], part 2. Gastropoda and Scaphop(Mla. Bull. mus. comp. zool., vol. 18, 492 pp., pi. 10-40. Dall, W. H. '89*>. A preliminary catalogue of the shell-bearing marine mollusks and brachiopods of the south-eastern coa.st of the United States, with illustra- tions of many of the species. Bull. 37, U. S. nat. mus., 221 pp., 74 pi. Davenport, C. B. '98. The fauna and flora about Coldspring Harbor, L. I. Science, n. s., vol. 8, p. 685-689. Fi.scher, H. '91. Recherches anatomitiues sur un molluscjue nudibranche appartenant au genre Corambe. Bull, scient. France et Belgique, vol. 23, p. 358-398, pi. 9-12. Fischer, Paul. '81-'87. Manuel de conchylioh)gie et de paldontologie conchyliologique. Paris, 1881-87. Gill, Theodore. '71. Arrangement of the families of mollusks. Smith.sonian misc. coll., vol. 10, art. 2, 16, 49 pp. Gould, A. A. '70. Report on the Invertebrata of Massachusetts. 2d ed. comprising the Mollusca, edited by W. G. Binney. Boston, 1870. Digitized by VjOOQ IC h\'2 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Kerbert, E. '85. [Corambe.] Tijdschr. nederl. dierk. vereen., ser. 2, dl. 1, p. 5-0, cxxxvii-cxxxviii. Lankester, E. Ray. '73. A contribution to the knowledge of haemoglobin. Proc. roy. soc, Lond., vol. 21, p. 70-81, pi. 1. Lea, H. C. '44. A description of some new species of marine shells Inhabiting the coast of the United States. Proc. Bost. soc. nat. hist., vol. 1, p. 204-206. Lea, H. C. '45. Descriptions of some new species of marine shells, inhabiting the coast of the United States. Bost. journ. nat. hist., vol. 5, p. 286-290, pi. 24. Pelseneer, P. '94. Introduction k I'Stude des moUusques. M^m. soc. malacol. Belg., vol. 27, p. 31-248. Pelseneer, P. '97. Tralt6 de zoologle, public sous la direction de Raphael Blauchard. Fasc. XVL MoUustiues, par Paul Pelseneer. Paris, 1897. Perkins, G. H. •69. The moUuscan fauna of New Haven. * * * Proc. Bost. soc. nat. hist., vol. 13, p. 109-130, 139-163. Prime, Tem])le. See Smith, Sanderson, and Prime, Temple. Smith, Sanderson, and Prime, Temple. '70. Report on tlie Mollusca of Long Island, N. Y., and of Its dependencies. Ann. lye. nat. hist. N. Y., vol. 9, p. 377-407. Stlmpson, Wm. '65. Researches upon the Hydrobilnae and allied forms. Smithsonian misc. coll., vol. 7, art. 4, 3, 69 pp. Verrill, A. E. '73. Report upon the Invertebrate animals of Vineyard Sound and adjacent waters, with an account of the physical characters of the region. Rept. U. S. comm. fish and fisheries for 1871-72, p. 295-778. Verrill, A. E. *80-'8l. Notice of recent atidltious to the marine Invertebrata of the north- eastern coast of Auierica, with descriptions of new genera and species and critical remarks on others ; part 2. Proc. U. S. nat. nuis., vol. 3, p. 356-406. Verrill, A. E. '81-'82. Catalogue of marine Mollusca added to the fauna of the New England region during the past ten years. Trans. Conn. acad. arts and sci., vol. 5. p. 447-587, pi. 42-44, 57. 58. Verrill, A. E. '84. Second catalogue of Mollusca recently atltled to the fauna of the New England coast and the a4ljacent parts of the Atlantic, consisting mostly of deei)-sea species, with notes on others previously recorded. Tran.s. Conn, aciul. arts and sol., vol. 6, p. 139-2J)4, pi. 28-32. Verrill, A. E. '85. Third catalogue of Mollusca recently fwlded, etc. Trans. Conn, acad arts and sci., vol. 6, p. 895-452, pi. 42-44. Prilled October, IS'jO. Digitized by VjOOQ IC Digitized by VjOOQ IC Balch. — Marine Mollusca. PLA.TE 1. Fig. 1. A part of the radula of Polycerella emertonii Verrill (from Bergli '83). Fig. 2. Two rows of teeth from the radula of Polycerella davenport ii. oc. 3.X oil immersion ^K. Fig. 3. A small specimen of Polycerella davenport ii creeping on a hydroid . stem. The animal drawn from a specimen of li mm. Fig. 4. Polycerella davenportii, doi-sal aspect, m. ^ mouth, o. p. = oral palps, rh. = rhinophore, d. pap. ^ dorsal pappilla, c = cardijic prominence. FroiA a si>ecimen of 3 mm. Fig. 5. Polycerella davenport ii, posterior aspect, a. ^ arms, f. — foot. Fig. 6. Polycerella davenportily enlarged outline of gills and cardiac promi- nence, g. m. ^ medial gill, g. r. =^ right gill, g. 1. = left gill, w. =^ web, r. a. - rudimentary anterior branch. Fig. 7. Polycerella davenportii, side view. From a specimen of 4 mm. Fig. 8. Shell of Assiminea mmlcs^ta lA*a. After camera sketch. X35. Fig. 9. Operculum of Assiminea niodesta Lea, drawn by W. Howe, after a camera sketch. XSo. Fig. 10. AsHiminia vKjdesta lje&. e. = eye spots, ped. = peduncle (fiL»T.ScK.:NAr.HisT. Voi^/il). H*;.-ivT'''"»"i»'i'*. ' ^''' Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. Memoirs. 4to. Tne developmebt, structure, and affinities of the genus Equisetum. By Edward C. Jeffrey. 3(5 pp., 6 plates. $1.00. Localized stages in development in plants and animals. By Robert T. Jacksdn. m pp., 10 plates. $2.00. ^ Prookbdings. 8vo. i Contributions from the Gray herbarium of Harvard university. New series, no. 17. 1. Revision of the genus Gymnolomia. 2. Supplementary notes upon Galea, Tridax, and Mikania. By B. L. Robinson and J. M. Greenman. 22 pp. 25 cts. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United States. By Garry de N. Hough. 10 pp. 10 cts. Notes on the reptiles and amphibians of Intervale, N. H. By Glover M. Allen. 13 pp. 16 cts. Variation and sexual selection in man. By Edwin Tenney Brewster. 17 pp. 25 cts. Moniloporidae, a new family of Palaeozoic corals. By Amadeus W. Grabau. 16 pp., 4 plates. * 25 cts. Studies in the gold-bearing slates of Nova Scotia. By J. Edmund Woodman. 33 pp., 3 plates. 50 cts. North American wood frogs. By Reginald Heber Howe, Jr. 6 pp. 10 cts. Some Hydroids from Puget Sound. By Gary N. Calkins. 35 pp., 6 plates. 50 cU. The Odonate genus Macrotheniis and its allies. By Philip P. Calvert. 32 pp., 2 plates. 50 cts. On the veins of the Wolffian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 25 cts. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 15 cts. The genus Antennaria in New England. By Merritt L. Femald. 13 pp. 15 cts. The land mammals of peninsular Florida and the coast region of Georgia. By Outram Bangs. 79 pp. 75 cts. A contribution to the petrography of the Boston Basin. By Theodore G. White. 40 pp., 5 plates. 65 cts. Clymene producta sp. nov. By Margaret Lewis. 5 pp., 2 plates. 15 cts. The Harvard geographical models. By W. M. Davis. 26 pp., 4 plates. 25 eta. The role of water in growth. By C. B. Davenport. 12 pp. 15 cts. Digitized by VjOOQ IC ' OCT 23 1899 Ho Proceedings of the Boston Society of Natural EUstory. ToL. 29, No. 8, p. 163-178. THE BLOOD VESSELS OF THE m:AllT IN CARCHARIAS, RAJA, AND AMIA. By G. H. Parker and Fredbrica K. Davis. With three plates. BOSTON: PRINTED FOR THE SOCIETY. October, 1899. Digitized by VjOOQ IC Digitized by VjOOQ IC OCT 28 1899 No. 8. — The Blood Vessels of the Heart i7i Carcharias^ I^qja^ and Amia} By G. H. Parker axd Frederica K. Davis, Cambridge, Mass. With three plates, Introductio7i. In mammals the blood vessels of the heart are usually disposed in the following way : A right and a left coronary artery take their origins from the base of the aorta and are distributed, in the main, to the ventricular and auricular walls of their respective sides. The blood from these arteries is collected by a set of super- ficial veins arranged in three systems : a right, a left, and a median one. The right system is represented by the right or small coro- nary vein ; the median by the middle cardiac vein ; the left by the left or great coronary vein and the posterior cardiac veins. All these veins enter the coronary sinus, which in turn enters tlie right auricle. In addition to these superficial veins, the walls of the heart contain the numerous but small deep-seated vessels of Thebesius which, according to Langer ('81), open from the ventricles and auricles into a system of fine branches that connect with the coro- nary arteries and veins by means of capillaries. In the case of the veins, but not of the arteries, these connections may be by vessels larger than capillaries, as demonstrated by Pratt ('98, p. 9*J) . The extent to which the blood vessels characteristic of the mam- malian heart occur in the lower vertebrates has never been fully ascertained. Presumably the most primitive conditions occur among fishes, and we have, therefore, undertaken the study of these vessels in three easily accessible species : Ilaja erinacea Mitchill, Carcharias littorctlis Alitchill, and Amia calva Linn. The work on Kaja anci Amia was done in the Zoological laboratory of Rad- cliffe college at Cambridge ; that on Carcharias was carried on at the laboratory of the United States fish commission at Woods Hole, Mass., and we take this opportunity of expressing our thanks to « Contributions from the Zoological laboratory of the Museum of comparative zoology at Haryard college. E. L. Mark, Director. No. 101. Digitized by VjOOQ IC 164 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. the Director of the laboratory, Prof. H. C. Bumpus, for much kind assistance in connection with this portion of the research. Coronary Arteries, The coronary arteries and the vessels with which they are directly connected reach a higher degree of complexity in the elas- mobranchs than in any other group of fishes, and consequently questions of terminology have arisen chiefly in connection with these forms. The system which we shall adopt is, in the main, an expansion of that employed by T. J. Parker ('84 and '87) in his contributions to this subject, and its components may be briefly defined in the following way : The irregular, longitudinal artery by which the ventral ends of some or all of the efferent branchial arteries of a given side are brought into communication may be called the lateral hypobranchial artery (PI. 1, fig. 1, IChni. I.). The arteries which leave the lateral hypobranchials on their median sides and, after more or less transverse courses, unite with one another in the median plane may be termed the commissural arter- ies (corns, iv-v). The longitudinal median trunk produced by the union of the commissural arteries may be designated the median hypobranchial artery (Ji'brn, ni.). From the posterior end of the median hypobranchial, the coronary arteries {cor. v.) pass off to the heart. In the skates, there are in addition postenor coronaiy arteries (PI. 1, fig, 2, cor.p.s.). These arise from a vessel which is a branch of the subclavian artery and which from its proximity to the coracoid portion of the pectoral girdle may be called the coracoid artery (cc\L). The coracoid artery, besides giving rise to the posterior coronary and certain small branches to the neigh- boring muscles, may anastomose with either the median or the lateral hypobranchial artery. The terms defined in the preceding paragraph agree in general with those used by Parker ('87), except in the case of the hypo- branchials. This author, to whom we are indebted for the name of these vessels, figin*ed and described them as branches from the subclavians. After leaving the subclavians they give rise to the posterior coronaries, and, according to him, either they retain lateral positions, as in the skate, where they extend anteriorly to connect with the ventral ends of the efferent branehials of either side, or they unite in the median plane and give rise to a single longitudi- Digitized by VjOOQ IC PARKER AND DAVIS : HEART BLOOD VESSELS IN FISHES. 165 nal vessel following the course of the ventral aorta. That the lat- eral and median vessels are not homologous, as is implied in Parker's account, is seen from the fact that in Mustelus, as Parker himself ('87, p. 697, PL 34, fig. 1) has shown, both sets of vessels may be present. We have, therefore, given them distinguishing names: median and lateral hypobranchials. Moreover, neither of these vessels can be properly considered a dependency of the sub- clavian, for the branch which leaves that artery, and which Parker regarded as their root, may be connected with them, as Hyrtl ('58, p. 17, Taf. 2) has shown, by only a relatively small vessel. The union, then, is not in the nature of a continuous trunk, but an anastomosis, and the vessel posterior to this union must be consid- ered in the light of an independent artery. This we have called the coracoid artery. In Carchxirias littoralis the lateral hypobranchial artery (PI. 1^ fig, 1, iVbrn. L) is a vessel irregular in its course but always con- nected with the efferent branchials of the second, third, fourth^ and fifth visceral arches (II-V).i It may extend to meet those of the sixth arch (VI), but when this occurs, the prolongation is usually on one side of the animal only, and the system as a whole is unsym metrical. Lateral hypobranchials, essentially similar to those in Carcharias, occur in Zygaena malltus and in Mnstelns stellatus according to the figures given by Hyrtl ('7*2, Taf. 3, fig. 2, and Taf. 2, fig. 2) and in Mustelns anUircticus as figured by Parker ('87, PI. 34, fig. 1). In these three species the vessels are figured as extending from the second to the sixth arch. Carcharias possesses two or at most three pairs of commissural arteries. The most anterior pair lies in the grooves between the second and third insertions of the ooracobranchial muscles (PJ. 1, fig. 1, cc^o hrn. 2 and 3) and parallel with the efferent arteries going to the fourth visceral arches. We have, therefore, called these vessels the commissural arteries of the fourth arch or more briefly the fourth commissural arteries (corns, iv). In a corre- sponding way, fifth commissural arteries (corns, r) can be dis- tinguished. The fourth and fifth arteries were found in all the 1 In numbering the visceral arches we liave followed the scheme laid down by Oegen- baur (".w. p. 457^, in which the first visceral arch is represented by the upper and lower jaws, the second by the hyf>ld arch, the third by the first branchial arch, the fourth by the second branchial arch. etc. Digitized by VjOOQ IC 166 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. specimens we dissected, while a sixth was less constant. When present, this last was. connected with either the anterior or the posterior efferent arteries of the sixth arch or with both. Occasion- ally, as shown in the figure, it failed to reach the vessels of any arch. The variations in this respect were often unsymmetrical. Judging from the descriptions and figures given by various authors, commissural arteries of the fourth arch occur in all sharks.* The only exception to this statement is the observation made by de Blaiuville ('11, p. 117) that in tSqualus peler/rinus the coronary arteries probably come from the efferent arteries of the posterior arch, presumably the sixth ; but as the origin of these vessels was not exactly determined, this may have been a mistaken surmise. From the evidence of previous figures and descriptions, commis- sural arteries of the fifth arch occur in Mustelus (Ilyrtl, '7*2, p. 271 ; Parker, '87, p. 697), in Zygaena (Ilyrtl, 7*2, p. 271), and in Scyl- lium (Hyrtl, '72, p. 267 ; Marsliall and Ilurst, '92, p. 242), where, however, they have been called by Hyrtl, Arteria nutriens recur- rens (branchialis). Commissural arteries of the sixth arch, such as occasionally occur in Carcharias, have been figured only in Zygaena by Ilyrtl ('72, Tab. 3, fig. 2). The median hypobranchial artery (PI. 1, fig. 1, JChrn, rn.) in Carcharias is formed on the ventral side of the ventral aorta by the union of the right and left fourth commissural arteries. This vessel has no anterior branch such as Hyrtl ('72, p. 271) has described in Zygaena under the name of Arteria thyreoidea impar, but extends entirely in a posterior direction, and, after giving off what Hyrtl ('72, p. 269) has called the epigastric branch (e'^a.), becomes the ventral coronary {cor. v.). A dorsal coronary artery (PL 2, fig. 4, cor. d.) is formed on the dorsal side of the ventral aorta by a corresponding union of the right and left fifth commis- sural arteries, supplemented by the sixth when they are present. The trunk thus formed lies so near the heart that it may be called the dorsal coronary artery {cor. d.), though it might without impro- priety be regarded as in part a median hypobranchial. Much the same condition as that found in Carcharias has been described for Zygaena (Hyrtl, '72, p. 271, Taf. 3, fig. 2) and Mus- telus (Parker, '87, PI. 34, fig. 2), except that in these fishes the iThey are found in Scylliiira according to Hyrtl ('72, p. 207) and to Marshall and Hurst (•1»2, p. 242); in Mustelus according to Hyrtl ('72, p. 271) and Parker ('87, p. 697); and in Squatina. Acantliias, and Zygaena accordinj^ to Hyrtl ('72, p. 268-269 and 271). Digitized by VjOOQ IC PARKER AND DAVIS : HEART BLOOD VESSELS IN FISHES. 167 ventral vessel is formed from the fourth and fifth commissural arteries instead of from the fourth only, and the dorsal vessel is not definitely shown to be present. In Scyllium, however, according to Hyrtl ('72, p. 267) a dorsal vessel is present; but it arises from the fourth commissural arteries which also give rise to jmired ventral vessels. In Acanthias and Squatina Hyrtl ('72, p. 269) describes only ventral vessels; these are, however, paired and only the right one extends to the ventricle. In Carcharias the epigastric artery is distributed to the muscles which surround the pericardial space. According to Ilyrtl (72, p. 269 and 271) this artery also occurs in Acanthias and Zygaena. It has been identified in Squatina (Hyrtl, '72, p. 269) and in Mus- telus (Parker, '87, p. 697), in both of which it anastomoses with what we have called the coracoid artery, thus establishing connec- tions between the subclavian and the median hypobranchial systems. The ventral coronary artery (PI. 1, fig. 1, cor, v.) in Carcharias divides into three vessels, a median, a right, and a left, and is thus distributed over the ventral surface of the ventricle. The dorsal coronary (PI. 2, fig. 4, cor. d.) also divides into three branches, one of which goes to the right side of the ventricle and to the auricle, another to the left side of the ventricle, and a third by passing around the conus to the left anastomoses with branches of the ventral coronary. The coronary arteries in sharks have not heretofore been described in detail except in the case of Selache, in which, according to the careful account given by Pavesi ('74, p. 67, 6S), the plan of distri- bution coincides almost exactly with that found in Carcharias. As in the cases of other sharks, Carcharias possesses no posterior coronary arteries. In Raja eruiacea the lateral hypobranchial arteries (PI. 1, fig. 2, h'brn. /.) show great diversity, extending in some instances over the branchial region from the second to the sixth arch, while in others they are restricted to the middle portion of this region. Any efferent branchial artery of any arch between the second and sixth may, or may not, connect with the lateral hypobranchial ; connection is, however, the rule with vessels near the middle of the series and the exception with those near the anterior and posterior extremes. So far as these connections were concerned, none of the specimens examined by us were bilaterally symmetrical. Digitized by VjOOQ IC 1G8 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. The lateral hypobranchial artery of the skate was early described by Monro (1785, p. 16, Tab. 1, figs. 4, 5), who stated that it was connected with all the efferent branchial arteries. Ilyrtl ('58, p. 16) described it in liaja clavata as coming exclusively from the vessels of the second branchial cleft, i. e., the cleft between the third and fourth visceral arches ; and Parker (*84, p. 61) figured it in Jiaja ruisiUa as connected with the efferent arteries of this cleft and the next posterior. In our experience these differences are quite as likely to be individual variations as to be of specific impoitauce. The commissural arteries in liaJa erinacta are of two kinds, dorsal and ventral, of which only the dorsal correspond to the commissurals in Carcharias. These dorsal vessels pass through the coracobranchial muscle, either in company with the afferent branchial arteries of the fourth arch (PI. 1, fig. 2, coins, d, n\), in which case they correspond to the fourth commissural arteries in Carcharias, or in company with those of the fifth arch, thus repre- senting the fifth commissurals. A vessel (coins, d, H,) whose root may possibly represent the sixth commissural is usually present, but has never been observed to be connected with the lateral hypo- branchial system. In no skate examined by us were both the fourth and fifth commissural arteries present, but notwithstanding this fact the two sets of vessels were so constant in their relation to afferent branchials that their serial homology cannot be doubted. The ventral commissural arteries in the skate (PI. 1, fig. 2, coms.v.) spring from the lateral hypobranchials and pass mediad between the coracobranchial and the coracohyoid muscles, i. e., lie ventral to the coracobranchial muscle instead of dorsal to it. They may be entirely absent and when present are usually unilateral, though a trace of a companion vessel may sometimes be present (Fig. 2). They occurred in six of the twelve specimens examined by us ; in four they were unilateral and associated with dorsal com- missural arteries on the same side ; in a fifth case the ventral artery was unilateral but unaccompanied by a dorsal vessel on the same side; and in the sixth instance the arteries were bilateral and asso- ciated with but one dorsal artery. Dorsal and ventral arteries, when both are present on a given side, unite near the ventral aorta (Fig 2). Heretofore dorsal and ventral commissural arteries have not been distinguished, but when the figures and descriptions of the earlier Digitized by VjOOQ IC PARKER AND DAVIS ; HEART BLOOD VESSELS IN FISHES. 169 authors are compared it must be admitted that both sets of arteries have been previously observed. Thus in Baja clavata, Hyrtl ('58, p. 16) described and figured the commissural arteries as lying between the coracobranchial and coracohyoid muscles ; hence they correspond to what we have called ventral commissurals ; and the same may be said of the commissural arteries figured in a species of skate by Martin (*94, Fig. 7). Dorsal commissurals have been figured by Parker ('84, p. 62, Fig. 20) in Haja nasvta and by Monro (1785, p. 16, Tab. 1, fig. 4) in the species of skate described by him. In Jiaja 7ia8uta they are the fifth commissurals and in the species figured by Monro they are double and represent commissurals of the fourth as well as of the fifth arch. The right and left commissural arteries in Maja erinacea converge on the root of the ventral aorta, where they may anastomose, as in i?q/a clavata (Hyrtl, '58, p. 16), or remain unconnected. From each commissural artery a coronary artery extends over the bulbus to be distributed to the ventricle. The left coronary (PI. 1, fig. 2, cor, 8.) is larger than the right (cor. dx.) and is distributed in the main to the ventral surface of the ventricle ; the right extends to the dorsal surface of the bulb and ventricle and to the auricle. This plan of distribution has already been observed in other species (jRaJa clavata, Hyrtl, '58, p. 16 ; Raja sp.?, Martin, '94, p. 25 ; Torpedo sp.?, Hyrtl, '58, p. 3). From the coracoid arteries in Baja erinacea posterior coronary arteries (PI. 1, fig. 2, cor.p,8.) may extend over the ventral face of the venous sinus to be distributed to the dorsal face of the ventricle. These vessels may be unilateral (Fig. 2) or paired (PI. 2, fig. 5). Similar vessels have been described and figured in Maja clavata (Hyrtl, '58, p. 17) and Raja nasuta (Parker, '84, p. 61). The coracoid arteries in Raja erinacea may give off branches which anastomose with what we have supposed may represent the sixth commissural arteries (PI. 1, fig. 2, corns, d, t'l.), a condition similar to what has already been found in other skates by Monro (1785, Tab. 1, fig. 4), Ilyitl ('58, Taf. 2), and Parker ('84, p. 62, Fig. 20) . In Amia calva no trace of lateral hypobranchials was discover- able. The arteries which supply blood to the coronaries come from the fourth visceral arches, and hence correspond to the fourth com- missural arteries (PI. 2, ^g. 3, corns, iv.). These unite in the Digitized by VjOOQ IC 170 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. median line on the ventral side of the ventral aorta and give rise to a median hypobranchial artery (h*bm,m,). This extends pos- teriorly, giving off a large epigastric branch (e^ga.), and then divides to give rise to a ring vessel surrounding the aorta and to two coronary arteries, one dorsal (PI. 2, fig. 6, cor. d.) and the other ventral (PI. 2, ^^. 3, cor, v,) . These extend posteriorly over the bulb to be distributed eventually to the respective faces of the ventricle. Each coronary shows evidence of division into right and left branches. Judging from the figures and descriptions of various authors, the coronary arteries of most teleostoraes conform to the plan in Amia.* Probably in all the higher fishes the vessels leading from the gills towards the heart are the fourth pair of commissural arteries. Stannius ('46, p. 101), however, described these vessels in the stur- geon as coming from the third branchial arch and hence corre- sponding to the fifth commissurals ; but this is probably a mistake, for, according to Ilyrtl ('55, p. 11), the vessels in the sturgeon come from the fourth visceral arches as in other higher fishes. The cod has been described by Jourdain ('07, p. 192) as receiving its coronary supply from the third branchial arch, while Parker ('84, p. 117) figures it as coming from the fifth. The vessel in the cod, however, is so small and the ventral ends of the efferent branchials are so crowded that the exact connections are rather matters of interpretation than observation. Since the sturgeon and the cod seem to be the only recorded exceptions to the general rule, and since these, as exceptions, are of doubtful value, it may well be that in all teleostomes the vessels that leave the gills for the heart come from the fourth visceral arches and represent fourth commissural arteries. Both the right and the left fourth commissural arteries are well developed in Amia, and the same is presumably true of the stur- geon (Ilyrtl, '55, p. 11) and of the pike (Mtlller, '41, p. 198). In Scomber and Pelarais the right artery is said (Hyrtl, '55, p. 11) iThe chief exception to this statement is to be found in Orthragoriscus mola, bs de8cril>e.)• The right and the left coronary veins are distributed to the whole of the ventral face and the outer edges of the dorsal face of the ventricle. The central portion of the dorsal face is cov- ered with a system of veins (Fig. 8, vn. crd. m.) which enter for the most part into a transverse trunk extending parallel to the posterior edge of the venous sinus. From this trunk small vessels pass across to the venous sinus into whose cavity they open. This system of vessels corresponds very closely to the cardiac veins of mammals. The openings of what are presumably the right and the left coro- nary veins in a skate were described and figured by Monro (1785, p. 18, Tab. 3, 37). The same was shown in Raja ritbus by Tiede- Digitized by VjOOQ IC PARKER AND DAVJS : HEART BLOOD VESSELS IN FISHES. 1 73 mann ('09, p. 9, Tab. 1, fig, 2), who also figured the openings of six small vessels belonging to the cardiac veins. In Amia calva the superficial veins of the heart (PI. 3, fig. 9) open into the venous sinus by a single orifice which lies posterior and slightly to the right of the sinu-auricular aperture (ap,) . From this opening two veins, the right (yn. cor, dx.) and the left (vn, cor. 8.) coronary veins, encircle the heart at the level of the coronary sulcus and anastomose so freely on the ventral side (Fig. 12) that a ring vessel is established. From the right coro- nary vein (Fig. 9, v?i, cor, dx,) two branches are given off, one anteriorly to the right side of the cone, and the other posteriorly to the same side of the ventricle. From the left coronary vein (vn, cor, 8,) a corresponding pair of branches is given off; that to the cone, however, is small and more ventral (Fig. 12) in posi- tion than its fellow of the opposite side. From the ventral anas- tomosis of the coronary veins a single vein extends over the median ventral surface of the ventricle (Fig. 12). This probably represents a branch from the left coronary vein. The coronary veins of higher fishes have generally escaped attention. Rose ('90, p. 35) mentions them as present in Pirnelodus catiis and Tetrodoii physa^ but absent from the eel; and Martin ('94, p. 21) states that in the salmon the right coronary vein only is present and this opens into the auricle. That there is a unilateral condition of the coronary veins as well as of the arteries in the higher fishes is not impossible. Vessels of Thebesius, The vessels of Thebesius seem heretofore never to have been sought for in the hearts of fishes. We have endeavored to ascertain whether they were present in the three species which we have studied. On inflating the left coronary vein of a fresh heart of Car- charxas littoralis by means of a blow^pipe, the auricle was gradually distended with air. As the entrance of air into the auricle through the sinu-auricular opening was carefully guarded against, such air as found its way into the heart must have come through some other aperture. If a heart whose auricle is distended with water be inflated as described above, bubbles will be seen forming on the inside of the left wall, and if the opposite wall be removed, these Digitized by VjOOQ IC 174 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. bubbles can be seen coming off freely from the inner surface of the uninjured wall. Similar experiments on the right coronary vein gave only negative results, and from neither the right nor the left vein could bubbling be produced from the inner surface of the ventricle. As this experiment can be successfully carried out with only a slight pressure of air, there is no reason to suppose that vessels were ruptured, and we believe we are justified in concluding that the left coronary veins have connections which open freely on the inner surface of the left auricular wall. These connections can be none other than the vessels of Thebesius. Attempts to blow from the inside of the auricle through to the left coronary vein always failed, doubtless because of the impediment offered by the spongy nature of the auricular wall. Experiments of a similar character canied out on the coronary arteries resulted in the production of small bubbles on the inside of the left auricular wall. This, however, was accomplished only after very vigorous blowing and consequently demonstrates that the con- nections between the coronary arteries and the veins of Thebesius are much more restricted than those between the coronary veins and these vessels, a condition already observed by Pratt ('98) in mammals. On inflating either the right or the left coronary vein of Hcya erinacea with air, bubbling could also be demonstrated from the uninjured inner surface of the auricle, but no bubbling was ever observed from the inner surface of the ventricle. If the single opening of the coronary veins in Amia calva be inflated, bubbling takes place from the inner surface of the ven- tricle as well as of the auricle. This fish was the most satisfactory of the three species for the demonstration of the vessels of Thebe- sius. These experiments, in our opinion, show that the hearts of fishes possess veins of Thebesius which open into the ventricles as well as into the auricles and which connect more freely with the coro- nary veins than with the coronary arteries. Conclusions. When the blood vessels of the heart in fishes are compared with those in mammals, the most noteworthy feature is the striking simi- Digitized by VjOOQ IC PARKER AND DAVIS : HEART BLOOD VESSELS IN FISHES. 175 laritj between the two sets of structures. Vessels of Thebesius may open into the single auricle and ventricle of a fish as they open into the paired auricles and ventricles of a mammal; and their freer communication with the coronary veins than with the coro- nary arteries in the fish recalls the condition found in mammals. Morphologically the vessels of Thebesius in fishes seem to corre- spond exactly to those in mammals. The superficial veins of the heart in fishes are also much like those in mammals. A right, a left, and a median system can be distinguished, and in Carchariae the vessels representing these are almost identical with those in the mammalian heart. In only one important respect do the superficial veins in fishes differ from those in mammals; in fishes these veins open into the venous sinus, in mammals into the right auricle. When, however, it is remembered that the venous sinus in mammals becomes incorporated in the right aui-icle this supposed difference disappears. There is then no reason for supposing that the veins of the mammalian heart are not homologous with those in the heart of the fish. The coronary arteries in fishes show less resemblance to those in mammals than has been noticed between the other classes of vessels, and this is particularly true of the way in which the coro- nary arteries originate. These arteries in mammals arise from the base of the aortic arch very near the heart ; in fishes they come from the efferent branchial arteries at places that would corresp6nd to positions well towards the dorsal side of the aortic arch of a mammal. The fact that the ventral aorta of a fish carries impure blood and the corresponding vessels in a mammal pure blood, is a suflicient physiological reason for this difference, but it leaves the question of the homology of these parts entirely open. Have the coronary arteries of mammals been derived from those of fishes or are they a new system of vessels ? The supposed absence of coro- nary arteries from the heart of amphibians has been urged in favor of the latter opinion, but Martin's ('94, p. 59-60) statement that in the tadpole a system of coronary arteries essentially like that in the fish is replaced towards the close of larval life by other coro- nary arteries is in reality almost the first piece of positive evidence bearing on this question. While this evidence is opposed to the homology of the coronary arteries of higher and lower vertebrates, it must not be forgotten that the vessels of Thebesius and the eoronary veins of the higher and lower forms show every evidence Digitized by VjOOQ IC 176 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. of being homologous, and since the arteries constitute an integral part of this system, it seems improbable that they alone should have undergone such fundamental replacements as is implied by Martin's observations on the frog. Possibly the condition in the Amphibia is to be explained on the basis of coenogenetic changes. However this may be, the whole question of the homology of the coronary arteries in higher and lower vertebrates seems to us to demand much more extensive comparative study, both anatom- ical and embryological, than has thus far been accorded to it, before a final answer can be reached. Summary, 1. Vessels of Thebesius have been found to open into the auri- cle of Carcharias, and of Raja and into the auricles and ventricles of Amia. These vessels communicate more freely with the coro- nary veins than with the coronary arteries. They are homologous with the similarly named vessels in mammals. 2. The superficial veins of the heart in Amia, and particularly in Raja and in Carcharias, are arranged in three groups corresponding to the right coronary vein, the left coronary vein, and the middle cardiac vein of mammals. These three groups of veins in fishes open into the venous sinus and thus agree in this respect with the similarly named mammalian veins which open into the right auri- cle into which the venous sinus has been incorporated. The above mentioned superficial veins of the fish's heart are homologous with those in the mammaPs heart. 3. The ventral ends of the efferent branchial arteries in fishes may be connected by a lateral hypobranchial artery. From this, commissural arteries may pass towards the median plane ; these may be either dorsal or ventral as in Raja, and the doi*sal ones may be serially arranged corresponding to the fourth, the fifth, and possibly the sixth visceral arches, as in Carcharias and Raja. The union of the right and left commissurals gives rise to, a median hypobratichial from which coronary arteries (anterior) are given off. These coronary arteries differ from those in mammals chiefly in the remoteness of their point of origin. This, however, does not necessarily preclude honiologizing them with the coronary arte- ries in mammals. Posterior coronary arteries were found only in Raja and have no homoloffues in mammals. Digitized by VjOOQ IC PARKER AND DAVIS : HEART BLOOD VESSELS IN FISHES. 177 LITERATURE. Aerassiz, L., et Vogt, C. '46. Aiiatoinie des salmones. Neuchatel, 196 pp., tab. A-0. Blainville, H. de. '11. M6moire sur le squale p61erin. Ann. mus. d'hist. nat. torn., 18, p. 88- 135, pi. 6. Cuvier, G., et Valenciennes, A. '28. Hist. nat. des poissons. Tom. 1. Paris, 14 + 424 pp., pi. 1-8. Oegenbaur, C. '98. Vergleichende anatomic der wirbelthiere. Band 1. Leipzig, 14 + 978 pp. Hyrtl, J. '55. tfber die selbststeuerung des herzens. Wien, 72 pp. Hyrtl, J. '68. Das arterielle gef ass-system der rochen. Denks. akad. wiss. Wien. Math. nat. cl. bd. 15, p. 1-36, taf. J -5. Hyrtl, J. '72. Die kopfarterien der haifische. Denks. akad. wiss. Wien Math. nat. cl. bd. 32, p. 263-275, taf. 1-3. Jourdain, S. '67. Sur la structure du cceur des poissous du genre gade. Comptes rendus acad. sci., torn. 64, p. 192-194. Langer, L. *81. Die foramina Thebesii iin herzen des menschen. Sitzber. akiwl. wiss. Wien. Matli. nat. cl. bd. 82, abt. 3, p. 25-39, 1 taf. Marshall, A. M., and Hurst, C. II. '92. A junior courst^ of practical zoology. Tliird edition. Lon Rohert E. Biukk. VViTU OXE 1*LATE. BOSTON: PRINTED FOR THE SOCIETY. April, li)00. Digitized by VjOOQ IC Digitized by VjOOQ IC APR 20 1900 No. 9. — The Occurrence of Fosaila in the Roxbury Conglomerate. By Henry T. Burr and Robert E. Burke. The age of the Boston Basin sediments has long been a matter for controversy. Two types of rock make up the mass of the sec- tion, a thick series of heavy conglomerates, and a series of compact, flinty slates. The former are known as the Roxbury conglomerates, and are usually assumed to be of the same age throughout. The slates are generally similar in character, but are now believed to represent at least two widely different horizons. The conglom- erates and much the greater portion of the slates are remarkably barren of fossils. In 1856 Paradoxides harlani Green was reported from the slates near Ilayward's Creek, Braintree, Rogers, *56, pp. 27-29, 40-41. This discovery established the age of that portion of the slates as Cambrian. As the greater part of the slates in the Basin resemble those of Braintree, the whole series was regarded as of the same age. Within recent years Lower Cambrian fossils have been found in the impure limestone at Nahant, Foerste, *89, pp. 261-263, and at Mill Cove, Weymouth, Grabau, '98, also Burr, 1900. In some parts of the Basin, at least, the conglomerates appear to underlie the slates ; hence they, too, were held to be of Cambrian age. The conglomerates are largely made up of fragments which appear to have been derived from the complex of granitic rocks to the south, Crosby, '89, p. 6. The granite, then, is older than the conglomerate. On the supposition that the conglomerate is below the slate, it is necessary to regard the granite as older than the slate, also. But the granite, as is now known, is intrusive into the slates of Braintree, Wadsworth, '83, p. 27 ; also Crosby, '89, p. 5, and is, ' therefore, later than that portion of the Cambrian series. The conglomerates, then, not only overlie the Middle Cambrian slates, but are separated from them by a great period of igneous action, and an interval long enough to allow the forces of erosion to penetrate deep into the granite mass. It is, therefore, no longer necessary to regard the conglomerates and associated slates as of Cambrian age. In the absence of positive evidence to the contrary, however, they are still held to be Cambrian by many observers, and are so mapped by Walcott in his correlation papers, Walcott, '91, p. 268, and map, p. 358. Digitized by VjOOQ IC 180 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Of late years the opinion has been gaining ground that the con- glomerates, at least, should be referred to the Carboniferous. This belief is based largely upon the strong lithological resemblance between the Boston Basin series and the Carboniferous conglom- erates of the neighboring Norfolk County and Narragansett Basins. The argument from analogy is not strictly valid, for it is quite pos- sible that very similar conglomerates might be formed at widely different horizons, particularly where the sources of supply remain the same. In fact, as has been pointed out, our glacial material, if worked over by the sea, would yield a deposit essentially like the Roxbury conglomerate. Until recently the idea of the Carboniferous age of these sedi- ments has been practically unsupported by fossil evidence. Some years ago Mr. J. B. Woodworth found, in the conglomerate near Franklin Park, a fragment which he regarded as a portion of a fossil plant, similar to forms occurring in the sandstones of the Narragan- sett Basin. Not then appreciating the importance of his discovery, he did not preserve the specimen. Other observers have reported the discovery of fossils in the conglomerates and associated slates, but no identifiable forms have ever been produced. During the past year the writers have made a careful search over a considerable portion of the area. Traces of possibly organic remains have been found in the slates and in the quartzite pebbles of the conglomerate, but none of these have had any determinative value. Recently, however, a sandy zone near the top of the con- glomerates has yielded fossils of a much more satisfactory nature. These are believed to be casts and moulds of the trunks or roots of tree-like forms. They are cylindrical m form, with circular cross- sections. The largest has a maximum diameter of four and eight tenths inches. They are marked by somewhat irregular transverse wrinklings (see Plate 1), which sometimes pass entirely aiound the form, sometimes die out, or become united. The organic matter has entirely disappeared. The casts are composed of compact sand- stone, which is indistinguishable from that of the rock in which they lie. The bedding of the sandstone is obscure, but is believed to lie. at right angles with the axes of the fossils. The species to which these forms should be referred cannot at present be determined. The genus, even, is highly problematical. The only markings which are at all characteristic are the transverse wrinklings referred to above. Quite similar wrinklings appear Digitized by VjOOQ IC BURR AND BURKE : FOSSILS IN ROXBURY CONGLOMERATE, 181 upon several species of the genus Artisia^ particularly upon Artisia distans, figured by Grand *Eury in Flore Carbonifer^ de la Loire. It is quite possible that the specimens figured are related to that form. It cannot, however, be denied that such vague markings have little determinative value and may even be of mechanical origin. While the writers are disposed to believe that these forms cannot be identified with certainty, they nevertheless feel confident that they are true fossils. The several geologists who have passed judg- ment upon them, have, with one exception, expressed themselves as satisfied of their organic origin. The other view is that they are due to mechanical action, — that they are akin to stylolites. It is difficult to see upon what this opinion is based. The speci- mens figured certainly bear little resemblance to the ordinary forms of stylolites. As generally defined, stylolites are forms produced in rocks by displacement or the development of slickensides about a portion protected by a shell or other hard capping. Such forms have characteristically slickensided surfaces, usually with strongly marked longitudinal stHations, and frequently with the development of secondary minerals. The forms under consideration show no trace of longitudinal striations. The surfaces are not smooth after the manner of rubbed surfaces. There is, so far as can be seen, no development of new minerals. In short, the phenomena of slicken- siding are altogether absent. Stylolites are of small size, seldom exceeding four inches in length or two in diameter. The specimens figured average about four inches in diameter, and the largest is over a foot in length, with the total length not known. There is no reason why stylolites should have a circular cross-section. It would be strange indeed if, as in this case, all the specimens found in a limited area should have this form. There seems then to be no reason whatever for thinking that the surfaces of these specimens were developed by differential movement. It should be added that stylolites usually occur in limestones or in fine-grained shales, and have never, so far as is known, l)een reported from coarse sandstones. It may be suggested that these forms are concretionary. Cylin- drical or rod-like concretions are not unknown, and, so far as the form is concerned, these specimens might be of such nature. If the forms are concretionary, the material of which they are made up should differ, in a determinable way, from the material of the matrix. This does not appear to be the case. In both cast and Digitized by VjOOQ IC 182 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. matrix the material is a compact quartzose sandstone, without trace of lime and with but enough iron to give it a reddish tinge. More- over, although similar sandstone is abundantly developed in portions of the conglomerate area, it nowhere shows signs of concretionary action. No other action comes to mind as capable of producing similar forms, unless it is assumed that they may be due to jointing. But MAP Showing tocd^lity of FOSSILS. SCALE SOO'-l". it seems extremely improbable that jointing should produce such regular forms. Moreover, the surfaces lack all the characteristic features of joints, Woodworth, '96, pp. 163-183. All things considered, the forms are best explained as casts and moulds, and may be fairly assumed to be of organic origin. Granting this, it is still true that these fossils do not definitely Digitized by VjOOQ IC BURR AND BURKE : FOSSILS IN ROXBURY CONGLOMERATE. 183 determine the age of the conglomerate. They might have come from the Devonian or the Trias as well as from the Carboniferous. But they surely do settle the question as between Cambrian and Carboniferous. And this is and has been the point at issue. The fossils were found in Forest Hills Cemetery, on the south- em edge of the wide belt of conglomerate extending through Brookline, Roxbury and Dorchester. The field relations show that this conglomerate mass is, structurally, a broad, flat-topped anticline. The fossil horizon is, therefore, at the top of the series. The slates surround this anticline, and everywhere dip and strike conformably with the adjacent beds of the conglomerate series. It is a safe assumption that they overlie the conglomerates and are conformable with them. The accompanying section shows the FOSSIL Conglomer^e with S*>nd»ton« bands in upper portfons. /HORIZON. •n««NI.U«S. JAMAICA ^LAIM. POReST HtkLl GENERAUZED SECTION ACROSS MAIN CONGLOMERATE MASS. structural relation of the fossil horizon to the conglomerates and the overlying slates. The evidence from these fossils seems appli- cable to the whole of this conformable series. It has not, as yet, been definitely proved that the other conglom- erates of the region are of the same age as those of the central belt, although such is generally assumed to be the case. They are much alike lithologically, show the same degree of secondary alteration, and have the same apparent relations with the igneous rocks. The outcrops of slate are so scattered that it is not possible to correlate them in a satisfactory manner. It seems probable, however, that the greater part of the slate is closely associated with the conglom- erate, and of approximately the same age. It is believed, then, that the discovery of these forms serves to establish the fact that Carboniferous sediments are present in the Boston basin, and to render it probable that the greater part of the area is occupied by sediments of this age. Digitized by VjOOQ IC 184 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. LITERATURE. Burr, H. T. 1900. A new Lower CanibriaD fauna from Eastern Massachusetts. Amer. geol., vol. 26, pp. 41-60. Crosby, W. O. '89. Physical history of the Boston basin. Boston, 1889, pp. 6, 6. Foerste, A. F. *89. Palaeontological horizon of the limestone at Nahant. Proc. Boston soc. nat. hist., vol. 24, pp. 261-263. Gtabau, A. W. *98. Guide to localities illustrating the geology, marine zoology and botany of the vicinity of Boston. Boston, 1898. (Refers to discovery by Prof. W. O. Crosby.) Rogers, W. B. , *56. Notes on Paradoxides from Braintree. Proc. Bosttm soc. nat. hist., vol. 6, pp. 27-29, 40-44. Wadsworth, M. E. '83. On the relation of the Quincy granite to the Primordial argillite of Braintree, Massachusetts. Proc. Boston soc. nat. hist, vol. 21, pp. 274-277. Walcott, C. D. '91. Correlation papers. — Cambrian. Bull. 81, U. S. geol. surv., p. 268, map, p. 858. Wood worth, J. B. '96. On the fracture system of joints. Proc. Boston soc. nat. hist., vol. 27, pp. 163-183. Printed, April, 1900. Digitized by VjOOQ IC ,Q ,0 en c h X c CQ Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. Memoirs. 4t(>. The development, structure, and affinities of the genus EquLsetum. By Edward C. Jeffrey, m pp., 5 plates. $1.00. Localized stages in development in plants and animals. By Robert T. Jackson. 05 pp., 10 plates. $2.00. Proceedings. 8vo. Tlie blood vessels of the heart in Carcliaria-s, Raja, and Amia. By G. H. Parker and F. K. Davis. 10 pp., 3 plates. 25 cts. List of marine mollusca of Coldspring Harbor, Long Island, with descriptions of one new genus and two new species of nudibranchs. By Francis Noyes Balch. 30 pp., 1 plate. 35 cts. The development of Penilia schmackeri Richard. By Menin T. Sudler. 23 pp., 3 plates. 30 cts. Contributions from the (iray herbarium of Harvard university. New series, no. 17. I. Revision of the genus Gymnolomia. 2. Supplementary notes upon Calea, Tridax, and Mikania. By B. L. Robinson and J. M. (ireenman. 22 pp. 25 cts. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United Slates. By Garry de N. Hough. 10 pp. 10 ct»s. Notes on the reptiles and amphibians of Intervale, N. H. By Glover M. Allen. 13 pp. 15 cts. Variation and sexual selection in man. By Edwin Tenney Brewster. 17 pp. 25 ct.s. Moniloporidae, a new family of Palaeozoic corals. By Amadeus W. Grabau. 10 pp., 4 plates. 25 cts. Studies in the gold-bearing slates of Nova Scotia.* By J. Edmund Woodman. 33 pp., 3 plates. 50 cts. North American wood frogs. By Reginald Heber Howe, Jr. 6 pp. 10 cts. Some Hydroids from Puget Sound. By Gary N. Calkins. 35 pp., 6 plates. 50 cts. The Odonate genus Macrothemis and itii allies. By Philip P. Calvert. 32 pp., 2 plates. 50 ct^. On the veins of the Wolffian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 25 cts. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 15 cts. The genus Autennaria in New England. By Merritt L. Femald. 13 pp. 15 cts. The land mammals of peninsular Florida and the. coast region of Georgia. By Outram Bangs. 71>pp. 75 cts. A contribution to the petrography of the Boston Bjisin. By Theodore G. White. 40 pp., 5 plates. 05 cts. Clymene producta sp. nov. By Margaret Lewis 5 pp., 2 plates. 15 cts. The Harvard geographical models. By W. M. Davis. 20 pp., 4 plates. 25 cts. The role of water in growth. By C. B. Davenport. 