Volume 177 THE Number 1 BIOLOGICAL BULLETIN Marine Biological Laboratory . LIBRARY SEP 13 1989 I Woods Hole, Mass. AUGUST, 1989 Published by the Marine Biological Laboratory THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY JAN 1 ? 1990 Editorial Board Woods Hole, Mass. GEORGE J. AUGUSTINE, University of Southern GEORGE M. LANC University of California North Carolina at Chapel Hill RUSSELL F. DOOLITTLE, University of California Louis LEIBOVITZ, Marine Biological Laboratory at San Diego RUDOLF A. RAFF, Indiana University Wn i IAM R. ECKBERG, Howard University HERBERT SCHUEL. State University of New York at ROBERT D. GOLDMAN, Northwestern University Buffalo VIRGINIA L. SroFiELD, University of California at EVERETT PETER GREENBERG, Cornell University Los Angdes Schoo| of Medlclne JOHN E. HOBBIE. Marine Biological Laboratory KENSAL VAN HOLDE, Oregon State University LIONEL JAFFE, Marine Biological Laboratory DONALD P. WOLF. Oregon Regional Primate Center /•,,/».«/ MICHAEL J. GREENBERG, The Whitney Laboratory. University of Florida Managing Editor: PAMELA L. CLAPP, Marine Biological Laboratory DECEMBER, 1989 Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory'. MBL Street. Woods Hole, Massachusetts 02543. Subscriptions and similar matter should be addressed to Subscription Manager. THE BIOLOGICAL BULLETIN. Marine Biological Laboratory. Woods Hole. Massachusetts 02543. Single numbers, $25.00. Subscription per volume (three issues). $57.50($1 15.00 per year for six issues). Communications relative to manuscripts should be sent to Michael J. Greenberg, Editor-in-Chief, or Pamela L. Clapp. Managing Editor, at the Marine Biological Laboratory. Woods Hole. Massachusetts 02543. Telephone: (508) 548-3705, ext. 428. FAX: 508-540-6902. POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory. Woods Hole, MA 02543. Copyright >c 1989, by the Marine Biological Laboratory Second-class postage paid at Woods Hole, MA, and additional mailing offices. ISSN 0006-3 1 85 INSTRUCTIONS TO AUTHORS The Biological Bulletin accepts outstanding original re- search reports of general interest to biologists throughout the world. Papers are usually of intermediate length (10-40 manu- script pages). Very short papers (less than 9 manuscript pages including tables, figures, and bibliography) will be published in a separate section entitled "Notes." A limited number of solicited review papers may be accepted after formal review. A paper will usually appear within four months after its accep- tance. The Editorial Board requests that manuscripts conform to the requirements set below; those manuscripts that do not con- form will be returned to authors for correction before review. 1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies of photographs are not acceptable for review purposes.) The original manuscript must be typed in no smaller than 12 pitch, using double spacing (in- cluding figure legends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 81/: by 1 1 inches. Please, no right justification. Manuscripts should be proofread carefully and er- rors corrected legibly in black ink. Pages should be numbered consecutively. Margins on all sides should be at least 1 inch (2. 5 cm). Manuscripts should conform to the Council of Biology Editors Style Manual. 4th Edition (Council of Biology Editors. 1978 (and to American spelling. Unusual abbreviations should be kept to a minimum and should be spelled out on first refer- ence as well as defined in a footnote on the title page. Manu- scripts should be divided into the following components: Title page. Abstract (of no more than 200 words). Introduction, Ma- terials and Methods, Results, Discussion. Acknowledgments, Literature Cited, Tables, and Figure Legends. In addition, au- thors should supply a list of words and phrases under which the article should be indexed. 2. Title page. The title page consists of: a condensed title or running head of no more than 35 letters and spaces, the manuscript title, authors' names and appropriate addresses, and footnotes listing present addresses, acknowledgments or contribution numbers, and explanation of unusual abbrevi- ations. 3. Figures. The dimensions of the printed page, 7 by 9 inches, should be kept in mind in preparing figures for publica- tion. We recommend that figures be about IV: times the linear dimensions of the final printing desired, and that the ratio of the largest to the smallest letter or number and of the thickest to the thinnest line not exceed 1:1.5. Explanatory matter gener- ally should be included in legends, although axes should always be identified on the illustration itself. Figures should be pre- pared for reproduction as either line cuts or halftones. Figures to be reproduced as line cuts should be unmounted glossy pho- tographic reproductions or drawn in black ink on white paper, good-quality tracing cloth or plastic, or blue-lined coordinate paper. Those to be reproduced as halftones should be mounted on board, with both designating numbers or letters and scale bars affixed directly to the figures. All figures should be num- bered in consecutive order, with no distinction between text and plate figures. The author's name and an arrow indicating orientation should appear on the reverse side of all figures. 4. Tables, footnotes, figure legends, etc. Authors should follow the style in a recent issue of The Biological Bulletin in preparing table headings, figure legends, and the like. Because of the high cost of setting tabular material in type, authors are asked to limit such material as much as possible. Tables, with their headings and footnotes, should be typed on separate sheets, numbered with consecutive Roman numerals, and placed after the Literature Cited. Figure legends should contain enough information to make the figure intelligible separate from the text. Legends should be typed double spaced, with consecutive Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to authors' current ad- dresses, acknowledgments or contribution numbers, and ex- planation of unusual abbreviations. All such footnotes should appear on the title page. Footnotes are not normally permitted in the body of the text. 5. Literature cited. In the text, literature should be cited by the Harvard system, with papers by more than two authors cited as Jones el ai, 1 980. Personal communications and mate- rial in preparation or in press should be cited in the text only, with author's initials and institutions, unless the material has been formally accepted and a volume number can be supplied. The list of references following the text should be headed Liter- ature Cited, and must be typed double spaced on separate pages, conforming in punctuation and arrangement to the style of recent issues of The Biological Bulletin. Citations should in- clude complete titles and inclusive pagination. Journal abbre- viations should normally follow those of the U. S. A. Standards Institute (USASI), as adopted by BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor differences set out be- low. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS (BIOSIS List of Serials; the most recent issue). Foreign authors, and oth- ers who are accustomed to using THE WORLD LIST OF SCIEN- TIFIC PERIODICALS, may find a booklet published by the Bio- logical Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K.) use- ful, since it sets out the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these differ. CHEMICAL ABSTRACTS publishes quarterly supplements of additional abbreviations. The following points of reference style for THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD LIST) usage: A. Journal abbreviations, and book titles, all underlined (for italics) B. All components of abbreviations with initial capitals (not as European usage in WORLD LIST e.g. J. Cell. Comp. Physiol. NOT/ cell. comp. Physiol.) C. All abbreviated components must be followed by a pe- riod, whole word components must not (i.e. J. Cancer Res.) D. Space between all components (e.g. J. Cell. Comp. Physiol., not J.Cell.Comp.Physiol.) E. Unusual words in journal titles should be spelled out in full, rather than employing new abbreviations invented by the author. For example, use Ril Visindajjelags Islendinga without abbreviation. F. All single word journal titles in full (e.g. Veliger, Ecol- ogy. Brain). G. The order of abbreviated components should be the same as the word order of the complete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. A few well-known international journals in their pre- ferred forms rather than WORLD LIST or USASI usage (e.g. Na- ture. Science, Evolution NOT Nature. Lond.. Science, N.Y.; Evolution, Lancaster, Pa.) 6. Reprints, page proofs, and charges. Authors receive their first 100 reprints (without covers) free of charge. Addi- tional reprints may be ordered at time of publication and nor- mally will be delivered about two to three months after the is- sue date. Authors (or delegates for foreign authors) will receive page proofs of articles shortly before publication. They will be charged the current cost of printers' time for corrections to these (other than corrections of printers' or editors' errors). Other than these charges for authors' alterations. The Biologi- cal Bulletin does not have page charges. CONTENTS No. 1, AUGUST 1989 Annual Report of the Marine Biological Laboratory DEVELOPMENT AND REPRODUCTION Bosch, Isidro Contrasting modes of reproduction in two Antarc- tic asteroids of the genus Punnua, with a description of unusual feeding and non-feeding larval types . . . Fuller, S. Cynthia, Richard A. Lutz, and Ya-Ping Hu Bilateral asymmetry in the shell morphology and microstructure of early ontogenetic stages of Ano- mxi simplex Parks, Annette L., Brent W. Bisgrove, Gregory A. Wray, and Rudolf A. Raff Direct development in the sea urchin Phyllacanthus pan'ispinu* (Cidaroidea): phylogenetii history and functional modification ECOLOGY AND EVOLUTION Enzien, Michael, Heather I. McKhann, and Lynn Margulis Ecology and lite history of an amoebomastigote, 1 77 83 96 P(irinh}/itil(i ................................. Keough, Michael J. Variation in growth rate and reproduction ol the bryozoan Btigti/a ni'iitina ....................... 247 254 277 ECOLOGY AND EVOLUTION Garthwaite, Ronald L., Carl J. Berg Jr., and June Harrigan Population genetics of the common squid Loligu pealei LeSueur, 1821, from Cape Cod to Cape Hat- teras 287 Maki, J. S., D. Rittschof, A. R. Schmidt, A. G. Sny- der, and R. Mitchell Factors controlling attachment of bryozoan larvae: a comparison of bacterial films and unfilmed sur- faces . . 295 PHYSIOLOGY Stickle, William B., Martin A. Kapper, Li-Lian Liu, Erich Gnaiger, and Shiao Y. Wang Metabolic adaptations of several species of crusta- ceans and molluscs to hypoxia: tolerance and micro- calorimetric studies 303 ABSTRACTS Abstracts of papers presented at the General Scientific Meetings of the Marine Biological Laboratory 313 No. 3, DECEMBER 1989 BEHAVIOR ECOLOGY AND EVOLUTION Shuster, Stephen M. Female sexual receptivity associated with molting and differences in copulatory behavior among the three male morphs in Paracerceis sculpta (Crustacea: Isopoda) 331 DEVELOPMENT AND REPRODUCTION Boyer, Barbara Conta The role of the first quartet micromeres in the de- velopment of the polyclad HoplnjAana inquilina .... Bradneld, James Y., Robert L. Berlin, Susan M. Rankin, and Larry L. Keeley Cloned cDN A and antibody for an ovarian cortical granule polypeptide of the shrimp Pendent va/iini- Mitton, Jeffry B., Carl J. Berg Jr., and Katherine S. Orr Population structure, larval dispersal, and gene flow in the queen conch, Strombus gigas, of the Car- ibbean ...................................... 356 PHYSIOLOGY Drewes, C. D., and C. R. Fourtner Hindsight and rapid escape in a freshwater oligo- chaete ........................ ........ 363 Fisher, Charles R., James J. Childress, and Elizabeth Minnich Autotrophic carbon fixation by the chemoauto- trophic symbionts of Ri/liu pachyptila ............ Freadman, M. A., and W. H. Watson III Gills as possible accessory circulatory pumps in Lim- 372 386 344 Pennington, J. Timothy, and Michael G. Hadneld Larvae of a nudibranch mollusc (Phc^iil/n \ibngue) metamorphose when exposed to common organic solvents . 350 Safranek, Louis, and Carroll M. Williams Inactivation of the corpora allata in the final instar of the tobacco hornworm, Manduca \c.\7«, requires integrity of certain neural pathways from the brain 396 Index to Volume 177 ........................... 401 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board GEORGE J. AUGUSTINE, University of Southern California RUSSELL F. DOOLITTLE, University of California at San Diego WILLIAM R. ECK.BERG, Howard University ROBERT D. GOLDMAN, Northwestern University EVERETT PETER GREENBERG. Cornell University JOHN E. HOBBIE, Marine Biological Laboratory LIONEL JAFFE, Marine Biological Laboratory GEORGE M. LANGFORD, University of North Carolina at Chapel Hill Louis LEIBOVITZ, Marine Biological Laboratory RIIDOLF A. RAFF, Indiana University HERBERT SCHUEL, State University of New York at Buffalo VIRGINIA L. SCOFIELD, University of California at Los Angeles School of Medicine KENSAL VAN HOLDE, Oregon State University DONALD P. WOLF, Oregon Regional Primate Center Editor-m-Chief: MICHAEL J. GREENBERG, The Whitney Laboratory, University of Florida Managing Editor: PAMELA L. CLAPP. Marine Biological Laboratory AUGUST, 1989 Printed and Issued by LANCASTER PRESS, Inc. Marine Biological Laboratory LIBRARY SEP 13 1989 Woods Hole, Mass. PRINCE & LEMON STS. LANCASTER, PA THE BIOLOGICAL BULLETIN THE BIOLOGICAL Bi 11.1.1 I IN is published six times a year by the Marine Biological Laboratory. MBL Street. Woods Hole. Massachusetts 02543. Subscriptions and similar matter should be addressed to Subscription Manager. THE BIOLOGICAL Blil.l i- TIN. Marine Biological Laboratory, Woods Hole. Massachusetts 02543. Single numbers. $25.00. Subscription per volume (three issues), $57.50 ($ 1 1 5.00 per year for six issues). Communications relative to manuscripts should be sent to Michael J. Greenberg. Editor-in-Chief, or Pamela L. Clapp. Managing Editor, at the Marine Biological Laboratory'. Woods Hole. Massachusetts 02543. POSTMASTER: Send address changes to THE BIOLOGICAL BULLE i IN. Marine Biological Laboratory, Woods Hole. MA 02543. Copyright i6' 1989. by the Marine Biological Laboratory Second-class postage paid at Woods Hole, MA, and additional mailing offices. ISSN 0006-3 185 INSTRUCTIONS TO AUTHORS The Biological Build in accepts outstanding original re- search reports of general interest to biologists throughout the world. Papers are usually of intermediate length ( 10-40 manu- script pages). Very short papers (less than 9 manuscript pages including tables, figures, and bibliography) will be published in a separate section entitled "Notes." A limited number of solicited review papers may be accepted after formal review. A paper will usually appear within four months after its accep- tance. The Editorial Board requests that manuscripts conform to the requirements set below: those manuscripts that do not con- form will be returned to authors for correction before review. 1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies of photographs are not acceptable for review purposes.) The original manuscript must be typed in no smaller than 1 2 pitch, using double spacing (///- eluding figure legends, footnotes, bibliography, etc. ) on one side of 16- or 20-lb. bond paper, 81/: by 1 1 inches. Please, no right justification. Manuscripts should be proofread carefully and er- rors corrected legibly in black ink. Pages should be numbered consecutively. Margins on all sides should be at least 1 inch (2.5 cm). Manuscripts should conform to the Council ol Biology Editors Style Manual. 4th Edition (Council of Biology Editors, 1978) and to American spelling. Unusual abbreviations should be kept to a minimum and should be spelled out on first refer- ence as well as defined in a footnote on the title page. Manu- scripts should be divided into the following components: Title page. Abstract (of no more than 200 words). Introduction, Ma- terials and Methods. Results, Discussion, Acknowledgments. Literature Cited, Tables, and Figure Legends. In addition, au- thors should supply a list of words and phrases under which the article should be indexed. 2. Title page. The title page consists of: a condensed title or running head of no more than 35 letters and spaces, the manuscript title, authors' names and appropriate addresses, and footnotes listing present addresses, acknowledgments or contribution numbers, and explanation of unusual abbrevi- ations. 3. Figures. The dimensions of the printed page, 7 by 9 inches, should be kept in mind in preparing figures for publica- tion. We recommend that figures be about 1 '/: times the linear dimensions of the final printing desired, and that the ratio of the largest to the smallest letter or number and of the thickest to the thinnest line not exceed 1:1.5. Explanatory matter gener- ally should be included in legends, although axes should always be identified on the illustration itself. Figures should be pre- pared for reproduction as either line cuts or halftones. Figures to be reproduced as line cuts should be unmounted glossy pho- tographic reproductions or drawn in black ink on white paper, good-quality tracing cloth or plastic, or blue-lined coordinate paper. Those to be reproduced as halftones should be mounted on board, with both designating numbers or letters and scale bars affixed directly to the figures. All figures should be num- bered in consecutive order, with no distinction between text and plate figures. The author's name and an arrow indicating orientation should appear on the reverse side of all figures. 4. Tables, footnotes, figure legends, etc. Authors should follow the style in a recent issue of The Biological Bulletin in preparing table headings, figure legends, and the like. Because of the high cost of setting tabular material in type, authors are asked to limit such material as much as possible. Tables, with their headings and footnotes, should be typed on separate sheets, numbered with consecutive Roman numerals, and placed after the Literature Cited. Figure legends should contain enough information to make the figure intelligible separate from the text. Legends should be typed double spaced, with consecutive Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to authors' current ad- dresses, acknowledgments or contribution numbers, and ex- planation of unusual abbreviations. All such footnotes should appear on the title page. Footnotes are not normally permitted in the body of the text. 5. Literature cited. In the text, literature should be cited by the Harvard system, with papers by more than two authors cited as Jones el a/., 1 980. Personal communications and mate- rial in preparation or in press should be cited in the text only, with author's initials and institutions, unless the material has been formally accepted and a volume number can be supplied. The list of references following the text should be headed Liter- ature Cited, and must be typed double spaced on separate pages, conforming in punctuation and arrangement to the style of recent issues of The Biological Bulletin. Citations should in- clude complete titles and inclusive pagination. Journal abbre- viations should normally follow those of the U. S. A. Standards Institute (USASI), as adopted by BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor differences set out be- low. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS (BIOSIS List of Serials: the most recent issue). Foreign authors, and oth- ers who are accustomed to using THE WORLD LIST OF SCIEN- TIFIC PERIODICALS, may find a booklet published by the Bio- logical Council of the U.K.. (obtainable from the Institute of Biology. 41 Queen's Gate, London, S.W. 7, England, U.K. (use- ful, since it sets out the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these differ. CHEMICAL ABSTRACTS publishes quarterly supplements of additional abbreviations. The following points of reference style for THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD LIST) usage: A. Journal abbreviations, and book titles, all underlined (for italics) B. All components of abbreviations with initial capitals (not as European usage in WORLD LIST e.g. J. Cell. C'omp. Physio/. NOT J. cell. comp. Physiol.) C. All abbreviated components must be followed by a pe- riod, whole word components must noi (i.e. J. Cancer Res.) D. Space between all components (e.g. ./. Cell. Comp. Pliysiol.. not J.Cell.Comp.Physiol.) E. Unusual words in journal titles should be spelled out in full, rather than employing new abbreviations invented h\ the author. For example, use Ril I 'isimlafielags Islemlinga without abbreviation. F. All single word journal titles in full (e.g. I'e/igcr, Ecol- ogy, Brain). G. The order of abbreviated components should be the same as the word order of the complete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. A few well-known international journals in their pre- ferred forms rather than WORLD LIST or USASI usage (e.g. Na- ture. Science. Evolution NOT Nature. Land.. Science, N.Y.; Evolution. Lancaster, Pa.) 6. Reprints, page proofs, and charges. Authors receive their first 100 reprints (without covers) free of charge. Addi- tional reprints may be ordered at time of publication and nor- mally will be delivered about two to three months after the is- sue date. Authors (or delegates for foreign authors) will receive page proofs of articles shortly before publication. They will be charged the current cost of printers' time for corrections to these (other than corrections of printers' or editors' errors). Other than these charges for authors' alterations. The Biologi- cal Bulletin does not have page charges. Editor's Note For ten years, as the editor of The Biological Bulletin, Dr. Charles B. Metz, worked to improve the technical quality and content of the journal, and to meet the rising standards expected of it. He was eminently successful, and I am pleased to express my appreciation, and that of the Editorial Board and staff, for his efforts; he has prepared the way well for those of us who follow. The Biological Bulletin will continue to publish outstanding papers for a general readership with interests in contemporary aspects of biology. — M.J.G. The Marine Biological Laboratory Ninety-First Report for the Year 1988 One-Hundred and First Year Officers of the Corporation Denis M. Robinson, Honorary Chairman of the Board of Trustees Prosser Gifford, Chairman of the Board of Trustees Harlyn O. Halvorson, President of the Corporation and Director Robert D. Manz, Treasurer Kathleen Dunlap, Clerk of the Corporation Contents Report of the President and Director 3 Report of the Treasurer 7 Financial Statements 8 Educational Programs Summer Courses 15 Short Courses 20 Summer Research Programs Principal Investigators 27 Other Research Personnel 29 Library Readers 31 Institutions Represented 33 Year-Round Research Programs 37 Honors 43 Board of Trustees and Committees 46 Laboratory Support Staff 50 Members of the Corporation Life Members 52 Regular Members 53 Associate Members 66 Certificate of Organization 70 Articles of Amendment 70 Bylaws 71 Centennial Events Calendar . 74 Report of the President and Director Our National Marine Biological Laboratory in 1988: The Years Ahead Are Here highlighting some of the milestones of our Centennial year, 1988. // will be an interesting place to watch, in the years ahead. In a rational world, things ought to go as well for the MBL as they have in the past, and it should become an even larger and more agile collective intelligence. — Lewis Thomas In his 1974 book. The Lives of a Cell, Lewis Thomas called the MBL America's unofficial "National Biological Laboratory." In the 14 years since Dr. Thomas honored the MBL with that title, we have pointed proudly to that passage on a number of occasions — because it is flattering, because it is true, and because it comes from such a distinguished participant in, and observer of. American biology. But when you read the chapter in which the passage occurs, you find Dr. Thomas was not simply passing out compliments. In fact, his MBL chapter is a meditation on the life of institutions, and a good, hard look (from the early 1970s) at where one of America's odder institutions — the MBL — was headed. In the same sentence in which he identified the MBL as the national center for biology. Dr. Thomas identified the major challenge facing the laboratory: "It is the National Biological Laboratory without being officially designated (or yet funded) as such," he wrote. In a bit of inspired prophecy. Dr. Thomas went on to list a series of challenges the MBL was facing — and continues to face. In this 10 1st director's report, I call your attention to Dr. Thomas' analysis of the MBL because we are now well into the future he anticipated so clearly back in 1974. 1 will return to Dr. Thomas' words shortlv, after Report of the year While the years leading to up to the centennial were a time of reassessment, the centennial year itself was, in part, a celebration of the Laboratory's long and useful life. Those celebratory activities are recorded at the end of this volume and in the Autumn 1988 Collecting Net (vol. 6, no. 2). Here, it is sufficient to note that the national science community joined us in celebrating the Laboratory's past and, simultaneously, in reaffirming its status as a national facility for research and teaching — a place where senior investigators can do important work and where new generations of experimental biologists can receive the most rigorous training. But if 1 988 was a year for celebrating, it was also a year for building and strengthening. In August, the Howard Hughes Medical Institute awarded us a $4 million grant in support of our courses and our library. Announcing the grant, Hughes Institute President Purnell W. Choppin called the MBL "a richly productive research institute and one of the nation's most prized assets for scientific education." The Hughes grant includes a $ 1 million portion earmarked for modernizing the MBL/WHOI Library. The grant comes at a time when we are facing a national crisis in information management, fired by the accelerating pace and changing nature of biological research, which has created an urgent demand for more efficient information processing and retrieval. The Hughes grant has allowed us to establish a Library Planning Committee, chaired by Dr. Edward A. Adelberg, deputy provost for the biomedical sciences at Yale. The planning committee is composed of two 4 Annual Report subcommittees: Information Management and Technology, Research and Education, chaired by Dr. Carl Bowin. a geologist from the Woods Hole Oceanographic Institution; and Facilities and Collections, chaired by Nina Matheson, director of The Johns Hopkins University's William H. Welch Medical Library. The grant has also allowed us to establish an Office of Library Planning, directed by Catherine Norton, formerly the MBL assistant librarian. Long a leader in the pursuit of biological knowledge, the MBL is now taking a leadership role in the exploration of computer technology and scientific information management. The same Hughes grant provides a $3 million foundation for our summer courses, including Neurobiology, Physiology, Embryology, and Microbiology. The generous grant will support those outstanding courses over a seven-year period. Further recognition of the Laboratory's national character came in September when Congress passed and the president signed a $2.2 million appropriation toward the construction of a Marine Biomedical Institute for Advanced Studies at the MBL. The MBIAS will include a new Advanced Studies Laboratory that will house year-round researchers working in areas related to traditional MBL strengths, including cell and developmental biology, neurobiology, microbial physiology, and molecular evolution. The MBIAS will also include a new marine resources building with facilities for culturing and studying marine animals important to biomedical research. The $2.2 million appropriation is an important first step. Beyond the initial appropriation. Congress passed a bill authorizing additional funds that will be required for construction of the new facilities. The authorization does not guarantee funding, but allows us to go back to Congress for a significant portion of the $25 million dollars that will be needed to complete the MBIAS. In addition to locating new sources of support, we strengthened programs, put together an increasingly professional and effective staff", and broadened our base in a number of areas: • Dr. Leslie Garrick arrived in June to direct the Office of Sponsored Programs. This critical department has prospered under his energetic, professional, and accurate style of management. One of his reorganization efforts was the redesign of the Annual Bulletin, with the goal of making it as coherent and useful as possible. • In the summer and early autumn, two new courses got oft" to rousing starts. Eminently worthy of MBL tradition, the Methods in Computational Neuroscience course and the Molecular Evolution course are both operating at the forefront of their fields. • Our summer research space rental was at or near capacity, despite the additional space opened by the departure of the year-round NINCDS laboratories. • We worked with the American Psychological Association to develop an NIMH-funded Special Seminar Program for Minority Neuroscience Fellows, which will bring as many as 1 2 fellows and 4 faculty to the MBL in the summer of 1989. • With our neighbors and colleagues at the Woods Hole Oceanographic Institution, we established an MBL/WHOI protein and nucleotide instrumentation core facility. The facility, which will open sometime in 1 989, is supported by a $375,000 NSF grant. • A committee chaired by Dr. Holger Jannasch and Dr. George Davis reviewed the Gray Museum collection to advise us on the best management of this resource. • A committee led by Dr. Robert Barlow reviewed The Biological Bulletin and carried out a successful search for an editor to replace Dr. Charles Metz, the retiring editor who led the Bulletin with distinction for a decade. The new editor will be announced in 1 989. • On the personnel front, we worked hard in 1988 to develop the strong, professional staff" the MBL needs to meet the challenges we face in the next few years. The employees of the Laboratory have long been one of its most important resources, and in the Centennial year we tried to turn special attention to employee issues and concerns. I am very happy with the results of our search for a personnel director: Susan Goux, an experienced personnel and affirmative action professional, became the MBL Human Resources Manager in December. Over the course of the year, we also improved the employees' retirement plan and made significant progress on an employee grievance procedure. • Recognizing the increasing financial difficulties facing young investigators who want to come to the MBL, we restored Steps Toward Independence: A Program of Support for Research and Scholarship by Junior Faculty Members. Beginning in the summer of 1989, the Steps program will once again help untenured faculty from across the country come to the MBL at a time when their research careers can be most enriched by our unique facilities, associations, and training. Report of the President and Director 5 • I mentioned earlier that we found new sources of support in 1988 — but one traditional source of support deserves special mention. The Annual Fund reached new heights in 1988, with a record number of Corporation members contributing to a record total that topped $ 100,000 for the first time in the Laboratory's history. This is the second consecutive year the annual Fund has broken previous records. Additionally, philanthropic support in 1988 reached record levels. So 1988 was a year for celebrating the past and building for the future — but it was also a year for honoring those who have supported the Laboratory over the years. We had the pleasure of thanking many individuals, families, and foundations, whose names are recorded elsewhere (see the Autumn 1988 issue of The Collecting Net [vol 6, no 2] for a list of Centennial awards). In August, we were pleased to be able to acknowledge the support of two families with the Ellen Grass, Albert Grass, and John Donling at the 22 July 1988 dedication of the Grass Reference Room. dedication of the Grass Reference Room (named for Ellen R. and Albert M. Grass, in appreciation of their long-standing support of neuroscience at the MBL) and the Bay Reading Room (named for Charles Ulrick Bay, whose foundation in 1987 made the generous gift to the library that enabled us to complete the Mellon challenge grant). I cannot close out the Centenial year without thanking the many volunteers who served on Centennial subcommittees, and the good friends of the Laboratory who led those subcommittees: Garland Allen (History); Jelle Atema and Olivann Hobbie (Cultural Events); Robert Barlow (Scientific Events); John Pfeiffer (Public Information). Finally, I want to cite Dr. James Ebert for the distinguished leadership he provided throughout the Centennial observances. Most Corporation members know that Dr. Ebert served as Centennial chairman: few can know how patient, steady, and valuable his leadership was. H 'here we are at the close of 1988 The close of our Centennial year finds us in the familiar favored position we've occupied for most of our 100-year history: we are firmly situated as a national biological facility, with important programmatic ties to other major institutions. We continued to be connected, for instance, to Boston University through the BUMP program, to the University of Pennsylvania through the Laboratory of Marine Animal Health, to our near- neighbors at the Oceanographic Institution and our distant neighbors in Naples and Japan. We are working hard to strengthen the formal, institutional ties we have and to develop new ones, where that is appropriate. At the same time, we continue to serve as a national laboratory to individual scientists — the hundreds of investigators and students who come from their home institutions to spend part of the year teaching, studying, and doing research in Woods Hole. Moving just outside the immediate scientific community, we are attracting the interest of a national audience of historians and philosophers of science. In a 1988 book titled The American Development of Biology, historian Phillip J. Pauley noted, "Unlike most investigators, who experienced the 'scientific community' largely as an abstraction that referred to the variety of contacts they made in schools, journals, and professional meetings, biologists structured their professional lives around one place — the Marine Biological Laboratory, in the village of Woods Hole." At the close of our Centennial year, the general public is a bit better acquainted with the MBL and its mission. Although the science at the MBL has long been reported in the popular press by science writers, the MBL as an institution caught the attention of the national press during the Centennial. Among other national press clippings from 1988, one PBS show (Newton's Apple) devoted a half-hour to the MBL and Smithsonian magazine ran a feature article about the MBL in their June issue ("A World Center of Basic Biology Celebrates a Century of Science by the Sea"). The years ahead: the challenges "There is no way of predicting what the future will be like for an institution such as the MBL," Lewis Thomas wrote back in 1974 — then proceeded to describe precisely the challenges we have faced in the last decade and a half, and are facing still today. "One way or another," Dr. Thomas wrote, "(the MBL) will evolve. It may shift soon into a new phase. 6 Annual Report with a year-round program for teaching and research and a year-round staff, but it will have to accomplish this without jeopardizing the immense power of its summer programs, or all institutional hell with break loose. It will have to find new ways for relating to the universities, if its graduate programs are to expand as they should. It will have to develop new symbiotic relations with the Oceanographic Institution, since both places have so much at stake. And it will have to find more money, much more — the kind of money that only federal governments possess — without losing any of its own initiative." That is the prophecy I mentioned in the opening of this report, and I cite it here because each point in it has come to pass. The MBL has evolved, and is evolving. For some time now, year-round programs have been an essential part of the MBL, and they will have to play a larger role in the years to come — without jeopardizing the summer programs, of course. We are looking for ways to strengthen existing ties and to develop new ties to the universities we serve. We are exploring joint initiatives with WHOI, working with that institution's enthusiastic new director. Dr. Craig Dorman, to develop new symbiotic relationships. And we are looking for money, much more. The federal government has provided some of what we need, but our needs remain large. We have aging facilities, and, like most academic institutions, we have too often deferred maintenance. We need to build major new facilities — the Advanced Studies Laboratory and a marine resources building. We need to find a permanent endowment that will support our educational program when the generous Hughes grant runs out in 1995. We must be concerned about the difficulties faced by new generations of MBL biologists — the high costs of housing and lab fees, the paucity of parking. These challenges are numerous enough and large enough that we need the help of a broad coalition of trustees and Corporation members. The years ahead: the case for unhricllecl optimism Looking at science and the national agenda, I am convinced that the last years of this century are a propitious time for the MBL to be re-dedicating itself to serving a national science community, in the classroom and in the laboratory. In the classroom: teachers, politicians, leaders of industry, and virtually everyone else is calling for improvements in science education and training, with the sort of fervor and sense of national urgency we've not seen since the days of Sputnik. Across the country, there is a call for more and better trained young scientists — a development that bodes well for our educational program. In the laboratory: biology's contributions to health care and agriculture (and to the nation's economy) are already well-recognized. New areas of research, such as molecular evolution, computational neuroscience, molecular parasitology. and marine microbiology are attracting the attention of industry, foundations, and governmental agencies. In an era of increasingly scarce national resources, I think we will nonetheless find that opportunities — and funding — for biological research and training are going to continue to be available to the institutions best situated to serve the nation's obvious need for better science. Increasingly, there is a national call not only for more science but also for a more science-literate public. In this effort, too, the MBL has a role to play. In 1 988, we hosted the third class of MBL Science Writing Fellows, supported by the Carnegie Corporation of New York, the Sloan Foundation, and the Foundation for Microbiology. In December, one of the 1987 fellows, Washington Post science writer Boyce Rensberger, published in the Post a remarkable five-part series on basic biology. This highly sophisticated yet completely accessible series stands as a testament to the usefulness of the MBL science writing fellowships and as a model of what science writing for the public can be. In another effort to boost the level of public dialogue about science-related issues, we announced in 1988 the establishment of an annual Lewis Thomas Award for excellence in communicating basic research in life sciences to the general public. The first winner will be announced in 1989. Dealing more directly with public education, we have established, with Simon's Rock of Bard College, a summer program to train high school teachers and teach marine biology to 30 high school students. In an attempt to continue our partnership with local schools, we are setting aside a number of those 30 slots for local students. The close of our Centennial year finds us in good shape, with formidable — but surmountable — challenges before us. Where we were strong, we're still strong. Where there are national needs for biological research and education, we're well-positioned to develop appropriate new programs and new ties to complement our existing programs and ties. Most hopefully. I can report that significant national sources of support appear to be available. We made good progress in 1988. We'll need to make more, much more, in 1989 and the years immediately beyond, but I can report this year that we are already well into the process of becoming the larger and more agile collective intelligence that Lewis Thomas predicted 14 years ago. — Harlvn Halvorson Report of the Treasurer The year 1988 was a successful one for the Laboratory. Unrestricted operations other than Housing showed a deficit of revenues to expenses of $28.637, considerably better than the budgeted deficit of $80,000. You will remember that the Executive Committee approved a one-time deficit budget for 1988, recognizing the challenges of that year of administrative transition and explicitly refusing to reduce services given the financial strength and good prospects of the Laboratory. The Housing enterprises fund showed an excess of revenues over expenses of $151, 786, of which $51,786 was applied to the reduction of debt principal and $ 100,000 was used to fund depreciation for Housing. The total of the MBL's unrestricted operations, including Housing, thus showed an excess of revenues over expenses of $123, 149. While this result is gratifying, we should not be satisfied until the excess of revenues over expenses has reached the total of our depreciation on plant— $561,239 in 1988. We saw some very positive trends in 1988. For the first time since 1985, grant support of direct costs of the year-round program increased. We expect that support to increase again in 1989 and to continue to grow as the MBIAS program takes shape. Our summer research program brought in considerably more revenues in 1988— $606,218 versus $52 1,338 in 1987. That strength at the core of the Laboratory's scientific program is very heartening. 1 988 saw the receipt of the first funds from the Howard Hughes Medical Institute under its $4 million grant to the Laboratory for support of instructional program overhead and modernization of Library systems. This seven-year grant will give us the time to seek endowments to underwrite the Laboratory's unique role in biological education, and will allow study of the best route to the twenty-first century for the Library. Total expenses of the Laboratory grew by more than 8% in 1988. Almost all of the real increase, i.e., in excess of the inflation rate of 4% was in administration. You will see a similar, though smaller, increase in 1989. My experience with the Laboratory leads me to conclude that that increase is fully warranted; I am confident that we have added "beef rather than "flab." As I survey the challenges ahead of us, I am more confident of success when I look at the excellent management team that Paul Gross started, and that Harlyn Halvorson and Ray Epstein have continued to build. The continued excellence of your Laboratory depends critically on the talent and motivation of its managers. There is one area of serious financial concern to which I want to draw your attention. As I have pointed out before, our plant renovation and replacement needs have exceeded our capital-generating capacity. The result has been that capital reserve funds have been drawn down. The Laboratory's unrestricted quasi- endowment — monies set aside by the Laboratory in the past to serve as endowment until and unless otherwise needed— have declined from $ 1 ,072. 1 86 in 1 985 to $409.997 in 1 988 and are budgeted to decline to $390.000 in 1 989. Any further reductions would be imprudent, but the demands of our plant for repair and modernization will not cease. I know that we will do a better job in the near future at generating capital funds for our plant repair and replacement. By continuing to improve our financial discipline, we will generate capital through operations that can be used to renew our existing plant. In addition, as we add new facilities to our plant, we will keep our eyes on the need to assure ourselves that the necessary funds are in place, whether through operations or endowment to properly maintain them. —Robert D. Manz Financial Statements Coopers &Lybrand cedified public accouniants REPORT OF INDEPENDENT ACCOUNTANTS To the Trustees of Marine Biological Laboratory Woods Hole, Massachusetts We have audited the accompanying balance sheet of Marine Biological Laboratory as of December 31, 1988 and the related statement of support, revenues, expenses and changes in fund balances for the year then ended. We previously examined and reported upon the financial statements of the Laboratory for the year ended December 31, 1987, which condensed statements are presented for comparative purposes only. These financial statements are the responsibility of the Laboratory's management. Our responsibility is to express an opinion on these financial statements based on our audit. We conducted our audit in accordance with generally accepted auditing standards. Those standards require that we plan and perform the audit to obtain reasonable assurance about whether the financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures in the financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management, as well as evaluating the overall financial statement presentation. We believe that our audit provides a reasonable basis for our opinion. In our opinion, the financial statements referred to above present fairly, in all material respects, the financial position of Marine Biological Laboratory at December 31,1 988, and its support, revenues, expenses and changes in fund balances for the year then ended in conformity with generally accepted accounting principles. C_OOOtno t oCubnomd Boston, Massachusetts April 15, 1989 (Except for the information presented in Note J. for which the date is May 23, 1989.) O O T Tf O ""' — OO SO ON w> r i — 1 — rsi i — ON oo r-- r- so oo S ON r-1 — ON u"i sO S oo w> r*-, ON r— «^> O O IT-! ~ ON ri oo 0 (N oo oo 1 oo W, — OO SO r^- "/~i so O JC] r-l so ON SO r- r^ oo t — f*~l — •" rn r^l so ON -t — ' — . — i ri r**i >n r*-{ Tj' 0 C -T *, PI as ^ oo r i ON •/"> oc r- *n OO Tj- O — ri ri ^ S? /-. ON "/~J oo — 00 f 1 — r- ON C ON so r o g - K. ; 0 so' OO ON ON r — •: *•' oo' C sO f' o r-" s 0 '. 6 o-' 7- ~t O z ^ 0 J fe " "1 oo in 1 ( n ri 0-' r~~- *. r--' • r, *3" - PI O oa "f 5 y pj /"l -' o oo oo' -r oo i2 ^£ Ov „ [ — U", oo -r -r , PI -f f*\ C S -' ( ON oc — so' — PI 1 00 «e S V) "2 'o C M C _o 2 «3 1 p ^ — "n •o c U O C s 73 •a S c c o. 3 G a 15 X b ^^ oa •o c a O ub c 0 X •o ! c f— CQ i/> en 1 n m § c a U t/5 C/3 < 2 OJ O 15 "o D O c 0 uouidmt c _0 ra 1 notes an O '^ c o 01 ^ Z o. OO a "C u o U ^ •a c 13 u 3 (2 n a E c ea •a c 00 g C — ._ 3 'c ^ oo r3 D. '> rj i> M •a S^ .^ 3 *«* E 3 c3 t! ^ 3 tn (L> 3 ^ "2 E s o O "g 6 J2 o c y P3 — g J3 u o ?3 S rt 4> 0 « - — OJ Q ^ T3 CO O .C •§ c 0 o 0 -C U c Efl H H £ 3 i O-- O U. 00 0 £ Pi £ £ So1 § 5 — rn r j r- **•! O O *~l — ri m r i \0 rj oo r-i £ ^ 8 £ 2 £ - r.| O — 0 X V) — OO ,C r | — iy-> n-, .30 CN ^ fl rj S^ r-j O O sO OC O T -T w". 3> ^o r-- — tN «/•»(*•! r\i S 2 ri r^, £ m-3-O^OOOO I W> w-i m oo i^ O oc *t ^c r- r- T OO \O ri — u-, — ^O — r*-, r- OO f~i — r- O ft 00 ri ^ — ""^ O^ sC ^O «". 00 CT c •o UJ Lu "JS 0 ^ (C Z 0^ < CL "O taJ S s CQ •c t/5 Q Z 3 C U. Z u ^ c o ^> S I £ X. c £ oc 1 -S Z oo „ -o e o < c£ < os r~- I — oo .i u _.-£ i o C O oa Q ^ u 0 § & <2 c D ^^ -O V5 _j C/J C ~* LJJ C 03 "S O ^5 o O "° i a « Pi U > S o •c u a; O >< -P '•= £ uj 1-1 2 E *o o tn" C S 0 C _ uj D, z ^ .g e | s - 3 6 3 8 u nj C •a c u ;i.!PPORT ANC Grant reimbui direct cosl Instruction Research Recovery of in related to instructio Summer PR C g Ed It 's- 1 io IP ill fe 1 § 1 1 1 i'l 11 s 1 1 1 Il||ll3llli!|l H (5 = 0 S Total suppo revenues EXPENSES: Instruction Research Scholarships a Support activi Dormitor> Dining hall ^ 5 s S 1 eo n e •- f S c i • .9 S3 5 -o oa a: ^ < II) O — T n I I I I I I I 8 = 8 'Si S! ^H •5 H D Q O c a c. Z X s u 3 u O 1S> o O It a o s_ T3 U -£ _c 1> •o c g i2 •_• u c ess (defk nd reven vpenses FERSA sition of /ment of u o 'o1 D. ng transi ction tra c M CJ UJ n u TRANS n 5" < « B, TJ o: O. 3 3 0 p U _c o 11 = - C .= s e .2 O a a ^= a -3 Z £ 12 Annual Report Notes to Financial Statements A. ljiii/>n\ei'/ l lie Laboratory: I he purpose of'the Marine Biological Laboratory (the "Laboratory") is to establish and maintain a laboratory or station for scientific study and investigations, and a school for instruction in biology and natural history. B. Significant accounting policies Basis of presentation — -fund accounting In order to ensure observance of limitations and restrictions placed on the use of resources available to the Laboratory, the accounts of the Laboratory are maintained in accordance with the principles of fund accounting. This is the procedure by which resources are classified into separate funds in accordance with specified activities or objectives. Externally restricted funds may only be utilized in accordance with the purposes established by the donor or grantor of such funds. However, the Laboratory retains full control over the utilization of unrestricted funds. Restricted gifts, grants, and other restricted resources are accounted tor in the appropriate restricted funds. Restricted current funds are reported as revenue when received and as related costs are incurred. Endowment funds are subject to restrictions requiring that the principal be invested with income available for use for restricted or unrestricted purposes by the Laboratory. Quasi-endowment funds have been established by the Laboratory for the same purposes as endowment funds; however, the principal of these funds may be expended for various restricted and unrestricted purposes. Fixed asset': Fixed assets are recorded at cost. Depreciation is computed using the straight-line method over estimated useful lives of fixed assets. Contracts and grunt* Revenues associated with contracts and grants are recognized in the statement of support, revenues, expenses and changes in fund balances when received and as related costs are incurred. The Laboratory reimbursement of indirect costs relating to government contracts and grants is based on negotiated indirect cost rates with adjustments for actual indirect costs in future years. Any over or underrecovery of indirect costs is recognized through future adjustments of indirect cost rates. Investments purchased by the Laboratory are carried at market value. Money market securities are carried at cost which approximates market value. Investments donated to the Laboratory are earned at fair market value at the date of the gift. For determination of gain or loss upon disposal of investments, cost is determined based on the average cost method. The Laboratory is the beneficiary of certain endowment investments which are held in trust by others. These investments are reflected in the financial statements. Every ten years the Laboratory's status as beneficiary is reviewed to determine that the Laboratory's use of these funds is in accordance with the intent of the funds. The market value of these investments are $3,551.482 and $3,334,500 at December 31. 1988 and 1987, respectively. Investment income and Jislrihiilion The Laboratory follows the accrual basis of accounting except that investment income is recorded on a cash basis. The difference between such basis and the accrual basis does not have a material effect on the determination of investment income earned on a year-to-year basis. Investment income includes income from the investments of specific funds and from the pooled investment account. Income from the pooled investment account is distributed to the participating funds on the basis of their proportionate share at market value adjusted for any addition or disposals to pooled funds. C. Land, building, and equipment The following is a summary of the unrestricted plant fund assets: /W>Vf W Land $ 6X9.660 $ 689,660 Buildings 16.694,233 16,385,099 Equipment 2.299.094 2,229.651 19,682,987 19,304,410 Less accumulated depreciation (8.268.927) (7.707.689) $11,414,060 $11,596,721 Notes to Financial Statements 13 D. Retirement jund: The Laboratory has a noncontributory defined benefit pension plan lor substantially all employees. Contributions are intended to provide for benefits attributed to the service date, but also those expected to be earned in the future. Actuarial present value of accumulated benefit obligation including vested benefits of $1.604,313 as of January 1, 1988 $1.659,926 Projected benefit obligation 2,368,957 Plan assets at fair value 2,41 1,253 Projected benefit obligation less than plan assets 42,296 Unrecognized net (gain) or loss (61.000) Prior service cost not yet recognized in net periodic pension cost Unrecognized net obligation at December 3 1 , 1988 (270,946) Prepaid pension cost (pension liability) recognized in the statement of financial position $ (289.650) Net pension cost for fiscal year ending December 31, 1988: Service cost-benefits earned during the period 160,293 Interest cost on projected benefit obligation 1 57,323 Actual return on plan assets ( 1 78,653) Net amortization and deferral (30,835) Net periodic pension cost $ 108,128 The actuarial present value of the projected benefit obligation was determined using a discount rate of 7.3% and rates of increase in compensation levels of 6%. The expected long-term rate of return on assets was 8%. In addition, the Laboratory participates in the defined contribution pension program of the Teachers Insurance and Annuity Association. Expenses amounted to $130,677 in 1988 and $103. 386 in 1987. E. Pledges and grants: Asot December 31. 1988, the Laboratory reported active pledge and grant commitments outstanding of $3,757,085 (unaudited) to be received as follows: Restricted 1989 $2.887.085 1990 810.000 1991 60,000 $3,757,085 F. Interfund borrowings: Interfund balances at December 3 1 are as follows: Current Funds 1988 1987 Due from plant funds $ 3,240 Due to restricted endowment fund $(31,600) (120,875) Due to restricted quasi-endowment funds (3,000) Due from current restricted fund 64,3 1 8 $(31,600) G. Mortgage and notes payable The mortgage note payable with a term of 26 years is in the amount of $1.3 million bearing interest based on the bank's prime rate plus three quarters percent (.75%) on a floating basis for the initial five-year period with a floor of 1.50% and a ceiling of 13.00%.. The interest rate at December 31, 1988 was 12.5'°.. The mortgage loan is collateralized by a first mortgage on the land and properties known as Memorial Circle, with recourse in the event of default limited to this land and property and the related revenue. Monthly principal and interest payments of $15,000 commenced January 19. 