12 pp. 15 cts. Digitized by VjOOQ IC APR 20 1900 Prooeedings of the Boaton Society of Natural History. Vol. 29, No. 11, pp. 217-222. A REVISION OF THE SYSTEMATIC NAMES EMPLOYED BY WRITERS ON THE MORPHOLOGY OF THE ACMAEIDAE. B^ M. A. Wii,Lr«»x, Ph.D. BOSTON: PRINTED FOR THE SOCIETY. - Ai'itiL, UK)0. Digitized by VjOOQIC Digitized by VjOOQ IC k^^'^tm No. 11. — A Jievision of the Systematic Names employed by Writers on the Morphology of the Acmaeidae. By M. a. Willcox, Ph. D., Professor of Zoology in Wellesley College, Wellesley, Mass. In view of the small extent of our knowledge of the Acroaeidae, it would seem especially desirable that what is known should be rendered available by the employment of uniform terminology. This is, however, no less an unattained ideal in this small group than in other parts of the animal kingdom. It is the purpose of this brief paper to so arrange and compare the systematic names employed by various authors as to enable the reader to orientate himself with the least possible delay. From 1758 down to the early part of this century all true limpets appear to have been included under the generic name of Patella, About 1830 several investigators independently separated off from the remaining members of this genus a group differentiated by the possession of a cervical gill, or ctenidium. Eschscholtz, '30,^ called the new genus Acmaea\ Audouin and Milne Edwards (Cuvier, '30, p. 326) named it Tect%ire\ Gray, '33, p. 800, termed it Lottia. This name was obviously of later date than the others, and Gray himself, '47, p. 158, abandoned it in favor of that introduced by Audouin and Milne Edwards, which he seems to have Latinized into Tectura,^ As to the use of Acmaea or Tecturay scientists are divided. The French hold to the latter, while the rest of the zoological world has agreed upon Acmaea, I have not found in 1 This reference, which I have been unable to verify, I take from Watson, '86, p. 29. DaU, '71, p. 237, quotes the same work with the date of Dorpat, 1828. As he gives the name in its English form, as in a later paper, *78, p. 342, he states that the ** English reprint," which was published in the spring of 1830, was dated by the author Dorpat, Jan. 7, 1828, and, Anally, as Watson states that he has made unavaUing search for any publication to which DalVs reference could refer, I am led to the belief that Eschscholtz's description did hot issue from the press before 1830. This, however, would not aifect the question of priority, since, as Watson points out, the Ann. des sci. nat., t. 21, con- taining as it does reports of meetings held as late as Dec. 13, 1830, could not have issued from the press before 1831. > Dall, 71, p. 239, states that he has failed to ftnd any earlier publication of the Latin form. I have not been more fortunate. Marschairs " Nomenclator " lists Teciura as published in 1830 by Audouin and Milne Edwards; but this, so far as the form of the word goes, is an inaccuracy, due undoubtedly to Marschairs habit of Latinizing names. Digitized by VjOOQ IC 218 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. French writers any attempt at a justification of this practice, which would seem to have arisen from the fact that Tectura, being pro- posed — at least in its French fonn — at about the same time as Acf/i/xea, secured a position in the land of its authors which has never been successfully disputed, although its rival has undoubted right of priority. The arguments have been fully stated by Dall and Watson. Bouvier, '87, p. 22, offers one argument unmentioned in either of these statements — the great similarity between the names Acmaea and Actnea, The latter was proposed by Ilartmann in 1821, but, according to Watson (op. cit.), was abandoned by him the same year. I have l>een unable to consult Hartmann's paper; but I am informed on excellent authority that his Acmea is derived from ^aKfiT^; Acmaea, on the other hand, comes from ^atcfialoi (Rathke, '33, p. 16), and should therefore stand. Some twenty years laj-er this group of ctenidium-bearing limpets was itself subdivided. In 1847 Gray, '47, p. 158, apparently with some hesitation, separated from the others those which have both a ctenidium and a branchial cordon, giving to this new genus the name of Scurria, and restricting the name of Ttctura (z=Lott%a) to those which lack the brancliial cordon. The name of Lottia Gray was thus abandoned to be revived after nearly twenty years. At this time Carpenter, '65, pp. 140-141, erected a new genus for a limpet provided with both ctenidium and branchial cordon, but having the latter absent in the region of the head. As it appeared that this very animal had been figured by Sowerby, '20-'25, vol. 1, pi. 141, as the first mentioned example, and therefore inferentially as the type of Lottia Gray, for which Gray himself had mentioned no type species, the name given to the new genus was of course Lottia. So that Lottia (xray — or more correctly, Lottia Cpr. ex Qray — represents but a small part of the original I^ottia Gray. It should be said also that some at least of the more recent authori- ties have reckoned Lottia as a subgenus of Scnrrla, so that the systematic rank of tlie group is a matter upon which opinions vary. And finally it may be remarked that Carpenter, '60, p. 3, had already listed this species, though without a description, under the name of Tecturella (jrandis. This name fell, owing to preoccupation. In 1834 Broderip described a new limpe^like shell, giving no account of the animal, and entitled it Scutella, As this name had already been employed by Lamarck, Gray, '47, p. 168, replaced it Digitized by VjOOQ IC WILLCOX: SYNONYMY OF THE ACMAEmAE. 219 by ScuteUina, Gray placed the new genus among the Patellidae ; but H. and A. Adams, '58, vol. 1, p. 460, included it among the Tecturidae. This appears to have been done on^ the strength of their own investigation ; at least they give the earliest figure which I have been able to find of the animal as distinct from the shell, and mention, although they do not figure, the gill. Scutellina retained its position in the Acraaeidae (^ Tecturidae) either as a genus or as a subgenus (of Acmaea) until Dall, '89, p. 410, after a careful study of the animal, transferred the genus from the Doco- glossa to the Rhipidoglossa, — a change which has been approved by Tryon and by Simroth. It is interesting to note that one of the points which decided D^l to make this transfer — the fact that the apex of the shell is posterior instead of anterior — was noted by Broderip in his original description. Although Scutellina has thus been removed by most authorities from the Acmaeidae, it is still occasionally included in that group. It may, therefore, be worth while to call attention to a brief paper by Pilsbry, '91, p. 88, in which he points out that /Scutellina, having been employed in 1841 by Agassiz for a genus of echinoderms, is preoccupied, and must therefore fall as the name of a molluscan genus. He suggests as a substitute Phenacolepas, The Acmaeidae then include the following groups : — Acmaea Eschscholtz — cervical gill, but no branchial cordon. Scurria Gray — cervical gill and continuous branchial cordon . Lottia Cpr. ex Gray — cervical gill and interrupted branchial cordon. If now we examine works dealing with the moi*phology of the family, we find these names often incorrectly used. I enumerate all the papers known to me which treat, otherwise than incidentally, the anatomy of any one of the Acmaeidae, noting errors of nomen- clature where I have found them. 1. Rathke, '33, gives a somewhat full description of the anatomy of Acmaea. 2. Bouvier, *87, p. 22, gives a brief account of the nervous sys- tem of Tectura (= Acmaea) testtidinalia. 3. Bernard, *90, pp. ^\l~^'lfi^ deals with two species of Tectura (= Acmaea) which are thus described (p. 217) : " C'est d'abord la Tectura (Acmaea) pileopsis, qui diff^re des Patelles par la pre- sence d'une branchie bipectin^e et la Tectura fontainesi, qui poss^de a la fois une branchie bipectin^e et des lamelles branchiales circumpall^ales." Digitized by VjOOQ IC 220 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. If we are to infer from this description that T. pileopsis differs from PateUa^ not only in the presence of a cervical gill, but also in the absence of a branchial cordon, it may possibly be identified with Acmaea pileopsis Q. and G., although this species is recorded by Tryon as belonging to the New Zealand, Indo-Pacific and Austra- lian region, while Bernard (p. 217) states that his species came from Chili. Tectura fontainesi, however, is obviously incorrectly named, and i s very probably a JScurria. I have not been able to find trace of any member of the family bearing this specific name. Bernard has also investigated Lottia peUuciday which he describes in the following terms (p. 225) : " Les Lottia ou Patina sont de petits Patellid^s d^pourvua de branchies proprement dites, mais munis de lamelles pall^ales et de tentacules lat^raux." This is obviously Patina pdlucida Linn. I have found no justification for applying the name of Lottia to this genus. 4. Von Erlanger, '92, p. 604, describes the nephridia of an unde- termined species of Tectura (^ Acmaea) . 5. Haller, '94, describes four Acmaeidae : Scurria, two species of Lottia Gray, and Scutellina, If Dallas classification of Scutelr Una be accepted, it is probable, as has been suggested by Thiele, that we have here to do with a case of incon*ect determination. The description (pp. 26-27) shows clearly that the animal (sup- posing it to belong to a described genus of the Docoglossa) is an Acmaea. The two species of Lottia Gray- are also Acmaeas ; one lacks the branchial cordon altogether ; the other has irregular outgrowths which Haller (p. 27) regards as the anlagen of the cyclobranch gills. They probably correspond with A, viridtUa Lam. and A. scutum Orb. 6. Pelseneer, '91, p. 61, describes the eye of Acmaea testu- dinalis. 7. Thiele, '92, p. 231, describes the structure of the mantle-edge in CoUisella (Acmaea) digitalis, CoUisella is one of the sub- genera of Acmaea, and according to both German and American rules the subgeneric name, when required, should be placed in parenthesis and interpolated between generic and specific appella- tions. The name of Thiele's species should therefore read Acmaea ( CoUisella) digitalis. 8. Willcox, '98, treats the general anatomy of Acmaea fragilis. Digitized by VjOOQ IC WILLCOX: SYNONYMY OF THE ACMAEIDAE. 221 LITERATURE. Adams, H. and A. '58. The genera of recent moUusca. 3 vols. London. Bernard, F^liz. '90. Recherches sur les organes pall^ux des gast^opodes prosobranches. Ann. des sci. nat., (7) t. 9, pp. 89-404, 10 pi. Bouvier, E. L. '87. Systtoie nerveux morphologie et classification des gast^ropodes pro- sobranches. Ann. des sci. nat., (7) t. 3, pp. 1-510, 19 pi. Broderip, W. J. '34. Description of a new genus of Gasteropoda. Proc. zool. soc., pt. 2, pp. 47-49. Carpenter, P. P. '60. Checklists of the shells of North America. Smithsonian misc. coll. '65. Diagnoses de mollusques nouveaux proyenant de Califomie et faisant partie du mus6e de I'institution Smithsonienne. Journal de conchylio- logic, t. 13, pp. 129-149. Cuvier, Georges. '30. Rapport sur trois mtooires de MM. Victor Audouin et Milne- Edwards, relatifs aux animaux sans vert^bres des c6tes de la France. Ann. des sci. nat., t. 21, pp. 317-829. Dall, W. H. '71. On the limpets ; with special reference to the species of the west coast of America and to a more natural classification of the group. Amer. journ.^of conch., vol. 6, pp. 227-282, 4 pi. '78. Report on the limpets and chitons of the Alaskan and Arctic regions, with descriptions of genera and species believed to be new. Proc. U. S. nat. mus., vol. 1, pp. 281-344. *89. Report of the mollusca of the Blake. II : Gastropoda and Scaphopoda. Bull. mus. comp. zool., vol. 18. Erlanger, R. von. '92. On the paired nephridia of prosobranchs. Quart, journ. micr. sci., vol. 33, pp. 687-623, 2 pi. Eschscholtz, F. '30. Appendix Kotzebue's Neue Reise. Weimar. Oray, J. E. '33. Some observations on the economy ot molluscous animals and on the structure of their shells. Phil, trans, roy. soc. of London, vol. 123, pp. 771-819. '47. A list of the genera of recent mollusca, their synonyma, and types. Proc. zool. soc., pt. 16, pp. 129-219. Digitized by VjOOQ IC 222 PROCEEDDfGB: BOSTON SOCIETY NATURAL HISTOBY. Haller, B^la. '94, Studien tt. dooogloase u. rhlpidoglosse Prosobranchier. Pp. 173, 11 pi. Leipzig. Pelseneer, P. '91. Sur I'oeil de quelques gastropodes. Ann. de la soc. beige de micro- scople, 1. 16, pp. 69-76. Pilsbry, H. A. '91. On the use of the generic name Scutellina. NaatilnB, yoI. 5, pp. 88-89. Rathke, M. H. '33. Eschscholtz's zoologischer Atlas. Ftinftes Heft, herauBgegeben yon D. [ate] Martin Heinrich Rathke. Berlin. Sowerby, James and George B. '20-'25, The genera of recent and fossil shells. 2 vols. London. Thlele, J. '92. BeitrMge zur Kenntnis d. Mollusken. 2, U. d. MoUoskenschale. Zeitr schr. wiss. zool., Bd. 55, pp. 220-251. Watson, Robert Bogg. '86. Report on the Sci4>hopoda and Gasteropoda collected by H. M. S. Chal- lenger during the years 1873-1876. Challenger reports, vol. 16. Pp. 766, 60 + 3 pi. Willcox, M. A. '90. Zur Anatomie yon Acmaea fragilis Chemnitz. Jen. Zeitschr., Bd. 32, pp. 411-466, 3 pi. PrirUed AprUy 1900. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC '^'• Boston Society of Natural History. RECENT PUBLICATIONS. ' Memoirs. 4to. The development, .structure, and affinities of tlie genus Etjtiisetuni. By Edward C. Jeffrey. 36 pp., 5 plates. $1.00. Localized stages iu development in plants and animals. By Robert T. Jackson. • 66 pp., 10 plates. $2.00. Frock EDiNOS. 8vo. Tlie blood ve.s.sels of the heart in Carcharias, Raja, and Amia. By G. H. Parker and F. K. Davis. 16 pp., 3 plaWs. 26 cts. List of marine mollusca of Coldspriug Harbor, Long Island, with descriptions of t)ne new genus and two new species of nudibranchs. By Francis Noyes Balch. 30 pp., 1 plate. 35 cts. The development of Penilia schmackeri Richard. By Mervin T. Sudler. 23 pp., 3 plates. 30 cts. (Contributions from the Gray herbarium of Harvard university. New series, no. 17. 1. Revision of the genus (Jymnolomia. 2. Supplementary notes upon Calea, Tridax, and Mikania, By B. L. Robinson and J. M. (ireenman. 22 pp. 26 cts. Studies iu Diptera Cyclorhajiha. 1. The Pipunculidae of the United States. By Garry de N. Hough. 10 pp. 10 cts. Notes on the reptiles and amphibians of Intervale, N. II. By Glover M. Allen. 13 pp. 16 cts. Variation and sexual st»lection in man. By Edwin Tenney Brewster. 17 pp. 25 cts. Moniloporidae, a new family of Palaeozoic corals. By Anuvdeus W. Grabaii. 10 pp., 4 plates. 25 cts. Studies in the gold-bearing slates of Nova Scotia. By J. Edmund Woodnian. 33 pp., 3 plates. 50 cts. North American wood frogs. By Reginald Heber Howe, Jr. 6 pp. 10 cts. Some Ilydroids from Puget Sound. By Gary N Calkins. 36 pp., 6 plates. 50 cts. The Odonate genus Macrothemis and its allies. By Philip P. Calvert. 32 pp., 2 plates. 50 cts. ( )n the veins of the Wolffian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 25 cts. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 15 cts. The genus Antennaria in New England. By Merritt L. Fernald. 13 pp. 15 cts. The land mammals of peninsular Florida and the coast region of Georgia. By Outram Bangs. 70 pp. 75 cts. A contribution to the petrography of the Boston Basin. By Theodore G. White. 40 pp., 5 plates. 05 c\s. Clymene products sp. nov. By Margaret Lewis 5 pp., 2 plates. 15 cts. The Harvard geographical models. By W. M. Davis. 20 pp., 4 plates. 25 eta. The role of water in growth. By C. B. Davenport. 12 pp. 15 cts. Digitized by VjOOQ IC JUN 4 1900 Prooeedlngs of the Boston Society of Natiiral History. Vol. 29, No. 12, pp. 223-240. PROCEEDINGS OF THE ANNUAL MEETING, MAY 2, 1000. BOSTON: PHIN TED FOR THE SOCIETY. May, 1000. Digitized by VjOOQ IC Digitized'by VjOOQ IC JUN 4 1900 No. 12. — Proceedings of the Annual Meeting, May 2, 1900. REPORT OF THE CURATOR, ALPHEUS HYATT. The most notable event of the past year was the retirement of the Secretary, Mr. Samuel Henshaw, who resigned in order to accept a position in the Museum of Comparative Zoology at Cambridge. His resignation was noticed by a formal resolution, but naturally this did not allude to his long and efficient services during the time that he was an assistant in our Museum. This gentleman's name appears first in the Annual Report of 1876-77 when he took charge of our insects, and it occupies thereafter an annually increasing importance in the records of the Museum, until he became general assistant in 1883. A large part of the time from 1876 to 1883 he had worked for us without pay or with a merely nominal salary, and during those years had not only accomplished much for the insects which were under his charge, but had laid the Society under obligations for important work upon most of the other collections in the Museum. From 1883 to his election as Secretary in May, 1892, Mr. Henshaw fulfilled the duties of general assistant with excep- tional ability. His active connection with the Museum consequently has lasted for about twenty-three years, and his labor has been felt in every department and always greatly to the advantage of the Society. The Curator is, therefore, very glad to be able to say that, although his work now lies almost wholly in Cambridge, he still remains connected with our Museum. He has completed the admirable cycle of his life with us by consenting to remain in charge of the insects as a voluntary assistant, thus returning to the position with which he began so many years ago. The Society has long been in need of some separate room where the meetings of the Council could be held, and this year, princi- pally through the efforts of the President, the northwest basement room was fitted up for this j)urj>ose and also furnished with a large blackboard and settees, so that it can be used for meetings of sec- tions of the Society, if any are formed, or by natural history clubs or small societies that may find it convenient to meet in our build- ing. Improvements usually have accomj)anying inconveniences, and this one obliged the Curator to crowd the collections heretofore stored in two rooms in the cellar into one room and to make other changes that are not yet entirely completed. Digitized by VjOOQ IC 224 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. The urgent need of means iiith which to do work has been insisted on in these reports year after year ever since the first year of the Curator's connection \i4th the Society, and yet in spite of this, and often repeated appeals in more public ways, the impression is still prevalent in Boston that we are not in need of money. Another impression that needs correction is still more unfavorable to our progress. This is the prevailing opinion that we are not an educational institution. In spite of all the lecturing to teachers and to students and to the public that has taken place systematically and constantly in this building for thirty years past, and in defiance of moi'e or less frequent newspaper notices of the kind of work done here, people at large regard us as a Society whose functions, out- side of a museum that is open to the public for two days in the week, are strictly private and for the benefit of members alone. These false impressions lead wealthy people who are continually giving in other directions to neglect us and even to say, if requested to give, that a Society ought to take care of itself. A few years ago I heard one of the most prominent Boston merchants and lib- eral givers to the cause of education say the same thing about the Institute of Technology. His complete conversion to the o]i])osite side within a few years leads me to hope that a brighter future may be in store for us. Certainly nothing can be more without justifica- tion than that an institution like ours, devotee willing to tax themselves to the amount of our annual dues; but when only a very minute i)roportion of the citizens is willing to do this, it is practically absurd to expect us to maintain ourselves by any such means. At present the money for membei-ship is all used uj), and in fact is not sufticient to meet our annual ai)])ropriation for publications alone. In view of this general neglect of our needs it is pleasant to notice that this year we have received a donation of ten thousand dollars that the Rev. R. C. Watei-ston left by his will, to be paid after the decease of Mrs. Waterston. Mr. Waterston was elected a member in January, 18G0, and his death occurred Feb. 21, 1898. For about twenty years he was a helpful member of this Society, and he was selected to give one of the addresses at the celebration of our fif- tieth anniversary in 18S0. During the last ten years or so of his life, his increasing disabilities kept him from very active participa- Digitized by VjOOQ IC HYATT : REPORT OF THE CURATOR. 225 tion in the affairs of the Society, but he was always ready to assist in every way that he could, and his last generous action shows an appreciation which is very grateful, since he was thoroughly con- versant with what had been done and with the future aims of our work. The Secretary's Report gives the special puq)ose of this donation, and it is only necessary for the Curator to add that Mr. Waterston also left his collection under such conditions that we were able to take out of it whatever was considered suitable for our uses. It was a miscellaneous collection, but we found in it a con- siderable number of natural history specimens that could be used in the laboratory and a few for our Museum. The remainder of this report is given under special titles designating the different departments of the Museum in which work has been done. Mineralogy and Geology. Professor Crosby has continued the work on the general collec- tion of minerals which was interrupted last spring by poor health. This consists chiefly in weeding out duplicate specimens, which is in itself a great improvement of the collection, besides making room for new material. He has gone over the entire collection in this way, and will use a part of this material in exchanges. A large amount of work has been done in preparation of Part 3 of the Geology of the Boston Basin, which, it is expected, will soon be ready for distiibution. Teaching in the Museum. A lady of Boston, as stated in a former report, carried on this department for a number of yeai*s, and Mr. Grabau's work as the lecturer and guide to the collections and in other connected lines of public instruction was fruitful in results that justified the hope of making this undertaking a permanency. It was a really unique and successful effort to make collections effective instru- ments of instruction in j)lace of the lifeless and comparatively inefficient assemblage of facts they now are in museums, and would have met with substantial aid if the good work that was done could have been made known to the proper persons. The Curator there- fore, feels that apj^eals for the re-establishment of this department should not be dropi)ed. Digitized by VjOOQ IC 226 PROCEEDINGS: BOSTON SOCIETY NATURAL fflSTORY. Dynamical Zoology. If it were not for the great personal interest taken in thi» depart- ment by the Curator, the difficulties attending its installation would have long ago led to its abandonment. Considerable work has been done in this direction by the Curator and Miss Bryant, but as usual the results appear to be small, owing to the })eculiar difficul- ties that have to be encountered in selecting specimens that will show the relations of organisms to their surroundings. Synoptic Zoology. The gratuitous work of Mrs. Sheldon in this department, as noted in previous Reports, still continues, and this and other aid received from her makes it ponsible to anticii)ate the final completion of this unique collection and the text of the Guide. The amount of good work put into this undertaking will then be in such a form that it can be understood and properly appreciated. The following sum- mary gives only the bare facts of what has been done, but does not convey any idea of the amount of work involved in the study of the literature and the careful judgment and investigation needed for the selection of the dra^^nngs and specimens mentioned. The prin- cipal work of the year has been uj)on the Crustacea, Arachnozoa, and Myriapoda. The ty})e8 of all of these have been described, eighty for Crustacea, twenty-two for Arachnozoa, and eight for nuTiapods. The figures selected for illustrations of structure and develo])ment are as follows: forty-five for Crustacea, thirty-eight for Arachnozoa, and five for M\niaj)oda. Eighteen pages of the text on Arachnozoa have been written. A large amount of work has also been done upon the Insecta, in comparing different systems of classification, by the same assistant, who has also selected and described thirty types of this class and has picked out twenty-^ight figures to illustrate the fossils, the i)i'imitive fonns, and early stages of development ; and she has also written fifty-one pages of the text of the Guide for this class. Forty-eight figures of various grou})s, coelente rates, ecliinoderms, i)elecy]>ods, cephalopods, and pteropods, have been completed, and the text of the Guide relating to these has been revised. A beginning has also been made upon theVertebrata, four ty})es and ten figures of primitive forms having been selected and described, and some pages of the text written. Miss Maitin has spent considerable time in making colored draw- inijs for this collection, under the direction of 3Irs. Sheldon. Digitized by VjOOQ IC HYATT: REPORT OF THE CURATOR. 227 Botany. Fortunately the sickness of Miss Carter as mentioned in my last Annual Report did not prevent her return to duty during the early part of 1900; and since then this lady has worked in the Museum and reporte as follows. The small collection received from the Boston Museum has been catalogued, labeled, and incorporated in the Museum. The special labeling of the Lowell collection, begun some years ago, has been completed. Considerable progress has been made in the systematic arrangement of the collection of dupli- cates. Fifty specimens of economic fungi have been received from Seymour and Earle. Twenty-two persons have been permitted to consult and study in the herbarium. Paleontology. Miss Bryant has unpacked and named and catalogued the Cura- tor's collection of Anticosti fossils, which have been stored in the cellar for several years. These specimens had been loaned to the late Prof. James Hall, but were found, upon being opened, not to have been labeled while in Albany. The same assistant has also taken care of a small collection received from the estate of Mr. Waterston, and has spent some time in the identification of the corals recently purchased from Mr. G. K. Greene. MOLLUSCA. Miss Martin has been occupied mainly in the effort to bring together all of the collections in this department, which, owing to the absence of proper facilities for storage, have been hitherto neces- sarily kept in several different places. The Mollusca room has been furnished with suitable cases for this purpose, and the work of incor- porating all the different lots of shells into one single systematic collection has made some progress. Tliis necessarily involves a large amount of labor that will probably last for several years. The same assistant and Miss Bryant have worked over, and placed in vials and proper boxes, and numbered, the single shells, and have cata- logued all of the Gasteropoda of the Roper collection. The Cyreni- dae, the most valuable part of the Roper collection, containing a considerable number of t\T)es of new species and rare shells, were brought here by Mrs. Roper on her return from the West, and are Digitized by VjOOQ IC 22H PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. now in our collection. A full report on this collection is deferred until the whole has been i)ro[)erly catalogued and numbered. The shelves and specimens in Room K, containing? most of the shells, and those in the gallerj' adjoininij, have l>een dusted and placed in as good order as practicable, but this labor is really thrown away as long as the old cases remain in their present condition, with loosely fitting doors. Miss Brj-ant has looked over our special collection of New Eng- land shells and made a list of the species needed to fill gaps in this series, and has filled a few of these gaps with shells found elsewhere in the Museum. The Curator has continued his work upon the Achatinellidae, especially the ground shells of this family, and has practically completed the detailed descriptions in manuscript of all the species of the genera of this division throughout the whole chain of the Hawaiian Islands. In pursuance of the plan of this work, applica- tion has been made for the collections of these ground shells stored in the principal museums of this country. This part of the work has so far included only the small collections of the Yale University Museum and the Smithsonian Institution ; but, as it has been going on for a short time only, and the progress has been rapid, it is thought that it will not take many months. The Curator will undoubtedly be able to enrich the collection by exchanges, and he has already foun«l some exceedingly rare shells and some distinctly new species in the two collections so far studied. Rev. II. W. Peck has very generously placed his collection of Achatinellidae as a loan in the Society's Building, and his Amastras and other land shells have been named and described. Mr. Oleson has withdrawn his collection of Achatinellidae, to offer them for sale elsewhere. There still remain in the neighborhood of 40,000 shells in this building, and nearly a complete set of all the species of this family ; so there is sufficient material. Special observations had also been ma«le upon all of the Oleson shells, and tlie only loss is in the ability to revise manuscript relating to these from time to time, an omission which will tell more decidedly upon the value of Mr. Oleson 's collection than upon the memoir in which they are mentioned. CRrSTACEA. Professor Kingsley has coni}>leted the naming of fifty-three lots of the Amphipoila loaned to him some years since, and these have Digitized by VjOOQ IC HYATT : REPORT OF THE CURATOR. 229 been returneer of new (lia^ams have been made by Miss Martin. The Curator, assisted by Mr. Coles and Miss Martin, has rearranged the entu^ collection, a task requiring considerable labor, owing to the confusion into which all parts of this collection had fallen during the last ten years. Remarks. • An unusually large amount of time has been expended this year upon miscellaneous work not reporteecimens each^ and about 2,000 specimens were useublic schools. Fifty- nine meml)ers of this class were from Boston, and the balance represented thirty-four neighboring towns. Dr. R. W. Greenleaf, who had given the lessons on botany for Digitized by VjOOQ IC HYATT : REPORT OF THE CURATOR. 233 six years beginning in the winter of 1891-92, wa« obliged to resign on account of the pressure of his professional engagements. The Curator greatly regretted this, since Dr. Greenleaf had been a most successful teacher and his courses were very attractive to the best class of teachers and productive in other results noted from time to time in previous Reports. Mr. B. II. Van Vleck continued Dr. Green- leaf's four-year course, completing the third year with sixteen les- sons of two hours each, beginning Nov. 18, 1899, and ending March 24, 1900. The number of persons registered was forty, and the average attendance thirty-two. Thirty took the examination, and all passed. Two students were for adequate reasons allowed to be absent and to have an examination later by Mr. Van Vleck. The subject was the structure and physiology of algae, and numer- ous preparations and specimens were used, fully illustrating the more important morphological and physiological facts which it was desirable to demonstrate clearly on account of their general bearing in relation to the higher orders of plants. Mr. Van Vleck was assisted effectively by the work of Miss Cora H. Clarke ; and through her kindness the class was able to do its work more advantageously and to receive pressed mounts to the number of 700, representing 25 genera and thirty species. These were gratefully received by members of the class, who highly appreciated ^liss Clarke's generosity. The Curator gave the last series of lessons in a five years' course, consisting of twenty-two lessons of two hours each, altogether forty- four hours of instruction, beginning on the 21 st of October, 1899, and ending on the 15th of April, 1900. The examination has not yet been held, having been postponed until the second Saturday in May. The number of lessons exceeds that of any previous year, but it was necessary in order to cany out the plan of the whole course and finish it properly. The subjects were some of the higher orders of Insecta not finished last winter and the Vertebrata, ending with a special lesson on man. Thei-e were forty-eight tickets issued, and the average attendance was thirty-six. This course was fully illustrated as usual with speci- mens, and the lessons consisted partly of lectures and partly of obser\'ations made by the students themselves under the direction of the Curator, the object being instruction in the broad general facts of structural and functional relations of different animals, with only enough systematic w ork to enable the pupils to recognize the natural relations of the types use The meetings, attendance and communications have been as follows : — May 3, 1899. Annual meeting. Thirty-four persons present. Reports of the Curator, Secretary, Librarian, Treasurer, and Trustees. Mr. W. L. Tower. A quantitative study of the migration and variation of the Colorado potato-beetle. May 17, 1899. General meeting. Thirty-one persons present. Prof. Alpheus Hyatt. Exhibition of Gage's series of brook and lake lampreys from New York. Mr. A. W. Grabau. Evolution of the Fusidae. November 1, 1899. General meeting. Seventy-seven persons present. Prof. W. M. Davis. Geographical notes of a year in Europe. Dr. H. S. Pratt. The embryonic history of imaginal discs in the brachyceran Diptera. (By title.) Mr. Glover M. Allen. The species of Evotomys of eastern North America. (By title.) Mr. Arthur M. Edwards. Diatoms of the U. S. geological survey of the Territories. (By title.) November 15, 1899. General meeting. Sixty-one persons present. Mr. J. G. Jack. Forest aspects and problems in central Colorado. December^, 1899. General meeting. Ninety-three persons present. Mr. William Brewster. Nesting habits of some New England birds. 3Ir. H. T. Burr. The discoveiy of fossils in the Roxbury con- glomerate. Dr. R. P. Bigelow. Anatomy and development of Cassiopea xamachana, (By title.) December 20, 1899. General meeting. Forty-five persons present. Mr. A. W. Grabau. Notes on a geological excursion in the Rocky Mountains of Colorado. Mr. R. H. Howe, Jr. Description of a new race of Horizopua virena (Linn.) . (By title.) Digitized by VjOOQ IC 236 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Jaiiuary 3, 1900. (Teneral meeting. Fifty-six persons present. Prof. F. G. Wright. New light on the age of the Niagara gorge. Mr. R. T. Young. A brief report on the mammals of Prince Edward Island. (By title.) January 17, 1900. (xeneral meeting. Thirty j>erson8 present. Dr. R. T. Jackson. Some cases of oUl age characters in inver- tebrates. February 7, 1900. General meeting. 'Thirty persons present. Dr. C. S. Minot. Notes and illustrations of mammalian devel- opment. Dr. G. H. Parker. The correlation between the size of litters and the number of mammar}' glands in the swine. Dr. R. P. Bigelow. Notes on the development of Cassiopea, Dr. C. S. Minot. On the solid or closed condition of the intestine in the chick. Dr. C. S. Minot. On a hitherto unrecognized form of blood circulation without capillaries, in the organs of vertebrata. (By title.) Prof. M. A. Willcox. A revision of the systematic names employed by writers on the morphology of the Acmaeidae. (By title.) February 21, 1900. (xeneral meeting. Forty-five })ersons present. Prof. C. II. Fernald. The g\l>sy moth in America. March 7, 1900. General meeting. Fifty-eight }>erson8 present. Dr. Frank Russell. The Moki snake dance. March 21, 1900. General meeting. Eighty-six persons present. Prof. W. M. Davis. Glacial erosion in the Alps and in Norway. April 4, 1900. General meeting. Twenty-eight j)ersons present. Dr. G. n. Parker. The neurone theory in the light of recent investigations. April 18, 1900. General meeting. Thirty-seven }>ersons present. Mr. J. II. Enierton. The common species of American spiders. PUBLK ATIONS. During the year the following publications have been issued : — Proceedings of the annual meeting, May 8, 1899. Proceedings, vol. 29, no. 1, 43 pp. Digitized by VjOOQ IC BATCHELDER : REPORT OF SECRETARY AND LIBRARIAN. 237 Variation and Bexual selection in man. By Edwin Tenney Brewster. Proceedings, vol. 29, no. 2, 17 pp. Notes on the rejitiles and amphibians of Intervale, New Hamp- shire. By Glover M. Allen. Proceedings, vol. 29, no. 3, 13 pp. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United States. By Garry de N. Hough. Proceedings, vol. 29, no. 4,10 pp. Contributions from the Gray Herbarium of Harvard University. New Series. — No. 17. By B. L. Robinson and J. M. Greenman. Proceedings, vol. 29, no. 5, 22 pp. The development of Penilia achmackeri Richard. By Memn T. Sudler. Proceedings, vol. 29, no. 6, 23 pp., 3 pis. List of marine mollusca of Coldspring Harbor, Long Island, with descriptions of one new genus and two new species of nudibranchs. By Francis Noyes Balch. Proceedings, vol. 29, no. 7, 30 pp., 1 pi. The blood vessels of the heart in Carcharias, Raja, and Amia. By G. H. Parker and Frederica K. Davis. Proceedings, vol. 29, no. 8, 16 pp., 3 pis. The occurrence of fossils in the Roxburj' conglomerate. By Henry T. Burr and Robert E. Burke. Proceedings, vol. 29, no. 9, 6 pp., 1 pi., 2 cuts. On a hitherto unrecognized form of blood circulation without capillaries in the organs of vertebrata. By Charles Sedgwick Minot, LL.D. Proceedings, vol. 29, no. 10, 31 pp., 12 cuts. A revision of the systematic names emi)loyed by writers on the morphologA^ of the Acmaeidae. By M. A. Willcox, Ph.D. Pro- ceedings, vol. 29, no. 11, 6 i)p. In connection with the publications mention should be made of the bequest of 810,000, received this year from the late Robert C. Waterston, the income of which is to be devoted to the Society's publications. Library. The additions to the library have been : — 8vo. 4to. Folio. Total. Volumes Parts Pamphlets Maps Total 3S8 77 3 408 lS-21 355 1 2177 4G7 29 3 499 40 40 tioTG 461 47 '31H4 Digitized by VjOOQ IC 238 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. The library contains 25,629 volumes, 1401 current, or otherwise incomplete, volumes, and 13,311 pamphlets. Among the more important gifts to the library during the year have been the bequest of the late Robert C, Waterston (120 vol- umes) , some books from Mr. James M. Barnard (29 volumes) , and Field's Bibliography, received from the Trustees of the Thompson Fund. New exchanges have been arranged with the Philosophical Society of Washington, the Kongliga Universitets-Biblioteket, Upsala, Swe- den, the Royal Geographical Society of Australasia, and the Bemice Pauahi Bishop Museum, Honolulu. One exchange has ceased. The Society now exchanges its pub- lications with 435 institutions and periodicals. Eight hundred and ninety-two books have been borrowed by 182 persons ; 441 have been borrowed for use in the building ; the library has been consulted about 350 times. Four hundred and one volumes have been bound in 270 covers. Twelve volumes of the Proceedings of the U. S. National Mu- seum have been indexed. Current volumes of serials pre\nously indexed, are indexed as received. Walker Prizes. The subjects for competition appointed for 1900 were: — 1. Stratigraphy and correlation of the sedimentary formations of any part of New England. 2. A study in palaeozoic stratigraphy and correlation. The Committee has reported the following awards : A prize of one hundred dollars for the essay entitled, " A study in palaeozoic stratigraphy and correlation : the Hudson River beds of the neighborhood of Albany and their taxonomic equivalents," by Rudolf Ruedemann, Ph. D. A i)rize of fifty dollars for the essay entitled '* Cephalopod zones in the Carboniferous of Xoith America : a study in interregional coiTclation,'* by James Perrin Smith. The subject for the award in May, 1901, is: — A monograph on any problem connected with, .or any group belonging to, the Xoith American fauna or flora. Digitized by VjOOQ IC BOUVE : REPORT OF TREASURER. 239 > o H H O H o a. » aO 1-1 r^ 28 S.;:8^S^SS88S8$^8^ ^ •Sc5o-*a>i-3 *^ CO o Q lo Qooeo o> o o og t>. s «o ei CO o> o Q o *^ 35 eo »N «♦ o •-• 8^" cS^^ m .1= "2-3 :i : :=^ : SQOC ; 2 ••I ^ a £ ' - - - ^ - •s ^ 3 (2 3 1° •§8 I V* a s I %^^ n 3 « V o V E 4) 9 e2 ^S II « o »2 ell S- g " -i Digitized by VjOOQ IC 240 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. The reportH of the Trustees and of the Auditing Committee were presented, and it was voted to a(x*.ept the several reports. The Society then proceeded to ballot for officers for 1900-1901. Messrs. J. H. Blake and J. B. Woodworth were appointed to collect and count the ballots. They reported the election of PRESIDENT, CHARLES SEDGWICK MINOT. VICE-PRESIDENTS, CHARLES p. BOWDITCH, HENEY W. HAYNES. WILLIAM G. FARLOW. CURATOR, ALPHEUS HYATT. SECRETARY, CHARLES F. BATCHELDER. TREASURER, EDWARD T. BOUVfe. LIBRARIAN, CHARLES F. BATCHELDER. COUNCILLOR FOR ONE YEAR, Mis8 Clara E. Cummings. COUNCILLORS FOR THREE TEARS, S. L. AimoT, Mi.sH Catharine I. Ireland, William S. Bryant, Benjamin Joy Jeffries, William M. Davis, N. T. Kidder, Samuel Henshaw, William H. Niles. Printed, May, l-^oo. Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. Mkmuiuh. 4t(>. The development, structure, and affiaities of the p'nus E(}ui8etum. By Edward C. Jeffrey. 30 pp., 5 plates. $1.(X). Localized staples in development in plants and animals. By Robert T. Jack.son. 65 pp., 10 plates. $2.00. raocKEDiNOs. 8vo. A revision of tlie systematic names employed by writers on the morphology' of the Acmaeidae. By M. A. Willcox. 0 pp. 10 cts. On a hitherto unreco^niiz*Hl fonn of blood cifculation without capillaries in the organs oi veruebrata. By Charles Sedgwick Minot. 31 pp. 36 cts. The occurrence of fos^^ils in the Roxbury con.^lomerate. By Henry T. Burr and Robert K. Burke. 0 pp., 1 plate. 20 cts. Tae blood vessels of the heart in Carcharias, Raja, and Amia. By G. H. Parker and F. K. Davis. 10 pp., 3 plates. 26 cts. List of marine mollusca of Coldspring Harbor, Long Island, with d'CHcriptions of one new genus and two new sjxjcie.s of nudibranchs. By Francis Noyes Balch. 30 pp., 1 plate. 36 cts. Tiie development of Penilia schmackeri Richard. By Mervin T. Sudler. 23 pp., 3 plates. 30 cts. (\»ntributi{)ns from tlie (»ray herbarium of Harvard university. New series, no. 17. 1. Revision ot the genus (iymnoiomia. 2. Supplementary' not^s uimn Calea, Tridax, and Mikania. By B. L. Robuison and J. M. Ciieenman. 22 pp. 26 cts. Studies in Diptera Cyclorhapha. 1. 'Hie Pipunculidae of the United States. By Garry de N. Hr)ugh. 10 pp. 10 cts. Notes on the reptiles and amphibians of Intervale, N: H. By Glover M. Alien. 13 pp. 16 cts. Viiriation and st^xual selection in man. By Edwin Tenney Brewster. 17 pp. 26 cts. MoniloporidcU% a new family of Palaeozoic corals. By Amadeus W. Grabau. l(i pp., 4 plates. 26 cts. Studies in the gold-bearing slates of Nova Scotia. By J. Edmund Woodman. 33 pp., 3 plates. 60 cts. North American wood -frogs. By lieginald Heber Howe, Jr. 6 pp. 10 cts. Some Hydroids from Puget Sound. By Gary N Calkins. 36 pp., 0 plates. 60 CIS. The Odonate genus Macrothemis and its allies. By Philip P. Calvert. 32 pp., 2 plates. 60 cts. On the veins of the Wolffian bodies in the pig. By (Miarles Sedgwick Minot. 10 pp., 1 plate. 26 cts. Notes on a (Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 16 cts. The genus Antennaria in New England. By Merritt L. Fernald. 13 pp. 16 cts. The laud mammals of peninsular Florida and the coast region of Georgia. By Outram Bangs. 70 pp. 76 cts. A contribution to the petrography of the Boston Basin. By Theodore G. White. 40 pp., 6 plates. 06 cts. Clymene producta sp. nov. By Margaret Lewis. 6 pp., 2 plates. 16 cts. The Harvard geographical models. By W. M. Davis. 20 pp., 4 plates. 25 cts. The role of water in growth. By C. B. Davenport. 12 pp. 16 cts. Digiti ized by Google Ho Prooeadln^ Of the Boston Sooietj of Natural History. Vol. 29, No. 13, pp. 241-272. JUL 17 ISM THE EMBRYONIC HISTORY OF IMAGINAL DISCS IN MELOPHAGUS OVINUS L., TOGETHER WITII AN ACCOUNT OF THE EARLIER STA(;ES IN THE DEVELOPMENT OF THE INSECT. By H. S. Pratt., Ph. D. WlTH^SEVKN I'LATKS. BOSTON: PRINTED FOR THE SOCIETY. JlTNE, 1900. Digitized by VjOOQ IC Digitized by VjOOQ IC JUL 17 1900 No. 13. — The Embryonic History of Imaginal Discs in Melo- phagHfS ovinus X., together with aii Account of the Earlier Stages in the Development of the Insect} By H. S. Pratt, Ph.D. With seven plates, CONTENTS. Page Introduction 241 Historical : 1. Imaginal Discs in the Larva and Pupa . ... 242 2. Imaginal Discs in the Embryo 248 The Earlier Developmental Stages of Melophagus : 1. The Development of the Egg to the Completion of the Blasto- derm 251 2. The Formation of the Mesoderm and of the Proctodeum and Stomodeum 255 The Origin of the Imaginal Discs : 1. The Cephalic Discs 259 a. The Early Development of these Discs . . . • 259 6. The Involution of the Head 264 2. The Thoracic Discs 267 3. The Discs of the External Genitalia 268 4. The Discs of the Internal Organs 269 Methods 269 Bibliography 270 Introduction. In 1897 I published in " Psyche " a preliminary account of the origin and early development in Melophagus ovinus of those funda- ments of the imaginal head, wings, and appendages, which may be present in the larva of holometabolic insects and are known as imaginal discs. The present paper is an extended account of the same matter. Melophagus belongs to the small group of brachycerous Diptera 1 Contributions from the Zoological Laboratory of the Museum of Comparative Zoology £X Harvard College, under the direction of £. L. Mark, No. 111. Digitized by VjOOQ IC 242 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. called the Pupipara, which are closely related to the muscids. Leuckart (*S7) early showed the striking similarity of structure of the larval and pupal forms of the Pupipara to those of the muscids ; in a former paper ('93) I have emphasized the same fact ; MOggen- burg ('92) has homologized the mouth parts of all the pupipars with the fly's proboscis ; Brauer ('85) has shown that the Pupipara are degenerate flies and has placed them next to the Muscidae in his classification of the Diptera. The Muscidae are classical objects for the study of imaginal discs. It was with them that Weismann inaugurated the study of these interesting bodies, and, since the appearance of his first paper in 1803, Ganin, Kowalevsky, Van Rees, and many others have publidied the results of investigations of them. All of these studies, however, have been on the development of the discs during the larval and pupal period of the insect's life. The embryonic history of the discs, which includes their origin and the first stages of growth, has not been studied ; and no positive information exists concerning them except a short statement contained in a paper by Graber ('89), which will be spoken of later on, and the preliminary paper by myself, mentioned above. Historical. 1. Imagimd Discs tfi the Larva and Pupa. — Swammerdam (1737-38) was the earliest investigator to observe that fundaments of the imaginal thoracic appendages in the higher insects, and even of the head in some cases, do not appear first in the pupa, but are present in the larva. lie was thus the first observer of imaginal discs. He showed that in the larva of Culex, Apis, and Pieris the fundamentd^ of all the legs and wings lie beneath the thoracic integument. From his time down to the present generation no additions were made to the knowledge of the subject. Numerous investigators, however, made observations similar to those of Swammerdam. Lyonnet (1760) described and figured the two pairs of imaginal discs in the dorsal portion of the meso- and metathoracic segments of the caterpillar, and added the supposition that they were the funda- ments of wings. Herold ('15) described the same discs, and cor- rectly interpreted them. Burmeister ('35) also very accurately described the imaginal wing-discs of the caterpillar, as did Louis Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 243 Agassiz ('51) somewhat later. Leuckart ('57) mentions the tho- racic and cephalic imaginal discs of Melophagus, but without describ- ing them or knowing their ultimate fate. It was in the years 1863 to 1866 that Weismann ('63, '64, '66) laid the foundations of our present knowledge of imaginal discs in a series of investigations on the development of Musca and other Diptera, both brachycerous and nematocerous. He ('63, p. 229) found that the dipterous larva, which is apodous, and in the case of the Brachycera, acephalic, contains within its thorax six pairs of disc-like bodies. In Simulia, a nematocerous dipter, which is the first form he studied, they are situated near the animal's integument, although apparently not in connection with it. There are three pairs ven- trally located, each pair belonging to one of the thoracic segments, and three pairs dorsally located, these being similarly distributed among the thoracic segments. All of these bodies he found in close relation to nerves or tracheae or both. He further found that they remain function less during the life of the larva, although increas- ing greatly in size with the growth of the larva, and that during the metamorphosis they develop into certain organs of the imago. He called them, consequently, imaginal discs. He found that the three pairs of ventral discs develop into the imaginal legs, the dorsal metathoracic pair into the balancers, the dorsal mesothoracic pair into the wings, and the dorsal prothoracic pair into the anterior pupal spiracles, when these are present. In Corethra (Weismann, '66), also a nematocerous dipter, he found similar conditions. In Musca he ('63, '64) found the conditions very much more complicated. The six pairs of discs just mentioned he found pres- ent ; but, instead of being located near the integument, they were sunk into the centre of the animal's body. An additional pair of discs was also i)resent in the forward portion of the thorax, directly in front of and closely applied to the brain -garfglia ; these he found were destined to develop into the imaginal head. Weismann also found that only a small portion of the larval body passes directly into the imaginal body, the greater part of it undergoing disintegra- tion, so that the tissues entirely lose their identity, and afterwards the imaginal body is built up anew from the imaginal discs. To this process, the entire significance of which, however, was not understood until later, he gave the name " histolysis P These early papers of Weismann have furnished the starting- Digitized by VjOOQ IC 244 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. point for a large number of investigations upon the imaginal discs. He confined his studies to the Diptera ; other insect groups were soon investigated. Landois (*71) following Herold, Agassiz, and most of the older authors, studied the imaginal wing- discs to be found beneath the dorsal' thoracic integument of the caterpillar. Ktinckel d'llerculais (*75) showed that in the larva of Volucella, a muscid, the imaginal discs, although situated at a dis- tance from the integument, are connected with it by a delicate chord, the remnant of an invagination. He also discovered two pairs of imaginal discs near the hinder end of the body of the larva, which develop into the external genital organs. Ganin (*76) studied imaginal discs in several groups of insects, — namely, the Hymenoptera, Neuroptera, Coleoptera, Lepidoptera, and Diptera, — in the larvae of all of which he found wing-discs, and in those which are apodous, leg-discs as well. He added to these observations the discovery, in the larvae of brachycerous Diptera, of other discs than those described by Weismann. That author believed that the hypodermis of the larval abdomen went directly with qiodifications to form that of the imago. Ganin now showed that in, and forming a part of, the hypodermis of each of the eight abdominal segments of the muscidian larva, are four discs, two dorsal and two ventral, the tissue of which resembles that of the thoracic discs, and that they form the starting point for the grovrth of the imaginal hypodermis of the abdomen. Ganin likewise discov- ered similar discs in the epithelium of the larval mid-gut, whose fate it is to form in the same way the imaginal mid-gut ; and he also dis- covered the important fact that each imaginal disc is made up of two kinds of embryonic tissue, ectoderm and mesoderm. Further- more he discovered the amoeboid mesoderm cells which destroy the larval organs during histolysis. Dewitz ('78) took up the study of imaginal discs for the purpose of reviewing the work of Ganin, Landois, and Weismann. A few years later Viallanes ('82) studied afresh the post-embryonic devel- opment of the Muscidae, and laid the basis of our present knowledge of the histological details of the process of histolysis in the meta- morphosis of insects. In 1882 and 1883 Metschnikoff published the first of his epoch- making studies on the destruction of tissues in certain invertebrates by leucocytes, or, as he called them, phagocytes. He discussed Ganin's observations and especially that of the destruction of the Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 245 larval organs by amoeboid mesoderm cells during histolysis in the pupal muscid. These cells, he suggests, are none other than phago- cytes. Shortly afterward. Van Rees ('84) and Kowalevsky ('85) proved the correctness of this position. They showed that the process of histolysis consists of the ingestion and digestion of the functional larval tissues by phagocytes, and the building up of imaginal tissues from imaginal discs. Kowalevsty ('87) in another paper took up the investigation of the histolysis of the internal organs of Musca where Ganin and Viallanes had left it, and gave the first complete account of these processes. He showed the exact method by which the muscles, digestive tract, and hypodermis of the lai*va are destroyed by phagocytes and the imaginal organs reconstructed from imaginal discs. And in the following year Van Rees ('88) published his extensive paper on the post-embryonic development of muscids, and completed our knowledge of this phenomenon. He showed that when the muscidian larva enters upon the pupal stage, histolysis is inaugurated by the destruction of the larval muscles, which become unfunctional du-ectly after pupation and a natural prey to the phagocytes. Soon the thoracic hypodermis and the inner organs are a^ttacked, and at the same time the imaginal discs begin to grow and widen out, supplying the place of the tissues which are being destroyed. The continuity of the hypodermis and of most of the internal organs is thus at no time broken, an obsei-vation which Kowalevsky ('87, p. 585) also made, correcting at the same time the statement to the contrary made by Viallanes ('82, p. 221) . As these processes go on, the two large cephalic imaginal discs, which form two irregularly shaped sacs extending as diverticula from the dorsal wall of the pharynx back to the brain, begin to move for- ward, dragging the brain with them. Their anterior ends bend and pass ventrally, embracing the pharynx between them. At the same time their communications with the pharynx enlarge and their lumina fuse more and more completely with the pharyngeal lumen until they meet in the median line and form one single median opening, which, ever increasing in size, finally extends the entire length of the discs. The lumina of the discs and of the pharynx thus become completely merged and form together a single con- tinuous space, and the walls of the discs and of the pharynx a single continuous vesicle. This is the head vesicle or " Kopfblase," which is destined to become the imaginal head. This vesicle remains Digitized by VjOOQ IC 246 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. buried within the pupal thorax Until near the end of the pupal period, when it evaginates and forms the completed head of the insect. This evagination has been obsei'ved by Weismann ('64, p. 259) to be the consequence of the pressure of blood, which at the right moment is forced from the abdomen into the thorax and pushes the head vesicle forward. The metamorphosis of the thorax goes on simultaneously with the formation of the head vesicle. In proportion as the larval hypoderrais disappears under the attacks of the phagocytes, as already mentioned, the edges of the imaginal discs grow and take its place, forming the imaginal hypodermis. As we have seen, there are six pairs of these discs, a doi-sal and a ventral pair being present in each thoracic segment. They lie at a dis- tance from the integument, near the centre of the larva. Each disc is, however, connected by a very fine hollow cord with that portion of the hypodermis of the segment to which it genetically belongs, and where it is destined to appear as an extremity. This cord, first discovered by Ktlnckel d'llerculais, gradually shortens, and its lumen enlarges. The disc is thus brought nearer the surface, and as it advances it increases in size. The lumen of the cord then opens through the hypodermis, and the cord itself fin- ally becomes so wide and short that the disc is brought through the hypodermis to the outside. The hollow cord is of course obliterated by this process, and the edges of the proximal end of the disc are brought into direct contact with the hypodermis. The disc has now assumed its position as an extremity. It is an appendage of the body-wall ; it has become irregularly cylindrical in shape, and possessed of a number of constrictions and folds, which in the case of the ventral discs are equivalent to the joints of the future leg. The metamoq^hosis of the abdomen is retarded ; it does not begin until that of the head and thorax is well advanced. Then in each abdominal segment the two ventral and four dorsal discs (Van Rees found two additional dorsal discs in each segment) begin to grow and take the place of the disappearing larval hypodermis. In a paper of my own ('93), which contains the results of a study of the larva of Melophagus ovinus, is contained a full description of the imaginal discs of this insect. Melophagus, being a pupipar, and closely allied to the muscids, we should expect to find the same Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 247 imaginal discs in its larva as in the muscidian larva. And we do, in- dac. prHhz. dsc. ms'thx. dsc. mVthx. A B A, ventrcU frontal section ; B, dorsal frontal section ; dsc. pr^thx., dsc. msHhx.^ and dsc. mVthx.^ pro-, meso-, and metathoracic imaginal discs. deed, find similar conditions in general ; but there are some interest- ing and instructive differences. The larva is apodous and acepha- lous, like the muscidian, but in many ways it is less highly special- d 0 » 0 0 0 olo e oooooooooo ized ; indeed, it seems, in some respects, to occupy an intermediate position between Corethra and Musca. In the position of the thoracic discs, for instance, it closely resem- bles Corethra. These discs are found just beneath the integument in two very regular rows, and not near the centre of the larva, as in the muscidian larva. The accompany- ing cuts represent frontal sections through the anterior end of an old larva (Fig. A being a ventral and Fig. B a dorsal section), show- ing the position of the thoracic discs. In struc- ture the meso- and metathoracic discs stand exactly halfway between the same discs in Co- rethra and in Musca. In Corethra, according to Weismann ('66, p. 78), all the thoracic discs are of larval origin, and each is a double fold of the hypodermis, of which it remains a part, as is shown by Figure C. In Melophagus, on the other hand, these discs arise in the embryo; they are also Digitized by VjOOQ IC 248 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. double folds of the hypodermis, but become constricted off from It, as shown in Figure D. In Musca the discs are also of embryonic origin. They also become constricted off from the hypodermis; but, instead of remaining where they have originated, they suffer removal toward the centre of the larva, as is shown in Figure'E, and the peripodal membrane, as Van Rees calls the outer wall of the disc, lengthens to form the hollow cord which connects it with it3 old position at the hypodermis. In the cephalic discs the conditions are similar to those in Musca, but even more complicated. Instead of a single pair of head-discs there are two pairs present, one dorsal and one ventral. The dorsal pair corresponds to the muscidian head-discs in every respect ; they are destined to form the dorsal and lateral portion of the imi^inal head together with the compound eyes. The ventral head-discs have no counterpart in Musca. In the embryo they arise as a single median thickening, from which paired diverticula develop. From the bottom of each of these diverticula, which in the larva project from the ventral pharyngeal wall, there extends into its lumen a long projection. The diverticula fuse in the median line during the latter portion of the larval period, and the wall thus formed between them gradually disappears, so that in the full-grown larva the ventral discs appear as a single ventral diverticulum of the pharynx, at the bottom of which a pair of long projections extends towards the wide opening. The fate of these discs is to form the ventral portion of the head, the paired projections being the funda- ments of the proboscis. The formation of the head- vesicle during the metamorphosis proceeds in a way similar to that in Musca. The ventral disc fuses early at its lateral edges with the dorsal pair; the communications between both ventral and dorsal discs and the pharynx become rapidly larger (in the old larva they have already become very wide), and soon the discs and the pharynx form together a single vesicle, the head-vesicle. 2. Imaginal Discs in the Embryo. — As already stated, the em- bryonic history of imaginal discs has not been studied in the higher insects. The following are the speculations of authors as to their origin. In Corethra, a nematocerous dipteron, Weisraann ('66) found that the imaginal discs do not make their appearance until after the last larval moult. In the Brachycera, on the other hand, he found the cephalic and thoracic discs present in the youngest larvae, and he Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 249 concluded that they must arise in the embryo. Concerning, however, the exact manner in which they take their origin, he was not able to make any positive statements, but was of the opinion that they first appear within the body-cavity as proliferations of the epithelial coverings of tracheae and of the neurilemma of nerves. In Core- thra, on the other hand, he observed that the discs arise as folds of the larval hypodermis, and he calls attention to the fundamen- tally different method of origin of imaginal discs in these represent- atives of the two great divisions of the Diptera. Landob ('71) , investigating the development of wings in the larva of Lepidoptera, follows Weismann in his conclusions as to their ori- gin in the embryo. Ganin ('76) also supports Weismann as to the embryonic origin of imaginal discs in all the different groups of insects he studied. KUnckel d'Herculais ('75) was the first one who discredited Weismann 's conclusions. He found each disc in the muscidian larva to be connected with the hypodermis by means of a cord, and he rightly concluded that it resulted from an invagina- tion and that the discs are therefore of ectodermic origin. Weis- mann (*64, p. 139) also, it is true, saw these connecting cords, but failed to interpret them correctly. Dewitz ('78, '81) after a study of lepidopterous larva came to the same conclusion concerning the origin of the imaginal discs as Kttnckel d*Herculais, namely, that they are ectodermal invaginations. Pancritius ('84), also studying the Lepidoptera, reached the same conclusion. Balfour, in his text- book ('80) declared that, notwithstanding the authority of Weis- mann to the contrary, the cephalic and thoracic imaginal discs of Musca must be derivatives of the ectoderm, as they are in Corethra. Kowalevskjr ('86), after a study of the embryo of Musca, declared himself unable to determine the method of origin of the imaginal discs. He arrived at the negative result, however, that they do not arise as growths from the epithelium of tracheae, but that tracheae and nerves unite with them while they are still young. Van Rees ('88) in his studies of the muscidian larva, demonstrated the existence of a fine lumen, a continuation of the peripodal space, in the cord connecting the disc with the hypodermis, and showed that both lumen and peripodal space are lined with a fine cuti- cula. He asserts that this discovery is anatomical proof that the cord, the peripodal membrane, and the disc itself have all been parts of a single invagination of the embryonic ectoderm in exactly the same way as the imaginal discs in Corethra arise as in vagina- Digitized by VjOOQ IC 250 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. tions of the larval hypodermis. Graber ('77- '79), however, thought otherwise. In his valuable treatise ** Die Insekten," part 2, p. 563, following the results of the first paper of Dewitz ('78;, which had just been published, he introduces a diagram (Fig. 206, E) repre- senting the thoracic discs taking their rise as deep invaginations of the hypodermis. lie does not, however, commit himself to this interpretation, but says " Eine solche Annahme entbehrt aber yorlaufig jeglicher BegrUndung, und so dtirfte es doch besser sein, das Vorkommen einer endogenen Insek ten-Metamorphose einfach anzuerkennen als sie mit Gewalt zu einem Vorgang umzudeuteln der mit den bisherigen Beobachtungen nicht Ubereinstimmt." Shortly after the publication of the above-mentioned paper of Van Rees, Graber ('89) published an extensive series of studies on the embryonic development of the Muscidae and certain other insects, in which the origin of the imaginal discs is touched upon. He adheres to the conclusions of his former work just quoted, and takes exception to those of Van Rees. Supported by certain obser- vations on the embryonic development of Lucilia, he asserts that the anatomical evidence of that author does not prove the cephalic and thoracic discs to be hypodermal invaginations. Graber's obser- vations were made on the cephalic discs and were as follows : The first traces of these discs which he found were a pair of thick plates composed each of a siiiffle layer of epithelium situated at the right and left of the pharynx of the larva and in connection with its lat- eral walls (see his Figs. 116, 117, 117*). The origin of these plates, therefore, he did not observe, but the fact that when he first noticed them they were i7iter7ial structures, and the further fact that they were not at this early stage in the form of sacs, but of plates, led him to think that the sac-form these discs possess in their later and larval developmental stages is a secondary adaptation. In other words, he believed these discs to arise in the body-cavity of the animal as epithelial thickenings which afterward assume a sac-form. He asserts also that the possibility cannot be denied that the thoracic imaginal discs have a similar origin. In fact, this is the interpretation of the matter he himself believes. In later years, Verson ('90) and Gonin ('94) , after studying Lepi- doptera, Bugnion ('91), after studying a hymenopteron, and Wahl ('99), a dipteron (Eristalis) , have adopted the opinion of Van Rees, that imaginal discs are of ectodermal origin^ My own investigations on the embryology of Melophagus will Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 251 show that Graber's conclusions were in great part false. I shall show that in the Brachycera the imaginal discs of the head and thorax all arise in the embryo first as thickenings of the ectodermic body-wall, and not in the form of endogenic plates; that these thickenings soon invaginate and form simple pockets in the ecto- derm. I shall show that the cephalic invaginations are formed before the involution of the embryonic head characteristic of brachycerous development, and that as a result of this involution, whereby the so-called pharynx of the larva is formed from the walls of the embryonic head, the discs come to form diverticula of the pharynx, in which condition they are found during the entire larval and the first part of the pupal period. It will also be shown that the thoracic invaginations finally become separated from the embry- onic ectoderm, and form the thoracic discs as we find them in the larva; also that the external genital organs take their origin in the form of imaginal discs in the same way as do the thoracic extremities. The Earlier Developmental Stages of Melophagus. 1. The Development of the Egg to the Completion of the Blasto- derm,, — The formation of the blastoderm in the embryos of brachy- cerous Diptera has been observed and described by the following authors, — in Musca vomitoria by Weismann (*64, '82) , Kowal- ovsky ('86), Blochmann ('87), Ilenking ('88), Voeltzkow ('89), and Graber ('89) , and in LuciUa by Graber (*89). The accounts differ considerably from one another, although they are descriptions of the same process and all the authors mentioned are well-known, expe- rienced, and competent observers. On this account, the details of blastoderm formation in Melophagus will be a useful contribution to the subject. Melophagus is not, however, a favorable object for the study of the earlier stages of insect development, because of the impossi- bility of obtaining considerable numbers of its eggs and of deter- mining the age of the embryos within them. The female insect produces only one egg at a time and at intervals of several weeks; each egg must be dissected from the maternal uterus, where development goes on, and the age of the embryo can be ascertained only after the egg has been properly stained, and then Digitized by VjOOQ IC 252 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. only approximately by comparing it with other embryos. I should not, therefore, have selected the eggs of Melophagas for the inves- tigation of this phase of insect development. My problem, as the title indicates, was quite a different one, the study of imaginal discs ; but while searching for embryos containing these structures, I collected a considerable number illustrating all the different phases of the animal's development. The egg of Melophagus has almost exactly the shape of that of Musca, as observed by Weismann ('64) and Blochmann ('87). It b an elongated cylindrical object, tapering at the poles, being blunter at the posterior than the anterior end, with ^ length of 1.2 mm. and an average breadth of 0.3 mm. at its widest part. As seen from the side, its ventral outline is convex, its dorsal outline slightly con- cave. It is covered by a two-layered chorion, the outer layer being much thicker than the inner one, and by a delicate vitelline mem- brane. As it lies in the maternal uterus the main body-axes of thfe developing embryo correspond to those of the mother's body. The micropyle forms a large funnel-shaped depression in the anterior end of the chorion (PI. 1, Fig. 4), which sinks deeply into the yolk, and the deep dent it makes deforms the anterior end of the developing embryo until near the termination of the embryonic period. The micropyle is always found filled with a dense mass of spermatozoa. The very young uterine egg has also a structure similar to that of the egg of Musca, but differs from it in one important particular. It consists of a web of granular protoplasm within which lies a mass of spherical yolk granules, but the peripheral layer of clear protoplasm is exceedingly thin (PI. 1, Fig. 1). It has thus no " Keimhaut- blastem," as Weismann has called the thick peripheral layer of clear protoplasm first found by him in the muscine egg. This is rather remarkable, as the presence of the Keirahautblastem is characteris- tic of the higher insects, having been demonstrated in Musca by Weismann ('64) and others, in the Coleoptera by Heider (*89) and Wheeler ('89), in the Lepidoptera by Bobretzky ('78;, in the Hymenoptera by Grassi ('84) , and in the Hemiptera by Witlaczil ('84), and the absence of it is also characteristic of the lower insects^ as shown by Ayers ('84), Heymons ('95), Wheeler ('89), and Korot- neff ('85) in the Orthoptera, and by Brandt ('69) in the Pseu- doneuroptera. The only other instance with which I am acquainted of the failure of this layer in one of the higher insects is in Lucilia^ Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 253 a muscid, in which, according to Graber ('89), the hinder portion of the egg is provided with it, while the forward portion lacks it. Graber does not find even the thinnest peripheral protoplasmic layer in the forward half of the egg of this fly, but asserts that the yolk-balls abut immediately on the vitelline membrane, — an obser- vation the accuracy of which seems to me on general principles extremely doubtful. Melophagus, however, as will be seen, acquires a Keimhautblas- tem before the completion of the blastoderm. The processes leading to the maturation and fertilization of the egg were not observed. The segmentation-nuclei were first ob- served forming an irregular group near the centre of the egg. Figure 1 (PI. 1) represents a section of an egg in which about ten of these nuclei were counted, all of which were migrating through the yolk towards the periphery. Around each nucleus is a zone of clear protoplasm, possibly the result of the absorption of yolk granules near it. Each nucleus, moreover, as it moves toward the surface, leaves behind it a path of clear protoplasm. Only four of these nuclei are seen in the section, but portions of the paths of others are visible. It is necessary to call attention to the irregular nature of the migration of these nuclei. In Musca the earlier seg- mentation-nuclei are described by Blochmann and Voeltzkow ('89) as advancing in very regular order towards the surface of the egg from the centre and arranging themselves in positions approxi- mately parallel to the surface as they advance. Kowalevsky (*86) observed that the segmentation-nuclei in Musca arrive at the sur- face of the egg first in the hinder portion, then in the forward, and lastly in the central portion. Voeltzkow ('89) observed in the same animal that at all portions of the egg they arrive at the same time. In Melophagus, as will be seen in Figure 1, most of the nuclei are near the centre of the egg, but are advancing quite irreg- ularly towards the surface, as is indicated by their plasma-paths. The penpheral plasma-layer in this egg is very thin, being but 5 ^ thick. Figure 2 shows a section of an egg in which blastoderm formation has advanced much farther. The paths of the advancing nuclei form here a network among the yolk granules. Some of the nuclei are seen to be in the process of division ; but the great majority of them are spherical bodies with distinct nuclear membranes and few chromatin granules. Many have reached the periphery of the eggy Digitized by VjOOQ IC 254 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. and it will be readily seen that there is not the slightest order in their arrangement, nor is there any regularity in the way they advance. Figure 3 represents a portion of the same blastoderm on a larger scale. The peripheral protoplasmic layer has the same thickness as in the last egg mentioned, except at those places where nuclei have entered it. Here its thickness is increased by the diameter of the nucleus, which on the average is 12.5 /i. Figure 4 represents a section of an egg in which the formation of the blastoderm is still farther advanced. The nuclei begin to crowd one another at the surface of the egg, and, not finding room for themselves in the narrow, peripheral, protoplasmic layer, they begin to project outward, and thus form protuberances on the outer surface of the egg, giving a section the wavy outline shown in the figure. This protoplasmic layer has increased somewhat in thickness (probably as a result of the consumption of yolk granules by the rapidly dividing nuclei), and is now 7.5^ thick between the nuclei. As will be seen in Figure 5, which shows a much enlarged view of a portion of this blastoderm, the nuclei have nearly the same diameter as those shown in Figures 2 and 3, and are prolifer- ating much more rapidly. The protoplasmic paths are not so numerous as in the egg last described, and will become less frequent from now on, showing that the yolk granules move together again, thus obliterating the paths. The gradual thickening of the peripheral protoplasmic layer also probably goes on at the expense of these paths. These processes all continue. The segmentation nuclei at the surface increase very rapidly in number, and begin to decrease in size as they become more and more numerous, and the protuber- ances which they occupy begin to crowd one another (PI. 1, Fig. 6). These do not, however, show a tendency to merge with one another. Each protuberance becomes rather more clearly defined and dis- tinct, and its sides, from forming acute angles with the plane of the former surface of the egg, as shown in Figure 4, finally come to form right angles with it (Fig. 6) . The surface of the egg then presents a curious appearance, being like the pavement of a street the blocks of which are separated by deep narrow spaces. The average diameter of the nuclei in the egg represented in Figure 6 is 5 /A. The peripheral protoplasmic layer has also increased in thickness, and now measures 15 ^ at points between the nuclei. It now constitutes a blastema, such as is present in Musca before segmentation begins. Digitized by Google PRATT: IMAGINAL DISCS. 255 The nuclear paths in this egg have entirely disappeared. Their disappearance undoubtedly accounts largely for the increase in the thickness of the blastema. The steps in the formation of the blastoderm between this stage and the completed blastoderm I have not observed, but I believe the concluding process to be as follows : The nuclei and their pro- tuberances still further increase in number, and the former decrease in size ; the latter are brought into immediate contact with one an- other and their walls fuse and become the lateral boundaries of the future blastodei-m cells; these boundaries are carried still farther toward the inner surface of the blastema, and the cells are finally completed by the ^ formation of a wall bounding their inner ends. In the completed blastoderm (PL 1, Fig. 7) there is still a narrow blastema present, and the diameter of the nuclei has fallen to 2.5 IX. No inner or secondary blastema, such as is described in Musca by Graber ('89) and other authors, is present. I did not observe the formation of the pole-cells. 2. The JFormation of the Mesoderm and of the Proctodetim and Stomodeam, — At the time of its completion, the blastoderm is composed of narrow cells of equal height throughout. In the centre of the ^^o^ are numbers of so-called yolk-nuclei. Inasmuch as these yolk-nuclei are homologous to the primitive endoderm in the gastrulation of the majority of animals, as has been determined by Heymons (*95, *97) and other authors, the stage of development in the ontogeny of Melophagus in which the blastoderm is completed, as represented by Figure 7, would be the gastrula-stage. The next step in the development is the formation of the germinal plate and the median mesodermal band. The cells on the concave (dorsal) side of the ^^'g diminish somewhat in height (PI. 2, Fig. 8); those on the convex (ventral) side rapidly proliferate along the median line ; a slight depression appears in the blastodeim along the mid- ventral line, on the inner surface of which a ridge of cells, the primitive mesoderm {cr8,ms\lrm.)^ is raised, projecting into the yolk. The ectodermic cells of this ventral region become elongated, and with the mesoderraic cells constitute the germinal plate. The plate does not, however, confine itself to this portion of the egg, but, as is common in the Diptera, quickly extends itself to portions of the dorsal side, the thickened ectodermic portion encircling both poles of the egg and occupying about a third of the dorsal surface at each end (PI. 2, Fig. 9). The primitive mesodermal plate also extends Digitized by VjOOQ IC 256 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. beneath these dorsal ectodermic thickenings, but is not formed at the same time at the two poles of the egg. At the anterior end it extends dorsally more rapidly than at the posterior portions (Fig. 9) . I found no trace of the lateral gastrular invaginations which Graber ('89) describes in Calliphora as accompanying the median invagi- nation, and which, as Korschelt and Ileider ('92, p. 812) have remarked, undoubtedly mark the edges of the germinal plate. The germinal plate, as in other Diptera, is superficial in position and unprotected by embryonic envelopes, the amnion being repre- sented by only the merest rudiments at the anterior and posterior extremities of the germinal plate (PI. 2, Fig. 9, am.), on the dorsal surface of the egg. These rudiments mark the limits of the head- fold and the tail-fold, the former being the dorsal portion of the germinal plate at the anterior end of the egg {pli, ce,) , and the latter the dorsal portion of the germinal plate at the posterior end of the egg {pli, ca.) . Even these rudiments of an amnion disappear in the stages immediately following. Figure 9 represents a sagittal section of an embryo somewhat older than that shown in Figu^:e 8. The anterior end of the e^'g is marked by the presence of the micropyle (riir^jyy,) . The head- fold (pli, ce.) extends as far as the anterior amniotic ru^liment and contains mesodenn, whereas the tail-fold (pli, ca.), which extends to the posterior amniotic rudiment, consists as yet exclusively of a thickened ectodermic plate. The proctodeum and stomodeum both make their appearance as ectodermal depressions on the dorsal poition of the germinal plate. In common with Musca, as observed by Graber ('89) and Voeltzkow (*89) , the proctodeum appears first. Figure 10 (PI. 2) represents a sagittal section of an embryo somewhat older than the one previ- ously mentioned; in it the proctodeum has the form of a deep invagination in the tail-fold, while the stomodeum has not yet appeared. The point in the head-fold where the stomodeal invagi- nation will appear is, however, plainly shown in the thickening near the end of the fold (eras, stmd,). It will also be noticed in this embryo that the primitive mesoderm has extended beneath the tail-fold, that the posterior amniotic rudiment has disappeared, and that the germinal plate has begun to show traces of segmentation. Tracheal invaginations have not yet made their appearance. Figure 12 (PI. 2) shows a stage still older than the one just discussed, in which both proctodeum and stomodeum are present. Digitized by VjOOQ IC PRATT; IMAGINAL DISCS. 257 The proctodeal invagination {prd.) has become much deeper, and at its inner end the beginnings of the Malpighian tubules have made their appearance (tb, mpg,). The stomodeal invagination (stnuL) forms a deep depression in the head-fold a third of a milli- meter from the anterior end of the egg, A surface view of the dorsal aspect of an egg in this stage of development is represented in Figure 11 (PI. 2), and one of the lateral aspect, in Figure 13. In both these figures the head-fold (pit. ce,) is seen to be sharply marked off from the remainder of the dorsal integument by the presence of the rudiment of the amnion (am). The tail-fold is not marked off with the same distinctness for two reasons : the posterior rudiment of the amnion, which appeared in Figure 9, has now dis- appeared ; moreover the fold itself does not occupy as much of the dorsal surface of the egg as it did in the previous stages, as is shown by the position of the proctodeum nearer the posterior pole of the The primitive mesoderm also shows a marked advance in develop- ment in eggs of this age. The band or ridge of mesoderm which took its origin in the mid-ventral line of the germinal plate has sep- arated into two lateral bands which, as in other insects, occupy positions to the right and left of the mid-ventral line. Figure 13 shows these bands (tae. 7n8\lrm) as they appear in a lateral view of the embryo ; Figure 15 {tae. ms'drm) /in a cross-section. It will be seen that the mesodermal cells have entirely abandoned the mid- ventral portion of the egg, a 8[)ace being left between the yolk and the ectoderm which represents the fundament of the coelom (coel.). The formation of these lateral mesodermal bands, as just described, takes place, however, only in the middle portions of the egg. At the two ends of the egg and in the head- and tail-folds the meso- derm does not entirely abandon the median portions of the egg, although it expands into the lateral portions of it. The mesoderm of the ventral side is continuous at the ends of the egg with that of the head- and tail-folds. Thus at each end of the egg there is a continuous layer of mesoderm lining the entire inner surface of the ectoderm (Fig. 12), instead of two lateral bands, such as are found in the middle portions of the egg. In embryos of this stage segmentation is distinctly indicated in the ventral portions of the germinal plate, ten or eleven segments being represented (PI. 2, Figs. 12, 13). The tracheal invagina- tions also appear (Fig. 13), there being eleven pairs of them, of Digitized by VjOOQ IC 258 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. which the first two pairs are near the anterior pole of the egg. The endoderm has now made its appearance and is in process of surround- ing the yolk. As has been demonstrated by Voeltzkow (*89) and Graber ('89) in Musca, by Heymons ('95) in Forficula and various orthopterous insects, and by L^^caillon ('96) in Coleoptera, so in Melophagus the endoderm is a derivative of the inner ends of the stomodeal and proctodeal invaginations. The first indication of it is a proliferation of cells at the inner end of the proctodeum. This extends itself forward, in the form of a single layer of epithelium, along the dorsal surface of the yolk (the proctodeum having a dorsal position) to the head-fold (Fig. 12), and also around the sides of the yolk toward the ventral side of the egg (Fig. 15, en^lrm,). As is to be seen in Figure 10 (prd,)^ the proctodeal invagination, when it first appears, is bounded on its inner end by the mesodermic layer of the tail-fold. The boundary between mesoderm and ectoderm in this region is always perfectly easy to determine, because of the very different character of the cells of the two germ-layers. As the proctodeal invagination increases in depth and the fundaments of the Malpighian tubules begin to appear, the mesodermic layer gradually becomes thmner, until it entirely disappears. The cells of the anterior (deep) portion of the ectodermic invagination then proliferate rapidly and give rise to endoderm (Fig. 12, en\lrm.). The proctodeum, since it appears earlier than the stomodeum, apparently gives rise to the greater part of the endoderm, — an observation which was also made by Graber ('89) on Musea. In the stage represented by Figure 12 the endodermal epithelium has extended forward from the proctodenn to the head -fold, and lies beneath the stomodeum, but without having fused with it. I do not believe, however, that the stomodeum fails to cooperate in the production of the endoderm. As shown in the figure last men- tioned, the stomodeum has just pushed its way through the meso- dermic layer, which at first bounds its inner surface ; in the stages succeeding this, as will be shown very soon, this inner surface fuses with the endoderm and apparently aids in its formation. No trace of coelomic sacs appears in the mesodertn at any time. Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 259 The Origin of the Imaginal Discs. 1. The Cephalic Discs: — a. The Early Developnient of these Discs. — Soon after the stage of development represented by Figures 11-13 (PI. 2), but before the lateral mesodermal bands have extended to the dorsal side of the embryo and joined each other in the mid-dorsal line, and before the endoderm has extended around from the dorsal to the ventral side of the embryo (Fig. 15, e7i^drr)i), those imaginal discs which are destined to form the imaginal head have made their appearance. Three crescentic thickenings of the ectoderm appear in the vicinity of and partly encircling the stomo- deum, a median one in front of the stomodeum, and two paired ones behind it (PL 3, Fig. 17). The median thickening (dsc, ce. m,) is destined to form the ventral cephalic disc, which during the meta- morphosis develops into the ventral portions of the imaginal head together with the mouth-parts ; the paired thickenings (dsc, ce,) form the dorsal cephalic discs, which are homologous with the cephalic discs of the Muscidae, as described by Weismann, Van Rees, and others. These develop during the metamorphosis into the dorsal and lateral portions of the imaginal head. The median disc has no homologue in the Muscidae. There is no period in the development of the embryo of Melopha- gus when a distinct head is present, although it seems probable that the head-fold, since it forms the forward end of the germinal band, represents the head. It never shows, however, the slightest trace of segmentation. In Musca, according to Weismann ('64) the head is, at a certain stage of the embryo's development, quite as distinctly segmented as the trunk. As will be seen in Weismann's Figure 71, the embryo is divided into fourteen or fifteen segments, of which eleven belong to the body, and the remainder to the head. The anterior portion of the germinal band does not as in Melophagus, extend onto the dorsal side of the egg, the stomodeum heing sub- terminal. As the development of the embryonic fly advances, the cephalic portion gradually becomes reduced in size relatively, and its segmentation becomes less distinct. When the young larva is finally born its head has been reduced to a mere rudiment at the forward end of the body, without a trace of segmentation, and is smaller than any of the body segments. Consequently, of the twelve segments composing the fly larva the first alone represents the head. Digitized by VjOOQ IC 260 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. In the development of the embryonic Melophagus the head-fold has a history similar to that of these cephalic segments of Mosca. As the embryo advances in its development the head-fold diminishes in size, and the stomodeum, which first appeared at some distance from the anterior end of the egg on the head-fold, migrates forward towards its definitive position at the anterior end, probably as the result of a longitudinal concentration or shortening of the germ- band. The history of the tail-fold is the same. It also becomes reduced in size as the proctodeum migrates towards the posterior end of the embryo. This migration b much more rapid than that of the head-fold. Thus, when the young Melophagus larva is finally born, it resembles externally in a marked degree the new- bom fly larva. It consists of twelve segments, of which eleven belong to the body, and one, which represents the rudiment of the head-fold, to the head ; the head- and tail-folds have disappeared, the germ-band having entirely retreated from the dorsal side of the embryo. The further development of the cephalic discs goes on band in hand with this migration of the stomodeum towards the anterior pole of the egg. The paired discs develop much more rapidly than the median one. The latter remains a mere thickening of the ecto- derm, without showing signs of invagination, during the entire migration of the stomodeum, so that when this organ has reached its final position at the anterior end of the egg the median disc has simply changed its position. It is now an ectodermal thicken- ing on the ventral side of the embryo just beneath the mouth opening (PI. 3, Fig. 22, and PI. 4, Fig. 29, dac. ce. m.). The paired discs, on the contrary, early in the course of their movement forward begin to invaginate. The convex margin of each of the crescentic thickenings becomes much thicker than the concave, and along this outer (convex) margin a groove-like invagi- nation is formed (PI. 2, Fig. 14, dsc. ce.). At the stage of devel- opment represented by Figure 14, the right and left halves of the germinal band in the region of the body have not proceeded more than half way towards the dorsal side of the eggy as will be seen in Figure 15 (tae, ms'dmi)^ which represents a cross-section through the middle of the same embryo as the one shown in Fig- ure 14. The endoderm has not yet extended to the ventral side of the yolk (Figs. 14, 15, en'drm,)^ the tracheal invaginations form deep sacs (Figs. 14, 16, i'vag, tr,), and the nervous system has not yet made its appearance. Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 261 An embryo which has advanced somewhat beyond the stage just discussed is illustrated in Figures 18-21 (PI. 3), which repre- sent sections through a single animal. The stomodeum (Fig. 18, 8tmd,) has become,a deep invagination, which is situated on the dor- sal side of the embryo, 0.16 mm. from its anterior end. It has thus moved 0.17 mm. nearer the anterior end than it was at the stage represented in Figure 12. The inner end of the invagination has now fused with the endoderm. The forward ends of the paired cephalic discs appear in this section (Fig. 18, dec, ce,). At a distance of 0.08 mm. back of this section is the section represented by Figure 19, which shows these discs where the invaginations are deepest. The invaginations are here long, curved, deep slits, one on each side of the stomodeum, which have a diagonal position in the head- fold, the median end of each being directed posteriorly. It will also be obsei-ved that in this embryo the endoderm has completely enclosed the yolk. At the anterior end of the embryo it has fused with the stomodeum (Fig. 18), while at the posterior end it is seen to be continuous with the proctodeum (Fig. 21, en'drm.). The nerv- ous system has made its appearance in the form of a pair of longi- tudinal cords of cells on the ventral wall of the body, one on each side of a mid-ventral ectodermal ridge (PI. 3, Figs. 18-20, n, v.). In the head-fold a longitudinal nerve-cord makes its appearance immediately beneath each cephalic invagination and on the median side of the slit-like invagination (Figs. 18, 19, g7i, cb.) ; these con- nect with the ventral nerve-cord at the forward end of the embryo. These nervous fundaments in the head-fold are undoubtedly the beginnings of the cerebral ganglia. This early establishment of a relation between the brain of the animal and the paired cephalic discs is important, since the relation becomes more intimate as development advances. In the larva and pupa the cephalic discs are so firmly joined to the cerebral ganglia that Weismann ('64), who first discovered the corresponding discs in Musca, gave them the name " Hirnanhange." A pair of large spherical bodies, apparently of nervous tissue, appears in the head-fold at this stage of development (PI. 3, Fig. 18, ffn, ala.) at the right and left of the stomodeum, and near the for- ward ends of the cephalic invaginations. The origin of these very noticeable bodies I did not observe ; but they arose between the stages represented by Figures 14 and 18, and probably from the ectoderm. They are at this stage quite unconnected with any other Digitized by VjOOQ IC 262 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. organs. It is very difficult to identify them satbfactorily, but ap- parently they belong to the pharyngeal (sympathetic) ganglionic system, and I have called them ganylia allata (see Heymons, 'dS, p. 49), inasmuch as they do not arise from the dorsal stomodeal wall, but probably from the outer ectoderm, whence they migrate dorsally to the position they finally occupy. In other insects in which these or at least similar ganglia appear, they originate as thickenings or invaginations of the ventral body-wall in the neighborhood of the maxillae, and migrate dorsally either to fuse with each other in the mid-dorsal line just above the stomodeum (Forficula) or to remain separate (Gryllus). In Melophagus, since the head-fold and the stomodeum occupy the dorsal portion of the germinal band, these ganglia probably do not have a ventral, but rather a lateral, origm (PI. 3, Fig. 18, gn, ala.) , But, as will be seen in subsequent stages, they migrate to a position immediately beneath the dorsal body- wall, fuse together, and enter into relations with important pharyn- geal organs. No neuroblasts appear in any part of the nervous system at this early stage of development, the cells being spherical bodies of nearly equal size. The germinal band has not, in the stage just discussed, grown over the dorsal side of the animal (except, of course, on the head- and tail- folds) to complete the formation of the back. In the next stage the stomodeum (PL 3, Fig. 22, atmd,) has reached the forward pole of the body. The cephalic discs have also changed their positions and migrated farther foi-ward. The median disc is now situated immediately ventrad of the mouth opening {dsc. ce. 7n.) ; the paired discs have moved still farther forward, and are now 0.15 mm. from the anterior end of the animal ; but their essential character has not changed. Figure 23 (PI. 4) represents a parasagittal section of the same embryo of which Figure 22 represents a slightly oblique sagittal section. It, together with Figure 26, — which represents a cross-section of an embryo of the same age, — shows the relation of the paired discs to the surrounding organs. These discs (dsc. ce.) are seen to be no longer dorsal to the embryonic intestine, as in the previous stages, but to have moved to a position in front of it. The nervous system has developed considerably. The paired nerve-cords (n. v.) have fused with the mid- ventral ectodermic ridge, and neuroblasts appear throughout their entire extent. The Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 263 paired cerebral ganglionic fundaments {gn, cb,) beneath the paired discs have increased in size. Figure 23 shows their connection with the ventral cords. The ganglia allata ([/n, ala.) have moved to the dorsal side of the embryo, where their anterior ends have fused with each other, and occupy a median position (PI. 3, Fig. 22, and PI. 4, Fig. 26, gn, ala.)^ while their paired posterior portions lie back of the paired discs (Fig. 28, gn. ala.), A still better idea of the position of these bodies is given in Figures 43 and 44 (PI. 7), although these represent sections of a much older embryo, Figure 44 (gn. ala.) showing the posterior portions of the ganglia, and Figure 43 (g7^. ala.) their juncture. Their anterior fused portion, even in the earlier stage (PI. 3, Fig. 22, gn, ala.) , has much elon- gated, and extends forward to a mass of mesodermic tissue, dorsal to the forward portion of the stomodeum, which is fast developing into a group of muscle-libres (Ing.), whose later history will be found to be interesting. In the dorsal wall of the stomodeum is to be seen a slight evagination (gn./., PL 3, Fig. 22; PI. 4, Fig. 26), which I take to be the fundament of the ganglion frontale. It is a very transitory structure, soon disappearing without leaving a trace in the later history of the animal. Segmentation has become much more strongly marked in the forward portion of the embryo, but has almost disappeared from the hinder portion of it. In Figure 22, which represents a somewhat oblique sagittal section, we see three deep grooves in the ventral ectoderm, which mark the boundaries between head, prothorax, mesothorax, and metathorax, respectively. Two following shallow grooves mark the boundaries of the metathorax, the first, and the second abdominal segments respectively. The arrival of the stomodeum at the anterior end has transformed that part of the body of the embryo. The mouth is a transverse, slit-like opening (PI. 4, Figs. 25 and 21jStnid.). Its ventral lip is the median cephalic disc, which, however, is still only a thick- ened portion of the ectoderm. Its dorsal lip is a conical struc- ture, and projects freely forward above it (PI. 3, Fig. 22 ; PI. 4, Fig. 27, big.). This structure is of great importance in the lar- val life of the insect, for it acts as a sucking tongue by means of which the animal ingests its milk-like food. (For a description of it in the larva, see Pratt, '93.) When it first appears, this organ, as will be seen in Figure 22, forms the dorsal wall of the anterior end of the stomodeum, and its anterior tip projects beyond the Digitized by VjOOQ IC 264 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. remainder of the body. These relations are also to be seen in figure 27 and in the cross-sections shown in Figures 24 and 25, Fig- ure 24 being a section through the extreme anterior end of the ani- mal, in which the ectoderm is the only tissue cut, and Figure 25 being a section just back of it. The sucking tongue is composed of two distinct tissues, an outer epithelium and /an inner muscular core (Figs. 22, 25, Ing.), At the stage represented by these figures the muscle-fibres are just beginning to differentiate. The hinder sur- face of this muscle mass is in direct contact with the fused median portions of the ganglia allata (Fig. 22). Figures 27 and 28 repre- sent surface views of an embryo in this stage of development. It will be noticed in Figure 22 that a communication between the intestine and the proctodeum is established. The growth of the proctodeum has thus advanced faster than that of the stomodeum, as no communication has yet developed between the stomodeum and the forward end of the intestine. The opening between the mid- intestine and the end-intestine is, however, closed again before the birth of the larva and remains so during the larval life of the animal, a peculiarity of structure which the larva of Melophagus shares with that of the honey-bee. In the animal represented by these figures, the germinal band has extended over the dorsal side, and the back is thus closed. The heart is present in the form of a delicate tube extending from the ganglia allata to the hinder end of the body. In embryos in this stage of development (and not in any other) a pair of deep invaginations appears in the outer ectoderm below the mouth (PI. 4, Figs. 24, 25 and 27, gl, saL); these are probably fundaments of the salivary glands. They are entirely transitory structures. b. The Involution of tlie Head. — An important change now takes place in the develoj)ment of the for\Yard end of the body. The paired discs, which hitherto have consisted of two ectodermic thickenings embracing each a simple, diagonally placed slit (PI. 4, Fig. 26, else, ce.), now move forward and towards the median line, where the slits finally meet and unite, forming thus a single, trans- verse, slit-like opening which extends nearly across the embryo. It* is only the upper or ectal portions of the invaginations, however, which thus fuse ; their inner or ental portions increase very much in volume and depth, though remaining free from each other, and extend posteriorly in the body-cavity to the vicinity of the Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 265 cerebral ganglia, with which they lie in contact. The paired discs have thus become a single structure with a single, median, dorsal, anterior opening, backward from which extends an irregu- larly Y-shaped invagination. Figure 30 (PI. 5), which represents a sagittal section, shows the median opening and the median por- tion of the invagination. Figure 31, which represents a parasag- ittal section, shows one of the branches of the Y-shaped invagina- tion (dec. ce.) ; and Figure 32, a parasagittal section laterad of the one last mentioned, shows the irregular structure of the disc. The ventral median disc (Fig. 30, dsc. ce. m.) has not changed its position or character, except to become much thicker. A comparison of the structures of the embryo represented in Figure 22 (PI. 3) with those of the one represented in Figure 29 (PI. 4) , which is the same as that of the last three figures dis- cussed, shows that the development in the latter has been consid- erable. The muscle fibres of the sucking tongue are distinctly developed, but the nerve which proceeds from its base to the ganglia allata, and, also, these ganglia themselves, have not changed their character. The paired ventral nerve-cords and the cerebral gan- glionic fundaments (Fig. 31) have developed considerably, the former having fused with the median ectodermic ridge, which has now effected a separation from the ectoderm, and thus come to form with the paired nerves a single structure. Neuroblasts are present throughout the entire extent of both ventral and cerebral nerve-masses, but are not present in the ganglia allata nor in the median nerve proceeding from it. It will also be noticed that the anus has shifted its position from the hinder end of the animal ' to the ventral side near the hinder end of the raid-intestine, the position it occupies in the larva. A communication has also appeared between the storaodeum and the intestine (Fig. 30) . The involution of the head of the embryo now takes place. An ectodermic fold starts back of the cephalic discs, both dorsally and ventrally (PI. 5, Fig. 30), and grows rapidly forward towards and over the mouth. The mouth, together with the ventral disc {dsc, ce. m.) just below it, the muscular tongue {Ing.)