1987. Other notes payable consist of the following: Unsecured note with interest at 7.90%, with monthly principal and interest payments of $22 1 .20 $2.522 Unsecured note with interest at 6.90%, with monthly principal and interest payments of $394.7 1 2,704 $5,226 14 Annual Report At December 31,1 988, these mortgages and notes payable had aggregate future annual principal payments as follows: 1989 1990 1991 1992 1993 Thereafter Less current portion Amount $ 49,968 38.641 43,262 48,990 55,477 969.119 1.205,457 49.968 $1,155,489 H. Investments: The following is a summary of the cost and market value of investments at December 31, 1988 and 1987, and the related investment income and distribution of in vestment income for the years ended December 31, 1988 and 1987. Endowment and quasi-endowment U.S. Government securities Corporate fixed income Common stocks Preferred stock Money market securities Real estate Total Less custodian fees Restricted current fund Certificates of deposit Money market securities Total Total investments Cost 1988 $1,328,927 3,124,493 3,717,850 916,280 15.749 9.103,299 638,603 1,750,000 2.388.603 $11,491,902 1987 $1,299,763 2,136,500 4,088,042 9,611 1,292,097 15.749 8.841.762 850,000 850.000 $9,691,762 Market 1988 1987 $1,323,105 $ 1.270.857 3,131,404 2,134.363 5 375 980 5.545.523 14,973 916.280 1,287,520 15,749 15,749 10.762,518 10.268,985 638,603 1,750.000 2.388.603 $13.151,121 850.000 850,000 Investment Income $11,118,985 1988 1987 $ 95,234 257,692 211,319 250 49,995 $218,919 104,583 204,776 1,019 42.911 614,490 (44,073) 570,417 572.208 (54,290) 517,918 38,603 66,248 104,851 40,360 40,360 $675,268 $558,278 I. (Jilt support lor instructions: Unrestricted gifts includes $341,015 of gifts for the support of the Laboratory's instruction program available for indirect costs attributable to the instruction program. J. Subsequent event On May 23, 1989, the Laboratory terminated its defined benefit pension plan. Approval of the termination is currently being reviewed by the IRS. All benefits earned by employees under the terminated plan become fully vested. The plan assets will be distributed to plan participants. Educational Programs Summer Courses Biology of Parasitism (June 12 to August 12) Directors Paul Englund, Johns Hopkins School of Medicine Alan Sher, NIAID, National Institutes of Health Faculty Stephen Beverley, Harvard Medical School Jennie Blackwell, University of Alabama School of Medicine Stephen L. Hajduk, University of Alabama School of Medicine Gerald W. Hart, Johns Hopkins School of Medicine Carole Long, Hahnemann Medical College David Sacks, National Institutes of Health Phillip Scott, National Institutes of Health Larry Simpson, University of California, Los Angeles Mervyn J. Turner, Merck, Sharp & Dohme Laboratory Donald L. Wassom, University of Wisconsin Lecturers Barry Bloom, Albert Einstein School of Medicine Andre Capron, Pasteur Institute Monique Capron, Pasteur Institute Tony Cerami, Rockefeller University Alan Cochrane, NYU Medical Center Robert Coffman, DNAX Palo Alto, California John Donelson, University of Iowa Ronald Germain, National Institutes of Health Judith Glaven. George Washington University Michael Good, National Institutes of Health Michael Gottlieb, Johns Hopkins School of Hygiene and Public Health Russell Howard, DNAX, Palo Alto, California Paul Knopf, Brown University Wayne Masterson, Johns Hopkins School of Medicine Ted Nash, National Institutes of Health Tom Nutman. National Institutes of Health Onesmo Ole-Moiyoi, ILRAD Edward Pearce, NIAID, National Institutes of Health Elmer Pfefferkorn. Dartmouth Medical School Tom Quinn, National Institutes of Health, Johns Hopkins School of Medicine Jonathan Ravdin, University of Virginia Jose Ribeiro, Harvard University School of Public- Health Theresa Shapiro, Johns Hopkins School of Medicine Ethan Shevach, National Institutes of Health Barbara Sollner-Webb, Johns Hopkins School of Medicine Andrew Spielman, Harvard School of Public Health C. C. Wang, University of California, San Francisco Kenneth Warren, Rockefeller Foundation Rick Young, Massachusetts Institute of Technology Fidel Zavala, New York University Medical Center Students Myrna C. Bonaldo, Fundacao Oswaldo Cruz, Brazil Clotilde K. Carlow, New England Biolabs 15 16 Annual Report Glenn R. Frank, Colorado State University Ricardo T. Gazzinelli, Federal University of Minas Gerais, Brazil H. U. Goringer, Max-Planck Institute fur Molekulare Biologie, FRG Jean-Murie J. Grzych. Institut Pasteur, France Maria Lucia S. Guther, Paulista School of Medicine, Brazil Eric R. James, Medical University of South Carolina Jamil Kanaani, The Hebrew University, Israel Michel E. Ledizet, Rockefeller University Anna M. Lyles, Princeton University Rona J. Mogil, University of Alberta, Canada Jeffrey B. Moore, State University of New York, Albany Silvia N. Moreno, Institute of Microbiology, Federal University of Rio de Janeiro, Brazil Peter C. Sayles, University of Wisconsin, Madison Mariane M. Stefani, Federal University of Goias, Brazil Embryology: Cell Differentiation and Gene Expression in Early Development (June 19 to July 30) Director Eric H. Davidson, California Institute of Technology Assistant director J. Richard Whittaker. Marine Biological Laboratory Other faculty, lecturers, and staff Emmeline J. Chiao, Bryn Mawr College Gary Freeman, University of Texas at Austin John C. Gerhart, University of California, Berkeley Judith E. Kimble. University of Wisconsin Marc W. Kirschner, University of California, San Francisco Mary LaGrange, University of Massachusetts Elias Lazarides, California Institute of Technology Howard D. Lipshitz, California Institute of Technology Anthony P. Mahowald, Case Western Reserve University Dennis L. Smith, University of California, Irvine Dari K. Sweeton, Princeton University Eric Wieschaus, Princeton University Keith R. Yamamoto, University of California, San Francisco Course assistants Roberta R. Franks, California Institute of Technology Michael Garabedian, University of California, San Francisco Andrew D. Johnson, University of California, Irvine David Kimelman, University of California, San Francisco Andrew Ransic, University of Texas, Austin Teresa R. Strecker, California Institute of Technology Students Miguel Allende. University of Pennsylvania Kristin P. Carner, Scripps Institute of Oceanography Wayne F. Daugherty. San Diego State University James M. Denegre, Wesleyan University Dali Ding, California Institute of Technology Janice P. Evans, University of North Carolina Johan J. Geysen, University of Leuven, Belgium Neal R. Glicksman, University of North Carolina Jill K. Hahn, Boston L'niversity Marine Program Jeanene P. Hanley, University of Oklahoma Alicia Hidalgo, University of Oxford, England Jacquelyn Jarzem, University of Texas Christopher J. Kelly, University of Chicago Connie Lane, University of Iowa Valeria Matranga, Institute di Biologia dello Sviluppo del Consiglio Nazionale dello Richerche, Italy Roberto A. Mayor, University of Chile, Chile Brenda Ann Peculi, Johns Hopkins University Patricia A. Pesavento, Harvard University Wendy L. Richardson, Baylor College of Medicine Susan M. Smith, Rensselaer Polytechnic Institute Elly M. Tanaka, University of California, San Francisco Henry S. Tillinghast. U. S. Air Force Academy Kellie L. Watson, University of California, Irvine Dineli M. Wickramasinghe, Tufts University Joseph E. Zahner, Johns Hopkins University Marine Ecology1 (June 19 to August 13) Director Richard Osman, Academy of Natural Sciences Other faculty, lecturers, and staff Josephine Aller, State LIniversity of New York, Stony Brook Robert Aller, State University of New York, Stony Brook William Ambrose, East Carolina University Julie Berwald, University of Massachusetts, Amherst Cheryl Ann Butman, Woods Hole Oceanographic Institution Nina Caraco, Institute for Ecosystem Studies, New York Educational Programs 17 Hal Caswell, Woods Hole Oceanographic Institution David Checkley, North Carolina State University Jonathan Cole, Institute for Ecosystem Studies Joseph Costa, Boston University Marine Program Fred Dobbs. State University of New York, Stony Brook Stuart Findlay. Institute for Ecosystems Studies Marvin Freadman, Marine Biological Laboratory Brian Fry, Marine Biological Laboratory Steven Gaines, Brown University Eugene Gallagher, University of Massachusetts, Boston Anne Giblin, Marine Biological Laboratory Patricia Glibert, Horn Point Environmental Laboratories Diane Gould, University of Massachusetts, Boston Frederick Grassle, Woods Hole Oceanographic Institution Judith Grassle, Marine Biological Laboratory Victoria Hatch, Smith College Anson Hines, Smithsonian Environmental Research Center Brian Howes, Woods Hole Oceanographic Institution William Jenkins, Woods Hole Oceanographic Institution Lisa Levin, North Carolina State University Anton McLachlan, Oregon Institute for Marine Biology Lauren Mullineaux, Woods Hole Oceanographic Institution Richard Olson, University of New Hampshire Bruce Peterson, Marine Biological Laboratory Marshall Pregnall, Vassar College Maribel Pregnall, Vassar College David Policansky, National Research Council James Porter, University of Georgia Karen Porter, University of Georgia Donald Rhoads, Boston University Marine Program Donald Rice, Chesapeake Biological Laboratories Ivan Valiela, Boston University Marine Program John Waterbury, Woods Hole Oceanographic Institution James Weinberg, Woods Hole Oceanographic Institution Robert Whitlatch, University of Connecticut Craig Young, Harbor Branch Oceanographic Institution Roman Zajac, University of New Haven Students Conchita Avila, University of Barcelona, Spain Aaron P. Corkum, Ferrum College Sabine Dittmann, Zoologisches Institute, FRG Saxon M. Gibson, Vanderbilt University Todd H. Gillmore, State University of New York, New Paltz Beatrix B. Hoecker, University of Hamburg, FRG Susan B. Kane, University of Massachusetts, Boston Dagmar Lackschewitz, Max-Plank Institute, FRG Donald J. Morrisey. University of Bristol, England Cheryl A. Ondeka, University of Massachusetts, Amherst James L. Pickney, University of South Carolina Rosana M. Rocha, University of Estadual de Campinas, Brazil Robert G. Rowan, Johns Hopkins University Roberto Sandulli. University of Naples, Italy Carola Schmager, Institut fur Meereskunde an der Universitat Kiel, FRG Wesley W. Toller, University of California, Irvine Mark S. Willcox, University College, Swansea, England Microbiology: Molecular Aspects of Cellular Diversity (June 12 to July 28) Directors E. Peter Greenberg, University of Iowa Ralph Wolfe, University of Illinois Other Faculty, staff, and lecturers Tom Baldwin, Texas A & M University Dennis Bazylinski, Woods Hole Oceanographic Institution Richard Blakemore, University of New Hampshire Andreas Brune, University of Tubingen, FRG Steven Clegg, University of Iowa Martin Dworkin, University of Minnesota Elizabeth Henry, Harvard University Ralph Isberg, Tufts Medical School Jim Ferry, Virginia Polytechnical Institute Richard Frankel, California Polytechnical Ronald Gibbons, Forsyth Dental Institute Robert Gunsalus, University of California, Los Angeles Elliot Juni, LJniversity of Michigan Andrew Kropinski, Queens University, Canada Carla Kuhner, LJniversity of Illinois Margaret McFall-Ngai, Scripps Institution of Oceanography Edward Meighan, McGill University, Canada Karl Olson, University of Illinois L. Nicholas Ornston, Yale University Randy Rothmel, University of Illinois, Chicago Rolf Schauder, Universitat Ulm, FRG Bernhard Shink, Universitat Tubingen, FRG Karen Sment. Universitv of Illinois 18 Annual Report Scott Smith, University of Illinois Friedrich Widdel, Philips Universitat, FRG Students Bern Bendinger, University of Osnabruck, FRG Bette Jo Brown, Michigan Tech. Kevin R. Carmin, Florida State University Frank J. Cynar, Scripps Institute of Oceanography Eric Eisenstat, Harvard University Mark A. Fahrina, Queens University, Canada Kin Y. Fung, Wesleyan University Charles Hirsch, Merck & Company Kenneth H. Kerrick, University of Pittsburgh Jordon Konisky. University of Illinois Michael Lamontagne, Boston University Kerstin Laufer. Philips University, FRG Susanne Neuer, University of Washington Judith A. Palmer, Oklahoma University Sarah L. Storck, University of Illinois John P. Walsh, University of Massachusetts Lorraine G. Wilde, Scripps Institute of Oceanography Fitnat Yildiz. Indiana University Neural Systems and Behavior (June 12 to July 30) Directors Thomas Carew, Yale University Darcy Kelley, Columbia University Other Faculty, staff, and lecturers Axel Borst, Max-Plank Institut fur Biologische Kybernetik, FRG Jack Byrne, University of Texas Medical School Ronald Calabrese, Emory University Ellen Elliot, University of North Carolina Russell Fernald, University of Oregon Sally Hoskins, City University of New York Richard Levine, University of Arizona Eduardo Macagno, Columbia University Laurie Tompkins, Temple University Janis Weeks. University of California, San Diego Students Deborah J. Baro, University of Illinois Glen D. Brown, University of Iowa Sumantra Chattarji, Johns Hopkins University Andrew K. Cheng, Johns Hopkins University Jacqueline B. Fine-Levy, Georgia State University Thomas M. Fischer, University of California, Riverside Siglinde Gramol, The Hebrew University of Jerusalem, Israel Fred J. Helmstetter, Dartmouth College Robert Huber, Texas Tech University Kaaren L. Johanson, Vanderbilt University Ann M. Lohof, University of California, Los Angeles Joseph A. Mindell, Albert Einstein College of Medicine Lawrence I. Mortin. Brandeis University Naomi Nagaya, University of Southern California Clifford A. Opdyke. Emory University Jennifer L. Raymond, University of Texas Elian Scemes, Universidade De Sao Paulo, Brazil Erin M. Schuman, Princeton University Eric T. Vu, University of California, Los Angeles, School of Medicine Laura R. Wolszon. State University of New York, Buffalo Neurobiology (June 12 to August 20) Director Arthur Karlin, Columbia University Other facility, lecturers, and staff Linda Amos, Cambridge University, England Brian Andrews. National Institutes of Health Katie Armstrong, Rice University Cynthia Czajowski, Columbia University Nicholas Dale, Columbia University Douglas Fambrough, Johns Hopkins University Gerald Fischbach, Washington University School of Medicine Eric Frank, LJniversity of Pittsburgh School of Medicine Robert French, University of Calgary, Alberta, Canada SaraGarber, Stanford University Linda Hall, Albert Einstein College of Medicine Thomas Jessell, Howard Hughes Medical Institute Steve Jones, Case Western Reserve University Educational Programs 19 Robert Kass. University of Rochester School of Medicine Dennis Landis, Case Western Reserve University Story Landis, Case Western Reserve University Rudolfo R. Llinas, New York University Medical Center Craig Malbon, State University of New York, Stony Brook Gail Mandel, Tufts University School of Medicine Steven Matsumoto, University of Arizona College of Medicine Christopher Miller, Brandeis University Priscilla E. M. Purnick, Columbia University Tom Reese, NINCDS, National Institutes of Health Lewis P. Rowland, Columbia University Bert Sakmann, Max-Plank Institute fur Biophysikalische Chemie. FRG Bruce Schnapp, Marine Biological Laboratory Thomas P. Segerson, New England Medical Center Michael Sheetz, Washington University School of Medicine Stephen Siegelbaum, Columbia University Israel Silman, Weizmann Institute of Science. Rehovot, Israel Carolyn Smith, University of Pittsburgh Stefano Vicini. Georgetown University Hsien-Yu Wang, State University of New York, Stony Brook Monte Westerfield, University of Oregon Students Edward M. Blumenthal, Yale University Kerris E. Bright, University of Sussex, England Stephen C. Cannon, Massachusetts General Hospital Dan P. Felsenfeld, Columbia University Maura L. Hamrick, Johns Hopkins University Staffan Johansson, Karolinska Institutet, Nobel Institute, Sweden Chaya Joshi, Mayo Clinic Margaret A. Thompson, Harvard University Katja A. Wehner, Max-Plank Institute, FRG Xian-Chen Yang, State University of New York, Buffalo Rafael M. Yuste, Rockefeller University Dayao Zhou, Harvard Medical School Physiology: Cell and Molecular Biology (June 12 to August 5) Director Robert Goldman, Northwestern University Other faculty, staff, and lecturers Chris Amemiya, Showa Institute George Bloom, Worcester Foundation Kerry Bloom, University of North Carolina Kay O. Broshat, University of Miami David Burgess, University of Miami James Calvet, University of Kansas Christine Collins, Worcester Foundation George Dessev, Northwestern University Yoshio Fukui, Northwestern University Anne Goldman, Northwestern University Elizabeth Goodwin, Brandeis University John Hammarback, Worcester Foundation Margaret Kenna, University of North Carolina Leslie Leinwand, Albert Einstein College of Medicine Gary Litman, Showa Institute Sandra Mayrand, Worcester Foundation Elizabeth McNally, Albert Einstein College of Medicine Robert Obar, Worcester Foundation Mark Paradise, Colorado College Thoru Pederson, Worcester Foundation Andrew Szent-Gyorgyi, Brandeis University Richard Vallee, Worcester Foundation Students Susan Beckwith, Purdue University Amy Boardman, University of Alabama Alice Brown, Vanderbilt University Isabel Cintron, University of Puerto Rico Ian Crossley, University College of London, England Patrick Dunn, University of Pennsylvania David Elliot, Glasgow University, Scotland Michael Fautsch, Mayo Clinic Foundation Rob Fleischer, University of North Dakota Chris Georgiou, University of Iowa Felicia Houser, Harvard University Jennifer Johnston, Dartmouth College Diana Jones, Purdue University Steve Keller, Scripps Institution/University of California. San Diego Andreas Kiener, Harvard Medical School David Klatte, Northwestern University Alison Krufka, University of Wisconsin Denis Larochelle, Stanford University Ellen Lemosy, Duke University Jian Li, State University of New York, Syracuse Guohong Long, University of Massachusetts Salim Mamajiwalla, LIniversity of Miami Lisa Mendoza, Scripps Institution/University of California. San Diego Dino Messina, State University of New York, Syracuse Naomi Morrissette, University of Pennsylvania 20 Annual Report John Oblong, University of Chicago Scott Olsen, University of Minnesota Matthew Rounseville, George Washington University Alice Rushforth, University of Wisconsin Sukanya Suhramanian, Albert Einstein College of Medicine Anthony Vattay, Rutgers University Kevin Vaughan, Cornell University Jin Wang, Worcester Foundation Joseph Wolenski, Rutgers University Connie Wolfe. Scripps Institute/University of California, San Diego Lin Yue, Johns Hopkins University Short Courses Analytical and Quantitative Light Microscopy in Biology, Medicine, and Materials Science (May 12 to 19) Co-directors Edward D. Salmon, University of North Carolina, Chapel Hill D. Lansing Taylor, Cargegie-Mellon University Other Faculty, staff, and lecturers Brad Amos, Medical Research Council, Cambridge, England Gordon Ellis, University of Pennsylvania Fred Fay, University of Massachusetts Medical School Ralph Gonzalez, Perceptics Corporation Shinya Inoue, Marine Biological Laboratory Fred Lanni, Carnegie-Mellon University Katherine Luby-Phelps, Carnegie-Mellon University Michel Nederlof, Carnegie-Mellon University Tim O'Brien, Duke Medical Center Alan Waggoner, Carnegie-Mellon LIniversity Richard Walker, University of North Carolina. Chapel Hill Commercial faculty Richard Aikens, Photometries, Ltd. Dyon Anniballi, Universal Imaging Corporation Michael Bady, Nikon Inc. Richard Baucom, Olympus Corporation Peter Baurschmidt, Carl Zeiss, Inc. Gerald S. Benham, Bio-Rad Laboratories Doug Benson, Inovision Corporation Steve Boyd, Universal Imaging Corporation Mel Brenner, Nikon Inc. Cynthia Brown, Bio-Rad Laboratories Tim Bruchman, Photometries, Ltd. Donald L. Commare, Photonic Microscopy, Inc. Toni Feder, Bio-Rad Laboratories Steve Floyd. Perceptics Corporation Marc Friedman, Olympus Corporation ClifTGlier, Perceptics Corporation Dan Green. Universal Imaging Corporation Scott Henderson, Nikon Inc. Jan Hinsch, Wild Leitz, USA, Inc. Mike Howard, Perceptics Corporation Richard Inman, Inovision Corporation Theodore Inoue, Universal Imaging Corporation Ryozo Ito, Nikon Inc. Ernst Keller, Carl Zeiss, Inc. Jerry Kleifgen, Dage-MTI, Inc. Mark J. Kuno, Wild Leitz USA, Inc. Lloyd London. Bio-Rad Laboratories Seth Miller, Carl Zeiss, Inc. Masafumi Oshiro, Photonic Microscopy, Inc. Dave Patek, Perceptics Corporation Phillip Presley, Carl Zeiss, Inc. Jerry Rubinow, Universal Imaging Corporation Kevin Ryan, Perceptics Corporation Kurt Scheier, Olympus Corporation Stanley Schwartz, Nikon Inc. Dan Terpstra, Perceptics Corporation Thomas Tharp, Carl Zeiss, Inc. Paul Thomas, Dage-MTI, Inc. Brad Whitman, Carl Zeiss, Inc. Robert A. Wick, Photonic Microscopy, Inc. Richard Woods, Perceptics Corporation Students Karl E. Adler, E. I. DuPont De Nemours & Company Tobias Baskin, University of California, Berkeley Tamie J. Chilcote, The Rockefeller University Jeffrey T. Corwin, University of Hawaii Christopher Cullander, University of California, San Francisco Phillip N. Dean, Lawrence Livermore National Laboratory Jan R. De May, European Molecular Biology Laboratory, FRG Josef Eisinger, Mount Sinai School of Medicine Donald A. Fischman, Cornell University Paul M. Horowitz, University of Texas Health Science Center Bruce D. Jensen, Smith, Kline, and French Laboratories James P. Kelly, Columbia University Hallie M. Krider. LIniversity of Connecticut Stephen J. Kron, Stanford University Greta M. Lee, Duke University Philip M. Lintilhac, University of Vermont Educational Programs 21 Nancy A. O'Rourke, University of California, Irvine Lionel I. Rehbun, University of Virginia Thomas F. Robinson, Albert Einstein College of Medicine Frederick Sachs, State University of New York, Buffalo Edwin H. Smail, University Hospital, Boston Barbara J. Varnum-Finney, Howard Hughes Medical Institute, UCSF Cell and Molecular Biology of Plants (July 31 to August 19) Co-directors Leon S. Dure, University of Georgia Joe L. Key, University of Georgia Other faculty, lecturers, and staff' Frederick M. Ausubel, Massachusetts General Hospital Anthony Cashmore, University of Pennsylvania Gloria Coruzzi, The Rockefeller LIniversity Martha L. Crouch, Indiana University Alan Darvill, USDA Russell Lab Robert Fed, University of Florida Robert T. Fraley, Monsanto Company Wilhelm Gruissem, University of California, Berkeley Tom Guilfoyle, University of Missouri Richard B. Hallick, University of Arizona Robert Haselkorn, University of Chicago R. B. Horsch, Monsanto Company Gary Kochert, University of Georgia C. S. Levings III, North Carolina State University Elliot Meyerowitz, California Institute of Technology Peter H. Quail. USDA Plant Gene Expression Center Clarence A. Ryan, Washington State University Carolyn D. Silflow, University of Minnesota William Timberlake, University of Georgia Hans D. Van Etten, Cornell University Susan R. Wessler, University of Georgia Students Benita A. Brink, Marquette LIniversity Laura G. Catignani. Hyannis, MA Eunice A. Cronin, Belmont Abbey College S. P. Dinesh Kumar. University of Kentucky Sue H. Kadwell, CIBA-GEIGY Biotechnology Jingqui Li, Marquette University William J. Mathews, City University of New York Lorraine Mineo, Lafayette College Rosevelt L. Pardv, University of Nebraska Yves Poirier, Clinical Research Institute, Canada Daniel J. Prochska, Miami University James Ramsay, University of Utah Teresa Snyder, Pennsylvania State University Frederica Sponga, Enichem Americas William L. Steinhart, Bowdoin College Salma N. Talhouk, Ohio State University Stephen J. Tarapchak, Great Lakes Environmental Research Laboratory Joanne A. West, ZOECON Research Institute Mariculture: Culture of Marine Invertebrates for Research Purposes (May 22 to 28) Director Carl J. Berg. State of Florida, Bureau of Marine Research Faculty Philip Alatalo, Woods Hole Oceanographic Institution Thomas Capo, Howard Hughes Medical Institute Jeffrey Fisher, Eastern Connecticut State University David Egloff, Oberlin College Patrick M. Gaffney, University of Delaware Scott M. Gallager, Woods Hole Oceanographic Institution Robert Guillard, Bigelow Laboratories of Ocean Sciences Roger Hanlon, University of Texas Louis Leibovitz. Marine Biological Laboratory James P. McVey. NOAA Katherine Orr, Marathon, FL Neal Overstrom, Sea Research Foundation, Mystic Marinelife Aquarium Philip H. Presley, Carl Zeiss, Inc. David Turner, Eastern Connecticut State University Students Ervin J. Bass, National Institutes of Health Joseph S. Couturier, Enviro Systems Inc. Christopher V. Davis, University of Maine Robert I. Davidson, Trio Laboratories Tien-Lai Hsu, University of Rhode Island David Kohan, Hebrew University of Jerusalem. Israel Suzanne M. Lussier, EPA Environmental Research Laboratory Melanie Meade, Long Island University Gloria D. Royall, George Mason University Rafael Sarda, Boston University Marine Program Jan Taschner. Yale University Edward Wade, Aquatic Research Organisms Roger G. Zirk. CFM Environmental Services Inc. 22 Annual Report Methods in Computational Neuroscience (August 14 to September 3) Directors James M. Bower, California Institute of Technology Christof Koch, California Institute of Technology Faculty Paul Adams. State University of New York, Stony Brook Thomas Albright, The Salk Institute Daniel Alkon, NINCDS, National Institutes of Health Thomas Anastasio, University of Southern California Richard Andersen, Massachusetts Institute of Technology John Hildebrand, University of Arizona John Hopfield, California Institute of Technology Stephen Lisberger, University of California, San Francisco, School of Medicine Rudolfo Llinas, New York University Medical Center Nicos Logothetis, Massachusetts Institute of Technology Eve Marder, Brandeis University Mark Nelson, California Institute of Technology Alexander Pentland, Massachusetts Institute of Technology V. S. Ramachandran, University of California, San Diego John Rinzel, National Institutes of Health Idan Segev, Hebrew University of Jerusalem, Israel Terrence Sejnowski, Johns Hopkins University Roger Traub. IBM Watson Research Center David Van Essen, California Institute of Technology Course assistants Upinder S. Bhalla, California Institute of Technology John Uhley, California Institute of Technology Matthew Wilson, California Institute of Technology Students Werner Backhous, Freie Universitat Berlin, FRG Jim Cummings, University of Pennsylvania Allen C. Dobbins. McGill University, Canada William N. Frost, University of Iowa Edward W. Kairiss, Yale University Alexander Kirillov, USSR Academy of Sciences, Moscow, USSR Guy Major, University of Oxford, England Yair Manor, Hebrew University, Israel Hiroyoshi Miyakawa, New York Medical College Barbara Moore, Massachusetts Institute of Technology Laura J. Reece, University of Washington Clay Reid. The Rockefeller University John R. Rose. State University of New York, Stony Brook Franklin H. Schuling. University of Groningen. The Netherlands Jeffrey E. Segall, Max Planck Institute for Biochemistry, FRG Brian H. Smith, University of Arizona William Skaggs, University of Colorado Xiao-Jing Wang, University of California. Berkeley Donald J. Weir, Oxford University, England Hagit Zabrodsky. Hebrew University, Israel Molecular and Cellular Immunology (July 31 to August 19) Director Darcy B. Wilson, Medical Biology Institute Other faculty, lecturers, and staff Steven Abramson, New York University Medical Center Jenny Blackwell, London School of Hygiene and Tropical Medicine. England Peter Brodeur, Tufts LIniversity School of Medicine Charles Dinarello, Tufts New England Medical Center Ann J. Feeney, Medical Biology Institute Steve Hedrick, University of California, San Diego Nancy Hogg. Imperial Cancer Research Fund Richard Larson, Dana Farber Cancer Institute Sidney Leskowitz, Tufts University Medical School Donald Mosier, Medical Biology Institute Lee Nadler, Dana Farber Cancer Institute David Parker, University of Massachusetts Medical School Jonathan Sprent, Scripps Clinic & Research Foundation Jack Strominger, Harvard Lfniversity Geoffrey Sunshine, Tufts University School of Veterinary Medicine Fred T. Valentine. New York University Medical Center Gerald Weissmann, New York University Medical Center Robert Winchester. New York University Medical Center Students Gina L. Adel, University of Iowa Hospital Anthony L. Back, University of Washington Educational Programs 23 Kenneth R. Brown, Merck, Sharp, & Dohme Nino M. Dobrovic, Stevens Institute of Technology Verna C. Gibbs, University of California, San Francisco Patricia A. Grady, University of Maryland Wendy L. Niebling, Northwestern University Maria J. O'Shea, Mount Holyoke College Cynthia A. Pise, University of Massachusetts Medical School Robert C. Rickert, University of North Carolina Elizabeth M. Southard, University of Texas N. Kanaga Sundaram, New Jersey Medical School Susan B. Sylvers, University of Massachusetts, Amherst Charles W. Taylor, University of Arizona Guillermo Torre. University of Chicago Samuel P. Wertheimer, New York University Optical Microscopy and Imaging in the Biomedical Sciences (March 6 to 11) Co-directors Nina Stromgren Allen, Wake Forest University Colin S. Izzard, State University of New York. Albany Faculty, lecturers, and staff Shinya Inoue, Marine Biological Laboratory Kenneth A. Jacobson, University of North Carolina Kenneth OrndorrT, Dartmouth College Greenfield Sluder, Worcester Foundation for Experimental Biology Stephen J. Smith, Yale University Medical School Kenneth R. Spring, National Institutes of Health, National Heart, Lung, and Blood Institute Course assistant Joseph De Pasquale, State University of New York, Albany Student faculty Bruce Faison Holifield, University of North Carolina Commercial faculty • Rob Ashmead, Nikon Inc. Michael Bady, Nikon Inc. Gerald S. Benham, Bio-Rad Laboratories Donald L. Commare, Photonic Microscopy, Inc. Hermann J. Esser, Videoscope International, Ltd. Charles Fanghella, Nikon Inc. Barbara Foster, Reichert-Jung David Hillman, ADCO Aerospace, Inc. Theodore Inoue. Universal Imaging Corporation Ernst Keller, Carl Zeiss, Inc. Jerry Kleifgen, Dage-MTI, Inc. Rob Klueppel, Polaroid Corporation Mark J. Kuno, Wild Leitz USA, Inc. Jeffrey Larson, Nikon Inc. Instrument Group Eric Marino. Image Processing Solutions Hugh Mellaly, Eastman Kodak Company Patrick Moore. Molecular Probes Raj Mundag, Bio-Rad Laboratories Phillip Presley, Carl Zeiss, Inc. Martin L. Scott, Eastman Kodak Company Thomas Tharp, Carl Zeiss, Inc. Paul Thomas. Dage-MTI, Inc. Robert A. Wick, Photonic Microscopy. Inc. S indents Barbara C. Boyer, Union College Kathryn I. Casteel, University of North Carolina, Chapel Hill Linda N. Curtis. University of North Carolina. Greensboro Sarah C. Elgin, Washington LIniversity Joseph Farley. Princeton University Clare Fewtrell, Cornell University Lorraine A. Fitzpatrick, University of Texas Health Science Center, San Antonio Lawrence C. Katz, The Rockefeller University Nicholas F. LaRusso, Mayo Clinic James M. Mcllvain, State University of New York, Syracuse P. Scott Pine, Uniformed Services University of Health Sciences Martin Reers, Harvard Medical School Joseph F. Rizzo, Massachusetts Eye and Ear Infirmary Allyson A. Simons, Federal Bureau of Investigation Laboratory Mark Terasaki, Marine Biological Laboratory Daniel Ts'o, The Rockefeller University Michael Tytell, Wake Forest University Hsien-yu Wang, State University of New York at Stony Brook Patrick Weyer, Massachusetts General Hospital R. Reid Zeigler, Merck Sharp & Dohme Research Labs Workshop on Molecular Evolution (September 18 to 30) Director Mitchell L. Sogin, National Jewish Center for Immunology and Respiratory Medicine 24 Annual Report Other faculty, lecturers, ami staff Robert Cedegren, University of Montreal, Canada Michael Clegg. University of California, Riverside John W. Drake, National Institute of Environmental Health Sciences Daniel E. Dykhuuzen, State University of New York, Stony Brook Joseph Felsenstein, University of Washington Katherine G. Field, Oregon State University Walter B. Goad, Los Alamos National Laboratory Morris Goodman, Wayne State University School of Medicine Barry G. Hall, University of Connecticut John Lawrie, GENE-TRAK Systems Lynn Margulis, University of Massachusetts, Amherst Roger Milkman, University of Iowa Gary Olsen, University of Illinois Norman R. Pace. Indiana University Philip J. Regal. University of Minnesota David Shub, State University of New York, Albany Maxine F. Singer, Carnegie Institution of Washington Michael Syvanen, University of California, Davis Bruce Walsh, University of Arizona Michael S. Waterman, University of Southern California Mark Wheelis, University of California, Davis Carl R. Woese, University of Illinois Students Rogelio Alonso-Morales, Harvard School of Public Health James W. Ammerman, Lamont-Doherty Geological Observatory of Columbia University Ann Antlfinger, University of Nebraska Nathalie Antoine, University of Montreal, Canada Edgardo Ariztia, National Jewish Center for Immunology and Respiratory Medicine Wendy J. Bailey. Wayne State University William V. Baird, University of Georgia David Begun, Arizona State University Michael J. Braun, University of Cincinnati Colleen M. Cavanaugh, Harvard University Michael P. Cummings, Harvard LJniversity Lloyd Demetrius, Max-Plank Institute Paul Desjardins, Universitede Montreal, Canada Daniel L. Distel, Woods Hole Oceanographic Institution Hille Elwood, National Jewish Center for Immunology and Respiratory Medicine David H. Fitch, Wayne State University Jane C. Gallagher, City College of New York Mark A. Gallo, Cornell University Johann P. Gogarten, University of California, Santa Cruz Spencer J. Greenwood, University of Guelph, Canada Sidney Grimes, VA Medical Center, Shreveport, Louisiana Thaddeus A. Grudzien Jr.. Oakland University John H. Gunderson. Vanderbilt University Winston Hide, Temple University Llewellya Hillis-Colinvaux. Ohio State University Holly H. Hobart, University of Arizona Volker Huss, National Jewish Center for Immunology and Respiratory Medicine Piroska Huvos, Southern Illinois University Thomas S. Kantz, Louisiana State University Christopher J. Kelly, University of Chicago Cordula Kirchgessner. University of Massachusetts Medical Center Robert Kraft, Albert Einstein College of Medicine Antonio A. Lazcano, National University of Mexico, Mexico Roger D. Longley, Pacific Sciences Institute Denis H. Lynn, University of Guelph, Canada George M. McCorkle, Yale University Tammy S. McCormick, University of Cincinnati Mara A. McDonald, University of Cincinnati J. Mitchell McGrath, University of California at Davis Nancy Moncrief. Wayne State University Paul A. Nelson. University of Chicago Gokaldas C. Parikh. Quinnipiac College Aloysius Philips, Temple University Douglas Prasher, Woods Hole Oceanographic Institution A. Rajaguru, University of Maryland Thomas Redlinger, University of Massachusetts at Amherst Rebeca Rico-Hesse, Yale University Margaret Riley, Harvard University Jan E. Rines, University of Rhode Island Michael S. Roberts, Wesleyan LIniversity Lori Sadler, University of California, Los Angeles Patricia Sawaya. University of Cincinnati Martin Schlegel, Universitat Tubingen, FRG Susan Sell, Hopkins Marine Station M. Andrew Shenk, Yale LIniversity George N. Sideris, New York University Scott Smiley. University of California, San Francisco Joseph L. Staton, University of Southwestern Louisiana Robert E. Steele, LJniversity of California, Irvine Tamalyn Stockton, University of Massachusetts, Amherst Educational Programs 25 Diane K. Stoecker, Woods Hole Oceanographic Institution John F. Stolz, University of Massachusetts, Amherst Danilo Tagle, Wayne State University John B. Waterbury, Woods Hole Oceanographic Institution Steve A. Wolfe, Louisiana State University Hong Y. Yan, University of Maryland Clarice M. Yentsch, Bigelow Lab for Ocean Sciences Workshop on Plant and Animal Cell Microinjection Techniques (May 20 to 22) Co-directors Robert B. Silver, University of Wisconsin Edward B. Tucker, City University of New York Other faculty, lecturers, and staff' Jan Blaas, Research Institute Ital. The Netherlands Shinya Inoue, Marine Biological Laboratory Lionel Jaffe, Marine Biological Laboratory William Jeffrey, University of Texas Douglas Kline, University of Connecticut Health Center W. Langridge, Cornell University Katherine Luby-Phelps, Carnegie Mellon University P. McNeil, Harvard Medical School James Mailer, University of Colorado W. Muller, Harvard Medical School P. M. Pechan, Plant Research Center, Canada Martin Poenie, Universitv of Texas D. Lansing Taylor, Carnegie Mellon University B. R. Terry, The Flinders University of South Australia R. Tsien, University of California, Berkeley Commercial faculty Charles Fanghella, Nikon Inc., Instrument Group Barbara Foster, Cambridge Instruments Daniel Green, Universal Imaging Corporation Stuart F. Havel, Medical Systems Corporation Theodore O. Inoue, Universal Imaging Corporation Jerry Kleifgen, Dage-MTI, Inc. Hans Koczinski. Carl Zeiss, Inc. Mark John Kuno, Wild Leitz USA, Inc. Victor P. Laronga, Atlantex & Zeiler Instrument Corporation Charles McDonough. Cambridge Instruments Seth Miller, Carl Zeiss, Inc. Victor Olszewski, David Kopf Instruments Phillip Presley, Carl Zeiss. Inc. J. Kemp Randolph, Medical Systems Corporation William Reid, Technical Manufacturing Corporation Jim Schindele, Carl Zeiss, Inc. Rick Smith, Eppendorf Paul Thomas, Dage-MTI, Inc. Melina Vratny, Nikon Inc., Instrument Group Students (lectures, demonstrations and laboratory) Lee-Ann Allen, University of Wisconsin John Andersland, Cornell University Thomas Bjorkman, University of Washington Mingxin Che, Fordham University John R. Coleman, Brown University Frank J. Dye, Western Connecticut State University Miles F. Epstein, University of Wisconsin Edward A. Fisher, Medical College of Pennsylvania Paul E. Gallant, National Institutes of Health Stella M. Hurtley, Yale University George M. Langford, University of North Carolina Willy Lin, E. I. DuPont Company Andrew Maretzki, Hawaiian Sugar Planters Association Paul J. Millard, Cornell University Corey Nislow, Dartmouth College Jo Ann Render, Hamilton College Jen Sheen, Massachusetts General Hospital Jonathan M. Shenker, University of California Cuthbert O. Simpkins, Howard University Robert Turgeon, Cornell University J. Richard Whittaker, Marine Biological Laboratory Wayne Yunghans, State University of New York, Fredonia 26 Annual Report Students (lectures and demonstrations) Rosario Agnosti, University of Zurich, Switzerland Hassan Amjad, Jafary Medical Clinic Winston A. Anderson, Howard University Elena Armandola, New York Medical College Ann C. Burke, Harvard University Andrew R. Burns, Cornell University Robert M. Coleman, University of Lowell Colleen Curran, University of Minnesota Carol Dorworth, Louisiana State University Julie Downey, City University of New York Richard A. Fluck, Franklin and Marshall College Donald J. Fujita, University of Calgary, Canada Irene Garcia, University of Geneva Edwin H. Gilland. Harvard University Maria A. Giovino, University of Lowell William Gordon-Kamm, DeKalb Pfizer Genetics Thomas C. Gore, Salsbury Laboratories Inc. Helana Hoover-Litty, Yale University Subhash C. Juneja, Old Dominion University P. G. Kadkade, Biotechnica International Karen Kurvink, Moravian College Cheryl Laursen, Molecular Genetics, Inc. Catharina S. Lee, E. I. DuPont Company Kersti K. Linask, Thomas Jefferson Medical College MingS. Lok, Denver General Hospital Rosalind Lowen, City University of New York Burra V. Madhukar, Michigan State University David E. McClain, Armed Forces Radiobiology Research Institute Raman Mocharla, Indiana University Suresh Savarirayan, Mayo Clinic Lewis Tilney, University of Pennsylvania Nemat Ullah, Washington State University Gabriele Weitz. National Institutes of Health Yang Dar Yuan, The Upjohn Company Summer Research Programs Principal Investigators Adelman, Jr., William J., NINCDS, NIH Alkon, Daniel L., NINCDS/NIH Anderson, Peter A. V., University of Florida Armstrong, Clay M., University of Pennsylvania Medical School Armstrong, Peter B., University of California Augustine, George J., University of Southern California Baker, Robert, New York University Barlow, Robert B. Jr., Syracuse University Barry, Susan R., University of Michigan Bass, Andrew H., Cornell University Beauge, Luis A., Institute M. y M. Ferreyra, Argentina Begenisich, Ted, University of Rochester School of Medicine and Dentistry Bennett, Michael V. L., Albert Einstein College of Medicine Bezanilla, Francisco, University of California Bloom, George S., University of Texas Southwestern Medical Center Bodznick, David, Wesleyan University Borgese, Thomas A., Lehman College Boron, Walter F., Yale University School of Medicine Borst, David W., Illinois State University Boyer, Barbara C., Union College Brady, Scott T., University of Texas Southwestern Medical Center at Dallas Brehm. Paul, Tufts University School of Medicine Brenner, Sydney, Molecular Genetics Unit, Cambridge, England Brown, Joel E., Washington University School of Medicine Burdick, Carolyn J.. Brooklyn College Burger, Max M., University of Basel, Switzerland Chang, Donald C., Baylor College of Medicine Chappell, Richard L., Hunter College of the City University of New York Charlton, Milton P., LIniversity of Toronto, Canada Clay, John R., NINCDS/NIH Cohen, William D., Hunter College Cohen, Avis H., Cornell University Cohen, Lawrence B., Yale University School of Medicine Cooperstein, Sherwin J., University of Connecticut Health Center Cresti, Mauro, University of Siena, Italy Dowling, John E., Harvard University Dunlap, Kathleen, Tufts Medical School Dykens, James A., Grinnell College Ehrlich, Barbara E., University of Connecticut Health Center Feinman, Richard B., State University of New York Health Science Center at Brooklyn Fine, Alan, Dalhousie University, Canada Fink, Rachel, Mount Holyoke College Fishman, Harvey M., University of Texas Medical Branch, Galveston Gadsby, David C., Rockefeller University Gainer, Harold, NIH, NINCDS Gaze, Raymond M., University of Edinburgh, U.K. Giuditta, Antonio, University of Naples, Italy Glynn, Paul, AT&T Bell Laboratories Gonzalez-Serratos, Hugo, University of Maryland School of Medicine Gould, Robert M., Institute for Basic Research in Developmental Disabilities Govind, C. K., University of Toronto, Canada Graf, Werner, Rockefeller University Green, Douglas R., LIniversity of Alberta, Canada Haimo, Leah T., University of California, Riverside Hepler, Peter K., University of Massachusetts, Amherst Highstein, Stephen M., Washington University School of Medicine 27 28 Annual Report Hill, Susan D., Michigan State University Hoskin. Francis C. G., Illinois Institute of Technology Hoy, Ronald R.. Cornell University Jeffery, William R., University of Texas, Austin Josephson, Robert K., University of California Kaminer. Benjamin, Boston University School of Medicine Kao, C. Y., State University of New York Downstate Medical Center Kaplan, Barry B.. University of Pittsburgh Western Psychiatric Institute and Clinic Katz. Paul S., Cornell University Kirk, Mark, Boston University Kreibel, Mahlon E., State University of New York Health Science Center Landowne, David, University of Miami Langford, George M., University of North Carolina School of Medicine Lasek, Raymond J., Case Western Reserve University Laufer, Hans, University of Connecticut Lee, Helen M.. University of Massachusetts at Amherst Levin, Jack, Veterans Administration Medical Center Levis, Richard A., Rush Medical Center Levitan, Irwin, Brandeis University Lin, Stephen S., Brandeis University Linck. Richard, University of Minnesota Lipicky. Raymond J., Food and Drug Administration Lisman. John, Brandeis University Llinas. Rudolfo, New York University Loewenstein, Werner R., University of Miami Matteson, Donald R., University of Maryland McClintock, Timothy S.. The Whitney Laboratory Metuzals, Janis, University of Ottawa, Canada Miller. Christopher, Brandeis University Montgomery, John C., University of Auckland, New Zealand Moore, John W., Duke University Mooseker, Mark S., Yale University Narahashi, Toshio, Northwestern University Nasi, Enrico, Boston University Medical School Nelson, Leonard, Medical College of Ohio Noe, Bryan D., Emory University School of Medicine Nolen, Thomas G., University of Miami Obaid, Ana Lia, University of Pennsylvania School of Medicine Oberhauser-Bittig. Andres. University of Pennsylvania School of Medicine Ohki. Shinpei, State University of New York, Buffalo Pala/zo, Robert E., University of Virginia Pappas, George D., University of Illinois College of Medicine at Chicago Quigley. James, State University of New York Rafferty. Nancy S., Northwestern University Medical and Dental Schools Rakowski, Robert F., Chicago Medical School Rebhun, Lionel I., University of Virginia Regehr, Wade G., California Institute of Technology Rickles, Frederick R., University of Connecticut Health Center Ripps, Harris, University of Illinois College of Medicine at Chicago Rose, Birgit, University of Miami School of Medicine Ross, William N., New York Medical College Ruderman, Joan V.. Duke University Russell, John M., University of Texas Medical Branch Saez. Juan C., Albert Einstein College of Medicine Salzberg, Brian M., University of Pennsylvania School of Medicine Sanger, Jean M.. University of Pennsylvania School of Medicine Sanger, Joseph W., University of Pennsylvania School of Medicine Scofield, Virginia L., University of California, Los Angeles School of Medicine Segal, Sheldon J., Rockefeller Foundation Shumway, Carolyn A.. University of California, San Diego Silver. Robert B.. University of Wisconsin Sloboda, Roger D., Dartmouth College Smith, Stephen J., Yale University School of Medicine Spiegel, Evelyn. Dartmouth College Spiegel, Melvin, Dartmouth College Stadler, Herbert, Max-Planck-Institute of Biophysical Chemistry. FRG Steinacker, Antoinette, Washington University Stracher, Alfred, State University of New York Health Science Center at Brooklyn Stuart, Ann E., University of North Carolina Suprenant, Kathy A., LJniversity of Kansas Tanguy, Joelle, Ecole Normale Supereure, France Telzer, Bruce R., Pomona College Thibault, Lawrence E.. University of Pennsylvania Tilney, Lewis G., University of Pennsylvania Treistman, Steven N., Worcester Foundation for Experimental Biology Trinkaus, John P., Yale University Troll, Walter. New York University Medical Center Tucker, Edward B., Baruch College Tytell, Michael, Bowman Gray School of Medicine Vogel, Steven S., Columbia University Waxman, Stephen, Yale Medical School Wegmann, Thomas G.. University of Alberta, Canada Other Research Personnel 29 Weiss, Dieter G., Technical University Munich, FRG Weissmann, Gerald, New York University Medical Center Wonderlin, William F., University of Calgary, Canada Yeh, Jay Z., Northwestern University Medical School Zigman, Seymour, University of Rochester School of Medicine Zottoli, Steven J., Williams College Zukin, R. Suzanne, Albert Einstein College of Medicine Other Research Personnel Abramson. Charles I., State University of New York Health Science Center at Brooklyn Adler, Elizabeth M., University of Toronto, Canada Adra, Chaker N., University of Ottawa, Canada Alberghinia, Mario, University of Catania, Italy Albert, Daniel, University of Chicago Medical Center Albrecht. Kenneth, University of Connecticut Allen, Nina S., Wake Forest University Altamirano, A. A., University of Texas Medical Branch Anderson, Bruce, Medical College of Virginia Anderson, Matthew, Howard Hughes Medical Institute Badgerow, John P., Eastern Michigan University Bamrungphol, Wattana, University of Pennsylvania Bernal-Martinez, Juan. University of Connecticut Health Center Bleakman. J., NINCDS, NIH Brawley, Jennifer, University of Virginia Breitwieser. Gerda E., Johns Hopkins University Brown, Anthony, Case Western Reserve University Medical School Brozen, Reed, Yale University Buchanan, Jo Ann, Yale University Medical School Butler, Derrick. Morehouse College Callaway, Joseph C., University of Washington Caputo. Carlo, Centre Biofisica y Bioquimica. Venezuela Carbine, Larry, University of Virginia Cariello, Lucio, Stazione Zoologica, Italy Celli, Giulia, Wellesley College Chen, Chong, California Institute of Technology Chen, Fukuen E., Northwestern University Medical School Chun, Xiao, Yale University School of Medicine Church, Paul, Boston University Cohen, Avrum, Yale University Collin, Carlos, NINCDS. NIH Colton, Carol A., Georgetown University Medical School Cooper, Mark. Yale University Medical School Cornwall. M. Carter, Boston University School of Medicine Cottrell, Glen, New York University Medical Center Couch, Ernest F., Texas Christian University Davis. Graeme, Williams College Defendi, Germaine, New York University Medical School Dermietzel, Rolf, Albert Einstein College of Medicine DiPolo. Reinaldo. Institute Venezolano de Investigaciones Cientificas, Venezuela Dome, Jeffrey S.. University of Pennsylvania School of Medicine Dostrovsky, Jonathan O.. University of Toronto, Canada Dovvdall, Michael J., University of Nottingham, U.K. Dworkin, Jonathan, Swarthmore College Etcheberrigaray, Rene, NINCDS/NIH Ferkowicz, Michael, Michigan State University Flores, Roberto. INSERM. Paris. France Floyd, Carl. Morehouse College Forman, Robin. Medical College of Virginia Galbraith. James, University of Pennsylvania Gao, Pei-Qin, Baylor College of Medicine Gee, Christine. University of Toronto, Canada Gerosa, Daniela, Friedrich-Miescher-Institut. FRG Gilbert, Daniel L.. NINCDS/NIH Gilbert, Susan P., Pennsylvania State University Gill-Kumar, Pritam, Food and Drug Administration Graff, Tracy, Syracuse University Grant, Alan J., Worcester Foundation for Experimental Biology Grant, Philip, NIH, NINCDS Grassi, Dan, Fort Lauderdale. Florida 30 Annual Report Graubard, Katharine, University of Washington Greif, Peter. Food and Drug Administration Greiner, Francine, Emory University School of Medicine Haneji, Tatsuji. Population Council Hernandez, M. R., University of Texas Medical Branch Hess, Stephen, University of Southern California Hice, Rita, Whitney Laboratory Hines, Michael. Duke University Hogan, Emilia, Yale University School of Medicine Holbrook, Pam, University of Birmingham, England Homola, Ellen, University of Connecticut Hopp. Hans-Peter, Yale University School of Medicine Hu, S. L., State University of New York Downstate Medical Center Huddie, Patrick, NINCDS, NIH Hunt, John R., Baylor College of Medicine Hunter, Catherine, University of Pennsylvania Hyrc, Krysztof, University of Miami School of Medicine Jarrad, Hugh, Haverford College Johansson, Steffan, Karolinska Institutet, Sweden Jumblatt. James E., University of Louisville School of Medicine Kadam, Arjun, Population Council Kahana. Alon. Brandeis University Kao, Peter N., Columbia University College of Physicians and Surgeons Kaplan. Ehud. Rockefeller LJniversity Klein, Kathryn, Emory University School of Medicine Knudsen, Knud. Food and Drug Administration Koide, Samuel S., Population Council Kojima, Hiroshi, Northwestern LIniversity Kolodnex, Michael, Washington University Medical School Kosik, Kenneth S., Brigham and Women's Hospital Kramer. Richard H., Brandeis University Kronidou. Nafsika. Dartmouth College Lahti, Carol, University of California at Los Angeles School of Medicine Leddy, Scott, University of Alberta Lederhendler, I. Izja, NINCDS, NIH Lehman, Herman K.. Syracuse University Leighton, Stephen B., NINCDS/NIH Li, Guan, Cornell University Lin, Jen-Wei, New York University Medical Center Llinas, Rafael, Washington University Lowe, Kris, University of Rochester Medical Center Lthi, Theres, Biozentrum, Switzerland Luca, Frank, Duke University Marchaterre, Margaret A., Cornell University Marrone, M., University of Toronto, Canada Martin. Melissa Marie, Eastern Illinois University Matzel, Louis D., NINCDS/NIH McDonald, John K., Emory University School of Medicine McKee. Juliet M., University of Kansas Menichini, Enrico, University of Naples, Italy Milgram, Sharon, Emory University School of Medicine Misevic, Gradimir, Friedrich-Miescher-Institut, FRG Montgomery, John, University of Auckland, New Zealand Murray, Sandra, University of Pittsburgh Nemec, Vaclav, University of Connecticut Northern, Sue, Williams College Nuno, Claudia, University of California Oberhauser, Andres, University of Pennsylvania Olds, James, NINCDS/NIH Pant, Harish, NINCDS/NIH Parsons, Thomas D., University of Pennsylvania School of Medicine Pearce, Joanne, University of Toronto, Canada Perozo, Eduardo, University of California, Los Angeles Poole, Thomas J., SUNY Health Science Center, Syracuse Queck, Francis, University of Michigan Rafferty, Keen A. Jr., University of Illinois Medical Center Rasgado-Flores, Hector, University of Maryland School of Medicine Raslavicus, Alex, Union College Regehr, Wade G., California Institute of Technology Robinson, Phyllis R., Brandeis University Romero, Adarli, Washington LIniversity School of Medicine Sanchez, Ivelisse, Hunter College Sands, Peter J., New York University School of Medicine Schiminovich, David. Yale University Schneider, Melissa R., University of Rochester Medical School Shibuya, Ellen. Duke University Silbaugh, Timothy H., Syracuse University Silverstein, Kurt, Baruch College Spires, Sherrill, University of Rochester Medical Center Spray, David C, Albert Einstein College of Medicine Standart, Nancy, University of Cambridge, U.K. Steffen, Walter, University of Minnesota Stein, Philip G., University of Texas Medical Branch, Galveston Library Readers 31 Stokes, Darrell, Emory University Sugimori, Mutsuyuki, New York University Medical Center Svoboda, Katherine, University of California, Los Angeles, School of Medicine Swalla. Billie J., University of Texas, Austin Sydlik, Mary Anne, Eastern Michigan University Tabares, Lucia, University of Pennsylvania Takashima, Shiro, University of Pennsylvania Tewari, Kirti P., I. G. Medical College, Simla, India Thompson, Jennifer, Lehman College Tricas, Timothy, Washington University Tyndale, Clyde, NINCDS/NIH Uchiyama, Hiroyuki, Syracuse University Ueno, Hiroshi, Rockefeller University Vandenberg, Carol A., University of California. Los Angeles Vautrin, Jean, State University of New York Health Science Center Villanova, Lynnea, University of Miami Vogel. Jacalyn M., Illinois State University Wu, Jian-Young, Yale University School of Medicine Yamoah, Ebenezer, University of Alberta, Canada Yao, Jibin, Georgetown University Medical School Young, Steven R., New York Medical College Yuan, Liu, University of Basel, Switzerland Zakevicius, Jane, University of Illinois College of Medicine at Chicago Zavilowitz, Joseph, Albert Einstein College of Medicine Zecevic. Dejan, Institute for Biological Research, Yugoslavia Zigman, Bunnie R., University of Rochester Medical Center Library Readers Allen, Garland, Washington University Anderson, Everett, Harvard Medical School Apter, Nathaniel, Nova University Baccetti, Baccio, Institute of Zoology, Naples. Italy Bang, Betsy, Marine Biological Laboratory Barrett, Dennis. University of Denver Benjamin. Thomas, Harvard Medical School Bernheimer, A. W., New York, NY Boyer, John F., Union College Buck. John, National Institutes of Health Burr, A. H., Simon Eraser University Candelas, Graciela C., University of Puerto Rico Cape Cod Planning and Economic Development Carriere, Rita, Downstate Medical Center Child, Frank, Trinity College Chinard, Francis P., New Jersey Medical School Clark, Arnold, Marine Biological Laboratory Clark, John, University of Massachusetts Cobb. Jewel Plumm. California State University, Fullerton Cohen, Leonard A.. American Health Foundation Cohen, Seymour S., Marine Biological Laboratory Constantine, Betsy J., Arthur D. Little, Inc. Corliss, Bruce H., Duke University DeToledo-Morrell, Leyla, Rush Presbyterian, St. Lukes Medical Center Eder. Howard A.. Albert Einstein College of Medicine Ellner, Jerrold, Case Western Reserve University Farb, David, SUNY Farmanfarmian, A., Rutgers University Finkelstein, Joan Kent, The Rockefeller University Fisher, Saul H., NYU Medical Center Frenkel, Krystyna, NYU Medical Center Friedler, Gladys. Boston University School of Medicine Fussell. Catharine P., Pennsylvania State University Galatzer-Levy, Robert, University of Chicago German, James L., The New York Blood Center Goldfarb. Ronald H., Pittsburgh Cancer Inst. \nnualReport Goldstein, Jr., Moise, John Hopkins University Goodgall, S. H., University of Pennsylvania Gordon, Erlinda. Case Western University Grossman, Albert. NYU Medical Center Gruner, John. NYU Medical Center Guiseppe. D'Allesio, University of Naples, Italy Guillemin-Meselson. Jeanne, Boston College Guttenplan. Joseph B.. NYU Dental Center Harding, Clifford V., Wayne State University Harrington, John P.. University of Alaska Hatten, Mary, Columbia University Herskovits, Theodore T., Fordham University Hill, Richard W., Michigan State University Humphreys, Tom, University of Hawaii Ilan, Joseph, Case Western Reserve University Ilan, Judith, Case Western Reserve University Inoue, Sadayuki, McGill University Issidorides, Marietta, Athens University Medical School Johnson, William, Goucher College Johnston, Ardis, Harvard University Kaltenbach, Jane C, Mount Holyoke College Kaplan, Ilene M., Union College Kass-Simon, Gabriele. University of Rhode Island Kelly, Robert, University of Chicago, College of Medicine Kemlow, Kenneth M., Wilkes College King, Kenneth, Jr., Children's Hospital Kirk. Mark D., Boston University Krane. Stephen M., Massachusetts General Hospital Kravitz, Edward A., Harvard Medical School Laderman, Aimlee, Marine Biological Laboratory Lau, Dian, University of Pennsylvania Lazarow, Normand H., Rochester, MN Lee, John, City College of CUNY Leighton, Joseph, Medical College of Pennsylvania Levine, Rachmiel, City of Hope Medical Center Levitz, Mortimer, NYU Medical Center Lewis, Larry, Millersville University Lorand, Laszlo, Northwestern University Lustig, Robert H., The Rockefeller University Marine Research, Inc. Matsumura, Fumio, University of California Mautner, Henry G., Tufts University School of Medicine Mauzerall, David, The Rockefeller University McCann-Collier, Marjorie, Saint Peter's College Meyer, Kenneth, R. E. Consulting Inc. Mercurio, Arthur, Harvard Medical School Miki-Noumura, Taiko, Ochanomizu University Mitchell, Ralph, Harvard University Mizell, Merle, Tulane University Morrell, Frank, Rush Presbyterian, St. Lukes Medical Center Moyer, Carolyn F., EG&G Mason Research Institute Musacchia, X. J., University of Louisville Nagel, Ronald L., Albert Einstein College of Medicine Nicaise, Mari-Luz H., University of Nice. France Nichol, Charles A.. Glaxo Research Laboratories Nickerson, Peter A.. State University of New York, Buffalo Nowotny, Alois H., University of Pennsylvania Olins, Ada, University of Tennessee, Oak Ridge Olins, Donald E., University of Tennessee, Oak Ridge Olufemi, Ogunnowo, Marine Biological Laboratory Palmer, Claude Irene, New York University Parsons, Katharine C., Manomet Bird Observatory Paton, David, Marine Biological Laboratory Person, Philip, VA Medical Center, Brooklyn, NY Prusch. Robert D., Gonzaga University Rabinowitz, Michael, Marine Biological Laboratory Reynolds, George T.. Princeton LIniversity Robinson. Denis, Marine Biological Laboratory Rome, Larry, Marine Biological Laboratory Rosenbluth, Raja, Simon Eraser University Roth, Jay S.. University of Connecticut Russell-Hunter. W. D., Syracuse University Salman, Edward D., LIniversity of North Carolina Schippers, Jay, Marine Biological Laboratory Schuel, Herbert, State University of New York, Buffalo Schuel, Regina, State University of New York, Buffalo Sheppard, Frank. Woods Hole Data Base Shepro, David, Boston University Sherman, Irwin W., University of California, Riverside Shriftman, Mollie Starr. North Nassau Mental Health Center Simon, Keiko O., University of Pittsburgh Sluder, Greenfield, Worcester Foundation for Experimental Biology Sonnenblick. Benjamin P., Rutgers University Southeastern Massachusetts University Library Speck, William T., Case Western Reserve University Spector, Abraham, Columbia University Stevens, E. Don, University of Guelph, U.K.. Tashiro, Jay Shiro, Bard College Taylor, Rachael, Harvard University Tilney, Lewis, University of Pennsylvania Trager, William, The Rockefeller University Tweedell, Kenyon S., University of Notre Dame Van Holde, Kensal E., Oregon State University Domestic Institutions Represented 33 Wagenbach, Gary, Carleton College Wagner, Robert R., University of Virginia Warren, Leonard, Wistar Institute Webb, H. Marguerite, Marine Biological Laboratory Weidner, Earl H., Louisiana State University Weiner, Jonathan, Doylestown, PA Weiss, Leon, University of Pennsylvania Wheeler, George, Brooklyn College Whittaker, J. Richard, Marine Biological Laboratory Wichterman, Ralph. Marine Biological Laboratory Wilber, Charles G., Colorado State University Domestic Institutions Represented Wittenberg, Beatrice, Albert Einstein College of Medicine Wittenberg, Jonathan, Albert Einstein College of Medicine Wolken, Jerome J., Carnegie Mellon University Worth, Dyan, Harvard University Yow, F. W., Kenyon College Zigmond. Richard E., Harvard Medical School Zimmerman, Morris, Zimmerman Associates Academy of Natural Sciences ADCO Aerospace. Inc. Alabama. University of Albert Einstein College of Medicine of Yeshiva University Aquatic Research Organisms Arizona State University Arizona, University of Arizona, University of. College of Medicine Armed Forces Radiobiology Research Institute Atlantex & Zeiler Instrument Corp. AT&T Bell Laboratories Axon Instruments. Inc. Baruch College Belmont Abbey College Bethesda Research Laboratories Bigelow Laboratories for Ocean Sciences Bio-Rad Laboratories Biotechnica International Boston University Boston University Marine Program Boston University School of Medicine Bowdoin College Bowman Gray School of Medicine Brandeis University Brigham & Women's Hospital Brooklyn College Brown University Bryn Mawr College California Institute of Technology California Polytechnical California, University of. School of Medicine, San Francisco California, University of. San Francisco California, University of, Berkeley California, University of, Davis California, University of, Irvine California, University of. Los Angeles California, University of, Los Angeles, School of Medicine California, University of. Riverside California, University of San Diego California, University of, San Francisco California, University of, Santa Cruz Cambridge Instrument, Inc. Carnegie Institution of Washington Carnegie-Mellon University Case Western Reserve University Case Western Reserve University School of Medicine CFM Environmental Services. Inc. Chesapeake Biological Laboratories Chicago, University of Chicago. University of. School of Medicine Ciba-Geigy Biotechnology Corp. Cincinnati, University of. School of Medicine Colorado, College Colorado State University Colorado, University of Columbia University Columbia University College of Physicians and Surgeons Connecticut, University of Connecticut, University of. Health Center Cornell University Cornell University Medical College Dage-MTI. Inc. Dana Farber Cancer Institute Dartmouth College Dartmouth College Medical School David Kopf Instruments Dekalb Pfizer Genetics Delaware, University of Denver General Hospital DNAX Corporation DuPont. E. I. DeNemours & Co. Duke University Duke University Medical Center E-C Apparatus Corporation E.G.&G Enviro Systems, Inc. Earlham College East Carolina University Eastern Connecticut State University Eastern Illinois University Eastern Michigan University Eastman Kodak Company Emory University Emory University School of Medicine Environmental Protection Agency Research Laboratory Enichem Americas Eppendorf Federal Bureau of Investigation Laboratory Ferrum College Florida. University of Florida State University Food and Drug Administration Fordham University Forsyth Dental Institute GENE-TRAK Systems George Mason University George Washington University Georgetown University Medical School Georgetown University Georgia State University Georgia, University of Great Lakes Environmental Research Lab Gnnnell College Hahnemann Medical College & Hospital Hamilton College Harbor Branch Oceanographic Institution Harvard Graduate School of Arts and Sciences Harvard Medical School Harvard School of Public Health Harvard University Harvard University School of Medicine 34 Annual Report Haverford College Hawaii, University of Hawaiian Sugar Planters Association Hopkins Marine Station Horn Point Environmental Laboratories Howard Hughes Medical Institute Howard University IBM. Watson Research Center Illinois Institute of Technology Illinois State University Illinois, University of Illinois. University of. College of Medicine Illinois. University of. Medical Center Image Processing Solutions Indiana University Inovision Corporation Institute for Basic Research in Developmental Disabilities Institute for Ecosystems Studies. New York Iowa. University of Iowa, University of. Hospital JaFary Medical Clinic Johns Hopkins University Johns Hopkins University School of Medicine Johns Hopkins University School of Hygiene & Public Health Kansas. University of Kentucky. University of LaFayette College Lawrence Livermore National Laboratories Long Island LIniversity Los Alamos National Laboratory Louisville University of. School of Medicine Louisiana State University Lowell, University of Maine, University of Marquette University Marine Biological Laboratory' Maryland. University of Maryland. University of. School of Medicine Maryland, University of, Chesapeake Biological Laboratory Massachusetts Eye and Ear Infirmary Massachusetts General Hospital Massachusetts Institute of Technology Massachusetts, LIniversity of, Amherst Massachusetts, University of, Boston Mayo Clinic, Rochester. MN Mayo Graduate School of Medicine Medical Biology Institute Medical College of Ohio Medical College of Pennsylvania Medical College of Virginia Medical Systems Corporation Merck, Sharp & Dohme Research Laboratory Miami, University of Miami University of Ohio Miami. LIniversity of. School of Medicine Michigan State University Michigan Tech Michigan. LJniversity of Minnesota. University of Molecular Genetics, Inc. Molecular Probes, Inc. Moravian College Morehouse College Mount Sinai School of Medicine Mystic Marinelife Aquarium National Aeronautical & Space Administration National Institute of Allergy and Infectious Diseases National Institute of Environmental Health Sciences National Institutes of Health National Heart. Lung. & Blood Institute National Institutes of Mental Health National Institutes of Health National Institute of Neurological and Communicative Disorders and Strokes National Jewish Center For Immunology and Respiratory Medicine National Oceanographic and Aeronautical Administration (NOAA) National Research Council New England Biolabs New England Medical Center New Haven, University of New Hampshire, University of New York University New York University Medical Center New York, City University of, Bernard M. Baruch College New York, City University of, Brooklyn College New York. City University of. City College New York, City University of. Hunter College New York, City University of, Lehman College New York, City University of. Ml. Sinai School of Medicine New York, State University of, Albany New York, State University of, Buffalo New York, State University of, Downstate Medical Center New York, State University of. Health Sciences New York, State University, Health Science Center at Brooklyn New York, State University. Health Science Center at Stony Brook New York. State University, Upstate Medical Center New York, State University of. Fredonia New York, State University of. New Paltz New York, State University of, Syracuse New York, State University of. Stony Brook Nikon. Inc. North Carolina State University North Carolina. University of North Carolina, University of, at Chapel Hill North Carolina, University of, at Greensboro North Carolina, University of. School of Medicine North Dakota, University of Northwestern University Medical School Northwestern University Medical and Dental Schools Northwestern LIniversity Oakland University Oberlin College Oklahoma, LIniversity of Oklahoma, University of. Health Science Center Ohio State LJniversity Old Dominion University Olympic Corporation of America Oregon Institute for Marine Biology Oregon State University Oregon, University of Pacific Sciences Institute Pennsylvania State University Pennsylvania, University of Pennsylvania, LIniversity of. School of Medicine Perceptics Corporation Photometries. Ltd. Photonic Microscopy. Inc. Pittsburgh, University of. School of Medicine Pittsburgh, LIniversity of. Western Psychiatric Institute Foreign Institutions Represented 35 Polaroid Corporation Pomona College Population Council Princeton University Puerto Rico, University of Purdue University Quinnipiac College Rensselaer Polytechnic Institute Reichert-Jung Rhode Island, University of Rice University Rochester, University of Rochester, University of. Medical Center Rochester, University of. School of Medicine and Dentistry Rockefeller Foundation Rockefeller University Rush Medical Center Rush University Rutgers University San Diego State University Salk Institute Salsbury Laboratories, Inc. Scripps Clinic and Research Foundation Scripps Institute of Oceanography Sea Research Foundation Showa University Research Institute Smith College Smithsonian Environmental Research Center Smith, Kline, and French Laboratories South Carolina, University of, Medical School Southern California, University of Southern California, University of. School of Medicine Southern Louisiana State University Southwestern Louisiana. University of Stanford University Stanford University Medical School State of Florida, Bureau of Marine Research Stevens Institute of Technology Swarthmore College Syracuse University Technical Manufacturing Corporation Technical Products International, Inc. Temple University Texas Agriculture and Mines Texas Tech. University Texas Christian University Texas, University of, Austin Texas, University of. Health Science Center, Dallas Texas, University of. Health Science Center, Houston Texas, University of. Health Science Center, San Antonio Texas. University of. Medical Branch, Galveston Texas, University of. Medical School Thomas Jefferson University Thomas Jefferson Medical College Trio Laboratories Tufts New England Medical Center Tufts University Tufts University School of Medicine Tufts University School of Veterinary Medicine Uniformed Services University of Health Sciences Union College United States Air Force Academy L'nited States Department of Agriculture, Russell Lab United States Department of Agriculture, Plant Gene Expression Center United States Environmental Protection Agency United States Food and Drug Administration LIniversal Imaging Corporation University Hospital. Boston University Hospitals of Cleveland LIpjohn Company, The LJtah. University of Vanderbilt University Vassar College Veterans Administration Medical Center. Shreveport, LA Videoscope International, Ltd. Virginia Polytechnic Institute and State University Virginia, University of Wake Forest University Washington State University Washington University Washington University School of Medicine Washington, University of Washington, LIniversity of. School of Medicine Wayne State University School of Medicine Wesleyan University Western Connecticut State University Whitney Laboratory Wild Leitz USA. Inc. Wisconsin, University of Wisconsin, University of. School of Veterinary Medicine Woods Hole Oceanographic Institution Worcester Foundation for Experimental Biology Yale University Yale University School of Medicine Yale University School of Public Health Zeiss, Carl, Inc. Zoecon Research Institute Foreign Institutions Represented Alberta. LIniversity of. Canada Auckland. University of. New Zealand Barcelona, University of, Spain Basel, University of, Switzerland Berlin. University of, FRG Biozentrum, Switzerland Birmingham, University of, U.K. Bristol, University of, U.K. Calgary, University of, Canada Cambridge, University of, U.K. Catania, University of. Italy Catholic University of Leuven. Belgium Central University of, Venezuela Centre National de la Recherche Scientifique, France Centra BioFiscia y Bioquinica, Venequela Centra Nacional Patagonico, Argentina Chile University, Chile Clinical Research Institute, Canada Cuyo, University of, Mendoza. Argentina Dalhousie LIniversity, Canada Ecole Normale Superieure, France Edinburgh. University of, U.K. European Molecular Biology Laboratory, FRG Federal University of Goias, Brazil Federal University of Minas Gerais. Brazil Flinders University of South Australia. The 36 Annual Report Freie Universitat of Berlin. FRG Friedrich Miescher-Institut. Switzerland Glasgow, University of, Scotland. U.K. Groningen, University of. The Netherlands Guelph, University of, Canada Hamburg. University of, FRG Hebrew University, The, Israel Hebrew University of Jerusalem, Israel Heidelberg, University of, FRG Imperial Cancer Research Fund, U.K. Imperial College of Science and Technology, U.K. Institut fur Meereskunde. FRG Institut fur Tierphysiologie, FRG Institut Pasteur, France Institute of Biological Research, Yugoslavia Instituto de Biologia, Argentina Institute de Histologia y Embriologia, Argentina Instituto de Investigacion Medica, Argentina I.G. Medical College, Simla, India Inserm, France Instituto di Biologia Cellulare. Italy Instituto de Microbiolgie, Brazil Instituto Venezolanode Investigaciones Cientifican, Venezuela International Lab. for Research on Animal Diseases, Kenya Karolinska Institutet, Sweden Leuven, University of. Belgium London School of Hygiene and Tropical Medicine, U.K. Max-Planck Institute, FRG McGill University. Quebec, Canada Medical Research Council, England, U.K. Montreal. University of. Canada Naples Zoological Station. Italy Naples, University of, Italy National University of Mexico, Mexico Nottingham, University of, LI.K. Osnabruck, University of, FRG Ottawa, University of, Canada Oxford University, U.K. Pasteur Institute, France Paulista School of Medicine. Brazil Philipps LIniversitat, FRG Plant Research Center. Canada Puerto Rico, University of Research Institute, The Netherlands Siena, LIniversity of. Italy St. Andrews University, Scotland, U.K. St. Georges Hospital Medical School, U.K. Station Zoologique. France Stazione Zoologica, Italy Stockholm. University of, Sweden Strathclyde, University of, Scotland. U.K. Sussex University. U.K. Swansea. University College at. U.K. Sweden, University of, Stockholm Technical University of Munich. Garching. FRG Toronto. University of, Canada Toyohashi University of Technology. Japan Tromso. University of Norway Tubingen, University of, FRG Ulm. University of, FRG Universidad Nacional de Mar del Plata. Argentina Universidad de Sao Paulo, Brazil Universidad de Estadual de Campinas. Brazil Universidade Federal do Espirito, Brazil Universita Degli Studi Di Napoli, Italy Universitat Kiel, FRG University College, Ireland University College of London, England. U.K. University College of North Wales, U.K. Uppsala University, Sweden USSR Academy of Sciences. USSR Utrecht, University of. The Netherlands Warwick, University of, LI.K. Weizmann Institute of Science, Israel Yamaguchi University, Japan Zoological Station, Italy Zoologisches Institut der Universitat, Heidelberg, FRG Zurich, LIniversity of, Switzerland Year-Round Research Programs Boston University Marine Program Faculty Strickler, J. Rudi, Professor of Biology, Program Director Atema, Jelle, Professor of Biology Humes, Arthur G., Professor of Biology Emeritus Tamm, Sidney L.. Professor of Biology Valiela, Ivan, Professor of Biology Visiting investigators and instructors Baldwin, Christopher (Boston University) Bloomer, Sherman (Boston University) D'Avanzo, Charlene (Hampshire College) Deegan, Linda (University of Massachusetts, Amherst) Freadman, Marvin (Visiting Assistant Professor) Guerrero, Ricardo (University of Barcelona) Margulis, Lynn (University of Massachusetts, Amherst) Marrase, Celia (University of Barcelona) Mitchell. James (University of Barcelona) Muscatine, Leonard (University of California, Los Angeles) Peckol, Paulette (Smith College) Rhoads, Donald Richman. Sumner( Lawrence University) Rietsma, Carol (SUNY. New Paltz) Sarda, Rafael (University of Barcelona) Research staff' Costa, Joseph. Research Associate Costello, Jack, Research Associate Tamm, Signhild, Senior Research Associate Van Etten, Richard. Research Assistant Voigt, Reiner, Research Associate Zinn, Margery, Research Assistant Staff Corker, Amy, Course Assistant Harm, Dorothy. Senior Administrative Secretary Hinckle, Greg, Course Assistant Sunley. Madeline, Administrative Manager Weis, Virginia. Course Assistant Graduate students A I her. Merryl Banta. Gary Brooks, Cydney Corotto, Frank Coughlin, David Cowan, Diane Elskus, Adria Foreman, Kenneth Gallager, Scott Hahn.Jill Hersh, Douglas Krieger, Yutta Hwang, Jiang-shiou LaMontagne, Michael Lavalli, Kari Mazel, Charles Merrill, Carl Moore. Paul Mulsow, Sandor Piotrowski, Michael Scott. Marsha Siddiqui, Pirzada Tamse, Armando Trager, Geoffrey Trott, Thomas White, David Wood. Susan Undergraduate students Batjakas. loanis Bergles, Dwight Brewer. Matthew C. Buckley, Joseph P. 37 38 Annual Report Call, Christopher A. Casteline, Jennifer Cochran, Wendy Eleuterio. Daniel Forrest, Davina Forrester. Amy L. Halczyn. William C. Kennedy. S. Blain Kenzora, Kristen T. Kleinhans, Julie (Cornell) Kreuger, Dana (Lawrence University) LaPusata, David T. Lynch. Helen M. MassofT, Daniel A. Mayer. Marilyn A. (LIniversity of Maryland) Michmerhuizen, Cathy (Lawrence University) Myers, Monique Melvin, Mary Kay Pacioni. Thomas D. (SUNY. Buffalo) Parker, Todd A. Pizzelanti, Donna M. Rugoletti, Steven J. Rutka, Timothy Scholz, Nathaniel L. Short. Graham Siwko, Robert P. Stone, Gayle (Skidmore) Zeller. Robert (special honors work) Summer undergraduate interns Butler, Elizabeth (Hampshire College) Coburn. Cara Forrest, Davina (Boston University) Gerardo, Hortense Townsend, Susan Rogers, Ruth Rutka, Timothy Sammon, Leslie Short, Graham Laboratory ofJelle Atema Organisms use chemical signals as their main channel of information about the environment. These signals are transported in the marine environment by turbulent currents, viscous flow, and molecular diffusion. Receptor cells extract signals through various filtering processes. Currently, the lobster with its exquisite sense of taste and smell is our major model, to study the signal filtering capabilities of the whole animal and its narrowly tuned receptor cells. Research focuses on amino acids (food signals) and pheromones (courtship), neurophysiology of receptor cells, behavior guided or modulated by chemical signals, and computer modeling of odor plumes and neural filters. Laboratory of Arthur G. Humes Research interests include systematics, development, host specificity, and geographical distribution of copepods associated with marine invertebrates. Current research is on taxonomic studies of copepods from invertebrates in the tropical Indo-Pacific area, and poecilostomatoid and siphonostomatoid copepods from deep-sea hydrothermal vents and cold seeps. Laboratory ofRudi J. Strickler Use high-speed cinematography and special laser light optical systems with target tracking devices to observe zooplankton-algae. carnivorous-herbivorous zooplankton. and fish-zoo-plankton interactions. Lab and field results show the degree to which abiotic forces influence the evolution of species, feeding guilds, and predator-prey interactions. Additional topics in the feeding ecology ofcrinoids. bryozoans and other suspension feeding invertebrates enhance perception of the first consumer level in the aquatic food chain. Laboratory of Ivan Valiela Research emphasis is on structure and function of salt marsh ecosystems and coastal embayments, including the processes of predation, herbivory, decomposition, and nutrient cycles. A parallel line of work, with more applied aspects, is eutrophication in coastal marine communities and interactions between watersheds and coastal waters. The Ecosystems Center The Center was established in 1975 to promote research and education in ecosystems ecology. Eleven scientists study the terrestrial and aquatic ecology of a wide variety of ecosystems ranging from northern Europe (trace gas emission from acid- rain affected forests) and the Alaskan Arctic (long-term studies on the controls of tundra, lake and stream biota), to the Harvard Forest (long-term studies of the effects of disturbance on forest ecosystems) and Buzzard's Bay (controls of anaerobic decomposition). Many projects, such as those dealing with sulfur transformations in lakes and nitrogen cycling in the forest floor, investigate the movements of nutrients and make use of the Center's mass spectrometry laboratory (directed by Brian Fry) to measure the stable isotopes of carbon, nitrogen, and sulfur. The research results are applied wherever possible to questions of the successful management of the natural resources of the earth. In addition, the ecological expertise of the staff is made available to public affairs groups and government agencies who deal with such problems as acid rain, groundwater contamination, and possible carbon dioxide-caused climate change. Stall and consultants Melillo, Jerry M., Acting Director Hobbie. John E., Director (on sabbatical) Banta, Gary T. Bauman, Carolyn H. Boutwell, Anne M. Bowles. Francis P. Brooks, Marilyn Year-Round Research Programs 39 Davis, Sarah J. Danforth, Carolyn Dornblaser, Mark Downs, Martha Fry, Brian Giblin, Anne E. Griffin, Elisabeth A. Helfrich, John V. K. Hooper, David Hullar, Meredith Jordan. Marilyn Kicklighter, David W. Knudson. Heather N. Laundre, James McKerrow, Alexa Michener, Robert H. Nadelhoffer, Knute J. O'Brien, Margaret Pallant, Julie Peterson, Bruce J. Piterman, Oksana Rastetter, Edward B. Regan, Kathleen M. Ricca, Andrea Russell, Anne E. Saupe, Susan Schwamb, Carol Semino, Suzanne J. Shaver, Gaius R. Steudler, Paul A. Tucker, Jane Postdoctorals Bowden, Richard D. KJing, George W. Mclvor, Carole C. Raich, James Ryan, Michael G. Visiting investigators Joyce, Linda, U.S.D.A. Forest Service Waring, Richard, University of Oregon Laboratory for Marine Animal Health The laboratory provides diagnostic, consultative, research and educational services to the institutions and scientists of the Woods Hole community concerned with marine animal health. Diseases of wild, captive, and cultured animals are investigated. Staff Leibovitz. Louis. Director, Laboratory for Marine Animal Health, MBL, and Professor, Department of Avian & Aquatic Animal Medicine, New York State College of Veterinary Medicine Abt, Donald A., Co-Investigator, University of Pennsylvania Bullis, Robert A.. Senior Research Associate. Cornell LIniversity Hansen, Sandra B., Secretary, Cornell University McCafferty, Michelle, Histological Technician, Cornell University Moniz, Priscilla C., Administrative Secretary Wadman, Elizabeth A.. Microbiological Technician, Cornell University Visiting investigators Garvey, Margaret, University of Pennsylvania Laboratory of Aquatic Biomedicine Our laboratory is studying hematopoietic neoplasia, a leukemia-like disease of soft shell clams. Monoclonal antibodies developed by this laboratory and techniques in molecular biology are used to investigate the differences between normal and leukemic cells and their ontogeny. Staff Reinisch. Carol L., Investigator, MBL, and Chairperson, Department of Comparative Medicine, Tufts University School of Medicine Miosky, Donna, Laboratory Technician Turano, Brian, Research Assistant 40 Annual Report Laboratory of Cell Biochemistry This laboratory studies developmental, metabolic, and environmental influences on the genetic regulation of cellular enzymes. Current emphasis is on hepatic heme biosynthesis and utilization in marine fish. These processes are responsive to hormonal and nutritional signals as well as to environmental pollutants such as hydrocarbons. This work is being conducted with primary cultures of hepatocytesand with cDNA probes that permit quantitation of gene activity. Other research is concerned with translocation of proteins between various subcellular compartments both in fish hepatocytes and in invertebrate eggs before and after fertilization. Staff Neal W. Cornell, Senior Scientist Grace Bruning, Research Assistant Michael Ferkowicz. Research Assistant Laboratory ofD. Eugene Copeland Electron microscopy of luminescent organs (photophores) in deep-sea fish; gas secretion in swimbladders of deep-sea fish; and osmoregulatory tissue in Limulus. Laboratory of Developmental Genetics This research group studies the early gene control of cellular differentiation pathways (cell lineage determination) in embryos of tunicates and other marine invertebrate species. Staff Whittaker, J. Richard, Senior Scientist Crowther, Robert, Research Assistant Loescher, Jane L., Research Assistant Meedel, Thomas H., Assistant Scientist \ 'isiting investigators Arnold, John M., University of Hawaii Collier, J. R., Brooklyn College Heady, Judith E., University of Michigan-Dearborn (Sabbatical year. 1987-88) Johnson, Carl D., Cambridge NeuroScience Research Laboratory of Judith P. Grassle Studies on the population genetics and ecology of marine invertebrates living in disturbed environments, especially of sibling species in the genus Capita/In ( Polychaeta). Grassle, Judith P.. Senior Scientist Mills, Susan W., Research Assistant Laboratory^ ofHarlyn O. Halvorson The first project involves the study of spore germination in the bacterium Bacillus sithlilis. A lambda genomic library has been screened and clones corresponding to the Ger] locus have been isolated. Ger J is involved in late spore germination. We wish to characterize Ger] and flanking sequences. In collaboration with J. Weinberg(WHOI), we are also studying genome evolution in the marine polychaete Nereis acwninata. Several reproductively isolated field populations will be analyzed by comparative physical mapping, using cosmids and artificial chromosome vectors. Staff Halvorson, Harlyn, Principal Investigator, Director Chikarmane. Hemant. Assistant Scientist Pratt, Sara S., Research Assistant I 'isiting investigators Kornberg, Hans. University of Cambridge, UK Keynan, Alex, Memorial Sloan Rettering Cancer Center Vincent, Walter, University of Delaware Laboratory ofShinya Inoue Mechanism of mitosis and related motility. Development of high resolution 3-D video microscope systems. Staff Inoue. Shinya, Distinguished Scientist. MBL. and University of Pennsylvania Anniballi, Dyon, Programming Engineer. Cornell University College of Engineering Boyd, Steven, Programming Engineer, Cornell University College of Engineering Inoue, Theodore, Programming Engineer. Cornell University College of Engineering Shimomura, Sachi, Research Assistant, Stanford University Taracka. Robert. Research Assistant Woodward, Bertha M., Laboratory Manager Visiting investigators Bajer, Andrew S., University of Oregon Burgos, Mario H., University of Cuyo, Mendoza, Argentina Fukui, Yoshio, Northwestern University Medical School Kiehart, Daniel P., Harvard University Salmon, Edward D., University of North Carolina Sardet, Christian, Biol. Cell. Marine. Ville France-Sur-Mer Silver, Robert B., University of Wisconsin Laboratory of Sensory Physiology Since 1973, the laboratory has conducted research on various aspects of vision. Current studies focus on photoreceptor cells; on their light-absorbing pigments; and on their biochemical reactions initiated by light stimulation. Microspectrophotometric and biochemical techniques are used to study the receptors of both vertebrates (amphibia, fish, and mammals) and invertebrates (horseshoe crab and squid). Year-Round Research Programs 41 Staff Harosi, Ference, Director. Associate Scientist, MBL, and Boston University School of Medicine Szuts, Ete. Assistant Scientist, MBL, and Boston University School of Medicine Trapp, Susan. Research Assistant Zahajszky. Tibor, Research Associate I 'isiting investigators Clay, John R., Laboratory of Biophysics, NINCDS/NIH Cornwall, Carter, Boston University School of Medicine Eckberg. William R., Howard University Hawryshyn, Craig W., McMaster University, Canada Nicaise, Ghislain, University of Nice, France Petry, Heywood M., SUNY, Stony Brook Wood, Susan F., Boston University Marine Program Laboratory of Neurobiology The Laboratory of Neurobiology is concerned with the secretory mechanism underlying synaptic transmission, the mechanism of organelle movement underlying axoplasmic transport, and the organization of neural cytoplasm. stuff Reese, Thomas S., Chief, Laboratory of Neurobiology, MBL. NINCDS/NIH Andrews, S. Brian, Research Associate Bechtold-Imhof, Ruth. Research Assistant Cheng, Toni, Research Associate Chludzinski. John, Research Technician Coyle, Jo-Anne, Secretary Gallant, Paul, Research Associate Garbus-Gooch, Cynthia, Research Assistant Hammar, Katherine, Research Assistant Khan, Shahid, Visiting Research Associate Reese, Barbara F., Research Technician Sheetz, Michael P., Visiting Research Associate Schnapp, Bruce J., Research Associate Tatsuoka, Hozumi, Research Associate Terasaki, Mark, Research Associate Laboratory of Neuroendocrinology This laboratory studies the molecular and cellular bases of two neural programs that regulate different important behaviors in the model mollusc Aplysia. Research is conducted on the mechanisms of the neuronal circadian oscillators located in the eyes. These circadian oscillators drive the circadian activity rhythm of the animal, which is concerned with the daily timing of food gathering and of prolonged rest. Additional research is conducted on a group of neuroendocrine cells that produce a peptide, "egg-laying hormone," that initiates egg laying and associated behaviors. The laboratory is interested in how the three-dimensional shape of this peptide hormone allows a highly specific interaction with its receptor and the intracellular processes that are triggered by it. In another project, the laboratory has discovered and is continuing research on an anti-toxin protein that inhibits ADP-ribosylation of G-proteins induced by bacterial exotoxins. Staff Strumwasser, Felix, Director, Laboratory of Neuroendocrinology, Boston University School of Medicine, and MBL Cox, Rachel L., Senior Research Assistant Eason, Barbara. Laboratory Assistant Click, David, Senior Postdoctoral Fellow Hellmich, Mark, Graduate Student (Boston University School of Medicine) Viele, Daniel P., Senior Research Assistant Laboratory of Michael Rabinowitz Measurement of lead in baby teeth to see if lead exposure at different ages, recorded at different sites within teeth, are related to child development. Staff Rabinowitz, Michael, Investigator, MBL, and Instructor in Neurology, Harvard Medical School Lewandowski, Ann, Research Assistant, Harvard Medical School Laboratory ofOsamu Shimomura Biochemical studies of the various types of bioluminescent systems. Preparation of the improved forms of aequorin for measuring intracellular free calcium. Staff Shimomura, Osamu, Senior Scientist, MBL, and Boston University School of Medicine Shimomura, Akemi, Research Assistant [ 'isiting investigators Musicki. Branislav, Harvard University Nakamura, Hideshi, Harvard University 42 Annual Report Laboratory of Raymond E. Stephens Biochemistry of microtubules in cilia, tlagella. and the cytoplasm; mechanosensitivity and the control of ciliary' movement. Staff Stephens, Raymond E., Investigator, MBL, and Boston University School of Medicine Good, Michael J.. Research Assistant Oleszko-Szuts. Susan, Research Associate Stommel, Elijah W., Research Associate, St. Elizabeth's Hospital Visiting investigator Holz. George G., Tufts University School of Medicine Laboratory ofRaquel Sussman Investigation of the molecular mechanism of DNA damage- inducible functions. Present studies deal with the structure- function relationships of X represser analyzed by immunological techniques. Staff Sussman, Raquel, Associate Scientist McLaughlin, Jane, Research Assistant Cornuel. Catherine, Research Assistant National Vibrating Probe Facility We are exploring the roles of ionic currents, gradients, and waves in controlling development. We focus on controls of pattern and by calcium ions. Staff Jaffe, Lionel, Senior Scientist and Facility Director Kuhtreiber, Wiel, Physiologist Miller. Andrew, Research Associate Sanger, Richard, Technician Shipley, Alan, Technician Williams, Phillip C., Engineer Speksnijder, Annelies, Research Associate I 'isiting investigators Asman, Sally, Harvard University Baumann. Steve, EPA, Research Triangle Park, NC Biggers, John, Harvard Medical School Bowdan, Elizabeth, University of Massachusetts Chen, Tsung-Hsien, Academia Sinica, Taiwan Crawford. Karen, Swarthmore College Cullander, Chris, University of California, S. F. Devlin, Leah, University of Rhode Island Dickey, Joe, Clemson College Diehl-Jones. W. L., University of Manitoba. Canada Fink, Rachael, Mount Holyoke College Holtug, Klars, Royal Veterinary University, Denmark Kunkel, Joseph. Llniversity of Massachusetts Mladenov. Phillip, Mount Allison University, New Brunswick. Canada Payne, Richard, Llniversity of Maryland Pet nig, Ron, University College of North Wales, U. K. Sardet, Christian, Station Marine Ville franche-sur-mer, France Trinkaus, John, Yale University Troxell. Cindy. University of Colorado Ver Achtert, Barend, University of Leuven, Belgium Wyman, Robert, Yale University Lionel Jaffe in his laboratory. Honors Centennial Evening Lectures Meredith Applebury, University of Chicago, 24 June "Hecht and Wald: Towards the Molecular Basis of Vision" Daniel Koshland Jr., University of California, Berkeley, 1 July "The Future of Biological Research: What is Possible and What is Ethical" Torsten Wiesel. The Rockefeller University, 8 July, Lang Lecture ' 'Neural Mechanisms of I 'isual Perception: The Legacy of Hart line and Kuffler Joshua Lederberg, The Rockefeller University, 15 July "Genetic Maps: Fruit Flies, People, Bacteria, and Molecules — A Tribute to Morgan and Sturtevant" Gerald Weissmann, New York University, 1 7 July, Dedication Day "Loeb: The Mechanistic Conception of Life Revisited Sydney Brenner, Medical Research Council, 21, 22, July, Forbes Lectures "Reading the Genetic Script " "Genes and the Nervous System " Clay Armstrong, University of Pennsylvania, 29 July "Ionic Channels That are Gated by I 'oltage: A Tribute to Cole and Hodgkin John Hobbie, Marine Biological Laboratory, 5 August "Ecology at Woods Hole: Baird, Bigelow, and Red field" Donald Kennedy, Stanford University, 6 August, Monsanto Biotechnology Lecture "The Regulation oj Science: How Much Can We Afford'.'" Robert Solow, Massachusetts Institute of Technology, 7 August, Hiroshima Day Lecture "Military Spending and Economic Health: A Great Place to Start?" Edward Wilson, Harvard University, 12 August "Analyzing the Superorganism: The Legacy of Whitman and Wheeler" Clifford Slayman, Yale University, 16 August 1 'Channels, Pumps, and Motors in Plants: A Tribute to Osterhoiil" Shinya Inoue, Marine Biological Laboratory, 19 August ' 'Porter and the Fine A rchitectiire of Dividing Cells" John Gurdon, Cambridge University, 26 August "Determinants of 'Development: From Conklin and Li/lie to the Present Fellowships Joshua Lederberg delivering his Centennial Evening Lecture. Robert Day Allen Fellowship Suprenant, Kathy A., University of Kansas Frederik B. Bang Fellowship Fund Dykens, James A., Grinnell College Frank A. Brown, Jr., Memorial Readership Hill, Richard W., Michigan State University 43 44 Annual Report Founders Fellowship Beckvvith. Susan M., Purdue University Brown, Alice B.. Vanderbilt University Cintron, Isabel G.. University of Puerto Rico Dunn, Patrick J.. University of Pennsylvania Fautsch, Michael P., Mayo Clinic Foundation Georgiou, Chris, University of Iowa Hayden-Baille Fellowship Crossley, Ian Brian, University College of London, England Willcox, Mark S., University College at Swansea, England William Randolph Hearst Fellowship Gramoll, Siglnde, The Hebrew University of Jerusalem, Israel Stephen W. Kuffler Fellowships Nolen, Thomas G., University of Miami Saez, Juan C., Albert Einstein College of Medicine Frank R. Lillie Fellowship Ding, Dali, California Institute of Technology Dittmann, Sabine, Zoologisches Institute, FRG Geysen, Johan J., University of Leuven, Belgium Hidalgo, Alicia, University of Oxford, England Lackschewitz, Dagmar, Max-Planck Institute, FRG Mayor, Robert A., University of Chile, Chile Morrisey. Donald J., University of Bristol, England Rowan, Robert G., Johns Hopkins University Wickramasinghe. Dineli M., Tufts University Jacques Loeb Fellowship Watson, Kellie L., University of California at Irvine MBL Summer Fellowships Fine, Allen, Dalhousie University, Canada Fink, Rachel, Mount Holyoke College Herbert W. Rand Fellowship Gibson, Saxon M., Vanderbilt University Kiener, Andreas M., Harvard Medical School Mendoza, Lisa M., Scirpps Institute, University of California at San Diego Science Writing Fellowships Amato, Ivan, Science News Crabtrec, Margo, Science World Golden. Frederic, Freelance, San Francisco, CA Kanigel. Robert, Freelance. Baltimore, MD Keller, Eric, Smithsonian Knudson, Mary, Baltimore Sim Kornberg, Warren, Mosaic Raymond, Chris Anne. Journal of the American Medical A ssot -iat ion H. B. Steinbach Fellowships Palazzo, Robert E., University of Virginia Scholarships Biology Club of City University of New York Houser. Felicia, Harvard University Father Arsenius Boyer Scholarship Fund Johnson, Jennifer A., Dartmouth College C. Lalor Burdick Scholarship Blumenthal, Edward M., Yale University Fautsch, Michael P., Mayo Clinic Foundation Larochelle, Denis A., Stanford University Gary N. Calkins Memorial Scholarship Dunn, Patrick J., University of Pennsylvania Lemosy, Ellen K., Duke University Frances S. Claff Memorial Scholarship Bright, Kerris E., University of Sussex, England Cannon, Stephen C., Massachusetts General Hospital Edwin Grant Conklin Memorial Scholarship Jones, Diana C., Purdue LIniversity Lucretia Crocker Endowment Fund Cintron, Isabel G., University of Puerto Rico William F. and Irene Diller Scholarship Fund Beckwith, Susan M., Purdue University Brown, Alice B., Vanderbilt University Caswell Grave Scholarship Felsenfeld. Dan P., Columbia University Keller, Steve M., University of California, San Diego/ Scripps Institute of Oceanography Honors 45 Aline D. Gross Scholarship Klatte, D. H., Northwestern University Merkel H. Jacobs Scholarship Johansson, Staffan, Karolinska Institutet, Nobel Institute, Sweden Arthur Klorfein Fund Scholarship Krufka, Alison, University of Wisconsin Pesavento, Patricia A., Harvard University Scemes, Elina, Universidad de Sao Paulo. Brazil Thompson, Margaret A., Harvard University Vu, Eric T., University of California, Los Angeles Wehner, Katja A., Max-Planck Institute, FRG Lucille P. Markey Charitable Trust Scholarship Elliott, David D., Glasgow University, Scotland Georgiou, Chris, University of Iowa Hoecker, Beatrix B., University of Hamburg, FRG Mamajiwalla, Salim N., University of Miami Messina, Dino A., State University of New York, Syracuse Morrissette, Naomi S., University of Pennsylvania Oblong, John E., University of Chicago Olsen, Scott G., University of Minnesota Rocha, Rosana M., University of Estadual de Campinas, Brazil Rounseville, Matthew P., George Washington University Rushforth, Alice M., University of Wisconsin Sandulli. Roberto, University of Naples, Italy Schmager, Carola, Institut fur Meereskunde, an der Universitat Kiel, FRG Subramanian, Sukanya V., Albert Einstein College of Medicine Vattay, Anthony, Rutgers University Vanghan, Kevin T., Cornell University Wang. Jin. Worcester Foundation for Experimental Biology Wolenski, Joseph S., Rutgers University Wolfe, Connie J., Scripps Institute of Oceanography, L'niversity of California, San Diego Yue, Lin, Johns Hopkins University S. O. Mast Founders Scholarship Kane, Susan B., University of Massachusetts, Boston Allen R. Memhard Scholarship Hamrick, Maura L., Johns Hopkins University James S. Mountain Memorial Fund Scholarship Beckwith, Susan M., Purdue University Brown, Alice B., Vanderbilt University Cintron, Isabel G., University of Puerto Rico Dunn, Patrick J., University of Pennsylvania Fautsch, Michael P., Mayo Clinic Foundation Planetary Biology Internship Carman, Kevin R., Florida State University Lackschewitz, Dagmar, Max-Planck Institute. FRG Marjoire W. Stetten Scholarship Fund Yang, Xian-Cheng, State University of New York, Buffalo Yuste, Rafael M., Rockefeller University Surdna Foundation Scholarship Huber, Robert, Texas Technical University Li, Jian, State University of New York, Syracuse Long, Guohong, University of Massachusetts Mindell, Joseph A., Albert Einstein College of Medicine Ondeka, Cheryl A., University of Massachusetts at Amherst Pinckney, James L., University of South Carolina Zhao, Dayao, Harvard Medical School William Morton Wheeler Family Founders' Scholarship Gillmore, Todd H., State University of New York at New Paltz Howard A. Schneiderman (center) accepts an MBL Distinguished Leadership Award from MBL Director Harlyn Ilalvorson (left) and Chairman Prosser Gifford (right). Board of Trustees and Committees Corporation Officers and Trustees Ex officio Honorary Chairman of the Board of Trustees, Denis M. Robinson, Key Biscayne, FL Chairman of the Board of Trustees, Prosser Gifford, Washington, DC President of the Corporation and Director, Harlyn O. Halvorson, Marine Biological Laboratory, Woods Hole, MA Treasurer, Robert D. Manz. Helmer & Associates, Waltham. MA Clerk of the Corporation, Kathleen Dunlap, Tufts University School of Medicine, Boston, MA Class of 1992 Norman Bernstein, Bernstein Management, Inc., Washington, DC Ellen R. Grass, Grass Foundation, Quincy, MA Warren G. Hathaway, Hathaway Publishing, Somerset, MA Sir Hans Kornberg, Christ's College, Cambridge, UK George Langford, University of North Carolina. Chapel Hill,NC Jack Levin, University of California School of Medicine, San Francisco, CA Evelyn Spiegel, Dartmouth College, Hanover, NH Andrew G. Szent-Gyorgyi, Brandeis University, Waltham, MA Kensal E. VanHolde, Oregon State University, Corvallis, OR Stanley W. Watson, Associates of Cape Cod, Inc., Falmouth, MA Class of 1991 Robert B. Barlow Jr., Syracuse University, Syracuse, NY Dieter Blennemann. Carl Zeiss, Inc., Thornwood, NY James M. Clark, Woods Hole, MA Wensley G. Haydon-Baillie, Porton, Int., London, UK Laszlo Lorand, Northwestern University, Evanston, IL Lionel I. Rebhun, University of Virginia, Charlottesville. VA Carol L. Reinisch. Tufts University School of Veterinary Medicine, Boston, MA Brian M. Salzberg, University of Pennsylvania, Philadelphia, PA Howard A. Schneiderman. Monsanto Company, St. Louis, MO Sheldon J. Segal. The Rockefeller Foundation. New York, NY Class of 1990 John E. Dowling, Harvard University, Cambridge, MA Gerald D. Fischbach, Washington University School of Medicine, St. Louis, MO Robert D. Goldman. Northwestern LJniversity, Chicago, IL John E. Hobbie. Marine Biological Laboratory, Woods Hole, MA Richard E. Kendall, East Falmouth, MA Irving W. Rabb, Boston, MA Joan V. Ruderman. Duke University, Durham, NC Ann E. Stuart, University of North Carolina, Chapel Hill, NC D. Thomas Trigg, Wellesley, MA Trustee Committees 47 Class of 1989 Garland E. Allen, Washington University, St. Louis, MO Peter B. Armstrong, University of California, Davis, CA Robert W. Ashton, Gaston, Snow, Beekman & Bogue, New York, NY Jelle Atema, Marine Biological Laboratory, Woods Hole, MA John G. Hildebrand, University of Arizona, Tucson, AZ Thomas J. Hynes Jr., Meredith and Grew, Inc., Boston, MA Robert Mainer, The Boston Company, Boston, MA Birgit Rose, University of Miami School of Medicine, Miami, FL Gerald Weissmann, New York University, New York, NY Emeriti John B. Buck, National Institutes of Health, Bethesda, MD Aurin Chase, Princeton University, Princeton, NJ Seymour S. Cohen, Woods Hole, MA Arthur L. Colwin, Key Biscayne, FL Laura Hunter Colwin, Key Biscayne, FL D. Eugene Copeland, Marine Biological Laboratory, Woods Hole, MA Sears Crowell, Indiana University, Bloomington, IN Alexander T. Daignault, Boston, MA William T. Golden, New York, NY Teru Hayashi, Woods Hole, MA Lewis KJeinholz, Reed College, Portland, OR Maurice E. Krahl, Tucson, AZ Charles B. Metz, Miami, FL Keith R. Porter, University of Pennsylvania, Philadelphia, PA C. Ladd Prosser, University of Illinois, Urbana, IL S. Meryl Rose, Waquoit, MA John Saunders, Jr., Waquoit, MA Mary Sears, Woods Hole, MA Homer P. Smith, Woods Hole, MA W. Randolph Taylor, University of Michigan, Ann Arbor, MI George Wald, Cambridge, MA Executive Committee of the Board of Trustees Prosser Giftbrd, Chairman* Robert B. Barlow Jr., 1991 John E. Dowling, 1990 Ray L. Epstein* * ex offitio Gerald D. Fischbach, 1989 Harlyn O. Halvorson* Robert D. Manz* Irving W. Rabb, 1991 Sheldon J.Segal, 1989 D. Thomas Trigg, 1990 Trustee Committees 1988 Audit Robert Mainer, Chairman Ray L. Epstein* Robert D. Manz* Sheldon J. Segal Andrew G. Szent-Gyorgyi D. Thomas Trigg Kensal E. Van Holde Stanley W. Watson Compensation Thomas J. Hynes Jr., Chairman James M. Clark John E. Dowling Investment D. Thomas Trigg, Chairman Ray L. Epstein* William T. Golden Maurice Lazarus Werner R. Loewenstein Robert D. Manz Irving W. Rabb W. Nicholas Thorndike Centennial James D. Ebert, Chairman Pamela L. Clapp, Coordinator* Garland E. Allen Robert B. Barlow Jr. Paul R. Gross Harlyn O. Halvorson* Olivann Hobbie Richard E. Kendall John Pfeiffer Keith Porter Frank Press C. Ladd Prosser John S. Reed D. Thomas Trigg John Valois 48 Annual Report Standing Committees for the Year 1988 Animal Care Committee Carol L. Reinisch, Chairman Ray L. Epstein* Leslie D. Garrick* Linda Huffer Edward Jaskun Andrew Mattox* Roxanna Smolowitz J. Richard Whittaker Buildings & Grounds Kenyon Tweedell, Chairman Lawrence B. Cohen Richard Cutler* Alan Fein Daniel Gilbert Clifford Harding, Jr. Ferenc Harosi Donald B. Lehy* Thomas Meedel Philip Person Lionel Rehhun Thomas Reese Evelyn Spiegel Employee Relations Ellen Binda. Chairwoman Ida Baker Peg Corbett H. Tom Fischer Lionel Hall Alan Shipley Frank Sylvia John MacLeod Fellowships Thoru Pederson, Chairman Ray L. Epstein* Leslie D. Garrick* Judith Grassle John G. Hildebrand George M. Langford Eduardo Macagno Carol L. Reinisch Housing, Food Service, and Child Care Jelle Atema, Chairman Robert B. Barlow, Jr. Gail Burd Stephen Highstein Lou Ann King* Thomas Reese Joan Ruderman Institutional Biosafety Raquel Sussman, Chairman Paul DeWeer Paul Englund Harlyn O. Halvorson* Paul Lee Donald B. Lehy* Joseph Martyna Andrew Mattox* Alfred W. Senft Instruction Judith Grassle, Chairman Ray L. Epstein* Brian Fry Leslie D. Garrick* John G. Hildebrand Ron Hoy Tom Humphreys Hans Laufer Brian Salzberg Roger Sloboda Andrew Szent-Gyorgyi Library Joint Management Harlyn O. Halvorson, Chairman* Garland Allen George Grice, WHOI John W. Speer* John Steele, WHOI Gary Walker, WHOI Library Joint Users Garland Allen, Chairman Wilfred Bryan, WHOI A. Farmanfarmaian Jane Fessenden* Lionel Jafte Catherine Norton* John Schlee, USGS John Teal, WHOI Edward Sholkovitz, WHOI Carole Winn, WHOI* Oliver C. Zanriou Standing Committees 49 Marine Resources Robert Goldman, Chairman William Cohen Richard Cutler* Louis Leibovitz Toshio Narahashi George Pappas Roger Sloboda Melvin Spiegel Antoinette Steinacher John Valois* Radiation Safety Paul DeWeer, Chairman David W. Borst Richard L. Chappell Sherwin J. Cooperstein Louis M. Kerr* Andrew Mattox* Ete Z. Szuts Walter Vincent Research Services Birgit Rose, Chairman Peter Armstrong Robert B. Barlow Jr. Richard Cutler* Barbara Ehrlich Ehud Kaplan Samuel S. Koide Aimlee Laderman * ex offiao Andrew Mattox* Bryan Noe Bruce J. Peterson Rudi Strickler Research Space Joseph Sanger, Chairman Clay Armstrong Ray L. Epstein* Leslie D. Garrick* David Landowne Hans Laufer Laszlo Lorand Eduardo Macagno Jerry Melillo Joan V. Ruderman Roger Sloboda Evelyn Spiegel Steven Treistman Ivan Valiela Safety John Hobbie. Chairman Daniel Alkon D. Eugene Copeland Richard Cutler* Edward Enos* Louis Kerr* Alan Kuzirian Donald B. Lehy* Andrew Mattox* Edward Sadowski Ray Stephens Paul Steudler Laboratory Support Staff' Biological Bitllt'tin Metz, Charles, B., Editor Bauer, Diane Clapp, Pamela L. Mountford, Rebecca J. Controller's Office Speer, John W., Controller Accounting Services Binda, Ellen F. Blain, Harriet Campbell. Ruth B. Davis, Doris C. Gilmore, Mary F. Godin, Frances T. Hobbs, Roger W.. Jr. Hough. Rose A. O'Brien-Sibson, Patricia J. Oliver, Elizabeth Poravas, Maria Cheni Room Clough. Lisa A. LaMontagne. Michael Mercurio, Kimberley Ross, Darcy Sadowski, Edward A. Computer Services Tollios. Constantine Purchasing Evans, William A. Hall, Lionel E., Jr. Copy Service (.'enter Gibson. Caroline F. Jackson, Jacquelyn F. Mountford, Rebecca J. Dcvc/o/niicnl Office Ayers, Donald E., Director Berthel. Dorothy Lvons. Elaine D. * Including persons who joined or slafl during I9KX O'Hara, Aqua Thimas, Lisa M. Director's Office Halvorson, Harlyn O., President and Director Berthel, Dorothy Epstein, Ray L. Kinneally, Kathleen R. Watkins, Joan E. Hitman Resources Goux, Susan P. Gray Museum Bush. Louise. Curator Armstrong, Ellen P. Montiero, Eva Housing King, LouAnn D., Conference Center and Housing Manager Crocker, Susan J. Eddy, Kristine A. Johnson, Frances N. Klopfer. Katherine Krajewski, Viola I. Kuil, Elisabeth Leach. Adele McNamara, Noreen Morrill, Barbara J. Perry. Ann R. Potter, Maryellen Price, Dale L. Roderick, Cynthia F. Sadovsky, Sebastian Taylor, Tobey Tedeschi, Christine R. Library Fessenden, Jane, Librarian Ashmore, Judith A. Costa. Marguerite E. Mirra, Anthony J. Mountford, Rebecca .1. Nelson, Heidi Norton, Catherine N. Pratson, Patricia G. deVeer, Joseph M. MBL Associates Liaison Scanlon, Deborah Public Information Office Liles, George W.. Jr., Director Anderson. Judith L. Dzierzeski. Michelle J. Stone, Beth R. Services. Projects, and Facilities Cutler, Richard D.. Manager Buildings and Grounds Lehy, Donald B.. Superintendent Anderson, Lewis B. Baldic, David P. Blunt, Hugh F. Bourgoin, Lee E. Carini, Robert J. Collins, Paul J. Conlin, Henry P. Conlin, Mary E. Fish, David L., Jr. Fuglister. Charles K. Gibbons. Roberto G. Gonsalves, Walter W., Jr. Klinger, Michael Krajewski, Chester J. Lochhead. William M. Lunn, Alan G. Lynch, Henry L. MacLeod, John B. McAdams, Herbert. Ill Mills. Stephen A. Rattacasa, Frank D. Rossetti, Michael F., Jr. Schoepf, Claude Schwamb. Peter .1. Simonelli. Bernard R. Simonelli.Guy S. deVeer, Robert L. Ward. Frederick Weeks, Gordon W. Wilson. Mitchell.!. Windle. Irvin Machine Shop Sylvia, Frank E. Laboratory Support Staff 51 Marine Resource* Center Valois, John J., Manager DeGiorgis. Joseph Enos. Edward G.. Jr. Enos, Joyce B. Fisher. Harry T., Jr. Frank, Donald S. Hanley, Janice S. Moniz, Priscilla C. Revellese, Christopher Sullivan. Daniel A. Tassinari, Eugene Photolab Golder. Linda M. Golder, Robert J. Rugh. Douglas E. Sponsored Programs Garrick, Leslie D.. Assistant Administrator Casiles. Phyllis B. Dwane, Florence Hurler. Linda Lynch, Kathleen F. Telephone Office Baker. IdaM. Geggatt, Agnes L. Ridley. Alberta W. Animal Care Facility Briggs, Kimberly A. Hamilton. Kathryn R. Povio, Sandra C. Sohn, Marcus J. Tripp. Gretchen Radiation Safety Mattox, Andrew H., Safety Officer Apparatus Barnes, Franklin D. Haskins. William A. Martin, Lowell V. Nichols, Francis H.. Jr. Sanger, Richard H., Jr. Shipping and Receiving Geggatt, Richard E. Illgen, Robert F. Monteiro, Dana Electron Microscopy Lab Hodge, Alan J. Kerr, Louis M. 1988 Summer Support Staff Allen, Carin T. Allen. Tania L. Amon, Tyler C. Anger, Susan R. Ashmore, Lynne E. Avery, Kenneth C. Bezanilla, Pilar M. Bolton. Hugh Braunlich. Denise E. Cadwalader, George, Jr. Campbell. Andrew Capobianco, James A. Child. Malcolm S. Dettbarn, Donata A. Dino, Victor H. Dodge. Michael F. Dodge, Susan A. Donovan. Christine B. Donovan. Jason P. Dooley, Kimberly A. Francisco. Manuel A. Green, Pamela A. Grossman, Howard Hadamard, John Haldiman, Susan M. Hamilton. Elizabeth R. Hines, Eric M. Hines, Kristen A. Jones, Leeland A. Kinneally. Kara J. Manheim. Francesca Marini, Michael F. McMenamin-Balano, Jonathan Minner, Eugene W. Peal, Richard W. Relyea. Timothy J. Remsen, Andrew W. Revellese, Christopher Rickles, Andrew H. Riemer, Daniel J. Roche. Theseus Roderick. Ann J. Rook, Kellyann Rossetti, Michael P. Rudin, Erik Rugh. Douglas E. Schauer, Caroline L. Schopf, Kenneth M. Shaw, Trevor P. Showalter, Carl J. Silverstein, Susan M. Sintoni, Michael A. Sohn, Marcus J. Valois, Francis X. Valyou. Patricia M. Wetzel. Ernest D. Members of the Corporation' Members of the MBL community natch the Coast Guard's nautical salute to the MBL's Centennial. Life Members Abbott, Marie, c/o Vaughn Abbott, Flyer Rd., East Hartland, CT06027 Adolph, Edward F., University of Rochester. School of Medicine and Dentistry, Rochester, NY 14642 (deceased) Beams, Harold VV., Department of Biology. University of Iowa. Iowa City, IA 52242 Behre, Kllinor, Black Mountain, NC 2871 1 (deceased) Bernheimer, Alan \\., Department of Microbiology, New York University Medical Center. 550 First Ave., New York, NY 10016 Bertholf, Lloyd M., Westminster Village #2114, 2025 E. Lincoln St., Bloomington, IL 6 1 70 1 Bishop, David \V., Department of Physiology, Medical College of Ohio, C. S. 10008. Toledo. OH 43699 Bold, Harold C., Department of Botany, University of Texas, Austin. TX 787 12 Bridgman, A. Josephine, 7 1 5 Kirk Rd., Decatur, GA 30030 Buck, John B., NIH. Laboratory of Physical Biology, Room 112. Building 6 Bethesda, MD 20892 Burbanck, Madeline P., Box I 5 I 34. Atlanta, GA 30333 Burbanck, William D., Box 1 5 1 34. Atlanta, GA 30333 Carpenter, Russell L., 60-H Lake St.. Winchester, MA 01890 Chase, Aurin, Department of Biology, Princeton University, Princeton, NJ 08544 Clark, Arnold M., 53 Wilson Rd., Woods Hole, MA 02543 Cohen, Seymour S., 10 Carrot Hill Rd.. Woods Hole, MA 02543-1206 Colwin, Arthur, 320 Woodcrest Rd.. Key Biscayne. FL 33149 Colwin, Laura Hunter, 320 Woodcrest, Key Biscayne, FL 33149 Copeland, D. E., 41 Fern Lane. Woods Hole, MA 02543 Costello, Helen M., Carolina Meadows, Villa ! 37, Chapel Hill, NC 275 14 * Including action of the 1988 Annual Meeting. Crouse, Helen, Institute of Molecular Biophysics, Florida State University. Tallahassee. FL 32306 Diller, Irene C., Rydal Park. Apartment 660. Rydal, PA 19046 (deceased 2/88) Elliott, Alfred M., 428 Lely Palm Ext., Naples, FL 33962- 8903 (deceased 1/20/88) Failla. Patricia M., 2149 Loblolly Lane. Johns Island, SC 29455 Ferguson, Frederick P., National Institute of General Medical Science, NIH. Bethesda, MD 20892 (deceased 9/27/88) Ferguson, James K. \V., 56 Clarkehaven St.. Thornhill. Ontario L4J 2B4 CANADA Fries, Erik F., 41 High Street. Woods Hole. MA 02543 Graham, Herbert, 36 Wilson Rd.. Woods Hole, MA 02543 Green, James W., 409 Grant Ave.. Highland Park. NJ 08904 Grosch, Daniel S., 1222 Duplin Road, Raleigh, NC 27607 Hamburger, Viktor, Department of Biology. Washington University. St. Louis. MO 63130 Hamilton, Howard L., Department of Biology. University of Virginia, Charlottesville, VA 22901 Hisaw, F. L., 5925 SW Plymouth Drive. Corvallis. OR 97330 Hollaender, Alexander, Council for Research Planning. 1717 Massachusetts Ave. NW. Washington, DC 20036 Humes, Arthur G., Marine Biological Laboratory, Woods Hole. MA 02543 Johnson, Frank II., Department of Biology. Princeton University, Princeton, NJ 08540 Kaan, Helen V\ '., Royal Megansett Nursing Home, Room 205, P. O. Box 408. N. Falmouth, MA 02556 Karush, Fred, Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104- 6076 Kille, Frank R., 1 1 1 1 S. Lakemont Ave. #444, Winter Park. FL32792 52 Regular Members 53 Kingsbury, John M., Department of Plant Biology, Cornell University. Ithaca, NY 14853 kleinholz, Lewis, Department of Biology, Reed College. Portland, OR 97202 Laderman, Ezra, P. O. Box 689, 18 Agassiz Road, Woods Hole, MA 02543 Lauffer, Max A., Department of Biophysics, University of Pittsburgh, Pittsburgh, PA 1 5260 LeFevre, Paul G., 1 5 Agassiz Road, Woods Hole, MA 02543 Levine, Rachmiel, 2024 Canyon Rd., Arcadia, CA 91006 Lochhead, John H., 49 Woodlawn Rd.. London SW6 6PS. England. LI. K. Lynn, W. Gardner, Department of Biology. Catholic University of America, Washington. DC 2001 7 Magruder, Samuel R., 270 Cedar Lane, Paducah, KY 42001 Manwell, Reginald D., Syracuse University, Lyman Hall, Syracuse, NY 13210 Miller, James A., 307 Shorewood Drive. E. Falmouth, MA 02536 Milne, Lorus J., Department of Zoology. University of New Hampshire, Durham. NH 03824 (deceased) Moore, John A., Department of Biology, University of California. Riverside, CA 9252 1 Moul, E. T., Woodbriar, 339 Gifford St., Falmouth, MA 02540 Naee, Paul F., P. O. Box 529. Cutchogue. NY 1 1935 Page, Irving H., Box 5 1 6. Hyannisport. MA 02647 Pollister, A. W., 8 Euclid Ave.. Belle Mead, NJ 08502 Prosser, C. Ladd, Department of Physiology and Biophysics, Burrill Hall 524, University of Illinois. Urbana. IL 61801 Provasoli, Luigi, Via Stazione 43. 21025 Comerio(VA), ITALY Prytz, Margaret McDonald, Address unknown Renn, Charles E., Route 2, Hempstead, MD 21074 Richards, A. Glenn, 942 Cromwell Ave.. St. Paul, MN 55 1 14 Richards, Oscar W., Route 1, Box 79F. Oakland, OR 97462 Rockstein, Morris, 600 Biltmore Way, Apt. 805. Coral Gables. FL 33 134 Ronkin, Raphael R., 32 1 2 McKinley St., NW, Washington. DC 200 15 Sanders, Howard, Woods Hole Oceanographic Institution, Woods Hole. MA 02543 Scharrer, Berta, Department of Anatomy. Albert Einstein College of Medicine. 1300 Morris Park Avenue, Bronx, NY 10461 Schlesinger, R. Walter, University of Medicine and Dentistry of New Jersey, Department of Molecular Genetics and Microbiology, Robert Wood Johnson Medical School, Piscataway. NJ 08854-5635 Schmitt, F. O., Room 1 6-5 1 2, Massachusetts Institute of Technology, Cambridge. MA 02 1 39 Scott, Allan C., 1 Nudd St.. Waterville. ME 04901 Shemin, David, 33 Lawrence Farm Rd.. Woods Hole, MA 02543 Smith, Homer P., 8 Quissett Ave.. Woods Hole. MA 02543 Smith, Paul F., P. O. Box 264. Woods Hole, MA 02543 Sonnenblick, B. P., 91 Chestnut St.. Millburn. NJ 07041 Steinhardt, Jacinto, 1 508 Spruce St.. Berkeley. CA 94709 Stunkard, Horace W., American Museum of Natural History. Central Park West at 79th St., New York, NY 10024 Taylor, Robert E., 20 Harbor Hill Rd., Woods Hole. MA 02543 Taylor, W. Randolph, The Herbarium, North University Bldg.. University of Michigan. Ann Arbor, MI 48109 Taylor, W. Rowland, 152 Cedar Park Road, Annapolis. MD 21401 TeWinkel, Lois E., 4 Sanderson Ave., Northampton, MA 01060 Trager, William, The Rockefeller University, 1230 York Ave., New York. NY 10021 W aid, George, 2 1 Lakeview Ave., Cambridge, MA 02 1 38 Waterman, T. H., Yale University. Biology Department. Box 6666, New Haven, CT 065 1 1 Weiss, Paul A., Address unknown Wichterman, Ralph, 3 1 Buzzards Bay Ave., Woods Hole. MA 02543 Wiercinski, Floyd J., Department of Biology. Northeastern Illinois University, Chicago. IL 60625 Wilber, Charles G., Department of Zoology. Colorado State University. Fort Collins, CO 80523 Young, D. B., 1 1 37 Main St., N. Hanover. MA 02339 Zinn, Donald J., P. O. Box 589, Falmouth, MA 02541 Zorzoli, Anita, 18 Wilbur Blvd.. Poughkeepsie, NY 12603 Zweifach, Benjamin W., 88 1 1 Nottingham Place, La Jolla, CA 92037 Regular Members Abt, Donald A., University of Pennsylvania School of Veterinary Medicine, 3800 Spruce Street, Philadelphia, PA 19104-6044 Ache, Barry W., Whitney Marine Laboratory, 9505 Ocean Shore Blvd.. St. Augustine, FL 32086 Acheson, George H., 25 Quissett Ave.. Woods Hole. MA 02543 Adams, James A., Department of Natural Sciences, University of Maryland. Princess Anne, MD 2 1 853 Adelberg, Edward A., Provost's Office, 1 1 5 Hall of Graduate Studies, Yale University, New Haven, CT 065 10 Adelman, William J., Jr., NIH. Bldg. 9, Rm. 1 E- 1 27. Bethesda, MD 20892 Afzelius, Bjorn, Wenner-Gren Institute, University of Stockholm. Stockholm. SWEDEN Alberte, Randall S., Oceanic Biology Program. Code 1 122B, Office of Naval Research. 800 North Quincy St., Arlington. VA 222 1 7-5000 Alkon, Daniel, Laboratory of Cellular and Molecular Neurobiology. NINDDS/NIH. Bldg. 5, Rm. 435. Bethesda. MD 20892 Allen, Garland E., Department of Biology. Washington Universitv. St. Louis. MO 63 1 30 54 Annual Report Allen, Nina S., Department of Biology, Wake Forest University, Box 7325. Winston-Salem. NC 27 109 Allen, Suzanne T., Department of Medical Oncology. Boston University Medical Center, 75 E. Newton Street, Boston, MA 02 11 8-2393 Amatniek, Ernest, 4797 Boston Post Rd., Pelham Manor, NY 10803 Anderson, Everett, Department of Anatomy. LHRBB. Harvard Medical School. 45 Shattuck St., Boston, MA 02115 Anderson, J. M., 1 10 Roat St., Ithaca, NY 14850 Armet-Kibel, Christine, Biology Department, University of Massachusetts-Boston, Boston, MA 02 125 Armstrong, Clay M., Department of Physiology, Medical School, University of Pennsylvania, Philadelphia, PA 19174 Armstrong, Peter B., Department of Zoology, University of California, Davis. CA 956 1 6 Arnold, John M., Pacific Biomedical Research Center. 209 Snyder Hall, 2538 The Mall, Honolulu, HI 96822 Arnold, William A., 102 Balsam Rd.. Oak Ridge, TN 37830 Ashton, Robert VV., Gaston Snow Beekman and Bogue. 14 Wall St., Suite 1600 New York. NY 10005 Atema, Jelle, Marine Biological Laboratory. Woods Hole, MA 02543 Atwood, Kimball C., P. O. Box 673, Woods Hole. MA 02543 Augustine, George J., Section of Neurobiology. Department of Biological Sciences, University of Southern California. Los Angeles, CA 90089-037 1 Austin, Mary L., 506 1/2 N. Indiana Ave., Bloomington, IN 47401 Ayers, Donald E., Marine Biological Laboratory, Woods Hole, MA 02543 Bacon, Robert, P. O. Box 723. Woods Hole, MA 02543 Baker, Robert G., New York University Medical Center, 550 First Ave., New York, NY 10016 Baldwin, Thomas O., Department ot Biochemistry and Biophysics, Texas A&M University, College Station, TX 77843 Bang, Betsy, 76 F. R. Lillie Rd.. Woods Hole, MA 02543 Barlow, Robert B., Jr., Institute for Sensory Research, Syracuse University. Merrill Lane, Syracuse. NY 1 32 10 Barry, Daniel I., Department of Physical Medicine and Rehabilitation. ID204. Llniversity of Michigan Hospital, Ann Arbor. MI 48 109-0042 Barry, Susan R., Department of Physical Medicine and Rehabilitation, ID204, University of Michigan Hospital, Ann Arbor. MI 48 109-0042 Bartell, Clelmer K., 2000 Lake Shore Drive, New Orleans, LA 70122 Bartlett, James H.. Department of Physics. Box 870324, University of Alabama. Tuscaloosa. AL 35487-0324 Bass, Andrew II., Seely Mudd Hall, Department of Neurobiology, Cornell University, Ithaca, NY 14853 Battelle, Barbara-Anne, Whitney Laboratory, 9505 Ocean Shore Blvd., St. Augustine, FL 32086 Bauer, G. Eric, Department of Anatomy, University of Minnesota, Minneapolis, MN 55455 Beauge, Luis Alberto, Instituto de Investigacion Medica, Casilla de Correo 389. 