^ and the common opening of the dorsal discs (of. m.) just above it, is rolled in by this process. A new mouth is thus formed (PI. 6, Fig. 34, or,), and back of it a new portion of the digestive tract (phy.)^ the so- called pharynx of Weismann and Van Rees, described by them in the muscidian larva. Digitized by VjOOQ IC 266 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. The cephalic imaginal discs now grow rapidly, and soon attain the form which characterizes them in the young larva. The median, ventral disc has begun to invaginate (PL 6, Fig. 34, dsc, ce. m.), and a cross-section through this region (PL 6, Figs. 36 and 37, dsc, ce, m,) shows that the invagination is paired. (Compare these cross- sections with Figure 34, where the positions of the cross-sections shown in Figures 36-44 are indicated by the corresponding numbers, 36-44.) The later history of this* disc (see Pratt, '93) shows that these paired outgrowths of the ventral invagination develop during the larval period into deep pockets, that a long projection springs from the bottom of each of them, and that subsequently the pockets fuse in the median plane and the projections develop into the proboscis of the imago. The muscular sucking tongue is now an internal organ (as may be seen in Figures 34, 37 and 38), and projects free into the pharynx. Figure 39 shows the base of the sucking tongue where it is continuous on each side with the pharyngeal wall. The open space dorsal to it is the forward end of the median portion of the lumen of the dorsal discs (of, m.), where it communicates with the pharynx. In the stage of development represented by Figure 30 (PL 5) , this opening (of, in.) leads to the outside of the animal's body. The paired nervous masses below the oesophagus are portions of the circum-oesophageal or cerebral commissures (Fig. 39, corns, ch) . The section shown in Figure 40 (PL 7) also passes through the base of the sucking tongue. The median portion of the fused dor- sal discs (dsc, ce,) is much broader here, extending quite across the embryonic body, while between it and the oesophagus {oe.) are the muscle-mass of the tongue {lug,) and the cerebral nerve- mass {(/n, cb.). The latter is in close contact with the ventral sur- face of the disc, and with the tongue muscles quite encloses the oesophagus. Figure 41 represents a section made immediately back of the sucking tongue. The median portion of the fused dorsal discs {dsc, ce,) is still shown, and between it and the oesophagus are the median nerve {n, m,) — which joins the sucking tongue with the . ganglia allata — and the two cerebral nerve-masses {(/n, cb,). Fig- ure 42 shows a section considerably farther back (compare Fig. 34) , passing through the paired portions of the dorsal discs (dsc, ce.), and also exhibiting the structures seen in the previous section. The next figure (Fig. 43) shows the paired portions of the dorsal discs {dsc. ce.) distinct from each other, though still in contact with the Digitized by VjOOQ IC PR AIT: IMAGINAL DISCS. 267 cerebral nerve-masses {gn, cb.), as is also shown in the parasagittal section (PI. 5, Fig. 33) . The ganglia allata (^n. ala,) and their fusion to form the median nerve are shown in Figure 43. The forward end of the heart (cr,) is also cut. Figure 44 represents a section posterior to the dorsal discs and through the hinder part of the ganglia allata. Mesoderm does not tnake its appearance in the cephalic discs, nor do nerves or tracheae enter them. The process of the involution of the embryonic head of Melopha- gusj by which the ventral cephalic disc, the sucking tongue, and the median opening of the dorsal cephalic discs, are changed from external to internal organs, and the head of the animal telescoped into the thorax, is really the final act of a longer operation. The head of the embryo, which, in my opinion, is represented by the head-fold of the germinal band, begins to disappear when the stomo- deum begins its migration towards the anterior end of the embryo. The involution of the head has been observed in only one other representative of the brachycerous Diptera besides Melophagus, viz. in Musca vomitoria by Weismann ('64) . Here the disappear- ance of the head is also a gradual process. The embryonic head gi*adually grows shorter and at the same time loses its segmenta- tion, its posterior edge moves forward until finally its anterior portion invaginates into the mouth, forming the so-called larval pharynx. The dorsal cephalic discs then appear as appendages of the dorsal pharyngeal wall. But Weismann did not observe the method by which these structures originate. 2. 27ie Thoracic Discs. — Six pairs of thoracic imaginal discs make their appearance in the embryo of Melophagus. Three pairs* are dorsal and three pairs ventral, a dorsal and a ventral pair belonging to each of the pro-, meso-, and metathoracic segments, respectively. The three ventral pairs give rise during metamor- phosis to the three pairs of imaginal legs ; the dorsal metathoracic pair to the rudiments of the balancers, and the dorsal meso- and pro thoracic pairs to no adult structiu^es, they being rudimentary structures. All of these discs first appear late in the embryonic life of the insect; namely, at about the time of the involution of the head, — at a time when the foimation of the dorsal cephalic discs is prac- tically complete, when the back of the embryo is closed, and the wall of the intestinal tract is formed. The earliest appearance of the thoracic discs figured is that shown for the embryo represented in Figure 32 (PI. 5), in which the Digitized by VjOOQ IC 2r)8 PROCEEDINGS: BOSTON SOCIETY NATUR.\X HISTORY. involution of the head is in progress. Three pairs of thickenings of nearly equal size appear in the ventral ectoderm of the forward end of this embryo. In the next stage shown (PL 6, Fig. 35), in which the involution of the head is completed, these thickenings have begun to invaginate, and three additional pairs of thickenings have made their appearance in the dorsal ectoderm at the forward end of the embryo. Only the dorsal j)rothoracic tliickening ap- pears in Figure 35 ; it lies immediately in front of the dotted line leading from the letters dsc, ce. The invagination of each ventral oisc begins at its posterior border, as shown in the ventral meta- thoracic disc in Figure 35 ; then the anterior border sinks in, as shown in the mesothoracic disc in this figure ; finally the entire disc sinks beneath the surface, as is shown in the prothoracic disc in the same figure. As this figure clearly indicates, the invagination of the three discs is not simultaneous, but the more anterior the disc the earlier the invagination. Complete invagination rapidly follows (PI. 7, Fig. 45) ; the disc at once separates itself from the ecto- derm, and the opening made in the ectoderm by the invagination closes (PL 7, Fig. 46) . It is while the discs are in this condition that the embryo leaves the egg-envelopes, and the discs remain in this condition during the entire larval life of the insect. The dorsal thoracic discs do not invaginate during the embryonic life of the animal. The meso- and metathoracic discs invaginate in the young larva, and become detached from the ectoderm, as do the ventral discs (PL 7, Fig. 40). The prothoracic discs do not advance beyond the stage of development represented by figure 45, but remain thick-walled pockets of the ectoderm. They are also lined with a cuticula, continuous with that of the rest of the larval ectoderm, and the different moultings take place from the pockets as from the rest of the insect. In my preliminary paper (Pratt, '97) I have stated that the dorsal prothoracic discs have a larval origin, but later examination of the material has shown this to be an error. As is the case of the cephalic discs, mesoderm does not appear in the thoracic discs during the embryonic life of the insect, nor do nerves or tracheae enter them. 3. The Discs of the External Genitalia. — These discs first appear in the embryo shortly before it leaves the egg-membranes. Two pairs of ectodermal thickenings then appear immediately in front of the anus, — a larger posterior and a smaller anterior pair. The former lie just in front of, and partly embracing, Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 269 the anus and rectum (PL 2, Fig. 48, dsc, gen, p,) , and the latter inimediately in front of it. These two pairs of discs do not make their appearance simul- taneously, the larger, posterior pair being the first to appear. These have the form of two invaginations, one on each side of the anus (Fig. 48), which project forward and fuse on the ventral (an- terior) side of the end-intestine, forming a single flattened sac, which partly encircles it (Fig. 49, dsc, gen, p,) , At its anterior end this sac separates into two parts, being in this region again paired (Fig. 50, dec, gn, p.) . It is in the outer ectoderm at this point that the anterior discs finally make their appearance in the form of a pair of ectodermal thickenings (Fig. 50, dec. gn. a,) , These do not, however, develop further in the embryo ; but in the larva they assume a sac-like form and are detached from the ectoderm. A description of the development of these discs in the larva has already been published (Pratt, '93) . 4. The Discs of the Internal Organs, — The imaginal discs of the internal organs and of the abdominal hypodermis do not appear during the embryonic life of the insect. Methods. The greater part of this investigation was carried on in the Zoological Laboratory of Harvard University under the direction of Prof. E. L. Mark. It was completed at Haverford College. Most of the material was obtained in Germany, but a portion of it came from the vicinity of Cambridge, Mass:, and of Haverford, Pa. The eggs were obtained by dissection from the maternal uterus, in which they develop. It was found to be impossible to remove them in the fresh condition without injury, on account of the extreme delicacy of the chorion. Consequently, in each case the mother insect was killed by decapitation ; the abdomen was slit open and then plunged into a warm corrosive-sublimate solution. This fixed the %g^y and at the same time hardened it so that it could be removed from the uterus. Most of the staining was done with borax carmine, which was found sufficient for all ordinary purposes, but Mayer's acid carmine and Ehrlich's haematoxylin also were used. Digitized by VjOOQ IC 270 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. BIBLIOGRAPHY. Agassiz, L. '51. The classification of insects from embryological data. Smithsonian contributions, etc., vol. 2, art. 6, pp. 1-28, pi. 1. Ayers, H. '84. On the development of Oecanthus niveus and its parasite Teleas. Mem. Boston soc. nat. hist., vol. 3, no. 8, pp. 225-281, pis. 18-26. Balfour, F. M. '80. A treatise on comparative embryology. London. Vol. 1. Blochmann, F. '87. Ueber die Richtungskorper bei Insecteneiem. Morph. Jahrb., Bd. 12, pp. 644-674, Taf. 20-27. ' Bobretzky, N. '78. Ueber die Bildung des Blastoderms und der Keimbl&tter bei Inaecten. Zeitschr. f. wiss. Zool., Bd. 31, pp. 196-216, Taf. 14. Brandt, A. '69. BeitrAge zur Entwicklungsgeschichte der Libelluliden und Hemipteren. M6m. acad. St. Petersbourg, torn. 13, no. 7, pp. 1-33, pl^. 1-3. Brauer, F. '85. Systematisch-zoologische Studien. K. Akad. Wiss. Wien, Sitzungsber. math.-natui'w. CI., Bd. 91, pp. 237-418, Taf. 1. Bugnion, E. '91. Recherches sur le developpement postembryonnaire, I'anatomie et les moeurs de TEncyrtus fuscicoUis. Recueil zool. Suisse, tom. 6, pp. 435-634, pis. 20-26. Burmeister, H. '32-'47. Handbuch der Entomologie. Berlin. Dewitz, H. '78. Beitrflge zur Kenntniss der postembryonalen Gliedmaassenbildung bei den Insecten. Zeitschr. f. wiss. Zool., Bd. 30, Suppl., pp. 78-105, Taf. 5. '81. Ueber die Fltigelbildung bei Phryganiden und Lepidopteren. Berl. ent. Zeitschr., Bd. 26, pp. 63-66, Taf. 3-4. Ganin, M. '76. [The postembryonic development of insects.] (Russian.) Warsaw. Reviewed in Zeitschr. f. wiss. Zool., Bd. 28, 1877, pp. 386-389. Gonin, J. '94. Recherches sur la metamorphose des l^pidopt^res. Bull. soc. vau- doise d. sci. nat., tom. 30, pp. 1-62, pis. 1-6. Graber, V. '77-'79. Die Insekten. Zweiter Theil. Mttnchen. '89. Vergleichende Studien tiber die Embryologie der Insecten und besond- ers der Musciden. Denkschr. k. Akad. Wiss. Wien. math.-naturw. CI., Bd. 66, pp. 267-314. Taf. 1-10. Grassi, B. '84. Intomo alio sviluppo delle api nelP uovo. Atti accad. gloenia scL nat. Catania, vol. 18, pp. 146-222, Tav. 1-10. Digitized by VjOOQ IC PRATT: IMAGINAL DISCS. 271 Heider, K. '89. Die Embryonalentwicklung von Hydrophilus piceus L. Jena. 98 pp., 13Taf. Henking, H. '88. Die ersten Entwicklungsvorgftnge im Fliegenei und freie Kernbild- ung. Zeitsclir. f. wiss. Zool., Bd. 46, pp. 289-336, Taf. 23-26. Herold, M. '15. Entwicklungsgeschiclite der Schmetterllnge anatomisch und physiolo giscli bearbeitet. Cassel und Marburg. 118 + 34 pp., 33 Taf. Heymons, R. *95. Die Embryonaleutwickelung von Dermapteren und Orthopteren, etc. Jena. 136 pp., 12 Taf. '97. Ueber die Zusammeusetzung der Insecteu-Kopf. Sitzungsber. Ges. naturf. Fr. Berlin, no. 7, pp. 119-123. Korotneff, A. '85. Die Embryologie der Gryllotalpa, Zeitschr. f. wiss. Zool., Bd. 41, pp. 670-604, Taf. 29-31. Korschelt, E., & Heider, K. '92. Lehrbuch der vergleichenden Entwicklungsgeschichte der wirbellosen Thiere. Jena*. Kowalevsky, A. '85. Beitrftge zur nachembryonaleu Entwicklung der Musciden. Zool. Anz., Jahrg. 8, pp. 98-103, 123-128, 153-157. '86. Zur embryonalen Entwicklung der Musciden. Biol. Centralbl., Bd. 6, pp. 49-64. '87. Beitrftge zur Kenntniss der nachembryonaleu Entwicklung der Mus- ciden. Theil 1. Zeitschr. f. wiss. Zool., Bd. 45, pp. 642-694, Taf. 26-30. Ktinckel d' Ilerculais, J. '75. Recherches sur 1' organisation et le dfiveloppement des volucelles. Pt. 1. Paris. 208 pp. 12 pis. Landois, H. '71. Beitrftge zur Entwicklungsgeschichte der SchmetterlingsflUgel in der Raupe und Puppe. Zeitschr. f. wiss. Zool., Bd. 21, pp. 306-324, Taf. 23. L6caillon, A. '98. Recherches sur I'oeuf et sur le dfeveloppenient embryonnaire de quelques Chrysom61idae. Theses pr6sent6es ft la Faculty des Sciences de Paris, 86r. A, no. 299. Paris. 219 pp., 6. pis. Leuckart, R. '58. Die Fortpflanzung und Entwicklung der Pupiparen. Abh. der naturf. Ges. iu Halle, Bd. 4, pp. 1-81, Taf. 1-3. Lyonnet, P. 1760. Traits anatomique de la chenille qui ronge de saule. La Haye. Metschinkoff, E. *83. Untersuchungen tiber die intracellulftre Verdauung bei wirbellosen Thieren. Arb. zool. Inst. Wien, Bd. 6, Heft 2. pp. 141-168, Taf. 13, 14. Mliggenburg, H. '92. Der Rtissel der Diptera pupipara. Arch. f. Naturg., Jahrg. 58, Bd. 1, pp. 287-332, Taf. 15, 16. Digitized by VjOOQ IC 272 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Pancritius, P. '84. Beitrftge zur Kenntniss der FltiKelentwickelung bei den Insecten. Inaug.-Dissertation. Konigsberg. 37 pp., 2 Taf. Pratt, H. S. '93. Beitrftge zur Kenntniss der Pupiparen. Arch. f. Naturg., Jahrg. 59, Bd. l,pp. 151-200, Taf. 6. '97. ImiiginaJ Discs in Insects. Psyche, vol. 8, pp. 15-30. Rees, J. van. '84. Over intra-cellulaire spijsverteering en over de beteekenis der witte bloedlichaampjes. Maandblad voor natuurwetenschappen, Amsterdam, Jaarg. 11, no. 6-8, 28 pp. '88. Beitrfige zur Kenntniss der inneren Metamorphose von Musca vomi- toria. Zool. Jahrb., Abth. f. Anat. u. Ontog., Bd. 3, pp. 1-134, Taf. 1-2. Swammerdam, J. 1737-38. Bybel der natuure, etc. Leydae. Verson, E. '90. Der SchmetterljnKsfltif^el und die sogeuannte Imaginalscheibe der- selben. Zool. Anz., Jahrg. 13, pp. 116-117. Viallanes, H. '82. Recherches sur I'histologie des Insectes et sur les phfinomfenes hlstol- ogiques qui accompagnent le dfeveloppement post-embryonnaire de ces animaux. Ann. sci. uat., zool., s§r. 6, torn. 14, pp. 1-348, pis. 1-18. Voeltzkow, P. A. '89. Entwicklung im Ei von Musca vomitoria. Arb. zool.-zoot. Inst. WUrzburg, Bd. 9, pp. 1-48, Taf. 1-5. Wahl, B. '99. Ueber das Tracheen-system und die Imaginalscheiben der Larve von Eristalis tenax L. Arb. zool. Inst. Wien, Bd. 12, Heft 1, pp. 46-98. Taf. 4-8. Weisraann, A. '63. Ueber die Entstehung des vollendeten Insectes in Larve und Puppe. Abh. der senckenb. natiirf. Ges., Bd. 4, pp. 227-260, Taf. 10-12. '64. Die nachembryonale Entwicklung der Musciden nach Beobachtungen an Musca vomitoria und Sarcophaga carnaria. Zeitschr. f. wiss. Zool., Bd. 14, Heft 3, pp. 187-336, Taf. 21-27. '66. Die Metamorphose von Corethra plumicoruis. Zeitschr. f. wiss. Zool., Bd. 16, pp. 46-127, Taf. 3-7. '82. Beitrfige zur Kenntniss der ersten Entwicklungsvorgfinge im Insectenei, Beitrfi,ge zur Anatomic und Embryologie als Festgabe Jacob Henle, zum 4. April, 1882, dargebracht von seiuen Schlilem. Bonn. Pp. 80-111. Taf. 10-12. Wheeler, W. M. *89. The Embryology of Blatta germanica and Dorj'phora decemlineata. Joiirn. morph., vol. 3, pp. 291-374, pis. 15-21. Witlaczil, E. '84. Entwicklungsgeschichte der Aphiden. Zeitschr. f . wiss. Zool., Bd. 40, pp. 559-696, Taf. 28-34. Printed June, 1900. Digitized by VjOOQ IC Pratt.— Imaginal discs. EXPLANATION OF PLATES. All figures are from preparations of Melophagus ovinus L., and were made with the aid of a camera lucida. The mesoderm and organs derived from it are colored blue, the yolk yellow. Abbreviations. a. anterior. Vvag. tr. tracheal invagination. am. amniotic rudiment. la. g. germinal plate. an. anus. In0. sucking tongue. bVdrm. blastoderm. mr'py. micropyle. chn. chorion. ms^drm. mesoderm. coel. body-cavity. n. m. median nerve. corns, cb. cerebral commissures. n. V. ventral nerve chord. cr. heart. nl. segmentation nuclei. eras. stmd. stomodeal thickening. nl. vt. yolk nuclei. crs. ma'drm. mesodermic ridge. oe. oesophagus. crs. m-v. mid-ventral ectodermic of. m. median opening of paired ridge. cephalic disc. d. dorsal. or. mouth. dsc. al. wing disc. P- posterior. (Uc. ce. paired cephalic disc. phy. pharynx. dsc. ce. m. median cephalic disc. pli. ca. taU-fold. dsc. gen. a. anterior genital disc. pli. ce. head-fold. dsc. gen. p. posterior genital disc. prd. proctodeum. dsc. pd. 1, prothoracic leg disc. stmd. stomodeum. dsc. pd. 3. mesothoracic leg disc. St. pr'pl. pi. peripheral protoplasmic dsc. pd. 3. metathoracic leg disc. layer. en'drm. endoderm. sul. m-v. mid-ventral groove. gl. sal. salivary glands. tae. rns^drm. mesodermic bands. gn. dla. ganglia allata. tb. mpg. Malpighian tubules. gn. cb. cerebral ganglia. tr. tVachea. gn.f. ganglion froutale. trt. pr^pl. protoplasmic paths. gran. vt. yolk granules. V, ventral. Digitized by VjOOQ IC Pbatt. — Imag^inal discs. PLATE 1. Fig. 1. Longitudinal section of a segmenting egg, in a plane nearly coinciding with the sagittal plane of the embryo. X 78. Fig. 2. Similar section of a somewhat older segmenting egg. X 78. Fig. 3. A portion of the section shown in the preceding figure, seen under a higher magnification. X 760. Fig. 4. Sagittal section of a segmenting egg in which all parts of the peripheral layer of protoplasm are occupied by nuclei. X 78. Fig. 6. A portion of the preceding section seen under a higher magnification, many of the nuclei in process of division ; the peripheral layer of protoplasm with shallow grooves surrounding the nuclei. X 760. Fig. 6. A portion of a section of a segmenting egg showing the deepening of the grooves in the peripheral layer of protoplasm to form the cells of the developing blastoderm. X 760. Fig. 7. A portion of a section showing the completely formed blasttxierm. X876. Digitized by VjOOQ IC Pratt- 1 MA GiNAL Discs. Plate 1. nl ''nipt - H- -j'.-iui/ .,r.„..L : T;: ^ rkft. .>^/v. ii r^^D thf ftf dm. t^rff/t if. Ht/jr)/^ I ( ^^ Phuc Boston Soc Na7. Kist Voi . l9. K%;^ .tiOiii -■ Digitized by Google Digitized by VjOOQ IC Digitized by VjOOQ IC Pbatt. — Imaginal discs. PLATE 2. Fig. 8. Cross-section of an embryo showing the origin of the mesoderm. X130. Fig. 9. Sagittal section of an embryo showing the formation of the head- and tail-folds. X 100. Fig. 10. Sagittal section of an embryo showing formation of proctodeum. XIOO. Fig. 11. Dorsal aspect of an embryo in which both proctodeum and stomodeum have been formed. X 100. Fig. 12. Semi-diagrammatic sagittal section of an embryo in which both proc- todeum and stomodeum have been formed. X 100. Fig. 13. Surface view, lateral aspect, of an embryo of the same stage as that of Fig. 12. X 100. Fig. 14. Cross-section through the stomodeum of an embryo of about the same age as that shown in Figs. 12 and 13. X 130. Fig. 15. Cross-section through the middle of the same embryo as the one shown in the preceding figure. X 130. Fig. 16. Cross-section through the proctodeum of the same embryo as that shown in Fig. 14. X 130. Fig. 48. Cross-section of an old embryo, passing through the anus and showing the invaginations of the posterior genital discs. X 220. ¥ig. 49. Cross-section passing through the end-intestine and middle portion of the genital discs. From the same en^bryo. X 220. Fig. 60. Cross-section passing through the anterior genital discs of the same embryo. X 220. Digitized by VjOOQ IC ''^; tl^r,fm4f n ♦ « 'ift :fe 'j^f*.fy ny ./.,. .^^^/I Digitized by Google 4 1 \ i L Digitized by VjOOQ IC Digitized by VjOOQ IC Pratt. — Imaginal disci. PLATE 8. Fig. 17. Surface view, dorsal aspect, of the head-fold of an embryo, showing the first appearance of the cephalic imaginal discs, x 100. Fig. 18. Cross-section through the stomodeum of an embryo somewhat older than that represented in Fig. 17. X 130. Fig. 19. Cross-section of the same embryo as that shown In Fig. 18, taken at some distance further back, showing the cephalic discs where they are deepest. X 130. Fig. 20. Cross-section through the middle of the same embryo. X 130. Fig. 21. Cross-section through the proctodeum of the same embryo. X 130. Fig. 22. A somewhat oblique, but nearly sagittal, section of an embryo, the stomodeum of which has reached a terminal poflition. X 130. Digitized by VjOOQ IC Pratt- I MAG] NAL Disc: Plate 3. //" Sfu,(f..., ■\ ^ ... ry.srcfjf/ . !L<^- 'Crii \ <'\ ■rhi. inr'py 17. (/.sr/'f'. — - ^. .J..*i.-yi;,J^'9^ ////^/. 'j/sa'. ./A.-^..-:. ^4^ 2S, \shnr/. Pruc ^.(^'ST'jn Soc 'Ja: !1;3t \'oi l'j .v///^ '/. 27. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Pbatt. — Imagina) discs. PLATE 6. Fig. 30. Anterior portion of the section represented in Fig. 29, seen under a higher magnification. X 200. Figs. 31 and 32. Parasagittal sections of the same embryo, Fig. 32 being from the more lateral section. X 200. Fig. 33. Parasa^ttal section of the same embryo as that shown in Figs. 34 and 36. It is more lateral in position than Fig. 36. X 200. Digitized by VjOOQ IC PkATT- iMA-iiN'AL 'JlSC: P,.ATt 5 ' I: ^>^^ ay. .A<'./v// .y^.yw'/^^ Vff ■ ^^ p [J.: F'ROc B^^'STON Soc Nai Hist V\ji. ;^9 Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Pratt. — Imaginal discs. PLATE 6. Fig. 34. A slightly oblique sagittal section of an embryo in which the Involu- tion of the head lb completed. X 200. Fig. 36. Parasagittal section intermediate in position between those of Figs. 83 and 34 and from the same embryo as those figures. X 200. Figs. 36-39. Cross-sections of the forward end of an embryo of about the same age as that shown in the preceding figures. The region of the embryo from which each of the sections is taken is indicated in Fig. 34 by the number of the figure representing the section. X 200. Digitized by VjOOQ IC pHATr-lr-;A.r: D:.,c': P;_ATh 6 .■»• *'V, I : ■:r-rP^''- rtShr^ / . '^V _f/srrr f(st\pa - iSS''.' '■-^'Wli^^i^Mti t: '^m .^^ .n fAf phr a ■5>' ji; :io;: 4 i;^^ 'iLft^,^ Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Pratt.— imaginal discs. PLATE 7. Tigs. 40-44. Cross-sections of the forward end of the same embryo as that shown in the preceding four figures. The region of each section is indicated in Fig. 34, PI. 7, by the number of the figure which repre- sents the section, x 200. Fig. 45. Frontal section passing through an invaginating thoracic disc. X 340. Fig. 46. Similar section of a completed thoracic disc. X 340. [There is no Figure 47. Figures 48-60 are on Plate 2.] Digitized by VjOOQ IC Pk-\T7- In'a.-.'NAI. i)fsc:-; ;,ArE7 rrWM ,/.-. u. /■■■.■ '.'!w. U M ./v L //^ ^*'4 ^ il<^.' t /^■S/W , i ^CS^ '^^' ^ ''^\' \ Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. Memoirs. 4to. The development, structure, and affinities of the genus Equisetum. By Edward C. Jeffrey. 80 pp., 6 phites. $1.00. Localized stages in development in plants and animals. By Robert T. Jackson. 65 pp., 10 plates. $2.00. Procbedinoh. 8vo. A revision of the systematic names employed by writers on the morphology of the Acmaeidae. By M. A. Willcox. 6 pp. 10 cts. On a hitherto unrecognized form of blood circulation without capillaries in Uie organs of vertebrata. By Charles Sedgwick Mi not. 31 pp. 35 cts. The occurrence of fossils in the Uoxbury conglomerate. By Heni7 T, Burr and Robert E. Burke. 6 pp., 1 plate. 20 cts. The blood vessels of the heart in Carcharia-s, Raja, and Aniia. By G. II. Tarker and F. K. Davis. 10 pp., 3 platt^s.. 26 cts. List of marine mollusca of (^oldspring Harbor, L(mg Island, with descriptions of one new gei\.us and two new species of nudibranchs. By Francis Noyes Balch. 30 pp., 1 plate. 35 cts. The development of Penilia schmackeri Richard. By Mervin T. Sudler. 23 pp., 3 plates. 30 cts. . Contributions from the (iray herbarium of Harvard university. New series, no, 17. 1. Revision of the genus Gymnolomia. 2. Supplemfentary notes upo^^ Calea, Tridi\x, and Mikania. By B. L. Robinson and J. M. Greenman. 22 pp. 25 cts. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United States. By (tarry de N. Hough. 10 pp. 10 cts. Notes on the reptiles and amphibians of Intervale, N. H. By Glover M. Allen. 13 pp. 16 cts. Variation and sexual selection in man. By Edwin Tenney Brewster. 17 pp. 25 cts. Moniloporidae, a new family of Paba^ozoic corals. By Amadeus W. Grabau. . 10 pp., 4 plates. 25 cts. - Studies in the gold-bearing slates of Nova Scotia. By J. Edmund Woodman. 33 pp., 3 i)late.s. 60 cts. North American w^od frogs. By Reginald Ileber Howe, Jr. 6 pp. 10 cts. Some Hydroids from Puget Sound. By Gary N. Calkins. 36 pp.. 0 plates. ' 50 cts. The Odonate genus Macrothemis and its allies. By Philip P. Calvert. 32 pp., 2 plates. 50 ct,s. On the veins of the Wolffian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 25 cts. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 15 cts. The genus Antennaria in New England. By Merritt L. Femald. 13 pp. 15 cts. The land mammals of peninsular Florida and the coast region of Georgia. By Out ram Bangs. 7J> pp. 75 cts. A contribution to the petrography of the Boston Basin. By Theodore G. White. 40 pp., 5 plates. 05 cts. - * Clymene producta sp. ;i()v. By Margaret Lewis. 5 pp., 2 plates. 15 cts. The Harvard g(?ograi)hical models. By W. M. Davis. 26 pp., 4 plates. 26 cts. The role of water in growth. B}' C. B. Davenport. 12 pp. 16 cts. Digitized by VjOOQ IC 1% Proceedings of the Boston Society of Natural EUstorj. Vol. 29, No. 14, pp. 273-322. GLACIAL EROSION IN FRANCE, SWITZERLAND AND NORWAY. By William Morris Davis. WiTHM THREK PLATKS. BOSTON: PRINTED FOR THE SOCIETY. July, 1900. Digitized by VjOOQ IC Digitized by VjOOQ IC NO. 14. — GLACIAL EROSION IN FRANCE, SWITZER- LAND AND NORWAY. By William Morris Davis. ' Introductory, — Eighteen years ago I presented to this Society an essay on Glacial Erosion, in which my own observations were supplemented by a review of all that I could find written on the subject, in the hope of reaching some safe conclusion regarding what was then (as it is still) a mooted question. Although recof^- nizing effective erosion to depths of "a moderate number of feet" where ice pressure was great and motion was rapid, in contrast to . deposition where pressure and motion were reduced and where the amount of subglacial drift was excessive, I could not at that time find evidence to warrant the acceptance of great glacial erosion, such as was advocated by those who ascribed Alpine lakes and Norwegian fiords to this agency. In a retrospect from the present time, it seems a^ if one of the causes that led to my conservative .position' were the extreme exaggeration of some glacialists, who found in glacial erosion a destructive agency competent to accom- plish any desired amount of denudation — an opinion from which I recoiled too far. Since the publication of my pi-evious essay I had gradually come to accept a greater and greater amount of . glacial erosion in the regions of active ice motion ; but in spite of this slow change of opinion, the maximum measure of destructive work that, up to last year, seemed to me attributable to glaciers was moderate ; and it was therefore with great surprise that I then came upon certain facts in the Alps and in Norway which demanded wholesale glacial erosion for their explanation. The desire of some years past to revise and extend my former essay then came to be a duty, which it is the object of this paper to fulfil. My former revision of the problem divided the arguments for glacial erosion under four headings: observations on existing glaciers and inferences from these observations; the amount and arrangement of glacial drift; the topography of glaciated regions; and the so-called argument from necessity, — that is, the belief that glaciers must have done this and that because nothing else Digitized by VjOOQ IC 274 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. competent to the task could be found. It is not possible for me at present to review all the new material pertinent to the whole problem; attention can be given here chiefly to a few examples under the third heading. • A Glaciated Valley in Central France, — It is evident that, if it were possible to obtain a definite idea of the preglacial topography of a glaciated district, the amount of glacial work might be readily determined as the difference between the preglacial and the present form ; independent evidence sufficing to prove that general denu- dation of the rocky crust in the brief postglacial epoch had been inconsiderable. This method leads one to conclude that in gen- eral the topography of southern New England has not been strongly modified by glacial action ; for we find here on the whole the same maturely dissected upland that prevails in regions of simi- lar structure outside of the glacial boundary; the uplands being explained as parts of an uplifted peneplain of late Mesozoic date, and the valleys as the work of ordinary erosion in a part of Terti- ary time : but this method of measuring glacial erosion by dating topographic forms had not been developed twenty years ago. Strong glacial erosion may, however, be expected in New England where ice motion was locally accelerated, as through the notches of the White Mountains. Again, in the glaciated area of the Cen- tral Plateau of France, I had opportunity in January, 1899, of seeing a valley that had been locally modified to a determinate amount by a glacier that once descended northwest from the Cantal along the valley of the Rhue to the junction of the latter with the Dordogne. Outside of the glaciated area, the valleys of the plateau — an uplifted and sub-maturely dissected peneplain, mostly of crystalline rocks — frequently follow incised meandering courses, in which the steep concave slopes are regularly opposed to the gentler convex slopes ; the latter being spur-like projections of the uplands, advancing alternately from one and the other side of the valley. Valleys of this kind are singularly systematic in form, as the result of the combined downward and outward cutting by their streams which, already winding or meandering when the erosion of the valleys began, have increased the width of their meander belt while they deepened their valleys. On entering tlie glaciated valley of the Rhue, it is found that the regularly descend- ing spurs of the non- glaciated valleys are represented by irregular knobs and mounds, scoured on their up-stream and plucked on the down-stream side ; and that the cliffs formed where the spurs are Digitized by VjOOQIC DAVIS: GLACIAL EROSION. 275 cut off, as in Fig. 1, are sometimes fully as strong as those which naturally stand on the opposite side of the valley. The spurs generally remain in sufficient strength to require the river to follow its preglacial serpentine course around them, but they are some- times so far destroyed as to allow the river to take a shorter course Fig. 1. The glaciated valley of the Rhue. through what was once the neck of a spur.^ The short course is not for a moment to be confounded with the normal cut-offs through the narrowed necks of spurs, such as are so finely exhibited in the meandering valleys of the Meuse and the Moselle. The short courses are distinctly abnormal features, like the rugged knobs to which the once smooth- sloping spurs are now reduced. It was thus possible in the valley of the Rhue to make a definite restoration of preglacial form, and to measure the change produced by glaciation. The change was of moderate amount, but it was highly significant of glacial action, for it showed that while a slender, fast-flowing stream of water might contentedly follow a serpentine course at the bottom of a meandering valley, the clumsy, slow- moving stream of ice could not easily adapt itself to so tortuous a path. The more or less complete obliteration of the spui-s was the result of the effort of the ice stream to prepare for itself a smooth- sided trough of slight curvature ; and if the rocks had been weaker, or if the ice had been heavier, or if the glacial period of the Cantal had lasted longer, this effort might have been so successful as to have destroyed all traces of the spurs. Fortunately the change actually produced, only modified the spurs, but did not entirely 1 The short-cuts are sometimes narrow gorges incised in the half-consumed spurs ; and in such cases, tlie displacement of the Rhue from its former roundabout course is probably to be exi)Iained by constraint or obstruction by ice. Digitized by VjOOQ IC 276 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. destroy them ; and their rugged remnants are highly significant of what a glacier can do. Rocky Knobs in Glaciated Areas, — On thus generalizing the lesson of the Rhue, it is seen that just before the complete oblitera- tion of the spurs some of their remnant knobs may be isolated from the uplands whence these preglacial spurs descended. It is out of the question to regard the ruggedness of such knobs as an indication of small change from their preglacial form, as has been done by some observers. The ruggedness is really an indication of the manner in which a glacier reduces a larger mass to smaller dimen- sions, by plucking on the down-stream side as well as by scouring on the up-stream side. It is possible that knobs in other glaciated valleys than that of the Rhue may be of this origin ; they should then be regarded not as standing almost unchanged and testifying to the incapacity of glacial erosion, but as surviving remnants of much larger masses, standing, hke raonadnocks above a peneplain, as monuments to the departed greater forms. The two knobs at Sion (Sitten) and the Maladeires, all detached from Mont d'Orge in the upper valley of the Rhone, the hills of Bellinzona in the valley of the Ticino, the rocks of Salzburg where the Sal- zach emerges from the Alps, and even the Borromeo islands in Lake Maggiore, may perhaps be thus interpreted. Rugged as these knobs may be on the down-stream side, it would be an unreasonable contradiction of the conclusions based on observations of many kinds to maintain that their ruggedness was of preglacial origin. The ice stream from the Cantal at one time expanded sufficiently to flood the uplands bordering the valley of the Rhue,^ where it produced changes of a most significant kind. The neighboring unglaciated uplands are of systematic form ; broad, smoothly arched masses rise, round-shouldered, between the narrow valleys that are incised beneath them ; the uplands are as a rule deeply soil-covered, and bare ledges prevail only on the stronger slopes of the young valleys that have been eroded since the peneplain was raised to its present upland estate. But within the glaciated area near the Rhue, the broadly rounded forms of the uplands are replaced by a succession of most irregular rocky knobs, from which the preglacial soils have been well scom'ed away, as in Fig. 2. This seems to be a form most appropriate to glacial action on a surface that had been 1 According to Boule ('96), the glaciation of the uplands and of the valleys was sepa- rated by an interplacial epoch, but I did not have occasion to inquire particularly inta this aspect of the problem Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 277 weathered to variable depths in preglacial time. The ice action sufficed to rasp away the greater part of the weathered material and to grind down somewhat the underlying rock, often giving the knobs a rounded profile; but it did not nearly suffice to reduce the Fig. 2. Glaciated knobs on the Central Plateau of France. rocky surface to an even grade. The ice action seems here to have resembled that of a torrent which might sweep away the waste on a flood plain and lay bare and erode the rock ledges beneath ; but whose duration was not sufficient to develop a graded floor appro- priate to its current. Another example of this kind seems to occur where the huge glacier of the Inn, escaping from its well enclosed channel within the mountains, once spread forward in a great fan of ice over the foot-hills at the northern border of the Alps and crept out upon the piedmont plain. The glance that I had at this foot-hill district from a passing train gave me the impression that its ruggedness was much greater than usually obtains along the mountain flanks ; as if the rolling hills of preglacial time had been scoured to an increas- ing roughness by an overwhelming ice-flood that would, if a longer time of action had been permitted to it, have worn down all the inequalities to a smooth, maturely graded floor. The Vcdley of the Ticino, — My first entrance into the Alps last year was from the south by the valley of the Ticino. Thirty-one years before I had followed the same valley and admired its bold sides and its numerous waterfalls; but at that time nothing was noticed that seemed inappropriate to the general idea of the erosion Digitized by VjOOQ IC 278 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. of valleys by their rivers. Thirty years is a long enough time for one to learn something new even about valleys, and on my second visit it was fairly startling to find that^the lateral valleys opened on the walls of the main valley of the Ticino five hundred feet or more above its floor, and that the side streams cascaded down the steep main-valley walls in which they have worn nothing more than nar- row clefts of small depth. This set me wondering, not only as to the meaning of so peculiar an arrangement of valleys and streams, but also as to the reason why so peculiar an arrangement should not have sooner attracted attention as an exceptional characteristic of Alpine topography. Playfair long ago, when describing the relation of side valleys to their trunk, showed clearly that they had " such a nice adjustment of their declivities that none of them join the principal valley either on too high or too low a level : a circumstance which would be infinitely improbable if each of these vallies were not the work of the stream that flows in it" ('02, 102); yet the whole course of the passing century has hardly sufficed to make full application of this law. So much latitude is usually allowed in the relation of branch and trunk valleys that hun- dreds of observers, many of whom must have been cognizant of Playf air's law, have made no note of the extraordinary exceptions to it that prevail in the glaciated valleys of the Alps. Even the most pronounced advocates of glacial erosion, with a few exceptions to be noted below, have been silent regarding the remarkable failure of adjustment between the declivities of lateral and main glaciated valleys. Indeed, in reviewing the writings of those who have accepted a ''^^2^£^^' "I Fii?. 3. Val d'Osogiia, a hanging lateral valley of the Ticino. Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 279 large measure of glacial erosion, one must be struck with the undue attention that they have given to lake basins and the relative inat- tention to valleys. This disproportion is probably to be explained as a result of the greater contrast that prevails between a river and* a lake than between a river and its branch ; it is perhaps for this reason that the attention of geologists and geographei-s has gener- ally been directed to the origin of lakes rather than to the relation of branch and trunk streams, even when the former cascade from their lateral valleys into the main valley. That glacial erosionists made so little claim for the general deepening of glaciated valleys while they demanded a great deepening of those parts of valleys which have been scoured down to form lake basins, has always seemed to me a difficulty in the way of accepting the demanded measure of lake-basin erosion ; and this difficulty was supported by the well-attested observation that the side slopes of glaciated valleys manifest no marked or persistent increase of declivity in passing from above to below the limit of glaciation. If glaciers had scoured out deep lake basins, like those of Maggiore and ^ ^^/nmh Geneva, they ought to have significantly deepened the valleys up-stream from the lakes; and if the valleys were thus significantly deepened, it seemed as if their slopes Fig. 4. Section of a glaciated valley. ^\iOM\^ be steeper below than above the limit of glacial action. The denial of the latter requisite seemed to me to carry with it the denial of the two preceding suppositions. Features of Strongly Glaciated Valleys, — It is true that the uppermost limit of glaciation, Q R, Fig. 4, in Alpine valleys is not attended by a persistent change in the steepness of the valley sides, A E, C J ; but on descending well within the glaciated val- ley, a very strong change may usually be found in the slope of the valley walls. The larger valleys, once occupied by heavy glaciers from the lofty central snow fields, are chai-acterized by " basal cliffs,*' E F, J II, that rise several hundred or even a thou- sand feet above their broad floors, and thus enclose what may be called a " bottom trough,*' E F II J, half a mile or a mile wide. The bottom trough of the Ticino, as seen when one looks up stream towards Giornico, is shown in Plate 1, Figure A, The Digitized by VjOOQ IC 280 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. basal cliffs are comparatively straight- walled ; they have no sharp spurs advancing into the valley floor. The rock floor, G, Fig. 4, is buried by gravels, F H, to an unknown depth. It is only from the benches above the basal cliffs that the valley sides flare open with maturely inclined slopes; and it is at a moderate depth beneath the level of the benches at the top of these basal cliffs that the lateral valleys, D K, open on the walls of the main valley. The bottom trough within the basal cliffs and beneath the lateral valleys seems to be of glacial origin. It is in the first place a characteristic feature of all the larger glaciated Alpine valleys, as I am assured by Professors Penck, Brttckner and Kichter, with whom the matter was carefully discussed last summer. The non- glaciated valleys manifest no such peculiar form. It is not simply that the terminal portion, J B K, of a lateral valley has been cut off by the glacial widening of the paain valley floor; the main valley has been strongly deepened, as is assured by the relation of its floor, F H, to the prolongation of the floor of the lateral valley, KB. The first may be several hundred feet — indeed in some valleys, a good thousand feet — below the second. The lateral valleys must have once entered the main valley at grade, for the flaring sides of the main valley indicate maturity ; the side slopes, AE, CJ, must have once met at B. Even the lateral valleys have an open V section, proving that their streams had cut down to a graded slope, D B, that must have led them to an accordant junction with the main river. Nothing seems so competent as glacial erosion to explain the strong discordance of the existing valleys. The lateral as well as the main valleys have been glaciated, but the former do not exhibit changes of form so distinctly as the latter: in the Ticino system the lateral valleys did not, as far as I saw them, seem to have been much affected by glaciation, a fact that may be attributed to the small size of their branch glaciers in contrast with the great volume of the trunk glacier. There is no sufficient evidence that the valley floor between the basal cliffs has been faulted down, after the fashion of a grahen ; for although this origin is advocated by Rothpletz (*98, 237) for the Linththal, the evidence that he adduces for the limiting faults is not agreed to by Alpine geologists in general, and the persistent association of the bottom troughs with the crooked course of pre-existent, maturely open valleys involves special conditions of faulting that cannot be accepted without the strongest evidence. Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 281 It is not satisfactory to explain the bottom trough as having been worn out by normal trunk-river erosion, leaving the side streams as it were hanging or suspended above them, for to admit such an origin would be to go counter to all that has been learned regarding the systematic development of valleys. Here it is with regret that I must differ from the opinion of two eminent Swiss geologists who explain the deepening of the main valleys by a revival in the erosive power of the rivers as a result of a regional uplift, while they regard the hanging lateral valleys as not yet accordantly deepened by their smaller streams. It is true that narrow trenches are cut in the floors of the hanging valleys, show- ing that their streams have made some response to the erosion of the bottom though in the main valley, and if the bottom trough were a narrow canyon, this relation of trunk and branch streams might be considered normal; but if the breadth as well as the depth of the bottom trough had been acquired by normal river erosion, the side valleys should now, it seems to me, have been trenched much deeper than they are, to some such slope as ST, Fig. 4. The opinions of Rtitimeyer and Heim on this question are as follows : — Rtitimeyer gave an excellent account of hanging lateral valleys thirty years ago in his description of the valley of the Reuss (*69, 13-24). He recognized benches or ThaUtufen on each side of the valley above the basal cliffs of the existing bottom trough, and regarded them as the remnants of a former, wide open valley floor. Side valleys of moderate fall enter the main valley about at the level of the Thalstufen^ and their waters then cascade down over the basal cliffs to the Reuss. Glacial erosion is dismissed as incompetent to erode the bottom trough ; indeed, the time of glacial occupation of the valley is considered a period of rest — a sort of "pupa stage" — in its development. The discordance of main and lateral valleys is ascribed entirely to the differential erosion of their streams. Heim's views on this matter are to be found in his ^^ Mechanismus der Gehirgshildung^^ ('78, 1, 282-301) and in an article *' JJcher die Erosion im Gehiete der Reuss " ('79). He recognizes that the bottom troughs have been excavated in the floors of pre-existing valleys, whose stream lines had been reduced to an even grade (profile of equilibrium, " GleichgercichtsUnie'') and whose lateral slopes had been maturely opened. The side streams must at that time have eroded their valleys deep enough to enter the main valley at accordant grade as stated above. Since Digitized by VjOOQ IC 282 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. then, it is concluded that an elevation of the region has caused a revival (Neubelebimg) of the main river ; and the present greater depth of the main valley is, according to Heim, merely the natural result of this revival, while the smaller side streams have not yet been able to deepen their valleys. The height of the ThcUstuJen or remnants of the former valley floor, seen in the benches above the basal cliffs of the bottom trough, is taken as a measure of the elevation that the mountain mass has suffered. Apart from the improbability that the deepening of a bottom trough by a revived main river could truncate so many lateral valleys with so great nicety as is repeatedly the case, leaving their streams to cascade down in clefts but slightly incised in the main valley walls, the following considerations lead me to reject the possibility of explaining the discordance between side and main streams by a normal revival of river action. Mekition of Trunk and Branch Valleys, — The general accord- ance of maturely developed main and lateral valleys in non-glaciated regions, as recognized by Playfair, is today fully established by innumerable observations in many parts of the world. Truly, during the attainment of mature development, it is possible that a large river may outstrip a small branch stream in the work of deep- ening its valley, but the discordance thus produced can prevail only during early youth ; for as soon as the main river approaches grade the further deeping of its valley is retarded, while at the same time the steepened descent of the lateral streams at their entrance into the main valley accelerates their erosive work. Hence, even if a large trunk river has for a time eroded its valley to a significant depth beneath the tributary valleys, this discordance cannot endure long in the history of the river. Examples of such normal discord- ance are to be found in non-glaciated regions only in the branch streams of rivers that occupy very narrow canyons ; and even rivers in canyons sometimes receive their branches at accordant grade, as seems to be usually the case with the Colorado, if one may judge by photogi-aphs. The narrow postglacial gorges cut by active streams, habitually receive their branches — when they receive any — from hanging side gorges ; and an excellent example has long since been on record in the gorge of Cattaraugus Creek in wegtem New York, where a branch, the Canaserowlie, falls into the main gorge from a side gorge of much less depth. Referring to this. Hall wrote: — " In the more recently excavated channels we find the streams fall- ing over the very edge of the cliff, having produced no perceptible Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 283 recession in the margin of the fall" (*43,380). But however ap- propriate a discordance of branch and trunk may be in early youth, it cannot endure long enough to be associated with maturely opened main valleys. It should be noted that discordance of side and main valleys may also be found where a large river has lately been turned to a new path, as in the normal progress of the capture of the upper course of one river by the headward gnawing of a branch of another river (see reference to Russell below) , or in the new arrangements of drainage lines in a region from which a glacial sheet has lately withdrawn. Furthermore, the valleys of very small wet-weather streams are frequently discordant with the valley of a serpentine river, if they enter it from the upland that is under-cut by the con- cave bank of the river. But these cases cannot find application in the hanging valleys of the Alps. The hanging valleys that open on sea cliffs, such as those of Normandy, are of course quite another matter. Overdeepened Main Valley a and Hanging Lateral Valleys, — Now it is characteristic of the bottom troughs of the glaciated Alpine valleys that they are broad-floored ; they cannot be described as canyons in any proper sense of that word : the walls are steep enough, but they are too far apart. If the existing breadth of the troughs had been acquired in the ordinary manner by the lateral swinging of the main stream and by the lateral weathering of the walls, the long time required for such a change would have amply sufficed for the lateral streams to cut down their valleys to grade with the main valley ; and their persistent failure to do so indicates the action of something else than normal river work in the widening of the main valley. This is the very kernel of the problem. If a main valley were excavated along a belt of weak rocks, the side valley might stand for some time at a considerable height above the main valley floor. Certain hanging vaUeys in the Alps seem at first sight to belong to this class, but such is not really the case. For example, where the Linth flows into the Wallen See, the well- defined bottom troughs of the river and of the lake both pass obliquely through a syncline of strong lower Cretaceous limestone, which forms cliffs on their walls. Side streams drain the high synclinal areas; one such stream cascades from the west into the Linth trough back of the village of Nafels ; another cascades from the north into the Wallen See near its western end. The first explanation for such falls is that they are normally held up on the resistant limestone ; but it should be noted that the bottom troughs Digitized by VjOOQ IC 284 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. of the Linth and the Wallen See have been cut down and broadly opened in the same limestones. If the troughs were of normal river origin, the side streams also should have by this timB trenched the limestones deeply, instead of falling over the limestone cliiTs at the very side of the larger troughs. In the Ticino valley where the side streams are most discordant, massive gneisses prevail ; the structure is so nearly uniform over large areas that it affords no explanation of the strong discordance between side and main valleys. It thus seems obligatory to conclude that the bottom troughs of the larger Alpine valleys were deepened and widened by ice action. This belief is permitted by the abundant signs of glacial erosion on the spurless basal cliffs, and required by the persistent association of over-deepened bottom troughs and discordant hanging lateral valleys with regions of strong glaciation. The valley of the Ticino manifests these peculiarities very distinctly, and I have recently described them in some detail in a paper in Appalachia (1900) . Subaerial Erosion during the Gldcial Period, — It should not be imagined that the glacial erosion of troughs in valley floors was necessarily so rapid that no significant subaerial erosion was accom- plished during its progress. Ordinary weathering and down-hill transportation of rock waste must have been in active operation on the valley sides above the border of the ice-filled channels; and the very fact that on the upper slopes of the mountains, pre- glacial, glacial and postglacial erosion was similarly conditioned, makes it difficult to distinguish the work done there in each of these three chapters of time. In the diagrams accompanying this article no indication of change from preglacial to postglacial outline on the ui>per mountain slopes is indicated, because no satisfactory meas- ure can be given to it. Lake Lugano, — In the presence of a variety of evidence col- lected for some years previous to my recent European trip, it had been my feeling that the best explanation offered for the large lakes that occupy certain valleys on the Italian slope of the Alps was that they had resulted from what has been called valley-warping, as set forth by Lyell, Heira and others. It was my desire to look es- pecially at Lakes Maggiore, Lugano and Como with this hypothesis in mind, and to subject it to a careful test by means of certain asso- ciated changes that should expectedly occur on the slopes of the neighboring mountains, as may be explained as follows. On the supposition of moderate or small glacial erosion, a well- Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 285 matured stage of dissection must have been attained in the district of the Italian lakes in preglacial time ; for the main valleys are widely opened, and even the lateral valleys have flaring slopes. In a mature stage of dissection mountains should exhibit a well- advanced grading of their slopes ; that is, their sides should be worn back to a comparatively even declivity with little regard to diversity of structure ; the descending streams of waste being thus seen to correspond to the flood plains of graded rivers. The agencies of weathering and transportation are delicately balanced wherever graded slopes prevail ; and a slight tilting of the moun- tain mass might suffice to disturb the adjustment between the sup- ply and the removal of waste ; then all the steepened slopes would soon be more or less completely stripped of their waste cover; their rock ledges would be laid bare, although still preserving the comparatively even declivity that had been gained under the slowly moving waste. If the lakes had been formed by warping, it is possible to deduce with considerable accuracy the localities where the mountain slopes would be steepened and stripped; namely, the northern slopes about the southern end of the lakes, and the southern slopes about the northern end ; but as far as I was able to examine the district about Lake Lugano, no effects of such a warping and tilting were to be detected. The submergence of lateral valleys about the middle of the lakes is also, as has been well pointed out by Wallace, a necessary consequence of the theory of warping; but although the main valley floor is now deep under water, the side valleys are not submerged. Failing to find evidence of warping, and being much impressed with the evidence of deep glacial erosion as indicated by the hanging lateral valleys of the overdeepened Ticino, I examined the irregular troughs of Lake Lugano for similar features, and found them in abundance. One of the reasons why Lake Lugano had been selected for special study was that it did not lie on the line of any master valley leading from the central Alps to the piedmont plains ; hence, if influenced by ice action at all, its basin must have been less eroded than those of Como and Maggiore on the east and west. But in spite of this peculiarity of position, Lugano received strong ice streams from the great glaciers of the Como and Maggiore troughs (see Glacial Distributaries, below), and its enclosing slopes possess every sign of having been strongly scoured by ice action. The sides of the lake trough are often steep and cliff-like for hun- Digitized by VjOOQ IC 286 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. dreds of feet above present water level, thus simulating the basal cliffs of the Ticino valley ; while at greater heights the valley sides lean back in relatively well-graded slopes, as in Plate 1, Figure B^ where the southern side of the northeastern arm of the lake, near Porlezza, is shown. The angle at the change of slope is often well defined, but it is independent of rock structure. Narrow ravines are frequently incised in the basal cliffs, and alluvial fans of greater or less size are built into the lake waters from the base of the ravines. The northeastern arm of the lake, extending from the town of Lugano to Porlezza, receives several cascading streams from hanging valleys on its southern side, one of which is here shown in Plate 2, Figure A, The side slopes of the hanging valleys are for the most part flaring open and well graded, from which it must be con- cluded that their streams had, under some condition no longer exist- ing, ceased to deepen their valleys for a time long enough to allow the valley sides to assume a mature expression ; and that since then the bottom trough of the main arm of the lake has been eroded deep and wide, with a very small accompanying change in the lateral valleys. In other words, the side valleys were, in preglacial time, eroded to a depth accordant with the floor of the master valley that tliey joined, and since then the bottom trough has been eroded in the floor of the master valley by a branch of the Como glacier. In postglacial time the side streams have begun to trench their valley floors, eroding little canyons ; but much of this sort of work must be done before the side valleys are graded down even to the level of the lake waters, much less to the level of the bottom of the lake. The two southern arms of the lake lead to troughs whose floors ascend southward to the moraines of the foot-hills, beyond which stretch forward the abundant overwashed gravels of the great plain of the Po. I do not mean to imply that every detail of form about Lake Lugano can find ready explanation by the mature glacial modifi- cation of a mature preglacial valley system ; but a great number of forms may be thus explained, and a beUef in strong glacial erosion was forced upon me here as well as in the valley of the Ticino. A detailed study of the Italian lakes with the intention of care- fully sorting out all the glacial modifications of preglacial forms would be most profitable. Various Examples of Glaciated Valleys, — My excursions of last summer showed me a number of over-deepened main valleys and hanging lateral valleys in the Alps ; for example, those of the Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 287 Fig. 6. True-scale cross-section of the Lauterbrunnen valley. Inn and of the Aar. Lakes Thun and Brienz receive numerous cascades from hanging valleys that stand high above the water surface. The valley of Lauterbrunnen also affords a conspicuous illustration of a deep bottom trough enclosed by high basal cliffs that rise to the edge of more open upper slopes ; the celebrated Staubbach fall is the descent of a small lateral stream from its lofty hanging valley (see extract from an article by Wallace, cited below) , and the picturesque village of JVItlrren, M, Fig. 5, stands on the flaring slope or Thalstufe of the preglacial valley, just above the great basal cliff of glacial origin. A mile or so south of the village of Lauterbrunnen, the Trummelbach, T, Fig. 5, descends the precipitous eastern wall from a hanging valley whose floor is hundreds of feet above that of the Ltltschine ; it is roughly sketched in Fig. 6. Although the lateral Trummelbach brings a large volume of water to the main valley, it descends by a very narrow cleft in the rock face, a trifling incision in the valley wall; while the main valley, whose trunk stream did not seem to be more than five times the volume of its branch, is half a mile or more broad, wide open and flat-floored. The cross-section of the main valley is over a thousand times as large as that of the lateral cleft. Such a disproportion of main valley and lateral cleft is entirely beyond explanation by the in- equality of their streams ; and for those who feel that they must reject glacial erosion as the cause of the disproportion, there seems to be no refuge but in ascribing the main valley to recent down- faulting : a process that can hardly be called on to follow systematically along the floors of the larger glaciated valleys of the Alps, and to avoid the non-glaciated valleys and the mountain ridges. ^ig. 0. Diagram of the gorge of the Trummelbach, Lauterbrunnen vallev. Digitized by VjOOQ IC 288 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Certain well-known Alpine glaciers may be instanced as reaching just beyond the end of a hanging lateral valley and thence cascading into the deeper main valley. One is the Mer de Glace by Chamou- nix ; its cascading end is known as the Glacier des Bois. Another is the neighboiing Glacier des Bossons, from whose upper amphi- theatre a steep tongue descends far below; like the waterfalls of Norway, the tongue may be seen lying on the side slope from a considerable distance up or down the main valley. A third example is the Glacier of the Rhone, whose splendid terminal cascade is so consjncuous from the road to the Furka pass. (These three I have seen some years ago.) Possibly the Vernagt glacier is another of the same kind ; its catastrophic overflows into the lower Rofen valley have often been described. Doubtless many other examples of this class might be named. While engaged upon these observations in the Alps in the spring of 1899, I sent a brief note about them to my esteemed friend, Mr. G. K. Gilbert of Washington, telling him that all the lateral vallej's seemed to be "hung up" above the floors of the trunk valleys. His reply was long in coming to Europe, and, on arriving at last, it was dated Sitka, Alaska, where Mr. Gilbert had gone as a mem- ber of the Ilarriman Alaskan Expedition, and where my note had been forwarded. He wrote that, for the fortnight previous to hearing from me, he and his companions had been much impressed with the discordant relations of lateral valleys over the waters of the Alaskan fiords, and he suggested that such laterals should be called "hanging valleys" — a term which I have since then adopted. He fully agreed that hanging valleys presented unanswerable tes- timony for strong glacial erosion, as will be stated in his forthcoming report on the geology of the Expedition. After leaving Switzerland, I had a brief view of the lake district in northwest England, before crossing to Norway. The amount of glacial erosion in the radiating valleys of the English lakes has been much discussed, and as usual directly opposite views have been expressed. Rugged rocky knobs were seen in abundance about Ambleside and along the ridge separating the valley of Thirlmere from St. John's Vale ; and the latter receives a hanging valley from the east near Dalehead post-oflice. The famous falls of Lodore seemed to descend from the mouth of a hanging valley into Derwentwater. A model of the lake district, on exhibition at Keswick, showed some other examples of lateral valleys that seemed to stand above the floors of their main valleys, notably one Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 289 coming from the south near the northeast end of UUeswater. Since coming home I read the following in Marr*s " Scientific Study of Scenery": — "We find in the Lake District a number of tribu- tary valleys occurring in the hearts of the ridges, and opening out far above the bottoms of the main valleys, discharging their waters down the slopes in cascades. They are specially well marked on the east side of Helvellyn, and a number of them also open into the upper branches of Borrowdale." The explanation is that of Rtlti- meyer and Heim : — " For a considerable period after the deepening of the main valley, the minor valleys will end as definite gorges some height above the floor of the main valley, and discharge their waters in a series of cascades or falls down the side of the main valley" (1900, 136). One of my former students, Mr. W. B. Lloyd, has recently shown me a number of photographs of the fiords of southern New Zea- land, which he brought back from a visit to that distant country. High cascades, plunging from hanging lateral valleys into the broad waters of the fiords, are repeatedly shown ; the most striking view is here reproduced in Plate 2, Figure £ showing Sterling Fall leaping into Milford Sound. Fiords and Hanging Valleys in Norway, — In Norway I had the pleasure of making a ten days' cross-country excursion in com- pany with Dr. Reusch, Director of the Norwegian Geological Survey. We entered from Bergen through Hardanger fiord, and crossed the highlands by the Ilaukelisaetr road to Skien on the southeastern lowlands, thus making a general cross-section on which many char- acteristic features were seen. Norway has long been known as a land of waterfalls, but it is not generally stated with sufiicient clear- ness or emphasis that many or most of the falls are formed by the descent of streams from maturely opened trough-like hanging val- leys which are abruptly cut off by the walls of the fiords. The discordance between main and side streams is simply amazing. The fiord valleys are frequently one or two miles wide ; the waters of the fiords are of great depths, reaching 3000 feet in some cases. Even when a side valley stands but little above sea-level, its floor may be half a mile above the floor of the fiord. On passing inland beyond the head of the fiord water, where the whole depth of the fiord valley is visible, the side valleys may open more than a thousand feet above the main valley floor. In many cases where the fiords are enclosed by smooth walls, the cascading side streams have not yet incised a cleft in the bare rock surface, so that their foaming Digitized by VjOOQ IC 290 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. waters are visible for many miles up and down the fiord. Streams of considerable size sometimes plunge down from the rolling uplands in whose edge they seem to have just begun to cut a cleft. Ab- normal discordance of trunk and branch stream is, therefore, a strongly marked characteristic of the Norwegian drainage. The necessity for appealing to strong glacial erosion in explanation of this prevailing discordance, may be set forth as follows. Measure of Glacial Erosion in Norwegian Eiords, — The deep valleys of Norway, partly occupied by sea water, are incised beneath an uneven highland which bears so many hills and moun- tains that it makes little approach to a peneplain, yet which here and there shows so many broadly opened uplands between the bills and mountains that it may be taken to represent the well-advanced work of a former cycle of denudation when the region stood much lower than it stands now. As a whole, a mature or late mature stage seems to have been reached before a movement of uplift introduced the present cycle. Let us now make two suppositions regarding the work of normal river erosion in the preglacial part of the present cycle, in order to determine, if possible, how much addi- tional erosion must be attributed to ice in the production of existing forms. First, let it be supposed that the revived main rivers had incised their valleys to the depth of the present fiords in preglacial time, and that the discordance of main and side valleys now visible is the appropriate result of the youth of the present cycle. If we recall only the steepness of the fiord walls, this supposition might be justi- fied, and thus the amount of glacial erosion needed to develop exist- ing forms would be small. But it must not be forgotten that the fiords, although often steep-walled, are always broad, much broader than a young preglacial valley could have been at that stage of early youth when its side streams had not cut down to its own depth. Hence glacial erosion must, under this supposition, be appealed to for the widening of preglacial canyons, steep-walled and narrow, into the existing fiord troughs, steep- walled and broad. At the middle of the fiord troughs, the lateral erosion thus demanded would often measure thousands of feet, and that in the most massive and resistant crystal- line rocks. A second supposition leads to no greater economy of glacial action. Let it be supposed that the revived streams of preglacial time had reached maturity before the advent of the glacial period. In that case, the side streams must have entered the main streams Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 291 at accordant grade, and hence the main valleys could not then have been cut much deeper than the side valleys are now cut ; not so deep, indeed, for the side valleys have been somewhat deepened by glacial action, if one may judge by their trough-like form as well as by the evidence of intense glacial action all over the uplands, even over most of the surmounting hills and mountains. Hence, to develop the existing discordant valley system frpm a mature pre- glacial valley system of normal river erosion, requires a great deep- ening of the fiords by ice action, again to be measured in thousands of feet. Thus there seems to be no escape from the conclusion that glacial erosion has profoundly modified Norwegian topography. As far as I could judge from my brief excursion over the highlands, either one of the two suppositions above considered is permissible, provided only that strong glacial erosion comes after the river work of the current cycle. If the Hardanger fiord may be taken as the type of its many fellows, one may say that hanging lateral valleys are the rule, not the exception, in Norway. Furtheimore, the smoothed, spurless walls of the larger fiords, composed of firm bare rock from the upland to water edge, do not resemble the ravined and buttressed sides of normal valleys. The marks of downward water erosion are replaced by what seem to be marks of nearly horizontal pluck- ing and scouring. Blunt-headed valleys and corries (botner) both seem beyond production by normal weathering and washing. Yet, striking as these features are, they do not seem to me so compul- sory of a belief in strong glacial erosion as the hanging valleys that have so little relation to the fiords beneath them, and the flaunting waterfalls that descend so visibly from the hanging valleys, instead of retiring, as is the habit of falls all over the un glaciated pai*ts of the world, into ravines where they are hid to sight from most points of view. The rocky islands that rise from the shallower parts of the fiords should not be taken as signs of feeble glacial erosion, but rather as remnants surviving from the destruction of larger masses in virtue of some slight excess of resistance. A well-known example of this kind is near Odde at the head of the large southern arm (Sorfjord) of the upper Hardanger fiord ; but in the same neighborhood are several fine hanging valleys, one of which is shown in Plate 3, Figure Ay its open floor is high above the fiord level ; its cascading stream, the Strandfoa, descends into Sandven Lake, just south of the side valley occupied by the well-known Buer glacier. Digitized by VjOOQ IC 292 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Correlation of River Valleys a7id Olacier Valley a, — Thus far the consequences of glacial erosion have been described as if thej were unlike those of river erosion, especially in respect to the pro- duction of hanging valleys. A just comparison of the two agen- cies will show that their resemblances are more marked than their differences, when due allowance is made for their individual peculiarities. The likeness of glaciers and rivers has been frequently con- sidered. The motion of water streams and ice streams is retarded by bottom and banks, and is fastest in mid-channel where farthest removed from all hindrances. The motion is faster on strong than on gentle slopes, and in large than in small streams : the line of fastest motion departs from the medial axis towards the concave bank. Forel ('97, 204) and Gannett ('98, 422) have justly com- pared ordinary valley glaciers, not to rivers that mouth in the sea> but to rivers that descend from mountains to wither away on pied- mont deserts. The terminal moraine of the glacier corresponds to the terminal delta-like fan of withering rivers. The fluctuation of a withering river following changes of weather or season corresponds to the secular fluctuations of glaciers, as during the period of about thirty-five years in the Alps. The advance and retreat at the end of large glaciers does not occur synchronously with the advance and retreat of small glaciers, although both large and small glaciers accomplish their periodic variations of length in the same interval ; and it is probable that the same contrast obtains in withering^ rivers of different length, although I cannot find any direct state- ments to this effect. Meunier ('97, 1043) has suggested that certain peculiaV successions of drift deposits in Switzerland may be the result of the enlargement of the drift-bringing glacier by the capture of the head reservoirs of another glacier, after the analogy of rivers. Gannett and Penck (see abstracts below) have gone further still and have shown that the hanging valleys, so char- acteristic of strongly glaciated drainage systems, have a perfect analogy in the valley systems of ordinary rivers in non-glaciated areas. This comparison is so instructive that it deserves full state- ment here. The ** nice adjustment of declivities " that characterizes the main and the side valleys of a river system is found only in maturely developed valleys. The adjustment or accordance between main trunk and lateral branch obtains only with respect to the surface of the streams or to the floor of their valleys. The beds of the trunk Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 293 and the branch channels may be discordant at their junction, and this discordance will increase with the difference in volume of trunk and branch stream. Truly, the discordance of stream beds is sel- dom noted, because the beds are hidden by the streams; but if a river system were laid dry, we may be assured that the beds of the smaller tributaries would open in the banks of the main river a number of feet from its bottom. In the case of the Mississippi, the discordance might easily measure fifty or more feet. All this applies equally to glacial streams. The surface of a trib- utary glacier is adjusted to the surface of the trunk glacier that it joins ; but the depth of the beds may be very different. As long as the glaciers occupy their channels, the discordance of their beds may not be often considered, but when a climatic 'change causes the glaciers to melt away, their channels are called " valleys," and the discordance of main and lateral glaciated "valley floors" is taken as an abnormal feature. In reality the discordance is per- fectly normal to the peculiar system of ice drainage by which it was produced, however discordant it may be to the system of water drainage now in possession of the valleys. Let us compare the maturely developed channels of rivers and glaciers. ChamieU of Mature Rivers and Glaciers, — A river flows rap- idly ; and the cross-section of its channel is but a small fraction of the cross-section of its valley. The river channel is U-shaped, very broad compared to its depth, while the valley sides flare open, V-like, above the river banks. The water surface slopes steadily down-stream, but the channel bed has many small inequalities in the form of bars and basins, and the water in the bottom of the basins must ascend a little to get out of them. If the river should dry away, the deeper parts of the bed would be occupied by pools of standing water, while the bars would show lines suggestive of flowing water. The banks of the river channel are smoothly worn in nearly horizontal lines, parallel to the flow of the river current, while the sloping sides of a river valley are buttressed with spurs and scored by the down-hill ravines of descending streams. At the junction of trunk and branch streams, a moderate discordance in the level of the channel beds is to be expected ; but this is seldom con- sidered, because the channels are usually occupied by water and the beds are hidden. A glacier moves slowly, and the cross-section of its channel may be a considerable part of the cross-section of the valley that it drains. Forel estimates that the glacier of the Rhone, even where Digitized by VjOOQ IC 294 PROCEEDINGS: BOSTON SOCIETY NATURAL fflSTORY. descending its steep cascade, has only 1 : 12,000,000 of the velocity of a large river on a similar slop% ('97, 203). The glacial channel is U-shaped, broad and deep, while the valley flares open, Y-like, above the ice surface. The ice surface slopes steadily down-stream, but the bed of its channel is unevenly scoured, here rising in knobs, there sinking in hollows or basins from which the bottom ice must ascend a little as it moves forward. When the ice melts away, lakes occupy the rock basins; the rocky knobs are seen to be rounded and plucked in a manner suggestive of heavily moving ice. The banks of the channel are scoured and fluted parallel to the ice motion ; but above the ice- worn channel the flaring valley sides are ravined by descending water streams. At the junction of trunk and branch glaciers a strong discordance in the level of the channel beds may be expected ; and the discordance becomes conspicuous when the glaciers melt away and leave their " channels " to be called "valleys." Hanging side valleys are therefore appropriate as well as characteristic features of glaciated main valleys. They must come to be considered even more significant of glacial erosion than lake basins. The Cycle of Glacial Denudation, — The points of resemblance between rivers and glaciers, streams of water and streams of ice, are so numerous that they may be reasonably extended all through a cycle of denudation. Let us then inquire if glaciers may not, during their ideal life history, develop as orderly a succession of features as that which so well characterizes the normal development of rivers. The " life history of a glacier *' need not be taken only in the sense so well illustrated in the last chapter of Russell's " Gla- ciers of North America," where the glacier is called young when it is small at the beginning of a glacial climatic epoch; mature when it is largest during the full establishment of the glacial climate ; and old when it is vanishing under the re-establishment of a milder cli- mate. Let us here consider the life history of a glacier under a constant glacial climate, from the beginning to the end of a cycle of denudation, just as Russell has considered the "life history of a river " under a constant pluvial climate, in his " Rivers of North America." Thus young glaciers will be those which have been just established in courses that are consequent upon the slopes of a. newly uplifted land surface ; mature glaciers will be those which have eroded their valleys to grade and thus dissected the uplifted surface ; and old glaciers will be those which cloak the whole low- land to which the upland has been reduced, or which are slowly Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 295 fading in the milder climate of the low levels appropriate to the close of the cycle of denudation. Imagine an initial land surface raised to a height of several thou- sand feet, with a moderate variety of relief due to deformation. Let the snow line stand at a height of two hundred feet. As ele- vation progresses, snow accumulates on all the upland and highland surfaces. Glaciers are developed in every basin and trough ; they creep slowly forward to lower ground, where they enter a milder climate (or the sea) and gradually melt away. At some point between its upper heads and its lower end, each glacier will have a maximum volume. Down stream from this point, the glacier will diminish in size, partly by evaporation but more by melting ; and the ice water thus provided will flow away from the end of the glacier in the form of an ordinary stream, carving its valley in normal fashion. Some erosion may be accomplished under the upper fields of snow and n^v^, but it is believed that more destruc- tive work is done beneath the ice. The erosion is accomplished by weathering, scouring, plucking and corrading. Weathering occurs where variations of external temperature penetrate to the bed-rock, as is particularly the case between the s^racs of glacial cascades, and again along the line of deep crevasses or bergschrunds that are usually formed around the base of reservoir walls, which are thus transformed into corries (cirques, karen, botner) as has been sug- gested by several observers; scouring is the work of rock waste dragged along beneath the glacier, by which the bed-rock is ground down, striated and smoothed ; plucking results from friction under long-lasting heavy pressure, by which blocks of rock are removed bodily from the glacier bed and banks ; corrading is the work of subglacial streams, which must be well charged with tools, large and small, and which must often flow under heavy pressure and with great energy. All these processes are here taken together as " gla- cial erosion." Let it be assumed that at first the slope of a glacier's path was steep enough to cause it to erode for the greater part or for the whole of its length. Each young glacier will then proceed to cut down its consequent valley^ at a rate dependent on various fac- tors, such as depth and velocity of ice stream, character of rock bed, quantity of ice-dragged waste, and so on ; and the eroded ^A valley is understood to Include the channel that is eroded along its floor. The channel, with its bed and banks, Is therefore that i)art of a valley which is occupied by the stream. Digitized by VjOOQ IC 296 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORT. channel in the bottom of the valley will in time be given a depth and width that will better suit the needs of ice discharge than did the initial basin or trough of the uplifted surface. The upper slopes of the glacial stream will thus be steepened, while its lower course will be given a gentler descent. Owing to the diminution of the glacier toward its lower end, the channel occupied by it will dimin- ish in depth and breadth downwards from the point of maximum volume; this being analogous to the decrease in the size of the channel of a withering river below the point of its maximum volume. A time will come when all the energy of the glacier on its gentler slope will be fully taxed in moving forward the waste that has been brought down from the steeper slopes ; then the glacier becomes only a transporting agent, not an eroding agent, in its lower course. This condition will be first reached near the lower end, and slowly propagated headwards. Every part of the glacier in which the balance between ability to do work and work to be done is thus struck may be said to be " graded " ; and in aU such parts, the sur- face of the glacier will have a smoothly descending slope. Maturity will be reached when, as in the analogous case of a river, the nice adjustment between ability and work is extended to all parts of a glacial system. In the process of developing this adjustment, a large trunk glacier might entrench the main valley more rapidly than one of the smaller branches could entrench its side valley ; then for a time the branch would join the trunk in an ice-rapid of many s^racs. But when the trunk glacier had deepened its valley so far that further deepening became slow, the branch glacier would have opportunity to erode its side valley to an appropriate depth, and thus to develop an accordant junction of trunk and branch ice surfaces, although the cJuinnds of the larger and the smaller streams might still be of very unequal depth, and the channel beds might stand at discordant levels. If the glaciers should disappear at this stage of the cycle, their channels would be called valleys, and the discordance of the channel beds might naturally excite surprise. The few observers who, previous to 1898, commented upon a dis- cordance of this kind, explained it as a result of excessive erosion of the main valley by the trunk glacier; while the hanging lateral valleys were implicitly, if not explicitly, regarded as hardly changed from their preglacial form. When the trunk and branch glaciers have developed well-defined, maturely graded valleys, the continuous snow mantle that covered the initial uplands of early youth is exchanged for a discontinuous Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 297 cover, rent on the steep valley sides where weathering comes to have a greatly increased value, and thickened where the ice streams have established their courses. This change corresponds to that belween the ill-defined initial drainage in the eai-ly youth, and the well-defined drainage in the maturity, of the river cycle. It is probable that variations in rock structure will have permitted a more rapid development of the graded condition in one part of the glacial valley than in another, as is the case with rivers of water. Steady-flowing reaches and broken rapids will thus be produced in the ice stream during its youth ; and the glacial channel may then be described as " broken-bedded." But all the rapids must be worn down and all the reaches must become confluent in maturity. It is eminently possible that the reaches on the weaker or more jointed rocks may be eroded during youth to a somewhat greater depth than the sill of more resistant or less jointed rock next down stream ; and if the glacier should vanish by climatic change while in this condition, a lake would occupy the deepened reach, while the lake outlet would flow foi-ward over rocky ledges to the next lower reach or lake. Many Norwegian valleys today seem to be in this con- dition. Indeed some observers have described broken-bedded val- leys as the normal product of glacial erosion, without reference to the early stage in the glacial cycle of which broken-bedded glacial chan- nels seem to be characteristic. Truly, it is not always explicitly stated that the resistance of the rock bed varies appropriately to the change of form in a broken-bedded channel ; but the variations of structural resistance or firmness that the searching pressure and fric- tion of a heavy glacier could detect might be hardly recognizable to our superficial observations ; and on the other hand the analogy of young ungraded glaciers with young ungraded rivers seems so nat- ural and reasonable that broken-bedded glacial channels ought to be regarded only as features of young glacial action, not as persistent features always to be associated with glacial erosion. If the glaciers had endured longer in channels of this kind, the " rapids " and other inequalities by which the bed may be interrupted must have been worn back and lowered, and in time destroyed. If a young glacier erodes it« valley across rocks of distinctly dif- ferent resistances, a strong inequality of channel bed may be devel- oped. Basins of a considerable depth may be excavated in the weaker strata, while the harder rocks are less eroded and cross the valleys in rugged sills. Forms of this kind are known in Alpine valleys ; for example, in the valley of the Aar above Meiringen Digitized by VjOOQ IC 298 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. (Wallace, '96, 176) and in the lower Gastemthal near ite junctioii with the Kanderthal ; in both these cases the basins have been aggraded and the sills have been trenched by the postglacial streams. In the lower Gastemthal the height and steepness of the rocky sill, when ai)proached from upnstream, is astonishing ; its contrast to the basin that it encloses is difficult enough to explain even for those who are willing to accept strong glacial erosion. It should, how- ever, be noted that river channels also are deeper in the weaker rocks up-stream from a hard rock sill ; if the river volume should greatly decrease, a small lake would remain above the sill, drained by a slender stream cutting a gorge through the sill. If an initial depression occurred on the path of the glacier, so deep that the motion of the ice through it was much retarded, an ice-lake would gather in it. Then the waste dragged into the basin from up-stream might accumulate upon its floor until the depth of the basin was sufficiently decreased and the velocity of the ice through it sufficiently increased to bring about a balance betw-een ability to do work and work to be done. Here the maturely graded condition of the ice stream would have been attained by aggrading its bed, instead of degrading it ; this being again closely analogous to the case of a river, which aggrades initial depressions and degrades initial elevations in producing its maturely graded course. Water streams subdivide toward the headwaters into a great number of very fine rills, each of which may retrogressively cut its own ravine in a steep surface, not cloaked by waste. But the branches of a glacial drainage system are much more clumsy, and the channels that they cut back into the upland or mountain mass are round-headed or amphitheatre-like ; but the beds of the branch- ing glaciers cannot be cut as deep as the bed of the large glacial channel into which they flow : thus corries, perched on the side- walls of large valleys, may* be produced in increasing number and strength as glacial maturity approaches, and in decreasing strength and number as maturity passes into old age. As maturity ap- proaches, the glacial system will include not only those branches that are consequent upon the initial form, but certain others which have come into existence by the headward erosion of their n6v6 reservoirs following the guidance of weak structures ; thus a ma- turely developed glacial drainage system may have its subsequent as well as its consequent branches. It is entirely .conceivable, as has been suggested by Meunier, that one ice stream may capture the upper part of another. The conditions most favorable for such a Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 299 process resemble those under which river diversions and adjustments take place; namely, a considerable initial altitude of the region, allowing a deep dissection ; a significant difference of drainage areas or of slopes, whereby certain glaciers incise deeper valleys than others ; a considerable diversity of mountain structure, permitting such growth and arrangement of subsequent glaciers as shall bring the head reservoir of a subsequent ice stream alongside of and somewhat beneath the banks of a consequent ice stream. Thus glacial systems may come to adjust their streams to the structures upon which they work, just as happens in river systems. The load transported by a glacial system may at first be supplied largely by waste plucked and scoured from the beds of the glacial channels as well as by waste detached from the enclosing slopes ; but in time, when the graded condition of the chief channels is reached and their further deepening almost ceases, by far the largest share of load will be supplied from the subaerial valley sides, where weathering of the ordinary kind will ravine the slopes, thus produc- ing a topography that is strongly contrasted with the smooth walls of the glacial channels. If the initial glacial system should incise its channels so deeply beneath a lofty highland that the supply of waste from the valley sides continued to increase after the development of graded glacial channels, it is conceivable that the channel- beds might have to be aggraded for a time, as is believed to be the case \^4th river channels under similar conditions; but owing to the receipt on the glacial surface of waste from the valley sides, it is also conceivable that this analogy may not closely obtain. Toward the end of the ice stream it may well happen that the diminution of its volume and the consequent diminution of its capacity to do work will result in the aggradation of its bed by waste that cannot be carried further forward. At the same time, the outflowing river may be unable to wash away all the waste that is delivered to it, and so, for a time through later youth and early maturity, the river may act as an aggrading agent and build up a broad, flat alluvial fan, such as fronts the terminal moraines of the Alpine glaciers that once descended to the plain of Lombardy. Some response to the change thus produced in the altitude of the end of the glacier may be expected far up its channel, whose bed would thus come to be aggraded with till. Similarly, the ice sheets that spread from the Scandinavian and Laurentian highlands over the lowlands on the south changed their behavior from degrading agents in the central area to aggrading agents on the peripheral area. Hence, a belief in Digitized by VjOOQ IC 300 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. effective erosion is not antagonistic to a belief in effective deposition in the case of glaciers any more than in the case of rivers. In each case the action varies appropriately to its place in the drainage system and to its stage in the cycle. But there will be a later stage, when the wasting of the superglacial slopes reduces them to mod- erate declivity, so that the waste delivered from them decreases in quantity ; then the outflowing water stream at the end of the glacier may become a degrading agent ; the altitude of the end of the gla- cier may be slowly lessened ; and a very slow and long-continued deepening of the whole glacial channel will take place, without requiring a departure from an essentially graded condition. As the general denudation of the region progresses, the snow fall must be decreased and the glacial system must shrink somewhat, leaving a greater area of lowland surface to ordinary river drainage. When the upland surface is so far destroyed that even the hill tops stand below the 200-foot contour, the snow fields will be represented only by the winter snow sheet, and the glaciers will have disap- peared, leaving normal agencies to complete the work of denudation that they have so well begun. If a snow line at sea-level be assumed, glaciation would persist even after the land had been worn to a submarine plain of denuda- tion at an undetermined depth beneath sea-level. The South Polar regions offer a suitable field for the occurrence of such a surface. Whether glaciers of the Norwegian or of the Alpine type shall occur, is dependent partly on initial conditions, partly on the stage of advance through the cycle of denudation. If the initial form offer broad uplands, separated by deep valleys, snow fields of the Norwegian type may have possession of the uplands during the youth of the glacial cycle ; but when maturity is reached, the up- lands will be dissected, and the original confluent snow field will be resolved into a number of head reservoirs, separated by ridges. On the other hand, as the later stages of the cycle are approached, the barriers between adjacent reservoirs will be worn away, and they will tend to become confluent, here and there broken only by Nunatuker. If the snow line lay low enough, a completely con- fluent ice and 8no\<^ shield would cover the lowland of glacial denu- dation when old age had been reached. If the glacial conditions of Greenland preceded as long as they have followed the glacial period over the rest of the North Atlantic region, who can say how far the ice of the Greenland shield has modified the forms on which its work began I Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 301 Glacial Distributaries, — If a maturely dissected mountain range were occupied by snow-fields and glaciers of large size, certain peculiar results might be expected near the mountain base. Under normal preglacial conditions, a small low ridge sufiices for the com- plete separation of two river systems, because the channels of rivers are so small in comparison to their valleys. But glacial channels are a large part of their valleys, and when great glaciers from the lofty mountain centres descend by the master valleys to the mountain flanks or even to the piedmont plains, distributary ice streams or outflowing branches may naturally enough be given off wherever the ice surface rises high enough to overtop the ridges by which the master valleys are separated from adjacent minor valleys. If a dis- tributary branch has suflicient strength and endurance, it may wear down the ridge that it crosses and thus increase and perpetuate its lateral discharge ; but it cannot usually be expected to erode a channel as deep as that of the main glacier from which it departs. On the disappearance of the ice, a hanging valley will be left above the floor of the master valley ; but in this case, the drainage of the hanging valley will be away from, not toward, the master. Here we probably have the explanation of those broad hanging valleys which lead from the valley of Lake Maggiore on the west and, less distinctly, from that of Lake Como on the east to the compound basin of the intermediate Lake Lugano. On going southward by rail from BeUinzona to Lugano, along a stretch of the St. Gotthard route between the great tunnel and Milan, the railway obliquely ascends the southeastern wall of the trough-like valley of the Tieino just above the head of Lake Maggiore ; and at a height of several hundred feet over the delta flood-plain the line turns off to a well- marked hanging valley in which the stream runs away from the Tieino to Lake Lugano. The notch made by this supposed glacial distributary is a conspicuous feature in the view from BeUinzona and thereabouts. The anomalous forking of Lake Como and the open branch from the main valley of the Rhine at Sargans through the trough of Wallen See to Lake Zurich appear to be the paths of large glacial distributaries which eroded their channels deeply across divides that presumably existed in preglacial time. The west wall of the main valley of the Is^re in the Alps of Dauphiny, southeastern France, is deeply breached by passes that lead northwest to the troughs of Lakes Annecy and Bourget, through which the distributaries of the Is^re glacial system must have flowed. Lug^on ('97, 62-70) has Digitized by VjOOQ IC 302 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. explained the breaches as marking the former northwest paths of transverse members of the Is^re system, from which they have been diverted by the subsequent growth of the main longitudinal valley — that northeast- southwest part known as the Gr^sivaudan — above Grenoble. HLs discussion of the problem takes, however, no account of modifications of valley depth by glacial erosion ; and as this must have been considerable (for the valleys hereabouts have superb basal cliffs, as appears in the valley of the Romanche by Bourg d'Oisans in Plate 3, Fig. J5, after a photograph by Neurdin Fr^re^ of Paris), it may well be that the rearrangement of river courses in this interesting region is not altogether the work of river action. Similarly, the various modifications of the Rhine system in eastern Switzerland, explained by Ileim as the work of streams alone, may come to be at least in part referred to ice erosion. It may be further supposed that if the preglacial valleys were so arranged that a glacial distributary found a shorter and steeper course to the piedmont plain or to the sea than that