5000 Cordoba. ARGENTINA Beck, L. V., School ot Experimental Medicine, Department of Pharmacology. Indiana University. Bloomington. IN 47401 Begenisich, Ted, Department of Physiology, University of Rochester, Rochester. NY 1 4642 Begg, David A., LHRRB. Harvard Medical School, 45 Shattuck St.. Boston. MA 02 1 1 5 Bell, Eugene, Organogenesis. Inc., 83 Rogers St., Cambridge, MA 02 142 Benjamin, Thomas L., Department of Pathology. Harvard Medical School, 25 Shattuck St.. Boston, MA 02 1 1 5 Bennett, M. V. L., Albert Einstein College of Medicine, Department of Neuroscience, 1 300 Morris Park Ave.. Bronx. NY 10461 Bennett, Miriam F., Department of Biology, Colby College, Waterville, ME 04901 Berg, Carl J., Jr., Bureau of Marine Research, 1 3365 Overseas Highway, Marathon, FL 33050 Berne, Robert M., University of Virginia, School of Medicine, Charlottesville, VA 22908 Bezanilla, Erancisco, Department of Physiology. Llniversity of California. Los Angeles. CA 90052 Biggers, John D., Department of Physiology, Harvard Medical School, Boston, MA 02 1 1 5 Bishop, Stephen H., Department of Zoology, Iowa State University, Ames, IA 50010 Blaustein, Mordecai P., Department of Physiology, School of Medicine, Llniversity of Maryland. 655 W. Baltimore Street, Baltimore, MD 2 1 20 1 Bloom, Kerry S., Department of Biology, University of North Carolina, Chapel Hill, NC 275 14 Bodian, David, Department of Otolaryngology, 1721 Madison, Johns Hopkins University, Baltimore. MD 2 1 205 Bodznick, David A., Department of Biology. Wesleyan University, Middletown, CT 06457 Boettiger, Edward G., 29 Juniper Point, Woods Hole, MA 02543 Boolootian, Richard A., Science Software Systems. Inc., 3576 WoodcliffRd., Sherman Oaks, CA 9 1403 Borei, Hans G., Long Cove, Stanley Point Road, Minturn, ME 046 5 9 Borgese, Thomas A., Department of Biology, Lehman College, CUNY. Bronx. NY 10468 Borisy, Gary G., Laboratory of Molecular Biology, University of Wisconsin, Madison, WI 53706 Borst, David W., Jr., Department of Biological Sciences, Illinois State University, Normal. IL 6 1 76 1 Bosch, Herman F., 1 7 Damon Drive, Falmouth, MA 02540 Bowles, Francis P., P. O. Box 674. Woods Hole. MA 02543 Boyer, Barbara C., Department of Biology, Union College, Schenectady. NY 12308 Brandhorst, Bruce P., Biology Department, McGill Llniversity, 1205 Avenue Dr. Pentield, Montreal, P. Q. H3A IB I.CANADA Brehm, Paul, Department of Physiology, Tufts Medical School, Boston, MA 02 1 1 1 Brinley, E. J., Neurological Disorders Program, NINCDS, 812 Federal Building, Bethesda, MD 20892 Regular Members 55 Brown, Joel E., Department of Ophthalmology, Box 8096 Sciences Center, Washington University, 660 S. Euclid Ave., St. Louis, MO 63 110 Brown, Stephen C., Department of Biological Sciences, SUN Y, Albany, NY 12222 Burd, Gail Deerin, Department of Molecular and Cell Biology, University of Arizona, Tucson, AZ 8572 1 Burdick, Carolyn J., Department of Biology, Brooklyn College, Brooklyn, NY 1 1210 Burger, Max, Department of Biochemistry, Biocenter. Klingelbergstrasse 70, CH-4056 Basel, SWITZERLAND Burky, Albert, Department of Biology, University of Dayton, Dayton, OH 45469 Burstyn, Harold Lewis, Melvin and Melvin, 700 Merchants Bank Bldg.. Syracuse, NY 1 3224 Bursztajn, Sherry, Neurology Department, Program in Neuroscience, Baylor College of Medicine, Houston. TX 77030 Bush, Louise, 7 Snapper Lane, Falmouth. MA 02540 Calabrese, Ronald L., Department of Biology, Emory University, 1555 Pierce Drive, Atlanta. GA 30322 Candelas, Graciela C., Department of Biology. University of Puerto Rico. Rio Piedras. PR 0093 1 Carew, Thomas J., Department of Psychology. Yale University. P. O. Box 1 1A, Yale Station, New Haven, CT 06520 Cariello, Lucio, Stazione Zoologica, Villa Comunale, 80120 Naples, ITALY Carlson, Francis D., Department of Biophysics, Johns Hopkins University, Baltimore, MD 21218 Carriere, Rita M., Department of Anatomy and Cell Biology, Box 5, SUNY Health Science Center, 450 Clarkson Ave., Brooklyn, NY 11203 Case, James, Office of Research Development, Cheadle Hall, University of California, Santa Barbara, CA 93 106 Cassidy, Rev. J. D., Pope John Center, 1 86 Forbes Rd., Braintree, MA02184 Cebra, John J., Department of Biology, Leidy Labs. G-6. University of Pennsylvania, Philadelphia, PA 19174 Chaet, Alfred B., University of West Florida, Pensacola, FL 32504 Chambers, Edward L., Department of Physiology and Biophysics, University of Miami, School of Medicine, P. O. Box 01 6430, Miami, FL 33101 Chang, Donald C., Department of Physiology and Molecular Biophysics, Baylor College of Medicine. One Baylor Plaza, Houston, TX 77030 Chappell, Richard L., Department of Biological Sciences, Hunter College, Box 210, 695 Park Ave., New York, NY 10021 Charlton, Milton P., Physiology Department MSB, University of Toronto, Toronto, Ontario. M5S 1 A8 CANADA Chauncey, Howard H., 30 Falmouth St.. Wellesley Hills, MA 02181 Child, Frank M., Department of Biology, Trinity College, Hartford, CT 06 106 Chisholm, Rex L., Department of Cell Biology and Anatomy, Northwestern University Medical School, 303 E. Chicago Avenue, Chicago, IL 6061 1 Citkowitz, Elena, 410 Livingston St.. New Haven, CT0651 1 Clark, Eloise E., Vice President for Academic Affairs, Bowling Green State University, Bowling Green. OH 43403 Clark, Hays, 26 Deer Park Drive, Greenwich, CT 06830 Clark, James M., Shearson Lehman Brothers Inc.. 14 Wall St., 9th Floor, New York, NY 10005 Clark, Wallis H., Jr., Bodega Marine Laboratory, P. O. Box 247, Bodega Bay, CA 94923 Claude, Philippa, Primate Center, Capitol Court, Madison, WI 53706 Clay, John R., Laboratory of Biophysics, NIH, Building 9, Room I E- 1 27, Bethesda. MD 20892 Clowes, George H. A., Jr., The Cancer Research Institute, 194 Pilgrim Rd.. Boston, MA 02215 (deceased 9/10/88) Clutter, Mary, Office of the Director, Room 518, National Science Foundation, Washington, DC 20550 Cobb, Jewel Plummer, California State University, Fullerton, CA 92634 Cohen, Adolph I., Department of Ophthalmology, School of Medicine, Washington University, 660 S. Euclid Ave., St. Louis, MO 63 110 Cohen, Carolyn, Rosenstiel Basic Medical Sciences Research Center, Brandeis University, Waltham. MA 02254 Cohen, Lawrence B., Department of Physiology. Yale University School of Medicine, B-106 SHM, P. O. Box 3333, New Haven, CT 065 10-8026 Cohen, Leonard A., 279 King St.. Chappaqua, NY 105 14 Cohen, Maynard, Department of Neurological Sciences. Rush Medical College, 600 South Paulina, Chicago, IL 60612 Cohen, Rochelle S., Department of Anatomy, University of Illinois, 808 W. Wood Street, Chicago, IL 60612 Cohen, William D., Department of Biological Sciences. Hunter College, 695 Park Ave., New York, NY 10021 Cole, Jonathan J., Institute for Ecosystems Studies, Cary Arboretum, Millbrook. NY 12545 (resigned 3/7/88) Coleman, Annette W., Division of Biology and Medicine, Brown University. Providence, RI 02912 Collier, Jack R., Department of Biology, Brooklyn College, Brooklyn, NY 11210 Collier, Marjorie McCann, Biology Department, Saint Peter's College, Kennedy Boulevard. Jersey City, NJ 07306 Cook, Joseph A., The Edna McConnell Clark Foundation, 250 Park Ave.. New York. NY 10017 Cooperstein, S. J., LIniversity of Connecticut. School of Medicine, Farmington Ave., Farmington, CT 06032 Corliss, John O., P. O. Box 53008, Albuquerque, NM 87 1 53 Cornell, Neal W., Marine Biological Laboratory, Woods Hole, MA 02543 Cornwall, Melvin C.. Jr., Department of Physiology L7 1 4. Boston University School of Medicine, 80 E. Concord St.. Boston. MA 02 118 Corson, David Wesley, Jr., 1034 Plantation Lane. Mt. Pleasant, SC 29464 Corwin, Jeffrey T., Department of Otolaryngology, University of Virginia Medical Center. Box 430, Charlottesville, VA 22908 56 Annual Report Costello, Walter J., College of Medicine, Ohio University. Athens, OH 45 701 Couch, Ernest F., Department of Biology, Texas Christian University, Fort Worth, TX 76129 Cremer-Bartels, Gertrud, Universitats Augenklinik, 44 Munster. WEST GERMANY Crow, Terry J., Department of Physiology, University of Pittsburgh, School of Medicine, Pittsburgh, PA 25261 Crowell, Sears, Department of Biology, Indiana University, Bloomington, IN 47405 Crowther, Robert, Marine Biological Laboratory, Woods Hole, MA 02543 Currier, David L., P. O. Box 2476, Vineyard Haven, MA 02568 Daignault, Alexander T., 280 Beacon St., Boston, MA 02 1 1 6 Dan, Katsuma, Tokyo Metropolitan Union, Meguro-ku, Tokyo. JAPAN D'Avanzo, Charlene, School of Natural Science, Hampshire College, Amherst, MA 01002 David, John R., Seeley G. Mudd Building. Room 504. Harvard Medical School, 250 Longwood Ave., Boston, MA 02 1 1 5 Davidson, Eric H., Division of Biology, 156-29. California Institute of Technology, Pasadena, CA 91125 Davis, Bernard D., Bacterial Physiology Unit, Harvard Medical School. Boston. MA 02 1 1 5 Davis, Joel P., Seapuit, Inc., P. O. Box G. Osterville, MA 02655 Daw, Nigel \V., 78 Aberdeen Place. Clayton, MO 63 105 DeGroof, Robert C., E. R. Squibb & Sons, P. O. Box 4000, Princeton, NJ 08543-4000 DeHaan, Robert L., Department of Anatomy and Cell Biology. Emory University School of Medicine. Atlanta, GA 30322 DeLanney, Louis E., Institute for Medical Research. 2260 Clove Drive, San Jose, CA 95 1 28 DePhillips, Henry A., Jr., Department of Chemistry, Trinity College, Hartford, CT06106 DeTerra, Noel, 2 1 5 East 1 5th St., New York, NY 10003 Dettbarn, Wolf-Dietrich, Department of Pharmacology, School of Medicine, Vanderbilt University, Nashville, TN 37127 DeWeer, Paul J., Department of Physiology. School of Medicine, Washington University, St. Louis, MO 631 10 Dixon, Keith E., School of Biological Sciences, Flinders University, Bedford Park, South Australia, AUSTRALIA Donelson, John E., Department of Biochemistry, University of Iowa. Iowa City, IA 52242 Dowdall, Michael J., Department of Zoology, School of Biological Sciences, University of Nottingham, LIniversity Park, Nottingham N672 UH, England, UK Dowling, John E., The Biological Laboratories, Harvard University, 16 Divinity St., Cambridge, MA 02 138 DuBois, Arthur Brooks, John B. Pierce Foundation Laboratory. 290 Congress Ave.. New Haven. CT06519 Dudley, Patricia L., Department of Biological Sciences, Barnard College, Columbia University, New York, NY 10027 Duncan, Thomas K., Department of Environmental Sciences. Nichols College, Dudley. MA 01 570 Dunham, Philip B., Department of Biology. Syracuse University. Syracuse. NY 13244 Dunlap, kathleen, Department of Physiology, Tufts University Medical School, Boston, MA 02 1 1 1 Ebert, James D., Office of the Director. Chesapeake Bay Institute. The Johns Hopkins University. Suite 340. The Rotunda, 77 1 West 40th St.. Baltimore. MD 2 1 2 1 1 Eckberg, William R., Department of Zoology, Howard University. Washington. DC 20059 Edds, Kenneth T., Department of Anatomical Sciences, SUN Y. Buffalo. NY 14214 Eder, Howard A., Albert Einstein College of Medicine. 1300 Morris Park Ave.. Bronx. NY 1046 1 Edstrom, Joan E., 25 1 5 Milton Hills Drive, Charlottesville, VA 22901 Edwards, Charles, University of Southern Florida College of Medicine, MDC Box 40, 12901 Bruce B. Downs Blvd., Tampa, FL 336 12 Egyud, Laszlo G., 18 Sky view. Newton, MA 02150 Ehrenstein, Gerald, NIH, Bethesda. MD 20892 Ehrlich, Barbara E., Division of Cardiology. University of Connecticut Health Center, Farmington, CT 06032 Eisen, Arthur Z., Division of Dermatology, Washington University, St. Louis, MO 631 10 Eisenman, George, Department of Physiology, University of California Medical School, Los Angeles, CA 90024 Elder, Hugh Young, Institute of Physiology, University of Glasgow, Glasgow, Scotland, UKG12 8QQ Elliott, Gerald F., The Open University Research Unit, Foxcombe Hall, Berkeley Rd., Boars Hill, Oxford, England. UK Englund, Paul T., Department of Biological Chemistry. Johns Hopkins School of Medicine, Baltimore, MD 21205 Epel, David, Hopkins Marine Station. Pacific Grove. CA 93950 Epstein, Herman T., Department of Biology, Brandeis University. Waltham, MA 02254 Epstein, Ray L., Marine Biological Laboratory, Woods Hole, MA 02 543 Erulkar, Solomon D., 318 Kent Rd.. Bala Cynwyd. PA 19004 Essner, Edward S., Kresge Eye Institute. Wayne State University. 540 E. Canfield Ave., Detroit, Ml 48201 Farb, David H., SUNY Health Science Center. Brooklyn. NY 11203 Farmanfarmaian, A., Department of Biological Sciences. Nelson Biological Laboratory. Rutgers University. Piscataway, NJ 08854 Fein, Alan, Physiology Department. University of Connecticut Health Center. Farmington, CT 06032 Feinman, Richard D., Box 8, Department of Biochemistry, SUNY Health Science Center, Brooklyn. NY 1 1203 Feldman, Susan C., Department of Anatomy, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, 100 Bergen St., Newark, NJ 07 103 Fessenden, Jane, Marine Biological Laboratory. Woods Hole. MA 02543 Regular Members 57 Festoff, Barry W., Neurology Service (127). Veterans Administration Medical Center, 4801 Linwood Blvd., Kansas City, MO 64 128 Fink, Rachel D., Clapp Biology Laboratory, Mount Holyoke College. South Hadley. MA 01075 Finkelstein, Alan, Albert Einstein College of Medicine, 1 300 Morris Park Ave., Bronx, NY 10461 Fischbach, Gerald, Department of Anatomy and Neurobiology, Washington University School of Medicine, St. Louis, MO 63 110 Fischman, Donald A., Department of Cell Biology and Anatomy, Cornell LIniversity Medical College, 1 300 York Ave.. New York, NY 10021 !• is h 111:111, Harvey M., Department of Physiology, University of Texas Medical Branch. Galveston, TX 77550 Flanagan, Dennis, 12 Gay St., New York, NY 10014 Fox, Maurice S., Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02 1 38 Frank, Peter \\ '., Department of Biology, University of Oregon. Eugene. OR 97403 Franzini-Armstrong, Clara, Department of Biology G-5, School of Medicine, University of Pennsylvania, Philadelphia, PA 19174 Frazier, Donald T., Department of Physiology, University of Kentucky Medical Center. Lexington, KY 40536 Freeman, Gary L., Department of Zoology, University of Texas, Austin, TX 78172 (resigned 3/31/88) Freinkel, Norbert, Center for Endocrinology, Metabolism & Nutrition, Northwestern University Medical School, 303 E. Chicago Avenue, Chicago, IL 606 1 1 French, Robert J., Department of Medical Physiology, University of Calgary. 3330 Hospital Dr., NW, Calgary, Alberta, T2N 4N 1 , CANADA Freygang, Walter J., Jr., 6247 29th St., NW, Washington, DC 20015 Fry, Brian, Marine Biological Laboratory, Woods Hole, MA 02543 Fukui, Yoshio, Department of Cell Biology and Anatomy, Northwestern University Medical School, Chicago. IL 60201 Fulton, Chandler M., Department of Biology, Brandeis University, Waltham, MA 02 1 54 Furshpan, Edwin J., Department of Neurophysiology, Harvard Medical School, Boston, MA 021 15 Fuseler, John W., Department of Biology, University of Southwestern Louisiana, Lafayette, LA 70504 Futrelle, Robert P., College of Computer Science, Northeastern University, 360 Huntington Avenue, Boston, MA 02 115 Fye, Paul, P. O. Box 309. Woods Hole, MA 02543 (deceased 3/11/88) Gabriel, Mordecai, Department of Biology, Brooklyn College, Brooklyn, NY 1 1210 Gadsby, David C., Laboratory of Cardiac Physiology. The Rockefeller University, 1230 York Avenue, New York, NY 10021 Gainer, Harold, Section of Functional Neurochemistry, NIH, Bldg. 36, Room 4D-20, Bethesda, MD 20892 Galatzer-Levy, Robert M., 180 N. Michigan Avenue, Chicago, I L 60601 Gall, Joseph G., Carnegie Institution. 1 1 5 West University Parkway, Baltimore, MD21210 Gallant, Paul E., Laboratory of Preclinical Studies, Bldg. 36, NIAAA/NIH, 1250 Washington Ave., Rockville, MD 20892 Gascoyne, Peter, Department of Experimental Pathology, Box 85E. University of Texas System Cancer Center, M. D. Anderson Hospital and Tumor Institute, Texas Medical Center, 6723 Bertner Avenue, Houston, TX 77030 Gelfant, Seymour, Department of Dermatology, Medical College of Georgia, Augusta, GA 30904 Gelperin, Alan, Department of Biology, Princeton University, Princeton. NJ 08540 German, James L., Ill, The New York Blood Center. 310 East 67th St., New York, NY 1002 1 Gibbs, Martin, Institute for Photobiology of Cells and Organelles, Brandeis University, Waltham, MA 02254 Giblin, Anne E., Ecosystems Center, Marine Biological Laboratory, Woods Hole. MA 02543 Gibson, A. Jane, Wing Hall, Cornell University, Ithaca, NY 14850 Gifford, Prosser, 560 N Street. SW, N705, Washington, DC 20024 Gilbert, Daniel L., NIH, Bldg. 9, Room IE- 124. Bethesda. MD 20892 Giudice, Giovanni, Via Archirafi 22, Universita di Palermo. Palermo, ITALY Glusman, Murray, New York State Psychiatric Institute, Box 70. 722 W. 168th St.. New York. NY 10032 Golden, William T., 40 Wall St.. Room 4201. New York, NY 10005 Goldman, David E., 63 Loop Rd., Falmouth, MA 02540 Goldman, Robert D., Department of Cell Biology and Anatomy, Northwestern University, 303 E. Chicago Ave., Chicago, I L 606 1 1 Goldsmith, Paul K., NIH, Bldg. 10. Room 9C-101. Bethesda. MD 20892 Goldsmith, Timothy H., Department of Biology. Yale University. New Haven, CT 065 10 Goldstein, Moise H., Jr., ECE Department, Barton Hall, Johns Hopkins University. Baltimore. MD 2 1 2 1 8 Goodman, Lesley Jean, Department of Biological Sciences. Queen Mary College, Mile End Road, London, El 4NS, England, UK Goudsmit, Esther, M., Department of Biology, Oakland University, Rochester, MI 48309 Gould, Robert Michael, Institute for Basic Research in Developmental Disabilities, 1050 Forest Hill Rd.. Staten Island, NY 10314 Gould, Stephen J., Museum of Comparative Zoology, Harvard University, Cambridge, MA 02 1 38 Govind, C. K., Zoology Department — Scarborough, University of Toronto. 1265 Military Trail, West Hill, Ontario, M 1 C 1 A4 CANADA Graf, Werner, Rockefeller University. 1230 York Ave., New York. NY 10021 Grant, Philip, 1 20 Center Drive #308, Bethesda. MD 208 14 58 Annual Report Grass, Albert M., The Grass Foundation. 77 Reservoir Rd., Quincy, MA 02 170 Grass, Ellen R., The Grass Foundation, 77 Reservoir Rd.. Quincy, MA 02 170 Grassle, Judith, Marine Biological Laboratory, Woods Hole. MA 02543 Green, Jonathan P., Department of Biology. Roosevelt University. 430 S. Michigan Avenue. Chicago. IL 60605 (resigned 2/5/88) Greenberg, Everett Peter, Department of Microbiology. Stocking Hall. Cornell University, Ithaca, NY 14853 Greenberg, Michael J., C. V. Whitney Laboratory, 9505 Ocean Shore Blvd., St. Augustine. FL 32086 Griffin, Donald R., The Rockefeller University. 1230 York Ave.. New York, NY 1002 1 Gross, Paul R., Office of the Vice President and Provost, University of Virginia, Charlottesville, VA 22906-9014 Grossman, Albert, New York LIniversity. Medical School, New York, NY 10016 Gruner, John, Department of Neurosurgery. New York University Medical Center, 550 First Ave., New York, NY 10016 Gunning, A. Robert, P. O. Box 165. Falmouth, MA 0254 1 Gwilliam, G. P., Department of Biology, Reed College, Portland. OR 97202 Hall, Linda M., Department of Molecular Genetics. Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx. NY 10461 Hall, Zack W., Department of Physiology. University of California, San Francisco, CA 94143 Halvorson, Harlyn O., Marine Biological Laboratory, Woods Hole. MA 02543 Hamlett, Nancy Virginia, Department of Biology, Swarthmore College, Swarthmore, PA 19081 Hanna, Robert B., College of Environmental Science and Forestry, SUNY. Syracuse, NY 1 32 10 Harding, Clifford V., Jr., Wayne State University School of Medicine. Department of Ophthalmology, Detroit, MI 48201 Harosi, Ferenc I., Laboratory of Sensory Physiology, Marine Biological Laboratory, Woods Hole, MA 02543 Harrigan, June F., 74 1 5 Makaa Place, Honolulu. HI 96825 Harrington, Glenn \V., Division of Cell Biology and Biophysics, 403 Biological Sciences Building, University of Missouri, Kansas City, MO 641 10 Harris, Andrew L., Department of Biophysics, Johns Hopkins University. 34th & Charles Sts., Baltimore. MD 21218 Haschemeyer, Audrey E. V., 21 Glendon Road, Woods Hole, MA 02543 Hastings, J. W., Harvard University, 1 6 Divinity Street, Cambridge. MA 02 138 Hauschka, Theodore S., RD1. Box 781, Damariscotta, ME 04543 Hayashi, Teru, 7 105 SW 1 1 2 Place. Miami. FL 33 1 73 Hayes, Raymond L., Jr., Department of Anatomy, Howard University. College of Medicine, 520 W St., NW, Washington, DC 20059 Henley, Catherine, 5225 Pooks Hill Rd.. #1 127 North, Bethesda, MD 20034 Hepler, Peter K., Department of Botany, University of Massachusetts, Amherst, MA 01003 Herndon, Walter R., University of Tennessee. Department of Botany. Knoxville. TN 37996-1 100 Hessler, Anita Y., 1360 Tourmaline Ave.. San Diego, CA 92109-1915 Heuser, John, Department of Biophysics, Washington University. School of Medicine. St. Louis, MO 631 10 Hiatt, Howard H., Brigham and Women's Hospital, 75 Francis Street. Boston, MA 02 1 1 5 Highstein, Stephen M., Department of Otolaryngology, Washington University. St. Louis, MO 63110 I lildebrand, John G., Arizona Research Laboratories. ARL Division of Neurobiology, 603 Gould-Simpson Science Building, University of Arizona, Tucson, ,AZ 8572 1 Hill, Richard W., Department of Zoology, Michigan State University, E. Lansing, MI 48824 Hill, Susan D., Department of Zoology, Michigan State University. E. Lansing. Ml 48824 Hillis-Colinvaux, Llewellya, Department of Zoology. Ohio State University, 484 W. 1 2th Ave., Columbus, OH 432 1 0 Hillman, Peter, Department of Biology, Life Sciences & Neurobiology, Hebrew University, Jerusalem, ISRAEL Hinegardner, Ralph T., Division of Natural Sciences. University of California. Santa Cruz, CA 95064 Hinsch, Gertrude, W'., Department of Biology, LIniversity of South Florida. Tampa. FL 33620 Hobbie, John E., Ecosystems Center. Marine Biological Laboratory, Woods Hole, MA 02543 Hodge, Alan J., Marine Biological Laboratory, Woods Hole. MA 02543 Hoffman, Joseph, Department of Physiology, School of Medicine, Yale University, New Haven, CT 065 10 Hollyfield, Joe G., Baylor School of Medicine, Texas Medical Center, Houston, T.X 77030 I lull /man. Eric, Department of Biological Sciences, Columbia University, New York, NY 10017 Holz, George G., Jr., Department of Microbiology, SUNY, Syracuse, NY 13210 Hoskin, Erancis C. G., Department of Biology, Illinois Institute of Technology, Chicago. IL 60616 Houghton, Richard A., Ill, Woods Hole Research Center. P. O. Box 296, Woods Hole, MA 02543 Houston, Howard E., 2500 Virginia Ave.. NW. Washington. DC 20037 Howarth, Robert, Section of Ecology & Systematics, Corson Hall. Cornell University, Ithaca, NY 14853 (resigned 3/9/88) Hoy, Ronald R., Section of Neurobiology and Behavior. Cornell University, Ithaca, NY 14850 Hubbard, Ruth, 67 Gardner Road. Woods Hole, MA 02543 Hufnagel, Linda A., Department of Microbiology, University of Rhode Island, Kingston, RI 0288 I I 1 iiinmoii. \Yilliam I)., Department of Zoology, Ohio University, Athens, OH 45701 Humphreys, Susie H., 810 Waukegan Rd., Glenview. IL 60025 Regular Members 59 Humphreys, Tom D., University of Hawaii. PBRC, 41 Ahui St., Honolulu, HI 968 13 Hunter, Robert D., Department of Biological Sciences, Oakland University, Rochester. MI 48309-4401 Hunter, \V. Bruce, Box 32 1 , Lincoln Center, MA 01773 Hunziker, Herbert E., Esq., P. O. Box 547. Falmouth. MA 02541 Hurwitz, Charles, Basic Science Research Lab, Veterans Administration Hospital, Albany, NY 12208 Hurwitz, Jerard, Sloan Kettering Institute for Cancer Research, 1275 York Avenue, New York, NY 1 102 1 Huxley, Hugh E., Department of Biology, Rosenstiel Center, Brandeis University, Waltham, MA 02 1 54 Hynes, Thomas J., Jr., Meredith and Grew, Inc., 1 60 Federal Street, Boston, MA 02 1 1 0- 1 70 1 Ilan, Joseph, Department of Anatomy, Case Western Reserve University, Cleveland, OH 44106 Ingoglia, Nicholas, Department of Physiology, New Jersey Medical School, 100 Bergen St.. Newark. NJ 07 103 Inoue, Saduyki, McGill University Cancer Centre, Department of Anatomy, 3640 University St., Montreal, PQ, H3A 2B2. CANADA Inoue, Shinya, Marine Biological Laboratory, Woods Hole, MA 02543 Issadorides, Marietta, R., Department of Psychiatry, University of Athens, Monis Petraki 8, Athens, 140 GREECE Isselbacher, Kurt J., Massachusetts General Hospital, 32 Fruit Street, Boston, MA 02 114 Izzard, Colin S., Department of Biological Sciences, SUNY, Albany, NY 12222 Jacobson, Antone G., Department of Zoology, University of Texas. Austin, TX 787 12 Jaffe, Lionel, Marine Biological Laboratory, Woods Hole, MA 02543 Jahan-Parwar, Behrus, Center for Laboratories & Research, New York State Department of Health, Empire State Plaza, Albany. NY 12201 Jannasch, Holger \V., Department of Biology, Woods Hole Oceanographic Institution. Woods Hole, MA 02543 Jeffery, William R., Department of Zoology. University of Texas. Austin. TX 787 12 Jones, Meredith L., Division of Worms, Museum of Natural History, Smithsonian Institution, Washington, DC 20560 Josephson, Robert K., School of Biological Sciences, University of California, Irvine, CA 92664 Rabat, E. A., Department of Microbiology, College of Physicians and Surgeons Columbia University, 630 West 168th St., New York, NY 10032 Kaley, Gabor, Department of Physiology, Basic Sciences Building, New York Medical College, Valhalla, NY 10595 kaltenbach, Jane, Department of Biological Sciences, Mount Holyoke College, South Hadley. MA 01075 Kaminer, Benjamin, Department of Physiology, School of Medicine, Boston University, 80 East Concord St., Boston, MA02118 Kammer, Ann E., Department of Zoology, Arizona State University, Tempe, AZ 85281 Kane, Robert E., PBRC, University of Hawaii, 41 Ahui St.. Honolulu, HI 968 13 Kaneshiro, Edna S., Department of Biological Sciences, University of Cincinnati. Cincinnati. OH 4522 1 Kao, Chien-yuan, Department of Pharmacology, Box 29, SUNY, Downstate Medical Center, 450 Clarkson Avenue, Brooklyn, NY 1 1 203 Kaplan, Ehud, The Rockefeller University, 1230 York Ave., New York. NY 10021 Karakashian, Stephen J., Apt. 16-F, 165 West 91st St., New York, NY 10024 Karlin, Arthur, Department of Biochemistry and Neurology, Columbia University. 630 West 168th St., New York, NY 10032 Katz, George M., Fundamental and Experimental Research, Merck Sharpe and Dohme, Rahway, NJ 07065 Kean, Edward L., Department of Ophthalmology and Biochemistry, Case Western Reserve University, Cleveland, OH 44101 (resigned 10/27/88) Kelley, Darcy Brisbane, Department of Biological Sciences. 1018 Fairchild. Columbia University, New York, NY 10032 Kelly, Robert E., Department of Anatomy. College of Medicine, University of Illinois, P. O. Box 6998, Chicago, IL 60680 Kemp, Norman E., Department of Biology, University of Michigan, Ann Arbor, MI 48104 Kendall, John P., Faneuil Hall Associates, Suite 1620, One Boston Place. Boston. MA 02108 Kendall, Richard E., 26 Green Harbor Rd., East Falmouth, MA 02536 Keynan, Alexander, Hebrew University, Jerusalem, ISRAEL Kiehart, Daniel P., Department of Cellular and Developmental Biology, Harvard University, 16 Divinity Street. Cambridge, MA 02 1 38 Klein, Morton, Department of Microbiology, Temple University, Philadelphia, PA 19103 Klotz, Irving M., Department of Chemistry, Northwestern University, Evanston, IL 60201 Koide, Samuel S., Population Council, The Rockefeller University, 1230 York Avenue, New York, NY 10021 Konigsberg, Irwin R., Department of Biology, Gilmer Hall, University of Virginia, Charlottesville, VA 22903 Kornberg, Sir Hans, The Master's Lodge, Christ's College, Cambridge CB2 3BU, England, UK Kosower, Edward M., Ramat-Aviv, Tel Aviv, 69978 ISRAEL Krahl, M. E., 2783 W. Casas Circle, Tucson. AZ 85741 Krane, Stephen M., Massachusetts General Hospital, Fruit Street. Boston, MA 021 14 Krassner, Stuart M., Department of Developmental and Cell Biology, University of California, Irvine, CA 927 1 7 Krauss, Robert, FASEB, 9650 Rockville Pike, Bethesda, MD 20814 Kravitz, Edward A., Department of Neurobiology, Harvard Medical School, 25 Shattuck St.. Boston. MA 021 15 Kriebel, Mahlon E., Department of Physiology, B.S.B., Upstate Medical Center, 766 Irving Ave., Syracuse, NY 13210 (ill Annual Report kristan, William B., Jr., Department of Biology B-022. University of California San Diego, La Jolla, CA 92093 Kiilins. William J., Hospital for Sick Children, Department of Biochemistry Research, Toronto, Ontario M5G 1X8, CANADA Kusano, Kiyoshi, NIH, Bldg. 36. Room 4D-20. Bethesda. MD 20892 Ku/iriaii. Alan M., Marine Biological Laboratory, Woods Hole. MA 02543 Laderman, Aimlee, P. O. Box 689. Woods Hole. MA 02543 LaMarche, Paul H., Eastern Maine Medical Center, 489 State St.. Bangor. ME 04401 I and is. Dennis M. D., Department of Neurology, Case Western Reserve School of Medicine. Cleveland. OH 44 106 I .and is. Story C., Department of Pharmacology, Case Western Reserve Llniversity School of Medicine, Cleveland. OH 44106 Landowne. David, Department of Physiology, P. O. Box 016430, University of Miami School of Medicine, Miami, FL33101 Langford, George M., Department of Physiology, Medical Sciences Research Wing 206H, University of North Carolina, Chapel Hill, NC 27599-7545 Lasek, Raymond J., Case Western Reserve University, Department of Anatomy. Cleveland, OH 44106 Laster, Leonard, University of Massachusetts Medical School. 55 Lake Avenue, North, Worcester, MA 01655 I.aufer, Hans, Biological Science, Molecular and Cell Biology, Group U- 1 25, University of Connecticut, Storrs. CT 06268 Lazarow, Paul B., The Rockefeller University, 1230 York Avenue. New York, NY 10021 Lazarus, Maurice, Federated Department Stores. Inc., 50 Cornhill, Boston, MA 02 108 Leadbetter, Kdward R., Department of Molecular and Cell Biology. U- 1 3 1 . Llniversity ot Connecticut, Storrs, CT 06268" Lederberg, Joshua, The Rockefeller Llniversity, 1 230 York Ave., New York, NY 10021 Lederhendler, Izja I., Laboratory of Cellular and Molecular Neurobiology. NINCDS/NIH, Park 5 Building. Room 435, Bethesda, MD 20892 Lee, John J., Department of Biology, City College of CUNY, Convent Ave. and 1 38th St.. New York. NY 1003 1 Lehy, Donald B., Marine Biological Laboratory, Woods Hole, MA 02543 Leibovitz, Louis, 3 Kettle Hole Road, Woods Hole, MA 02543 I .eighton, Joseph, 1201 Waverly Rd., Gladwyne, PA 19035 Leighton, Stephen, NIH, Bldg. I 3 3W I 3, Bethesda. MD 20892 Leinwand, Leslie Ann, Department of Microbiology and Immunology. Albert Einstein College of Medicine, 1300 Morris Park Ave.. Bronx, NY 10461 Lerman, Sidney, Eye Research Lab, Room 41, New York Medical College. 100 Grasslands Ave., Valhalla, NY 10595 Lerner, Aaron B., Yale University, School of Medicine. New Haven, CT 065 10 Lester, Henry A., 156-29 California Institute of Technology, Pasadena. CA 9 1 125 Levin, Jack, Clinical Pathology Service, VA Hospital, 1 1 3A, 4 1 50 Clement St.. San Francisco. CA 94 1 2 1 Levinthal, Cyrus, Department of Biological Sciences, Columbia University. Broadway and I 16th Street. New York, NY 10026 Levitan, Herbert, Department of Zoology, University of Maryland. College Park, MD 20742 Linck, Richard W., Department of Anatomy. Jackson Hall, University of Minnesota, 32 1 Church Street. S. E.. Minneapolis, MN 55455 Lipicky, Raymond J.. Department ofCardio-Renal/HFD 1 10, FDA Bureau of Drugs. Rm. 16B-45. 5600 Fishers Lane, Rockville. MD 20857 Lisman, John E., Department of Biology, Brandeis University. Waltham, MA 02254 Liuzzi, Anthony, 55 Fay Rd., Box 184, Woods Hole, MA 02543 Llinas, Rodolfo R., Department of Physiology and Biophysics, New York University Medical Center, 550 First Ave.. New York. NY 10016 Loew, Franklin M., Tufts University School of Veterinary Medicine, 200 Westboro Rd.. N. Grafton, MA 01536 Loewenstein, Werner R., Department of Physiology and Biophysics, Llniversity of Miami. P.O. Box 016430. Miami. FL33101 Loewus, Frank A., Institute of Biological Chemistry, Washington State University, Pullman. WA 99164 Loftfield, Robert B., Department of Chemistry, School of Medicine. University of New Mexico, 900 Stanford, NE, Albuquerque, NM 871 31 London, Irving M., Massachusetts Institute of Technology, E-25-55 1, Cambridge, MA 02139 Longo, Frank J., Department of Anatomy, University of Iowa. Iowa City, IA 52442 Lorand, Laszlo, Department of Biochemistry and Molecular Biology, Northwestern University, Evanston, IL 60208 Luckenbill-Edds, Louise, 155 Columbia Ave., Athens, OH 45701 I in ia. Salvador E., 48 Peacock Farm Rd., Lexington. MA 02173 Macagno, Eduardo R., 1003B Fairchild. Department of Biosciences. Columbia Llniversity, New York, NY 10022 MacNichol, E. E., Jr., Department of Physiology. Boston University School of Medicine. 80 E. Concord St., Boston, MA 02 1 18 Maglott-Dufh'eld, Donna R. S., American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852- 1776 Maienschein, Jane Ann, Department of Philosophy, Arizona State University, Tempe, AZ 85287-2004 Mainer, Robert, The Boston Company. One Boston Place, 5-D, Boston, MA 02 108 Malbon, C'raig C'urtis, Department of Pharmacology, Health Sciences Center, SLINY, Stony Brook, NY 1 1 794-865 1 Malkiel, Saul, Allergic Diseases. Inc.. 130 Lincoln St., Worcester, MA 01605 Regular Members 61 MM ii;ihs. Richard S., Department of Biological Sciences, Indiana University — Purdue University at Fort Wayne, Fort Wayne, IN 46805 IMangum, Charlotte P., Department of Biology. College of William and Man. Williamsburg, VA 23185 Margulis, Lynn, Botany Department. University of Massachusetts, Morrill Science Center. Amherst, MA 01003 Marinucci, Andrew C, 102 Nancy Drive. Mercerville, NJ 08619 Marsh, Julian B., Department of Biochemistry and Physiology. Medical College of Pennsylvania, 3300 Henry Ave., Philadelphia. PA 19129 Martin, Lowell V., Marine Biological Laboratory, Woods Hole, MA 02543 Martinez-Palomo, Adolfo, Seccion de Patologia Experimental, Cinvesav-ipn, 07000 Mexico. D.F. A. P., 140740. MEXICO Maser, Morton, P. O. Box EM. Woods Hole Education Assoc.. Woods Hole, MA 02543 Mastroianni, Luigi, Jr., Department of Obstetrics and Gynecology. Hospital of the University of Pennsylvania. 106 Dulles, Philadelphia. PA 19174 Mathews, Rita W., Department of Medicine. NYU Medical Center, 550 First Ave.. New York, NY 10016 Matteson, Donald R., Department of Physiology, G4, School of Medicine. University ot Pennsylvania, Philadelphia, PA 19104 Mautner, Henry G., Department of Biochemistry. Tufts University. 136 Harrison Ave., Boston, MA 02 1 1 1 Mauzerall, David, The Rockefeller University, 1230 York Ave., New York. NY 10021 Mazia, Daniel, Hopkins Marine Station. Pacific Grove, CA 93950 Mazzella, Lucia, Laboratorio di Ecologia del Benthos. Stazione Zoologica di Napoli, P.ta S. Pietro 80077. Ischia Porto (NA), ITALY McCann, Frances, Department of Physiology. Dartmouth Medical School, Hanover, NH 03755 McLaughlin, Jane A., Marine Biological Laboratory. Woods Hole. MA 02543 McMahon, Robert F., Department of Biology, Box 19498, University of Texas, Arlington, TX 76019 Meedel, Thomas, Marine Biological Laboratory, Woods Hole, MA 02543 Meinertzhagen, Ian A., Department of Psychology, Life Sciences Center, Dalhousie University, Halifax, Nova Scotia B3H 45 1 . CANADA Meiss, Dennis E., 462 Soland Avenue, Hayward, CA 94541 Melillo, Jerry A., Ecosystems Center. Marine Biological Laboratory. Woods Hole. MA 02543 Mellon, Richard P., P. O. Box 1 87. Laughlintown. PA I 5655 Mellon, DeForest, Jr., Department of Biology, University of Virginia, Charlottesville. VA 22903 Metuzals, Janis, Department of Anatomy. Faculty of Medicine, University of Ottawa. Ottawa, Ontario KIN 9A9, CANADA Metz, Charles B., 7220 SW 1 24th St.. Miami, FL 33 1 56 Milkman, Roger, Department of Biology. University of Iowa, Iowa City, IA 52242 Mills, Eric L., Oceanography Dept., Dalhousie University, Halifax, Nova Scotia B3H 4J 1 . CANADA Mills, Robert, 10315 44th Avenue, W 12 H Street, Bradenton, FL 33507- 1535 Mitchell, Ralph, Pierce Hall, Harvard University, Cambridge, MA 02 1 38 Miyamoto, David M., Department of Biology. Drew University. Madison, NJ 07940 Mizell, Merle, Laboratory of Tumor Cell Biology. Tulane University, New Orleans. LA 701 18 Moore, John \\ ., Department of Physiology, Duke University Medical Center, Durham, NC 27710 Moore, Lee E., Department of Physiology and Biophysics. University of Texas Medical Branch, Galveston, TX 77550 Morin, James G., Department of Biology, University of California, Los Angeles, CA 90024 Morrell, Frank, Department of Neurological Sciences, Rush Medical Center. 1753 W. Congress Parkway. Chicago, IL 60612 Morse, M. Patricia, Marine Science Center. Northeastern University. Nahant. MA 01 908 Morse, Richard S., 193 Winding River Rd.. Wellesley. MA 02181 (deceased 7/1/88) Morse, Robert W., Box 574. N. Falmouth, MA 02556 Morse, Stephen Scott, The Rockefeller University. 1230 York Ave., Box 2, New York. NY 1002 1-6399 Moscona, Arthur A., Department of Molecular Genetics and Cell Biology, University of Chicago, 920 East 58th St.. Chicago. IL 60637 Mote, Michael L, Department of Biology. Temple LJniversity, Philadelphia. PA 19122 Mountain, Isabel, Vinson Hall #112, 6251 Old Dominion Drive, McLean, VA 22101 Mullins, Lorin J., University of Maryland. School of Medicine. Baltimore, MD 2 1 20 1 Musacchia, Xavier J., Department of Physiology and Biophysics, University of Louisville School of Medicine, Louisville, KY 40292 Nabrit, S. M., 686 Beckwith St.. SW, Atlanta, GA 30314 Nadelhoffer, knute, Marine Biological Laboratory, Woods Hole. MA 02543 Naka, Ken-ichi, PHL 821, Department of Ophthalmology. NYU Medical Center. 550 First Avenue. New York, NY 10016 Nakajima, Shigehiro, University of Illinois College of Medicine at Chicago. 835 S. Wolcott Avenue, Chicago, IL 60612 Nakajima, Yasuko, LJniversity of Illinois College of Medicine at Chicago, Department of Anatomy and Cell Biology, M/C 512. Chicago. IL 606 12 Narahashi, Toshio, Department of Pharmacology. Medical Center, Northwestern University, 303 East Chicago Ave., Chicago. I L 606 11 Nasatir, Maimon, Department of Biology, University of Toledo. Toledo. OH 43606 Nelson, Leonard, Department of Physiology. Medical College of Ohio. Toledo. OH 43699 62 Annual Report Nelson, Margaret C., Section of Neurobiology and Behavior, Cornell University, Ithaca. NY 14850 Nicholls, John G., Biocenter, Klingelbergstr. 70, Basel 4056, SWITZERLAND Nicosia, Santo V., Department of Pathology, University of South Florida. College of Medicine, Box 1 1. 12901 North 30th St.. Tampa, FL 336 12 Nielsen, Jennifer B. K., Merck Sharp & Dohme Laboratories, RY 80-2 10. Rahway, NJ 07065 Noe, Bryan D., Department of Anatomy, Emory University, Atlanta, GA 30345 Obaid, Ana Lia, Department of Physiology and Pharmacy, University of Pennsylvania, 4001 Spruce St., Philadelphia, PA 19104-6003 Oertel, Donata, Department of Neurophysiology, University of Wisconsin, 28 1 Medical Science Bldg., Madison, WI 53706 O'Herron, Jonathan, 45 Swifts Lane. Darien, CT 06820 Olins, Ada L., University of Tennessee-Oak Ridge, Graduate School of Biomedical Sciences, Biology Division ORNL. P. O. Box Y, Oak Ridge, TN 37830 Olins, Donald E., University of Tennessee-Oak Ridge, Graduate School of Biomedical Sciences, Biology Division ORNL, P. O. Box Y, Oak Ridge, TN 37830 O'Melia, Anne F., 16 Evergreen Lane, Chappaqua, New York 10514 Oschman, James L., 9 George Street, Woods Hole, MA 02543 Palmer, John D., Department of Zoology. University of Massachusetts, Amherst. MA 01002 Palti, Yoram, Rappaport Institution. Technion. POB 9697, Haifa. ISRAEL Pant, Harish C, NINCDS/NIH, Bldg. 36, Room 4D-20, Bethesda, MD 20892 Pappas, George D., Department of Anatomy. College of Medicine, University of Illinois. 808 South Wood St.. Chicago. I L 606 1 2 Pardee, Arthur B., Department of Pharmacology, Harvard Medical School, Boston. MA 02 1 1 5 Pardy, Roosevelt L., School of Life Sciences, University of Nebraska, Lincoln, NE 68588 Parmentier, James L., Becton Dickinson Research Center. P. O. Box 12016, Research Triangle Park, NC 27709 Passano, Leonard M., Department of Zoology. Birge Hall. University of Wisconsin. Madison, WI 53706 Pearlman, Alan L., Department of Physiology, School of Medicine, Washington University, St. Louis, MO 631 10 Pederson, Thoru, Worcester Foundation for Experimental Biology, Shrewsbury, MA 0 1 545 Perkins, C. D., 400 Hilltop Terrace, Alexandria, VA 22301 Person, Philip, Research Testing Labs. Inc., 167 E. 2nd St.. Huntington Station, NY 1 1746 Peterson, Bruce J., Ecosystems Center. Marine Biological Laboratory, Woods Hole. MA 02543 Pethig, Ronald, School of Electronic Engineering Science, University College of N. Wales, Dean St., Bangor, Gwynedd. LL57IUT, UK Pfohl, Ronald J., Department of Zoology, Miami University, Oxford, OH 45056 Pierce, Sidney K., Jr., Department of Zoology, University of Maryland. College Park, MD 20740 Poindexter, Jeanne S., Science Division. Long Island Llniversity, Brooklyn Campus. Brooklyn. NY 1 1201 Pollard, Harvey B., NIH. N1DDKD. Bldg. 8. Rm. 401. Bethesda. MD 20892 Pollard, Thomas D., Department of Cell Biology and Anatomy, Johns Hopkins University, 725 North Wolfe St., Baltimore, MD 2 1205 Pollock, I , eland \V., Department of Zoology, Drew University, Madison, NJ 07940 (resigned 12/17/88) Poole, Alan F., P. O. Box 533. Woods Hole, MA 02543 Porter, Beverly H., 13617 Glenoble Drive, Rockville. MD 20853 Porter, Keith R., Department of Biology, Leidy Laboratories, Rm. 303, University of Pennsylvania, Philadelphia, PA 19104-6018 Porter, Mary E., Department MCD Biology, Campus Box 347. University of Colorado. Boulder. CO 80309 Potter, David. Department of Neurobiology, Harvard Medical School, Boston. MA 02 1 1 5 Potts, William T., Department of Biology, University of Lancaster, Lancaster, England, UK Pratt, Melanie M., Department of Anatomy and Cell Biology, University of Miami School of Medicine (R124). P. O. Box 016960. Miami. FL 33 101 Prendergast, Robert A., Wilmer Institute. Johns Hopkins Hospital. Baltimore. MD 2 1 205 Presley, Phillip H., Carl Zeiss. Inc., 1 Zeiss Drive, Thornwood. NY 10594 Price, Carl A., Waksman Institute of Microbiology, Rutgers University. P. O. Box 759, Piscataway, NJ 08854 Prior, David J., Department of Biological Sciences, NAU Box 5640, Northern Arizona University, Flagstaff, AZ 8601 1 Prusch, Robert D., Department of Life Sciences, Gonzaga University, Spokane, WA 99258 Przyb) Iski, Ronald J., Case Western Reserve University. Department of Anatomy. Cleveland. OH 44104 Purves, Dale, Department of Anatomy. Washington University School of Medicine. 660 S. Euclid Ave.. St. Louis, MO 63 110 Quigley, James, Department of Microbiology. Box 44, SUNY Downstate Medical Center, 450 Clarkson Ave., Brooklyn. NY 1 1 203 Rabb, Irving \V., P.O. Box 369. Boston. MA 02101 Rabin, Harvey, DuPont Biomed. Products. BRL-2. 331 Treble Cove Road. No. Billerica, MA 01862 Rahinonit/., Michael B., Marine Biological Laboratory, Woods Hole, MA 02543 Raff, Rudolf A., Department of Biology. Indiana University, Bloomington, IN 47405 Rako\vski, Robert F., Department of Physiology and Biophysics. UHS/The Chicago Medical School. 3333 Greenbay Rd., N. Chicago. IL 60064 Ramon, Fidel, Dept. de Fisiologia y Biofisca. Centre de Investigation y de Estudius Avanzados del ipn, Apurtado Postal 14-740, D.F. 07000, MEXICO Regular Members 63 Ranzi, Silvio, Sez Zoologia Sc Nat, Via Coloria 26. 1 20 1 33, Milano, ITALY Rastetter, Edward B., Ecosystems Center, Marine Biological Laboratory. Woods Hole. MA 02543 Ratner, Sarah, Department of Biochemistry, Public Health Research Institute. 455 First Ave.. New York. NY 10016 Rebhun, Lionel I., Department of Biology, Gilmer Hall. University of Virginia. Charlottesville. VA 22901 Reddan, John R., Department of Biological Sciences, Oakland University, Rochester, MI 48063 Reese, Barbara F., NINCDS/NIH. Bldg. 36. Room 3B26. Bethesda. MD 20892 Reese, Thomas S., NINCDS/NIH, Bldg. 36. Room 2A27. Bethesda, MD 20892 Reiner, John M., Department of Biochemistry, Albany Medical College of Union University. Albany. NY 12208 Reinisch, Carol L., Tufts University School of Veterinary Medicine. 203 Harrison Avenue. Boston. MA 02 1 1 5 Reuben, John P., Department of Biochemistry. Merck Sharp and Dohme. P. O. Box 2000, Rahway, NJ 07065 Reynolds, George T., Department of Physics. Jadwin Hall. Princeton University, Princeton, NJ 08540 Rice, Robert V., 30 Burnham Dr., Falmouth, MA 02540 Rich, Alexander, Department of Biology, Massachusetts Institute of Technology. Cambridge, MA 02 1 39 Rickles, Frederick R., University of Connecticut. School of Medicine, VA Hospital. Newington, CT061 1 1 Ripps, Harris, Department of Ophthalmology, University of Illinois College of Medicine, 1855 W. Taylor Street. Chicago, IL 606 1 1 Robinson, Denis M., 200 Ocean Lane Drive #908, Key Biscayne. FL33149 Rose, Birgit, Department of Physiology and Biophysics, R- 430. University of Miami School of Medicine. P. O. Box 016430, Miami. FL 33 101 Rose, S. Meryl, 32 Crosby Ln.. E. Falmouth. MA 02536 Rosenbaum, Joel L., Department of Biology. Kline Biology- Tower, Yale University, New Haven, CT 06520 Rosenberg, Philip, School of Pharmacy. Division of Pharmacology, University of Connecticut, Storrs, CT 06268 Rosenbluth, Jack, Department of Physiology, New York University School of Medicine, 550 First Ave., New York. NY 10016 Rosenbluth, Raja, 3380 West 5th Ave.. Vancouver 8, BC V6R 1R7, CANADA Roslansky, John, Box 208. Woods Hole. MA 02543 Roslansky, Priscilla F., Box 208, Woods Hole. MA 02543 Ross, William N., Department of Physiology, New York Medical College, Valhalla. NY 10595 Roth, Jay S., 1 8 Millfield Street. P. O. Box 285, Woods Hole, MA 02543 Rowland, Lewis P., Neurological Institute, 710 West 168th St., New York, NY 10032 Ruderman, Joan V'., Department of Zoology, Duke University, Durham, NC 27706 Rushforth, Norman B., Case Western Reserve LJniversity. Department of Biology, Cleveland, OH 44106 Russell-Hunter, W. D., Department of Biology. Lyman Hall 029. Svracuse Universitv. Svracuse. NY 13210 Saffo, Mary Beth, Institute of Marine Sciences, 272 Applied Sciences, University of California, Santa Cruz, CA 95064 Sager, Ruth, Dana Farber Cancer Institute. 44 Binney St.. Boston, MA 021 15 Salama, Guy, Department of Physiology, University of Pittsburgh, Pittsburgh, PA 15261 Salmon, Edward D., Department of Zoology, University of North Carolina, Chapel Hill, NC 275 1 4 Salzberg, Brian M., Department of Physiology, University of Pennsylvania, 40 10 Locust St.. Philadelphia. PA 19104- 6085 Sanborn, Richard C., 1 1 Oak Ridge Road, Teaticket, MA 02536 Sanger, Jean M., Department of Anatomy. School of Medicine. LIniversity of Pennsylvania. 36th and Hamilton Walk. Philadelphia, PA 19174' Sanger, Joseph, Department of Anatomy, School of Medicine, LIniversity of Pennsylvania. 36th and Hamilton Walk. Philadelphia, PA 19174 Sato, Eimei, Department of Animal Science, Faculty of Agriculture. Kyoto University. Kyoto 606, JAPAN (resigned 10/20/88) Sato, Hidemi, Sugashima Marine Biological Laboratory, Nagoya University, Sugashima-cho, Toba-shi. Mieken 5 1 7, JAPAN Sattelle, David B., AFRC Unit-Department of Zoology. University of Cambridge, Downing St., Cambridge CB2 3EJ. England. UK Saunders, John W., Jr., P. O. Box 381, Waquoit Station, Waquoit. MA 02536 Saz, Arthur K., Department of Immunology, Georgetown University Medical School. Washington, DC 20007 Schachman, Howard K., Department of Molecular Biology, University of California, Berkeley, CA 94720 Schatten, Gerald P., Integrated Microscopy Facility for Biomedical Research, University of Wisconsin, I 1 17 W. Johnson St.. Madison. WI 53706 Schatten, Heide, Department of Zoology. University of Wisconsin. Madison. WI 53706 Schiff, Jerome A., Institute for Photobiology of Cells and Organelles, Brandeis University, Waltham. MA 02 1 54 Schmeer, Arline C., Mercenene Cancer Research Institute, Hospital of Saint Raphael. New Haven, CT 065 1 1 Schnapp, Bruce J., Department of Physiology, Boston University Medical School, 80 East Concord Street, Boston. MA 02 118 Schneider, E. Gayle, Department of Obstetrics and Gynecology, Yale University School of Medicine. 333 Cedar St.. New Haven. CT 065 10 Schneiderman, Howard A., Monsanto Company, 800 North Lindbergh Blvd.. D1W, St. Louis, MO 63 166 Schotte, Oscar E., Department of Biology, Amherst College. Amherst, M A 0 1 002 (deceased 4/12/88) Schuel, Herbert, Department of Anatomical Sciences, SUNY. Buffalo, Buffalo, NY 14214 Schuetz, Allen W., School of Hygiene and Public Health. Johns Hopkins LIniversity, Baltimore, MD 2 1 205 64 Annual Report Schwartz, James H., Center for Neurobiology and Behavior. New York State Psychiatric Institute — Research Annex. 722 W. 168th St.. 7th Floor, New York. NY 10032 Scofield, Virginia Lee, Department of Microbiology and Immunology, UCLA School of Medicine, Los Angeles, CA 90024 Sears, Mary, P. O. Box 152, Woods Hole, MA 02543 Segal, Sheldon J., Population Division. The Rockefeller Foundation, 1 133 Avenue of the Americas. New York, NY 10036 Seliger, Howard H., Johns Hopkins University, McCollum- Pratt Institute, Baltimore, MD 2 1218 (resigned 1/31/88) Selman, Kelly, Department of Anatomy. College of Medicine, University of Florida, Gainesville, FL 32601 Senft, Joseph, Biology Department, Juniata College, Huntingdon, PA 16652 Shanklin, Douglas R., Department of Pathology, Room 584, University of Tennessee College of Medicine, Memphis, TN 38163 Shapiro, Herbert, 6025 North 13th St., Philadelphia. PA 19141 Shaver, Gaius R., Ecosystems Center. Marine Biological Laboratory. Woods Hole, MA 02543 Shaver, John R., Department of Biology, Cayey Regional Campus, University of Puerto Rico, Cayey, PR 00633 Sheelz, Michael P., Department of Cell Biology and Physiology, Washington University Medical School, 606 S. Euclid Ave., St. Louis, MO 631 H) Shepard, David C., P. O. Box 44, Woods Hole, MA 02543 Shepro, David, Department of Biology, Boston LIniversity. 2 Cummington St.. Boston. MA 022 1 5 Sher, F. Alan, Immunology and Cell Biology Section, Laboratory of Parasitic Disease. NIAID, Building 5, Room 1 14. NIH. Bethesda, MD 20892 Sheridan, William F., Biology Department, University of North Dakota, Box 8238, University Station, Grand Forks, ND 58202-8238 Sherman, I. W., Division of Life Sciences, University of California, Riverside, CA 92502 Shilo, Moshe, Department of Microbial and Molecular Biology, The Hebrew University, 91904 Jerusalem. ISRAEL Shimomura, Osamu, Marine Biological Laboratory, Woods Hole. MA 02543 Shoukimas, Jonathan J., 45 Dillingham Avenue, Falmouth, MA 02540 Siegel, Irwin M., Department of Ophthalmology, New York University Medical Center, 550 First Avenue, New York, NY 10016 Siegelman, Harold W., Department of Biology, Brookhaven National Laboratory, Upton. NY 1 1973 Silver, Rohert B., Department of Physiology. Cornell Llniversily, 822 Veterinary Research Tower, Ithaca, NY 14853-6401 Silverstein, Arthur M., Wilmer Institute, The Johns Hopkins Hospital, Baltimore, MD21205 Sjodin, Raymond A., Department of Biophysics, University of Maryland. Baltimore, MD 21201 Skinner, Dorothy M., Oak Ridge National Laboratory. P. O. Box 2009. Biology Division. Oak Ridge, TN 37830 Sloboda, Roger D., Department of Biological Sciences, Dartmouth College, Hanover, NH 03755 Sluder, Greenfield, Cell Biology Group, Worcester Foundation for Experimental Biology. 222 Maple Ave., Shrewsbury. MA 01 545 Smith, Homer P., 8 Quissett Avenue. Woods Hole. MA 02543 Smith, Michael A., J 1 Sinabung. Buntu #7. Semarang. Java. INDONESIA Smith, Ralph I., Department of Zoology, University of California, Berkeley, CA 94720 Sorenson, Martha M., Depto de Bioquimica-RFRJ, Centra de Ciencias da Saude-I. C. B., Cidade Universitaria-Fundad, Rio de Janeiro, BRASIL 21.910 Speck, William T., Case Western Reserve University, Department of Pediatrics, Cleveland, OH 44 1 06 Spector, Abraham, College of Physicians and Surgeons, Columbia University, 630 West 168th Street, New York, NY 10032 Speer, John W., Marine Biological Laboratory. Woods Hole. MA 02543 Spiegel, Evelyn, Department of Biological Sciences. Dartmouth College, Hanover, NH 03755 Spiegel, Melvin, Department of Biological Sciences, Dartmouth College, Hanover, NH 03755 Spray, David C., Albert Einstein College of Medicine. Department of Neurosciences, 1 300 Morris Park Avenue, Bronx. NY 10461 Steele, John Hyslop, Woods Hole Oceanographic Institution, Woods Hole, MA 02543 Steinacher, Antoinette, Dept. of Otolaryngology, Washington University, School of Medicine, 49 1 1 Barnes Hospital, St. Louis. MO 63 110 Steinberg, Malcolm, Department of Biology, Princeton University, Princeton, NJ 08540 Stephens, Grover C., Department of Developmental and Cell Biology, LIniversity of California, Irvine, CA 927 1 7 Stephens, Raymond E., Marine Biological Laboratory, Woods Hole, MA 02543 Stetten, DeWitt, Jr., NIH. Bldg. 16. Room 1 18. Bethesda. MD 20892 Stetten, Jane Lazarow, 2 W Drive, Bethesda, MD20814 Steudler, Paul A., Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA 02543 Stokes, Darrell R., Department of Biology, Emory University, Atlanta. GA 30322 Stommel, Elijah W., Marine Biological Laboratory'- Woods Hole. MA 02543 Stracher, Alfred, Downstate Medical Center, SUNY. 450 Clarkson Ave.. Brooklyn. NY 1 1203 Strehler, Bernard I.., 2235 25th St., #217, San Pedro, CA 90732 Strunwasser, Felix, Marine Biological Laboratory, Woods Hole, MA 02543 Stuart, Ann E., Department of Physiology, Medical Sciences Research Wing 206H, University of North Carolina. Chapel Hill. NC 275 14 Regular Members 65 Sugimori, Mutsuyuki, Department of Physiology and Biophysics, New York University Medical Center, 550 First Avenue, New York, NY 10016 Summers, William C., Huxley College of Environmental Studies, Western Washington University. Bellingham, WA 98225 Suprenant, Kathy A., Department of Physiology and Cell Biology, 4010 Haworth Hall, University of Kansas, Lawrence, KS 66045 Sussman, Maurice, 72 Carey Lane. Falmouth, MA 02540 Szabo, George, Harvard School of Dental Medicine, 188 Longwood Avenue, Boston, MA 021 15 Szent-Gyorgyi, Andrew, Department of Biology, Brandeis University. Waltham, MA 02254 Szent-Gyorgyi, Eva Szentkiraly, Department of Biology, Brandeis University, Waltham, MA 02254 (deceased 3/22/88) Szuts, Ete Z., Laboratory of Sensory Physiology, Marine Biological Laboratory, Woods Hole, MA 02543 Tamm, Sidney L., Boston University Marine Program, Marine Biological Laboratory, Woods Hole, MA 02543 Tanzer, Marvin L., Department of Oral Biology, Medical School, University of Connecticut. Farmington. CT 06032 Tasaki, Ichiji, Laboratory of Neurobiology, Bldg. 36, Rm. 2B- 16, NIMH/N1H, Bethesda. MD 20892 Taylor, Douglass L., Biological Sciences, Mellon Institute, 440 Fifth Avenue, Pittsburgh, PA 15213 Teal, John M., Department of Biology, Woods Hole Oceanographic Institution, Woods Hole, MA 02543 Telfer, William H., Department of Biology, University of Pennsylvania, Philadelphia, PA 19174 Thorndike, W. Nicholas, Wellington Management Company, 28 State St., Boston, MA 02109 Trager, William, Rockefeller University, 1230 York Ave., New York, NY 10021 Travis, D. M., Veterans Administration Medical Center, 2101 Elm Street, Fargo, ND 58 102 Treistman, Steven N., Worcester Foundation for Experimental Biology, Shrewsbury, MA 01545 Trigg, D. Thomas, 1 25 Grove St.. Wellesley, MA 02 1 8 1 Trinkaus, J. Philip, Department of Biology. Box 6666, Yale University, New Haven, CT 065 1 0 Troll, Walter, Department of Environmental Medicine, College of Medicine, New York University. New York, NY 10016 Troxler, Robert F., Department of Biochemistry, School of Medicine, Boston University, 80 East Concord St., Boston, MA 02 118 Tucker, Edward B., Department of Natural Sciences, Baruch College, 17 Lexington Ave.. New York, NY 10010 Turner, Ruth D., Mollusk Department, Museum of Comparative Zoology, Harvard University, Cambridge, MA 02 138 Tweedell, Kenyon S., Department of Biology, University of Notre Dame, Notre Dame, IN 46656 Tytell, Michael, Department of Anatomy, Bowman Gray School of Medicine, Wake Forest University, Winston- Salem,NC27103 lleno, Hiroshi, Department of Biochemistry, The Rockefeller University, 1230 York Ave., New York, NY 10021 Uretz, Robert B., Division of Biological Sciences, University of Chicago, 950 East 59th St., Chicago, IL 60637 Valiela, Ivan, Boston University Marine Program, Marine Biological Laboratory, Woods Hole, MA 02543 Vallee, Richard, Cell Biology Group, Worcester Foundation for Experimental Biology, Shrewsbury, MA 01 545 Valois, John, Marine Biological Laboratory, Woods Hole, MA 02543 Van Holde, kensal, Department of Biochemistry and Biophysics, Oregon State University, Corvallis, OR 9733 1 Villee, Claude A., Parcel B, Room 122, Harvard Medical School, 25 Shattuck St., Boston, MA 02 1 1 5 Vincent, Walter S., School of Life and Health Sciences, University of Delaware, Newark, DE 1971 1 Waksman, Byron, National Multiple Sclerosis Society, 205 East 42nd St., New York, NY 1 00 1 7 Wall, Betty, 9 George St., Woods Hole, MA 02543 Wallace, Robin A., Whitney Laboratory, 9505 Ocean Shore Blvd., St. Augustine, FL 32086 Wang, An, Wang Laboratories, Inc., One Industrial Ave., Lowell, MA 01 851 Wang, Ching Chung, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA 94143 Warner, Robert C., Department of Molecular Biology and Biochemistry, University of California, Irvine, CA 927 1 7 Warren, Kenneth S., MacMillan, Inc., 866 Third Avenue, New York. NY 10022 Warren, Leonard, Wistar Institute, 36th and Spruce Streets, Philadelphia, PA 19104 Watson, Stanley, Associates of Cape Cod, Inc., P. O. Box 224, Woods Hole, MA 02543 Waxman, Stephen G., Department of Neurology, LCI 708, Yale School of Medicine, 333 Cedar Street, New Haven, CT 06510 Webb, H. Marguerite, Marine Biological Laboratory, Woods Hole, MA 02543 Weber, Annemarie, Department of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania, Philadelphia, PA 19104 Webster, Ferris, Box 765, Lewes, DE 19958 Weidner, Earl, Department of Zoology and Physiology, Louisiana State University, Baton Rouge, LA 70803 Weiss, Dieter, G., Institut fur Zoologie. Technische Universitat Munchen, 8046 Garching, FRG Weiss, Leon P., Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104 Weissmann, Gerald, New York University Medical Center, 550 First Avenue, New York, NY 10016 Werman, Robert, Neurobiology Unit, The Hebrew University, Jerusalem, ISRAEL Westerfield, R. Monte, The Institute of Neuroscience, University of Oregon, Eugene, OR 37403 Wexler, Nancy Sabin, 15 Claremont Avenue, Apt. 92, New York, NY 10027 66 Annual Report White, Roy L., Department of Neuroscience, Albert Einstein College. 