followed by the master glacier, the distributary^ might under a long enduring glaci- ation become the main line of glacial discharge ; and if so, it could be eroded to a greater depth than the former master valley at the point of divergence. In such a case, the postglacial river drainage would differ significantly from the preglacial. There is reason for believing that examples of this kind are to be found in Norway, the evidence of which will soon be published in an essay by Barrett (1900). The diversion of the head of a stream in the Sierra Costa of northwestern California to a deei)er-l\ing valley through a gorge cut by a glacial distributarj^ has lately been described by Hershey (1900, 47). The Depth of Mature Glacial Chaiinels, — The depth with respect to sea-level to which the channels of a glacial system may be eroded when the graded condition is reached, is a subject of special interest. ' For many miles along the lower course of a branchless trunk glacier, its volume is lessened by melting and evaporation, and at its end the ice volume is reduced to zero ; slow ice motion being progressively replaced by rapid water motion. In such a case the law of continuity does not demand that the ice velocity shall be inversely proportional to the area of the cross section, as is the case in the normal river (where it is assumed that there is no loss by evai)oration). Indeed, in the lower trunk of a mature glacier, it may well be that the velocity of ice movement is in a rough way directly proj)ortional to cross-section area. This appears to be veri- Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 303 fied by meafluremente of the Rhone glacier, where the mean annual movement is 110 met. in the heavy trunk above the cascade, 27 met. just below the cascade, and only 5 met. close to the melting front (Forel, '97, 203) . Evidently then, the erosion of the glacial bed, in so far as it is determined by the pressure and motion of the ice stream, will have its maximum some distance up-stream from the end of the glacier (J. Geikie, '98, 236). The glacial channel must therefore become narrower and shallower as its end is neared, as has already been stated. If the glacier ends some distance inland from the sea, its action will be conditioned by the grade and length of the river that carries away the water and waste from its lower end. The deepening of the distal part of the channel accomplished in youth might be followed by a shallowing for a time during matu- rity, when the accumulation of mor^nal and washed materials in front of the glacier compelled its end to rise. Now it may well be conceived that the surface slope of such a glacier near its end is less than the angle between the surface and the bottom of the glacier ; and in this case, the glacial floor must become lower and lower for a certain distance up-stream. If such a glacier should. melt away, the distal part of its channel would be occupied by a lake, although even the head of the lake may not reach to the locus of maximum glacial erosion. Lakes Maggiore, Como and Garda seem to occupy basins whose distal enclosure by heavy moraines and sheets of over- washed gravels has added to the depth produced by erosion further up-stream. It would ^eem, however, that a lake basin thus situated must be only a subordinate incident in the general erosion of the whole length of the glacial channel. Too much attention has, as a rule, been given to lakes of this kind, and not enough to the other effects of glacial action ; it seems especially out of proportion to sup- pose that the maximum erosion by a glacier takes place near its end, as has been done by some authors, on account of the prevalent occurrence of lakes in this situation. If a glacier advances into the sea and ends in an ice cliff, from which ice blocks break off and float away, something of a basin-like form of its lower channel may be produced ; but the dimensions of this basin will be determined by the climate at the termination of of the glacier. If the climate is such as to allow the glacier to enter the sea in maximum volume, then a basin is not to be expected. The more the glacier diminishes towards its end, the less erosion and the more deposition may occur beneath it, and the more of a basin may be developed inland from its end. Digitized by VjOOQ IC 304 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORT. The depth to which a glacier may cut its channel when it enters the sea is of particular importance. If the glacier is a thousand feet thick at its end, it must continue to press upon and scour its bed until only about 140 feet of ice remain above sea4evel ; ita channel will thus be worn more than 800 feet beneath sea-level. Truly, the latter part of this work will be performed with increasing slowness ; but if time enough be allowed the work must be accomplished, just as is the case with rivers. If a glacier should melt away from its deep entrenchment, its channel would be occupied by an arm of the -sea or fiord, reaching many miles into the land. The fiord might be shallower at its mouth than further inland, if differential erosion and deposition had occurred along its channel. Yet even this result is analogous to the case of a river ; for if the Mississippi were to dis- appear in a prolonged drought, a slender arm of the sea would invade the river channel many miles up-stream from the delta front. Indeed, the Mississippi offers an excellent example of a channel that is ba- sined inward from the river mouth ; for while it is only a score of feet deep at the passes where most of its sediment is deposited, it is several score of feet in depth further up-stream ; and the slender arm of the sea that would occupy its channel if it should disappear by climatic change, would be truly fiord-like in having a less depth at its mouth than further inland. An important corollary from this conclusion — perhaps not so much of a novelty to glacial erosionists as to their confreres of the opposite oi)inion — is that the depth of water in the fiords of a strongly gla- ciated coast is not a safe guide to the movement of the land since preglacial time. If there had been a still-stand of the earth's crust through the whole glacial period, the preglacial river channels that were graded down a little below sea-level at their mouths would be replaced by glacial channels that might be eroded hundreds of feet below sea-level. The depth of fiords thus seems to depend on the size of their ancient glaciers, on the height of the mountain back- ground, and on the duration of the glacial period, as well as on movements of the land. If liberal measures of glacial erosion and glacial time are allowed, no depression of glaciated coasts sinc^ pre- glacial time' is needed to account for their peculiar features. The glacial channels may have been simply invaded by the sea, as the ice melted away, without any true submergence. Even the advocates of strong glacial erosion do not seem to have expUcitly recognized the full importance of this possibility. James Geikie, for example, writes : " The fiords of high latitudes and the Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 305 narrow inlets of non-glaciated lands are simply submerged land- valleys; the intricate coast-lines of such regions have been deter- mined by preceding subaerial denudation." Again: "In a word, fiords are merely the drowned valleys of severely glaciated moun- tain-tracts." ('98, 263, 250.) The deep waters in the valley of the Hudson through the Highlands of southeastern New York are the most fiord-like in the eastern United States, and they are universally explained as the result of submergence of a normal river valley ; but the constricted ice current that must have flowed through the High- land gorge may have been energetic enough to deepen its bed beneath sea-level, and since the ice melted away, who can say how much submergence beneath preglacial levels has taken place. I do not know how far this view of the matter has been taken by ear- lier advocates of strong glacial erosion, but for my own part, the acceptance of such a possibility means a complete reversal of the belief that I held two years ago. The reversal is, however, accom- panied by the memory that it was always difficult to understand why submergence and glaciation were so closely associated : even if glaciation had caused depression, it was diflicult to understand why the relief from ice pressure in postglacial time had not now been followed by a rise of the land much nearer to its former altitude than would be the case if the greater part of the depth of fiords is explained by submergence. The Origin of Carrie Hasins. — On pursuing the above line of consideration a little further, it may give some light on the occur- rence of the small rock basins that are so often found in the floor of cliff-walled corries. Imagine that a large glacial system has become maturely established, and that it " rises " in many blunt head- branches that have excavated corries in a preglacial mountain mass, and have cut down channels, at their junction with the larger branches or trunk glacier, to a depth appropriate to their volume. Unless the erosion of the corries has been guided by differences of rock structure, there does not seem to be reason for their possessing a basined floor at this stage of development; but if a change of climate should now cause the trunk glacier to disappear, while many of the blunt head-branches remain in their corries, each little glacier thus isolated will repeat the conditions of erosion above inferred for the tnink glacier ; and if this style of glaciation linger long enough, rock basins may very generally characterize the floors of the corries when the ice finally melts away. Figure 7 may make this clearer. Let the broken line, ABC, be the slope of a preglacial lateral ravine Digitized by VjOOQ IC 306 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. which reaches a trunk stream at C, while ADC is the profile of an adjoining lateral spur. After vigorous and mature glaciation, the dotted line, GE, may represent the surface slope of a lateral glacier, and GHJ that of the lateral gla<*ier bed ; while EFL is the surface of the trunk glacier, and EKL the bed. The lower part of the lateral spur has been cut off to make the basal cliff beneath D. On Fig. 7. Diagram-section of a lateral valley with a corrie basin. the disappearance of the trunk glacier at this stage, the shrunken side glacier, GNJH, occupies its corrie or hanging valley, which opens at J on the oversteepened wall, DJK, of tlie evacuated channel of the trunk glacier. Let the maximum erosion of the corrie glacier, as conditioned by pressure and motion, be at H. Then after some time the weathering of the cliff walls and the erosion of the floor will have transformed the corrie and its glaoier to a form, G'N' J'H', such that the deepening of the glacial bed should be a maximum at HIP. The continuous slope of the glacial bed, GHJ, appropriate to the time when the lateral glacier joined the trunk glacier, may thus be transformed into a basined curve, G'H'J', appropriate to a small glacier terminating at J' ; and on the disap- pearance of the small glacier, a tarn or rock-basin lake may occupy the depression at II'. It is on the basis of a supposition like this that a determination has been attempted of the altitude at which the shrinking remnants of an extensive glacial system endured for a time before their entire disappearance (J. Geikie, *98, 233). Richter's supposition that the uplands of Norway result from the Digitized by Google DAVIS: GLACIAL EROSION. 307 consumption of pre-existent mountains by the great extension of corrie-glacier floors, each similar to J' H', thus seems mechanically possible ; but it is nevertheless climatically very improbable, and it seems to me deficient in not attributing enough work to normal pre- glacial erosion. Overdeepened Valleys and Oversteepened Walls, — As in the case of the normal cycle of denudation in which the life history of river systems is involved, so in the glacial cycle, all manner of com- plications may arise, causing great departures from the ideal case. The assumed initial land form may be a surface previously more or less dissected by river erosion, on which glaciers must then proceed to develop a drainage system appropriate to their own peculiar needs, as has been partly considered above in connection with glacial distributaries. It will be instructive to make out a good series of examples illustrating different combinations of river and glacial action, and including young, mature and old river valleys, modified by young or mature glaciation. For example, the existing vaUey of the Rhue in the Central Plateau of France shows a submature river valley with incised meanders, moderately affected by young and relatively light glaciation ; the valley of the Ticino in the south- ern Alps is a well-matured preglacial river-valley system, modified by strong submature glaciation. The fiords of Norway result from the submature and intense glaciation of a river-valley system whose stage of preglacial development is not yet well determined. Interruptions of regular progress in the glacial cycle must, as in the river cycle, be occasioned by elevation, depression, or deforma- tion of the land mass ; but no examples of complications of this kind can be adduced. Variations of climate may replace creeping glaciers in young, mature or old stages of development, by flowing rivers ; and the early stages of such rivers are of much importance among existing geographic forms. Lakes, delaying the river flow, occupy the depressions of the glaciated surface, as has been known since Ramsay first pointed out the correlation of lacustrine and glaciated regions in 1861 ; but the analogy between lakes in the beds of melted glaciers and pools in the beds of dried-up rivers has perhaps not been suiRciently insisted upon. Waterfalls connect the streams that occupy the discordant beds of glacial channels, as has lately been clearly set forth. Landslides frequently occur after the sup- porting glacier withdraws from the oversteepened banks of its huge channel ; fallen masses of this origin have been repeatedly mistaken for moraines in Alpine valleys, as has been lately shown by Brtlck- Digitized by VjOOQ IC 308 PROCEEDINGS: BOSTON SOCIETY NATURAL mSTORY. ner. Every lake or fiord is an effective lowering of base-level for the stream above it ; for the level of a body of standing water is essentially the same at both ends. As fast as the inflowing river builds its delta forward at the head of the lake or fiord, its flood plain must rise up-stream and aggrade the valley floor. This process is very j)ronounced in many Alpine valleys, where the aggraded valley floor has a relatively rapid descent on account of the plentiful and coarse detritus furnished by the active side streams. Indeed, every ravine furnishes a great quantity of rock waste, whose descent is analogous to rej)eated landdides of small dimensions. The valley floor beneath the ravines is invaded by great alluvial fans, and the main stream is driven away toward the further valley wall by their rapid advance. At every flooen upon these from the fjelds. At that time the work of rain and running water could not have been carried on equally over the whole land, otherwise we should find now a completely developed hydro- graphic system — not a plateau intersected by profound chasms, but an undulating mountain-land with its regular valleys According to Dr. Richter, the remarkable contrast between the deep valleys of the fiords and the shallow side-valleys that open upon them from Digitized by Google DAVIS: GLACIAL EROSION. 317 the f jelds — the profound erosion in the former, and the arrest of erosion on the plateau — admits of only one explanation. While rivers and rapid icenstreams, flowing in previously excavated valleys, were actively engaged in deepening these, the adjacent f jelds were buried under sheets of n4v4 In short, while rivers and glaciers were deepening the great valleys and making their walls steeper, the intervening mountain-hfeights were gradually being reduced and levelled by denudation It was somewhat otherwise in the Alps, where the hydrographic system, perfectly regular in preglacial times, was only slightly modified by subsequent glacial action. Yet even there erosion proceeded most rapidly along the chief lines of ice-flow. Were the great rock-basins of the principal Alpine valleys pumped dry we should find the mouths or openings of the side valleys abruptly truncated, and their waters cascading suddenly into the ice-deepened main- valleys. For, as Dr. Wallace has shown, it is the present lake-surfacey not the lake-^oWom, that represents approximately the level of the preglacial valley. In a word, erosion proceeded most actively in the main valleys, the bottoms of which have been lowered for several hundred feet below the bottoms of the side-valleys. Precisely the same phenomena are repeated in Scot- land. Were all the water to disappear from the Highland lakes and sea-lochs, we should find waterfalls and cascades at the mouth of every lateral stream and torrent " ('98, 246-249). It is evident from these extracts that the deepening of valleys is regarded as greatest where lake basins have been eroded beneath the preglacial valley floors ; and this belief is explicitly expressed in the following extract from the latest edition of the same author's " Great Ice Age," the standard work on that subject : '* Take the case of a glacier creeping down an Alpine valley and spreading itself out upon the low ground at the foot of the mountains. Let us suppose that, in the upper part of its course, the incline down which it moves is greater than the slope of the lower reaches of the valley. When the glacier attains the more gently inclined part of its course, it is evident^that its flow must be retarded, and there will therefore be a tendency in the ice to accumulate or heap uj). Now we know that the pressure of a body in motion upon any given surface varies with the degree at which that surface is inclined ; as the inclination decreases the pressure increases. It follows from this that when the glacier leaves the steeper part of its course, and begins to creep down the gentler slope beyond, it will press with greater force upon its rocky bed, and this increased pressure will be further intensified Digitized by VjOOQ IC 318 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. by the greater thickness of the accumulated ice The result of all this is the formation of a rock-basin, the deeper portion of which lies towards the upper end, just where the grinding force of the glacier is greatest" ('95, 228, 229). It seems to me that too great emphasis is here placed on the ero- sion accomplished near the end of a glacier, as indicated by lakes, and not enough upon the deepening of the valleys up-stream from terminal lakes, as indicated by hanging valleys. It is also to be noted that de Lapparent, Richter, and J. Greikie all describe the hanging valleys of Norway as if their preglacial form had not been significantly changed, thus failing to bring clearly forward the fact that the valleys of today are the ice channels of the past, and that the larger and smaller channels must have normally discordant floors in a system of glacial drainage, just as they have in a system of river drainage, although to a much greater degree. The full analogy between ice and water channels, which throws so much light on the whole question of glacial erosion, was first clearly set forth by the two following observers. Gannett on Lake Chelan^ 1898, — The most complete statement of the general principles involved in the production of hanging valleys that I have found in print is in an article on Lake Chelan, in the Cascade Mountains of Washington, by Henry Gannett. Chelan is a long narrow lake occupying the distal two thirds of the deep U- shaped valley of the Stehekin River on the eastern slope of the mountains. It was occupied in the glacial period by a heavy ice stream, fifty or sixty miles long, and half a mile to a mile broad. The rock walls which enclose the valley are strikingly parallel to one another, without buttressing spurs ; they rise 4000 to 5000 feet above the lake waters. Nearly all the streams which flow into the valley tumble over its walls in a series of cascades. " From all indications it appears that the ice must have been at least 3000 feet deep in this gorge of the Stehekin, since several of the smaller branches joinr the main glacier at that height above its bed.'* Speaking of these features in a more general way, Gannett says : " A glacier is a river of ice, and it behaves almost precisely as a river of water does. Its effects upon its channel are almost pre- cisely similar to those of a river upon its channel, excepting in the fact that all its operations are on a vastly greater scale A word of caution must here be interpolated. The channel of a river, in wliich its water flows, must not be confused with its valley, which it drains. ^Phe above comparison refers to the channel of a river Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 319 [or of a glacier], not to its valley The glacier moves down the gorge, scouring and cutting the bottom and sides as it travels. The ends of the mountain spurs are planed off instead of being trimmed to sharp, angular points, as is done by streams in gorges cut by them Where the main glacier is joined by a branch, the bed of the branch is commonly found to be at a higher level than the bed of the main glacier, because being larger and heavder the main glacier has greater cutting power ; indeed, in many cases the beds of small branches are hundreds, or even thousands, of feet higher than that of the main glacier to which they are tributary. The parallelism between the glacier and the river in their channels is further illustrated by this fact. The surface of the ice in the main glacier and in the branch must have been at the same level, although the bottoms, as stated above, differ greatly in elevation. So it is with a river at the point of junction of branches. The surface of the water must be practically at the same level in all cases, but the bottoms of the channels differ by the difference in depth of the streams at their point of junction. This fact affords us a measure of the minimum thickness of the ice at any place. It cannot have been less than the vertical distance between the bed of the main glacier and that of the tributary, and, indeed, must in all cases have been greater, owing to the thickness of the tribu- tary" ('98, 417-428, especially 418-420). Penck on Alpine VcUleya^ 1899, — A no less explicit and detailed statement of the peculiar features of glacial channels and their rela- tions to river channels was made by Penck at the meeting of the International Greographical Congress in Berlin, September, 1899. The discordance of lateral and main valley floors wa« described as a general feature of all the larger Alpine valleys within their glaciated areas. The possibility of explaining the discordance by faulting, as suggested by Rothpletz for the Linththal, was considered, but rejected. The contrasts of the glaciated and non-glaciated Alpine valleys were strongly emphasized. The excess of the depth in the main valley beneath the floor of the hanging ^laterals was taken as a minimum measure for glacial erosion, and the term " over-deepened," already adopted on earlier pages of this essay, was applied to valleys thus worn to a greater depth than would have been possible to normal rivers. The publication of Penck's address is awaited with interest. Harker on Glacial Valleys in Skye, 1899. — A brief article by Harker on glaciation in Skye describes the valleys as eroded in massive gabbros, with U-shaped cross-section, especially in the upper Digitized by VjOOQ IC 320 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. stretches, and frequently heading in a corrie whose floor may hold a small rock-basin tarn. In longitudinal profile, the floor of a valley often consists of two or three stretches of relatively gentle slopes, or sometimes of basin form and then holding lakes, separated by relatively sudden descents. Tributary valleys mouth at a consid- erably higher level than the floor of the main valley, (*99, 196-199). Gilbert on Alaskan Valleys^ 1899. — A valuable contribution to the origin of hanging valleys will be found in a report on the Harri- man Alaskan Expedition of 1899, to be published shortly. A gen- eral statement of results wa« made by Gilbert during the session of the Geological Society of America in Washington, December, 1899, when the importance of the hanging lateral valleys in the Alaskan fiords, and their bearing on the problem of glacial erosion, was clearly set forth. Blanford on Scotch GlenSy 1900. — The only article that I have found on hanging valleys in Scotland is by Blanford, " On a partic- idar form of surface, apparently the result of glacial erosion, seen on Loch Lochy and elsewhere." The '* particular form " here referred to is the smoothness of the sides of the Great Glen of Scotland, a feature that may be held analogous to the smooth rock walls of the Norwegian fiords, and to the spurless basal cliffs of the glaciated Alpine valleys. It is inferred that in preglacial time the streams of the lateral glens were separated by advancing spurs which buttressed the sides of the Great Glen. Now the spurs seem to have been truncated, producing the smooth and even sides of the glen, to which attention is esj>ecially directed. The lateral glens are described as at present opening a thousand feet above the floor of the Great Glen, whose smoothed sides are very little eroded by the descending tributary streams. The change from the inferred preglacial form is conservatively taken to indicate a glacial erosion of "at least 250 or 300 feet of rock " (1900, 198-204). Hershey on Sierra Costa^ California^ 1900. An article by Her- shev, already referred to above, is the latest contribution to the sub- ject in hand. In following uj) a valley in the Sierra Costa in northwest California, it is at first V-shaj>ed, with jagged ledges between sharjMJut ravines on the sides, and hardly wider at the bottom than the stream that drains it. On reaching the stretch once occupied by a local glacier, the valley becomes an open U- shaped trough, with smooth slopes free from ravines and spurs. Above the limit of glacial smoothing, the mountain sides are still deeply scored with ravines and jagged with outcropping ledges. Digitized by VjOOQ IC DAVIS: GLACIAL EROSION. 321 The descent of a glaciated valley floor is effected by a series of steps ; the stretches of more gentle fall alternate with almost pre- cipitous falls where the floor is let down several hundred feet, Corries with tarns in their floors are well developed (1900, 42-57) ^ Several essays by Norwegian authors remain to be considered. It has not been possible to make reference to them without postponing the appearance of this paper, and consideration of them is there- fore deferred to another occasion. With all these new contributions to the subject, it may be expected that hanging lateral valleys and overdeepened main valleys will soon gain the importance that they deserve in geographical literature. LITERATURE. Barrett, R. L. 1900. The Sundal drainage system in central Norway. Bull. Amer. geogr. soc, 32, (in press). Blanford, W. T. 1900. On a particular form of surface, apparently the result of glacial ero- sion, seen on Loch Lochy and elsewhere. Quart journ. geol. soc, 66, pp. 198-204. Boule, M. '96. La topographic glaciaire en Auvergne. Ann. de g^ogr., 5, pp. 277- 296. BrUckner, E. '85. Die Vergletscherung des Salzachgebietes. Geogr. Abhandlungen, Vienna, 1, pp. 1-188. Davis, W. M. 1900. Glacial erosion in the valley of the Ticino. Appalachia, 9, pp. 136-156. Forbes, J. D. '53. Norway and its glaciers. Edinburgh. Forel, F. A. *97. Fleuves et glaciers. Bull. soc. vaud. sci. nat., 33, pp. 202-204. Gannett, H. '98. Lake Chelan. Nat. geogr. mag., 9, pp. 417-428* Geikie, J. '95. The great ice age. New York. 3d ed. '98. Earth sculpture. London. Digitized by VjOOQ IC 322 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Hall, J. *43. Geology of New York. R. 4, comprising the geology of the fourth dis- trict. Harker, A. '99. Glaciated valleys in the Cuillins, Skye. Geol. mag., 6, pp. 196-199. Helm, A. '78. Mechanifimus der Gebirgsblldung. BaseL '79. Ueber die Erosion im Gebiete der Reuss. Jahrb. echw. Alpenclab, 14, pp. 371-405. Hershey, O. H. 1900. Ancient alpine glaciers of the Sierra Costa Mountains in California. Journ. geol., 8, pp. 42-67. Lapparent, A. de. *96, *98. Le9on8 de geographic physique. Paris. 1st ed., 1896 ; 2d ed., 1898. Lubbock, J. *96. The scenery of Switzerland. New York. Lug6on, M. '97. t«e9on d'ouverture du cours de gfeographie physique profess^ a I'uni- ver8it6 de Lausanne. Bull. soc. vaud. sci. nat., 33, pp. 49-78. McGee, W. J. '83. Glacial canyons. Proc. Amer. assoc, 1883, p. 238. '94. Glacial canyons. Journ. geol., 2, pp. 350-364. Marr, J. E. 1900. The scientific study of scenery. London. Meunier, S. '97. Sur r allure g^n^rale de la denudation glaciaire. Comptes rendus, 124. p. 1043. Play fair, J. *02. Illustrations of the Huttonian theory of the earth. Edinburgh. Richter, E. '96. Geomorphologische Beobachtungen aus Norwegen. Sitzungsber. k. k. Akad. Wien, math, naturw. Classe, 105, Abth. 1, pp. 147-189. Rothpletz, A. '98. Das geolektonische Problem der Glamer Alpen. Jena. Russell, I. C. '89. Quaternary History of Mono Valley, California. 8th ann. rep. U. S. geol. surv., 1889, pp. 261-394. RUtimeyer L. '69. Ueber Thai- und See-Blldung. Basel. Tarr, R. S. *94. Lake Cayuga a rock basin. Bull. geol. soc. Amer., 5, pp. 339-356. Wallace, A. R. '93. The ice age and its work. Fortnightly rev., 60, pp. 616-633, 760- 774. '96. The gorge of the Aar and its teachings. Fortnightly rev., 66, pp. 175-182. Printed July^ 1900, Digitized by VjOOQ IC Digitized by VjOOQ IC Da VIA. — Glacial Erosion. PLATE 1. Fig. A. Valley of the Ticino, looking up-stream to Giomico. Fig. B. Cliffs beneath graded slopes, eastern arm of Lake Lugano. Digitized by VjOOQ IC a: .as % y. i S I J 1 Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Da viB. — GlftcUU Erosion. PLATE 2. Fig. A, Cascade at tlie mouth of a hanging valley, eastern arm of Lake Lugano. Fig. J5. Sterling Fall, Milford Sound, New Zealand. Digitized by VjOOQ IC 3J ■I. I H X 0 c a; 0. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Da VI*. — Glacial Erosion. PLATE 3. Fig. A. The Strandfos, a cai^cade descending from a hanging valley to Sandven Lake, Norway. Fig. B. Valley of the Romanches Alps of Dauphiny. Digitized by VjOOQ IC J h X Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Boston Society of Natural History. RECENT PUBLICATIONS. Mjemoirs. 4tO. Description of the human spines showing numerical variation, in the Wtrreo^ Museum of the Harvartl Medical School. By Thomas Dvrigfat 70 pp. $1.50. The anatomy and development of Cassiopea xamacbana. By Robert Payne Bigelow. 4(5 pp., 8 plates. $1.40. The development, structure, and affinities of the genus Equlsetum. By Edward C. Jeffrey. 36 pp., 6 plates. $1.00. Localized stages in development in plants and animals. By Robert T. Jackson. 05 pp., 10 plates. $2.00. Proceedings. 8vo. Glacial erosion in France, Switzerland and Norway. By William Morris Davis. 50 pp., 8 plates. 50 cts. The embryonic history of imaginal discs in Melophagus ovinus L., together with an account of the earlier stages in the development of the insect By H. 8. l*ratt. 32 pp., 7 plates, 75 cts. A revision of the systematic names employed by writers on the morphology of the Acmaeidae. By M. A. Willcox. 0 pp. 10 cts. On a hitherto unrecognized form of blood circulation without capillaries in the organs of vertebrata. By Charles Sedgwick Minot. 31 pp. 35 cts. The occurrence of fossils in the Roxbury conglomerate. By Henry T. Burr and Robert E. Burke. 6 pp., 1 plate. 20 cts. The blood ves-sels of the heart in Carcharias, Raja, and Amia. By G. H. Parker and F. K. Davis. 16 pp., 3 plates. 25 ctsi. List of marine mollusca of Coldspring Harbor, Long Island, with deacriptions of one new genus and two new species of nudibranchs. By Francis Noyes Balch. 30 pp., \ plate. 35 cts. The development of Penilia schmackeri Richard. By Mervin T. Sadler. 23 pp., 8 plates. 30 cts. Contributions from the Gray herbarium of Harvard university. Ne^v series, no. 17. 1. Revision of the genus Gymnolomia. 2. Supplementary noiei upon Calea, Tridax, and Mikania. By B. L. Robinson and J. M. Greenman. 22 pp. 25 cts. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the United States. By Garry de N. Hough. 10 pp. 10 cts. Notes on the reptiles and amphibians of Intervale, N. H. By Glover M. Allen. 13 pp. 15 cts. Variation and sexual selection in man. By Edwin Tenney Brewster. 17 pp. 25 cts. Moniloporidae^ a new family of Palaeozoic corals. By Amadeus W. Grabau, 10 pp., 4 phues. 25 cts. Studies in the gold-beaiiug slates of Nova Scotia. By J. Edmund Woodman. 33 pp., 3 plates. 50 cts. North American wood frogs. By Reginald Heber Howe, Junior. 6 pp. 10 cts. Some Hydroids from Puget Sound. By Gary N. Calkins. 35 pp., 6 i>lateB. 50 ct«. The Odonate genus Macrotheniis and its allies. By Philip P. Calvert 32 pp., 2 plates. 50 cts. On the veins of the Woltlian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 25 cts. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 15 cts. The genus Autennaria in New England. By Merritt L. Femald. 13 pp. 15 cts* The land mammals of peninsular Florida and the coast re^nof Georgia. By Oiuram Bangs. 70 pp. 75 cts. A contribution to the petrography of the Boston Basin. By Theodore G. White. 40 pp., 5 plates. 05 cts. Clymene producta sp. nov. By Margaret Lewis. 5 pp., 2 plates. 16 cts. The Harvard geographical models. By W. M. Davis. 26 pp., 4 plates. 25 cta^ The role of water in growth. By C. B. Davenport. 12 pp. 15 cts. Digitized by VjOOQ IC ^ tm -•< 1 1 ■1 ^H PioceedlilgB or the Bootoii Sooi«ty of Natural Hii ^^^^^^^^^^^H ^ft \'»t. ■ilt, N... i.\ ^^^^^^^1 ■ p- HHi -437. ^ ■ TIW PftLVniAF/rA nr TIJE rrtiKT SiH'NI* UEfilnN U\ fli.i:i(ki{i P\iu.«^ .luiTN>n9( With ^'^iMKTRRN i*i,vriJ» l*KtNTi:i> FiMi THE HiHIETY. Digitized by Google . Digitized by Google AUG 8d 1901 No. 18. — The Polychaeta of the Puget /Sound Region. By Herbert Parlin Johnson. The following account of the Polychaeta of Puget Sound and neighboring waters is primarily based upon a collection made by Nathan R. Harrington, lately deceased, a member of the Columbia University Zoological Expedition to that region in the summer of 1896. The collection, comprising thirty-four species belonging to seventeen families, was sent to me in January, 1898. Preliminary examination showed that the collection, although meagre, in part poorly preserved, and almost destitute of data, contained much of interest. It seemed best, however, to defer any publication of results until more and better material could be obtained. Thanks to the good efforts of several collectors, notably Prof. William E. Ritter and Miss Alice Robertson, both of the University of Cali- fornia, very substantial additions have been made to the original collection. These, together with two species from Victoria, B. C, kindly contributed by Prof. William A. Herdman, raise the total number of species to fifty-one, distributed in thirty-four genera and tw^enty-six families (see Table, p. 384) — practically all the Poly- chaeta known to occur in the Puget Sound region.^ They cannot reasonably be supposed to rej)re8ent more than a fourth or even a fifth of the actual Polj'chaete fauna. Nearly all the collecting thus far has been between tide-marks. Dredging has been small in amount and limited to very moderate depths (not over thirty fath- oms) ; and the entire absence of pelagic forms — with the possible exception of Arichhopsis megnlops — would indicate but slight use of the tow-net. 1 Ehlers ('68) describes eight species from the Pacific Coast ; live of these [Nereis agasfizi, N. procera, JV. vejrillosa, X. virens and XepMhys coeea) have been found in Puget Sound or vicinity. Balrd ('63) describes nine species, all from Esquimalt Harbor, Vancouver- Island. Four of these. Polynoe ( Lepldonohui) itisignls, lordi, frur/Uia, and XerelsfoHata (= X. virens Sars) have been identified, and an account is given of them. Lepidonotxis gntbei Is in all probability the same as Pohjtioe itinhjiiUs. There remain unidentified only four of Baird's species (Hannot/ute unicolor, Xereis hicamilicufata, Glycera cor- nigata, &nd Sabellaria aaxicara^ These it is practically impohsible lo identify with certainty from Baird's descriptions. Digitized by VjOOQ IC 582 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Although comparison has been carefully made with the descrip- tions of Japanese Polychaeta given by Grube ('77) and by von Marenzeller (79 and '84), no species, with the sole exception of the widely ranging Ilarmothoe imbricata, has been found to inhabit both sides of the Pacific. Thb is not surprising when we consider that nearly all of the sixty-two species of Grube's and Marenzeller's lists are from the southern portions of Japan, and therefore belong to the Indo-Pacific fauna, not to the circumboreaL But, on the other hand, comparison of Puget Sound Annelids wi^ those of Bering Sea (Wir^*n, '83; Marenzeller, '90) also shows absolute dissimilarity. So far as known, the Polychaeta of the more northern parts of Bering Sea — the only ones of that region dealt with by Wir^^n and von Marenzeller — are practically those of the Arctic and North Atlantic (i. e., are circumboreal) with very little ^mixture of forms peculiar to the North Pacific. In Wir^n's list -of twenty-nine species there is not one which can be regarded as ibelonging distinctively to the North Pacific. Yon Marenzeller's list of twenty-four (exclusive of eight previously given by Wiren) affords only two new species; the others are well-known inhabi- tants of the shores of Scandinavia, Iceland, Greenland, or north- hical Distribution. 2. 3. 4. 5. a. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1«. 17. 18. 10. 20. 21. 22. 23. 24. 25. 2(1 27. 28. POLYKOIDAE. Polynoe (Lepidonotus) squamata (L.) Aud. and M. -Edwards. P. insi^U (Baird) P. lordi (Baird) P. pulchra Johnson P. fragilis (Baird) Harmothoe imbricata (L.) Malm- gren. H. ipiiionelloides sp. nov. H. complanata sp. nov. H. paciflca sp. nov. H. tuta (Grube) SiGALIONIDAE. Sthenelais fiisca Johnson Hesionidak. Podarke pugetteiisis sp. nov. NivKEIDAE. Nereis virens M. Sars N. vexillo.sa Grube N. agassizi Khlers X. cyclurus Harrington N. procera Khlers Nefhthvdidae. Xephthys coeca (Fab.) Oerst. EuPHROi*YNn>AE. E uphrosyne heterobraiichia sp nov. Syllidae. Piouosyllis elonaata sp. nov. TiypaiKKsyllLs gemmipara sp.' nov. ! Onupiiididae. , Northia elegans sp. nov. ; N. iridescens s]). nov. j LCMBinCONEREIDAE. | Liunbricrmert'is zonata sp. nov., Glycerii>ae. I Glycera riigosa sp. nov. G. nana s]). nov. HeniipoJii£. 80. Magelonn lon^corQiJ^ Ap. iiov. + Capitelmc^ae. SL Capital h dizoiiata *p. nov. + M Chluha1vMII>ai:. 82. Trophonia papillata sp. noy. 4- Southward; Shelter Cove, CaL SS. FlabelUgera infundibularis sp. + nov. Stbrkaspidab. d4. Sterxuwpis foflsor (?) Stimp. + Victoria, B. C. (Herdman) ; Ber- ing Sea, Japan (Maren- zeUer); N. E. coast North America. Maldanidas. 86. Clymenella rubrocincta sp. noy. + 4- Southward to San Pedro, Cal. 86. Nicomache pereonata sp. noy. 4- Ammocharidab. 87. Ammochares occidentalU sp. noy. 4- Arbnicolidae. 88. + 4- Mediterranean. CiRRATCLIDAB. 89. Cirratulua cingulatus sp. nov. + 4- 40. C. robustus sp. nov. + 4- Amphictenidae. 41. Pectinarla brevicoma sp. nov. + Ampharetidab. 42. Sabellides anops sp. nov. + Terebellidae. 48. Amphitrite robusta sp. nov. + 4- 44. A. spiralis sp. nov. 4- 45. Lanice heterobranchia sp. nov. 46. Thelepus crispus sp. nov. + Southward to San Francisco. Sabellidae. 47. Bisplra polymorpha sp. nov. + ++ Southward to Monterey Bay. 48. Megachone aurantiaca gen. et. -f sp. nov. Eriographidae. 40. Myxicola pacifica sp. nov. H- Seupulidae. 60. Serpula Columbiana sp. nov. ++ Southward to San Francisco. 61. Serpula zygophom sp. nov. + Northward (?). Digitized by VjOOQ IC 386 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. The entire absence of Phyllodocidae and Opheliidae from a col- lection of this size and representative character is remarkable. The Eunicidae and Chaetopteridae are represented, the former by several headless fragments of some large species, and the latter by numerous long, annulated, empty tubes. For any zoologist devoted to the study of living animals, the examination and description of a collection of preserved spedmens, gathered by others in a region he has never visited, is likely to prove an irksome and unprofitable task. In the present instance, however, the task has been enlivened by the discovery of several very interesting forms. First and foremost should be mentioned Tri/proiosi/llia gemmipara, a Syllidian with alternation of genera- tions in which the sexual zooids, instead of forming a linear series, arise by collateral budding near the posterior extremity. Hamio- thoe tufa, with its great number of asymmetrical somites, also deserves notice ; and as regards zoogeography, the discovery of a Pacific species of Mai/elona, the finding of a genuine Hemipodia in the Northern Hemisphere, and the confirmation of Gamble and Ash worth's (^00) statement regarding the occurrence of Arenicola claparedei on the west coast of America, are worthy of note. I gladly avail myself of this opportunity to express my sincere and hearty thanks to Prof. C. O. Whitman, who generously placed at my disposal an investigator's room at the Marine Biological Labora- tory during the latter part of the summer of 1900 ; to Prof. E. L. Mark, of Harvard, to whom I am indebted for laboratory privileges at the Museum of Comparative Zoology; and to Dr. Wm. M. Woodworth for permission to make use of the Polychacte collec- tions of the same institution. POLTNOIDAE. 1. Polynoe squamata (L.) Aud. and M.-Edwards. Lepidonotics squamatus Leach. Zoological miscellany, London, 1816. This well-known circumboreal species is represented by three specimens, probably from the vicinity of Port Townsend. P. squamata occurs on the California coast as far south as Santa Monica, where a specimen was obtained for me by Mr. J. J. Rivers, the well-known entomologist. At Pacific Grove it is frequent in Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 387 dredgings at twelve fathoms and over, but T have never obtained it there between tide-marks. I have collected it just above low-water mark at Point Cavallo, on the northern shore of the Golden Gate, and even higher on the beach near Black Point, Sonoma County, about one hundred miles north of San Francisco. The specimens from Puget Sound are larger than any as yet collected from the California coast, but are inferior in size to New England specimens. It is probable that this species will be found to have a wide distri- bution along the eastern and western shores of the North Pacific, comparable to its dispersal on both sides of the North Atlantic. 2. PolynoeinsigxiiB (Baird). Xf€pidonotii8 insignis Baird. Proc. zool. soc. London, Apr., 1863, p. 106. II(do8ydna insignis Baird. Joum. Linn. soc. London, vol. 8. (Zoology), 1865, p. 188. Polynoe brerisetosa (Kinberg) Johnson. Proc. Cal. acad. sci- ences, 3d ser., Zoology, vol. 1, 1897, p. 167. Figs. 24, 31, 40, 40a, 46, 46a. This, the commonest Polynoid of the western coast of North America, is represented in the Puget Sound collections by several slender, darkly pigmented specimens commensal with Thelepus crispuSy and a single specimen with remarkably thick and tuber- culated elytra, obtained by the Columbia University Expedition; also by four specimens collected by Miss Robertson at Alki Point. With the exception of Ilarmothoe imbricata this species has the widest known distribution along our western sea-board of any of its family, ranging from San Diego to Kadiak.^ South of Point Con- cepcion it is rare, at least inshore, being almost wholly replaced by P. ccUi/ornica,^ Its remarkable variations according to habitat have been described elsewhere (vide Johnson, *97, p. 167). The identification of this species with the Halosydna brevisetosa of Kinberg ('55; '58, p. 18), as given by me in the "Preliminary Account" ('97), was undoubtedly characterized by too little confi- dence in the accuracy of Kinberg's figures and too much influenced by his statement that Haloaydna bremsetoaa was collected in San- (salito Bay, near San Francisco. As P, insig^iis is the only Poly- 1 A smaU collection of Polychaeta from Kadiak Island, including a single P. insignU^ wai gathered by Mr. Cloudsley Rutter, and kindly loaned to me for examination and de- scription by Stanford University. s Name changed from P. reticulata Johnson C97) as the latter name is preoccupied. Digitized by VjOOQ IC 388 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. noid with eighteen pairs of elytra I have ever seen from that portion of the coast, it is highly probable that Kinberg's species came from another part of the world. P. insignia was described by Baird in 1863 from specimens collected by J. K. Lord at Esquimalt, Vancouver Island. His Lejndonotiis grubei^ described at the same time (and subsequently) with the foregoing, is in all probability a mere color variety of P. insignia, 8. Polynoe lordi (Baird). Lepidon(Ptns lordi Baird. Proc. zool. soc. London, Apr., 1863. Halosydna lordi Baird. Journ. Linn. soc. Ldndon, vol. 8 (Zool- ogy), 1865, p. 190. Polynoe lordi Johnson. Proc. Cal. acad. sciences, 8d ser., Zool- ogy, vol. 1, 1897, p. 175. Figs. 35, 44, 61. As in the case of P. insignia and P. fragilis^ the Paget Sound region may be regarded as the type locality of the present species. All three were collected by J. K. Lord at Esquimalt, Vancouver Island. For an instructive and entertaining account of this curious Polynoid's habits and habitat, the reader is referred to Lord's << Naturalist in Vancouver Island and British Columbia,'' vol. 2, page 9 ('66). As this somewhat rare work is doubtless inaccessible to many, I quote somewhat at length from Lord's account. Speak- ing of P'iaaurella cratitia^ the host of P, lordi, he says: "I had found him at last and at home, so pounced upon him as a law- ful and legitimate prize. Knife and hammer soon severed his close attachment to the rocks ; and turning him up, to take a peep at his powerful ring of muscle and strangely-formed breathing appa- ratus, I spied a worm evidently very uneasy, about three inches long, brown, and in shape like an ancient dagger blade. He appeared to me to be wriggling out from betwixt the folds of the foot or the mantle, and apparently most anxious to escape .... In displacing other shells, I found in nearly every one a similar tenant: the secret was discovered, the worm was a parasite, that lived in peace and good-fellowship with the Keyhole. . . . That the parasite worm does no harm is clearly proved by the healthy state of the mollusc in whose shell it takes up its abode .... On more carefully examining the position of the worm I found it was invariably coiled away in a semi-circle under the foot, like a ribbon on its edge, never fiat. This seems to me a wise provision ; for the pressure of the muscles when the limpet grips the rock would crush a soft-bodied worm to death, if flat ; but by being edge on, which is the position Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 389 chosen, all risk of harm is avoided, as it fits in a cleft between two layers of soft material. ... At least four out of every six contained a parasite, and, what is rather strange, the worms ^-ere nearly all of one size." Later collectors have found it not uncommon in that region. The Columbia University collection contains four specimens, probably from Neah Bay, as the species is mentioned by Harrington and Grif- fin ('97) as having been found there "on JFissurdlay Miss Robertson's collections afford only a single fragmentary specimen from Alki Point. I have several from Anacortes, collected by Miss Louise M. Carpenter, of Berkeley, Cal. These last occurred under the mantle of specimens of Olyphia ctapera, pre8er> ed with the Anne- lids. This Gastropod is undoubtedly the usual host, but I have found it also in the gill-groove of CryptorMton ateUeri, and once on the Leather Star (Dermasterias imbricata), crawling on the aboral surface. The southern limit, so far as known, is Point San Pedro, about twelve miles south of San Francisco. Two specimens were found at that locality by Prof. W. J. Raymond of the University of California, who kindly placed them at my disposal. Although about eighteen specimens of Olyphis aspera were examined, only these two specimens of P. lordi were obtained. Like Pdynoe fragilis^ this spedes becomes more abundant northward. Every specimen of Olyphis cupera brought from Anacortes by Miss Carpenter had one or two examples of P. lordi under its mantle ; and Lord states the proportion to be "at least four out of every six." Two of the Columbia University specimens have almost no pig- ment. Even the brown zone on the ninth somite, so constant a feature of , this species, cannot be made out in one specimen. This example is remarkable also for its size, having 83 somites and 41 pairs of elytra. P. lordi J like Lepidametria commensalis (Webster, '79), Polynoe gigas Johnson, and Harmothoe tuta (Grube) often has asjinmet- rical somites in the posterior portion of the body. In one of the specimens from Anacortes the thirty-first somite is asymmetrical (cirriferous on the right side, elytrophorous on the left). In a specimen from Dillon's Beach, Sonoma Co., Cal., there are as many as nine asymmetrical somites, and yet there are the same number of elytra (27) on both sides. The elytra on the right side are borne on somites 2, 4, 5, 7 ... . 23, 26, 28, 29, 31, 33, 35, 37, 38, 40, 42, Digitized by VjOOQ IC 390 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 44, 46, 48, 50, 52 ; and on the left side on somites 2, 4, 5 23, 26, 28, 29, 31, 33, 35, 37, 39, 41, 43, 44, 46, 47, 48, 50, The asym- metrical somites are printed in heavier type. 4. Polynoe pulchra Johnson. Polytioe pulchra Johnson. Proc. Cal. acad. sciences, 3d ser^ Zoology, vol. 1, 1897, p. 177. Figs. 34, 43, 43a, 50, 50a, 50b. The species of scaly annelid referred to by Harrington and Griffin (*97) as occurring on Molothutia califarnica is without much doubt P. pidchra, as I have found it a frequent commensal or parasite of that Holothnrian. It is well represented in the Columbia coUec- tion. None of the specimens exhibits dark brown or black mark- ings on the elytra, but all appear to have had the protective reddish or flesh tints characteristic of the individuals found on Holothuria califorjiica. 5. Polynoe fragilis (Baird). PI. 1, fig. 1. Lepidonotus fragilis Baird. Proc. zool. soc. London, Apr., 1863. IlcUoaydna fragilis Baird. Journ. Linn. soc. London, vol. 8 (Zoology), 1865, p. 190. Polynoe fragilis Johnson. Proc. Cal. acad. sciences, 3d ser., Zoology, vol. 1, 1897, p. 179. Figs. 36, 45, 52, 52a, 52b. Numerous specimens from all three districts of the Puget Sound region. As stated in the "Preliminary Account" (Johnson, '97, p. 180) this species is much more plentiful in the Puget Sound region than on the California coast, where I have collected it only in San Francisco Bay, and only three or four specimens in as many years. Its great abundance at Port Orchard is noted by Miss Robertson as follows : " Twenty specimens were taken from twenty- seven or twenty-eight Star- fishes. Several times two and in one instance three, were found on a single Star-fish." The frequent absence of ventral cirri is a striking peculiarity. Even when present, the ventral cirrus is very diminutive ; but, on the other hand, its absence on all the parapodia seems to be a rare occurrence. Of the twenty-nine examples at my command, only one (and an imperfect specimen at that) is entirely destitute of ven- tral cirri. In not a single individual, however, is every parapodium provided with a ventral cirrus. 