1300 Morris Park Avenue, Bronx, NY 10461 Whittaker, J. Richard, Marine Biological Laboratory, Woods Hole, MA 02543 Wigley, Roland I.., 35 Wilson Road, Woods Hole, MA 02543 Wilson, Darcy B., Medical Biology Institute, 1 1077 North Torrey Pines Road, La Jolla, CA 92037 Wilson, Edward O., Museum, Comparative Zoology, Harvard University, Cambridge, MA 02 1 38 Wilson, T. Hastings, Department of Physiology, Harvard Medical School, Boston. MA 02 1 1 5 Wilson, Walter L., 743 Cambridge Drive, Rochester Hills, MI 48063 (deceased 3/1/88) Witkovsky, Paul, Department of Ophthalmology, New York University Medical Center, 550 First Ave., New York NY 10016 Wittenberg, Jonathan B., Department of Physiology and Biochemistry, Albert Einstein College, 1300 Morris Park Ave., Bronx, NY 10016 Wolfe, Ralph, Department of Microbiology, 131 Burrill Hall, University of Illinois, Urbana, IL 61801 Wolken, Jerome J., Department of Biological Sciences. Carnegie Mellon University, 440 Fifth Ave., Pittsburgh. PA 15213 Worgul, Basil V., Department of Ophthalmology, Columbia University, 630 West 168th St.. New York, NY 10032 Wu, Chau Hsiung, Department of Pharmacology. Northwestern University Medical School. Chicago, IL 60611 Wyttenbach, Charles R., Department of Physiology and Cell Biology. University of Kansas. Lawrence. KS 66045 Yen, Jay Z., Department of Pharmacology, Northwestern University Medical School, Chicago, IL 606 1 1 Young, Richard, Mentor O & O, Inc., 3000 Longwater Dr., Norwell. MA 02061 Zackroff, Robert, 80 Kersey Rd., Peacedale, RI 02883 Zigman, Seymour, School of Medicine and Dentistry. University of Rochester, 260 Crittenden Blvd., Rochester. NY 14620 Zigmond, Richard E., Center for Neurosciences, School of Medicine, Case Western Reserve University, Cleveland, OH 44106 Zimmerberg, Joshua J., Bldg. 1 2A, Room 2007, NIH, Bethesda, MD 20892 Zottoli, Steven J., Department of Biology, Williams College, Williamstown, MA01267 Zucker, Robert S., Neurobiology Division. Department of Molecular and Cellular Biology, University of California, Berkeley, CA 94720 Associate Members Ackroyd, Dr. Frederick W. Adams. Dr. Paul Adelherg, Dr. and Mrs. Edward A. Ahearn, Mr. and Mrs. David Alden. Mr. John M. Allard, Dr. and Mrs. Dean C., Jr. Allen. Miss Camilla K. Allen, Dr. Nina S. Amon, Mr. Carl H. Jr Anderson, Mr. J. Gregory Anderson, Drs. James L. and HeleneM. Antonucci, Dr. Robert V Armstrong, Dr. and Mrs. Samuel C. Arnold, Mrs. Lois Aspinwall, Mr. and Mrs. Duncan Atwood, Dr. and Mrs. Kimball C.. Ill Ayers. Mrs. Donald Backus, Mrs. Nell Baker, Mrs. C. L. Ball, Mrs. Eric G. Ballanune. Dr. and Mrs. H. T., Jr. Bang, Mrs. Frederik B. Bang, Miss Molly Banks. Mr and Mrs. William L. Barkan. Mr. and Mrs. Mel A. Barrows, Mrs. Albert W. Baum, Mr. Richard T. Baylor, Drs. Edward and Martha Beers, Dr. and Mrs. Yardley Belesir. Mr. Tasos Bennett, Dr. and Mrs. Michael V. L. Berg. Mr. and Mrs. C. John Bernheimer. Dr. Alan W. Bernstein, Mr. and Mrs. Norman Bicker, Mr. Alvin Bigelow. Mr. and Mrs. Robert O. Bird, Mr. William R. Bleck, Dr. Thomas B. Boche, Mr. Robert Bodeen, Mr. and Mrs. George H. Boettiger, Dr. and Mrs. Edward G. Boettiger, Mrs. Julie Bolton, Mr. and Mrs. Thomas C. Bonn, Mr. and Mrs. Theodore H. Borg, Dr. and Mrs. Alfred F. Borgese, Dr. and Mrs. Thomas Bowles, Dr. and Mrs. Francis P. Bradley. Dr. and Mrs. Charles C. Bradley, Mr. Richard Brown, Mrs. Frank A., Jr. Brown, Mr. and Mrs. Henry Brown, Mrs. James Brown, Mrs. Neil Brown. Mr. and Mrs. T. A. Brown, Dr. and Mrs. Thornton Broyles, Dr. Robert H. Buck, Dr. and Mrs. John W. Buckley. Mr. George D. Bunts, Mr. and Mrs. Frank E. Burt, Mrs. Charles E. Burwell. Dr. and Mrs. E. Langdon Bush. Dr. Louise Buxton, Mr. and Mrs. Bruce E. Mill Associates Joel and Kulh Davis. Associate Members 67 Buxton. Mr. E. Brewster Cadwalader, Mr. George Calkins, Mr. and Mrs. G. N., Jr. Campbell, Dr. and Mrs. David G. Carlson, Dr. and Mrs. Francis Carlton, Mr. and Mrs. Winslow G. Case, Dr. and Mrs. James Chandler. Mr. Robert Chase, Mr. Thomas H. Child, Dr. and Mrs. Frank M., Ill Chisholm, Dr. Sallie W. Church, Dr. Wesley Claff, Mr. and Mrs. Mark Clark, Dr. and Mrs. Arnold Clark, Mr. and Mrs. Hays Clark. Mr. and Mrs. James McC. Clark, Mr. and Mrs. Leroy, Jr. Clark, Dr. Peter L. Clarke, Dr. Barbara J. Clement, Mrs. Anthony C. Cloud, Dr. Laurence P. Clowes Fund. Inc. Clowes, Dr. and Mrs. Alexander W. Clowes. Mr. Allen W. Clowes, Mrs. G. H. A., Jr. Coburn. Mr. and Mrs. Lawrence Cohen, Mrs. Seymour S. Coleman. Drs. John and Annette Collum, Mrs. Peter Colt, Dr. LeBaron C., Jr. Connell, Mr. and Mrs. W. J. Cook. Dr. and Mrs. Joseph Cook, Dr. and Mrs. Paul W., Jr. Copel. Mrs. Marcia N. Copeland, Dr. and Mrs. D. Eugene Copeland, Mr. Frederick C. Copeland, Mr. and Mrs. Preston S. Costello, Mrs. Donald P. Cowan, Mr. and Mrs. James F., Ill Crabb. Mr. and Mrs. David L. Cram, Mr. and Mrs. Melvin C. Cramer, Mr. and Mrs. Ian D. W. Crane. Mrs. John O. Crane, Josephine B.. Foundation Crane, Mr. Thomas S. Crosby, Miss Carol Cross, Mr. and Mrs. Norman C. Crossley. Miss Dorothy Crossley, Miss Helen Crowell, Dr. and Mrs. Sears Currier. Mr. and Mrs. David L. Daignault, Mr. and Mrs. Alexander T. Daniels, Mr. and Mrs. Bruce G. Davidson, Dr. Morton Davis, Mr. and Mrs. Joel P. Day, Mr. and Mrs. Pomeroy Decker. Dr. Raymond F. DeMello, Mr. John DiBerardino, Dr. Marie A. DiCecca. Dr. and Mrs. Charles Dickson, Dr. Willim A. Dierolf. Dr. Shirley H. Donovan. Mr. David L. Dreyer, Mrs. Frank Drummey. Mr. and Mrs. Charles E. Drummey, Mr. Todd A. DuBois. Dr. and Mrs. Arthur B. Dudley, Dr. Patricia DuPont, Mr. A. Felix, Jr. Dutton. Mr. and Mrs. Roderick L. Ebert, Dr. and Mrs. James D. Egloff, Dr. and Mrs. F. R. L. Elliott, Mrs. Alfred M. Enos, Mr. Edward, Jr. Eppel, Mr. and Mrs. Dudley Epstein, Mr. and Mrs. Ray L. Estabrooks. Mr. Gordon C. Evans, Mr. and Mrs. Dudley Farley, Miss Joan Farmer, Miss Mary Faull, Mr. J. Horace, Jr. Ferguson. Mrs. James J., Jr. Fisher, Mrs. B. C. Fisher, Mr. Frederick S., Ill Fisher, Dr. and Mrs. Saul H. Fluck. Mr. Richard A. Folino, Mr. John W., Jr. Forbes. Mr. John M. Ford, Mr. John H. Fowlkes, Mr. Aaron Francis. Mr. and Mrs. Lewis W., Jr. Frenkel, Dr. Krystina Fribourgh, Dr. James H. Friendship Fund Fries, Dr. and Mrs. E. F. B. Frosch. Dr. and Mrs. Robert A. Fye, Mrs. Paul M. Gabriel, Dr. and Mrs. Mordecai L. Gagnon, Mr. Michael Gaiser, Mrs. David W. Gallagher. Mr. Robert O. Garcia, Dr. Ignacio Garneld. Miss Eleanor Gellis. Dr. and Mrs. Sydney Gephard. Mr. Stephen German. Dr. and Mrs. James L., Ill Gewecke, Mr. and Mrs. Thomas H. Gifford, Mr. and Mrs. Cameron Gifford, Mr. John A. Gifford, Dr. and Mrs. Prosser Gilbert, Drs. Daniel L. and Claire Gilbert. Mrs. Carl J. Gildea, Dr. Margaret C. L. Gillette, Mr. and Mrs. Robert S. Glad, Mr. Robert Glass, Dr. and Mrs. H. Bentley Glazebrook, Mr. James G. Glazebrook, Mrs. James R. Goldman, Mrs. Mary Goldring. Mr. Michael Goldstein, Dr. and Mrs. Moise H.Jr. Goodwin, Mr. and Mrs. Charles Gould, Miss Edith Grace, Miss Pnscilla B. Grant, Dr. and Mrs. Philip Grassle, Mrs. J. K. Green, Mrs. Davis Crane Greenberg, Noah and Mosher. Diane Greer, Mr. and Mrs. W. H., Jr. Griffin. Mrs. Robert W. Griffith, Dr. and Mrs. B. Herold Grosch, Dr. and Mrs. Daniel S. Gross. Mrs. Mona Gunning, Mr. and Mrs. Robert Haakonsen, Dr. Harry O. Haigh. Mr. and Mrs. Richard H. Hall. Mr. and Mrs. Peter A. Hall, Mr. Warren C. Halvorson, Dr. and Mrs. Harlyn O. Hamstrom, Miss Mary Elizabeth Harrington, Mr. Robert D., Jr. Harvey, Dr. and Mrs. Richard B. Hassett. Mr. and Mrs. Charles Hastings. Dr. and Mrs. J. Woodland Haubrich. Mr. Robert R. Hay, Mr. John Hays, Dr. David S. Hedberg, Mrs. Frances Hedberg, Dr. Mary Hersey, Mrs. George L. Hiatt. Dr. and Mrs. Howard Hichar. Mrs. Barbara Hill, Mrs. Samuel E. Hirschfeld, Mrs. Nathan B. Hobbie, Dr. and Mrs. John Hocker. Mr. and Mrs. Lon Hodge. Mrs. Stuart Hokin, Mr. Richard Hornor, Mr. Townsend Horwitz, Dr. and Mrs. Norman H. Hoskin, Dr. and Mrs. Francis C.G. Houston, Mr. and Mrs. Howard E. Howard, Mrs. L. L. Hoyle, Dr. Memll C. Huettner. Dr. and Mrs. Robert J. Hutchison, Mr. Alan D. Hyde, Mr. and Mrs. Robinson Hynes. Mr. and Mrs. Thomas J., Jr. Inoue, Dr. and Mrs. Shinya Issokson, Mr. and Mrs. Israel Jackson, Miss Elizabeth B. Jaffe, Dr. and Mrs. Ernst R. Janney. Mrs. F. Wistar Jewett, G. F., Foundation Jewett, Mr. and Mrs. G. F., Jr. Jones, Mr. and Mrs. DeWitt C., Ill Jones, Mr. and Mrs. Frederick. II Harlyn Halvorson presents Lilyan Saunders with a Centennial poster in appreciation for her efforts in the Associates' Gift Shop. 68 Annual Report Jones. Mr. Frederick S.. Ill Jordan. Dr. and Mrs. Edwin P. Kaan. Dr. Helen W. Kahler. Mrs. Robert W. Rammer, Dr. and Mrs. Benjamin Karplus. Mrs. Alan K. Karush, Dr. and Mrs. Fred Kelleher, Mr. and Mrs. Paul R. Kendall, Mr. and Mrs. Richard E. Keosian, Mrs. Jessie Keoughan, Miss Patricia Ketchum. Mrs. Paul Kien. Mr. and Mrs. Pieter Kinnard, Mrs. L. Richard Kirschenbaum, Mrs. Donald Kissam, Mr. and Mrs. William M. Kivy, Dr. and Mrs. Peter Keller, Dr. Lewis R. Korgen. Dr. Ben J. Kravitz. Dr. and Mrs. Edward Kufller, Mrs. Stephen W. Laderman. Mr. and Mrs. Ezra Lafferty, Miss Nancy Larmon, Mr. Jay Laster, Dr. and Mrs. Leonard Laufer, Dr. and Mrs. Hans Laufer, Jessica, and Weiss, Malcolm LaVigne. Mrs. Richard J. Lawrence, Mr. Frederick V. Lawrence, Mr. and Mrs. William Leach, Dr. Berton J. Leatherbee, Mrs. John H. LeBlond, Mr. and Mrs. Arthur Leeson, Mr. and Mrs. A. Dix LeFevre, Dr. Marian E. Lehman, Miss Robin Lenher. Dr. and Mrs. Samuel Leprohon, Mr. Joseph Levine, Mr. Joseph Levine, Dr. and Mrs. Rachmiel Levitz. Dr. Mortimer Levy, Mr. and Mrs. Stephen R Lindner, Mr. Timothy P. Little, Mrs. Elbert Livingstone, Mr. and Mrs. Robert Lloyd, Mr. and Mrs. James Loeb, Mrs. Robert F. Loessel, Mrs. Edward Lovell, Mr. and Mrs. Hollis R. Lovering, Mr. Richard C. Low, Miss Doris Lowe. Dr. and Mrs. Charles V. Lowengard. Mrs. Joseph Mackey, Mr. and Mrs. William K. MacLeish, Mrs. Margaret MacNary. Mr. and Mrs. B. Glenn MacNichoI. Dr. and Mrs. Edward F., Jr. Maher, Miss Anne Camille Mahler, Mrs. Henry Mahler. Mrs. Suzanne Mansworth. Miss Mane Maples. Dr. Philip B. Marsh, Dr. and Mrs. Julian Martyna, Mr. and Mrs. Joseph C. Mason, Mr. Appleton Mastroianni, Dr. and Mrs. Luigi, Jr. Mather, Mr. and Mrs. Frank J.. Ill Matherly, Mr. and Mrs. Walter Matthiessen, Dr. and Mrs. G. C. McCusker. Mr. and Mrs. Paul T. McElroy, Mrs. Nella W. Mcllwain. Dr. Susan G. McMurtne, Mrs. Cornelia Hanna Meigs. Mr. and Mrs. Arthur Meigs. Dr. and Mrs. J. Wister Melillo. Dr. and Mrs. Jerry M. Mellon, Richard King, Trust Mellon, Mr. and Mrs. Richard P. Mendelson, Dr. Martin Metz, Dr. and Mrs. Charles B. Meyers, Mr. and Mrs. Richard Miller, Dr. Daniel A. Miller, Mr. and Mrs. Paul Mixter. Mr. and Mrs. William J., Jr. Mizell, Dr. and Mrs. Merle Monroy, Mrs. Alberto Montgomery, Dr. and Mrs. Charles H. Montgomery, Mrs. Raymond B. Moore, Drs. John and Betty Morgan, Miss Amy Morse, Mrs. Charles L., Jr. Morse, Dr. M. Patricia Moul, Mrs. Edwin T. Murray, Mr. David M. Myles-Tochko, Drs. Christina J. and John Nace, Dr. and Mrs. Paul Nace, Mr. Paul F., Jr. Neall. Mr. William G. Nelson, Dr. and Mrs. Leonard Nelson, Dr. Pamela Newton, Mr. William F. Nickerson. Mr. and Mrs. Frank L. Norman, Mr. and Mrs. Andrew E. Norman Foundation Norris, Mr. and Mrs. Barry Norris, Mr. and Mrs. John A. Norns, Mr. William Norton, Mrs. Thomas J. O'Herron. Mr. and Mrs. Jonathan Olszowka, Miss Janice S. O'Neil, Mr. and Mrs. Barry T. O'Rand, Mr. and Mrs. Michael O'Sullivan, Dr. Renee Bennett Pappas. Dr. and Mrs. George D. Park, Mr. and Mrs. Malcolm S. Parmenter, Dr. Charles Parmenter. Miss Carolyn L. Peltz, Mr. and Mrs. William L. Pendergast, Mrs. Claudia Pendleton, Dr. and Mrs. Murray E. Peri, Mrs. Barbara Anne Perkins, Mr. and Mrs. Courtland D. Person, Dr. and Mrs. Philip Peterson, Mr. and Mrs. E. Gunnar Peterson, Mr. and Mrs. E. Joel Peterson, Mr. Raymond W. Petty, Mr. Richard F. Pfeiffer, Mr. and Mrs. John Plough, Mr. and Mrs. George H. Plough, Mrs. Harold H. Pointe, Mr. Albert Pointe, Mr. Charles Porter, Dr. and Mrs. Keith R. Pothier, Dr. and Mrs. Aubrey Press, Drs. Frank and Billie Proskauer, Mr. Joseph H. Proskauer. Mr. Richard Prosser, Dr. and Mrs. C. Ladd Psaledakis, Mr. Nicholas Psychoyos, Dr. Alexandre Putnam, Mr. Allan Ray Putnam, Mr. and Mrs. William A., Ill Raymond. Dr. and Mrs. Samuel Reese, Miss Bonnie Reingold, Mr. Stephen C. Reynolds, Dr. and Mrs. George Reynolds. Mr. and Mrs. Robert M. Reznikoff, Mrs. Paul Ricca, Dr. and Mrs. Renato A. Righter, Mr. and Mrs. Harold Riley, Dr. Monica Riina, Mr. John R. Robb, Mrs. Alison A. Roberts. Miss Jean Roberts, Mr. Mervin F. Robertson, Mrs. C. W. Robinson. Dr. Denis M. Root, Mrs. Walter S. Rosenthal, Miss Hilde Roslansky, Drs. John and Priscilla Ross. Dr. and Mrs. Donald Ross, Dr. Robert Ross, Dr. Virginia Roth. Dr. and Mrs. Stephen Rowan. Mr. Edward Rowe, Mr. Don Rowe, Mr. and Mrs. William S. Rugh, Mrs. Roberts Ryder, Mr. and Mrs. Francis C. Sager, Dr. Ruth Sardinha, Mr. George H. Saunders. Dr. and Mrs. John W. Saunders, Mrs. Lawrence Saunders. Lawrence. Fund Sawyer, Mr. and Mrs. John E. Saz. Mrs. Ruth L Schlesinger. Dr. and Mrs. R. Walter Schwamb, Mr. Peter Scott, Mrs. George T. Scott, Mr. and Mrs. Norman E. Sears, Mr. Clayton C. Sears. Mr. and Mrs. Harold B. Sears. Mr. Harold H. Seaver, Mr. George Segal, Dr. and Mrs. Sheldon J. Selby, Dr. Cecily Senft. Dr. and Mrs. Alfred Shapiro. Mr. and Mrs. Howard Shapley. Dr. Robert Shemin. Dr. and Mrs. David Shepro, Dr. and Mrs. David Siegel, Mr. and Mrs. Alvin Simmons, Mr. Tim Singer, Mr. and Mrs. Daniel M. Smith, Drs. Frederick E. and Marguerite A. Smith, Mrs. Homer P. Smith, Mr. Van Dorn C. Snyder, Mr. Robert M. Solomon. Dr. and Mrs. A. K. Speck, Dr. William T. Specht, Mr. and Mrs. Heinz Spiegel, Dr. and Mrs Melvin Spotte, Mr. Stephen Steele, Mrs. John H. Stein. Mr. Ronald Steinbach, Mrs. H. Bun- Stetson. Mrs. Thomas J. Stetten, Dr. Gail Stetten, Dr. and Mrs. H. DeWitt, Jr. Stunkard, Dr. Horace Sudduth, Dr. William Swanson. Dr. and Mrs. Carl P. Swope, Mrs. Gerard, Jr. Swope, Mr. and Mrs. Gerard L. Szent-Gyorgyi, Dr. Andrew Taber, Mr. George H. Taylor. Mr. James K. Taylor, Dr. and Mrs. W. Randolph Tietje, Mr. and Mrs. Emil D., Jr Timmins, Mrs. William Todd, Mr. and Mrs. Gordon F. Tolkan, Mr. and Mrs. Norman N. Trager, Mrs. William Trigg, Mr. and Mrs. D. Thomas Associate Members 69 Troll, Dr. and Mrs. Walter Trousof. Miss Natalie Tucker, Miss Ruth Tully, Mr. and Mrs. Gordon F. Ulbnch, Mr. and Mrs. Volker Valois. Mr. and Mrs. John Van Buren, Mrs. Harold Van Holde, Mrs. Kensal E. Veeder. Mrs. Ronald A. Vincent. Mr. and Mrs. Samuel W. Vincent, Dr. Walter S. Wagner, Mr. Mark Waksman, Dr. and Mrs. Byron H Ward, Dr. Robert T. Ware, Mr. and Mrs. J. Lindsay Warren, Dr. Henry B. Warren, Dr. and Mrs. Leonard Watt, Mr. and Mrs. John B. Weeks, Mr. and Mrs. John T. Weinstein. Miss Nancy B. Weisberg, Mr. and Mrs. Alfred M. Weissmann, Dr. and Mrs. Gerald Wheeler, Dr. and Mrs. Paul S. Whitehead, Mr. and Mrs. Fred Whitney, Mr. and Mrs. Geoffrey G., Jr. Wichterman. Dr. and Mrs. Ralph Wickersham, Mr. and Mrs. A. A. Tilney Wiese, Dr. Konrad Wilhelm, Dr. Hazel S. Wilson, Mr. and Mrs. T. Hastings Wilson, Mr. and Mrs. Leslie J. Winn, Dr. William M. Winsten, Dr. Jay A. Witting. Miss Joyce Wofinsohn. Mrs. Wolfe Woodwell. Dr. and Mrs. George M. Yntema, Mrs. Chester L. Young, Miss Nina L. Zinn, Dr. and Mrs. Donald J. Ztpf, Dr. Elizabeth Certificate of Organization Articles of Amendment Bylaws of the MBL Certificate of Organization (On File in the Office of the Secretary of the Commonwealth) No. 3170 We. Alpheus Hyatt. President. William Stanford Stevens. Treasurer, and William T. Sedgwick, Edward G. Gardiner, Susan Minis and Charles Sedgwick Mmot being a majority of the Trustees of the Marine Biological Laboratory in compli- ance with the requirements of the fourth section of chapter one hundred and fifteen of the Public Statutes do hereby certify that the following is a true copy of the agreement of association to constitute said Corporation, with the names of the subscribers thereto: We. whose names are hereto subscribed, do. by I his agreement, associate ourselves with the intention to constitute a Corporation according to the provisions of the one hundred and fifteenth chapter of the Public Statutes of the Commonwealth of Massachusetts, and the Acts in amendment thereof and in addition thereto. The name by which the Corporation shall be known is THE MARINE BIOLOGICAL LABORATORY. The purpose for which the Corporation is constituted is to establish and maintain a laboratory or station for scientific study and investigations, and a school for instruction in biology and natural history. The place within which the Corporation is established or located is the city of Boston within said Commonwealth. The amount of its capital stock is none. //; II //w\\ II hi'i'i'nl. we have hereunto set our hands, this twenty seventh day of February in the year eighteen hundred and eighty-eight, Alpheus Hyatt, Samuel Mills. William T. Sedgwick, Edward G. Gardiner. Charles Sedgwick Mmot. Wil- liam G. Farlow. William Stanford Stevens, Anna D. Phillips. Susan Minis, B. H. Van Vleck. That the first meeting of the subscribers to said agreement was held on the thir- teenth day ot March in the year eighteen hundred and eighty-eighl. /« HVi/icvs ll'licrcnl. we have hereunto signed our names, this thirteenth day of March in the year eighteen hundred and eighty-eight. Alpheus Hyatt. President. William Stanford Stevens. Treasurer. Edward G. Gardiner, William T. Sedgwick, Susan Minis. ( harles Sedgwick Mmot. (Approved on March 20. 1888 as follows: / herein- ivrnfy that it appears upon an examination of the within written certifi- cate and the records of the corporation duly submitted to my inspection, that the requirements of sections one. two and three of chapter one hundred and fifteen, and sections eighteen, twenty and twenty-one of chapter one hundred and six. of the Public Statutes, have been complied with and I hereby approve said certificate this twentieth day of March A.D. eighteen hundred and eighty-eight. Charles Endicott Commissioner of Corporations) A rticles of A mend men t (On File in the Office of the Secretary of the Commonwealth) We, James D. Ebert, President, and David Shepro. Clerk of the Manne Biological Laboratory, located at Woods Hole. Massachusetts 02543. do hereby certify that the following amendment to the Articles of Organization of the Corporation was duly adopted at a meeting held on August 15, 1975. as adjourned to August 29. 1975. by vote of 444 members, being at least two-thirds of its members legally qualified to vote in the meeting of the corporation: Voted: That the Certificate of Organization of this corporation be and it hereby is amended by the addition of the following provisions: "No Officer, Trustee or Corporate Member of the corporation shall be personally liable for the payment or satisfaction of any obligation or liabil- ities incurred as a result of, or otherwise in connection with, any commit- ments, agreements, activities or affairs of the corporation. "Except as otherwise specifically provided by the Bylaws of the corpora- tion, meetings of the Corporate Members of the corporation may be held anywhere in the United States. "The Trustees of the corporation may make, amend or repeal the Bylaws of the corporation in whole or in part, except with respect to any provi- sions thereof which shall by law, this Certificate or the bylaws ol the corpo- ration, require action by the Corporate Members." The foregoing amendment will become effective when these articles of amend- ment are tiled in accordance with Chapter 180. Section 7 of the General Laws unless these articles specify, in accordance with the vote adopting the amend- ment, a later effective dale not more than thirty days after such filing, in which event the amendment will become effective on such later date. 70 By laws of the MBL 71 In H 'itness whereof and Under the Penalties oj Perjury, we have hereto signed our names this 2nd day of September, in the year 1975, James D. Ebert, President; David Shepro, Clerk. (Approved on October 24, 1975. as follows: I hereby approve the within articles of amendment and. the filing fee in the amount of $10 having been paid, said articles are deemed to have been filed with me this 24th day of October. 1 975. Paul Guzzi Secretary a/the Commonwealth) Bylaws of the Corporation of the Marine Biological Laboratory (Revised August 14. 1987) I. (A) The name of the Corporation shall be The Marine Biological Laboratory. The Corporation's purpose shall be to establish and maintain a laboratory or sta- tion for scientific study and investigation, and a school for instruction in biology and natural history. (B) Marine Biological Laboratory admits students without regard to race, color, sex, national and ethnic origin to all the rights, privileges, programs and activities generally accorded or made available to students in its courses. It does not discnminate on the basis of race, color, sex, national and ethnic origin in employment, administration or its educational policies, admissions policies, scholarship and other programs. II. (A) The members of the Corporation ("Members") shall consist of persons elected by the Board of Trustees, upon such terms and conditions and in accor- dance with such procedures, not inconsistent with law or these Bylaws, as may be determined by said Board of Trustees. Except as provided below, any Member may vote at any meeting either in person or by proxy executed no more than six months prior to the date of such meeting. Members shall serve until their death or resignation unless earlier removed with or without cause by the affirmative vote of two-thirds of the Trustees then in office. Any member who has attained the age of seventy years or has retired from his home institution shall automatically be designated a Life Member provided he signifies his wish to retain his membership. Life Members shall not have the right to vote and shall not be assessed for dues. (B) The Associates of the Marine Biological Laboratory shall be an unincorpo- rated group of persons (including associations and corporations) interested in the Laboratory and shall be organized and operated under the general supervision and authority of the Trustees. III. The officers of the Corporation shall consist of a Chairman of the Board of Trustees. President. Director, Treasurer, and Clerk, elected or appointed by the Trustees as set forth in Article IX. IV. The Annual Meeting of the Members shall be held on the Fnday following the Second Tuesday in August in each year at the Laboratory in Woods Hole, Massachusetts, at 9:30 a.m. Subject to the provisions of Article VIII(2), at such meeting the Members shall choose by ballot six Trustees to serve four years, and shall transact such other business as may properly come before the meeting. Spe- cial meetings of the Members may be called by the Chairman or Trustees to be held at such time and place as may be designated. V. Twenty-five Members shall constitute a quorum at any meeting. Except as otherwise required by law or these Bylaws, the affirmative vote of a majority of the Members voting in person or by proxy at a meeting attended by a quorum (present in person or by proxy) shall constitute action on behalf of the Members. VI. (A) Inasmuch as the time and place of the Annual Meeting of Members are fixed by these Bylaws, no notice of the Annual Meeting need be given. Notice of any special meeting of Members, however, shall be given by the Clerk by mail- ing notice of the time and place and purpose of such meeting, at least 15 days before such meeting, to each Member at his or her address as shown on the records of the Corporation. (B) Any meeting of the Members may be adjourned to any other time and place by the vote of a majority of those Members present or represented at the meeting, whether or not such Members constitute a quorum. It shall not be neces- sary to notify any Members of any adjournment. VII. The Annual Meeting of the Trustees shall be held promptly after the An- nual Meeting of the Corporation at the Laboratory in Woods Hole. Massachu- setts. Special meetings of the Trustees shall be called by the Chairman, the Presi- dent, or by any seven Trustees, to be held at such time and place as may be desig- nated. Notice of Trustees' meetings may be given orally, by telephone, telegraph or in writing; and notice given in time to enable the Trustees to attend, or in any case notice sent by mail or telegraph to a Trustee's usual or last known place of residence, at least one week before the meeting shall be sufficient. Notice of a meeting need not be given to any Trustee if a written waiver of notice, executed by him before or after the meeting is filed with the records of the meeting, or if he shall attend the meeting without protesting prior thereto or at its commencement the lack of notice to him. VIII. (A) There shall be four groups of Trustees: ( 1 ) Trustees (the "Corporate Trustees") elected by the Members according to such procedures, not inconsistent with these Bylaws, as the Trustees shall have determined- Except as provided below, such Trustees shall be divided into four classes of six. one class to be elected each year to serve for a term of four years. Such classes shall be designated by the year of expiration of their respective terms. (2) Trustees ("Trustees-at-large") approved by members according to such procedures, not inconsistent with these Bylaws, as the Trustees shall have deter- mined. Except as provided below, such Trustees-at-large shall be divided into four classes of four, one class to be elected each year to serve for a term of four years. Such classes shall be designated by the year of expiration of their respective terms. It is contemplated that, unless otherwise determined by the Trustees for good reason. Trustees-at-large. shall be individuals who have not been considered for election as Corporate Trustees. ( 3 ) Trustees ex officio. who shall be the Chairman, the President, the Director, the Treasurer, and the Clerk. (4) Trustees emeriti, who shall include any Member who has attained the age of seventy years (or the age of sixty-five and has retired from his home institution) and who has served a full elected term as a regular Trustee, provided he signifies his wish to serve the Laboratory in that capacity. Any Trustee who qualifies for emeritus status shall continue to serve as a regular Trustee until the next Annual Meeting whereupon his office as regular Trustee shall become vacant and be filled by election by the Members or by the Board, as the case may be. The Trustees ex officio and ementi shall have all the rights of the Trustees, except that Trustees emeriti shall not have the right to vote. (B) The aggregate number of Corporate Trustees and Trustees-at-large elected in any year (excluding Trustees elected to fill vacancies which do not result from expiration of a term) shall not exceed ten. The number of Trustees-at-large so elected shall not exceed four and unless otherwise determined by vote of the Trustees, the number of Corporate Trustees so elected shall not exceed six. Corpo- rate Trustees shall always constitute a majority on the Board of those elected or approved by the Corporation. (C) The Trustees and Officers shall hold their respective offices until their suc- cessors are chosen in their stead. (D) Any Trustee may be removed from office at any time with or without cause, by vote of a majority of the Members entitled to vote in the election of Trustees; or for cause, by vote of two-thirds of the Trustees then in office. A Trustee may be removed for cause only if notice of such action shall have been given to all of the Trustees or Members entitled to vote, as the case may be, prior to the meeting at which such action is to be taken and if the Trustee so to be removed shall have been given reasonable notice and opportunity to be heard before the body proposing to remove him. 72 Annual Report (E) Any vacancy in the number of I rustees, however arising, may be tilled by the Trustees then in office unless and until filled by the Members at the next Annual Meeting. (F) A Corporate Trustee or u Tmstee-at-large who has served an initial term of at least two years duration shall be eligible for re-election to a second term, but shall be ineligible for re -election to any subsequent term until two years have elapsed after he last served as Trustee. IX. (A) The Trustees shall have the control and management of the affairs of the Corporation. They shall elect a Chairman of the Board of Trustees who shall be elected annually and shall serve until his successor is selected and qualified and who shall also preside at meetings of the Corporation. They shall elect a President of the Corporation who shall also be the Vice Chairman of the Board of Trustees and Vice Chairman of meetings of the Corporation, and who shall be elected annually and shall serve until his successor is selected and qualified. They shall annually elect a Treasurer who shall serve until his successor is selected and quali- fied. They shall elect a Clerk (a resident of Massachusetts) who shall serve for a term of four years. Eligibility for re-election shall be in accordance with the con- tent of Article VIII(F) as applied to Corporate or Board Trustees. They shall elect Board Trustees as described in Article VIII(B). They shall appoint a Director of the Laboratory for a term not to exceed five years, provided the term shall not exceed one year if the candidate has attained the age of 65 years pnor to the date ofthe appointment. They may choose such other officers and agents as they may think best. They may fix the compensation and define the duties of all the officers and agents ofthe Corporation and may remove them at any time. They may fill vacancies occurring in any of the offices. The Board of Trustees shall have the power to choose an Executive Committee from their own number as provided in Article X. and to delegate to such Committee such of their own powers as they may deem expedient in addition to those powers conferred by Article X. They shall from time to time elect Members to the Corporation upon such terms and conditions as they shall have determined, not inconsistent with law or these By- laws. (B| The Board of Trustees shall also have the power, by vote of a majority ofthe Trustees then in Office, to elect an Investment Committee and any other committee and, by like vote, to delegate thereto some or all of their powers except those which by law, the Articles of Organization or these Bylaws they are prohib- ited from delegating. The members of any such committee shall have tenure and duties as the Trustees shall determine; provided that the Investment Committee, which shall oversee the management ofthe Corporation's endowment funds and marketable secunties, shall include the Chairman ofthe Board of Trustees, the Treasurer of the Corporation, and the Chairman of the Corporation's Budget Committee, as ex officio members, together with such Trustees as may be required for not less than two-thirds ofthe Investment Committee to consist of Trustees. Except as otherwise provided by these Bylaws or determined by the Trustees, any such committee may make rules for the conduct of its business; but, unless otherwise provided by the Trustees or in such rules, its business shall be conducted as nearly as possible in the same manner as is provided by these Bylaws for the Trustees. X. (A) The Executive Committee is hereby designated to consist of not more than ten members, including the ex officio Members (Chairman ofthe Board of Trustees. President, Director, and Treasurer); and six additional Trustees, two of whom shall be elected by the Board of Trustees each year, to serve for a three- year term. Beginning with the members elected for terms ending in 1990, one of the Trustees elected to serve on the Executive Committee should be a Trustee-at- large. This procedure will be repeated in the class of 1991, and henceforth the Trustees will elect to the Executive Committee Trustees to ensure that the compo- sition ofthe Committee is four Corporate Trustees and two Trustees-at-large. (B) The Chairman of the Board of Trustees shall act as Chairman ofthe Execu- tive Committee, and the President as Vice Chairman. A majority ofthe members ofthe Executive Committee shall constitute a quorum and the affirmative vote of a majority of those voting at any meeting at which a quorum is present shall constitute action on behalf of the Executive Committee. The Executive Commit- tee shall meet at such times and places and upon such notice and appoint such subcommittees as the Committee shall determine. (C) The Executive Committee shall have and may exercise all the powers of the Board during the intervals between meetings ofthe Board of Trustees except those powers specifically withheld from time to time by vote ofthe Board or by law. The Executive Committee may also appoint such committees, including per- sons who are not Trustees, as it may from time to time approve to make recom- mendations with respect to matters to be acted upon by the Executive Committee or the Board of Trustees. (D) The Executive Committee shall keep appropnate minutes of its meetings and its action shall be reported to the Board ol'Trustees. (E) The elected Members ofthe Executive Committee shall constitute a stand- ing "Committee for the Nomination of Officers," responsible for making nomina- tions, at each Annual Meeting ofthe Corporation, and ofthe Board of Trustees, for candidates to fill each office as the respective terms of office expire (Chairman ofthe Board, President. Director. Treasurer, and Clerk). XI. A majority ofthe Trustees, the Executive Committee, or any other com- mittee elected by the Trustees shall constitute a quorum; and a lesser number than a quorum may adjourn any meeting from time to time without further notice. At any meeting ofthe Trustees, the Executive Committee, or any other committee elected by the Trustees, the vote of a majority of those present, or such different vote as may be specified by law, the Articles of Organization or these Bylaws, shall be sufficient to take any action. XII. Any action required or permitted to be taken at any meeting of the Trust- ees, the Executive Committee or any other committee elected by the Trustees as referred to under Article IX may be taken without a meeting if all ofthe Trustees or members of such committee, as the case may be. consent to the action in writ- ing and such written consents are filed w ith the records of meetings. The Trustees or members ofthe Executive Committee or any other committee appointed by the Trustees may also participate in meeting by means of conference telephone, or otherwise take action in such a manner as may from time to time be permitted by law. XIII. The consent of every Trustee shall be necessary to dissolution ofthe Ma- rine Biological Laboratory. In case of dissolution, the property shall be disposed of in such a manner and upon such terms as shall be determined by the affirmative vote of two-thirds ofthe Board of Trustees then in office. XIV. These Bylaws may be amended by the affirmative vote ofthe Members at any meeting, provided that notice ofthe substance ofthe proposed amendment is stated in the notice of such meeting. As authorized by the Articles of Organiza- tion, the Trustees, by a majority of their number then in office, may also make, amend, or repeal these Bylaws, in whole or in part, except with respect to (a) the provisions of these Bylaws governing (i) the removal of Trustees and (ii) the amendment of these Bylaws and (b) any provisions of these Bylaws which by law, the Articles of Organization or these Bylaws, requires action by the Members. No later than the time of giving notice ofthe meeting of Members next follow- ing the making, amending or repealing by the Trustees of any Bylaw, notice thereof stating the substance of such change shall be given to all Corporation Members entitled to vote on amending the Bylaws. Any Bylaw adopted by the Trustees may be amended or repealed by the Mem- bers entitled to vote on amending the Bylaws. XV. The account ofthe Treasurer shall be audited annually by a certified pub- lic accountant. XVI. Except as otherwise provided below , the Corporation shall, to the extent legally permissible, indemnify each person who is. or shall have been, a Trustee, director or officer ofthe Corporation or who is serving, or shall have served, at the request ol the Corporation as a Trustee, director or officer of another organiza- tion in which the Corporation directly or indirectly has any interest, as a share- holder, creditor or otherwise, against all liabilities and expenses (including judg- ments, fines, penalties and reasonable attorneys' lees and all amounts paid, other than to the Corporation or such other organization, in compromise or settlement ) imposed upon or incurred by any such person in connection with, or arising out ol, the defense or disposition ol any action, suit or other proceeding, whether civil or criminal, in which he or she may be a defendant or with which he or she may be threatened or otherwise involved, directly or indirectly, by reason of his or her being or having been such a Trustee, director or officer. Bylaws of the MBL 73 The Corporation shall provide no indemnification uith respect to any matter as to which any such Trustee, director or officer shall he finally adjudicated in such action, suit or proceeding not to have acted in good faith in the reasonahle belief that his or her action was in the best interests of the Corporation. The Cor- poration shall provide no indemnification with respect to any matter settled or compromised, pursuant to a consent decree or otherwise, unless such settlement or compromise shall have been approved as in the best interests of the Corpora- tion, after notice that indemnification is involved, by (i) a disinterested majority of the Board of Trustees or of the Executive Committee or. (ii) a majority of the Corporation's Members. Indemnification may include payment by the Corporation of expenses in de- fending a civil or criminal action or proceeding in advance of the final disposition of such action or proceeding upon receipt of an undertaking by the person indem- nified to repay such payment if it is ultimately determined that such person is not entitled to indemnification under the provisions of this Article XVI, or under any applicable law. As used in this Article, the terms "Trustee," "director" and "officer" include their respective heirs, executors, administrators and legal representatives, and an "interested" Trustee, director or officer is one against whom in such capacity the proceeding in question or another proceeding on the same or similar grounds is then pending. To assure indemnification under this Article of all persons who are determined by the Corporation or otherwise to be or to have been "fiduciaries" of any em- ployee benefit plan of the Corporation which may exist from time to time, this Article shall be interpreted as follows: (i) "another organization" shall be deemed to include such an employee benefit plan, including w ithout limitation, any plan of the Corporation which is governed by the Act of Congress entitled "Employee Retirement Income Security Act of 1974," as amended from time to timeC'ER- ISA"): (ii) "Trustee" shall be deemed to include any person requested by the Cor- poration to serve as such for an employee benefit plan where the performance by such person of his or her duties to the Corporation also imposes duties on, or otherwise involves services by, such person to the plan or participants or benefi- ciaries of the plan: (in) "fines" shall be deemed to include any excise taxes assessed on a person with respect to an employee benefit plan pursuant to ER1SA; and (iv) actions taken or omitted by a person with respect to an employee benefit plan in the performance of such person's duties fora purpose reasonably believed by such person to be in the interest of the participants and beneficiaries of the plan shall be deemed to be for a purpose which is in the best interests of the Corporation. The right of indemnification provided in this Article shall not be exclusive of or affect any other nghts to which any Trustee, director or officer may be entitled under any agreement, statute, vote of members or otherwise. The Corporation's obligation to provide indemnification under this Article shall be offset to the ex- tent of any other source ot indemnification or any otherwise applicable insurance coverage under a policy maintained by the Corporation or any other person. Nothing contained in this Article shall affect any nghts to which employees and corporate personnel other than Trustees, directors or officers may be entitled by contract, by vote of the Board of Trustees or of the Executive Committee or other- wise. Centennial Events Calendar Jean-Pierre Rampal and Jelle Alema perform in the 14 August 1988 Centennial Benefit Concert. March April May June 1 6 Hoya Crystal reception for the MBL. 29 Panel Discussion on Learning and Memory. Speakers included Daniel Alkon, Jerome Kagan, and David Hubel. 22-25 The Cellular Basis of Morphogenesis symposium honoring Dr. J. P. Trinkaus on his retirement from Yale University. 24 Futures in Science Student Presentations. 9, 10 MBL Update. Science writers and Science Writing Fellowship alumni met to learn about and discuss science as it is performed at the MBL. 1 5 "Science in a Social H 'orld" five- part, weekly lecture series, sponsored by the Boston University Marine Program Graduate Student Organization. 23-25 Ionic C 'hannels: Structure, [•'unction, and Morphology symposium. July 24 Centennial Evening Lecture: Meredith Applebury. speaker. 26 Associates ' Brunch. 30 Cape Cod: A Diversity of Life photo contest and exhibit (6/30 to 7/1 5). 1 Centennial Evening Lecture: Daniel Koshland, speaker. 1,2 "To See \\ 'hat Even '< me Has Seen, To Think What No One Has Thought. "A symposium honoring the late Albert Szent-Gyorgyi. 8 Centennial Evening Lecture (Lang Lecture): Torsten Wiesel, speaker. 8 Scientific Illustration — 1560- 1988 exhibit. 9 "Study Nature Not Books " botany field trip. 1 4 ' 'Roadblocks and Breakthroughs in Scientific Discovery" weekly history of science series opened. 1 5 C 'entennial Evening Lecture: Joshua Lederberg. speaker. 1 6 Associates ' Nautical Treasure Hunt. Centennial Calendar 75 1 7 Dedication Day (the 1 00th anniversary of the MBL's opening) Opening of Centennial Art Exhibit Centennial Evening Lecture: Gerald Weissmann, speaker MBL Birthday Party 21.22 Forbes Lectures: Sydney Brenner, speaker. 22 A Ibert M. and Ellen R. Grass Reference Room Dedication. 29 Centennial Evening Lecture: Clay Armstrong, speaker. (From left) Ratsuma Dan, former Dan Fellow Melanie Pratt. Daniel Ma/i.i. Shinya Inouc, former Dan Fellow Ron Vale, and (kneeling) for- mer Dan F'ellow Carl Johnson at the 1 1 August 1988 "Reflections with Katsuma Dan and Daniel Ma/ia" centennial event. 12 14 MBL Centennial Celebratory Week: Corporation Meeting Dedication of Charles Ulrick Bay Reading Room Trustees Meeting Opening Ceremony Centennial Evening Lecture: E. O. Wilson, speaker Open Rehearsal for Benefit Concert MBL Centennial Benefit Concert featuring new composition by Ezra Laderman. Performed by Jean-Pierre Rampal. Jelle Atema, and the Colorado Quartet August 4 Cape and Islands Chamber Music Festival Concert. 5 Centennial Evening Lecture: John Hobbie. speaker. 6 National Academy of Sciences Reception. Hiroshima Day Lecture: Donald Kennedy, speaker. 1 1 Reflections with Katsuma Dan and Daniel Mazia. Old Timers return from their 16 August 1988 collecting trip. 76 Annual Report 15 Biomedical Applications of Basic Research symposium 1 6 Old Timers' Collecting Trip Centennial Evening Lecture: Clifford Slaman, speaker 1 7 Old Timers' Day 19 Centennial Evening Lecture: Shinya Inoue, speaker Closing Ceremony September 20 Cape anil Islands Chamber Music Concert. 22-24 MBL General Scientific Meetings. 26 Centennial Evening Lecture: John Gurdon, speaker. 27-30 Developmental Biology of Sea Urchins symposium. 3-6 Ionic Currents in Development symposium. Reference: Biol. Bull 111: 77-82. (August, 1989) Contrasting Modes of Reproduction in Two Antarctic Asteroids of the Genus Porania, With a Description of Unusual Feeding and Non-feeding Larval Types ISIDRO BOSCH Institute of Marine Sciences and Biology Board of Studies, University of California, Santa Cruz, California 95064 and Department of Larval Ecology, Harbor Branch Institution Inc., Ft. Pierce, Florida 34946 Abstract. Reproduction and development were mark- edly different in two morphologically similar asteroids of the genus Porania that occur in shallow waters of Mc- Murdo Sound, Antarctica. Adults of the recognized spe- cies, Porania antarctica (Perrier, 1894), are large (Rmax = 70 mm) and have genital pores that are situated ab- orally on the disc; females broadcast spawn large num- bers (3-4 X 104) of buoyant eggs that measure 0.55 mm and develop into unusual, yolky planktotrophic larvae. In contrast, adults of the undescribed Porania sp. are considerably smaller (Rmax = 30 mm) and their genital pores are located orally on the disc. Female fecundity is low ( 100-3 10 eggs); the few eggs produced measure 0.55 mm, are heavier than seawater, and develop into demer- sal lecithotrophic larvae. These differences conform to general patterns reported forechinoderms with divergent types of reproduction. However, other differences con- tradict established trends; specifically, P. antarctica with planktotrophic development has a shorter embryonic and larval phase (65 days vs. 78 days) and a larger juve- nile size at metamorphosis (0.8 vs. 0.6 mm) than Porania sp., which has lecithotrophic development. The repro- duction of P. antarctica incorporates advantages of both planktotrophic and lecithotrophic strategies and may be particularly well-suited for environmental conditions in the Antarctic Ocean. Introduction The larval development of echinoderms can be classi- fied into general categories on the basis of mode of nutri- tion and habitat (Chia, 1974). Larvae may be lecitho- Received 24 March 1989; accepted 31 May 1989. trophic (non-feeding) or planktotrophic (feeding on par- ticulate material); lecithotrophic larvae can be pelagic, demersal, or brooded, whereas planktotrophic larvae are nearly always pelagic (Young and Chia, 1987). Species with lecithotrophic development typically have a larger maximum egg size, reduced fecundity, abbreviated lar- val development, and greater juvenile size at metamor- phosis than species with planktotrophic development (Strathmann, 1985; Emlet et ai. 1987). Intermediate re- productive strategies (e.g.. facultative planktotrophy), considered transitional and evolutionarily unstable by some theoretical modelers (Vance, 1973; Christiansen and Fenchel, 1979), are known or implied for several species of echinoids (Strathmann, 1979; Emlet, 1986) but seem to be generally rare among echinoderms. Phylogenetically proximate echinoderm species that are nearly indistinguishable by morphological criteria of- ten occur sympatrically and exhibit contrasting develop- mental strategies. With such strong similarities between species, these complexes provide useful systems for reli- able analyses of the interrelationship between different reproductive traits (e.g., egg size, larval type, juvenile size) and as such they have been the focus of several in- vestigations (Atwood, 1973; Menge, 1975; Lawson-Kerr and Anderson, 1978; Emson and Crump, 1979; Schei- bling and Lawrence. 1982; Emlet, 1986). A pair of morphologically similar asteroids of the ge- nus Porania co-occur in shallow waters of McMurdo Sound, Antarctica. The two types can be distinguished reliably by the position of the genital pores, which are found on the aboral surface in the recognized species, Porania antarctica (Perrier, 1894), and on the oral sur- face in the undescribed form, referred to here as Porania sp. This contrast was first recognized by H. E. S. Clark 77 78 I. BOSCH ( 1 963) and A. M. Clark ( 1 962), who argued that a higher taxonomic ranking was warranted for the undescribed form. However, their observations did not lead to the de- scription of a new species and at present the taxonomic standing of Porania sp. is unresolved (H. E. S. Clark, pers. comm.). This paper describes differences in reproduction and development between Porania antarctica and Porania sp. that substantiate Clark's (1962) recommendation of distinct species rankings for the two forms. The findings of this study are compared to general patterns of repro- duction in other echinoderms and discussed in the con- text of life-history evolution. Materials and Methods Adult sea stars were collected haphazardly by SCUBA divers at depths of 10-33 m beneath the annual sea ice in McMurdo Sound between Sept. 1984 and Dec. 1985. Immediately after collection, individuals were trans- ported to McMurdo Station in ice chests filled with am- bient seawater and maintained for up to several months in laboratory sea tables with flow-through seawater (-1.5°C). Spawning was induced by injection of 1-meth- yladenine(l-MAat 1 X 10~4M in seawater) into the coe- lom. For each species, the diameters of 30-35 eggs se- lected haphazardly from two spawning females were measured using a compound microscope equipped with an ocular micrometer. The annual fecundity of females was estimated by direct count of spawned eggs for Pora- nia sp. and by counting a known portion of the total spawn for P. antarctica. Spawned females were main- tained in the laboratory for at least one week after a spawning event and treated daily with 1-MA to assure that most mature oocytes had been shed. Embryos and larvae were reared at temperatures near their ambient (-1.5 to -1.0°C) in gently stirred or un- stirred 4 L culture vessels following procedures estab- lished by Strathmann (1971), as modified for Antarctic asteroid larvae by Pearse and Bosch (1986). Larvae of Porania antarctica were fed every two days with equal amounts of xenic Isochrysis galhana and Dnnaliella ter- tiolecta to a final concentration of 5- 10 X 103 cells/ml in 5 ^m-filtered seawater. No cultured food was added to vessels containing lecithotrophic larvae of Porania sp. Brachiolariae with well-developed rudiments were iso- lated in small glass dishes and induced to settle with sub- strates (e.g., shell debris, small rocks, sediment) collected from adult habitats. Since competent brachiolariae of P. antarctica were usually buoyant, glass slides covered with a bacterial-algal film were suspended horizontally in midwater to facilitate settlement. Results Distribution and spawning Porania antarctica were collected primarily at New Harbor (Fig. 1) during the austral spring and summer. 77°00' IS1 Ross Ice Shelf;,vX 78°00 163 Figure 1. Sites in McM urdo Sound, Antarctica, where Porania ant- arctica and Porania sp. were collected. 1984 and 1985, when weather and sea ice conditions per- mitted access to this remote site. Only a few specimens were collected at E. Cape Armitage and McMurdo Sta- tion, usually below depths of 30 m. Adults were large, measuring up to 70 mm from the tip of the longest ray to the center of the disc ( R ). The ovaries of all females (n = 11) examined during the austral spring, 1984 and 1 985, were small and contained only a few oocytes of the largest size class. Individuals (n = 9) collected in early February, 1985 did not respond to 1-MA; the dissected ovaries of one female were relatively small and devoid of fully grown oocytes. The remaining eight sea stars were maintained in laboratory sea tables until mid October and treated with 1-MA about every two months. Two of four females spawned in late September and the other two in early October. Males contained active sperm from early August to mid-October, as revealed by partial spawnings and biopsies of testicular lobes. Gametes were broadcast through aboral genital pores. Spawned oo- cytes, approximately 3-4 x 104 in number (n = 2 fe- males), were of moderate size (548 ± 10 ^m, n = 30), opaque yellow in color, and buoyant. Individuals of Porania sp. were generally rare and only 20 were collected during this study: 14 from East Cape Armitage, and 3, 2, and 1 from McMurdo Station, Cape Evans, and New Harbor, respectively. Adults were small, with R ranging from 10-30 mm. Females were induced to spawn with 1-MA in November (n = 2) and December (n = 1 ), 1 984, and in February (n = 1 ), April (n = 2), and July(n = 1), 1985. Males examined during each of these sampling dates contained active sperm. Although similar in size (554 ± 16/um. n = 35) and lipid content (50-60% of dry weight; J. McClintock, unpubl. data) to the eggs of Porania antarctica, the few eggs ( 100-3 10, n = 4) pro- duced by Porania sp. were free-spawned through oral DEVELOPMENT OF ANTARCTIC ASTEROIDS 79 Table I Embryonic and planktolrophic lan'iil development oj Porania antarctica reared at -1.5 to -1.0'C Developmental stage First appearance (days) Size* (mm) Zygote 0 0.55 Two-cell embryo 0.42 — Coeloblastula 3.2 — Hatched blastula 5.8 — Gastrula 12 0.70 Bi-lobed larva 20 0.76 Bipinnaria 22 0.80 Brachiolaria 40 1.1 Juvenile 65 0.79 * Refer to the diameter of zygotes and juveniles and the maximum length of all other stages. genital pores and immediately settled onto the bottom of culture dishes. Planktotrophic development of Porania antarctica With several important exceptions described below, the development from fertilization to metamorphosis of Porania antarctica is similar to that of the north temper- ate asteroid Porania pnlvillns (Gemmill, 1915) and gen- erally follows the typical pattern of asteroids with feeding larvae (Table I). Early cleavage divisions were slow at - 1 .5 to - 1 .0°C; morula and blastula stages were reached approximately 2.0 and 3.2 days after fertilization, respectively. By the fifth day of development, blastulae had developed a rich, active ciliary field. Hatched blastulae were highly buoy- ant and seemed to have little control over their vertical position; they usually floated near the water surface in stirred as well as unstirred culture vessels. Gastrulation by invagination was accompanied by substantial expan- sion of the larva, which attained a length of nearly 0.7 mm by the 1 2th day of development. Young bipinnariae were large, and judging from their coloration and buoyancy, retained a substantial propor- tion of the maternal yolk. At this stage the mouth is open and connects to the digestive tract, which is divided into an esophagus, stomach, and intestine. There followed a period of significant differentiation but of little apparent increase in size, as evidenced by the extension of the axo- hydrocoels, which by the 27th day had fused anteriorly forming a U-shape around the esophagus. After 32 days of development, an adhesive disc formed in the presumptive region of the brachiolarian complex. Brachiolarian arms were first observed on the 40th day of development. The various ciliated arms characteristic of planktotrophic asteroid larvae are lacking in Porania antarctica. Early brachiolariae measured 1.1 mm in length and grew very little to attain their final form (Fig. 2). Over the next 25 days, the brachiolarian arms in- creased gradually in size and developed an arrangement of adhesive papillae. Sixty-five days after fertilization, the larvae were competent to metamorphose. Some would swim along the bottom of dishes and temporarily attach themselves by their brachiolarian arms. However, most fully developed brachiolariae (83%, n = 53) were posi- tively buoyant and remained near the water surface. They attached and metamorphosed on the sides of dishes and underneath floating objects rather than on the bot- tom. Newly metamorphosed juveniles (0.79 ± .04 mm diameter, n = 9) had well-differentiated arms and re- tained the opaque, yellowish coloration of the egg. Lecithotrophic development of Porania sp. A chronology of development is given in Table II. Early cleavage followed the typical pattern of asteroids, but the rate of development was extremely slow; an early blastula stage was not reached until 12 days after fertil- ization. Embryos hatched as ciliated blastulae that were negatively buoyant and moved along the bottom of cul- ture vessels by ciliary action. Gastrulation was by invagi- nation, leading to the formation of a blastopore that was widely open at first, but gradually narrowed and disap- peared during later stages of development. The gastrula reached 0.6 mm by the 26th day of development, when a small lobe first became evident on the anterior end (Fig. 3A). Continued expansion in this fashion resulted in the formation of a generally ciliated, pear-shaped larva that consisted of a large, rounded posterior lobe and a narrow Figure 2. Feeding brachiolaria of Porania antarctica. The opacity of this larva is likely due to an abundance of yolk. Note the absence of ciliated larval arms, which are typically used in swimming and feeding by planktotrophic asteroid larvae. 80 I. BOSCH Table II Embryonic and lecithotrophic larval development oftheundescribed inii'ii'id Porania .«/>. reared at -J.5 lo-I.O'C Developmental stage First appearance (days) Size* (mm) Zygote 0 0.55 Hatched hlastula 1? — Earl\ gastrula 21 0.57 Late gastrula 26 0.60 Pear-shaped larva 38 0.76 Early hrachiolaria 47 0.80 Late brachiolaria 68 1.20 Juvenile 75 0.60 * Refer to the diameter of zygotes and juveniles and the maximum length of all other stages. anterior lobe and was entirely lacking feeding structures (Fig. 3B). These larvae were negatively buoyant and usu- ally swam near or on the bottom of stirred and unstirred culture vessels with the anterior lobe foremost and the anterioposterior axis in a horizontal attitude. Three bulbous arms first formed 47 days after fertiliza- tion: a single median anterodorsal arm, and a pair of ven- trolateral arms that occupy a region near the base of the narrow anterior lobe (Fig. 3C). Soon after, each arm was able to adhere temporarily to glass surfaces, and an adhe- sive disc had differentiated central to them. Therefore, the larva can be considered a modified brachiolaria. Fully developed brachiolariae (Fig. 3D) measured 1.2 mm in length. Attachment to a substratum was initially accomplished by the brachiolarian arms and the adhe- sive disc. Metamorphosis included the complete degen- eration of the anterior lobe, and lasted from one to two weeks. During this period, some larvae detached from the bottom and continued to swim for hours to days. Fi- nal attachment was facilitated by the tube feet of the ju- venile rudiment. The newly metamorphosed sea star, about 0.60 ± 0.03 mm across (n = 10), had two pairs of tube feet on each of its 5 arms. The arms were short and difficult to distinguish due to the presence of yolk on the aboral surface of the disc. Discussion In echinoderms, and particularly asteroids, there are considerable differences (e.g., egg size, larval morphol- ogy) between planktotrophic and lecithotrophic devel- opmental strategies (Strathmann, 1974; Emlet el at., 1987). One interpretation of this phenomenon is that types of development intermediate of planktotrophy and lecithotrophy are evolutionarily transitional and short lived (Vance, 1973). However, possible intermediate strategies do appear in some species, particularly among echinoids. Planktotrophic echinoplutei of Clypcastcr n>- seacens develop from relatively large eggs and do not re- quire particulate food to complete development to meta- morphosis (Emlet, 1986). A similar mode of develop- ment (i.e., facultative planktotrophy) is suggested for echinoplutei of the spatangoid Brisasterlatifrons(Stralh- mann, 1979) and judging from some egg sizes reported by Emlet el al. (1987) may occur in several other echi- noids (e.g., Sterechinus agassizi, Brisaster fragilis). Additional studies are necessary to adequately evalu- ate the ecological and evolutionary significance of the unusual development described here for Porania antarc- tica. Nonetheless, it is evident that features of both leci- thotrophic and planktotrophic strategies are manifested in this species. The larvae feed on bacteria (Rivkin el al.. 0.1. 0.4 mm Figure 3. Larval stages of Porania sp. (A) Post-gastrula 26 days after fertilization with a small anterior lohe. large posterior lobe, and blasto- pore. (B) Pear-shaped larva (38 days) showing increased development of the anterior lobe. (C) Early modified hrachiolaria 47 days after spawning. Three brachiolarian arms and an adhesive disc have formed on the preoral lobe. (D) Fully developed brachiolaria 68 days after spawning, ad, adhesive disc; al, anterior lobe; b. blastopore; ha. brachi- olarian arms. DEVELOPMENT OF ANTARCTIC ASTEROIDS 81 1986) but, unlike other planktotrophic asteroid larvae, they develop from buoyant eggs containing large lipid yolk reserves (J. McClintock, unpubl. data) and produce an unusually large juvenile at metamorphosis. The char- acters cited above may contribute to several other unique developmental features of P. antarctica. These include: (1) an absence of ciliated arms that aid swimming and feeding in other larvae (Strathmann, 1974), and (2) the metamorphosis of larvae almost exclusively on the sides of culture dishes and undersides of suspended glass slides. One possible advantage conferred on feeding larvae that develop from larger eggs is an abbreviated larval phase (Vance, 1973; McEdward, 1984;Emlet, 1986). Ac- cordingly, the period from fertilization to metamorpho- sis of P. antarctica is less than half that of another plank- totrophic antarctic asteroid, Odontaster validus (Pearse and Bosch, 1986). Indeed, at 65 days the developmental period of P. antarctica is similar to that of some temper- ate asteroids reared at their ambient temperatures (Emlet et ai. 1987). The production of a fully developed feeding larva from a larger, yolk-laden egg apparently requires less growth and differentiation. In low temperature envi- ronments such as the Antarctic, these differences are translated into a considerable reduction in development time, presumably leading to a concomitant decrease in the risk of mortality associated with a pelagic feeding lar- val phase. In contrast to Porania antarctica, eggs spawned by Porania sp. were negatively buoyant. The larvae were lecithotrophic, and judging from their distribution and swimming behavior in culture vessels most likely are de- mersal in nature. As in Porania sp., larvae ofAsterina minor also are heavier than seawater and pass through a pear-shaped brachiolaria stage (Komatsu et ai, 1979). Hatching of A. mmorbrachiolariae from an attached egg case occurs on the 9th day of development and is fol- lowed by a brief (ca. 1 day) demersal stage that precedes metamorphosis. In Porania sp., the free-swimming de- mersal larval phase lasts about two months. With the ex- ception of one other Antarctic sea star, Acodontaster hodgsoni (Bosch and Pearse, in press), this is the longest period of development to first metamorphosis reported for asteroids with lecithotrophic larvae (Emlet et a/.. 1987). In previous studies of morphologically similar, sym- patric asteroids with contrasting developmental strate- gies (Atwood, 1973; Lawson-Kerr and Anderson, 1978; Emson and Crump, 1979; Scheibling and Lawrence, 1982), distinction of the morphs or species has been made principally on the basis of egg size and embryonic or larval habitat (e.g., pelagic vs. brooded). By compari- son, the differences reported here for Porania sp. and P. antarctica (Table III), including sharp contrasts in larval form and mode of nutrition, represent a case of extensive Table III Summary oj major known differences between Porania antarctica ami Porania sp. Trait Porania antarctica Porania sp. 1) Morphology Aboral genital pores Oral genital pores 2) Adult size (R) Maximum 70 mm Maximum 30 mm 3) Timing of Seasonal (Sept.?-Oct.) Year-round reproduction 4) Fecundity ca. 35,000 100-310 (#eggs) 5) Larval type Modified planktotrophic, Lecithotrophic. pelagic demersal divergence. Such cases are relatively common in some invertebrate taxa, particularly polychaete annelids and gastropod molluscs (Perron, 1981; Hoagland and Rob- ertson, 1988), but seem to be unusual in the Echinoder- mata. Among related, sympatric echinoderms, those with larger adults tend to have an extended planktonic larval phase and produce relatively small juveniles at metamor- phosis. Small adult size is associated with brooding or an abbreviated non-feeding larval phase and the production of fewer, larger offspring which presumably have a greater chance of survival (e.g., Schoener, 1972; Menge, 1975; Lawson-Kerr and Anderson, 1978; Hendler, 1979). Strathmann and Strathmann (1982) reviewed several possible explanations for this phenomenon. One common explanation links brooding and other forms of non-pelagic development to energetic constraints on re- productive output (Menge, 1975). According to this hy- pothesis, small adults cannot produce sufficient offspring to benefit from the enhanced dispersal of a typically high- risk pelagic larval phase. In the Porania studied here, pe- lagic feeding development was associated with large adult size and high fecundity in P. antarctica. The smaller Porania sp. have considerably lower fecundity (ca. two orders of magnitude less) and develop non-pe- lagically. On the other hand, planktotrophic larvae of P. antarctica had a shorter developmental period and pro- duced larger juveniles. Therefore, Menge's (1975) hy- pothesis is applicable only if larval mortality rates in the plankton are much higher than near the bottom. Com- parisons of this type have not been made (see Strath- mann, 1985; Young and Chia, 1987). One possible con- sideration is the tendency for pelagic larvae to be swept by currents far from areas suitable for settlement and post-settlement survival, as shown, for example, for con- tinental slope ophiuroids by Gage and Tyler (1981). A second possibility is that predation is much lower on de- mersal larvae. However, possible larval predators such as polychaetes and small crustaceans are common in shal- low antarctic benthic habitats (Oliver, 1979; Marinovic, 82 I. BOSCH 1987). Indeed, several species, including the tanaid Noto- tanais dimorphus and the small infaunal actinian Ed- wardsia meridionalis often attain mean densities of more than 15,000 m'~ (Oliver, 1979). How a few larvae produced by Porania sp. escape these predators over sev- eral weeks of development and recruit successfully into benthic populations is a question of considerable in- terest. Acknowledgments I thank B. Marinovic, R. Britton, and J. Bernhard for assistance in collecting animals; H. E. S. Clark for invalu- able help in identifying sea stars; A. Russell for assistance with graphics; and the Antarctic Services Inc. of ITT, the Antarctic support services of the National Science Foun- dation, and the U.S. Navy Antarctic Support Force for their logistic support. M. Barker, J. McClintock, A. T. Newberry, and J. Pearse reviewed drafts of the manu- script and I thank them for their valuable suggestions. This work was supported by an NSF grant (#DPP- 83 1 7082) to John S. Pearse and by a Harbor Branch In- stitution postdoctoral fellowship. Harbor Branch Ocean- ographic Institution Contribution #713. Literature Cited Atwood, D. G. 1973. Larval development of the asteroid Echinaster echinophorous. Biol. Bull. 144: i-i i. Bosch, I., and J. S. Pearse. 1989. Developmental types of "shallow- water asteroids in McMurdo Sound. Antarctica. Mar. Biol.. in press. Chia, F.-S. 1974. Classification and adaptive significance of develop- mental patterns in marine invertebrates. Thalassia Jugosl. 10: 121- 130. Christiansen, F. B., and T. M. Fenchel. 1979. Evolution of marine invertebrate reproductive patterns. Theor. Popul. Bio/. 16: 267- 284. Clark, A. M. 1962. Asteroidea. British Museum Antarctic Exped. Rpls. Series B. 9: 1-143. Clark, H. E. S. 1963. The fauna of the Ross Sea. Part 3. Asteroidea. N. Z Dept. Sci. Ind. Res. Bull. 151: 1-84. Emlet, R., 1986. Facultative planktotrophy in the tropical echinoid Clypeaster roseaceus and a comparison with obligate planktotrophy in Clvpeaster subdepressus (Clypeasteroidea: Echinoidea). / Exp. Mar.Biol.Ecol.95: 183-202. Emlet, R., L. R. McEdward, and R. R. Strathmann. 1987. Echinoderm larval ecology viewed from the egg. Pp. 145-156 in Echinodenn Studies 2. M. Jangoux and J. M. Lawrence, eds. Bal- kema Press, Holland. Emson, R. H., and R. G. Crump. 1979. Description of a new species of Asterina (Asteroidea), with an account of its ecology. J. Mar. Biol. Assoc. U. K. 59: 77-94. Gage, J. D., and P. A. Tyler. 1981. Non-viable seasonal settlement of larvae of the upper bathyal brittle star Ophiocten gracilis in the Rockall Trough abyssal. Mar. Biol. 64: 153-161. Gemmill, J. 1915. The larva of the starfish Porania pulvillus. Q J Microsc. Sci. 61: 27-53. Hendler, G. 1979. Sex-reversal and viviparity in Ophiolepis keiri, n. sp.. with notes on viviparous bnttlestars from the Caribbean (Echi- nodermata: Ophiuroidea). Proc. Biol. Soc. H'ash. 92: 783-795. Hoagland, K. E., and R. Robertson. 1988. An assessment of poecilog- ony in marine invertebrates: phenomenon or fantasy? Biol. Bull 174: 109-125. Komatsu, M., V. T. Kano, H. Voshizawa, S. Akabane, and C. Oguro. 1979. Reproduction and development of the hermaphroditic sea star, Asterina minor Hayashi. Biol. Bull. 157: 258-274. Lawson-Kerr, C., and D. T. Anderson. 1979. Reproduction, spawning and development of the starfish Patiriella e.xigita (Lamarck) (Aster- oidea: Asterinidae) and some comparisons with P. calcar (La- marck). Aust. J Mar Freshwater Res. 29:45-53. Marinovic, B. 1987. Reproductive biology of the protogynous her- maphrodite Nolotanais dimorphus (Beddard), an antarctic tanaid. Masters Thesis, LJniv. of California, Santa Cruz, CA. Menge, B. A. 1975. Brood or broadcast? The adaptive significance of different reproductive strategies in the two mtertidal sea stars Lep- tastenas hexactis and Pisaster ochraceus. Mar. Biol. 31 : 87-100. McEdward, L. 1984. Some relationships between egg size and the al- lometry of larval growth in echinoid plutei. Am. Zoo/. 24: 46A. Oliver, J. S. 1979. Processes affecting the organization of marine soft bottom communities in Monterey Bay, California and McMurdo Sound. Antarctica. Doctoral Dissertation, Scripps Inst. of Ocean- ogr.. La Jolla. CA. Pearse, J. S., and I. Bosch. 1986. Are the feeding larvae of the com- monest Antarctic asteroid really demersal? Bull. Mar. Sci. 39: 477- 484. Perron, F. E. 1981. Larval biology of six species of the genus Conns (Gastropoda: Toxoglossa) in Hawaii, USA. .Mar. Biol 61:21 5-220. Rivkin, R. B., I. Bosch, J. S. Pearse, and E. J. Lessard. 1986. Bacterivory: a novel feeding mode in asteroid larvae. Sci- ence 233: 131 1-1314. Scheibling, R. E., and J. M. Lawrence. 1982. Differences in reproduc- tive strategies of morphs of the genus Echinaster (Echinodermata: Asteroidea) from the eastern Gulf of Mexico. Mar. BioL 70: 51-62. Schoener, A. 1972. Fecundity and possible mode of development of some deep-sea ophiuroids. Limnol. Oceanogr. 17: 193-199. Strathmann, R. R. 1971. The feeding behavior of planktotrophic echinoderm larvae: mechanisms, regulation and rates of suspension feeding. J. Exp. Mar. Biol. Ecol. 6: 1 09- 1 50. Strathmann, R. R. 1974. Introduction to function and adaptation in echinoderm larvae. Thalassia Jugosl. 10: 321-339. Strathmann, R. R. 1979. Echinoid larvae from the northeast Pacific (with a key and comment on an unusual type of planktotrophic development). Can. J. Zool. 57: 610-616. Slrathmann R. R. 1985. Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Ann. Rev. Ecol. Sysl. 16:339-361. Strathmann R. R., and M. F. Strathmann. 1982. The relation be- tween adult size and brooding in marine invertebrates. Am. Nat. 119:91-101. Vance, R. R. 1973. On reproductive strategies in marine benthic in- vertebrates. Am. Nat. 107: 339-352. Young C. M. and F.-S. Chia. 1987. Abundance and distribution of pelagic larvae as influenced by predation, behavior, and hydrogra- phy. Pp. 385-442 in Reproduction of Marine Invertebrates. Vol IX, A. C. Giese. J. S. Pearse and V. B. Pearse, eds. Blackwell/Boxwood, Palo Alto, CA. Reference: Biol. Bull. Ill: 83-95. (August. 1989) Bilateral Asymmetry in the Shell Morphology and Microstructure of Early Ontogenetic Stages of A nom ia simplex S. CYNTHIA FULLER, RICHARD A. LUTZ, AND YA-PING HU Institute of Marine and Coastal Sciences, Rutgers University, New Brunswick, New Jersey 08903 Abstract. Scanning electron microscopic examination of the valves of Anomia simplex larvae and postlarvae provides details of bilateral asymmetry in shell morphol- ogy and microstructure. Central provincular denticles and larger anterior and posterior hinge teeth develop in both valves. The umbo of the left valve becomes increas- ingly prominent with larval development, whereas no umbo is evident in the right valve. Thus, right and left valves of the prodissoconch differ markedly in height. In the left valve of the dissoconch, union of antero- and posterodorsal shell margins above the hinge causes lateral, exterior displacement of the umbo. The internal ligament extends ventrally to the right valve. Thin an- tero- and posterodorsal shell margins of the right valve extend centrally against the substrate and eventually unite exterior to the ligament. An increasingly larger bys- sal foramen, with a flexible, organic covering, is formed in the right valve. The outer layer of the left valve of the dissoconch is foliated calcite, whereas the outer layer of the right valve is composed of short, calcitic prisms. Inner shell layers consist of crossed lamellar and complex crossed lamellar microstructure, as well as myostracal prisms. Introduction The Anomiidae are noted for striking dissimilarity of the right and left valves. Bilateral asymmetry of the shell, which ultimately is an adaptation for secure attachment to the substrate (Yonge, 1977), begins in early larval stages. A prominent umbo characterizes the highly con- vex left valve; in contrast, an umbo is not apparent in the nearly flat right valve (Stafford, 1912; Miyazaki, 1935; Lebour, 1938; J0rgensen, 1946; Sullivan, 1948; Loosa- Received 28 November 1988; accepted 23 May 1989. noff et a/.. 1966; Chanley and Andrews, 1971; Yonge, 1977; Le Pennec, 1978; Booth, 1979). Larval hinge den- tition in the Anomiidae is taxodont; each valve has a se- ries of small, central teeth and 2-5 anterior and posterior teeth, which enlarge during the larval period (J0rgensen, 1946; Yonge, 1977; Le Pennec, 1978, 1980). In some an- omiids, formation of additional anterior and posterior provincular teeth in the left valve results in bilateral asymmetry of the larval hinge (Le Pennec, 1978, 1980). After metamorphosis, the right (lower) valve develops an expansive foramen through which a calcined byssus attaches to the substrate (except in some free-living an- omiids), while the left valve shows heightened shell growth along the antero- and posterodorsal margins, which eventually unite dorsally, above the hinge (Taylor et al. 1969; Yonge, 1977; Le Pennec, 1978; Prezant, 1984). This "supradorsal" growth in the left valve leads to lateral displacement of the umbo and ventral exten- sion of the ligament to the dorsal region of the crurum of the right valve (Yonge, 1977). [Crurum, used in the sense of Beu (1967) and Yonge (1977, 1980), refers to the prominent chondrophore in the right valve of the Anomiacea.] In some anomiids, calcitic foliated microstructure is the principal structural type in both right and left adult valves, while in other anomiids, the left valve is predomi- nantly foliated calcite, and the right valve is primarily prismatic calcite (Beu, 1967; Kobayashi, 1969; Taylor et al, 1969; Waller, 1978; Yonge, 1980). An inner layer of aragonitic crossed or complex crossed lamellar micro- structure surrounds the muscle scars in the left valve and the byssal foramen and muscle scar in the right valve (Taylor et al, 1969; Waller, 1978). As in other bivalves, the myostracum is prismatic aragonite (Taylor et al, 1969; Waller, 1978; Carter, 1980a). The present scanning electron microscopic study pro- 83 84 S. C. FULLER ET AL. vides a comprehensive description of the bilateral asym- metry in shell morphology and microstructure of early ontogenetic stages of Anomia simplex d'Orbigny, the common jingle shell of the western Atlantic. Details of shell morphological development are documented with micrographs of sequential stages. Comparison of early shell morphology in various North Atlantic anomiids re- veals features useful in studies concerning species identi- fication and taxonomy. A summary of the bilateral asymmetry in shell morphology and microstructure of early ontogenetic stages of four other common, inequi- valve pteriomorphs from the North Atlantic is included. Materials and Methods Adult specimens of Anomia simplex were collected in Wachapreague Inlet, Virginia, and were spawned by rais- ing the ambient water temperature from 25 to 30°C. Lar- vae were cultured in filtered (50 ^m mesh) baywater (sa- linity = 32-34 %o; temperature = 22.4-32.0°C) using standard techniques (Loosanoff and Davis, 1963). When larvae approached metamorphosis, a layer of eggshells was placed at the bottom of the culture tank. Spat on the eggshells and on the sides of the tank were attached loosely and were removed easily for sampling. Samples from larval cultures were treated with a 5.25% solution of sodium hypochlorite for 10 min to remove soft tissues (after Rees, 1950); disarticulated valves were rinsed with distilled water. Soft tissues were dissected from postlarval specimens with a small brush. The pli- able nature of postlarval right valves required careful placement as shells were mounted on silver tape. All specimens were coated with approximately 600 A of gold-palladium and were documented with an ETEC Autoscan scanning electron microscope. Positioning dis- articulated valves with selected points on the shell mar- gin aligned in a plane normal to the electron beam of the microscope resulted in consistent orientation for docu- mentation of shape. Photographs of a standard grid at the same magnification as each shell specimen provided accurate dimensions of shell features. Shell length is de- fined as the greatest anteroposterior shell dimension; shell height refers to the greatest measurement perpen- dicular to the hinge line. Terminology for juvenile and adult ligament components is that of Yonge (1977, 1 980). Microstructural terminology is taken from Carter (1980b) and Carter and Clark (1985). Microstructural varieties of inner shell layers were determined by exami- nation with reflected light and scanning electron micros- copy. Results Right and left valves of the prodissoconch I range from 75 to 80 Mm long (x ± SD = 76.4 ± 2.1; n = 10). The inequivalve nature of Anomia simplex is evident at a Figure 1. Scanning electron micrographs of disarticulated valves of Anomia simplex larvae. Numbers indicate shell lengths in ^m. shell length of 85-90 ^m, as a result of formation of a low, rounded umbo in the left valve (Figs. 1, 2). Left valves are equidimensional at approximately 140 /^m; shell height exceeds shell length during late larval stages. In the right valve, shell height is less than shell length throughout the larval period (Fig. 1 ). Provinculum length ranges from 54 to 62 /urn (x ± SD = 57.1 ± 2.8 pm; n = 18). A series of minute, irregular denticles extends across the central portion of the provin- culum in early larval valves; these interlocking denticles are slightly larger and more distinct in late larval stages (Fig. 2). Pairs of teeth, which also grow with larval devel- EARLY SHELLS OF ANOMIA SIMPLEX 85 Figure 2. Scanning electron micrographs of the hinge of disarticulated valves of Anomia simplex larvae seen in Figure 1. Numbers indicate shell lengths in pm. opment, form at the anterior and posterior extremes of the provinculum (Fig. 2). Articulation of the hinge teeth shifts gradually from a lateral to a dorsoventral orienta- tion; in the right valve at a length of approximately 155 nm, anterior and posterior provincular teeth extend dor- sally over the shell margin (Fig. 2). The first morphological evidence of postlarval devel- opment is formation of a ligament pit at a shell length of approximately 1 90 /urn, at which size central provincular denticles are mostly obliterated (Figs. 3, 4). Lengths of right and left valves are equal at this stage, but the left valve is approximately 24 ^m higher than the right valve. Supradorsal growth of the left valve begins at a shell length of approximately 240 ^m (Fig. 3). Bold margins of foliated microstructure expand first dorsally and then centrally across the umbo (Figs. 3, 5). By a shell length of approximately 1050 /urn, supradorsal extensions are as high as the umbo, and anterior and posterior provincular teeth are obsolete ( Figs. 3,4). Anterior and posterior shell margins extend approximately 200 nm above the hinge before they unite centrally (Fig. 6). Supradorsal union of the antero- and posterodorsal shell extensions is com- plete in specimens approximately 1600 /urn long (Fig. 3). The gap between the two margins is no longer visible in juvenile specimens approximately 5.5 mm long (Figs. 7, 8). At this size, several adult shell morphological features are recognizable, including the laterally displaced umbo, the differentiated inner and outer layers of the ligament, and three central muscle scars of myostracal prisms (Figs. 7-9). The inner layer surrounding the muscle scars is mostly fine and irregular complex crossed lamellar ara- gonite (Figs. 8, 10). The principal microstructural com- ponent of the left valve is foliated calcite (Figs. 7,11). A semicircular ligament pit, which is approximately 2 1 nm long, lies in the center of the hinge of the right valve of early postlarval specimens ranging from 190 to 86 S. C. FULLER ET AL. Figure 3. Scanning electron micrographs of disarticulated valves ofAmmiia ximplex postlarvae. Num- bers indicate shell lengths in ^m. 240 nm long (Figs. 3, 4). In specimens approximately 350 ^m long, anterior and posterior provincular teeth are obscured, and the ligament pit is positioned dorsally to the hinge (Fig. 4). At a shell length of 500-700 ^m. the anterodorsal and anteroventral shell margins are united immediately adjacent to the byssal notch but are sepa- rated anteriorly by the byssal foramen (Figs. 3, 12, 13). At a shell length of 700-900 ^m, the posteroventral re- EARLY SHELLS OF ANOMIA SIMPLEX 87 Figure 4. Scanning electron micrographs of the hinge of disarticulated valves ofAnoniui simplex post- larvae seen in Figure 3. Numbers indicate shell lengths i gion of the byssal foramen is covered by a flexible, or- ganic sheet, which is continuous with the adjacent pris- matic outer shell layer and is mostly calcined along the anterior edge (Figs. 14-16). Expansion of the anterodor- sal and anteroventral shell margins enlarges the byssal foramen anteriorly, while resorption of the shell extends Figures 5-11. Scanning electron micrographs of the left valve of postlarval specimens of Amnnia simplex. Figure 5. Posterior expansion of the anterodorsal shell margin across the umbo. Scale bar = 20 fim. Figure 6. Supradorsal extension of anterior and posterior shell margins. Scale bar = 100pm. Figure 7. Ligament and Supradorsal shell region after union of anterodorsal and posterodorsal shell margins in a juvenile specimen. Note the vertical pattern of outcropping edges of folia on the growth surface of the Supradorsal shell region. Scale bar = 100 ^m. POL, posterior outer layer of the ligament; AOL, anterior outer layer of the ligament; IL, inner layer of the ligament. EARLY SHELLS OF ANOAfIA SIMPLEX 89 the foramen in a posterior direction (Fig. 14). The extent of the covering, across the posteroventral portion of the foramen, remains constant during shell growth (Fig. 14). A calcitic outer layer of short prisms comprises most of the right valve of the early dissoconch (Fig. 17). An inner layer, which is mostly crossed lamellar aragonite, surrounds the byssal foramen and single muscle scar (Fig. 1 8). No obvious surface features were observed on a narrow band of the inner shell layer closest to the byssal foramen; a fractured section of this region is illustrated in Figure 19. No foliated microstructure was observed during thorough examinations of the right valve of post- larval specimens at several developmental stages. As the crurum increases in thickness and complexity, adjacent antero- and posterodorsal shell margins extend dorsally and centrally in a thin shell layer and are united at a shell length of 8.0 mm (Figs. 20, 2 1 ). In late postlarval stages, the crurum is comprised of several prismatic layers that overgrow and obscure the prodissoconch (Figs. 22, 23). Discussion Yonge (1977) described development of the ligament in the left valve of anomiids as a rounding of the anterior and posterior outer layers over the dorsal side of the in- ner layer. He pointed out that, in Anomia. the suprador- sally extended mantle lobes and antero- and posterodor- sal shell margins unite, but anterior and posterior outer layers of the ligament remain separate. As a result of in- tense growth in the supradorsal region of the left valve, the outer layers of the ligament lie laterally interior to the inner layer (Yonge, 1977). Development of the ligament in the right valve of an- omiids has been poorly understood; however, examina- tion of postlarval shells of Anomia simplex in the present study reveals a developmental process similar to that in the left valve. In the right valve, lateral (interior) displace- ment of the prodissoconch and dorsally extending liga- ment gives rise to the crurum (Fig. 3). Anterior and pos- terior outer layers of the ligament extend centrally around the crurum and thus are positioned laterally exte- rior to the inner layer and in vertical alignment with the corresponding layers in the left valve (Fig. 22). Conceal- ment of the prodissoconch during postlarval develop- mental stages is illustrated in Figure 22; layers of prisms mask the early shell and thicken the crurum. Shell morphological changes in the left valve of A. sim- plex during postlarval stages are similar to those de- scribed by Yonge (1977) for Pododesmus cepio (Gray). Early dissoconch growth is uniform along anterior, pos- terior, and ventral margins. Heightened dorsal extension followed by central growth across the umbo results in su- pradorsal union of shell margins, displacement of the umbo to an exterior position, and dorsoventral orienta- tion of the ligament. Shell morphological changes in the right valve of A. simp/ex during initial postlarval developmental stages also are similar to those described by Yonge (1977) for P. cepio in that the byssal foramen is formed between the anterodorsal and anteroventral dissoconch regions, which are united posteriorly. In the present study, docu- mentation of further postlarval developmental stages in A. simplex reveals that as adjacent shell margins expand, the byssal foramen is lengthened anteriorly, while shell resorption extends the foramen posteriorly. In late post- larval developmental stages, the anterodorsal shell mar- gin continues to expand slightly, but exaggerated growth of the anteroventral margin substantially increases the size of the byssal foramen (Figs. 3, 21). Further enlarge- ment of the byssal foramen to the adult size occurs by shell resorption (Jackson, 1890; Yonge, 1977). Similar expansion of the adult byssal notch is found in the pec- tinid Pedwn spondyloideiim (Gmelin) (Yonge, 1967; Waller, 1972). A sheet of decalcified prisms covers the posteroventral region of the byssal foramen in postlarvae of A simplex (Figs. 14-16). A similar structure has been described in some of the Propeamussiidae (Waller, 1984), a family with close ancestry to the Anomiacea (Waller, 1978). The covering in these glass scallops is an extension of the prismatic outer layer of the right valve; with growth of the byssal fascicle, the covering becomes "an applique on the surface of the fascicle" (Waller, 1984). Several features of the shell morphology and internal anatomy of adult A. simp/ex and Anomia ephippium L., the common jingle shell of Europe, are similar (Yonge, 1977, 1980). Some authors (e.g., Fischer-Piette, 1973; Porter, 1974; Rios, 1985) have listed these two species as synonymous, while others (e.g., Dall, 1898; Olsson and Harbison. 1953; Abbott, 1974) preserve separate species. Comparison of larval and early postlarval shell morpho- logical features of A. simplex with those of A. ephippium, described by Le Pennec (1978, 1980), suggests similarity in the two species. Shell length during the straight-hinge developmental stage is 75 ^m in A. ephippium (Le Pen- Figure 8. Shell after supradorsal union of anterior and posterior shell margins; specimen is the same as that in Figure 7. Scale bar = 1 .0 mm. Figure 9. Fracture section through the prismatic layer of the dorsal muscle scar. Scale bar = 5 jim. Figure 10. Fracture section through the inner shell layer in the region adjacent to the dorsal muscle scar; note complex crossed lamellar microstructure. Scale bar = 5 ^m. Figure 11. Folia of the interior surface of the outer shell layer. Scale bar = 5 urn. 15 Figures 12-16. Scanning electron micrographs of the right valve of postlarval specimens of Anemia simplex. Figure 12. Exterior surface of a shell 620 nm long. Arrow marks region of detail in Figure 13. Scale bar = 100 ^m. PRO, prodissoconch; DIS. dissoconch; BF. byssal foramen. Figure 13. Prismatic secretions at the byssal notch. Scale bar = 5 ^m. Figure 14. Interior surface at three sequential stages to illustrate enlargement of the byssal-foramen covering by shell resorption. Arrow marks the same spot on the three specimens, which are figured to scale. Scale bar = 100 Mm. 90 EARLY SHELLS OF ANOMIA SIMPLEX 91 nec, 1978); initial shell length of A. simplex ranged from 75 to 80 /urn in the present study, although individuals as small as 58-60 nm long have been reported (Loosanoff et cil., 1966; Chanley and Andrews, 1971). Le Pennec (1978) described formation of an early umbo in the left valve of A ephippiunt at a length of 100 pm and promi- nence of the umbo at a length of 1 10 ^m. In specimens of the left valve of A. simplex depicted in Figures 1 and 2, an umbo is evident at a shell length of 89 ^m and is well-developed at a shell length of 1 17 ^m. Chanley and Andrews (1971) described development of a rounded umbo in the left valve of A. simplex specimens 90-1 10 nm long. Larval hinge dentition in both species consists of a central region of denticles and a pair of larger teeth at the anterior and posterior ends of the provinculum. In the left valve of A. ephippium, a third tooth is added at the anterior and posterior extremes of the hinge at a shell length of approximately 100 ^m (Le Pennec, 1978, 1980). A third tooth is evident in the same position in A .simplex when shells are 117 ^m long (Fig. 2), but these new teeth are not well-developed in either species. In A. ephippium, formation of a ligament pit occurs when shell height and shell length average 190 ^m (Le Pennec. 1978). In specimens of A. simplex in the present study, a ligament pit is formed in shells 193 ^m long; these post- larvae also are equidimensional. Previous workers re- ported that metamorphosis in A. simp/ex occurred when animals were 180-215 /urn long (Loosanoff el ai, 1966; Chanley and Andrews, 1971). Finally, at a shell length of 250 nm, right and left valves of A. ephippium differ by 60 ^m in length and by 70 fim in height (Le Pennec, 1 978). Inequality of dimensions of right and left valves of the dissoconch of A. simplex is illustrated in an articulated specimen approximately 485 ^m long (Fig. 24). Right and left valves of this specimen differ by 70 ^m in length and by 50 nm in height. Despite similarities in shell morphologies of early on- togenetic stages in the two species, examination of shell microstructure of early stages reveals a significant differ- ence between A. simplex (= glabra) and A. ephippium and a basis for taxonomic separation (Jackson, 1 890). The right valve of the dissoconch of A. simplex is com- prised largely of prismatic microstructure, and the left valve is predominantly foliated (Jackson, 1 890; present study). On the other hand, both valves of the early disso- conch of A. ephippium are predominantly foliated; pris- matic microstructure is confined to a relatively thin outer layer of the right valve (Carpenter, 1848; Jackson, 1 890). Clearly, further studies are necessary to determine the genetic distance between A. simplex and A. ephip- pium. Historically, presence of a byssal notch has been a ma- jor distinguishing character used in the identification of early ontogenetic stages of anomiid species from the North Atlantic. Previously, a byssal notch was found in only the right valve of A. simplex and A. ephippium. whereas Anomia squamula (= aculeata) L. and Anomia patelliformis L. have a notch in both right and left valves (the notch in the left valve is shallower than the notch in the right valve) (Jackson, 1890; Ranson and Desjardin, 1941; J0rgensen, 1946; Merrill, 1962; Yonge, 1977; Le Pennec, 1978). However, data in the present study indi- cate that a notch sometimes occurs in the left valve of A. simp/ex (Fig. 25). Because of this variability, use of this feature for species identification may result in error. Other details of early shell morphology of. 4. squamula and A. patelliformis are not well-documented. Because larval anomiids are distinctive and easily separated from other bivalve larvae by their bilateral asymmetry and conspicuous byssal notch [or pedal sinus (Yonge, 1977)], most descriptions of larval stages are limited to these two features. However, larval hinge dentition has been de- scribed for. 4. squamiila; the provinculum of this species has central denticles and 3-5 larger teeth at the extremes ( Jergensen, 1 946). The anterior and posterior hinge teeth of A. squamula are more pronounced (see J0rgensen, 1946, Fig. 162) than those of A. simplex; larval hinge morphology, therefore, would provide a reliable means of distinguishing these two sympatric species during planktonic stages. Jackson ( 1 890) described the prodissoconch of A. sim- plex ( = glabra) as having "homogeneous" microstruc- ture and fine commarginal lines on the exterior surface. He described the left valve of the dissoconch as "subna- creous" (foliated), with an inner "porcelaneous" region surrounding the muscle scars, and the right valve of the dissoconch as "prismatic," with a white, "porcelaneous" band around the byssal foramen. The previously re- ported foliated and prismatic microstructures of the outer layers are confirmed in the present study; secretion of these layers begins at the clearly delineated prodisso- conch-dissoconch boundary (Figs. 12, 24, 25). In addi- tion, reflected light and scanning electron microscopic examination of the inner shell layers has enabled catego- rization of the "porcelaneous" microstructure. Most of the surface of the inner shell layer of the right valve of the early dissoconch has well-defined first order lamellae, and the microstructure is distinctly crossed lamellar. Ex- Figure 15. Interior surface view of the byssal-foramen covering of a specimen 1.5 mm long. Scale bar = 100 nm. CR, crurum. Figure 16. Interior surface at the junction of the byssal-foramen covering and the anteroventral shell margin. Scale bar = 10 ^m. Figures 17-23. Scanning electron micrographs of the right valve of postlarval specimens of Anomia simplex. Figure 17. Interior surface view of the prismatic outer layer. Scale bar = 5 urn. Figure 18. Surface view of the crossed lamellar inner layer. Scale bar = 20 urn. Figure 19. Fracture section of a specimen 2.0 mm long through the region of the inner shell layer closest to the byssal foramen. Scale bar = 5 ^m. Figure 20. Hinge region of the right valve before supradorsal union of shell margins. Scale bar = 50 ^m. 92 EARLY SHELLS OF ANOM1A SIMPLEX 93 Figures 24-25. Scanning electron micrographs of postlarval specimens ofAnomia simplex. Figure 24. Lateral view of articulated shell. Scale bar = 100 pm. RV. right valve; LV, left valve. Figure 25. Exterior surface of two left valves to show variation in byssal notch structure. Scale bar = 100 jum. PRO, prodissoconch; DIS. dissoconch. amination of the fractured section of the interior region of this inner layer (depicted in Fig. 19) indicates a variety of complex crossed lamellar microstructure with a low dip angle (J. Carter, pers. comm.). The inner layer of the left valve of the early dissoconch has fine and irregular complex crossed lamellae. The right valve of A sqitannda(= aculeata) also is pre- dominantly prismatic calcite (Jackson, 1890). In con- trast, both valves of adult specimens ofAnomia trigo- nopsis Hutton examined by Beu (1967) and of A. ep/iip- pium and Anomia peruviana d'Orbigny examined by Taylor et al. (1969) were mostly foliated calcite. Few studies subsequent to those conducted by Jackson ( 1 890) have reported prismatic microstructure in the right valve in Anomia. Beu (1967) found that the right valve of three species of Palm is comprised primarily of prismatic mi- crostructure and suggested that this character generally separates Patro from Anomia. Although Palm and Ano- mia have several other morphological differences in shell and ligament structure (Yonge, 1980), microstructure of the right valve is not a reliable character for separating these two genera. Several of the bilaterally asymmetric early shell fea- tures characteristic of A. simplex are seen in other pterio- morphs common in the Northwestern Atlantic. Prodis- soconch II specimens of Crassostrea virginica (Gmelin) and Ostrea edulis L. show greater convexity and umbo- nal protrusion in the left valve than in the right valve (Carriker and Palmer, 1979; Carriker ^^ ^ r,^t^ ^ |®^ m •.> -• f .*•' ^» " 1? » I *.* -»- 1%"- e^-'.. . M'M f^-VjiV •/•»•%' ^m fTlif "*$i, & r>s* H '^•-., ^f "<,*«?;' •*iS ', *.^.P>:C. «''•»••• \iii»"--. * r.--^--^. ,- _ i ^^^ Figure 3. A. Salt pond scum community dominated by sulftde oxidizing bacteria to the west of stake marked 4 on inset of Figure 1 . B. Salina Bido, Matanzas Cuba, site of collection of the Cuba strain of Paratetramitusjugosus. Mangrove trees form the barrier between the mats and the open ocean. C. Microco- leus mat from Salina Bido, hand sample. With the help of two Earthwatch teams (1983) the La- guna Figueroa sites were extensively surveyed and marked with a permanent stake ("metal surveyor's stake," Fig. 1 ). Wooden stakes were placed as indicated by the open circles, including three reference stakes ("stakes A, B, C," Fig. 1). Sediment samples from the five sites were placed on petri plates to enrich for P. jugosus in early May; the experiments were repeated six weeks later in June 1983. Collections involving samples from the five different sediment types at Pentapus were twice made again in March and October 1988. Agar plates were prepared with two kinds of thinly poured sterile enrichment media: "modified-K" and "manganese-acetate" (MnAc) (Margulis el a/., 1980; Read et al, 1983). These plates were taken to the field study site where a sample about 1 mm3 of each of the five sediment types was placed directly at the center of both kinds of sterile plates. Immediately after samples were in place, they were cov- ered with approximately 1 ml of sterile distilled water to suspend the organisms and initiate reproduction and per- haps excystment. When vigorous growth was evident, plates were monitored and scored for the presence of cysts and amoebae. These were subcultured as needed onto fresh medium by streaking with a sterile platinum loop. The organisms were routinely grown at room temper- ature on either modified K or MnAc media, both of which contain half-concentrated seawater. The food source for P. jugosus is a gram-positive, flagellated, facul- tatively aerobic rod (designated "B bacillus") that grows readily on both media. Because it is morphologically in- distinguishable, the B bacillus is likely to be a strain of the organism reported by Gong-Collins (1986). Maintenance and storage of stock cultures is described in Read et al. (1983). Light microscopy Culture slides. Hanging-drop culture slides were made for observation of live P. jugosus. A very small drop of LIFE HISTORY OF P. JUGOSUS 115 Table II Paratetramitus jugosus on field plates Observation scored on Day 1 Day 8 Day 10- 11 Day 16-18 Day 19 Day 28 Day 30 MEDIUM K MnAc K MnAc K MnAc K MnAc K MnAc K MnAc K MnAc Sites 1 . Gypsum crust 2. Gypsum mat 3. Mic rocoleus mat 4. Thiocapsa scum - -,:,- +,+,+ +,+,+ +,+,+ 1 . Gypsum crust 2. Gypsum mat 3. Microcoleus mat 4. Thiocapsa scum 5. black mud - :": : ++, - +, +, N ++:- +:+ :~: 7 -, F +, - ;+ £ Top: May-June 1983 Bottom: March 1988. except for day 28 which corresponds to August experiment Key: -, no or very few cysts; +, cysts abundant; ++ at 25X magnification every field on plate has cysts; N, F nematodes, fungi obscure readings Notes: Acanthamoeba sp. cysts (larger and more crenulated than those of Paratetramitus) also tend to appear on plates where P. jugosus cysts are abundant. Commas between entries indicate entirely different sets of experiments; each entry represents the value on 1 to 4 plates. MnAc medium was placed on a glass coverslip, inocu- lated with amoeba cysts, and then covered with a depres- sion slide. Petroleum jelly was used to adhere and seal the coverslip to the slide. Fresh preparations were viewed immediately with a Nikon Diaphot inverted microscope. The excysting amoebae grew on the coverslips mounted over the inverted glass depression slides. After amoebae were detected with the inverted scope, the slides were flipped upright and observed at higher magnification with the Nikon Fluophot phase-fluorescence micro- scope. Amoebomastigote transformation. Monoprotist cul- tures were grown from fresh microbial mat material col- lected by the above method at Salina de Bido, Matanzas, Cuba. The cultures, started within two weeks of collec- tion, contained various types of bacteria, including food bacteria, and were used to obtain amoebomastigote transformation. Plates were flooded with distilled water and allowed to stand for 10 min. The water was then pipetted offand placed in a sterile test tube. Several drops of this aqueous suspension were placed on a K plate, and the plate was monitored for the presence of amoebae and mastigotes by light microscopy over a 48-h period. Nuclear fluorescent staining Fixation. Cultures necessary for monitoring the repro- ductive processes of P. jugosus had to contain growing amoebae. Active trophic amoebae were acquired by the following technique: five glass coverslips were aseptically placed in a circle on a MnAc plate. The center of the circle made by the coverslips was then inoculated with a monoprotist culture of P. jugosus and flooded with ster- ile MnAc medium (lacking agar) up to the edge of the coverslips. One to two drops K medium (lacking agar) were added to the surface of the five coverslips. The amoebae were incubated at room temperature for one to four days, or until the growing edge of the culture could be detected on the surface of the coverslips using an in- verted microscope. The coverslips were removed and im- mediately immersed in Columbia jars containing a mod- ified Carnoy's fixative (70% ethanol, glacial acetic acid, in a 3:1 ratio). The coverslips containing the amoebae were then rinsed twice in 70% ethanol and stored at 4°C in 70% ethanol until they were stained. Unlike the agar culture slides used previously (Read et ai, 1983) it was not necessary to coat the coverslips with agar. No step in the staining procedure was needed to insure the adherence of the amoebae to the coverslips. Apparently enough adhesive substance from the K me- dium and any proteinaceous substances secreted by or from lysed bacteria caused excellent adhesion of fixed amoebae. Eliminating the necessity of coating coverslips with Parlodion® or other substances decreased the amount of background debris including spurious fluo- rescence in stained samples. Staining. Two fluorescent DN A stains were used in this study: 4'-6-diamidino-2-phenylindole (DAPI) and mith- ramycin (both purchased from Sigma Chemical Co., St. Louis, Missouri). Staining procedures were those of Cole- man et al. (1981). Stain concentrations were 0.5 Mg/ml DAPI in Mcllvaine's pH 4.4, or 50 Mg/ml mithramycin in Mcllvaine's pH 7.0 with 10 mM MgSO4. Fixed coverslips were rehydrated briefly through an ethanol series, washed twice in dH2O, and then twice in the appropriate buffer. 116 M. ENZIEN ET AL Figure 4. Photomicrographs of live Paratetramitus jugosus. A. Amoeba with visible nucleus and sev- eral small spherical and irregular-shaped bodies. Cyst in bottom left corner. B.C. Small spherical bodies frequently found in cultures. D-H. Sequential photos showing movement of small amoeba (arrow). The small amoeba closely resembles many other small irregular objects in our cultures. (Compare 4C with 4G) Bar scale = 10 jim. Coverslips were placed on blotting paper, sample-side up, and flooded with approximately 200 ^1 of stain solution. A clean glass slide placed over the coverslip was turned over and allowed to stand for one to three hours in the dark. The slide was then blotted dry and sealed around the edges with clear nail polish. Some slide preparations were treated with pancreatic RNase before staining to re- move background binding of DAPI to RNA (Coleman et a/.. 1981). Samples treated with bovine DNase I (Coleman el a/., 1981) allowed distinction of stained DNA from LIFE HISTORY OF P. JUGOSUS 117 Figure 5. Phase/fluorescence micrographs of DAPI-stained amoebae. A,B. Amoeba with metaphase nucleus. Phase micrograph shows nucleus clearly intact and under fluorescence, DNA is seen condensed in the center. C,D. Amoebae with telophase nucleus. D,F. Cytoplasmic DNA is faintly visible in the amoe- bae. E,F. Arrow indicates a structure which may be a chromatin body extruded by the amoeba. Its size and fluorescent properties differ from cytoplasmic fluorescence. Bar scale = 10 ^m. nonspecific brightly staining material in the preparations. Unstained amoeba slide preparations were also used as controls to detect and distinguish autofluorescence from authentic DNA staining. Excitation filter sets on the Ni- kon Fluophot in the UV and blue (approximately 365 nm and 490 nm) were used for epifluorescence observations of DAPI and mithramycin samples, respectively. Photo- micrographs were made with a Nikon Microflex AFX photomicrographic attachment. Films used for photomi- croscopy were Scotch 640 ASA tungsten film and Kodak Tri-X 400 ASA pushed to 1600. Electron microscopy Fixation and embedding of the Cuba strain of P. jugo- sus. Amoebae inoculated into aqueous suspension (de- 118 M. ENZIEN ET AL Figure 6. A,B. Phase/fluorescence micrographs of DAPl-stained amoebae with nuclei in telophase. C- H. Phase and fluorescence; Autofluorescence with blue filter (490nm) of small bodies, amoebae, and cysts. Bar scales = lOjim. scribed above) were allowed to grow for two days. Several milliliters of distilled water were added to the suspension and 3 ml of this water were transferred to a centrifuge tube to which 0.15 ml of glutaraldehyde was added to make the final concentration of the fixative 5%. After fixation overnight at room temperature and pelleting in a clinical centrifuge, the samples were washed twice in 0. 1 M cacodylate buffer (10 min each wash). The pellet was post-fixed for 10 min in 2% osmium tetroxide, washed twice with distilled water, and then stained with 0.5% uranyl acetate for 30 min. Samples were then washed with distilled water (10 min) and dehydrated through an ethanol series as follows: 70% (5 min), 80% (5 min), 90% (5 min), 95% (10 min), 100% 3X 10 min. For embedding, the pelleted sample was placed in pro- pylene oxide for 30 min, followed by 30 min in a 1:1 mixture of propylene oxide: Spurr's resin. After transfer- ring to a beam capsule, it was recentrifuged and left in Spurr's for 5 h after which the spent Spurr's was poured off and fresh Spurr's added; the resin was polymerized for 16hat70°C. Mat material from North Pond, Laguna Figueroa, Baja California, was fixed at the site according to the methods of Stolz( 1983). Sectioning and observations. The blocks were sec- tioned by Floyd Craft (Boston University) on a Porter Blum MT2B ultramicrotome, post-stained with 2% ura- nyl acetate for 15 min, lead citrate for 5 min, and then observed on a Philips model 410 transmission electron microscope at 20 kV. LIFE HISTORY OF P. JVGOSVS 119 Figure 7. A-D. Phase and fluorescence; DNase treatment of amoebae stained with mithramycin. Nuclear and cytoplasmic fluorescence is completely removed. E,F. Phase/fluorescence; DAPI stained amoeba and associated small spherical body (arrow). Small body has stronger fluorescence than the amoeba nucleus. G,H. Fluorescence and phase; Mithramycin staining clearly showing nucleus of amoeba and possibly other DNA containing organelles inside cell. I-N. Phase and fluorescence; Mithramycin staining of small spherically and irregularly shaped bodies. The fluorescence of these bodies is stronger than the nuclei of the amoebae. Bar scale = 10 nm. Isoenzyme analysis Slant cultures of the amoebomastigote isolate from Cuba were sent for isoenzyme analysis to the American Type Culture Collection (ATCC, Rockville, Maryland). Starch gel electrophoretic techniques for isoenzyme pat- terns were conducted by Tom Nerad (Nerad and Dag- gett, 1979). The Cuban strain was tested for three isoen- 120 M. ENZIEN ET AL Figure 8. A. Cysts of cultured Paratetramitus jugnsus in many ditferent stages: from completely desiccated to recently excysting. Ar- row indicates a nucleolus in a nucleus (N). Small bodies (b) are also seen which may represent released chromatin bodies or equivalent that develops into small amoebae. Bar scale = 5 /mi. B. Cyst showing inner and outer walls, mitochondria (m). storage granules (s). and chromatin body (C) in division (arrow). Bar scale = 2.5 ^m. Transmission electron micrograph (TEM) of material from Cuba strain ofP.jugosus. zyme systems: proprionyl esterase (PE), leucine amino- peptidase (LA), and acid phosphatase (AP) (Daggett and Nerad, 1983). Results and Discussion Field studies More than eight sets of field studies were undertaken in 1979 and 1980 after the discovery of the extraordinary abundance of small cysts on agar plates designed to de- tect the presence of algae, cyanobacteria, manganese-ox- idizing bacteria, and other organisms. In earlier studies, lack of consistency among samples, presence of nema- todes and contamination, especially by an unidentified black manganese-oxidizing fungus, precluded orderly collection of data. The last four sets of observations ( May and June 1983; March and August 1988) led to repeat- able results (Table II). The appearance of cysts smaller than those of Acanthamoeba was taken as putative evi- dence for P. jugosus. Three or four times (by high power light and once by electron microscopy) we verified that the cysts indeed were P. jugosus. Here we summarize the general experience after the eight sets of experiments in which each medium (K, MnAc) was represented by two to four plates per site. P. jugosus, or encysting small amoebae indistinguishable morphologically from P. jugosus, are invariably present in the laminated sediment when the Thiocapsa layer (the red layer. Fig. 21) is well-developed. P. jugosus is usually recoverable from the laminated Microcoleus mats such as that depicted in Figures 2E and 2H. P. jugosus popula- tions do develop but less frequently in the gypsum mat and black mud samples. They are least frequent in the white gypsum crust. Our practice now is to collect mats with Thiocapsa scums to insure recovery of large populations of healthy P. jugosus cysts, amoebae, and in the case of the Cuba samples, amoebae that readily transformed to mastigotes (Fig. 3C). In order to see abundant populations of P. jugosus, plates should be read several times, especially between 10 and 30 days after returning from the field. After this, plates become overgrown with many kinds of bacteria and some fungi as reported by Brown et al. (1985). Ex- cept for the common presence of Acanthamoeba, no other protists have been routinely seen on K plates. It is important to control the quantities of added distilled water, however; if water is too abundant a plethora of encysting ciliates appear that have not been studied. Due to the more limited nutrients on MnAc plates algae (for example the tiny encysting chlorella-like Mychonastes desiccatus BROWN, Margulis et al.. 1988) may appear on plates incubated in the light. We have always used two sets of media because more abundant growth of P. jugosus (and most other mi- crobes) develops on K plates whereas these amoebae are more easily recognized on the less permissive MnAc plates. We score the presence of P. jugosus at a site (e.g., "+" in Table II) only when the small cysts are present on all of the plates of both media. We assess P. jugosus to be a normal component of the microbial mat community whose population develops especially well in the layer dominated by Thiocapsa, be- low Microcoleus. Our data are consistent with the inter- pretation that during the spring rainy season, enormous populations develop and during the hot dry summer when halite and gypsum crystals dominate the mat sur- face, P. jugosus survives by encysting. We have observed excystment in 10-20 min. Upon desiccation, encyst- ment apparently takes place rapidly as well (certainly LIFE HISTORY OF P. JUGOSUS 121 |p^|-w.7^vv?*! ? •••••/•* ?H«f': 1 Figure 9. Healthy trophic amoebae, TEMs. A. Highly vacuolated (v) amoebae in a dense population of food bacteria. Small bodies (arrows) could be products of amoebae fission. Bar scale = 1 0 ^m. B. Amoeba with uroid(U), nucleus (N), vacuoles(v)and mitochondria (M); bacterial spores (S). Bar scale = 2.5 overnight). Because anaerobic bacteria abound in the Thiocapsa mat layer, it is also likely that both cysts and amoebae ofP.jugosiis tolerate low oxygen or even totally anoxic conditions. To develop techniques to most reli- ably detect the greatest number of small cysts on plates by low power microscopic examinations, we compared flooding the agar plates immediately upon collection of the field samples to drying them out entirely before flooding. Because early drying out selects for rapidly growing fungi and spore forming bacteria which may overgrow the amoebae, our best results were with sam- ples flooded with 1 ml distilled water in the field and per- mitted to dry for the weeks after collection. In conclu- sion, the optimal collecting conditions for P. jitgosus in- volve recognition of small cysts from plated samples of 1-mm cubes of distilled water-flooded Thiocapsa layer mat samples. The cysts should be recoverable on both K and MnAc media from 20 to 30 days after return from the field. For further purification to monoprotist cultures see Read?/ al. (1983). Light microscopy Culture slides. Observations of live material never re- vealed amoebae in division. Conditions for normal pro- mitotic amoebic division may not have been favorable under the growth conditions used, yet small bodies were frequently observed (Fig. 4A-H). These bodies were also seen frequently from cultures grown on agar plates with both light and electron microscopy (Fig. 6C, D; 7I-N; 8A; 9A). Many of these small bodies may be pieces of cytoplasm left behind by the amoeba or in EM sections through a small portion of the amoeba, however, on one occasion, a small body was observed budding off a parent amoeba followed by changes in its shape and monopo- dial movement (Fig. 4D-H) suggesting this body was a small amoeba. We cannot rigorously preclude the possi- bility that this small body was already present in the cul- ture and that the larger amoeba passed over it, making it look as if it were extruded. Amoebomastigote transformation. Aqueous suspen- sions examined 24 h after their preparation were ob- served to contain both amoebae and cysts. When the dis- tilled water suspension was plated, even six weeks after its preparation, it gave rise to viable P. jugosus amoebae and mastigotes; however, whether cysts, mastigotes, or amoebae dominated the suspension was not determined. Amoebae were dominant 24 h after plating, while masti- gotes dominated after 48 h. Mastigotes were conspicuous enough in these suspensions to warrant harvesting the cultures for electron microscopic analysis of kinetid structure. 122 M. ENZIEN ET AL '' •• . ^ ¥\ . ?-*?•- ~;is» M - .•.s.f+fX- • •- -,.^/S^, v.^: - ; - .- .-,.,;•:,. , ±^ ^: B > . :;^i . . . , ' ,5 it' . r ' ' ••• Figure 10. Rounded amoebae, TEMs. A. Dumbbell-shaped chromatin bodies (C) with eondensed ma- terial (arrow), which are clearly distinguishable from mitochondria (M). Bar scale = 1 .0 pm. B. Ribosome- LIFE HISTORY OF P. JUGOSUS 123 -.^a^r Figure 11. Amoeba with small bleb (arrow) off nucleus (N). Mito- chondria (M) are also seen; chromatin bodies are lacking. TEM. Bar scale = 1.0 ^m. Nuclear fluorescent staining. In both DAPI and mith- ramycin-stained preparations, the nuclei of amoebae were clearly visible with epifluorescence microscopy (Fig. 5; 6 A, B; 7E-H). Two stages of nuclear division were revealed: telophase (Fig. 5C, D; 6A, B) and meta- phase (Fig. 5 A, B). Cytoplasmic DNA was also seen in these preparations although less clearly. DAPI-stained slides tend to show high background fluorescence, mak- ing cytoplasmic DNA difficult to see (Fig. 5B, D, F; 7F). This problem has been reported in studies of other pro- tists as well (Coleman el a/.. 