6. Harmothoe imbricata (L.) Malmgren. This ubiquitous species was collected by the Columbia Expedi- tion, also by Miss Robertson at Alki Point and at other places. Most of the specimens are of the usual greenish gray tint with Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 391 mottled, iron-gray elytra. One of the largest specimens obtained at Alki Point has numerous undeveloped ova attached to the dorsal setae and in the intersegmental furrows. These bunches of ova are only partially covered by the elytra. 7. Harmothoe iphionelloides sp. nov. PI. 1, figs. 2-7. Form short, broad, and flattened ; width, including elytra, two fifths of length; dorsum and prostomium completely covered by elytra, which overlap extensively; somites, 36; elytra, 15 pairs, borne on somites 2, 4, 5, 7 23, 26, 29, 32. Prostomium (Fig. 2) broad, ti*ansverse diameter exceeding the longitudinal ; distinctly bilobed with median sulcus ; the lobes ovoid, rounded in front. Palpi stout, fusiform, minutely papillated, trans- versely ringed with grooves, acuminate. Tentacle with immense basal joint, nearly one half its length; terminal segment slightly papillate, subulate, unpigmented. Antennae with basal joints about as long as tentacle, slightly bulbous near tip, with subterminal dark band ; slightly papillated. Eyes large, black, the posterior pair smaller and nearer together than the anterior pair, Peristomicd cirri (Fig. 2) much stouter than antennae, and about twice as long ; with filiform papillae near tips ; basal joints very long; subterminal bulbous enlargement, with dark pigment-zone. Elytra thick, with large, rough, irregularly polygonal, flattened tubercles (Fig. 3), forming a pattern like alligator skin; tubercles increase in size from the concave (protected) side of elytron towards the convex and exposed portion ; elytra become larger towards middle of series, and diminish again towards posterior end ; ciliate on outer margin ; except first pair, all elytra more or less strongly reniform, the concave edge of each embracing the preceding elytro- phore. Parapodia (Fig. 4) rather long, thick, biramous; each ramus ending in a finger-like prolongation, into which the acicula extends. Dorsal cirrus extends beyond the setae ; basal joint nearly one fifth its length with pin-head-like papillae towards the incrassated tip. Ventral cirrus subulate, slightly papillate. Ventral setae straw- colored, hardly extending beyond the longist dorsal (Fig. 4, below the dorsal cirrus) , and only slightly stouter than the uppermost dor- sal (c/. Figs. 6 and 6); slightly hooked at tip, with 12-20 "frills" of usual form. Dorsal setae white, forming a graduated series from uppermost stout, short, strongly-curved ones (Figs. 4 and 6) to the lower elongated slender ones (Figs. 7 a, h) . The fine serrations ex- tend nearly one half the length of seta, whether it be long or short. Digitized by VjOOQ IC 392 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. XephricUal papMie very short, cylindrical, length and diameter about equal ; begin at 8th and extend to 30th somite. Length (approximately) 23 mm. ; width, including setae, 10.5 mm. ; dorso-ventral thickness, 8.5 nmi. ; proboscis, 4 mm. A single specimen from Pleasant Beach, near Seattle, collected by Professor Hitter. It is chiefly notable for its great, overlapping elytra, with tuberculation resembling the areolae of Samwtkoe hirsuta^ or of IphioneUa. 8. Harmothoe complanata sp. nov. PI. 2, figs. 8-13. Fortn of moderate length, flattened dorso-ventrally ; breadth, in- cluding setae, two sevenths of length ; number of somites, 36-38 ; elytra, 15 pairs, borne on somites 2, 4, 5, 7 23, 26, 29, 32. Prostamium (Fig. 8) approximately six-sided, width and length equal, deeply incised for reception of basal joint of tentacle, indis- tinctly bilobed. Eyes four, minute, posterior pair dorsal, anterior pair dorso-lateral, and further apart than the posterior. Tentacle with short basal joint, which extends a little beyond the '' peaks " of the prostomium, terminal portion nearly twice the length of the pro- stomium. Antennae very short, inserted below level of tentacle, two-jointed, less than one half the length of the prostomium. Palpi very thick at base, terete, gradually and uniformly tapering to an acute tip, barely papillate, as long as perittomial cirri Peristomial cirri long and slender, gradually and uniformly tapered, without subterminal dark band, slightly papillate. Elytra (Fig. 10) thin, oval, translucent, with minute, scattered conical tubercles. Parnpodia (Fig. 9) long, rami distinct, finger-like tips long and slender. Dorsal cirrus like peristomial, long and slender, extending far beyond tips of setae, papillate. Ventral cirrus long, evenly tapered, papillate. Dorsal setae of two sorts : (a) a supra-acicular fascicle of very stoat, minutely serrated setae (Fig. 11) which are the thickest in the foot ; they are arranged in a whorl, the shortest being the uppermost and most anterior; {h) a smaller tuft of very slender, elongated, serrulate setae (Fig. 12) inserted in the finger- Hke process ; these are considerably longer than the preceding, and much fewer (4-5 in number). The ventral setae (Fig. 9) are arranged in a graduated series of which the uppermost closely re- sembles in length, slenderness, and serrulation the dorsal set^ of fascicle h ; the middle and lower ones are of the more usual type (Fig. 13), with a series of "frills" near tip, beginning with very Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 393 minute ones and gradually increasing in size towards proximal end of seta. Nephridiai papillae begin at the 6th somite and extend to 36th ; with acuminate tips. Length of largest specimen, 21 mm.; width, including setae, 6 mm. This species is represented by two imperfect specimens, one from Puget Sound, collected by Harrington, the other from Coronado, Cal., collected by me in July, 1895. Its living color is stated in my notes to be "orange-yellow," but the alcoholic specimen is pale brown. Elytra and dorsar cirri are easily ^tached. The occur- rence of the species at localities so widely separated, and, so far as known, at no intermediate point, is certainly remarkable. 9. Harxnothoe pacifica sp. nov. PI. 2, figs. 14-17. Form rather flattened, diminishing towards both head and pos- terior end ; breadth, including setae, about four ninths of length. Somites, 37; elytra (wanting in specimen), 15 pairs, borne on somites 2, 4, 5 23, 26, 29, 32. Prostomiwn very strongly bilobed, nearly twice as broad as long; "peaks" short and blunt, divaricate; lobes of prostomium separated by a wide, shallow groove. Basal joint of tentacle very thick, about the length of the prostomium. Antennae slender, considerably linger than prostomium, inserted below the level of tentacle ; ciliate, their basal joints projecting beyond the peaks of the prostomium. Anterior pair of eyes laterally directed ; just in front of the bulge of the prostomium ; posterior pair dorsal, near base of prostomium, slightly nearer together than the anterior. Peristomial cirri villous, with9ut bulbous enlargement near tip. Dorsal cirri (Fig. 14) very similar. Dorsal ramus short and stubby without finger-like process, bearing setae of two different forms (Figs. 15, 16) which grade into each other. The short curved ones are the more dorsal. All the dorsal setae are much more slender than the ventral. Ventral setae (Fig. 17) stout, hastate, with straight or slightly curved tips, and from two to twelve serrations. Ventral cirrus fusiform, papillate. Length^'l^.b ram. ; width across middle, including setae, 11.5 mm. This species is represented by a single specimen, unfortunately without elytra and imperfect in other respects. It was collected by the Columbia Expedition. The species is probably scarce, as all efforts to obtain more specimens have thus far failed. Digitized by VjOOQ IC 394 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 10. Harmothoe tuta (Grube). PI. 2, figs. 18, 19. PI. 3, figs. 20-22. Polynoe tuta Grube. Arch. f. naturgesch., jahrg. 21,'bd. 1, 1855, p. 82. Haloaydna tuta Baird. Journ. Linn. soc. London, vol. 8 (Zool- ogy), 1865, p. 188. Form elongated, of nearly uniform width throughout anterior two thirds, gradually tapering in posterior third to extremity; somites numerous and variable in number (77-83) ; elytra not all paired, 35-38 on each side ; borne on somites 2, 4, 5, 7 ... . 23, 26, 29, 32 .... ; general color of formalin specimens, very pale brown ; a transverse band of darker brown on every somite of anterior half, and a median dark stripe. Proatommm (Fig. 18) distinctly bilobed, median furrow extend- ing to its base ; lobes rounded anteriorly, widely sundered by the thick napiform basal joint of tentacle ; eyes placed well forward, medium-sized, black ; the anterior pair laterally, the posterior pair upwardly, directed. Basal joints of antennae inserted below the basal joint of tentacle, on underside of prostomium ; distal segments of antennae and tentacle very short, nearly equal, not exceeding length of prostomium. Palpi also short, rapidly tapered to subulate tips ; constricted at intervals in contraction ; all cephalic appendages, like tentacular and dorsal cirri, sparsely beset with minute papillae. Tentacular cirri considerably longer and more slender than ten- tacle ; subterminal enlargement very slight. Elytra slightly variable in shape, nearly orbicular (Fig. 20), some broadly reniform, strongly imbricated, meeting across median line ; with very few microscopic tubercles, thin and translucent, suf- fused with smoky brown. Parapodia (Figs. 18, 19) elongate, nearly or quite equal to width of dorsum between elytrophores ; dorsal setae rather numer- ous, much shorter and more slender than the ventral (Fig. 19), divaricate, slightly curved, rather coarsely serrated along convex border near tip (Fig. 21). Ventral setae numerous (over twenty), with about twelve ** frills " near the slightly expanded tip. Point nearly straight, acute. Dorsal ramus short for this genus, with de- cided finger-like process, to tip of which the acicula extends. Ven- tral ramus stout, fleshy, with thick, stumpy, finger-like process which does not receive tip of acicula. Ventral cirrus short, abruptly taper- ing to a fine point. Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 3^5 Ne^jhridial papiUae extraordinarily long in large specimen, fully equal to ventral cirri ; begin on the sixth somite and extend to the eighty-iirst in the larger, and to the seventy-fourth in the smaller specimen. Length of larger specimen, 82 mm. ; width, including parapodia and setae, 12 mm. ; without setae, 10 mm. ; without parapodia, 5 mm. Length of smaller specimen, 61 mm. ; width, including para- podia and setae, 8 mm. ; without setae, 6 mm. ; without parapodia, 2 mm. This interesting Polynoid is represented by two specimens in the collections, one obtained by Harrington and stated by him to be commensal in the tube of the largest species of ^^Amphitrite " found by the Expedition, — undoubtedly Thdepus crtspus (see p. 428); the other collected by Miss Robertson at Alki Point. The latter is smaller and evidently younger than the former. The dimensions given in the diagnostic description show dearly the difference in the proportions of young and old. The nearly colorless condition, as well as the great length of the body and the thin, translucent, smooth elytra, indicates plainly its constantly commensal habit. Forms like P. insignis, P, calif or- nica,^ and Hamiothoe imbricata^ which are sometimes free-living, sometimes commensal, retain the pigment, often in heightened inten- sity, when they have either temporarily or permanently adopted the latter mode of life. In spite of the excessive number of somites and elytra, the true relationship of this form is with the species grouped under the genus Jfarmothoe (senau extensiori) , and not with Polynoe^ where numerous somites are much more frequent. The relationship with Hamnothoe is shown (1) by the structure of the prostomium, (2) by the finger-like processes of the rami of the parapodia, and (3) by the sequence of the elytra from the 23d to the 32d somite (23, 26, 29, 32). A fourth character of less importance is the sparse papulation of the cirri — a feature almost invariably present in liar- niothoe^ and absent in Polynoe. The excessive number of somites and pairs of elytra possessed by not a few commensalistic or parasitic Polynoids is no doubt correlated with their mode of life ; hence the unusual length attained by the commensal Ilamiothoe tuta need not surprise us, although such a length, and elytra in excess of 15 pairs, seem to be as rare in this genus as they are common in Polynoe, » SjnoDTm for P. reticulata Johnson which name is preoccupied by P. reticulata CUparide. Digitized by VjOOQ IC 39«J PROCEEDINGS : BOSTON SOCIETY NATURAL mSTORT. It is not unlikely that such highly modified fomiB as Poly not lordly P. pidchra, and P.fragilis have descended from Harnxothot stock and do not rightly belong in the genos Polynoe, The characters are a milange of those of Harmothoe and Polynoe (sensu exL), This may be taken to indicate that these commensals retain characters of the Polynoid ancestor from which typical Polynoe and Harmothoe have both descended. The structore of the pro- stomium, especially the presence of basal joints to the antennae, recalls Harmothoe \ while the stmcture of the foot is more like that of Polynoe. The sequence of the elytra differs from that of both genera. These species certainly do not belong in the genus Lepidasthenia^ where Darboux ('99) has placed them. The asymmetrical somites (dorsal cirrus on one side, elytra on the other) constitute the most striking peculiarity of this species, and one which, so far as present knowledge goes, it shares only with Polynoe yigas Johnson ('97, p. 174), JLepidametria com- mensalis Webster ('79, p. 210), and Polynoe lordi (Baird), In Polynoe yigas^ I have found at most two unsymmetrical somites; in P, lordi J from one to nine ; for Lejnd^nnetria the number is not stated. In two specimens of Hannothoe tuta I have found, respectively, 15 and 16 asymmetrical somites! As in P. giga^^ they are confined to the posterior part of the body (back of the 32d somite) whence the sequence of the elytra differs in different individuals. In the older specimen (Columbia University collec- tion) there are 38 elytra on the right side and 35 on the left In the younger 8])ecimen (No. 963) there are 38 elytra on the right side, 36 on the left. The asymmetrical somites of the former are : 38, 41, 43, 56, 57, 60, 61, 65, 69, 71, 77, 80, 81, 82, 83 — a total of 15 ; those of Xo. 963 are: 33, 37, 3S, 39, 40, 41, 43, 45, 60, 62, 64, 66, 68, 70, 72, 74 — 16 in all. The identification of the specimens from Puget Sound ^nth the Pohjnae tift'i of Grube ('55), described from specimens coUected at Sitka, seems reasonably safe, although his description takes no ac- count of the diagnostically important structures of the head. The seciuenoe of the elytra up to the thii-ty-second somite is precisely as stated by Grube. Beyond that point it differs in every different individual — a fact not perceived by Grube. In no Polynoid has a normal asymmetrical somite been found in front of the 32d. Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 397 SiGALlOXIDAE. 11. Sthenelais fusca Johnson. Sthetielais fusca Johnson. Proc. Cal. acad. sciences, 8d ser., Zooloiry, vol. i, 1897, p. 185. Figs. 60, 61, 64. A single imperfect specimen of huge dimensions, including pro- stomiuni and 74 somites, was collected by Miss Robertson at Pleasant Beach in July, 189X. It is far larger than any specimen yet taken on the California coast. It measures 107 ram. in length (the en- tire worm must have been over twice as long) and 12 mm. in trans- verse diameter. The exposed surface of the elytra is rusty, with the exception of the first six pairs and scattered ones along the body, which are unpigmented and translucent. IIesiomdae. 1*2. Podarke pugettensis sp. nov. PI. 3, figs. 23-25. Body gradually tapering towards both ends ; somites 50-54, the anterior three or four (Fig. 28) much shortened ; the rest about four times broader than long. Pro^toinutiii twice as broad as long, three-lobed in front, the lobes bearing the tentacle and 14. Pis. 10-18. "^riiis remarkable species has been ably described by Harrington ('97), and its extraordinary commensalistic relations with the Her- mit crab { Eiip'KjUrns iinnntas Dana) are discussed.' The scanty material at command r()Ctt'ti Ehlers. Die borstenwtlrmer, 1808, p. 557. PI. -28, tig. 2. A very slender Xereid obtained by the Columbia Expedition unes of a Chaetopterid. The extremely attenuated form of the body and Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 401 great number of somites suggest such a habitat ; and the stout dorsal seta shown in PI. 5, Fig. 59, is such as would properly belbng to a tube-dwelling Nereid. These setae remind one of the " hooded crotchets" of Nereis agassizi (PL 4, Ftg. 43), which is also a tube- dweller, but they are in fact compound setae with a much reduced appendage deeply sunk within the tip of the shaft. These setae are not found in the most anterior parapodia, but begin about the 40th foot. A seta of this form would be especially useful in clam- bering within the tube. These stout setae are only one or two in number in each foot (Figs. 54, 55, 56), and occur only in the dorsal ramus. There are no setae of the ordinary form in the dorsal ramus where these setae are present, except in two or three parapodia where the change is taking place (Fig. 54). Ehlers's description of this species is based upon a single specimen collected by Alexander Agassiz in the Gulf of Georgia in 1859. This type, now deposited in the Museum of Comparative Zoology, I have had an opportunity to ejcamine, and I find it of the same species as the specimens above mentioned. None of the specimens is complete. Ehlers's example had a length of 125 mm. and 179 somites; a nearly perfect specimen in the Columbian collection measures 146 mm. and has *260 somites. It is evidently not full grown, for its greatest transverse diameter, including parapodia, does not exceed 3 mm., whereas Ehlers's speci- men, which is the largest I have seen, has a diameter of 4 mm. Nephthydidae. 18. Nephthys coeca (Fabricius) ()rsted. Numerous examples from various localities — Neah Bay, Salmon Bay, and Pleasant Beach. The largest specimens measure *20 cm. and over in length, and 15 mm. across the thickest portion, including the parapodia. The 8i)ecies occurs northward along the Alaskan shores, in Bering Sea (Marenzeller, '90) and along the northern coast of Siberia (VViren, '83). It extends southward along the California, coast as far at least as San Francisco Bay, but the Cali- fornia specimens are pygmies as compared with those from Puget Sound and Alaska. The species was long ago collected by Alex. Agassiz in the Gulf of Georgia and identified by Ehlers (" Die borstenwttrmer,'* p. 588) as identical with Nphthys coeca of European waters. I have com Digitized by VjOOQ IC 402 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. jiared the Puget Sound specimens with sopie in the Museum of Comparative Zoologj' from Massachusetts Bay, and find them iden- tical in every respect. EUPHROSYNIDAE. 19. EuphroBjrne heterobranchia sp. nov. PI. 6, figs. 60-66 ty, sometimes tilled with coagulum. This seems to be a lymph space connected with the body cavity. In both species the walls of the gills are thick, but thicker in Euphrosyne than in Eurythot', The branchiae of Euphrosyne contain an axial strand of muscle fibres. Their surface is ciliated. SVLLTDAE. 20. Pionosyllis elongata sp. nov. PI. 6, fii^s. 07-70. PI. 7, fig. 71. /^or/?i slender, becoming much elongated with age; 140-200 so- mites ; diameter nearly uniform the entire length, tapered slightly towards head and tail ; intersegmental furrows are deeply incised ; somites average two^and one half times as broad as long. Digitized by VjOOQ IC 404 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Prostomiam (Fig. 67) about one and one half times as broad as long, in front obtuse, convex, the base nearly full width of peristo- mium. Palpi fused for nearly half their length. Antennae and median cirrus moniliform, similar to peristomial and dorsal cirri. Eyes four, the anterior pair twice the size of the posterior, and con- siderably further apart. Periston! i am with dorsal and ventral cirri, without parapodia or setae. ParSVa*fc.v distinct; no stolonization ; genital products develop only in posterior somites (10))d to lO-ltli in a female specimen with 198 somites), wliicli become much enlarged in consequence. (/oh)/\ in life, nearly white, translucent; ova rich yellow. Leniith of full-grown female (19>< somites) 5S.5 mm.; transverse diameter, iucludinii: parapodia, 1.1 mm. A single sjiecimcu was collected by Miss Rol>ert8on at Port Orchard in July, ls99. It is immature, measures 31 mm. in length, and has but 140 somites. This species occurs within tide-marks as far south as Pacific Grove, Cal, where I have taken sexually mature specimens in De- cember. I have also collected, in the month of February, sexually Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 405 mature specimenB of exceptionally large size at Point Cavallo, on the northern shore of the Golden Gate, where it occurs in small numbers on the underside of stones, inside of dead shells, etc. The much-swollen caudal segments, turgid with ripe ova, are very con- spicuous. I have examined the radial muscle-columns in the proventriculus, and find the structure as regards the central core of granular, undif- ferentiated protoplasm, in which the nuclei are located, and as regards the peripheral layer of muscle tissues, in harmony with the statements of Haswell (*86) and Malaquin ('93); I have not, how- ever, been able to discern striations in the muscular part, but this may be due to the fact that the material was not preserved with a view to histological study. 21. Trypanosyllis gemmipara sp. no v. PI. 7, figs, 72-76. Form elongated, much flattened dorso-ventrally, tapered towards both ends, abruptly towards the head, gradually towards the pygidium ; somites very short ; parapodia less than one sixth the width of the trunk in its widest portion ; somites very numerous (300 or more). Prostotniurtt (Fig. 72) comparatively small, broadest in front, distinctly bilobed, the lobes separated by a median fun*ow ; eyes four, the anterior pair larger and very slightly further apart than the posterior. Median cirrus nearly twice as long as the antennae ; these, as also the peristomial cirri and all the dorsal ciiTi, monili- form, with numerous short articulations, diminishing in size towards the tip. Entire surface of cirri covered with dark brown, easily detached, bud-like bodies (Figs. 72 and 74). Palpi (^>.) reniform, elongated, projecting far in front of the prostomium, widely sepa- rated their whole length. Peristominui extremely short, embracing the prostomium on its two sides ; bearing at its anterior corners two pairs of forwardly directed peristomial cim, of which the dorsals are twice the length of the ventrals. The arrangement of prostomium and peristomium, together with their appendages, closely resembles the collocation of these parts in the Polynoids. Parapoilia (Fig. 74) not prominent, ventral ramus fairly devel- oped, pointed at tip, with 7-9 setae of the form shown in Fig. 75 ; a small separate lobe covers the tips of the double aciculae. Ven- tral cirrus (v. c.) short, often curved, blunt at tip, non-moniliform. Dorsal cirrus very long, either straight or circinate at tip ; in the Digitized by VjOOQ IC 406 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. latter case incurved over the dorsum, in the former extending nearly at right angles to longitudinal axis of the body. iJiijestive systtjn exhibits a well-developed trepan (Fig. 73) of ten teeth ; these are surrounded by a circlet of elongated papillae (pep.). Oesophagus of moderate length (extending through 22 somites), strongly chitinized ; proventriculus cylindrical, of uniform diameter throughout; two well-defined lateral raphes ; radial muscle- columns very large. Alimentary canal back of proventriculus with extensive, segmental, paired diverticula. Dor Aid surf are elegantly marked with ^ne transverse dark lines which indicate the boundaries of the segments. Posterior extroiitty cajiable of producing successive crops of col- lateral, sexual buds or zooids (Fig, 7(5), which possess every exter- nal structure of the parent except mouth and anus. They lack, however, an alimentary canal and nephridia. Le)ujthy BS mm. ; transverse diameter, 3 ram. ; dorso-ventral diameter, 1 mm. A single specimen was collected by the Columbia University expedition, probably in the vicinity of Port Townsend. Unfor- tunately, no data accompany it. This individual possesses the remarkable sexual zooids, over fifty in number, presenting all stages of development. They arise as collateral buds from a prolif- erating somite near the posterior extremity (Fig. 76). At full maturity they evidently separate from the asexual stock and become frec-swimminix sexual zooids, provided with para])odia, antennae, eyes, and central nervous system, but destitute of an alimentary canal. Tliey will be fully described in volume 2 of the Biological Bulletin. OXI PHIDIDAE. 22. Northia elegans sp. nov. PI. s, figs. 77-85. Form stout, flattened dorso-ventrally, tapering towards head ; except in most anterior region, dorsal contour flattened and ventral convex ; branchiae simple, filiform, upraised, and slightly incurved over the back. ProstoiiuKm (Fig. 77) small, conical, considerably broader than long, and its sui-face monopolized by its large appendages ; three pairs of antennae ; most anterior j)air short, ovoid ; second pair with annulate basal joints and acute tenninal joints about one half the Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 407 length of the basal joints ; third pair three-jointed, the middle joint much the shortest, and the distal twice as long as the proximal ; this pair more than twice as long as the second pair, and reaching the sixth somite. Basal joints of second and third pairs of antennae with nine or ten obscure annulations. Median cirrus likewise three- jointed, one half as long as third pair of antennae, and with long terminal stylode ; its basal joint five-annulate. Palpi large, globose, approximate (Fig. 78,/>.). Eyes four, small, at bases of third pair of antennae, one pair directed forward, the other laterally (Fig. 78). Peristomium (Fig. 77, 78) shorter and narrower than the other somites, having on its antero-dorsal border a pair of small, slender, peristoraial cirri (Fig. 78,/>. c). Somites : first four or five back of peristomium longer and nar- rower than the succeeding ones, with parapodia (Figs. 77, 80) of different form from the rest, characterized by an elongated ventral ramus, with acute achaetous terminal portion, a large fusiform ventral cirrus (y. c), a dorsal cirrus {cL c.) of similar form and dimensions, and a cirriform gill {br.), likewise of similar aspect. The transition to the typical somite and foot is gradual {cf. Fig. 81, 28th foot). Branchiae unbranched throughout, tapered to an acute tip, increasing in length caudad, until they exceed half the transverse diameter of the trunk. The setae of the 4-5 anterior parapodia are different from those of the succeeding feet, and are of two forms, "hooded crotchets" and capillary bristles (Figs. 82, 83). Setae of dorsal rami throughout the series are buried in the foot ; at most, their tips protrude (Fig. S5) . Setae of ventral rami beyond fifth foot are (1) bordered capillary (Fig. 84), (2) a pair of stout, two-pronged uncini with flabellar expansions at tip — in all respects like uncini of succeeding species (Fig. 90) . Upper jaws (Fig. 79) asymmetrical, six pieces on right, seven on the left. LeiKjth of 85 somites, 66 mm. ; transverse diameter, 6 mm. The antero-dorsal portion of the trunk eleijantly marked with paired umber-brown spots })laced near the posterior border of each somite ; these tend to coalesce across the median line (Fig. 77) . Three or four specimens of this fine Northia occur in the Colum- bia University collection. Unfortunately, all lack the posterior portion of the trunk. There are no data as to depth or exact local- ity. The tubes also are lacking. Digitized by VjOOQ IC 408 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. 23. Northia irldescens sp. nov. PI. 8, figs. 86, 87. PI. 9, figs. 88-92. Form slender and nearly cylindrical, of almost uniform diameter in anterior region ; highly iridescent ; branchiae curved over dor- sum, not reaching median line ; slender, filiform, translucent (Fig. 88). ProHtoitiium (Figs. i^G, >^) projecting from its anterior bor- der. Jaics (Fig. 87) very similar to those of K. degans^ but with more teeth on the dental plates and more slender maxillae. Somites: four following the peristomium with modified parapodia (Fig. 88), similar to the same somites in X, elegatis; the succeed- ing ones with filiform gills and dorsal cirri, but no ventral cirri (Fig. 89). Hooded crotchvtfi (Fig. 91) of ventral fascicles of first four somites very similar to those of preceding species. Two stout, wing- tipped uncini (Fig. 90) in ventral fascicles of parapodia further back. Capillary setae with striated border (Fig. 92) . Lt^ngth of f)2 somites, 38 mm. ; transverse diameter, 3 mm. This species is represented by a single specimen, dredged b}" Prof. W. .V. Herd man at Victoria, B. C, in the summer of 1897, and by him kindly placed in niy hands for description. Unfortunately, all the posterior portion of the specimen is lacking. The tube is of parchment-like mateiial, opaque-white, flexible, and with adherent sand-grains. LUMBRICOXEREIDAE. 24. Lumbriconereis zonata sp. nov. PL 9, figs. 93-100. Form cylindrical, slightly tapered towards anterior end ; para- podia placed a little below the mid-lateral line ; somites three times as broad as long, each marked with a sharply-defined brown zone Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 409 which extends around the trunk, widening on each side, above and below, as it approaches the parapodiuni. Middle of each band marked by a sharp dark line (Figs. 93, 95, 96). Prostotaium (Fig. 93) acorn-shaped, of a lighter tint than the trunk ; perfectly smooth and glistening. Peristoiniuni (Fig. 93) tapered to the diameter of the base of the prostomium which is less than three fourths the diameter of the third somite. Second somite about one half the average length of the somites. Jtttrs as shown in Fig. 94. Farapodia (Figs. 97, 98) less than one half diameter of body, bi-lobed at tip, posterior lobe the longer. Setae inserted between the lobes. Setae, as usual, of two forms : winged capillary (Fig. 99) in antenor portion of body and ''hooded crochets" (Fig. 100) in the posterior portion. Length of *200 somites, 1()7 mm. ; diameter, including parapodia, 4.5 mm. ; without parai)0(lia, 3 mm. A single imperfect specimen lacking the posterior region, was col- lected by Professor Ritter at Salmon Bay, Puget Sound, May 29, 1><99. This specimen is remarkable for the possession of abnormal segmentation in as many as five places. In two instances the so- mites are spiral in the way shown in Figs. 95, 9i), representing re- spectively the dorsal and ventral sides. In another place, the spiral extends through nine somites, with a forked somite at each extremity. The other two instances are partially-divided somites without a spiral arrangement — in one case with the parapodium displaced towards the ventral side. The asymmetrical somites are not confined to any limited region, but are scattered for a long distance through the middle region of the body. Glyceridae. 25. Olycera rugosa sp. nov. PI. 10, figs. 101, 102. Form stout, terete, thickest about one third the distance from head to posterior extremity, tapering slightly cei)halad ; much more, though gradually, caudad. Number of somites 200-300, distinctly two-ringed, all setigerous except the pygidium. The rings are nearly equal, but the anterior one, which bears the parapodia, is often raised like a welt, giving the body a corrugated aspect. Digitized by VjOOQ IC 410 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Prostomium conical, tapering, 2-3 mm. long, length exceeding its breadth, obscurely twelve-ringed ; four minute tentacles at tip. Basal portion not sharply set off from the peristomium. Palpi (?) retracted. Pygidium small, globose, bearing a pair of slender subulate anal cirri. Anus minute, on dorsal side of pygidium. Parapodia (Fig. 101) stout, of nearly equal height and length, the largest equal to width of dorsum in its widest part ; except the two most anterior pairs, four-lobed, the two anterior lobes slightly longer than the posterior, but all lobes of nearly same form except in most anterior parapodia. Anterior dorsal lobe sometimes bifid. All the lobes more or less conical and pointed y their tips darkly pigmented ; simple capillary setae inserted between the dor. sal lobes ; the compound setae between the ventral. Ventral cirrus {v, c.) large, conical, strongly resembling the lobes; dorsal cirrus {d, c.) smaller, globose, much constricted at the point of attaclmient, placed high, at the base of the parapod. Branchiae (iP'ig. 101) begin at the 16th or 17th parapod and ex- tend to the 22d from the pygidium ; they consist of eight or nine finger-like, thin-walled lobes, sometimes bifurcated ; completely re- tractible into body-wall ; at sexual maturity crowded with reproduc- tive cells (Fig. 101). All branchiae are on posterior aspect of the parapodium ; the most anterior and most posterior of the series are single, sausage-shaped processes. Proboscis extremely variable in length (12 to 35 mm. or more), club-shaped, thicker than anterior portion of body, beset with mi- nute papillae of two forms, conical and ovate. Jaws (Fig. 102) strongly hooked; each bears a triangular appendage (.) with long falcate process. Color of alcoholic and formalin specimens variable, from tawny or buff to olive-brown. The color is due to abundant yellowish brown pigment grains in tlie liypoderrais. These are often ag- gre.ijated towards tips of the lobes of the parapodia. Lenyth of lartre specimen (much contracted), 170 mm. ; transverse diameter, inchiding parapodia, 9 mm. This species is probably abundant in the Puget Sound region, as it occurs in all the collections. It has been taken at Neah Bay, and at Salmon Bay (near Seattle) ; and there are a considerable number in the Columbia collection ( probably from the vicinity of Port Townsend) for which no locality is given. A Glyoerid col- Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 411 lected by J. K. Lord at Esquimalt and described under the name Glycera corrugata by Baird ('63) is probably identical with G. rugosa, but the description is too meager to admit of positive deter- mination. Many of the larger specimens are females with ripe or nearly ripe ova ; but I have seen no specimen which has undergone the atrophy of the proboscis incidental to sexual maturity mentioned by Arwidsson ('98, p. 6). As the Glyceridae are destitute of a vascular system, the gills are merely reversible pouches of the body- wall, into which the caelo- mic fluid passes. Reproductive cells also enter these thin-walled pouches. 26. G-lycera nana sp. nov. PI. 10, flgs. 103, 108a. Form short, thick, somites comparatively few (about 140 in one specimen) ; diameter nearly uniform for greater portion of length ; somites two-ringed ; the posterior ring slightly raised above the level. Pro8t(yniium conical, ten-ringed, four minute tentacles at tip. Pro- boscis short, club-shaped, beset with conical papillae. Jaw-appen- dage as shown in Fig. 1 08a. Parapodia (Fig. 103) rather slender, anterior lobes two, the ventral one the longer, both conical ; posterior lobe single, rounded ; ventral cirrus (w. c.) similar in shape to upper anterior lobe ; dorsal cirrus a rounded tubercle placed high above the foot on the side of body ; no gills. Setae elongate, of the usual two forms, capillary dorsal, and com- pound ventral. Length of larger specimen, 64 mm. ; greatest transverse diameter, 6 mm. ; without parapodia, 4 mm. The species is present in the Columbia collection and also in Miss Robertson's; she obtained it at Port Orchard in July, 1898. The exact locality is not recorded for the other specimen. In both examples the posterior portion is regenerating, so it is impossible to give accurately the normal number of somites. It probably lies between 180 and 200. 27. Hemipodia borealis s[). nov. PI. 10, figs. 104, 104a. Form terete, moderately long and slender, of nearly uniform thickness for the greater portion of the length, but tapered poste- riorly; somites three-ringed, 126 in number; two minute anal cirri. Prostomunn with conical, ringed process, tipped with four (?) ten- tacles. Proboscis beset with minute, ovate papillae. Jaws with notch near base; jaw-appendage (Fig. 104a) a simple rod. Digitized by VjOOQ IC 412 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Parapmlvt (F'ig. 104), as invariably in this genus, with only the lower ramus, a single acicula, and no simple setae. The dorsal cirrus {(J. r.) near the foot, ovate ; no gills. Parapodia consisting of an anterior, elongated lobe and a posterior, short, rounded one. LiHf/th, I'l mm. ; transverse diameter, including parapodia, 8 mm. Only a single female specimen of this ver\' interesting sj)ecies oc- curs in the Pnget Sound collections. It was gathered by the Co- lumbia I'niversity Exj»e). I'ntil tiie ])rese]it, no genuine species of IhinljuHlin has been added to the two original (uies, although two sj>ecies of Glyceridae — y/e//< />''>/*Itanir(i M'Intosh ('So, p. 349) and //. st-j'f^'ttfi'lnmifis Roule ('IHI, j). 4')*2) — have been erronef»usly attributed to tliis genus {ritJe Arwidsson 'OS, p. 28) . .\UH IIDAK. *is. Scoloplos elongata sp. nov. PI. 10, figs. 105-110. Form lonir and slender, somites short and very numerous ( 298 in one s])ecimen) ; tlattcnei.'), pp.) ; without eyes. Peristuiitinin increasing rapidly in width towards the second somite, which is the first to bear setae. Pharynx eversible, with leaf-shapeeyond, gradually becoming smaller and more rounded ; towards end of series the upper lobe becomes minute ; ventral lobe rounded, always smaller than the dorsal ; a filiform gill {br.) on somites 2-27, often with lanceolate tip. Setae (Figs. 112, 113, 114) of two sorts : capillary and " hooded crotchets " ; both kinds occur in the ventral rami back of the 17th somite; at Digitized by VjOOQ IC 414 iraOCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. f&nrt'Only one or two uncini (Fig. 114) among the capillarj bristles, ;gra(}ua1!ly increasing to ^five, while the capillary bristles decrease pari passu. Anal cirri two, short and stumpy. The present species undoubtedly belongs to the family LeTin- seniidae recently established by Mesnil and Caullery ('98) for the reception of a small group of peculiar little Polychaetes which show affinities to the Spionidae on the one hand, and to the Ariciidae on the other. The species under consideration most neafly resembles Aricidea, of which two species have been described from the Atlan- :tic coast by Webster and Benedict ('87). It differs enough how- ever from Aricideft to deserve generic distinction. The presence of parapodia and setae on tlie peristomium probably indicates a more primitive character than the allied genus exhibits. The large size of the eyes is also remarkable, and would seem to indicate a pelagic !habit. The foregoing description is based upon a single specimen from Port .(Orchard, jL'ollected by Miss Robertson in June, 1899. Magelonidae. 30. MagfilDna longicomis sp. nov. PI. 11, figs. 115-118. Fomn cylindrical, rather stout, of nearly uniform diameter, divided iatio two regions: (1) the ajiterioTy in front of ninth somite, with capillary, double-bordered setae (Figs. 116-117) in both fascicles ; (2) the posterior, back of and including the ninth somite, with uncini (Fig. 118) both above and below. Ninth somite (Fig. IIG) shorter than the others, and marked by a deeper constriction, with a pair of comb-like fascicles of short, stiff, iiapillary setae. ^ Prostornimn (Fig. 115) flattened, grooved in median dorsal line, anterior tip exj)anded ; no eyes. Peristomium beat's a pair of enormously long, flexile, tentacular cirri, beset with numerous cap- itate j)apillae on exterior aspect; showing rings of contraction near its base (Fig. 115). Proboscis (/>r.) rounded, without corrugations or surface differentiation ; extensible as far as tip of prostomium. Purnpodia (Figs. 115, 116) of anterior region slightly developed; dorsal and ventral cirri small; a small branchia between them; dorsal and ventral setae (Fig. 117) of same form, double-bordered capillary. In posterior region, branchiae and cirri are larger ; uncini (Fig. 118) inform of "hooded crotchets" bidentate at tip, in transveree rows of ten or eleven, rising high upon dorsal side. Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PI GET SOUND REGION. 415 Length of head and anterior region, 7 mm. ; length of first 12 somites of abdomen, 8 mm. ; greatest transverse diameter, 1.5 mm.; length of tentacular cirri, about 14 mm. Two imperfect specimens, consisting of only a few anterior somites, represent this interesting species. It was collected by Miss Robertson at West Seattle, June 23, 1899. Magdona papiUicoruia^ originally described by Fritz MttUer ('58) from the Island of Santa Catharina off the coast of Brazil, has since been found on both sides of the North Atlantic (vide Andrews, '91). Its anatomy, both external and internal, has been carefully studied by M'Intosh (78) and its remarkable blood has been investigated by Benham ('96). Hitherto it has remained a unique and isolated form, most closely related to the Spionidae but, as M'Intosh pointed out, having affinities also with the Chaetop- teridae. The present species differs from J/ pfipillicomis (1) in it« much greater size, (2) in the gi'eater length of its tentacular cirri and longer papillate areas of same, (3) in the comparative shortness of the prostomium, and (4) in the smoothness of the proboscis. Capitellidae. 31. Capitella dizonata sp. nov. PI. 11, figs. 119-121. Thorax thickest in region of 5th and Gth somites; smallest at 8th and 9th (Fig. 119), most of the thoracic somites two-ringed; abdominal somites three- to twelve-ringed; intersegmental con- strictions pronounced, especially in thorax. Prostomium short, conical, at base slightly more than one half the diameter of the peristomium; nuchal organs not discovered. Peristomium setigerous ; somites of thorax over three times as broad as long in the contracted state ; the 4th and 5tli each with a dark brown band passing around it in front of the fascicles. Female genital pore between the 7th and 8th somites ( 9 , Fig. 119). Abdominal somites notably longer than the thoracic, beginning at the 10th, which differs but slightly from the 8th and 9th of the thorax ; increase caudad in length and number of rings. Uncini- gerous tori placed near the posterior boundary of each segment ; the ventral the first to appear, and larger than the dorsal through- out anterior region of abdomen. Digitized by VjOOQ IC 416 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Capillary setae (Fig. 120) alone present in first seven segments; obtusely angled, with striated limb on convex border, arranged in dorsal and ventral widely separated fascicles ; persist in dorsal fasci- cles as far as 10th somite ; beyond this point replaced entirely by nncini. Uncini (Fig 121 ) begin in ventral fascicles at 8th somite ; in dor- sal fascicles at the 10th ; hooded, with four minute teeth above ros- trum; shaft strongly geniculate. Length of 39 anterior somites, 36 mm. ; greatest transverse diameter of thorax, 1.5 mm. This species is represented in the collection by a single incom- plete specimen, lacking the posterior portion. It is an immature female, and was collected by Miss Robertson at Port Orchard, July 2, 1898. The dorsal setae of the 8th and 9th somites appear to be entirely wanting. CULORAEHIDAE. 32. Trophonia papillata sp. nov. PI. 12, figs. 122, 123. Form rather long and slender, slightly tapered, thickest anteri- orly, abruptly diminishing toward mouth ; subcyhndrical ; inters^- mental constrictions distinct, but not noticeably deepened caudad ; entire surface papillate, slightly rough to the touch, but without adherent sand-grains; dorso-ventral differentiation slight, mainly expressed in differences between dorsal and ventral setae, and in the closer papillation of dorsum. Oral tentacles (Fig. 122) eight, of moderate length, pointed at tip; palpi (/>.) thicker and blunter, grooved on ventral aspect, distinctly constricted at regular intervals. iSetae of second, third, and fourth somites, both dorsal and ventral, elongated, flexible, capillary bristles, forwardly directed (Fig. 122), exhibiting the usual transversely-striate structure. Dorsal setae of remaining somites, capillary, three or four to each fascicle ; ventral setae (uncini), to the same number; stouter than dorsal setae, blackish, hooked (Fig. 123). Somites of only complete specimen, 89 in number; those in the anterior region twice as broad as long; posteriorly, length and breadth gradually becoming equal. Lengthy 88 ram. ; greatest transverse diameter, 4 mm. Two specimens, one incomplete, were collected at Port Orchard, July 2, 1898, by Miss Robertson. Digitized by VjOOQ IC JOHNSON; POLYCHAETA OF PUGET SOUND REGION. 417 33. naboUigera infandibnlaris ep. no v. PL 12, figs. 124- 127. Form (Fig. 124) rather stout, squarish, dorsal aspect more flattened than the ventral; enclosed in clear mucus; tapered in posterior third to a minute caudal extremity ; oral region (2d somite?) flared, with an almost complete circle of setae on the margin, formed by two broad fan-shaped fascicles ; intersegmental constrictions deep; parapodia (Figs. 124, 126) well developed, distinctly biramous, on every somite from the third. PeriatomiHm within the oral funnel ; bears numerous slender tentacles and two stout, lobulated palpi (/>.). Dorsal setae (Figs. 125, 126) longest ^d most numerous on second somite, where they form the funnel ; on the other somites, more slender, delicately curved, completely imbedded in the jelly, transversely stnate. Ventral setae (Fig. 126) begin on third somite, one or two in each ventral ramus, in form of long hooks, trans- versely striate, blackish towards tip. Numerous sensory papillae (Fig. 127) are borne at the tips of long varicose pedicels. Somites in four specimens are 42, 50, 56, and 71, increasing in number with size of animal. In contraction, somites are at least four times as broad as long. Length of specimen with 56 somites (about average size), 55 mm. ; greatest transverse diameter, 5 mm. According to the statements of Harrington and Griflin ('97, p. 162), this species is enormously abundant in Scow Bay, where it covers the muddy bottom over an area about half an acre in extent. The depth given for one lot of specimens is six fathoms. It does not appear in any of the littoral collections from the region of Seattle. The extraordinary elongation of the dermal sensory papillae in species of this genus is well exemplified in the present form. The thick coating of mucus which envelops every part of the animal except the anterior and posterior extremities (Fig. 124) (the funnel formed by the broad flabellate oral tentacles makes a passage to the mouth) apparently renders necessary this method of putting the animal in communication with the outer world. Digitized by VjOOQ IC 418 PROCEEDINGS: BOSTON SOCIETY NATURAL mSTORT. Stebnaspidae. 34. Stamaspis fossor (?) Stimpson. Three specimeDs, apparently of • this species, were collected at Victoria, Vancouver Island, by Prof. W. A. Herdman. The speci- mens are not sufficiently well preserved to admit of thorough and critical study ; but comparison of the ventral shields with those of a specimen of /S. /ossor from the Atlantic coast, and also with Marenzeller's ('90) figures, makes it reasonably certain that the above identification is correct. Specimens from Puget Sound collected by the naturalists of the Northwest Boundary Commission were doubt- fully described by Stimpson (*64) as a new species, which he named Sternaspia nffinis. Maldanidab. 85. Clymenella ^ rnbrocincta sp. nov. PI. 13, figs. 128-133. Form considerably elongated, cylindrical, narrowed in region of third and fourth somites, gradually enlarging to maximum diam- eter in region of somites 10-12, thence narrowing to somites 15-17, which are the longest and slenderest of the body ; the remaining somites (18-22) slightly thicker and progressively shorter; the 2l8t the shortest of the body. Cephalic plate (Figs. 128, 129) oval, concave on dorsal side, nearly bisected longitudinally by a median ridge continued back from the ovate palpode ; a distinct raised margin, and well-developed longitudinal nuchal organs (Fig. 129, n, o.). Mouth with thick- ened corrugated lower lip and crescentic outline. Peristomiwa achaetous ; capillary setae and uncini begin on sec- ond somite (Fig. 128), the latter with scarcely perceptible tori at first ; tori become distinct in fourth somite ; dorsal setae from wart- like papillae. tSohiites gradually increasing in length from fifth onward ; fifth to eighth marked with a broad indian-red band back of the setae and uncini, accentuated by a narrower whitish band in front. » Axiothea and Clymeyiella cover species too nearly alike to require generic distinc- tion. As recently pointed out by Verrill ( :00, p. 657). the name Axiothea being preoc- cupied, Clymenella {semu ext.) should cover the species formerly included under Axiothea. Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. * 419 Tori much enlarged in somites 12-21. Last 8 somites (including pjgidium) achaetous; pygidium funnel-shaped with eirrose border. Perianal cirri (Fig. 130) 18-30 in number, alternately long and short ; the mid-ventral one much elongated, and containing prolonga- tion of the ventral nerve-cord ; tips of all the longer cirri recurved. Anal rosette with alternate sectors raised, corresponding in position with the longer cirri. CapiUary setae of two forms, bordered and serrated, the latter much the more slender (Fig. 133). Both kinds in same fascicle; begin at second somite. Uncini (Figs. 131, 132) with five or six teeth, including rostrum, graduated in size ; bristles at base of rostrum lat- eral i^ position. Length of large specimen, 162 mm.; greatest transverse diam- eter, 3.5 mm. This fine species comes near Axiothea ccUenata Malmgren, a form of wide distribution in the North Atlantic and Arctic, and reported from Bering Sea by Marenzeller ('90). The present species, how- ever, differs from it in the form of the serrated setae and of the uncini. In A, catenata there are four preanal achaetous somites, in the present species only two ; C, ruhrocincta has 22 segments ; A, catenata, 24. The prfesent species was collected both by the Columbia Univer- sity Expedition and by Miss Robertson. I have found it in abundance at the entrance of Tomales Bay, and at San Pedro, California. It forms a tube of coarse sand, which is lined by a peculiarly tough, opaque, whitish membrane. 