1981 ). Because mithramycin stains DNA more specifically, preparations made with this stain displayed less background fluorescence (Fig. 7G) and allowed for clear identification of cytoplasmic DNA. Autofluorescence and DNase controls of amoebae showed no nuclear or cytoplasmic fluorescence verifying the presence of DNA (Fig. 6E-H; 7A-D). Fixed amoeba cultures always contained small irregu- lar and spherically shaped bodies that displayed stronger fluorescence than the nuclei of amoebae, cytoplasmic DNA or the nucleoids of the bacteria upon which the amoebae feed. Strong fluorescence was clearly removed by DNase treatment, indicating that at least some of the small bodies contain DNA. A large number of the small spherical bodies (Fig. 6G, H) and some of the irregularly shaped bodies (Fig. 6C, D) were autofluorescent. Small spherical bodies may be nuclei of ruptured cells or dried- out collapsed cysts. However the amount of autofluo- rescence in some of the irregularly shaped bodies was in- sufficient to account for the strong fluorescence seen in stained material (Fig. 7I-N). Several small bodies seen in close proximity to normal-size amoebae (Fig. 5E, F; 7E, F) differ in both extent and intensity from nuclear and cytoplasmic fluorescence: the intensity of fluorescence is greater and the fluorescent structures are intermediate in size between large nuclei and cytoplasmic DNA. Cy- toplasmic DNA in DAPI and mithramycin stained prep- arations is most easily interpreted to be due, at least in part, to mitochondria! nucleoids. Chromatin bodies (de- scribed below) detected in electron micrographs are ap- proximately the same size as mitochondria therefore DNA in these structures might also be represented by cytoplasmic fluorescence. However, the more diffuse cy- toplasmic DNA (regardless of its origin) stains less brightly and its fluorescence emanates from smaller structures relative to whatever DNA is causing the fluo- rescence of the five small bodies (as seen e.g., in Fig. 71- N). Specific correlation between cytoplasmic and the sources of extracellular (i.e., small body) fluorescence could not be made. The relative abundance of small bodies containing DNA was not measured; attempts to quantify were thwarted by small numbers of irregularly shaped and spherical bodies together in the same field, brightness of the preparation, and variations in population densities of both the amoebae and the bacterial lawns. Quantifi- cation would require synchronously grown amoebae with fixation at the same stages of development. Although their small size precludes definitive identifi- cation of these DNA-containing bodies, they may repre- sent the fate of chromatin bodies, i.e.. the released "chro- midia" reported in the early amoeba literature (Hogue, 1914). We hypothesize that these irregularly shaped bod- ies represent highly condensed packages of parental DNA that presumably contain information required for the full development of the organism. Although we rou- tinely see irregular bodies in our cultures of P. jugosus, including those purchased from the American Type Cul- ture Collection, we have never observed in a single speci- men the entire cycle of budding off, followed by develop- ment into small and then standard-sized trophic amoe- bae. Yet such bodies, uncannily like the "buds" Hogue studded mitochondria (M) adjacent to nucleus (N) with its outer membrane also studded with ribosomes. Chromatin bodies (c) shown at higher magnification. Bar scale = 0.5 ^m. C. Amoeba with highly vacuo- lated bodies which may be a form of chromatin bodies. Mitochondria (M), nucleus (N). Bar scale = 1.0 ^m. D. Amoeba with large food vacuoles (v) containing membranous material. Both mitochondria (M) and chromatin bodies (C) are also present. Bar scale = 1 .0 ^m. 124 M. ENZIEN ET AL. Figure 1 2. TEMs of'amoebae. A,B. Arrows pointing to cytoplasmic buds. Chromatin bodies (C), vacu- oles(V)and mitochondna (m) are also visible. Bar scale = A. 2.5 ^m. B. 2.0 ^m. (1914) described in larger amoebae parasitic in oysters, are always found if sought in cultures of P. jugosus. Pro- mitotic division of amoeba nuclei has also clearly been seen in our preparations (but never in live material) (Figs. 5A-D; 6A, B) suggesting more than one form of reproduction may be responsible for the growth of this organism. The definitive solution of the problem of multiple fis- sion in this tiny amoeba requires cloning in pure culture of a single isolated 2-4 j/m diameter "bud" after the ob- servation of its prior presence in the cytoplasm of the adult amoeba as a chromatin body. Cloning studies cou- pled with DAPI or mithramycin as a vital stain may eventually resolve the question of nucleus-derived cy- toplasmic chromatin bodies which are released as propa- gules to the medium under conditions of maximal growth on dense, well-fed cultures. Extrusion of these tiny bodies from the nucleus and then from the cell would have to be directly observed to confirm this sce- nario. Fluorescent DNA stains would facilitate these ob- servations by tracking nuclear DNA. Electron microscopy Cell struct lire and reproduction. Electron microscopy of cultures from Cuban mat samples revealed at least three different forms: cysts, amoebae, and mastigotes. Cysts were roughly spherical with a single, smooth outer layer, 500 nm thick (Fig. 8A.B). Cysts contained a nu- cleus with a prominent nucleolus, numerous dense mito- chondria with tubular cristae surrounded by ribosome- studded endoplasmic reticulum, vacuoles containing various materials (e.g., degrading spore-forming bacte- ria, stacks or whorls of membranes and unidentifiable structures, most likely degraded remnants of food). They also contain small electron-lucent spheres that may con- tain storage material (e.g.. at least 18 can be seen in Fig. 8B). Other cytoplasmic structures, distinguishable from mitochondria by their lack of cristae, darker staining, amorphous shape, and lack of surrounding endoplasmic reticulum, are here termed chromatin bodies. As indi- cated by their dumbbell shape, some of these bodies ap- pear to be in division. The combination of these struc- tures is indicative of a metabolically active rather than a dormant cell. Trophic amoebae characteristic of this species are monopodial ("Umax") (Fig. 9). The posterior region of the cell possesses an uroid (Fig. 9B) similar to other iso- lates of this organism. A more-or-less centrally located nucleus with a prominent nucleolus is seen in many thin sections. The nuclear membrane is surrounded by ribo- LIFE HISTORY OF P. JUGOSVS 125 - ' „•• •_ *i£^/"« ' Figure 13. A. Binucleated mastigote with tubules, probably derived from rhizoplast transverse micro- tubules that extend subcortically (see Fig. 16). B. Mastigote with four [9(2) + 0] kinetosomes showing approach of connecting rhizoplast to the nucleus. Mitochondria (M) interpreted to be in division. A,B. Bar scale = 1.0 ^m. somes (Fig. 10B). Densely staining mitochondria with tubular cristae are often surrounded by ribosome-stud- dedendoplasmicreticulum(Fig. 10A, B, D). Large vacu- oles containing bacteria and membranous whorls are of- ten present (Figs. 9; 10D; 17). In some amoebae, these vacuoles occupy a large portion of the cytoplasm (Figs. 9A; 10D; 12A; 17). Chromatin bodies are almost always found in rounded amoeba, but rarely in amoebae that clearly appear tro- phic. Rounded amoebae may represent an early stage of encystment in which degeneration of mitochondria to chromatin bodies is caused by the onset of metabolic dormancy. We think this is unlikely because encysted and rounded pre-cyst amoeba always contain both mito- chondria and chromatin bodies. Mitochondria including dumbbell-shaped forms (Fig. 1 IB, 13) are clearly distin- guishable from chromatin bodies. Their presence in amoebae, mastigotes, and cysts suggests P.jiigosus is fun- damentally aerobic even though it apparently tolerates high sulfide and anaerobic conditions. Some chromatin bodies contain centrally located, dark-staining masses re- sembling nucleolar material (Fig. 10A); others are argu- ably in the process of division (Figs. 8B; 10A, D). Some amoebae contain dark-staining structures that differ from chromatin bodies, e.g., the vacuolated structures of Figure IOC. These vacuolated structures, because they are similar in stain density, size, and number per cell, may represent some developmental form of chromatin bodies. We suggest that these bodies represent "blebs" of nuclear origin containing DNA. Nuclear blebbing was reported by Stevens el al. (1980) in Naegleria. We also observed nuclear blebs in P. jugosus (Fig. 1 1 ), although because they are smaller than the chromatin bodies, they may be early developmental stages that then provide the means for their formation. We have not been able to de- termine at the light microscopic level if any cytoplasmic fluorescent staining comes from these bodies, nor have we done ultrastructural autoradiography for DNA to dis- tinguish these from mitochondria, yet the electron mi- crographic appearance of the interior of these bodies is that of chromatin (Fig. 10B. D). Our interpretation is en- tirely consistent with that of the chromatin bodies in oys- ter parasitic amoebae called "chromidia" (Hogue, 1914). Cytoplasmic "budding," of interest because of the presence in cultures of numerous small bodies detectable by light microscopy (Figs. 4A-C; 5 E, F; 6C, D; 7E, F, I- M), is found in many sections (Fig. 12). Corresponding small bodies are difficult to identify at the ultrastructural level because the orientation of the amoebae cannot be determined with certainty; however, bodies that may 126 M. ENZIEN ET AL Figure 14. Cortical tubules (arrow) presumably originating from rtm of Figure 16. closely associated with the anteriorly positioned nu- cleus (N) of mastigote. Bracket indicates "rostral ridge" (IT). TEM. Bar scale = 1.0 f/m. correspond to those observed with light microscopy are depicted (Figs. 8A; 9A). The "buds," at least in Figure 12, are continuous with the cell cytoplasm and probably not an artifact resulting from overlapping amoebae. The "buds" lack both mitochondria and chromatin bodies in this particular micrograph, but this may simply be a function of orientation of this thin section. Because the details of mitotic division remain elusive, the case for multiple fission in this organism is incomplete and we still have no comprehensive view of how the luxuriant, rapid growth of this organism is achieved. The "buds" represent frequently observed oddities; they may or may not represent part of the amoeba reproduction process. If chromatin bodies originate from the nucleus, which is not yet clear, then budding could represent a means of their dispersion. Although mitosis has never been observed in ultra- structural studies of Parate tramitus jugoxus amoebae or mastigotes, some events associated with mitosis are evi- dent in thin sections. We have seen dumbbell-shaped chromatin bodies (Fig. 10A, D) and a single binuclear cell: a mastigote in which karyokinesis but not cytokine- sis apparently has occurred (Fig. 13). Mastigotes share many features with amoebae and cysts, including mitochondria surrounded by rough en- doplasmic reticulum (RER), large vacuoles, strands of rough endoplasmic reticulum, nuclei with a prominent nucleolus, and nuclear membrane surrounded by ribo- somes (Figs. 13, 14). However, the three forms do differ significantly. The nucleus, anteriorly located in masti- gotes and attached by rhizoplast microtubules to the nu- cleus, is in close proximity to the kinetosomes. This differs from previous descriptions of mastigotes in this organism (Darbyshire el al, 1976) in which nuclei were centrally located. The groove seen in the anterior region of the cell adjacent to the kinetid (Fig. 14, 15) may repre- sent a common feature observed in light micrographic studies called the "rostral ridge" of P. jugosus mastigotes. Chromatin bodies and the other dark-staining, vacuo- lated structures observed in thin sections of amoebae and cysts are apparently absent in micrographs of mastigotes. Kinetid structure. Kinetid structure, reconstructed Figure 15. Microtubules of axoneme are seen here in a tangential section, the distal portion of the axoneme is probably resorbing. The cortical tubules from the kinetid are in close association with the nu- cleus. Bracket indicates "rostral ridge" (rr). TEM of mastigote. Bar scale = 1.0 LIFE HISTORY OF P. JUGOSVS 127 Table III Kinetid structure Structure or orientation Axoneme I Axoneme 2 Kinetosome 1 Kinetosome 2 Basal sphere Lateral sphere Connecting fiber Nuclear fiber Rhizoplast Rhizoplast microtubules: transverse Rhizoplast microtubules: parallel Right Left Anterior Posterior Ventral Dorsal Nucleus Abbreviation axl ax2 Rl R2 bs Is cf nf RZ rtm rpm R L A P V D n from numerous electron micrographs (Figs. 13-15), and labeled according to Table III is summarized diagram- matically in Figure 16. The standard [9(2) + 0] kineto- somes (Fig. 13B), and [9(2) + 2] axonemes (Fig. 14) are present in this dikinetid that shows mirror image sym- metry. A microtubular rhizoplast containing both paral- lel and transverse tubules arises between the kinetosomes and runs posteriorly toward the nucleus (Figs. 13A, 14, 15). The parallel kinetosomes are underlain by a basal sphere connected to the nucleus; each possesses a pro- truding lateral sphere also connected to the nucleus by rhizoplast nuclear fibers (Kl/lsnf, K2/lsnf). The masti- gote in Figure 1 5 may either be extending or retracting its undulipodium. Because a large portion of degraded axonemal material is visible with cytoplasm at the distal extremity, resorption of the axoneme is more likely. A row of single microtubules (seen in cross section in Fig. 14 and tangential section in Fig. 15) originates from be- tween the kinetosomes and runs anteriorly and cortically along and past the nucleus. The sheet of microtubules presumably originating from rhizoplast transverse mi- crotubules (rtm, Fig. 16) and supporting the anterior end of the mastigote give the cell the conical appearance that prompted Darbyshire et al. (1976) to identify the "rostral ridge." The number of undulipodia, usually two as deter- mined by light microscopy, may vary greatly. While difficult to assess their precise positioning by transmis- sion electron microscopy, four [9(2) + 0] kinetosomes can be seen in a single cell (Fig. 1 3B). Amoebae in situ in a microbial mat. Although P.jugo- sus is infrequently observed in live mat during dry condi- tions, it was prevalent throughout the period of fresh wa- Figure 16. Rinetid structure, diagrammatic representation. See Table HI for description of labels. (Drawing by Sheila Manion Artz). 128 M. ENZIEN ET AL. Figure 17. Amoeba from embedded microbial mat with large food vacuoles (v), taken directly from a microbial mat sample at North Pond, Laguna Figueroa, Baja California, Mexico. TEM. Bar scale = 1 .0 ion. ter inundation. Electron microscopy of fresh mat from North Pond (Laguna Figueroa. Baja California Norte, taken October 1987) reveals an amoeba similar in form to P. jugosits (Fig. 17). The trophic amoeba appears to have uroid and large digestive vacuoles containing bacte- ria. The only other amoeba regularly seen in these field samples are larger acanthamoebae with acanthopodia. The organism greatly resembles in size and cell structure the limax amoebae in our cultures (Fig. 1 7); it is not an acanthamoeba. Thus, we interpret it to be a free-living P. jugosus photographed under natural conditions in as- sociation with purple phototrophic and other mat bac- teria. An encysted amoeba, with at least five areas of active- looking chromatin, can be seen in embedded material from the Thiocupsa layer at North Pond, Laguna Figue- roa (Fig. 18). Bacterial digestion is evident, emphasizing the rapidity with which temporary encystment can oc- cur. Photographed by John F. Stolz during his study of phototrophic bacteria, we interpret Figure 1 8 to be a sec- ond example of P. jugosits in situ in a stratified microbial community. The electron micrograph corresponds to the Figure 18. Organism from the Thioi-apsa .tp. layer of the microbial mat from North Pond interpreted to be an ectoplasmic cyst of P. jugo- sits. Although enclosed in a cyst wall (W), the ribosome-studded cyto- plasm (R), active digestion of bacteria (B) in vacuoles (V) and the well- developed chromatin (C) indicate metabolic activity in this amoebo- mastigote. TEM courtesy of John F. Stolz. Bar scale = 2.0 ^m. PJBC Cuba ATCC 30703 II 1 "& 0 ^ TJ n rr •-• M 11 1 1 Figure 19. Diagram of isoenzyme analysis of three different isoen- zymes. PjBC: the original bajacaliforniensisof Read el al. (1983) Cuba: the Cuban strain reported here. ATCC 30703: The standard soil isolate of Darbyshire cl ul ( 1976). Black represents a strong band on starch gel electrophoresis, cross-hatching a less conspicuous band, and white represents a very faint band. LIFE HISTORY OF P. JUGOSUS 129 "ectoplasmic cysts" reported in light micrographs in Read etal.( 1983; Figs. 7B-H, 13, and 14H). Isoenzyme analysis The results of the stock gel isoenzyme analysis run on amoebae and cysts of the Cuban P. jugosus strain com- pared to the ATCC-type strain (ATCC #30703) and P. jugosus bajacalifomiensis are shown in Figure 19. The pattern is unique for any amoebae in the ATCC collec- tion (T. Nerad, ATCC, pers. comm.). The propionyl es- terase (PE) and acid phosphatase (AP) each display bands resembling those enzyme patterns present in the Baja California and ATCC P. jugosus strains. The three narrow bands of the pattern of leucine aminopeptidase (LAP) of the Cuban strain seem to be unique. The most abundant protist in the purple phototrophic bacterial layer of microbial mat sediments, Paratetrami- tus jugosus displays a complex life history that correlates with rapid environmental changes in water abundance and salinity. Our evidence for this life history is consis- tent with Hogue's (1914) assertion that some amoebae reproduce by multiple fission and disperse by "chromi- dial" propagules. Acknowledgments We are grateful to Rene Fester, Gail Fleischaker, Wendy Lazar, Dorion Sagan, and Rae Wallhausser for assistance in manuscript preparation, including cartog- raphy. We are especially grateful to Annette Coleman, Floyd Craft, Tom Nerad, and Jane Gifford for their tech- nical assistance with fluorescence staining, electron mi- croscopy, isoenzyme analysis, and field photography, re- spectively. We are immensely grateful to the members of Earthwatch team 1 & 2 (May, June 1983) for aid in the surveying and mapping of Pentapus Salina and the col- lection of field data. We thank Sheila Manion Artz for drawing Figure 16 and John F. Stolz for use of his elec- tron micrograph (Fig. 18). We acknowledge helpful sug- gestions for observing live amoebae from Prof. F. Schus- ter. Support for this research came from NASA Life Sciences Offices (NGR-004-042 to L.M. at Boston Uni- versity), the Boston University MacDonald award and The Evolution Fund (to L.M.), the NASA Planetary Bi- ology Internship Program (administered through the Marine Biological Laboratory, Woods Hole, Massachu- setts), the University of Massachusetts Amherst Re- search Trust Funds, and the Richard Lounsbery Foun- dation. Literature Cited Brown, S., L. Margulis, S. Ibarra, and D. Siqueiros. 1985. Desiccation resistance and contamination as mechanisms of Gaia. Biosystems 17: 337-360. Coleman, A. W., M. J. Maguire, and J. R. Coleman. 1981. Mithramycin- and 4'-6-diamidino-2-phenylindole (DAPI)-DNA staining for fluorescence microspectrophotometric measurement of DNA in nuclei, plastids and virus particles. / Histochem. Cvto- cliem. 29: 959-968. Daggett, P. M., and T. A. Nerad. 1983. The biomedical identification of Vahlkampfid amoebae. J. Proiozool. 30: 126-128. Darbyshire, J. F., F. C. Page, and L. P. Goodfellow. 1976. Paratetramitus jugosus, an amoeho-flagellate of soils and fresh wa- ter, type-species of Paratetramitus nov. gen. Protistologica 12: 375- 387. Gong-Collins, E. 1986. A euryhalic, manganese- and iron-oxidizing Bacillus inegateriwn from a microbial mat at Laguna Figueroa. Baja California Mexico. Microbial 48: 109-126. Hogue, M. J. 1914. Studies of the life history of an amoeba of the Limax group. Arch. Protistenkd. 35: 154-163. Margulis, L. 1980. Undulipodia, flagella. and cilia. Biosvstems 12: 105-108. Margulis, L. 1988. Systematics: the view from the origin and early evolution of life. Secession of the Protoctista from the animal and plant kingdoms. Pp. 430-443 in Prospects in Systematics, D. L. Hawksworth. ed. Clarendon Press, Oxford. Margulis, L., E. S. Barghoorn, D. Ashendorf, S. Banerjee, D. Chase, S. Francis, S. Giovannoni, and J. Stolz. 1980. The microbial com- munity in the layered sediments at Laguna Figueroa, Baja Califor- nia, Mexico: does it have PreCambrian analogues? Precamb. Res. 11:93-123. Margulis, L., S. VV. Brown, G. Hinkle, H. McKhann, and B. Moynihan. 1988. Mychonastes desk-aims BROWN sp. nova (Chlorococ- cales, Chlorophyte) — an intertidal alga forming achlorophyllous dessication-resistant cysts. Arch. Hydrobiol. Suppl. 78: 425-446. Margulis, L., J. O. Corliss, M. Melknonian, and D. J. Chapman. 1990. Handbook ofProtoctista:the structure, cultivation, habitats and life cycles of the eurkaryotic microorganisms and their descen- dants exclusive of animals, plants and fungi. Jones and Bartlett Pub- lishers, Inc. Boston, MA. Margulis, L., and D. Sagan. 1 985. Order amidst animalcules: the Pro- toctista kingdom and its undulipodiated cells. Biosvstems 18: 141- 147. Nerad, T., and P. M. Daggett. 1979. Starch electrophoresis; an effec- tive method for separation of pathogenic and non-pathogenic Naegleria strains. / Pwtozool. 26: 6 1 3-6 1 5. Page, F. C. 1967. Taxonomic criteria for limax amoebae, with de- scriptions of 3 new species of 3 Hartmannella and of 3 I 'ahlkampfia. J. Protocol. 14:499-521. Page, F. C. 1983. Marine Gymnoamoebae. Institute of Terrestrial Ecology, Culture Centre of Algae and Protozoa, Cambridge, En- gland. 54 pp. Raikov, I. B. 1982. The Protozoan Nucleus: morphology and evolu- tion. Springer- Verlag, New York. Pp. 73, 1 12. Read, L. K., L. Margulis, J. Stolz, R. Obar, and T. K. Sawyer. 1983. A new strain of Paratetramitus jugosus from Laguna Figue- roa, Baja California, Mexico. Biol. Bull 165: 241-264. Stevens, A. R., J. De Jonckheere, and E. VVillaert. 1980. Naegleria lovaniensis new species: isolation and identification of six therrno- philic strains of a new species found in association with Naegleria fowleri. Inl J. Parasitol. 10: 51-64. Stolz, J. F. 1983. Fine structure of the stratified microbial commu- nity at Laguna Figueroa. Baja California Mexico. I. Methods of in sun study of the laminated sediments. Precamb. Res 20: 479-492. Reference: Bitil. Bull. 177: 130-140. (August, 1989) Bacterial Aggregates Within the Epidermis of the Sea Anemone Aiptasia pallida EDWARD E. PALINCSAR, WARREN R. JONES, JOAN S. PALINCSAR. MARY ANN GLOGOWSKI, AND JOSEPH L. MASTRO Department of Biology, Loyola University, Chicago, Illinois 60626 Abstract. Bacteria in cyst-like aggregates have been ob- served in the sea anemone Aiptasia pallida. Algal symbi- onts, common in certain Cnidaria, including Aiptasia, are located in the gastrodermis, while the bacteria de- scribed in the present study were found exclusively in the epidermis. They were gram-negative and packed closely together within what appeared to be a single cell. The bacterial sac varied in size according to the number of bacteria it contained. Ultrastructural features of the bac- teria included numbers of large vacuoles or inclusions often intertwined with web-like nucleoids in the central region. Aggregates /'// situ showed a whorled arrangement of the bacteria and maintained this pattern and their structural organization after extrusion from anemone epidermis. A fatty acid profile suggested that the bacteria may belong to the genus I Ibrio. Introduction Single bacterial cells have often been observed as endo- symbionts in cells of many types of organisms. Vesicles containing small numbers of bacteria and Chlorella were found in gastrodermal cells of Hydra (Margulis et ai, 1978; Thorington et al.. 1979). Wilkinson (1978) cul- tured and characterized bacteria associated with marine sponges, most of which inhabited the connective tissue- like mesohyl and thus were not usually intimately associ- ated with the sponge cells. Some individual bacteria, as well as cyanobacteria, were located inside sponge cells. In some animals, endosymbiotic bacteria inhabit spe- cific cell-like structures called bacteriocytes, such as cells noted by Vacelet (1975) in the marine sponge, Verongia. Palincsar et al. (1988) found many bacteria inside aggre- Received 12 October 1988: accepted 31 May 1989. gates in the epidermis of Aiptasia pallida. Peters et al. (1983) and Peters (1984) identified bacteria in "baso- philic ovoid bodies" in septa and calicoblast tissue of Acropora palmata and A. cen-icornis, and in mesoglea of Porites astreoides in the Caribbean. Gills of Spisula siibtnmcata have bacteriocytes containing a "bacterio- phore" with over 200 bacteria (Soyer et al.. 1987). Be- cause we found no nuclei directly associated with bacte- rial aggregates in A. pallida. and the origin of the mem- brane surrounding the bacteria is uncertain at present, we will use the term bacterial aggregate for large numbers of bacteria packed together, bounded by a discrete mem- brane, which are located either intracellularly or extra- cellularly. The bulk of trophosome tissue in vestimentarian tube worms, Riftia pachyptila. of the Galapagos Rift was ob- served by Jones ( 1 98 1 ) to be mostly bacteria; on inspec- tion under the electron microscope, Cavanaugh (1983) found that these bacteria were located in bacteriocytes. The tissues of other smaller pogonophorans not from hy- drothermal vent communities also contained bacteria, mostly within host cells (Southward et al., 1981). The gutless marine oligochaete P/iallodrillus (Felbeck et al., 1983) contained bacteria within its tissues. In addition, Cavanaugh (1983) described bacteriocytes in gills of the hydrothermal vent clams, Calyptogena magnified, C. pa- cifica, two non-vent clams, Solemya velum and Litci- noina annulata. and an additional pogonophoran. Fischer and Hand ( 1 984) studied bacteriocytes in Lucina Jloridana, a clam living in an aerobic high-sulfide envi- ronment. Dando et al. (1985) found several species of lucinid clams inhabiting low sulfide environments. The clams contained bacterial symbionts that accumulated sulfur as an energy source. Bouvy et al. (1986) and Soyer et al. (1987) described packets of bacteria which they 130 BACTERIA IN A1PTASIA 131 termed bacteriophores, in Spisula subtruncata, a surf clam living in a highly aerobic environment. These bac- teria were of two types, were spatially separated in the gill, and used reduced nitrogen and sulfur compounds as energy sources. Although it is well known that the gastrodermis of A. pallida harbors the symbiotic dinoflagellate Symhiodin- ium microadriaticum, here we describe bacteria in aggre- gates at both light and electron microscopic levels, in a different location in the epidermis of the anemone Aip- tasia pallida. In addition, we have begun chemical char- acterization of the bacteria with fatty acid analysis. We have also studied control of the bacterial population in situ to learn whether interrelationships exist between population levels of the two different symbionts, algal and bacterial. Materials and Methods Individuals of Aiptasia pallida were obtained from Carolina Biological Supply Company (Burlington, North Carolina 272 1 5). The anemones were kept in 38-1 aquaria in Instant Ocean (Aquarium Systems, Mentor, Ohio 44060) made with tap water at a specific gravity of 1.020 at 25°C. The anemones were cultured under fluo- rescent light with 16 h of light and 8 h of darkness, using a 35 W GE Lite-white and a 40 W Sylvania Grolux fluo- rescent tube at an intensity of 24 ^Em": s~', measured with a LI Photometer Model LI 1 85A with an underwater sensor (Lambda Instrument Corporation). The anemo- nes were fed to repletion with Anemia nauplii (San Fran- cisco Bay Brand, Newark, California 94560) twice a week. The H2S level in the culture tank was 0.015 mg/1 (Hach kit #2238); the sulfate level was 200 mg/1 (Hach kit #225 1 ); the nitrate level was 7.95 mg/1 and the nitrite level was 0.048 mg/1, measured with Model Nil 4 nitrate and nitrite kit (Hach Company, Loveland, Colorado 80539). A. pallida were maintained for two years in cul- ture. The tentacles contained clear ovoid structures as seen under the dissecting microscope. (In the ensuing 2 years, we have seen over 200 anemones with these struc- tures.) Higher magnification revealed bacteria inside the ovoid structures. New orders of Aiptasia from Carolina, and from Ward's Natural Science Establishment, exam- ined immediately upon arrival, also contained aggregates in the tentacular epidermis. Samples of Aiptasia pallida from the Shedd Aquarium, Chicago, and from Bermuda were examined; no aggregates were found (S. Kenney and M. Lesser, pers. comm.). A. pallida individuals col- lected immediately before shipment from their natural habitat at Morehead City, North Carolina by Lawrence Wallace (Carolina Biological Supply Company) had the largest numbers of bacteria seen to date, averaging 75 ag- gregates per sample (sampled according to methods given for antibiotic sensitivity determinations). For electron microscope studies, 10 anemones con- taining aggregates were anesthetized in 3% MgSO4 for 1 h. Tentacle tips 2-4 mm long were clipped and im- mersed for 1 h at 4°C in a modified Karnovsky's fixative (2.5% glutaraldehyde and 2% paraformaldehyde, Kar- novsky, 1967) made up in Instant Ocean, postfixed in 2% osmium tetroxide at room temperature for 30 min. dehydrated in a graded series of acetone, and embedded in Epon 812. Sections were cut on an LKB Nova Ultra- tome, stained with uranyl acetate (Hayat, 1969) and lead citrate (Reynolds, 1963), and viewed in a JEOL model 1200EX transmission electron microscope at 80 kV. One /urn sections from the same preparations were cut and stained in toluidine blue for light microscopic evalu- ation. Estimates of numbers of bacteria inside a small, typical (20 urn diameter) aggregate were made by comparing the volume of a aggregate with that of a bacterium, according to methods described by Boatman (1986). Bacterial vac- uole or inclusion numbers, diameters, and volumes were measured with the Bioquant Image Analysis System (R & M Biometrics, Inc., Nashville, Tennessee 37209) with Hipad electronic digitizer. All vacuoles in the sample were counted regardless of level of section through the sphere of the vacuole. Tentacles were removed and placed in two washes of Instant Ocean to obtain isolated aggregates for observa- tion, analysis, and culture. This procedure stimulated the release of intact aggregates into the medium. The isolated aggregates were also washed twice in Instant Ocean to minimize collection of any surface microflora. Both in- tact aggregates and individual bacteria released from bro- ken aggregates were stained with Gram's stain for type of wall structure and with Sudan Black for lipids (Huma- son, 1972). A sample of isolated aggregates was lyophilized and sent to Microbial ID, Inc. (Newark, Delaware 1971 1) for fatty acid analysis by gas chromatography. The cellular fatty acid profile, showing types and relative amounts of fatty acids by chain length, saturation, and side groups, was compared to a Hewlett Packard computer data bank of fatty acid profiles of bacterial genera and species. Iden- tification by fatty acid analysis is comparable to that by rapid biochemical strips, is effective for far more genera, and is often more specific than biochemical tests. Match- ing of similarities between the sample and known bacte- rial species provided a reliable index to the identification of bacteria (Moss and Lewis, 1967; Miller, 1984; Sasser etal, 1984). The effects of antibiotics on the control of the numbers of bacterial aggregates of Aiptasia were determined. 132 E. E. PALINCSAR ».$ "•*'-: • •':• " Figure 1. A. Phase contrast view of living tentacle containing numerous balloon-like aggregates in Aiptasia pullida epidermis. Note extruded aggregate (arrow) outside the tentacle. Bar = 100 ^m. B. Phase contrast view of two extruded aggregates. Bar = 25 ^m. C. Light microscope view of toluidine blue stained section of epidermis of anemone tentacle showing aggregates among epidermal cells. Aggregate (B), epider- mis (E), gastrodermis containing symbiotic dinoflagellates (Symbiodinium microadriaticum) (G), meso- glea(M). nematocyst (N). Bar = 100 ^m. M 2b Figure 2. A. Transmission electron microscope section of entire aggregate containing numerous bacte- ria. Note thin membrane surrounding aggregate and associated anemone membranes (facing arrows). Sev- eral anemone cells abut the bacterial enclosure. Anemone epidermal nucleus (A), aggregate (B), nemato- cyst (N). Bar = 5.0 /jm. B. Portion of aggregate-anemone interface. Note small amount of anemone cyto- plasm with mitochondria between membrane of aggregate vacuole and plasma membrane. Bacteria (B), mitochondrion (M); Arrows span width of anemone cytoplasm. Bar = 0.5 ^m. 134 E. E. PALINCSAR Figure 3. A. Portion of small aggregate. There is a relatively large amount of anemone cytoplasm in this specimen. Endoplasmic reticulum (E). Bar = 0.5 pm. B. High power view of membranes of isolated BACTERIA IN AIPTASIA 135 Three aquaria, each containing 10 anemones with their aggregates, were set up. One contained 1 25 mg/ml strep- tomycin sulfate in Instant Ocean, a second contained 25 mg/ml chloramphenicol (d'Agostino, 1975), and a third contained Instant Ocean alone, as a control. Conditions were the same as for the stock cultures. Aggregate popu- lations were sampled every 5 days over a 3-week period by clipping a tentacle from each anemone and counting bacteriocytes visible in the epidermis on one side of the tentacle from the tip toward the base for 0.6 mm. To determine whether decreased populations ofSym- biodiniwn would affect the population of bacteria, anemones were placed in 3 1 of aerated Instant Ocean at 17°C that contained 10^3 M 3-(3,4-dichlorophenyl)-l,l- dimethylurea (DCMU) (Sigma Chemical Company) in light of 580 MEirr: s"1 intensity from 2 GE 120 W re- flector floodlights, for 3 days. This procedure was adapted from the technique of Pardy ( 1 976) which, when applied over a longer period of time, or at a higher light intensity, resulted in nearly total destruction of algal symbionts. Pardy used this procedure to obtain aposym- biotic cnidarians for controls in symbiosis experiments. Aggregates were counted using the same procedure as that used for the antibiotic sensitivity determinations. Cell numbers of S. microadriaticum were sampled by counting the cells found in a 0.2-mm transect. The distri- bution of S. microadriaticum was homogeneous in the tentacle preparations. Results In intact tentacles viewed under phase contrast mi- croscopy, aggregates appeared as lucent spherical bodies within the epidermis ofAiptasia pallida (Fig. la). Their position varied from near the mesoglea to the free surface of the epidermis. While the aggregates were most com- mon at the tips of anemone tentacles, they were also ob- served proximally on the tentacles, on the oral disc, and at low density throughout the column of the anemones. The size of aggregates shown in Figures 1-3 varied from about 10 to 100 urn. The number of aggregates varied greatly from one anemone to another. Bacterial aggre- gates (counted according to the method for antibiotic sampling) from each of a sample of 44 different anemo- nes showed that 6(14%) anemones had no aggregates in the area of the sample, and 10 (23%) anemones had 25 or more aggregates in the same sample area. The mean number of aggregates for all 44 anemones was 14.86 per 0.6 mm sample. Under higher magnification, the aggre- gates had a characteristic "finger print" appearance due to the tendency of adjacent rod-shaped bacteria to orient parallel to each other, producing a whorled pattern (Fig. Ib). The organization of the aggregate, and the character- istic whorled arrangement of the bacteria, was main- tained after isolation of the aggregate from the epidermis. Toluidine blue-stained aggregates seen in tentacular epidermis of A. pallida under light magnification ap- peared as dark oval spots in the epidermal layer (Fig. Ic). Cnidocytes, ciliated cells, and other cells normally pres- ent in the epidermis surrounded the aggregates. An ag- gregate 20 ^m in diameter contained up to about 1200 individual bacteria. Examination of the ultrastructure of an aggregate in its place in anemone epidermis shows the bacteria packed closely together (Fig. 2a). The tendency of the bacteria to remain parallel to one another within the aggregate re- sults in many adjacent longitudinal or many adjacent cross sections. Even this relatively small aggregate is many times larger than the epidermal cells of the anem- one peripheral to it. Membranes of the anemone cells intersect the thin dark aggregate membrane and either merge with it or run parallel to it for some distance. The interface between the bacteria-containing structure and the anemone cell may contain cytoplasm and cell organ- elles (Fig. 2b). Very small aggregates may be surrounded by a wide band of cytoplasm (Fig. 3a), containing the usual complement of cell organelles, including a layer of endoplasmic reticulum parallel to the aggregate. Follow- ing extrusion from the anemone epidermis, isolated ag- gregates are surrounded by a double membrane strong enough to withstand the stress of up to 6 transfers to fresh Instant Ocean (Fig. 3b). Occasionally, when aggregates ruptured, the bacteria exhibited pronounced motility on contact with seawater. The size of the aggregate was close to the size of the intergrid space of the section support medium so that the ultrastructure of entire full-sized aggregates could not be observed. The size of the aggregate (observed range about 10-100 jurn) was determined by the number of bacteria it contained, since the bacteria were packed closely to- gether in every aggregate observed. The individual bacteria of the aggregates were gram- negative rods, averaging 1 .0 ^m X 4.0 nm in size. Viewed aggregate. Disruption of anemone cytoplasm was not due to fixation, since the same procedures were used on all tissues, but probably due to mechanical disruption during handling of individual aggregates. Bacteria (B), anemone cytoplasm (A), mitochondrion (M), membranes of aggregate (P) and (V). Arrowheads indi- cate membrane of aggregate vacuole. Bar = 0.5 jum. 136 E. E. PALINCSAR under the electron microscope, the cytoplasm contained numerous electron-lucent vacuoles or granules that were not membrane bound (Fig. 4a, b). Calculations from measurements of 370 vacuoles showed that they occu- pied up to 6.66 ± 0.74% of the volume of the cell, ap- peared randomly distributed and had a mean diameter of 162 ± 6.98 nm. Using the light microscope, we ob- served the bacteria accumulating the lipid stain, Sudan Black. Web-like nucleoid material was centrally located. The bacterial wall had the layered structure typical of Gram-negative bacteria, and stained negatively in Gram's stain. The outer wall of the bacterial cell ap- peared loose and undulating or ridged, leaving regular spaces between the wall and cell membrane (Fig. 4a, b). This appearance may be a fixation artifact in gram nega- tive bacteria, according to studies off. coli. as described by Dubochet et a/. (1983). Bacteria were regularly ob- served to be dividing by fission (Fig. 4c). The bacteria were observed only within aggregates in A. pallida, and we did not observe single bacterial cells in, or between, anemone cells. Some aggregates observed in electron mi- croscope sections also contained one or two less electron- dense bacteria that were about twice as large as the other bacteria. Southward (1986) also observed two different types of bacteria in the same aggregate in deep sea thya- sirids. Two types of bacteria were also seen by Soyer et al. (1987), although not within the same bacteriophore. Aggregate fatty acid analysis showed 1 1 fatty acids with chains from 9 to 18 carbons in length. Comparison of the aggregate fatty acid profile with the Hewlett Pack- ard bacterial classification data bank indicated most sim- ilarities to the genus l'ibrio(pers. comm., Microbial ID). Comparisons of aggregate fatty acids with \'ihiio mimi- cus (Microbial ID) and I '. angiiillarum (Boe and Gjerde, 1980) were made by setting the 16-carbon straight chain fatty acid in all three samples to 100%. Amounts of fatty acids having different chain lengths were calculated as percentages of this base value (Table I). During treatment with chloramphenicol, the aggregate population decreased from 17.5 aggregates per 0.6 mm sample of one side of a tentacle tip from each of 1 0 anem- ones, to 5.0 in 5 days, a decrease of 75% as compared to the control, reaching a minimum of 1 .4 after 10 days. In streptomycin, the number of aggregates decreased by 50% after 10 days from 14.9 to 7.8, and reached a mini- mum of 6.0 after 20 days. After 20 days, the number of aggregates was 4 times lower in chloramphenicol than in streptomycin. On day 1, the means of the three groups showed no significant differences (using one-way analy- sis of variance, the F statistic is 0.20 and the /'-value is 0.82). Also, this and subsequent observations on the con- trol group made on days 1.5. 10, 16. and 20, showed no significant changes in the mean of the control group over this time period (using one-way analysis of variance, the F statistic is 0. 1 5 and the P- value is 0.96 ). However, there was a sharp decline in both the mean and the variance of the number of bacteriocytes in the group treated with chloramphenicol when compared to the control group. A two-sample t test for independent samples with un- equal variances, using the Satterwaite approximation for the degrees of freedom in computing the P-value, indi- cated a significant difference in the mean number of ag- gregates in the anemones undergoing chloramphenicol treatment when compared to the control group of the fifth day (t = 4.54, d.f. = 17, P- value < 0.0001 ). Separate t tests comparing these groups on days 10, 16, and 20 also indicated significant differences (on the 20th day, t = 4.28, d.f. = 9, f-value = 0.002). The same method comparison showed no significant difference in the means of the streptomycin group and the control group until the 20th day. (On the 20th day t = 2.18, d.f. = 17, F-value = 0.043). Nonparametric methods, using Mann- Whitney statistics in place of the t statistics, gave similar results. For example, the difference in the control and the streptomycin group after 20 days remained statistically significant (F-value = 0.017). The anemones undergoing chloramphenicol treatment began to show deleterious effects during the third week, indicated by tentacular re- traction. Following treatment with light and DCMU, much of the Symbiodinium population was lost and the number of aggregates increased, from a mean of 8.8 ± 2.61 per 0.6 mm tentacle sample before treatment, to 1 1 .3 ± 4.86 after three days of treatment, to 27.4 ±3.15 three days following return to normal culture conditions. Mean numbers of Symbiodinium decreased during treatment from 26.6 ± 1.38 to 1 1.6 ± 1.73, and then to a further reduction of 5.2 ± 1.70 one week later. Seven days fol- lowing the return to normal culture conditions, the num- ber of aggregates was still elevated (15.1 ± 3.33 per tenta- cle). The standard errors are relatively high because of the previously mentioned great variation in numbers of aggregates from one anemone to another. Even so, t-test values demonstrated that the mean number of aggregates per tentacle at the beginning of the experiment was sig- nificantly different at the 0.01 confidence level from the mean number at day 6 and day 10. Discussion Bacteria in the aggregates from Aiptasia pallida speci- mens were not individually surrounded by host vacuoles and host cytoplasm as observed by Cavanaugh (1983) in So/cniya velum. Riftia pachyptila and Calyplogena mag- nifies or Fischer and Hand (1984) in Lucina floriclana but instead the bacteria were clumped together in one BACTERIA IN A1PTASIA 137 * -•••>,• <• - 4a • „ r ^ . 4c V N I •• Figure 4. A. Longitudinal section ot'hacterium inside aggregate. B. ctions of bacteria in aggre gate. C. Longitudinal section of bacteria undergoing fission (arrowheads) inside aggregate. Nucleosome (N). vacuole ( V). Bar = 0.5 138 E. E. PALINCSAR Table I Percentage distribution of fatly acids of aggregates found in Aiptasia Fatty acids as a percent of 1 6-C fatty acid Aggregate I 'ihrio mimicus1 ( ' anguillarunf Fatty acid chain length 9:0 3 5.30 0 0 12:0 8.00 0 16.50 12:0 3 OH 4.81 13.80 0 14:0 23.84 20.76 32.10 15:0anteiso 11.29 0 1.90 16:0iso 14.62 5.90 28.90 16:1 cis9 117.32 128.00 251.00 16:0 100.00 100.00 100.00 18:1 cis9 33.02 0 62.50 18:0 19.09 6.92 9.10 1 Data from Microbial ID Hewlett Packard data base. : Mean of 10 subcultures from Boe and Gjerde ( 1 980). ' The first number indicates the carbon chain length; the number following the colon gives the number of double bonds; any notations to the right give modifications in structure from a straight chain fatty acid. large vacuole. The appearance of the aggregate was more similar to those described by Peters (1983), Bouvy ct al. (1986), and Soyer el al (1987). The presence of numer- ous dividing bacteria (Fig. 4a) inside the aggregate indi- cated that they were metabolically active and that the ag- gregate was not a static cyst. The data suggest that there are two possible theories of the structural origin of aggregates in A. pallida. Our observations of small aggregates (Fig. 3a) suggest that the aggregate is an anemone cell that is distorted by a vacuole swollen by the multiplication of invading bacteria. Mi- crographs of isolated aggregates also showed that there were organized structures between two membranes sur- rounding the aggregate. Figure 3a shows endoplasmic re- ticulum next to the bacterial compartment. Chesnick and Cox (1986) suggest that the membranes of the bacte- riophore in the dinoflagellate Peridininm baltica may be derived from endoplasmic reticulum. These observa- tions indicate that the aggregate enclosure is not of bacte- rial origin, but an anemone cell with the cytoplasm stretched into a thin shell surrounding the vacuole con- taining the bacteria. On the other hand, pictures of larger aggregates (Fig. 2a) indicate that the bacteria are enclosed in a membrane separate from the anemone structure. Figure 2a shows several anemone cells around the pe- riphery of the bacterial enclosure, none of which enclose the aggregate. The great size difference between anemone cells and aggregates makes it seem unlikely that an anem- one cell could enlarge to the necessary degree. Also, no nuclei were observed in aggregates. No series of parallel intracytoplasmic membranes such as those characteristic of photosynthetic bacteria were seen in the bacteria of A. pallida. In this respect, bacteria from A. pallida were similar to bacteria from marine sponges (Vacelet, 1975; Wilkinson, 1978), R. pa- chvptila, other vestimentiferans, the clams Solemya ve- lum, C. magnified, C. pacifica, Lucinoma anmilala (Cav- anaugh, 1983), and L. floridana (Fischer and Hand, 1984), none of which possessed internal lamellae. In con- trast, Cavanaugh el al. (1981) described intracytoplasmic membranes in bacteria in formalin-fixed trophosome of R. pachyptila. In Hydra, individual bacterial symbionts are located in the gastrodermis in contrast to the epider- mal location in Aiptasia (Thoringlon et al., 1979). More- over, the symbiotic vesicles in Hydra contained very small numbers of both bacteria and Chlorella. Studies by Wilkerson (1980) conclude that the bacteria contribute to phosphate uptake in this organism. The bacteria in A. pallida differ from those in other marine animals in confinement to the epidermis and in containing numer- ous bacteria within a single aggregate except as noted by Peters ( 1983) in corals and Bouvy et al. (1986) in clams. The vacuoles or granules located in the bacterial cyto- plasm were a constant feature of the bacterium. Similar inclusions have been found in many other bacteria, and their contents have even been analyzed. Jones (1981) and Hand ( 1 987) identified intracellular vacuoles similar to those seen in the bacteria found in A. pallida, as loca- tions of sulfur deposits in R pachyptila. A common in- tracellular inclusion, which is similar in appearance to vacuoles found in bacteriocytes of A. pallida, is associ- ated with accumulation of poly-b-hydroxybutyrate. Lack of limiting membranes around the vacuoles does indicate that they might contain lipids. Sudan Black lipid staining under the light microscope also suggests this possibility. Lack of granular material in the vacuoles af- ter electron microscope preparation procedures used in- dicated that they did not contain glycogen. The computer matching process, based on fatty acid analysis, suggested that the bacterium found in A. pallida was a member of the genus Mbrio. This identification is supported by our structural findings, although the partic- ular bacteria found in A. pallida are large for this genus. The bacteria in the aggregates were consistently gram- negative, and not variable in their Gram staining charac- teristics. Therefore, even though the bacteria were in var- ious growth phases inside the aggregate, the consistency of staining shows that the cells are Gram-negative. This conclusion is further confirmed by the typical layered ap- pearance of the cell wall (Fig. 4), not found in Gram- positive bacteria. Moreover, members of the genus Vi- BACTERIA IN A1PTASIA 139 brio are frequently found in close association with ma- rine animals, are Gram-negative, and can be straight rods. Most I'ibrio require Na+ ions for growth, often grow at 20°-30°C, and tolerate moderately alkaline con- ditions. These requirements are consistent with the ma- rine environment of the aggregates. Some species store poly-b-hydroxybutyrate in intracellular granules. Many species form flagella, but others do not. They are sensi- tive to chloramphenicol and streptomycin, as were the bacteria in the aggregates. I 'ibrio is also sensitive to a rel- atively wide range of other antibiotics (Baumann et al, 1984). Because Aiptasia benefit from the presence ofSymbio- diniwn, the possibility that changes in the bacterial pop- ulation might affect the Symbiodinium population and consequently affect the well-being of Aiptasia, suggests an additional factor to be considered in studying how changes in one population may affect the other. It is un- known whether the increase of aggregates in anemones subjected to light-DCMU treatment was due to the use of substances from damaged Symbiodinium, or possibly to the increased availability of some substance not used by the decreased population of Symbiodinium. Decrease in bacterial numbers after 1 0 days suggests that the bacte- ria may have been using nutrients obtained from S1. mi- croadriaticum cells as they disintegrated. We have not observed detrimental effects of bacterial colonies on A. pallida. Individuals may develop shortened, thickened tentacles, but this condition has not been associated with certainty with increased numbers of aggregates, like Pe- ters ( 1 984) has associated pathological conditions in cor- als with the presence of bacteria. Work is in progress to isolate and culture the bacteria, characterize them physiologically and biochemically, and determine their metabolic relationships in A. pal- lida. We now have a three-way system to study. The in- terrelationships between Aiptasia and Symbiodinium are already well established. The combination of bacte- rial, dinoflagellate, and anemone cells living together suggests curious biochemical and ecological interrela- tionships, knowledge of which may prove useful as a model system in cellular ecology. Acknowledgments The authors express grateful appreciation to the fol- lowing: Drs. Myron Sasser and Margaret Roy of Micro- bial ID for helpful advice on sample preparation for the fatty acid analysis; Dr. Gerald Funk of the Department of Mathematics, Loyola University of Chicago, for statis- tical treatment of results of the antibiotics experiment; Lawrence Wallace of Carolina Biological Supply Com- pany for freshly collected Aiptasia pallida from the sea- shore; Susan Kenney of the Shedd Aquarium in Chicago for samples of Aiptasia pallida: and Michael Lesser of the University of Maine at Orono for examination of Aiptasia from Bermuda. Literature Cited d'Agostino, A. 1975. Antibiotics in cultures of invertebrates. Pp. 109- 133 in Culture of Marine Invertebrate Animals. W. L. Smith, ed. Plenum Press. New York. Baumann. P., A. L. Furniss, and J. V. Lee. 1984. Genus I. Vibrio. Pp. 518-537 in Sergey's Manual of Systematic Bacteriology, Vol. I, J. G. Holt, ed. Williams & Wilkins, Baltimore. Boatman, E. S. 1986. Digitizing and Quantitation. Chap. 1 3 in Ultra- structural Techniques for Microorganisms. H. C. Aldrich and W. J. Todd, eds. Plenum Press, New York. Boe, B., and J. Gjerde. 1980. Fatty acid patterns in the classification of some representatives of the families Enterobacteriaceae and Vi- brionaceae. J. Gen. Microbiol. 1 16: 41-49. Bouvy, M., G. Cahet, F. deBilly, J. Soyer, M.-O. Soyer-Gobillard, and C. Thiriot-Quievreux. 1986. Sur la presence de bacteries dans las branchie d'un mollusque bivelve littoral Spisula sublruncala. C R. Aead. Sci. Paris 303: 257-262. Cavanaugh. C. M., S. L. Gardiner, M. L. Jones, H. \V. Jannasch, and J.B.Waterbury. 1981. Prokaryotic cells in the hydrothermal vent tube worm Riftia pachyptila Jones: possible chemoautotrophic symbionts. Science. 213: 340-342. Cavanaugh, C. M. 1983. Symbiotic chemoautotrophic bacteria in marine invertebrates from sulphide rich habitats. Nature 302(3): 58-61. Chesnick, J. M. and E. R. Cox. 1986. Specialization of endoplasmic reticulum architecture in response to a bacterial symbiosis in Peri- dinium balticum. J Phycol. 22: 291-298. Dando, P. R., A. J. Southward, E. C. Southward, N. B. Terwiliger, and R. C. Terwiliger. 1985. Sulphur-oxidising bacteria and haemoglo- bin in gills of the bivalve mollusc Mvrlea spinifera. Mar. Ecol. Prog. Ser 23: 85-98. Dubochet, J., A. W. McDowell, B. Menge, E. N. Schmid, and K. G. Lickfeld, 1983. Electron microscopy of frozen-hydrated bacteria. J Bactenol. 155:381-390. Felbeck, H., G. Liebezeit, R. Dawson, and O. Giere. 1983. CO: fixa- tion in tissues of marine oligochaetes (Phallodrillus leitkodermatus and P. planus) containing symbiotic chemoautotrophic bacteria. Mar. Bioi 75: 187-191. Fischer, M. R. and S. C. Hand. 1984. Chemoautotrophic symbionts in the bivalve Lucina floridana from seagrass beds. Biol Bull 167: 445-459. Hand, S. C. 1987. Trophosome ultrastructure and the characteriza- tion of isolated bacteriocytes from invertebrate-sulfur bacteria sym- bioses. Biol. Bull. 173: 260-276. Hayat, M. A. 1969. Uranyl acetate as a stain and a fixative for heart tissue. Proc 27th Ann. Meeting Electron Microsc. Soc. Am. Claitors Publ., Baton Rouge. P. 412. Humason, G. L. 1972. Animal Tissue Techniques. Freeman, San Francisco. Jones, M. L. 1981. Rifiiapachyptilaiov.es: observations on the vesti- mentiferan worm from the Galapagos Rift. Science. 213: 333-336. Karnovsky, M. 1967. The ultrastructural basis of capillary permeabil- ity studied with peroxidase as a tracer. / Cell Biol. 98: 2 1 3-236. Margulis, L., G. Thorington, B. Berger, and J. Stolz. 1978. Endosymbiotic bacteria associated with the intracellular green algae of Hydra virtdis. Current Microbiol. 1:227-232. 140 E. E. PALINCSAR Miller, L. T. 1984. Gas-liquid chromatography of cellular fatly acids as a bacterial identification aid. Hewlett-Packard Application Note, Pp. 228-237. Moss, C. W. and V. Lewis. 1967. Characterization of Clostridia by gas chromatography. I. Differentiation of species by cellular fatty acids. Applied Aficr. 15: 390-397. Palincsar, E. E., VV. R. Jones, and J. S. Palincsar. 1988. Bacterial aggregntesvjiihinAiptasiapallida.Am Zoo/. 28(4): 149. Pardy, R. 1976. The production of aposymbiotic hydra by the photo- destruction of green hydra zoochlorellae. Bioi Bull 151: 225-235. Peters, E. C. 1984. A survey of cellular reactions to environmental stress and disease in Caribbean scleractinean corals. Helgol. Mecre- siinlurs. 37: 113-137. Peters, E. C., J. J. Oprandy, and P. P. Yevich. 1983. Possible causal agent of "White Band Disease" in Caribbean acroporid corals. J. Invert. Pathol. 41: 394-396. Reynolds, E. S. 1963. The use of lead citrate at high pH as an electron- opaque stain in electron microscopy. J. Cell Biol. 23: 27 1-288. Sasser, J. M., D. H. Fieldhouse, and C. N. Carter. 1984. Computer assisted identification of bacteria based on fatty acid analysis. Phyto- pathology74(T):S&2. Southward, A. J., E. C. Southward. P. R. Dando, G. H. Rau, H. Fel- beck, and H. Flugel. 1981. Bacterial symbionts and low 13 C/12 C ratios in tissues of Pogonophora indicate unusual nutrition and metabolism. Nature 293: 616-620. Southward, E. C. 1 986. Gill symbionts in thyasinds and other bivalve molluscs. J. Mar. Bio/. Ass. 66: 889-914. Soyer, J., M. Soyer-Gobillard, C. Theriot-Quievreux, M. Bouvy, and G. Cahet. 1987. Chemoautotrophic bacterial endosymbiosis in Spisula subtruncata. Ultrastructure, metabolic significance and ev- olutionary implications. Symbiosis 3'. 301-314. Thorington, G., B. Berger, and L. Margulis. 1979. Transmission of symbionts through the sexual cycle of Hydra viridis I. Observations on living organisms. Trans. Am. Microsc. Soc. 98(3): 401-413. Vacelet, J. 1975. Etude en microscopic electronique de 1'association entre bacteries et spongiaires du genre I'erongia. J Microscopic Biol. Cell. 23:271-288. VVilkerson, F. P. 1980. Symbionts involved in phosphate uptake by green hydra. Pp. 269-277 in Symbiosis and Cell Biology, Vol. 1 , W. Schwemmler and H. E. Schenk, eds. Walter de Gruyter, Berlin. Wilkinson, C. R. 1978. Microbial associations in sponges. Mar. Biol. 49: 161-185. Reference: Biol. Bull 111: 141-145. (August, 1989) FMRFamide-like Immunoreactivity in the Nervous System of the Starfish Asterias rubens MAURICE R. ELPHICK*. ROLAND H. EMSON**, AND MICHAEL C. THORNDYKE* ' * U 'olfson Laboratory, Biology Department, Royal Holloway and Bedford New College, University of London, Egham, Surrey, T\\'20 OEX, and** Division of Biosphere Science, King's College, University of London Abstract. The nervous system of the starfish Asterias rubens was subjected to immunocytochemical investiga- tion using antisera raised against the molluscan neuro- peptide FMRFamide. Immunoreactivity was detected in the radial nerve cords and the circumoral nerve ring, as well as in the sub-epithelial nerve plexus of the tube foot system. The hyponeural part of the radial cords con- tained numerous immunoreactive cell bodies. In the ec- toneural tissue, immunoreactive cells were present in the epithelium, with cell bodies especially abundant in the lateral parts of the nerve, close to the site of emergence of the innervation to the tube feet. The sub-epithelial nerve plexus of the tube feet contained immunoreactive fibers that were continuous with an extensive system of ecto- neural immunoreactive fibers in the radial nerve cords. Immunoreactive fibers were particularly evident in the regenerating radial nerves of previously sectioned arms. Introduction The isolation and characterization of FMRFamide (Phe-Met-Arg-Phe-NH:) from the clam Macrocallista nimbosa by Price and Greenberg (1977) heralded an era of intense investigation into both the occurrence and functional properties of this neuropeptide in molluscs. FMRFamide has a variety of effects on molluscan hearts, visceral and somatic muscles (reviewed by Greenberg et al.. 1983), and molluscan neurons (reviewed by Walker, 1986). Since its discovery, antisera raised to FMRFamide have been used to investigate the possibility that FMRF- Received 6 February 1989; accepted 31 May 1989. 