36. Nicomache personata sp. nov. PI. 13, figs. 134-139. Size small ; 25 somites, of which all except the peristoraium and the last two are setigerous; no cephalic plate; prostomium and peristoraium united to form a hood-shaped head (Fig. 134); mouth large, overhung by projecting upper lip ; a distinct crease in peristoraium back of mouth. Somites increasing in length back of second ; longest from the 8th to the 17th; those at the end very short; pygidium (Fig. 135) funnel-shaped, with a zone of 16-18 cirri on margin, quite uniform in length. Dorsal and lateral surfaces of head and first four or five somites beautifully mottled with chocolate-brown; ground-color, white ; somites 2-5 with contrasting white and brown bands ; on the head these bands have the aspect of a grotesque face. Setae of five kinds: (1) capillary double- bordered setae in all Digitized by VjOOQ IC 420 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. somites from second to twenty-third; (2) smaller serrated setae together with (1) in dorsal rami; (8) ventral setae in first three setigerous somites in form of a stont, acate spine (Figs. 134, 136) ; (4 ) very long, filamentous, spiral setae in dorsal rami beginning at 7th to 11th somites and continuing to 23d somite (Fig. 138); (5) uncini {Fig. 137) of the usual form in this family, in all setigerous somites back of the 4th. Length of much-contracted specimen (impossible to measure accurately on account of twists and flexures), 50 mm.; diameter through thickest portion, largest specimen, 1.75 mm. The only specimens of this odd little species were collected by Hiss Robertson at Alki Point, Aug. 3, 1898. The specie^ bears a close superficial resemblance to Xicmnache lumbricalis (Sars) Malmgren, but has one achaetous somite in front of the pygidium instead of two. The long, silky, spiral filaments shown in Fig. 138 have not, I believe, been, hitherto described in any Maldanid. Whether they are permanent structures peculiar to this species, or " nuptial setae '* (Pubitfttsborsten) which develop only at sexual maturity, is an interesting question. Filamentous nuptial setae of a similar form have been described by Michaelsen ('92, p. 6) in a Polynoid (Drieschia pelagica), Ammocharidae. 37. Ammochares occidentalis sp. nov. PI. 14, figs. 140-142. Form cylindrical, tapering towards posterior end ; 23 (?) somites, of which 20 are setigerous. Peristoinium produced into ten thick, bluntly-ramose '* tentacles" (Fig. 140) which rise to a level, giving anterior end of body a trun- cated appearance. Capillary setae in fascicles placed high on dor- sal surface (Fig. 142), very slender, acutely pointed, silvery by re- flected light, serrated. Uncini very minute, two-hooked (Figs. 141 a, ^), lG-18 horizontal rows in each band (6th somite), extend- ing three quarters of the way around the body ; begin at 4th somite. Second and third so)nites (Fig. 140) very short, third twice the length of the second, both with rudimentary parapodia (?) . Somites 4, 5, 6, 7 the longest of the body ; from the 7th diminishing gradu- ally to end of series. Digitized by VjOOQ IC JOHNSON: POLTCHAETA OF PUGET SOUND REGION. 421 Length, 22.5 mm. ; greatest transverse diameter, 1 mm. No species of Ammochar€8 has hitherto been reported as having chelate or bifid uncini. This character in fact has been considered so diagnostic of the allied genus Myriochele that M'Intosh did not hesitate to describe a species in the " Challenger '' collection as My- riochele pacifica from a fragmentary specimen lacking both anterior and posterior extremities — basing his diagnosis entirely on the structure of the uncini. The two specimens upon which the foregoing description is based were collected by Miss Robertson at Port Orchard, July 2, 1898. £ach was enclosed in a tube composed of sand-grains and minute particles of shell. The color of the formalin preserved specimens is nearly black. Arenicolidae. 38. Arenicola claparedei Levinsen. PI. 14, figs. 143, 144. In the excellent memoir of Gamble and Ashworth (:00) upon the Arenicolidae, the Mediterranean species, originally described by Clapar^de ('70, p. 300) as Arenicola marina, but afterwards erected as a separate species by Levinsen ('83, p. 137, footnote) under the name of A. claparedei, is attributed to the Pacific coast. After a careful examination of the Puget Sound Arenicolae at my disposal, and comparison with specimens of A, claparedei from Naples, I am convinced that Messrs. Gamble and Ashworth are cor- rect in their determination. The only notable points of difference between the Puget Sound specimens and those from Naples are the vastly greater size — at least eight times as great — of the former, and the smaller number of oesophageal coeca or "pouches" in the latter. Of the four specimens of A, clffjyaredei from Naples which I have examined, three have four pairs of pouches, and one only three pairs ; whereas, out of eight specimens from Puget Sound in four there are six pairs, in two fifteen pairs, in one sixteen pairs ; and in one there are sixteen coeca on the right and eighteen on the left ! In the Arenicolidae there are as a rule only two oesophageal coeca or " pouches," but in A. cla2xtredei they are not only numer- ous (as many as 32 in one instance, Fig. 144) but highly variable as to number and arrangement, and probably even differ as to function, if it is permissible to draw such an inference from the great size and thin-walled character of the most anterior pair (Figs. 143, 144). Digitized by VjOOQ IC 422 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. Ab, howerer, the whole qaesdon of the stmcture and functioii of the oesophageal pouches of Arenicola is still unsolved, this problem must be deferred for the present. The discovery of this Mediterranean form upon the Padfic coast of North America, and at no intermediate point, is certainly one of the most remarkable cases of discontinuous geographical distributioa ever recorded. With exception of a single specimen collected by Miss Robertson, I am indebted to Dr. C. M. Child of the University of Chicago for all the Puget Sound Arenicolcte I have. The specimens were col- lected by the Columbia University Expedition — exact locality not stated. While, as von Marenzeller has suggested, it is probable that Murdoch's ('84) ArenicoUi glacialis from arctic Alaska is none other than the circumboreal and widely-ranging A. marina, it is not at all certain that the latter species occurs so far south as Puget Sound, although it has been reported from Vancouver Island by von Marenzeller ('87). CiRBATULlDAE. 39. Cirratalns cingolatus sp. nov. PI. 14, figs. 145-148. Form stout, size moderate, tapered at both extremities, decidedly flattened on ventral aspect; dorsum rounded; seven anterior somites without setae or cirri ; two clusters of 17-18 tentacular fila- ments (Fig. 145, t.f.) each, on dorsal aspect of 8th (the first setig- erous) somite; when removed, an oval transverse scar is left; a series of similar cirri along each side, in the anterior and middle portions of body inserted low (Fig. 146), gradually rising to a higher level in the posterior region. SoNuteH very short, three-ringed above the dorsal setae, the middle ring raised welt-like above the level (Fig. 146). Prostortiimn (Fig. 145) acute, concave on ventral side towards mouth, which usually exhibits a partially everted pharjTix; eye- spots five or six, either in a group or transverse row. Pumpodhi slightly developed (Figs. 145, 146), both dorsal and ventral rami with slender, serrated setae (Fig. 147) ; these alone are present for 30 setigerous somites back of the head ; the uncini appear in the ventral rami (Fig. 148) at this point, and in the dorsal rami a few somites caudad ; they continue to end of series. Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 423 Cirri, especially those of the first cirriferous somite ("tentacular filaments^') of great length, frequently forming a tangled mass nearly as long as the entire body. Anus in pygidium, minute. Length of large specimen, 100 mm. ; greatest transverse diam- eter, 5 mm. There are several specimens in the Columbia University collec- tion, obtained " between tides." I have not found this species on the California coast. 40. Cirratulns robnstiui sp. nov. PI. 14, figs. 149, 150. Form short and thick ; ventral aspect broadly flattened, concave anteriorly and posteriorly ; dorsum rounded, tapered about equally towards both ends; 112 somites in one specimen, not dimded into rings (Fig. 150). Prostomitim (Fig. 149) shorter and thicker than in preceding species ; eye-spots in two oblique rows. First three somites achae- tous, fully twice the length of the rest ; fourth with two clusters of tentacular fllaments and with dorsal and ventral setae (Fig. 149, t. /.). Parapodia (Figs. 149, 150) slightly developed; anteriorly with capillary setae alone; at the 19th or 20th somite the ventral uncini begin, and two or three somites further back, the dorsal; cirri inserted low in anterior somites (Fig. 149), gradually rising to mid- lateral line (Fig. 150) towards middle of length, then in posterior region gradually approaching the dorsal setae. Anus terminal. Lengthy 59 mm. ; greatest transverse diameter, 5 mm. Only two mature and one yoiung specimen of this species are available for description. One adult specimen was obtained at Neah Bay by the Columbia University Expedition ; the other at Port Orchard by Miss Robertson. Both lack the cirri. The setae and uncini of this species are practically identical with those of Cirratidus clngulatus, Amphictenidae. 41. Fectinaria brevicoma sp. nov. PI. 15, figs. 151-156. Fortn conical, gradually widening towards anterior end ; cephalic disc nearly flat, plane, its edge entire ; scapha broadly ovate, not wider transversely than posterior end of thorax. Total number of somites, about 27 ; 21 in thorax (of which 17 are setigerous, and 13 are uncinigerous), and 6 (?) in scapha. Digitized by VjOOQ IC 424 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Pro8tom\um expanded into a large velnm (Fig. 151) with 33 fim- briae on edge. Periatornium with a pair of subulate, mpniliform cirri and numer- ous club-shaped tentacles (much contracted in preserved specimens). Paleolae short, stout, arranged in two distinct groups, 10-12 in each ; highly iridescent, either blunt or acutely pointed (lateral ones), strongly sloped towards ventral aspect. Second somite with a pair of cirri in all respects like the peristomial. Branchiae pectinate, borne on third and fourth somites. Setae broadly limbed and twisted near tip (Fig. 152) with serra- tions beyond the twist, or straight and without serrations (Fig. 153) . The latter are not so numerous. They diminish in size towards anterior and posterior extremities of thorax. Uncini with 4 teeth (Fig. 154) or occasionally five (Fig. 155) ; the latter begin about the 11th somite; none were found in front of this. Spines on scapha (Fig. 156) with a stout, laterally-bent hook at tip. Tube composed of coarse sand-grains, curved. Length of largest specimen, 28 mm. ; diameter of disc, 5 mm. Several specimens were dredged by the Columbia University Expedition at a depth of 10 fathoms. This species comes nearest to P, {Cistenidea) granulata (Malmgren), which has been found in Bering Sea (Marenzeller, *90), and was collected at Kadiak by Mr. Cloudsley Rutter, differ- ing from it only in the shortness of the paleolae and in the form of the setae and uncini. Upon examination of more abundant ma- terial, this form may prove to be identical with grannlata^ which is a wide-ranging, circumboreal 8i)ecies, and may therefore prove to be variable. While the uncini afford, on the whole, the surest diagnos- tic characters, they should be used with caution, as their variability in the present species clearlv indicates. Ampharetidae. 42. Sabellides anops sp. nov. PI. 15, figs. 157-161. PI. 16, figs. 162-163. Form stout, curv ed, thickest about midway of thorax, abdomen rather rapidly tapered, convex on dorsal, flat on ventral aspect (Fig. 157). Thirty to thirty-one somites, sixteen in thorax, four- teen to fifteen in abdomen ; fourteen setigerous somites in thorax ; Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 425 setae begin at third somite and cease at sixteenth. Uncini begin at sixth somite (the third setigerous somite), and extend to the pjgidium. Dorsal rami of all abdominal parapodia destitute of setae and uncini. Prostoinium (Figs. 158, 162, 163) broadly truncated with two anterolateral lobes ; no eyes ; a pair of transverse slits containing the nuchal organs at its base. Peristamuim (Figs. 158, 162, 1) forming, with second somite, a region destitute of parapodia and branchiae. The club-shaped tentacles, about fifteen in number, arise from the inner border of the peristomium ; they are destitute of papillae, but in contracted state are wrinkled (Fig. 163). Branchiae^ four on each side, smooth, terete, subulate, arise from dorsal aspect of the third, fourth, and fifth somites ; in length about twice the diameter of the body (Figs. 158, 162). Ventral rami or tori (Fig. 157) spatulate, increasing from the fom'th somite to the sixteenth, thence diminishing to end of body. Each bears a single row of pectinate, six-toothed uncini (Fig. 161). Dorsal thoracic rami contain fascicles of single and double- bordered, straight and slightly-curved setae (Figs. 159, 160); dorsal rami achaetous throughout the abdomen. Length of largest specimen, 27 mm. ; greatest transverse diameter of same, 5 mm. Several specimens are in the Columbia University collection, without data as to depth and locality. No tubes were preserved. This species comes close to SabeUidea {Amage) auricula (Malm- gren) but differs from it in having longer branchiae, attached to three somites, shorter tentacles, and differently shaped uncini. SdbeUides auricida^ however, has been reported from Japan by Marenzeller ('85), and its occurrence in any part of» the North Pacific would therefore not be surprising. I follow Theel ('79) in discarding Malmgren's genus Amage, as not being sufliciently distinct from Sabellides, Terebellidae. 43. Amphitrite robusta sp. nov. PI. 16, figs. 164-168. Form short, robust, thickest anteriorly in region between 5th and 12th somites, gradually and almost uniformly tapering thence to Digitized by VjOOQ IC 426 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. posterior end. Total number of somites, 63-90 ; in thorax, 20, of which 17 (4th to the 20th inclusive) are setigerous. Uncini begin at 5th somite and extend to the penultimate. Uncini nniseriai anterior to the 11th somite, and posterior to the 20th; biserial on thoracic somites 11-20 (Fig. 167). Prostomiiim with dorsal ridge and small lateral lobes formed by its continuation to the sides; no eyes. Peristomium bearing a semicircle of tentacles, rather thick in formalin specimens, decidedly grooved, spirally coiled in contracted state, and about one half the length of the thorax. Peristomium forms a thick prominent ventral lip, opposible to thin ventral edge of prostomium. Bratichiae (Fig. 164), three pairs, on somites 2-4, densely ramose, di- and tri-chotomously branched ; the main stems short and thick, ultimate branches subulate, slightly moniliform. Branchiae all nearly alike in form and size ; anterior pair slightly the largest ; all variable as to size and amount of branching. S€ti(feron8 lobes moderate, increasing in size from the first to the seventeenth (on 20th somite). Uncinigerous tori of nearly uniform length from 5th to 18th somites; thence gradually diminishing to end of body. Setae with striated limb on each side and curved, serrated tip (Fig. 165). Uncini avicular, with from 5 to 7 rows of teeth above the rostrum (Figs. 166-168). Xejt/iridi'al 2)apiUae twelve psirs^ on somites 4 to 15; first pair considerably the largest ; the rest of nearly uniform size ; placed ventrad, and a little posterior to setigerous lobes. Lenffth of larger specimen, 75 mm. ; greatest transverse diameter of thorax, 11 mm. ; average transverse diameter of abdomen, 4 mm. Several specimeim, collected both by the Columbia University Expedition and by MiHs Robertson. The species is doubtless com- mon. No tubes were preserved. The absence of eyes in this species and in the preceding is remarkable. They are evidently wholly lacking, as I was unable to find them even in serial sections. 44. Amphitrite spiralis sp. nov. PI. 16, figs. 169-171c. Form greatly elongated, abdomen slender, terete, and spirally coiled when out of the tube ; dorsal aspect of thorax high-arched ; ventral slightly convex. This condition is enhanced in anterior portion of abdomen,, where the somites are decidedly thicker on the dorsal than on the ventral aspect, producing thereby the spiral Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND BEGION. 427 coiling. Number of somites approximately 170 ; those of posterior portion of abdomen two-ringed. Thoracic somites, 43, of which 41 are setigerous and 39 uncinigerous. Prostomimn with dorsal crescentic groove separating a dorsal ridge from the ventral flap ; no eyes visible on surface. Peristomiiim with deep ventral groove, and dorsal semicircle of cirri. Branchiae two pairs, on third and fourth somites ; moderate ; anterior pair considerably the larger and m6re richly branched; main branches arising near the base ; beyond the first ramification, the branching is dichotomous (Fig. 169). Setir/erous lobes begin on fourth somite, increase slightly in size towards middle of thorax, then diminish ; the last few pairs very small. Setae of form usual in this genus (Fig. 170). Uncinigerous tori begin at fifth somite; first six pairs shorter than the rest ; gradually increasing in length to the seventh where they attain the maximum length, and retain it to the 16th or 17th, ^t which point they gradually diminish, becoming flattened and almost indistinguishable on the abdomen. Uncini small, avicular, with five rows of teeth above the rostrum (Fig. 171 c). They are uniserial on somites 5 to 10, and on 35 to end of series ; biserial, on somites 11 to 34. Lengthy 160 mm. ; greatest transverse diameter of thorax, 5 mm. ; average of abdomen, 2 mm.; dorso-ventral diameter of thorax, 4.5 mm. A single individual of this species was collected by Miss Robert- son near Seattle. 45. Lanice heterobranchia sp. nov. PI. 17, figs. 172-174. Thorax of 20 somites, 17 (?) setigerous. Prostonnuni of usual form ; no eyes. Branchiae three pairs, borne on somites 2-4; first pair (Fig. 172) much the largest and longest, with elongated main stem dendrit- ically branched ; ramifying branches very compact. Gills of second and third pairs short, without main stem. Setae (Fig. 173) with striated limb on each side ; tip entire, from fourth (?) somite onward. Uncini from fifth somite, uniserial, alter- nating ("rangee alterne," Clapardde), avicular, with three teeth in front of beak (Fig. 174). A single specimen in the Columbia University collection, too imperfect for complete description. The worm was enclosed in a mud tube. The difference in the size of the branchiae of different pairs is the most striking character. Digitized by VjOOQ IC 428 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. 46. Tholopns crispns sp. nov. PL 17, figs. 175-178^. Form rather stout, not greatly tapered posteriorly ; thorax pass- ing gradually into abdomen ; tapered considerably towarcis prosto- mium from tenth somite. Number of somites 88-147. Prostomium with ample dorsal flap, transversely corrugated on dorsal surface ; no eyes. Peristotnium with circlet of strongly grooved tentacles. Branchiae (Fig. 175) three pairs, branching from the base in numerous, slender, spirally curled filaments ; borne on somites 2-4. Sttae, begin at third somite and extend to penultimate in young specimens ; to fourteenth from pygidium in older ones ; with stri- ated limb on each side; sometimes slightly bent (Figs. 176, 177). Uncinigerous tori begin at fourth somite; nncini absent from extreme end of body. Uncini (Figs. 178 a, ^) single-ranked from fourth to seventh somites, inclusive, and gradually returning to this condition towards end of series; in flattened rings ^m eighth somite (^^rang6e parabolique," Clapar^e) onward. Tori attain their greatest length between the twelfth and twenty-fourth somites ; thence diminish very gradually to end of body ; those of the abdomen rounded and wart-like. Length of large female specimen, 270 mm. ; greatest transverse diameter (at sixteenth somite), 13 mm.; dorso-ventral diameter, 12 mm. This flne Terebellid is represented in the Columbia University collection by a single large female turgid with eggs. It occurs on the California coast as far south at least as San Francisco, and is abundant at Bolinas, Marin County. Its tube is formed of coarse sand or gravel. It frequently harbors commensal individuals of Pohjnoe insignis^ and northward, probably also Hamiothoe tuta (see p. 394). Sabellidae. 47. Bispira polymorpha sp. nov. PI. 17, figs. 179-183. Fl. 18, ?i^i^, 184, 185. Fonn nearly terete, dorsiira flattened in adult specimens; in young specimens, form slender; tapered gradually at posterior end to a minute pygidium; anus terminal. Somites, 170 or more. Thorax (Fig. 179) of nine somites, one sixteenth to one ninth of entire length (exclusive of branchiae) according to degree of con- traction and probably also the age of the specimen. Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 429 Branchiae (Fig. 179) about as long as thorax, dichotomously branched twice or thrice in some specimens, in others unbranched ; pinnate, radioles slender; 16-30 branchiae on each side, forming a spiral of 2-3 turns ; each rachis with 2-10 black eye-spots (Fig. 184). Tentacles flattened, lanceolate, about one fifth the length of branchiae. Fecal groove extending forward along mid-dorsal line of abdomen to thorax ; at posterior boundary of thorax passing on the left side to mid-ventral line of same, and thence to oral region. Peristoniium with raised anterior border or " collarette," deeply notched in mid-dorsal line, and produced into two pointed processes adjoining the ventral sulcus. Thoracic setae (Figs. 180, 181) begin on second somite; of two forms, winged-capillary and mucronate-spatulate ; the latter more numerous, forming a close series towards the torus. Uncinigerous tori of thorax dorsal to setigerous tubercles ; begin on third somite ; separated by full width of dorsum ; uncini biserial, of two sorts (Fig. 182), avicular and dilated-cuspidate ; both kinds with long manubria ; the points directed cephalad. Abdominal setae all of one kind (Fig. 185) arising from smaller tubercles, which are placed dorsad to the uncinigerous tori. Uncini (Fig. 183) uniserial, all avicular, with a shorter manubrium than the thoracic uncini ; rostra directed anteriorly. Tube cartilaginous, translucent, adherent to rocks, piles, etc. Length of average specimen (exclusive of branchiae), 96 mm.; greatest transverse diameter, 6 mm. Greatest transverse diameter of largest specimen at hand, 12 mm. Numerous specimens from the Puget Sound region, collected at Neah Bay and in the Port Townsend district by the Columbia Uni- versity Expedition, and at Alki Point and Port Orchard by Miss Robertson. It occurs also on the California coast as far south as Pacific Grove. This species is remarkable for the highly variable aspect which it presents, owing to the diverse coloration of the branchiae and the differences of shape caused by different states of contraction in which it has been killed. If killed within the tube, it is almost per- fectly cylindrical and often of great length, owing to the impossibil- ity of expansion within the rigid and tightly fitting tube. The longest specimen thus killed (posterior somites lacking) measures not less than 150 mm. in length and only 5 mm. in greatest trans- Digitized by VjOOQ IC 430 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. verse diameter. Even more striking, although not always percep- tible to the naked eye, is the variable character of the branchiae, which are in some specimens dichotoraously branched, and in others entirely unbranched. The coloration of the branchiae shows two distinct phases — pur- ple or wine-color and whitish or tawny. Either color may be present to the exclusion of the other, or the two may be in alternate, trans- verse bands. The eye-8[)ots may be few or many, but I have found no specimen without them. Although a lens is absent, the eye-spot j)roduces a wart-like elevation of the cuticula which covers it (Fig. 184). The eye-spots are of various sizes, the largest being over twice the diameter of the smallest. They are scattered at irr^ular intervals along the rachis, but are more numerous towards the base than towards the tip. None are found near the tip and very rarely any on the branches. The tube is adhesive, and is usually aUbced by its lower extremity or by its side to a stone or pile. At Pacific Grove, the species attains a large size and is abundant on rocky bottoms from low-water mark to a de})th of several fathoms. Megachone gen. nov. Form terete; no ventral shields; collarette liaring, interrupted only at ventral notch ; branchiae connected by a web ; no spatulate thoracic setae ; thoracic uncini with long manubrium, of one kind only ; abdominal uncini short avicular plates ; no ventral fissure at posterior end. 48. Megachone aurantiaca sp. nov. PI. 18, figs. 18G-192. Form subcylindrical, spiral in contraction, thickest in posterior jwrtion ; anterior end truncated; collarette broad (Figs. 186, 187) ; body narrowed in region of thorax and enlarged in abdomen ; pos- terior extremity abruptly tapered to a minute tip, which is curved dorsad ; anus terminal. Somites 75 in number. Branchine 20 on each side, unbranched (Figs. 187, 188), closely appressed, tips pointing ventrad ; radioles invisible until branchiae are raised. Th(jrer, short, biannulate; no collarette; two fleshy processes (tentacles?) on first somite, adjacent to mouth; thorax hardly dis- tinguishable from abdomen and composed of nine somites. Branclnae (Fig. 193) 14 on each side, connected by a web (w,) as far as radioles extend, /*. e,^ within 3 mm. of tip ; radioles slender, biserial; ti{)s of branchiae spirally coiled in retraction. Thoracic setae of two forms: (1) double-bordered capillary (Fig. 194), very slender, often twisted; and (2) blunt, spinous setae (Fig. 195) with conical tips, much fewer in number; both kinds occur together in tufted fascicles. Uncini of thorax with long manubria (Fig. 197), strong rostra, and minute teeth ; uniserial. Abdominal setae (Fig. 195) slender, minute, very broadly double- bordered ; alxlominal uncini (Fig. 198) very minute, avicular, 2-3 teeth above rostrum; arranged in tori which nearly encircle the body, being interrupted only by the setigerous papillae and a median stripe on doi*sum ; uniserial. Digitized by VjOOQ IC 432 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Length of large specimen (not including branchiae), 60 mm.; greatest transverse diameter of same, 9 mm. ; length of retracted branchiae, 21 mm. This species is represented by two female specimens, one col- lected by Miss Robertson at Port Orchard in June, 1899, the other by Professor Ritter at Pleasant Beach in May of the same year. Although exhibiting great difference as to size — one being more than twice the length of the other — they both contain nearly ripe ova, indicating sexual maturity. The transparent mucous envelope so characteristic of this genus was preserved with both specimens. Sebpulidae. oO. Serpula columbiana sp. nov. PI. 19, figs. 199-204. Form subterete; somewhat flattened dorso-ventrally ; gradually tapered towanis posterior extremity; fecal groove distinct, dorsal as far as thorax, there branching and passing to the ventral side under the thoracic membrane. Thorax with seven setigerous somites, well-developed collarette, and thoracijB membrane, the latter reaching to the tips of the setae. Abdominal somites, 250 or more. Branchial filaments 54 on each side, arranged in two spirals ascending from the ventral edges, each making a complete turn, then extending mesad and meeting each other just over the ibonth. Branchiae whitish, beautifully banded with scarlet or crimson ; distri- bution of color variable but usually a broad red band at or near base, followed by two narrow bands, and branchiae broadly tipped with the same. Operculum (Fig. 199) on right side, its mate on the left very short and rudimentary ; funnel-shaped, with about 100 ribs which form a notched border; deep, funnel-shaped cavity; pedicle geniculate just below the operculum. Bayonet-setue of first setigerous somite as in Fig. 200 ; the other thoracic setae broadly striate-bordered (Fig. 201). Uncini of thorax and abdomen similar in shape, 6-S toothed ; the tip of largest tooth often turned outward (Figs. 202, 203). Chisel-shaped setae of abdomen as shown in Fig. 204. Tube white, calcareous, more or less coiled ; anterior portion of old tubes often free from the substratum to which tube is attached. Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND REGION. 433 Length of large specimen, 55 mm. ; greatest transverse diameter of thorax, 7 mm. ; of abdomen, 6 mm. This beautiful Serpnlid is abundant in Puget Sound. Harrington and Griffin ('97, p. 103) mention a handsome Serpula^ probably this species, which forms " hard white tubes " on the rocky bottom of Hood's Canal ; and also as " whitening the rocks with its calca- reous tubes," along the beaches near Port Townsend. It was col- lected at Alki Point by Miss Robertson. It occurs also on the Cal- ifornia coast at Bolinas (Duxbury Reef) , and at Lime Point and Point Cavallo, on the northern shore of the Golden Gate. Its favorite habitat is the under side of a stone where the water flows freely. If not in a tide-pool, it is near extreme low-water mark. The uncini show considerable variation on the same individual, and even on the same torus. The number of teeth ranges from six to eight ; the upper border is high-arched or nearly straight. The tip of the large tooth may or may not be recurved. The coloration of the branchiae is also variable, both as to tint and distribution. It is either scarlet or damask-red, *and it may in- volve nearly the whole of the branchiae and operculum, or may be more restricted, so that the white predominates. If formalin speci- mens are not exposed to direct sunlight the color is retained for years in almost its original brightness. This may possibly be identical with Sevpvla jttkesii Baird, de- scribed by Grube ('77) from North Japan. His description is too brief, however, to put the question of identity beyond doubt. 51. Serpula zygophora sp. nov. PI. 19, figs. 205-208. Form nearly cylindrical ; abdomen strongly grooved on dorsal aspect, the somites marked on each side by transverse ridges ; seven thoracic setigerous somites. Branchiae spiral, thirty filaments on each side; carmine-red at base, and broadly barred with the same ; operculum (Fig. 205) on right side, funnel-shaped, moderately cupped, 26-ribbed, the ribs extending to the center of the concavity ; base yoke-shaped ; pedi- cle long and curved, geniculate just below the operculum; corre- sponding filament of left side club-shaped. Operculum and pedicle variegated with red. First setigerotis somite with bayonet-setae (Fig. 206). Thoracic uncini six-toothed (Fig. 207) ; abdominal uncini (Fig. 208) five- or six-toothed. Chisel-shaped abdominal setae almost identical with those of Serpula cohunhiana. Transverse diameter of thorax, 3.5 mm. ; of ablomen, 3 mm. Tube lacking. Digitized by VjOOQ IC 434 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. This species is represented by a single imperfect specimen (lack- ing posterior portions), collected by Miss Robertson at Alki Point. As it has never been collected, so far as known, on the California coast, it may be inferred that its distribution is northward. Digitized by VjOOQ IC JOHNSON: POLYCHAETA OF PUGET SOUND REGION. 435 LITERATURE. Andrews, E. A. *91. The distribution of Magelona. Johns Hopkins univ. circulars, vol. 10, no. 88, p. 96. Arwidsson, Ivar. *98. Studien tiberdie familien Glyceridae und Goniadidae. Bergens mu- seums aarbog, no. 11, pp. 1-69, 4 taf. Beiiham, W. B. ^ . *96. The blood of Magelona. Quart, joum. micros. 8ci.,vol. 89, pp. 1-17, Ipl. Clapar&de, Ed. •68-*70. Les Ann61ides Ch^topodes du (Jolfe de Naples. M6m. de la soc. de phys. et d* hist. nat. de Genfeve, t. 19, pp. 313-584, 16 pis.; lere part., pp. 1-225, 15 pis.; 2e part., pp. 365-542. Darboux, J. G. '99. Recherches sur les Aphroditiens. Travaux de Tinstitut do zoologie de Tuniversitfi de Montpellier, etc., nouv. sCr., m6m. no. 6, 276 pp., 83 figs. Dean, B., N. R. Harrington. G. N. Calkins, & B. B. Griffin. '97. The Columbia University zoological expedition of 1896, etc. Trans. N. Y. acad. sciences, vol. 16, pp. 33-42 (with map of Puget Sound). Ehlers, Ernst. *64-'68. Die borstenwUrmer (Annelida Chaetopoda), bd. 1, Nereidea, 4to, 748 pp., 24 taf. Gaml)le, F. W., & J. H. ^Vshworth. : 00. The anatomy and classification of the Arenicolidae, with some ob- servations on their post-larval stages. Quart, journ. micros, sci., vol. 43, pp. 419-569, 8 pis. Grube, Ed. '51. Middendorffs Reise in den ftussei-sten norden und osten Siberiens, bd. 2 (Zool.), th. 1. (Annelida.) '55. Beschreibung neuer oder wenig bekannter Anneliden. Arch. fUr naturgesch., jahrg. 21 (1855), bd. 1, pp. 81-136, 3 taf. '1?. Neue Anneliden aus Japan. 55 Jahresber. der Schles. gesellsch. ftir vaterl. cultur, pp. 104-106. Harrington, N. R, ' • '97. On Nereids commensal with Hermit Crabs. Trans. N. Y. acad. sci- ences, vol. 16, pp. 214-221, 3 pis. Harrington, N. R., & B. B. Griffin. '97. Notes upon the distribution and habits of some Puget Sound inver- tebrates. Trans. N. Y. acad. sciences, vol. 16, pp. 152-165. Haswell, W. A. '86. On the structure of the so-called glandular ventricle (drUsenmagen) of Syllis. Quart, joum. micros, sci., vol. 26, pp. 471-480, 1 pi. Digitized by VjOOQ IC 436 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. Johnson, H. P. "97. A preliminary account of the iflarine Annelids of the Pacific coast, with descriptions of new species. Proc. Cal. acad. sciences, 3d ser., (Zoology). voL 1, pp. 153-198, 6 pis. Kinberg, J. G. H. '55. Nya slftgten och arter af Annelider. 1. Aphroditea. Ofrereigt kongl. vetenskaps-akad. forhandl. Stockholm, 12, pp. 381-388. '58. Kongliga svenska fregatten Eugenies resa omkring jorden, Zool., 1, pp. 1-52, 8 pis. Stockholm. *65. Annulata nova. Ofversigt kongl. yetenskaps-akad. forhandl. Stock- holm, pp. 22, 167-179, 239-258. Lord, J. K. *66. The naturalist in Vancouver Island and British Columbia. 2 vols., 8vo. London. M»Int06h, W. C. '78. Beitrftge zur anatomic von Magelona. Zeitsch. fttrwissensch. zooL, bd. 31, pp. 401-472, 10 taf. '85. Report on the Annelida. Polychaeta. Voyage of H. M. S. " Chal- lenger,'" Zoology, vol. 12, pp. 86+564, 94 pis., and map. Malaquin, A. *93. Recherches sur les Syllidiens. Morphologic, anatomic, reproduction, d^veloppement. 477 pp., 14 pis. Lille: L. Danel. Malmgren, A. J. '65. Nordiska Hafs-Annulater. Ofversigt kongl. vetenskaps-akad. for- handl. Stockholm, 22, pp. 61-110, 181-192, 365-410, 8 tafl. v. Marenzeller, E^ '79. Stidjapanische Anneliden, 1. Denkschr. k. akad. wissensch. Wien, math.-naturwissensch. classe, bd. 41, pp. 109-164, 6 taf. '84. SOdjapanische Anneliden, 2. Denkschr. k. akad. wissensch. Wien, math.-naturwissensch. classe, bd. 49, pp. 197-224. '87. *Polychftten der Angra Pequena Bucht. Zool. jahrb., abth. fOr system- atik. bd. 3, pp. 1-24, 1 taf. *90. Annulaten des Berings-Meeres. Ann des. k. k. naturhist. Hofmu- seums, t, 5, pp. 1-8, 1 taf. AV^ien. Mesnil, F., & M. Caullery. 'd8. Etudes de morphologie exteme chez le.s Ann61ides. 4. La famille nouvelle des Levinsfenieus, etc. Bull, scient. de la France et de la Bel- gique, t. 31, pp. 126-160, 1 pi. Michaelsen, W, '92. Polychaeten von Ceylon. Jahrb. der Hamburg wissensch. anstalten, bd. 9, pp. 91-113, Itaf. Mttller. Fr. '58. Eiiiiges tiber die Annelidenfauna der Insel Santa Catharina an der brasilianischen kUste. Arch. fQr naturgesch., jahrg. 24 J1868), pp. 211- 220, 1 taf. Murdoch, J. '84. Description of seven new species of Crustacea and one new worm [Arenicola glacialis'} from arctic Alaska. Proc. U. S. nat. mus., vol. 7, pp. 618-622. Digitized by VjOOQ IC JOHNSON : POLYCHAETA OF PUGET SOUND BEGION. 437 Roule, L. *96. Rfisultats scientiflques de la Campagne du "Caudan" — 1896. An- n^lides, O^phyriens. Ann. de Tuniv. de Lyon, pp. 489-474, 7 pis. Paris: Masson et Cie. Stimpson, Wm. *64. Descriptions of new species of marine Invertebrata from Puget Sound, collected by naturalists of the Northwest Boundary Commission. Proc. acad. nat. sciences Phila., 1804, pp. 163-161. Verrill. A. E. : 00. Additions to the Turbellaria, Nemertina, and Annelida of the Ber- mudas, etc. Trans. Conn. acad. arte and sciences, vol. 10, pp. 695-672, Ipl. Webster, H. E. '79. On the Annelida Chaetopoda of the Virginian coast. Trans. Albany institute, vol. 9, pp. 202-272, 11 pis. Webster, H. E., & J. E. Benedict. '87. The Annelida Chaetopoda from Eastport, Maine. Report U. S. comm. iish and fisheries for 1886-87, pp. 707-766, 8 pis. Wir6n, A. *B3. Chaetopoder fr&n Sibiriska Ishafvet och Berings haf , insamlade under Vega-expeditionen, 1878-79. Vega-expedlt. vetensk. jakttagelser, bd. 2, pp. 381-428, 6 taf. Printed, August, 1901. Digitized by VjOOQ IC Johnson.— Polychmet*. PLATE 1.' Fig. 1. Anterior extremity, dirsal aspect, of PAynoe fragilis. The elytra have fallen ofif. X 8.0. FigH. 2-7. Harmothoe iphionelloidea. Fig. 2. Anterior extremity, dorsal aspect ; proboscis exserted ; anterior ely- tra removed. X 8.6. Fig. 3. Fifth elytron, right .side. X 8.5. Fig. 4. Third foot from right side, dorsal aspect. The setae above the dorsal cirrus all belong to tlie dorsal fascicle, x 28. Fig. 6. Ventral seta-tip, profile. X 200. Fig. 0. Stout, curved, dorsal seta. X 200. Fig. 7. a. Slender doi-sal seta. X •'>3. b. Tip of same, more magnified. X 200. ^With very few exceptions (in each instance, stated), the tlgares are from camera fh-awin^fl. In all the plates the drawings have been reduced one-half, and the mA^iflt cation, as piven with the explanation of each tlKure, has been corrected accordingly. ' Digiti ized by Google Johnson. — Polyciiaeta. Plate 1. M.Pj:^LHj:.del, Pkoc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JOHVSOK. " PolychMU. PLATE 2. Figs. 8-18. Harmothoe complanata. Fig. 8. Anterior extremity, dorsal aspect. X 13. Fig. 9. Twenty-first foot, anterior aspect. X 23. Fig. 10. Elytron from left side ; the nerves radiating from the elytrophore are distinctly seen. X 13. Fig. 11. Tip of stout dorsal seta. X 200. Fig. 12. Top of slender dorsal seta. X 100. Fig. 13. Tip of ventral seta, x 200. Figs. 14-17. Harmotkoe paciflca. Fig. 14. Anterior aspect of second foot. X 13. Fig. 16. Tip of slender, slightly curved dorsal seta. X 100. Fig. 16. Tip of strongly-curved dorsal seta. X 100. Fig. 17. Tip of ventral seta. X 100. Fig. 18. Harmothoe tuta ; anterior extremity, after removal of elytra. X 8.5. Fig. 19. Seventeenth parapod of the same. Anterior aspect. X 17. Digitized by Google JonNSON. — POLYCHAETA. Plate 2. TTM^dU Proc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHNHON. — Polychaeta. PLATE 3. Figs. 20-22. Harmothoe tuta. Fig. 20. Elytron from left .side. X (».'). Fig. 21. I)on>al seta from eighteintli foot. X 200. Fig. 22. Ventral seta-tip from eighteenth foot. X 200. Figs. 2.*)-2.'). Podarke jmgettensis. Fig. 28. Anterior extremity, probosci.s exserted. X 16. Fig. 24. Parapod from middle of the body, X 32. o. Fig. 25. Tip of a ventral (ecmipound) !*eta. X 200. Figs. 2(^-30. Ntreis virens. Fig. 20. Anterior extremity, dorsal aspect, of large specimen. X 0.5. Fig. 27. Foot from middle of body, anterior as] ect. X ^.5. Figs. 28-30. Tii»s of .**etae from ventral fascicle, sliowing graduation in length of ap])endage. X 150. Fig. 31. Tenth foot of Xereis vextllosa, posterior asi ect. X 23.5. Fig. 32. Natatoiy seta, female Heieronereis of X. vexillosa. X 2tX>. Digitized by VjOOQ IC Johnson. — Polyciiakta. Plate 8. HT.Jy^LH.J^deL Pboc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHirsoH . — Polycbaeta. PLATE 4. Figs. 33-38. Nereis vexillosa. Fig. 33. Anterior aspect of heteronereized foot. X 8.5. Fig. 34. Foot from posterior portion of body of young female. X 8.5. Fig. 35. Foot from pf. P0LY( HAKTA. Plate 4. HPj:.del. Puoc. Bust. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JomrsoK. " PolycbaeU. PLATE 6. Figs. 48-52. NtreU epelurus, Fig. 48. Anterior extremity, dorsal aepeet, showiag large, coUariform peri- stomium, within whicli the pro«tomius ia partially withdrawn. Young specimen. X 8.5. Fig. 49. Tenth foot, anterior aspect. X 13. Fig. 50. Falcate seta from lower i^acicle, fentral ramiB, of foot from anterior region of body. X 200. Fig. 51. Falcate seta from upper fascicle of ventral ramus. X 200. Fig. 52. " Fish-bone *' seta, upper fascicle^ rentral ramus. X 200. Figs. 53-59. Nereis proeera. Fig. 53. Anterior extremity, doreal ajq>ect. X 8.6. Fig. 54. Forty -first foot, showing 8tout» dorsal seta and three slender ones. X23. Fig. 55. Foot further back ; slender dorsal setae no longer present. X 23. Fig. 66. Foot from posterior region, near pygidium. X 23. Figs. 57, 58. '* Fish-bone " and falcate setae from Tentral ramus. X 200. Fig. 59. 8tout dorsal seta. X 200. Digitized by VjOOQ IC JOIIXSON. — POLYCHAKTA. J*LATE 5. HPJ.del Proc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by Google JoHKtOK.— PolxcbAeta. PLATE 0. Figs. 60-66. Euphroayne heterobranchia. Fig. 60. The prostomium, dorsal aspeUN»«>M. — Folychacta PLATE 9. Figs. 88-92. Northia irUUscena. Fig. 88. Anterior extremity^ dorsal aspect Free-hand drawing, x 3. Fig. 89. Foot from middle of series, showing the filiform gill and sabolate dorsal cirrus, x 23.5. Fig. 90. Wing-tipped uncinos, from a middle parapod. x 200. Fig. 91. Hooded crotchet from ventral fascicle of third foot. X 190. Fig. 92. Capillary seta with striated border. X 200. Figs. 93-100. Lutnbriconereis zonata. Fig. 93. Anterior extremit>% dorsal aspect X 8.6. Fig. 94. Maxillae, somewhat spread apart, and seen from the dorsal side. X 23.6. Figs. 96, 96. A "spiral" somite from posterior fegion of body, seen from the dorsal (Fig. 95) and from the ventral aspect (Fig. 96); 1, 2, 3. the somites involved, x 8.6. Fig. 97. Fifteenth foot, posterior aspect X 40. Fig. 98. Foot from middle of series, posterior aspect Unclni only are pres- ent X 23.6. Fig. 99. Double-bordered seta from an anterior foot. X 237. Fig. 100. " Hooded crotchet" from a posterior foot. X 237. Digitized by VjOOQ IC Johnson. — Polychaeta. Plate 9. H.VJ.dcl^ P»o(\ BosT. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHirsoir. ~ PolychMta. PLATE 10. Fig. 101. Posterior aspect of foot of Glycera rugosa, from middle of body fihowiug 6-lobed gill, and dorsal cirrus {d. c.) seen by transparency through uppermost lobe of gill (the tip of which has been cut off). The minute ova have entered the gills, c. c. Ventral cirrus, x 23.i>. Fig. 102. Jaw and jaw-appendage (ap.) of the same. X 13. Fig. 108. Posterior aspect of foot of Glycera nana. Dorsal cirrus is not shown. V. c. Ventral cirrus. X 23. Fig. 108a. Jaw-appendage of the same. X 40. Fig. 104. Posterior aspect of a foot from middle of series of He mipodia borealis. Setae are all compound, and several are almost entirely withdrawn witliin the foot. The single acicula is indicated by dotted lines. d. c. Dorsal cirrus, v. c. Ventral cirrus. X 75. Fig. 104a. Jaw-appendage of the same. The attached end is somewhat ex- panded. X 75. Figs. 105-110. Scoloplos elongata. Anterior extremity, dorsal aspect pp. Palpode. X 13. Anterior extremity, ventral aspect, with proboscis everted, showing foliaceous expansions. X 18. Profile of seventeenth foot, anterior aspect. X 23.6. Vertical section of foot from middle of length, showing the dorsally- directed parts, br. Gill. X 52.6. Portion of one of the deeply-serrated, anterior setae, x 340. Portion of a ventral seta. X 340. Anterior extremity of Aricideopsis megalops. X 39. The same. Foot from posterior portion, x 89. Fig. 105. Fig. 106. Fig. 107. Fig. 108. Fig. 100. Fig. 110. Fig. 111. Fig.ai2. Digitized by VjOOQ IC Johnson. — Polychaeta. Plate 13. Proc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHKSox.— Polycbaeta. PLATE 14. Figs. 140-142. Ammockares occidentalis. Fig. 140. ADterior portion, ventral aspect. The tori are represented, but the uncini are invisible at this magnification. X 13. Figs. 141 a, b. Uncini, frontal aspect and profile. When in situ, only the por- tion distal to the constriction is exposed, x 600. Fig. 142. Seta, showing double serration, x 460. Fig. 143. Portion of digestive tract of Arenkola claparedei^ showing lower sec- tion of oesophagus (oes.); oesophageal coeca of two kinds (c, c')« and the anterior extremity of the chlorogogous tract (cA.). The most anterior pair of coeca (c.) are always much larger than the rest, thin-walled and highly vascular. The other coeca (c'.) are thick- walled and the surface is striated. They are not always paired, and the number is highly variable (ten in the present instance, which is about the minimum). Free-hand drawing, x 2. Fig. 144. Oesophageal coeca of another specimen, showing the thin-walled, vascular pair (c.) in a much-contracted condition. The smaller coeca (c'.) number fifteen on each side (with one exception the highest number hitherto observed) and are <|uite uniform as to size. Several are attached at or near the median line. Free-hand drawing, x 2. Figs. 145-148. Cirratulus cingulatus. Fig. 145. Anterior portion, showing tentacular filament {tf.) and three most anterior pairs of cirri, all cut short. X 0. Fig. 14tJ. Two somites from middle of body. The dorsal cirri are broken off, but the points of attachment are shown. X 13. Fig. 147. Ventral setae, and two uncini in situ, x 187. Fig. 148. Ventral hook, middle region, x 150. Fig. 149. Anterior portion of Cirratulus robustus. The tentacular filament {t. /.) and dorsal cirri are broken off at point of attachment x 9. Fig. 150. Somites Si^^S (anterior third of body), left side of the same. Points of attachment of cirri are distinctly shown, x 13. Digitized by VjOOQ IC Johnson. — Polychaeta. Plate 14. fJUl>\\ f4z. HM.del. Paoc. BosT. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by Google JoH5tK)N — Polvchaeta PLATE 16. Fip*. 161-1 5- Polychaeu. PLATE 16. Fig. 102. Anterior extremity of Sabellides anops, dorsal aspect Branchiae on right side have been cut off at the base. The somites 1-5 are numbered. X 9. Fig. 103. Protstomium and tentacleH of the same, dorsal aspect. X 15. Figs. 104-108. Amphitrite robusta. Fig. 104. Gill, mounted in glycerine. X 0. Fig. 106. Seta from thorax. X 187. Fig. 100. Uncinus from eighteenth somite, in profile. X 321. Fig. 107. Biserial arrangement of uncinl, thoracic region. X 321. Fig. 108. Uncinus from near posterior end. X 321. Figs. 169-171. Amphitrite spiralis. • Fig. 109. One of the main branches of a gill, x 15. Fig. 170. Seta from eighteenth (thoracic) somite. X 321. Fig. 171 a, b. Uncinl from eighteenth somite, profile. X 321. Fig. 171 c. Uncinus, anterior aspect, showing five rows of teeth above rostrum. X 350. Digitized by VjOOQ IC Johnson. — PoLYniAETA. Plate 16. H.PJ.del Proc. Bost. See. Nat. Hist. Vol. 29. Digitized by Google Digitized by VjOOQ IC Digitized by VjOOQ IC JoHxsoM.— Polychaeu PLATE 17. Figs. 172-174. Lanice heterobranchia. Fig. 172. One of the first pair of gills ; ouly one branch ts given in detail. X 9. Fig. 178. Capillary, double-bordered thoracic seta. X 187. Fig. 174. Uncinus from nineteenth (thoracic) somite, x 321. Figs. 175-178. Thelepus crispus. Fig. 175. Two gill-tilaments, showing place of attachment to body. In glycer- ine. The bloixl- vessels show through the translucent walls. X 0. Figs. 17«, 177. Two capillary double-bordered setae of differing form. X loO. Fig. 178. Uncinus from thirty-second (thoracic) somite, profile view. X 321. Fig. 178 a, h. Two thoracic uncini of slightly different form. X 350. Figs. 179-183. Bispira polymorpha. Fig. 179. Anterior portion, ventral aspect, including thorax and first six so- mites of abdomen. Free-hand drawing, x 2 (circa). Fig. 180. Winged seta from dorsal portion of sixth fascicle. X 150. Fig. 181. Spatulo-mucronate seta-tip from ventral portion- of sixth fascicle. X 276. Fig. 182. Avicular and mucronate uncini from sixth torus, in their normal re- lation to each other, x 150. Fig. 183. Avicular uncinus from abdominal region (twentieth torus), x 150. Digitized by VjOOQ IC Johnson. — Polychaeta. Platk 17. H7fd.del Proc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHNDOx.— PoIycbaetA. PLATE 18. Fig. 184. Portion of gill of Bispira polj/morpha, near base, showing gill-fila- uientfi and two eyes, x 39. Fig. 185. Double-bordered, capillary seta from abdominal region. X 150. Figs. 186-192. Megachone aurantiaccu Fig. 180. Thorax, and first somite of abdomen, right side. Collar is clearly indicated, x 0. Fig. 187. Anterior extremity in end-view, showing regenerating (?) branchiae, appressed upon the disc ; their tips point ventrad. X 9. Fig. 188. Gill showing filaments, the blood-vessels of which contain oval blood- clote. X 26. Fig. 189. Bordered seta, from thorax. X 187. Fig. 190. Uncinus from sixth torus, x 150. Fig. 191. Uncinus from nineteenth torus. X 275. Fig. 192. Uncinus near posterior extremity. X 275. Digitized by VjOOQ IC Johnson. — Polychaeta. Plate 18. HPd.del Proc. Bost. Soc. Nat. Hist. Vol. 29. Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC JoHNflox. — PolychaeU. PLATE 19. Figs. 193-198. Myxicola pacffica. Fig. 193. Tip of branchia. The web connecting the branchiae is shown as a border on the right side, and as ruptured tips at w, w. X 12.6. Double-bordered, thoracic seta. X 450. Stout, straight seta from thorax, x 460. Abdominal seta. X 450. Thoracic uncinus. X 321. Abdominal oncinus. X 321. Figs. 199>203. Serpula columbiana. Profile view of operculum, x 0. " Bayonet " seta from first fascicle. X 187. Thoracia seta with striated border, x 187. Thoracic uncinus (many are d- or 7-toothed). X 275. Abdominal uncinus. x 275. Figs. 204-208. Serpula zygophora. Fig. 204. Chisel-shaped, abdominal seta. X 321. Fig. 205. Pn)file view of operculum, x 9. Fig. 20rt. "Bayonet " seu from first fascicle. X 187. Fig. 207. Thoracic uncinus («-toothed). X 321. Fig. 208. Abdominal uncinus (usually 6-toothed). X 321. Fig. 1. Echinodeinis from Puget St)und : starvations ma^le on the Ikihino- denns collect(' siematic names employed Ly writers on the morphol- ogy of the Acmaeidae. By M. A. Willcox. 0 pp. 10 cts. No. 10. On a hitherto unrecognized form of blood circulation without capillaries in the organs of verlebrata. By Charles Sedgwick Minot. 31 pp. 36 cts. No. 0. The occurrence of fossils in the Roxbury conglomerate. By Henry T. Burr and Robert E. Burke. 0 pp., 1 plate. 20 cts. No. 8. The blood vessels of the heart in Carcharias, Raja, and Amia. By G. H. Parker and F. K. Davis. 10 pp., 3 plates. 26 cts. No. 7. List of marine mollusca of Coldspring Harbor, Long Island, with descriptions of one new genus and two new species of Nudibranchs. By Francis Noyes Biilch. 30 pp., 1 plate. 36 cts. No. 0. The development of Penilia schmackeri Richard. By Mervin T. Sudler. 23 pp., 3 plates. 30 cts. No. 6. Contributions from the Gray herbarium of Harvard university. New series, no. 17. 1. Revision of the genus Gymnolomia. 2. Supplement^or^' notes upon Calea, Tridax, and Mikania. By B. I^ liobhison and J. M. Cireenman. 22 pp. 26 cts. No 4. Studies in Diptera Cyclorhapha. 1. The Pipunculidae of the Unitetl States. By Garry de N. Hough. 10 pp. 10 cts. No. 3. Notes on the reptiles and amphibians of Intervale, N. H. By Glover M. Allen. 13 pp. 16 cts. No. 2. Variation and sexual selection in man. By Edwin Tenney Brewster. 17 pp. 26 cts. No. 1. Proceedings of the Annual Meeting, May 3, 1800. 43 pp. 16 cts. To/. J«v, No. 1(). Moniloporidae, a new family of Palaeozoic corals. By Ama- dous W. (irabau. 10 i)p., 4 plates. 26 cts. No. 1'). Studies in the gold-bearing slates of Nova Scotia,. By J. F^mund Woodman. 33 j)p., 3 plates. 60 cts. No. 14. North American wood frogs. By R. H. Howe, Junior. 0pp. 10 cts. No. 13. Some Ilydroids from Puget Sound. By Gary N Calkins. 36 pp..O platefi. 60 cts. No. 12. The Odoiiate genus Macrothemis and its allies. By Philip P. Calvert. 32 pp., 2 plates. 60 cts. No. 11. ProcccdUi^s of the Annual Meeting, May 4, 1808. 20 pp. 16 cts. No. 10. On the veins of the Wolffian bodies in the pig. By Charles Sedgwick Minot. 10 pp., 1 plate. 26 cts. No. 15. Notes on a Carboniferous boulder train in eastern Massachusetts. By Myron L. Fuller. 14 pp. 16 cts. No. 8. The genus Antennaria in New England. By M. L. Femald. 13 pp. 16 cts. No. 7. The land mannnals of peninsular Florida and tlie coast region of Georgia, By Out ram Bangs. 70 pp. 76 cts. Digitized by Google Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC Digitized by VjOOQ IC I miikMrnm 3 2044 103 126 1 Digitized by Google