1 To whom correspondence should be addressed. amide-like peptides occur in non-molluscan species. In- deed, such immunochemical studies have revealed that FMRFamide-like substances are present in members of most of the major animal groups, including coelenterates (Grimmelikhuijzen, 1983), platyhelminthes (Reuter et al, 1984;Gustafsson eta/., 1985), nemertines (Varndell and Polak, 1983), nematodes (Li and Chalfie, 1986; Cowden et al., 1987), annelids (Kuhlman et al., 1985; Porchet and Dhainaut-Courtois, 1988). crustaceans (Hooper and Marder, 1984; Jacobs and Van Herp, 1984), a chelicerate (Watson et al., 1984), insects (Boer et al.. 1980), and vertebrates (Boer et al.. 1980;Dockray el al., 1981). Subsequently, several of the peptides re- sponsible for this immunoreactivity have been identified and, at present, the peptides isolated from protostomian species appear to be quite distinct from those of non-pro- tostomes. The Echinodermata, a major invertebrate phylum, has so far been neglected by those interested in neuropeptide biology. Therefore, as the first step in an attempt to iden- tify FMRFamide-related peptides in echinoderms, we have carried out an immunocytochemical study of the distribution of FMRFamide-like material in the nervous system of the starfish, Asterias rubens. Materials and Methods Specimens of Asterias rubens were collected on the south coast of England, transported to Kings College, and maintained there in an aerated seawater system at ire. The starfish were narcotized in 3.5% magnesium chlo- ride, and the various parts of the nervous system were then dissected into cold (4°C) Bouin's fluid in seawater. After fixation for approximately 18 h at 4°C, the tissue 141 142 M. R. ELPHICK ET AL was embedded by routine methods in paraffin wax (58°C mp), sectioned at 7-15 ^m, and mounted on poly-L-ly- sine coated glass slides. The primary rabbit antisera to FMRFamide (1 171 from C. J. P. Grimmelikhuijzen, and LI 35 from G. J. Dockray) were applied at dilutions be- tween 1:100 and 1:1000. The 1 171 antiserum has been characterized by solid and liquid phase absorption tests with numerous, potentially cross- reactive peptides; it has high affinity for FMRFamide, as expected, but also some affinity for FLRFamide, FMKFamide, LTRPRYamide and RFamide (Grimmelikhuijzen, 1984). Two methods were used to visualize the bound pri- mary antibodies: application of a fluorescein isothiocya- nate (FITC)-labelled swine anti-rabbit second antibody; or treatment with a peroxidase conjugated goat anti-rab- bit second antibody, followed by rabbit peroxidase anti- peroxidase (PAP) complex and diaminobenzidine as the peroxidase substrate (the PAP method). Three controls were carried out: the primary antibod- ies were pre-absorbed overnight at room temperature with 10 nmol of FMRFamide per ml of diluted antise- rum; antibodies to other peptides (anti-insulin, anti-sub- stance P, anti-cholecystokinin) were tested; and non-im- mune serum was also examined. Results General morphology The major components of the starfish nervous system are the circumoral nerve ring and its five branches — the radial nerve cords — which extend along the ventral sur- face of each arm (Fig. la). These nerve tracts contain two distinct parts, the ectoneural and hyponeural systems (Fig. 1 b). The ectoneural system is further organized into an outer epithelial region containing cell bodies and sup- porting cells, and an inner axonal region traversed by fi- bers from the supporting cells (Fig. Ib). It is continuous with an extensive sub-epithelial nerve plexus of the skin, which is thickened locally to form the marginal nerve cords and the tube foot nerve ring (Fig. la). The hypo- neural system lies above the ectoneural tissue, separated from it by a thin basement membrane (Fig. Ib). A minor component of the starfish nervous system is the aboral nerve ring, which is continuous, in each arm, with the apical nerve, a small strand of tissue lying along the mid-dorsal region of the coelomic epithelium (Fig. la). Immunocytochemistry Positive FMRFamide-like immunoreactivity was de- tected in the circumoral nerve ring and radial nerve cords ofAsterias(Fig. 2a. b). Bipolar immunoreactive cell bodies were evident in papula subepithelial nerve plexus apical nerve r^rnr^^l epithelium ampulla subepithelial nerve plexus ( longitudinal nerv hyponeural cell bodies basement membrane Figure 1. Diagrammatic representation of the nervous system in Asterias. (a) Composite cross section of an arm showing the anatomy of the nervous system at the level of a tube foot (left side) and between tube feet (right side), (b) Detailed cross section of radial nerve cord. the ectoneural epithelium, interspersed between sup- porting cells (Fig. 2c). Longitudinal sections of the radial nerves show cell bodies along the entire length of the ec- toneural network, with beaded fibers throughout the ax- onal region (Fig. 4a). Transverse sections reveal that the cell bodies and associated axonal tracts are particularly concentrated laterally, close to the points where the tube feet receive innervation from the nerve ring and cords (Figs. 2b, d; 3a). Immunoreactive fibers occur throughout the sub-epi- thelial nerve plexus of the tube feet and are clearly con- tinuous with the system of immunoreactive fibers in the ectoneural part of the adjacent nerve cord (Figs. 2b; 3a, b). No immunoreactive cell bodies were detected in the nerve plexus of the tube feet. In the hyponeural part of the nervous system, immu- noreactive cell bodies are particularly abundant, and processes could occasionally be seen directed towards the FMRFAMIDE IN ASTERIAS NERVOUS SYSTEM 143 mm, « Figure 2. (a) Transverse section of radial nerve showing distribu- tion of immunoreactivity in the ectoneural (EN) and hyponeural (HN) systems. Primary antibody. 1 171, with PAP labelling. Scale bar = 100 (b) Longitudinal section ofcircumoral ring and part of tube foot epi- thelium showing lateral concentration of ectoneural cell bodies (triple arrows) and distinctive tube foot sub-epithelial nerve plexus (TFP). Pri- mary antibody. 1 171. with PAP labelling. Scale bar = lOO^m. (c) Longitudinal section of radial nerve cord showing bipolar ecto- neural cell bodies (white arrows) interspersed between supporting cells (black arrows) of the epithelium. Primary antibody, I 171, with PAP labelling. Scale bar = 30 ^m. (d) Transverse section of lateral region of the radial nerve cord near junction with tube foot showing increased concentration of immunore- active ectoneural cell bodies and fibers (arrows). Primary antibody, L 1 35, with PAP labelling. Scale bar = 30 Mm. basement membrane, although no fibers appeared to cross it in either direction (Fig. 4a, c). In preliminary experiments designed to investigate the pattern of neuronal regeneration in previously sectioned arms, the concentration of immunoreactive fibers in the ectoneural system of the regenerates was noticeably in- creased (Fig. 4b). Both FMRFamide antisera used gave positive results, but all of the control experiments, including those using antisera previously absorbed with FMRFamide, proved negative. Discussion This investigation records for the first time, the occur- rence of immunoreactive FMRFamide-like molecules in the nervous system of an echinoderm. These findings have implications for both neuropeptide phylogeny and echinoderm neurobiology. Neuropeptide phylogeny Over the last decade, FMRFamide-like peptides have been characterized in a variety of species and appear, at present, to fall into two distinct groups. First, those iso- lated from protostome phyla (Nematoda, Annelida. Mollusca, and Arthropoda) share with FMRFamide the general C-terminal sequence: F(X)RFamide, where X can be methionine, leucine, or isoleucine. Second, those peptides isolated from non-protostomes (Coelenterata and Chordata) usually share with FMRFamide only the C-terminal RFamide. Greenberg et al (1988) suggested that the protosto- mian peptides are homologous, whereas the sharing of the RFamide C-terminus with FMRFamide among the non-protostomian peptides "may merely reflect general characteristics of associations between peptides and pro- teins. If there is a homology, it is likely to reside with the class of membrane proteins comprising peptide recep- tors." Since the echinoderms are deuterostomian inver- Figure 3. (a) Oblique section through radial nerve cord. Hyponeu- ral cell bodies (HN) are clearly evident as is the sub-epithelial plexus in adjacent tube feet (arrows). Notice the high concentration of immuno- reactive cells where the nerve cord branches to innervate the tube feet (arrow heads). Primary antiserum, 1 171, with PAP labelling. Scale bar = 100 ^m. (b) Transverse section of marginal nerve cord (MN) and adjacent tube foot (TF). Immunoreactive fibers in the sub-epithelial plexus are clearly evident (arrow). Primary antiserum, 1 171, with PAP labelling. Scale bar = 30 ^m. 144 Figure 4. (a) Longitudinal section of radial nerve with cell bodies in both ectoncural ( EN ) and hyponeural ( H N ) systems as well as beaded fibers in the axonal region (arrows). BM, basement membrane. Primary antibody, L 1 35, with PAP labelling. Scale bar = 100 urn. (b) Transverse section of regenerating radial nerve to show dense concentration of immunoreactive fibers (arrows). The absence of im- munoreactive cell bodies suggests these fibers are derived from cells distal to the region of growth. Primary antibody, LI 35, with PAP label- ling. Scale bar = 100 ^m. (c) Longitudinal section of the radial nerve cord showing hyponeu- ral cell bodies, one with a process (arrow) directed toward the basement membrane (BM). Primary antibody, 1 171, with FITC labelling. Scale bar = 25 pm. tebrates. the characterization of the FMRFamide-Iike peptides in echinoderms would provide a further test of the notion that the F(X)RFamide peptides are peculiar to protostomes. M. R. ELPHICK ET AL Echinoderm neurobiology Our current understanding of echinoderm neurobiol- ogy is far behind that of most of the other major inverte- brate phyla (see Cobb, 1987, 1988). Only one native echinoderm neuropeptide has ever been thoroughly in- vestigated: gonad stimulating substance (GSS) (see Ka- natani. 1979). Discovered thirty years ago (Chaet and McConnaughy, 1959), GSS has only recently (and par- tially) been sequenced (Shirai, 1987). This is the first extensive study of peptidergic neurons in an echinoderm. The distribution of FMRFamide-like immunoreactivity in the nervous system may provide some clues as to the function of these peptides in starfish. The abundance of immunoreactivity and its presence in both the ectoneural and hyponeural systems suggests that the peptides may have a general transmitter-like role. However, the immunoreactivity is particularly as- sociated with the innervation of the tube feet. Thus the sub-epithelial nerve plexus of the tube feet contains nu- merous immunoreactive fibers, whereas the soma of these neurons appear to lie within the adjacent nerve cord or ring. Florey and Cahill (1980) demonstrated that the tube feet of sea urchins are under cholinergic motor control. Their evidence indicates that chemical transmission in- volves the diffusion of acetylcholine (ACh) from nerve terminals of the sub-epithelial plexus to the musculature, across the intervening connective tissue layer. Peptides produced by the immunoreactive neurons described here may be modulating the motor control of tube foot activity. Some circumstantial support for this idea comes from Unger's work (1960, 1962). Using simple chro- matographic methods, this author isolated two physio- logically active substances from the radial nerve cords of Asterias glacialis. In addition to effecting color changes, both factors induced movement in whole animals, as well as in isolated arms, and one of them strongly exci- tated the Helix heart. These effects were clearly distin- guishable from those of ACh, serotonin, adrenalin, nor- adrenalin, and histamine, and we speculate that the ex- tracts may have contained peptidic factors, including FMRFamide-like molecules. The immunocytochemical data presented in this re- port shows us relatively little about the chemical nature of the immunoreactive peptides since any peptide with a C-terminal sequence similar to that of FMRFamide might cross-react with the antisera. A good example is the family of pancreatic polypeptide-related peptides (PP-RP) which terminates in Arg-Tyr-amide (see Thorn- dyke, 1986, for a more complete discussion of this problem). Experiments designed to isolate and sequence the FMRFamide-like peptides in Asterias are currently un- FMRFAMIDE IN ASTERIAS NERVOUS SYSTEM 145 derway. Once one or more sequences are known and synthesized, the physiological roles of the native echino- derm molecules can be established. Note added in proof Three novel peptides, detected with antisera to FRMFamide, have now been purified from the radial nerves of the starfish Asterias, and sequenced (Elphick cl al.. 1989). Acknowledgments We thank Professor M. J. Greenberg, for reading the manuscript and offering helpful criticism, and Professor G. J. Dockray and Dr. C. J. P. Grimmelikhuijzen, for gifts of antisera. Thanks also to Pat Enser and Zyg Pod- horodecki for help with manuscript preparation. This work was partly supported by grants from the Science and Engineering Research Council, Nuffield Founda- tion, and Royal Society (to MCT). Literature Cited Boer, H. H., L. P. C. Schot, J. A. Veenstra, and D. Reichelt. 1980. Immunocytochemical identification of neural elements in the central nervous system of a snail, some insects, a fish and a mammal with an antiserum to the molluscan cardio-excitatory tet- rapeptide FMRF-amide. Cell Tissue Res. 213: 21-27. Chaet, A. B., and R. A. McConnaughy. 1959. Physiologic activity of nerve extracts. Biol. Bull 117: 407. Cobb, J. L. S. 1987. Neurobiology of Echinodermata. Pp 483-525 in Nervous Systems in Invertebrates. M. A. AH, ed. Plenum Press, New York and London. Cobb, J. L. S. 1988. Neurohumors and neurosecretion in echino- derms: a review. Comp Biocliem. Physio/. 91 C: 151-158. Cowden, C., P. Sithigorngul, J. Guastella, and A. Strctton. 1987. FMRFamide-like peptides in Ascaris xuuin. Am. Zoo/. 27: 127A. Dockray, G. J., C. Vaillant, R. G. Williams, R. J. Gayton, and N. Os- borne. 1981. Vertebrate brain-gut peptides related to FMRFam- ide and Met-enkephalm-Arg'Phe7. Peptides 2(suppl.2): 25-30. Elphick, M. R., D. A. Price, T. D. Lee, and M. C. Thorndyke. 1989. The SALMFamides: a new family of neuropeptides isolated from an echinoderm. Soc. Neurosci. Abstr. 15: (in press). Florey, E., and M. A. Cahill. 1980. Cholinergic control of sea urchin tube feet: evidence for chemical transmission without synapses. / Exp.Biol. 88:281-292. Greenberg, M. J., S. D. Painter, K. E. Doble, G. T. Nagle, D. A. Price, and H. K. Lehman. 1983. The molluscan neurosecretory peptide FMRFamide: comparative pharmacology and relationship to the enkephalins. Fed. Proc. 42: 82-86. Greenberg, M. J., K. Payza, R. J. Nachman, G. M. Holman, and D. A. Price. 1988. Relationships between the FMRFamide-related pep- tides and other peptide families. Peptides 9(Svpp\.l): 125-135. Grimmelikhuijzen, C. J. P. 1983. FMRFamide immunoreactivity is generally occurring in the nervous systems of coelenterates. Histo- ctiemistry78:361-3&l. Grimmelikhuijzen, C. J. P. 1984. Peptides in the nervous system of coelenterates. Pp 39-58 in Evolution ami Tumour Pathology ol Neuroendocrine System, S. Falkmer, R. Hakanson, and F. Sundler. eds. Elsevier, Amsterdam. Gustafsson, M. K. S., M. C. Wikgren, T. J. Karhi, and L. P. C. Schot. 1985. Immunocytochemical demonstration of neuropeptides and serotonin in the tapeworm DiphyHohothriiiin dendriticum. Cell Tis- sut'Res. 241: 171-182. Hooper, S. L., and E. Marder. 1984. Modulation of a central pattern generator by two neuropeptides. proctolin and FMRFamide. Brain Res. 305: 186-191. Jacobs, A. C., and F. Van Herp. 1984. Immunocytochemical local- ization of a substance in the eyestalk of the prawn, Palaemon serra- tiis. reactive with an anti-FMRFamide rabbit serum. Cell Tissue Res 235:601-605. Kanatani, H. 1979. Hormones in echinoderms. Pp. 273-307 in Hor- mones and Evolution. E. J. W. Bamngton, ed. Academic Press. London. Kuhlman, J. R., C. Li, and R. L. Calabrese. 1985. FMRFamide-like substances in the leech: 1. Immunocytochemical localization. J. Neurosci. 5:2301-2309. Li, C, and M. Chalfie. 1986. FMRFamide-like immunoreactivity in C. elegans. Soc. Neurosci. Abstr. 12: 246. Porchet M., and N. Dhainaut-Courtois. 1988. Neuropeptides and monoamines in annelids. Pp 219-234 in Neurohormones in Inver- tebrates, M. C. Thorndyke and G. J. Goldsworthy. eds. Cambridge University Press. Cambridge. U.K.. Price, D. A., and M. J. Greenberg. 1977. Structure of a molluscan cardioexcitatory neuropeptide. Science 197: 670-67 1 . Reuter, M., T. Karhi, and L. P. C. Schot. 1984. Immuno- cytochemical demonstration of peptidergic neurons in the central and peripheral nervous systems of the flatworm Microstomum li- neare with anti-serum to FMRFamide. Cell Tissue Res. 238: 431- 436. Shirai, H., P. Bulet, N. Kondo, M. Isobe, K. Imai, T. Goto, and I. Ku- bota. 1987. Gonad-stimulating substance of starfish. Gen. Comp. Endocrinol. 66: 50. Thorndyke, M. C. 1986. Immunocytochemistry and evolutionary studies with particular reference to peptides. Pp. 308-327 in Immu- nocytochemistry: Modern Methods and Applications. 3. M. Polak and S. Van Noorden. eds. J. Wright and Sons. Bristol. linger, H. 1960. Neurohormone bei seesternen Manhasterias gla- cialis. Symp. Biol. Hung 1: 203-207. linger, H. 1962. Experimented und histologische Untersuchungen iiber Wirkfaktoren aus dem Nervensystem von Asterias (Manha- sterias) glacialis. (Asteroidea. Echinodermata). Zool. Jahrb. Abt. Allg. Zool Physiol. Tiers J 69: 48 1-536. Varndell, I. M., and J. M. Polak. 1983. Glycolytic enzymes and a peptide closely related to FMRFamide are present in neurosecre- tory cells in nemertean cerebral ganglia. Am Zool. 23: 924. Walker, R. J. 1986. Transmitters and modulators. Pp 279-485 in The Mollusca, I'ol. 9. Pan 2. A.O.D. Willows, ed. Academic Press, New York. Watson, W. H., J. R. Groome, B. M. Chronwall, J. Bishop, and T. L. O'Donohue. 1984. Presence and distribution of immunoreactive and bioactive FMRFamide-like peptides in the nervous system of the horseshoe crab, Limu/us polyp/iemus. Peptides 5: 585-595. Reference: Bio/. Bull. Ill: 146-153. (August, 1989) Protein Kinase C Activators Enhance Transmission at the Squid Giant Synapse LUIS R. OSSES1 M, SUSAN R. BARRY24, AND GEORGE J. AUGUSTINE1 34* ' Section ofNeurobiology, Department of Biological Sciences, University of Southern California, Los Angeles, California 90089-037 J. "-Department of Physical Medicine and Rehabilitation, University of Michigan. Ann Arbor, Michigan 48109, ^Catalina Marine Science Center, P.O. Box 398, Avalon, California 90704. and ^Marine Biological Laboratory, Woods Hole. Massachusetts 02543 Abstract. We have examined the possible role of pro- tein kinase C in synaptic transmission by asking whether agents that activate protein kinase C affect transmission at the squid giant synapse. Several phorbol esters and a synthetic diacylglycerol that activates the kinase pro- duced a substantial enhancement of transmission at the squid synapse, while structurally related compounds in- capable of activating the kinase did not affect transmis- sion. These agents enhanced both postsynaptic poten- tials and postsynaptic currents, revealing that they were not enhancing transmission exclusively by increasing postsynaptic input resistance. The increase in transmis- sion produced by phorbol esters was either irreversible or reversed over a time course of one hour or longer. Kinase C activators also enhanced transmission at other syn- apses in the squid stellate ganglion. Our results are con- sistent with a general role for protein kinase C in synaptic transmission and indicate that the squid giant synapse is a favorable experimental system for further elucidation of the specific function of kinase C at synapses. Introduction Protein kinase C (PKC) is a family of calcium-sensi- tive, phospholipid-dependent protein kinases (Nishi- zuka, 1984; Coussens el at.. 1986; Knopf el ai, 1986; Jaken and Kiley, 1987; Ono etal.. 1987). Although PKC is found in high concentrations in many nervous sys- tems, including the mammalian brain (Nairn el at., 1985), its physiological functions are largely unknown. The high concentrations of PKC present in presynaptic Received 21 September 1988; accepted 23 May 1989. * To whom repnnt requests should be sent (at USC). terminals (Kikkawa £»/«/., 1982; Wu et al. 1982;Girard et al.. 1985; Unver et al.. 1986) suggests that this enzyme plays a role in synaptic transmission. Kinase C could play a role in mediating or regulating neurotransmitter release (Augustine et al., 1987). Be- cause activation of PKC enhances secretion from a vari- ety of non-neural cells (Knight and Baker, 1983; Pozzan el al. 1984; Pocotte el al.. 1985), PKC may be a neces- sary component of the molecular apparatus responsible for mediating exocytosis (Baker, 1984). Consistent with such a proposal, transmission at the guinea pig ileum (Tanaka et al., 1984), the frog neuromuscular junction (Publicover, 1985; Eusebi et al.. 1986; Haimann et al.. 1987; Shapira et ai. 1987), and certain synapses in the hippocampus (Malenkat'/c//.. 1986, 1987) is potentiated by activators of PKC. We have attempted to address the role of PKC in syn- aptic transmission by asking whether agents that activate PKC alter transmission at the squid giant synapse. The large size of the presynaptic terminal of this synapse makes it unusually suitable for detailed analysis of the physiological mechanisms underlying synaptic transmis- sion (Llinas, 1982; Augustine et al.. 1988). Protein ki- nase C is also found in squid nerve terminals (Unver et al.. 1986), making a potential involvement of PKC in transmission at the giant synapse more plausible. We have found that agents that activate PKC dramatically enhance transmission at the giant synapse and at other synapses in the squid stellate ganglion. These observa- tions provide further evidence in support of a general role for PKC in synaptic transmission and pave the way for a detailed analysis of the intracellular mechanisms that permit PKC to enhance synaptic transmission. A preliminary report of some of this work has appeared (Osseselal.. 1986). 146 K.INASE C AND SYNAPTIC TRANSMISSION 147 Materials and Methods Stellate ganglia of the squids Loligo pealei and L. opalescens were isolated and maintained by means of techniques described in detail previously (Augustine and Eckert, 1984; Augustine et a/.. 1985a). Conventional electrophysiological methods were used to stimulate the most distal "giant" presynaptic axon with extracellular wire electrodes. The connective containing the presynap- tic axon usually was dissected (Augustine and Eckert, 1984) to eliminate other synaptic inputs that innervate the postsynaptic axon (Martin and Miledi, 1986), and to permit examination of transmission at the giant synapse in isolation. We often recorded postsynaptic potentials (PSPs) from the most distal giant postsynaptic axon with an intracellular microelectrode, although in some experi- ments, postsynaptic currents (PSCs) were measured with a two-microelectrode voltage clamp. Usually these sig- nals were digitized (12-bit resolution) and stored on a Digital LSI-1 1/23+ computer system and analyzed with previously published procedures (Augustine et ai, 1985b). In a few experiments, signals were displayed on a storage oscilloscope and analyzed manually. Unless otherwise indicated, all results reported here were ob- tained in a minimum of five independent experiments. To examine the effects of various PKC activating drugs on transmission, stock solutions of these com- pounds in DMSO were prepared, and these were then mixed with squid saline (enclosed in a capped polyethyl- ene test tube and vigorously agitated with a Vortex mixer) to yield a final DMSO concentration of 0.1% or less. An identical concentration of DMSO routinely added to control salines had no obvious effect upon syn- aptic transmission. In some cases, stock solutions of 12- deoxyphorbol. 13-butyrate 20-acetate were prepared in 20 mM HEPES (pH 7.2) instead of in DMSO. Such solu- tions produced physiological responses indistinguishable from those obtained with solutions prepared in DMSO. Giant synapses bathed in normal squid saline (454 mMNaCl, 54 mjV/MgCl:, 1 1 mMCaC\2, 10 mA/KCl, 3 rnAf NaHCO,, 10 mM HEPES buffer, pH 7.2) release so much transmitter that the postsynaptic response is sufficient to produce an action potential in the postsyn- aptic cell. We therefore lowered the extracellular Ca con- centration to 2.2-2.8 mM(by equimolar substitution of MgCl2 for CaCl2) to reduce the amplitude of postsynap- tic responses below the level of action potential genera- tion, and to facilitate quantitative assessment of the effects of drugs upon transmission. Also in low Ca, the postsynaptic membrane potential was more easily con- trolled when the voltage clamp was used to measure PSCs. Under these conditions, transmission was often stable for many hours (e.g.. Fig. 2 of Augustine and Charlton, 1986). Experimental solutions were delivered to the giant synapse by three different methods. Usually, solutions were delivered via a cannula inserted into the artery that irrigates the giant synapse (Augustine and Charlton, 1986). This technique permits more rapid delivery of so- lutions to the synapse than is possible by simply adding the solutions to the bulk medium surrounding the stel- late ganglion. When using this technique, solutions often were dyed with Phenol Red to visualize the movement of solutions through the circulatory system and into the ganglion. Control experiments indicated that the effects reported in this paper were not caused by the presence of Phenol Red. In other experiments, solutions were deliv- ered by a focal pipette delivery method (Augustine et al.. 1985a), or by simple addition to the bulk medium. Very similar results were obtained when kinase C activators were delivered by any of these three methods. Results To assess the role of PKC in transmission at the squid synapse, we tested known kinase activating agents for their ability to alter synaptic transmission. The selected agents all mimic diacylglycerol, a product of membrane phospholipid breakdown that is thought to be an intra- cellular messenger responsible for PKC activation //; vivo (Kishimoto et al.. 1980; Nishizuka, 1984). Two classes of agents were examined. First, membrane-permeant phorbol esters, which apparently act at the diacylglycerol binding site of PKC to activate the enzyme (Bell, 1986; Kikkawa et al.. 1983; Nishizuka, 1984), were applied to the synapse. Second, we examined the action of a syn- thetic diacylglycerol, 1 ,2-oleoylacetylglycerol (OAG), which has a limited ability to permeate membranes and activate PKC when applied extracellularly (Kaibuchi et ai. 1982, 1983; Nishizuka, 1984). The structures of the compounds tested are shown in Figure 1 . DPBA enhances transmission The actions of the phorbol ester, 1 2-deoxyphorbol, 1 3- butyrate 20-acetate (DPBA) were studied in particular detail. When DPBA was applied to the squid synapse, the amplitude and rate of rise of the postsynaptic poten- tial (PSP) increased dramatically (Fig. 2 A). The increase often was so large that the membrane potential of the postsynaptic axon reached threshold, causing an action potential to be elicited even though the synapse was bathed in low Ca saline to attenuate transmission. Mea- suring PSPs under such conditions proved difficult, be- cause their peak and decay were obscured by the action potential. In such cases, we measured the initial rate of rise of the PSPs as an indicator of changes in transmis- sion (Miledi and Slater, 1966; Llinasetal., 1981; Augus- tine and Charlton, 1986). To reduce complications asso- 148 L. R. OSSES ET AL. CHj- O-C— (CH^- CH=CH— (C 0 II CH-O-C— CH, OH' DPBA H -"-c* 0 0 Figure 1. Structures of the compounds whose effects on synaptic transmission were examined in this study. These compounds include a variety of phorbol esters and related compounds (left) and the diacyl- glycerol, l-oleoyl-2-acetylglycerol (right). The phorbol ester derivatives are tetradecanoyl phorbol-acetate (TPA), 12-deoxyphorbol, 13-buty- rate 20-acetate (DPBA). 12-deoxyphorbol, 13-isobutyrate (DPB), and the phorbols, 4-« phorbol and 4-0 phorbol. The structures of 4-« and 4-0 phorbol are similar, except that the hydroxyl group indicated by the asterisk points down, out of the plane of the page, for 4-« phorbol but points up for 4-0 phorbol. All other phorbol esters indicated have the hydroxyl group in a configuration identical to that of 4-0 phorbol. elated with the production of postsynaptic action poten- tials, we often voltage clamped the postsynaptic terminal and directly measured the postsynaptic currents (PSCs) underlying the PSPs. Under such conditions. DPBA was still capable of increasing transmission (Fig. 2B), with changes in PSC amplitude being roughly equivalent to the increases in PSP amplitude produced by comparable concentrations of DPBA. Because DPBA increases PSCs as well as PSPs, DPBA does not enhance transmission solely by changing the input resistance of the postsynap- tic cell. Changes in transmission produced by DPBA usually were gradual in onset. An example of the time course of the changes in PSCs produced by DPBA, from the same experiment shown in Figure 2B, is illustrated in Figure 3. In this example, PSCs were maximally increased within 25 min of initiating DPBA perfusion. In other ex- periments, maximal responses were obtained at times ranging from 5 to 43 min. Part of the variability in the time course of the synaptic response to DPBA is related to the speed with which the DPBA-containing saline was delivered to the synapse via the circulatory system; i.e.. the speed of response could be correlated with the time required for the arrival of the tracer dye, phenol red, or high-Ca saline at the synapse. However, even after the DPBA solution had arrived at the preparation, there were delays until the maximum change in transmission occurred. Thus, the time course of the response of the synapse to DPBA treatment seems to be determined par- tially by the time required for DPBA to reach the giant synapse and partially by a subsequent step. Exposure of the synapse to DPBA by focal pipette delivery (Augus- tine el al. 1985a), or addition of DPBA to the bulk solu- tion in the recording chamber, produced similar in- creases in transmission, although changes in transmis- sion were very slow when DPBA was simply added to the bath. In all subsequent experiments, drugs were deliv- ered by arterial perfusion, unless otherwise indicated. The reversibility of DPBA at the squid synapse was variable. In the experiment shown in Figure 3, PSC am- plitude gradually recovered from DPBA treatment dur- ing a 1-h long rinse with drug-free saline. The decrease in PSC amplitude was due to a reversal of the effect of DPBA, rather than a progressive deterioration of trans- mission, because PSC magnitude roughly returned to its pre-treatment level and appeared to be stable for another 60 min of recording not shown on the graph. Such results were found in 7 out of 1 1 experiments in which stable recording conditions were maintained for more than 60 min following DPBA exposure. In the remaining experi- ments, the effects of DPBA did not reverse (i.e.. the incre- A 2 uM DPBA 2.5 mV B 0.5 uW DPBA 500 nA L 1 ms Figure 2. DPBA enhances transmission at the squid giant synapse. (A) Superimposed single traces of PSPs recorded in the absence of DPBA and after a 15-min exposure to 2 MA/ DPBA reveal that DPBA treatment produced a large increase in PSP magnitude. (B) Superim- posed single traces of PSCs elicited in the absence of DPBA and after a 30-min exposure to 0.5 n\l DPBA indicate that PSCs were increased to roughly the same extent as PSPs recorded from preparations exposed to the same concentration of DPBA. K.INASE C AND SVNAPTIC TRANSMISSION 149 jO.5 uM DPBA t 1.6 r 40 80 120 TIME (min) Figure 3. Time course and reversibility of DPBA effects on trans- mission. PSCs were elicited by stimulating the presynaptic axon every 60 s, and 0.5 pM DPBA was applied by arterial perfusion during the period indicated by the bar. DPBA produced an increase in PSC magni- tude (measured by integrating each PSC record) that reached a maxi- mum and then slowly recovered during prolonged exposure to DPBA- free saline. PSC integrals have been normalized by dividing by the mean value of the PSC integrals measured before DPBA treatment. ment in postsynaptic response declined less than 30% af- ter 60 min in DPBA-free saline). In these cases synaptic transmission was enhanced for hours after removal of DPBA from the saline in which the giant synapse was bathed. Such irreversibility is commonly observed with phorbol esters (e.g., Publicover, 1 985; Eusebi ?/«/., 1986: Shapirarfrt/., 1987;Storm, 1987 (and is not unexpected, given that these molecules are hydrophobic and are also not readily metabolized within cells (Bell. 1986). The experiments described above were performed with rather high concentrations of DPBA to optimize our ability to detect its actions on transmission. How- ever, DPBA could enhance transmission at substantially lower concentrations. Concentrations of DPBA as low as 50 ruA/, the lowest concentration that we tested, in- creased PSP and PSC amplitude. Results of experiments with DPBA concentrations ranging from 50 nAI to 2 ^M are summarized in the concentration-response curve shown in Figure 4. This curve shows little sign of satura- tion at 2 fiM, the highest DPBA concentration that we studied. Assuming that this curve can be described by a saturable function, the apparent KD for such a function would be 1 fiM or higher. Because of the hydrophobic nature of DPBA and the complex morphology of the squid stellate ganglion, we suspect that this value greatly over-estimates the actual affinity of the synapse for DPBA (see Discussion). In summary, we find that DPBA produces concentra- tion-dependent increases in transmission at the squid gi- ant synapse. These observations are consistent with a role for PKC in transmission at this synapse. 400 LJ 300 CO CJ 200 100 .01 0.1 1.0 10 [DPBA] (uM) Figure 4. Concentration-dependence of DPBA effects on transmis- sion at the squid giant synapse. Combined results of experiments exam- ining both PSPs and PSCs reveal that DPBA concentrations as low as 50 nAI increase transmission, with larger effects produced by higher concentrations of DPBA. Points are means of 4 to 12 replicates, com- bining measurements on both PSPs and PSCs. and error bars indicate ±S.E.M, when this value is larger than the symbol. Other PKC activators enhance transmission If kinase C is involved in transmission at the squid syn- apse, then other PKC activators should produce changes in transmission similar to those produced by DPBA. The phorbol ester, tetradecanoyl phorbol-acetate (TPA), a compound used to activate PKC in a variety of experi- mental systems (Castagna eta/., 1982; Publicover, 1985; Caratschetal.. 1986;Shapira*Yfl/., 1987), was also capa- ble of enhancing synaptic transmission (Fig. 5A). The effects of TPA upon transmission appeared similar in magnitude (Table I) and time course to those of DPBA. although the example shown in Figure 5A illustrates an A B I uU Phorbol . 2 uM OPBA Q_ 1/1 Q_ ct o Q_ CO Q_ 2 4 a: O 40 80 120 TIME (min) 0 40 80 120 TIME (min) Figure 5. Other kinase C activators, such as TPA. enhance PSPs (A), while compounds that do not activate the kinase, such as 4-alpha phorbol. have little effect (B). The relatively small effect of 4-alpha phorbol is not due to an inability of the preparation to respond to phor- bol esters, because subsequent treatment with 2 fiM DPBA produced a robust increase in transmission. Peak amplitude of PSPs were mea- sured in both experiments and were normalized by dividing by the mean amplitude of PSPs recorded prior to drug treatment. 150 L. R. OSSES ET AL Table I Effect* nl vunoiis compounds i»i tni at l he sqiiui xu/nl synapse Compound Concentration "(• Increase in synaptic response DPBA 1 M.I/ 229±58(n = 8) 2nM 321 ± 53(n = 12) DPB 1 ^M 86 ± 9(n = 5) TPA ZnM 298 + 84 (n = 4) 4-/i phorhol \ tiM 6± I2(n = 10) 4-n phorbol \ M.I/ l±23(n = 6) OAG 50 pM 20 (n = 1) 1 00 MM 45+ 18 (n = 8) Increases are expressed as mean change in PSPs or PSCs ± S.E.M. unusually slow response to TPA. Still another phorbol ester known to activate PKC, 12-deoxyphorhol, 13-iso- butyrate (Dunn and Blumberg. 1983). also increased PSP amplitude (Table I), but was less effective than DPBA. Another difference between the effect of DPB and the other kinase C activators was that, in three out of five experiments, the increase in transmission that it produced was transient. However, all three phorbol es- ters known to activate PKC enhanced transmission at the squid synapse. Conversely, structurally related compounds that do not activate PKC were not able to enhance transmission. We examined two stereoisomers of the parent com- pound, phorbol, both of which have very weak abilities to activate PKC (Castagna et a/.. 1982). Treatment of synapses with 1 ^A/ concentrations of 4-« phorbol some- times produced modest increases in transmission (Fig. 5B), but small decreases were also seen. On average, nei- ther 4-a phorbol nor 4-(3 phorbol produced consistent changes in the postsynaptic response (Table I). Thus, only compounds known to activate PKC enhance trans- mission at the squid synapse. To reinforce this conclusion, we tested the action of OAG, the synthetic diacylglycerol analog. OAG seemed to increase the size of evoked PSPs when applied to the synapse at concentrations of 50 ^AI or greater (Table I). The relatively poor efficacy of OAG, compared to the active phorbol esters, could reflect its relatively weak ability to approach and permeate synaptic membranes or the fact that phorbol esters are more potent than diac- ylglycerol in activating PKC (Bell, 1986). Nevertheless, the observation that OAG, too, enhances transmission strengthens the argument that activation of PKC in- creases transmission at the squid giant synapse. Taken together, our results obtained with six compounds sug- gest that activation of PKC underlies the potentiating effect of DPBA and other phorbol esters on transmission at this synapse. DPBA enhances transmission at non-giant synapses The postsynaptic axon of the squid synapse is inner- vated by at least three other presynaptic terminals, in ad- dition to the so-called giant terminal that has been the subject of this and many other physiological studies (Young, 1939; Martin and Miledi, 1986). In some exper- iments, these other presynaptic inputs could be unam- biguously identified and selectively stimulated. We could then ask whether the effects of PKC activators described here are restricted to the giant synapse, or are a more general feature of synapses in the squid nervous system. An example of an experiment in which the activity of both the giant and another synapse were examined is shown in Figure 6. In this experiment, an extracellular stimulus applied to the connective innervating the stel- late ganglion evoked two temporally dispersed PSCs. Ex- amination of the electrical activity of the "giant" presyn- aptic terminal with an intracellular microelectrode (lower trace in Fig. 6) showed that the earlier of the two PSCs had the appropriate synaptic delay and other fea- tures characteristic of the PSC produced by the giant syn- apse. Addition of 0.5 nM DPBA caused this PSC (single arrow) to approximately double in amplitude, as ex- pected (e.g., Fig. 2B). In addition. DPBA caused an even more substantial increase in the amplitude of the later, non-giant PSC (double arrows). In four other experi- ments, we were able to evaluate the effects of various concentrations of DPBA upon transmission at these other synapses, usually in preparations in which the giant PSC Figure 6. DPBA also enhances transmission at other squid syn- apses. In this experiment, extracellular stimulation of the viscero-stel- late connective activated both the giant presynaptic cell (Vptc) and an- other synaptic input on to the postsynaptic axon. Both the PSC pro- duced by the giant input (single arrow) and the non-giant PSC (double arrows) were enhanced by 0.5 /j.\l DPBA. The action potential of the giant presynaptic terminal also was broadened by DPBA treatment. K.1NASE C AND SYNAPTIC TRANSMISSION 151 presynaptic input had been damaged by microelectrode impalement or dissection trauma. In every case, the post- synaptic response produced by these other inputs was en- hanced by DPBA. Therefore, we conclude that PKC acti- vation regulates transmission at both the giant synapse and other synapses on the postsynaptic axon. Discussion Our results demonstrate that a number of activators of PKC, including three phorbol esters and a diacylglycerol compound, enhance transmission at the squid giant syn- apse. The effects of these compounds were concentra- tion-dependent and long-lasting. Because structurally re- lated compounds incapable of activating PKC had little or no physiological effect, we propose that activation of PKC is responsible for the ability of the active com- pounds to increase transmission. Kinase C activators en- hance transmission, not only at the squid giant synapse, but also at the frog neuromuscular junction (Publicover, 1985; Eusebi etal., 1986; Haimann et a!., 1987;Shapira etal.. 1987), guinea pig ileum (Tanaka /tf/.. 1985; Rane and Dun- lap, 1 986). This might cast some doubt upon our conclu- sion that the actions of these compounds are due to PKC activation. However, the concentrations effective at the squid synapse are comparable to those that potentiate synaptic transmission in other multicellular prepara- tions, including hippocampal slices (Malenka el al.. 1986, 1987; Storm, 1987). Higher concentrations are probably required in more structurally complex tissues because these compounds non-specifically partition into hydrophobic domains (e.g.. connective tissue) and are partially unavailable for action upon the cell under in- vestigation. Direct microinjection of PKC into the giant pre- and postsynaptic terminals, as done with a Ca/cal- modulin-dependent protein kinase (Llinas et ai. 1985), would provide the most definite test of the assertion that PKC activation enhances synaptic transmission. Site and mechanism of action of kinase C activators Our results do not allow us to state whether the action of PKC is pre- or postsynaptic. We have eliminated one possible source, namely an increase in postsynaptic in- put resistance, as the sole cause of the ability of DPBA to increase transmission. However, other postsynaptic ac- tions, such as an alteration in postsynaptic sensitivity to the transmitter (Eusebi el al.. 1985; Caratsch et al., 1986), could contribute to the effects reported here. The unknown identity of the transmitter at this synapse, combined with the small amplitude of its single-quanta! events (Miledi, 1967; Mann and Joyner, 1978; Augus- tine and Eckert, 1984), makes it difficult to assess post- synaptic contributions to the response. In other systems, PKC activators have been shown to increase the amount of neurotransmitter released by presynaptic action po- tentials (Tanaka et al.. 1984; Wakade et al.. 1985; Zurgil and Zisapel, 1985). Recent experiments suggest that DPBA also acts presynaptically at the squid giant syn- apse (Augustine et al.. 1986). PKC activation may affect transmitter release by a va- riety of mechanisms. PKC could increase the calcium- sensitivity of secretion, lowering the concentration of calcium necessary to stimulate transmitter release (Knight and Baker, 1983; Knight and Scrutton, 1984; Pozzan et al.. 1984; Pocotte et al., 1985). Because trans- mitter release appears very sensitive to intracellular pH (Drapeau and Nachshen, 1988), PKC activation might be altering presynaptic H+ regulation (Moolenar et al., 1984; Swann and Whitaker, 1985). PKC also may en- hance mobilization of transmitter (HochnertV ai. 1986). Finally, because PKC activators alter transmembrane ion currents (reviewed in Kaczmarek, 1986; Miller, 1986). a change in presynaptic ion currents could under- lie the effect of these agents upon release. Consistent with this possibility, preliminary results suggest that PKC acti- vation augments transmission at the squid synapse by decreasing presynaptic potassium current and conse- quently broadening the presynaptic action potential (Augustine et al.. 1986). Such an increase in presynaptic action potential duration is evident in Figure 6. These and other possible mechanisms of PKC action at syn- apses merit further attention. A potpourri of protein kinase actions Although the squid giant synapse has long been re- garded as the preparation of choice for biophysical analy- sis of chemical synaptic transmission, it is sometimes ne- glected by those interested in molecular aspects of synap- tic function. Our observation that PKC activators enhance transmission at this synapse, when combined with the pronounced effects observed when a Ca/calmo- dulin-dependent protein kinase is injected into the pre- 152 L. R. OSSES ET AL synaptic terminal of the squid synapse (Llinas el /r.w300- <376 yum shell length) and juvenile oysters (16 mm shell height) were measured by open-flow simultaneous calo- rimetry and respirometry. Larvae were sieved from their respective cultures, concentrated, and placed in the per- fusion chamber. The number of individuals within the chamber varied with larval size (14 to 20 X 101 for99|300-<376 nm larvae), whereas the juvenile oysters were measured as individuals. The number of prodisso- conch larvae available was only sufficient for three groups of 99 i/m larvae and one group of 133 ^m larvae to be measured by calorespirometry, whereas six groups of >300-<376 ^m larvae and five individual juvenile oysters were measured. The calorespirometric system established an equilib- rium within 3 h and the rates of heat dissipation and oxy- gen uptake under normoxic conditions (fully air-satu- rated) were continuously monitored overnight (about 8 h). By increasing the proportion of nitrogen gas to air the P0l in the reservoir of FSW was reduced step-wise through 8, 4, and 2 kPa (60, 30, and 1 5 mm Hg). These levels of hypoxia were maintained for a period of 2-3 h to establish steady-state conditions of heat flow and P0,. Below 2 kPa, only oxygen-free nitrogen gas was bubbled into the reservoir. This gradually reduced the P0, to zero over 6 h. After each experimental run, the calorespirometric system was cleaned with 10% V:V sodium hypochlorite EFFECTS OF ANOXIA ON CR.4SSOSTREA 157 solution, thoroughly washed with distilled water, and its baseline checked before the next experimental run. An additional group of juvenile oysters was held at 1 5°C and fed Isochrysis galbana for 1 4 days before mea- suring the rate of oxygen uptake in air-saturated water ( 1 2%o). To quantify the emersed anoxic rate of heat dissi- pation, these individuals were transferred to the calorim- eter chamber, which was purged with nitrogen gas. Measurement of ingest ion rate in response to hypoxia and anoxia Ingestion rates by prodissoconch (>73-<140Mm)and pediveliger (>300-<376 /urn) larvae during exposure to hypoxia and anoxia were quantified using 3.44 ^m diam- eter 'Fluoresbrite' polystyrene fluorescent microspheres (Polyscience Inc.). To encourage their phagostimulatory nature, the microspheres were added to autoclaved 12%o seawater containing an algal extract. This was prepared from a pellet of Isochrysis galbana cells that had been homogenized and centrifuged to remove cell debris. The experimental protocol involved allowing about 50 larvae to filter and ingest the microspheres for 10 min under controlled partial pressures of oxygen, as de- scribed below. After the exposure period, further inges- tion of microspheres was prevented by fixing the larvae in 4% buffered formaldehyde. Preliminary experiments demonstrated that fixation did not cause the larvae to egest or defecate any of the microspheres within the di- gestive system. Larvae were then washed three times in distilled water to remove loose microspheres and stored in 4% buffered formaldehyde. The number of microspheres ingested was counted us- ing an inverted microscope fitted with a fluorescence light source. The prodissoconch larvae had a sufficiently small gut and thin shell that the number of ingested mi- crospheres could be counted directly. However, for the pediveligers, especially those that were actively feeding under normoxic conditions, the microspheres aggre- gated within the digestive system such that they could not be enumerated. For these heavily feeding larvae the percentage of individuals feeding was initially deter- mined from the entire group. From this group, about 50 randomly selected individual larvae were transferred, us- ing a micropippette, to wells of a flat-bottomed tissue cul- ture plate. After the water had been evaporated at 40°C. 10 n\ HC1 was added to each well to dissolve the larval shell. The acid was then evaporated and 30% H:O: added to dissolve most of the tissue. This digestion and disrup- tion process was enhanced by placing the plates in a Brinkman sonicator bath. The microspheres at the bot- tom of each well were then counted on the inverted mi- croscope after the H:O: had evaporated. Preliminary ex- periments determined that these treatments did not alter the particle fluorescence or size. The fluorescent microsphere method of assessing lar- val feeding rates was compared to a traditional method of measuring suspension feeding activity that involves estimating the logarithmic decline in algal cell concentra- tion over time (Coughlan, 1 969). In this comparative ex- periment, 1 500 pediveliger larvae were maintained for 3 h in 3 replicate 250 ml flasks containing Isochrysis gal- bana at a concentration of 30 X 106 cells 1~'. The initial algal cell concentration was counted using a Coulter Counter and again at hourly intervals for each of the rep- licates. From this, a mean filtration rate of 54.1 ± 7.8 (S.E.) n\ h"1 individual"' was estimated. Filtration rate was measured using the microsphere method for 100 in- dividual larvae from the same culture, measured at the same time. The rate was 66.2 ±21.3 (S.E.)jillr1 individ- ual"1. This indicates that the use of algal cells and micro- spheres gave comparable filtration rates. The larger vari- ance associated with the microsphere method reflects the examination of 100 individual larvae. However, to ob- tain a measurable decrease in algal cell concentration by the Coulter Counter method, many larvae are required in each flask. This masks any high individual variability in the feeding activity of larvae. Larvae were exposed to anoxic conditions in 5 ml glass syringes using the procedure described above (anoxia survival experiments). Ten minutes before the end of the incubation period, the anoxic seawater in the syringe was replaced by anoxic seawater containing fluorescent mi- crospheres at a concentration of 1 8 X 1 06 1 ~ ' . Larvae were exposed to hypoxic conditions in 5 ml vials through which 22°C water of the appropriate partial pressure of oxygen was pumped (50 ml fr')from a reser- voir. Isochrysis galbana cells (20 x 106 cells 1~') were added to the reservoir, providing the larvae with a source of food. This ensured that the ingestion rate measured with the microspheres was a typical steady state value and was not an enhanced rate due to rapid gut filling by starved larvae suddenly presented with particles. Ingestion rates in relation to exposure time were mea- sured at three different levels of hypoxia (0.8-1.7 kPa, 2.2-3.0 kPa, and 4.0-5.6 kPa) and compared to inges- tion rates by larvae at full air saturation (21 kPa). The partial pressures of oxygen within the reservoir was con- trolled by regulating the flows of nitrogen and com- pressed air; P0, was not maintained at an absolute level, but only within the stated ranges. The outflow from the 5 ml vials was passed through the Radiometer flow cell to provide a continuous record of the P0,. Ten minutes before the exposure period was complete, the flow was stopped and 10 ^1 of concentrated suspension of micro- spheres was injected through a silicone septum in the top 158 J. WIDDOWS ET AL Crassoslrea virgmica I larvae) 24 36 48 60 Hours of anoxia r 50 o 5 35 Crassosfrea virgmica I spall 100 Hours of anoxia Figure 1. Relationship between mortality and duration of anoxia for (A) larval stages (A prodissoconch, • veliconcha. • pediveliger) and (B) juvenile or spat (*) of Crdssoslrca virginiai. Open symbols repre- sent level of mortality in the respective controls. of each vial to give a final concentration of 18 X 106 mi- crospheres 1~'. Following a 10-min incubation with mi- crospheres, the larvae were sampled and the ingestion rate quantified using the procedures described above. Additional vials of larvae were prepared in order to ex- amine the ability of pediveliger larvae to recover from 24 h of exposure to hypoxic conditions. The rate of inges- tion of microspheres by larvae was then measured fol- lowing the return of water to full air-saturation. Results Anoxia tolerance Anoxic tolerance of Crassoslrea virginiea increased with developmental stage and body size (Fig. 1 ). The me- dian mortality time (MMT; i.e., the time required to reach a 50% mortality) was approximately 11, 18, and 5 1 h for the three larval stages (82 /urn prodissoconch, 167 nm veliconcha, and 312 ^m pediveliger, respectively). The MMT for the juvenile oysters was about 150 h. There was close agreement among replicates, and the normoxic controls typically showed only 6-10% mortal- ity at the time of total mortality in the experimental (an- oxic) groups. During the initial phase of anoxia, we ob- served that the larvae maintained swimming activity for at least 30 to 60 min before settling to the bottom of the syringe. Effect ofhypoxia and anoxia on the rate of heat dissipation and oxygen uptake The effect of a reduction in the partial pressure of oxy- gen (P0J on the rate of heat dissipation (Q) and oxygen uptake (N0;) by the three larval stages (99, 1 33, and 300- 376 ^m) and juveniles are shown in Figures 2A-D, re- spectively. The results demonstrate that the metabolic rate of larvae declined at reduced P0, values and the re- sponse to short-term hypoxia and anoxia (i.e.. duration of several hours) changed with body size or developmen- tal stage. The prodissoconch larvae (Fig. 2A) maintained their rates of heat dissipation and oxygen uptake independent of P0, down to 2 kPa ( 1 5 mm Hg), whereas the larger larvae (Fig. 2B, C) and juveniles (Fig. 2D) maintained their rates of heat dissipation and oxygen uptake down to only about 8 kPa (Table I). Below this P0l, generally referred to as the critical oxygen partial pressure (Pc) (Herreid. 1980), the rates of heat dissipation and oxygen uptake become dependent upon P0, . Table I also in- cludes the oxygen partial pressure that results in a reduc- tion in the heat dissipation rate to 50% of the normoxic and typically maximum rate. The P0, at 0.5 Qnormoxia was only 0.27 kPa for 99 /urn larvae, increasing to 2.4 kPa for the 133 and 300-376 ^m larvae and 3. 1 kPa for the juveniles. These results indicate that the larvae, espe- cially the early larval stage (99 /*m), can maintain their total metabolic rate down to low P0, values. Further- more, a comparison of the rates of heat dissipation and oxygen uptake (Fig. 2A-D) shows that oxygen consump- tion continues even at low P0, and forms a significant component of the total metabolic rate. The oxycaloric equivalent in aerobic catabolism ranges from -440 to -480 kJ mol~' O2 in aquatic ani- mals (Gnaiger, 1983). When experimentally derived oxycaloric equivalent values exceed the range of theoret- ical values, then partial anaerobiosis is indicated. The higher the experimental oxycaloric value the greater the reliance on anaerobic metabolism. Experimental oxyca- loric equivalent values, which can be derived from the calorespirometric measurements, are presented in Table II. The values under normoxia (20.5 kPa) and above 2 kPa were not significantly different from the theoretical range of oxycaloric equivalents, commonly used to con- vert rates of oxygen consumption into rates of catabolic EFFECTS OF ANOXIA ON CR.4SSOSTREA 159 r-20 Crassostrea virgmica (99 /jm larvae) 14 kPa 21 — 0 40 80 -6- --20 -2 12*0 160 mm Hg 40 20 (133 ^im larvae) t 0-3 h anoxia I -^ 6-15 h anoxia -, kPa 14 21 o 0. 48 36 24 12 40 80 Partial pressure of O, 120 160 mm Hg 400 o a -200 a 40 Partial pressure of O -0.5- •o .E -0.4- -0.3- » -0.2- -0.1- D Crassosfrea virgmica (spat) 14 kPa 21 4 _ 1 — 0 40 80 120 Partial pressure of 02 Figure 2. Effect of partial pressure of oxygen on rates of heat dissipation (•) and oxygen uptake (O) by prodissoconch larvae (A, means ± range; n = 3); veliconch larvae (B); pediveligers (C, mean ± S.E.; n = 5) and juveniles (D, mean ± S.E.; n = 5) of Crassostrea virginica. The rates of heat dissipation and oxygen uptake are plotted on corresponding scales using an oxycalonc equivalent of -450 kJ mol Or ' (or 1 nmoles O.rT1 = -0.125MW;Gnaiger, 1983). 160 mm Hg heat dissipation. However, the results indicate that the experimental oxycalonc equivalents were slightly above the theoretical range at many P0l levels. This may be due to the valve closure and quiescence of a proportion of the larvae and their reliance on anaerobic metabolism. At the lowest P0, level (0.67 kPa), the rate of heat dissipa- tion by all larval stages had a significant anaerobic com- ponent, and 40% of the juveniles were closed and totally anaerobic. At 1.33 and 2 kPa the juveniles also had a significant (ANOVA; P < 0.05) anaerobic contribution to the total heat dissipation. In Table I, the anoxic rates of heat dissipation (Qan0xia) during the initial 3 h of anoxia are expressed as a propor- tion of the normoxic rates of heat dissipation (Qn0rmoxia)- The 99 ^m and 1 33 /urn larvae had relatively high values, which were 34% and 23% of Qn0nnom5 respectively. How- ever, after 6 h of anoxia, the Qanoxm of 133 jum larvae declined to 7% of Qnormoxia and was maintained at this 160 J. WIDDOWS ET AL. Table I Descriptors oflhe effect of oxygen partial pressure (P0,) on the metabolic rate of three sizes of oyster (Crassostrea virginica) larvae and juveniles Oyster larvae Spat (Shell length) (Shell height) Size: 99 tim 133 /on >300-<376Mm 16 mm Pc (kPa)a 2 8 8 8 Po2(kPa)at0.5Qlnormmia)b 0.27 2.4 2.3 3.1 P0,(kPa)at0.5N0,,normoxla) 1.05 2.4 2.3 3.5 v300-<376 ^m) were recovered after > 1 1 h of anoxia. The P0, in the calorespirometric chamber was allowed to rise rapidly to normoxic levels while rates of heat dissi- pation and oxygen uptake were being monitored contin- uously. In each case, the response was a very rapid rise in both heat dissipation and oxygen uptake rates (Fig. 3). Due to the rapid changes in rates of heat dissipation un- der these conditions the apparent rate was converted to the instantaneous rate (Gnaiger, 1983), using an expo- nential delay correction (i.e.. 10 min time constant for the calorimetric system). In contrast to the slight over- shoot (32%) in the rate of heat dissipation, the larvae showed a marked overshoot (ca. 300%) in the rate of oxy- gen uptake (termed the oxygen debt payment). Oxygen uptake by the larvae returned to a near steady rate after 2-3 h. The mean oxycaloric equivalent during the first hour was -180 kJ mol ' O: and increased to -377 kJ mor1 O: and -500 kJ mol~' O: after 2 and 3 h, respec- tively. This was followed by a period (3-6 h recovery) when the oxycaloric equivalent was maintained at -583 Table II Experimental oxycaloric equivalents (kJ mol ' O;; derived by simultaneous calonmelry and respirometry) describing the nature of energy metabolism of oyster (Crassostrea virginica) lan'ae and juveniles P0,(kPa) Oyster larvae (Shell length) Size: 99 ^m 133/jm Oxycaloric equivalents (kJ mol >300-<376nm J Mean ± range for 99 ^m larvae. b Mean ± S.E. for 300-376 ^m larvae and spat. Spat (Shell height) 16 mm 20.5 -450 ± 14J -477 -469 ± llh -486+ 5b 4 -505 ±33 -524 -548 ± 53 -508 ± 15 2 -504 ±57 -462 -490 ± 22 -6I6±21 1.33 -584 ±41 -472 -549 ±37 -658 ± 46 0.67 -694 ± 54 -644 -598 ± 50 a 0 « a « a EFFECTS OF ANOXIA ON CR.4SSOSTREA 161 Table III Comparison between anoxic and normoxic metabolic rates (mean ± S.E.) by juveniles oj "Crassostrea virginica (30 mg dry tissue mass) at 15°C(n = 9) and22°C(n = .V 15°C 22°C Normoxia Respiration rate (^molesO2h ') 1.564 ±0.144 2.789±0.152 Respiration rate (J h ') 0.713 ±0.066 1.360 + 0.071 Anoxia Heat dissipation rate (J h ' ) Anoxic rate as a proportion of normoxic rate 0.037 + 0.005 0.035 + 0.005 0.05 0.03 kJ mol ' O2 and then declined to -486 kJ mol ' O: after 6 h. Effect oj hypoxia and anoxia on the rate ofingestion The general response of prodissoconch and pediveliger larvae to hypoxia and anoxia was a reduction in the pro- portion of the larvae actively feeding and ingesting mi- crospheres over 24 h exposure, and a marked decline in the rate ofingestion of microspheres within the initial 2- 3 h (Fig. 4A-D; Fig. 5A-D). Although the data for pro- dissoconch larvae were more variable, with the percent- age of larvae feeding being inexplicably depressed at cer- tain times (e.g., control or time zero and 2.75 h for P0, of 4.0-5.6 kPa, Fig. 4A; 2.25 h and 4 h for P0; of 0.8- 1 .7 kPa, Fig. 4C), the reduction in the rate ofingestion was small compared to the reduction shown by pediveligers under hypoxic conditions. The ingestion rates by prodis- soconch larvae were generally between 5 and 8 micro- spheres larva" ' per 10 min under normoxic conditions. Ingestion rates declined to about 2 microspheres lar- va"' 10 min ' after 2-3 h of hypoxia (<5.6 kPa) and these rates were then maintained for the 24 h of exposure to hypoxia (Figs. 4A-D). However, prodissoconch larvae exposed to anoxia showed no evidence ofingestion after 10 h of anoxia (Fig. 4D). Pediveligers also showed a gradual decline in the pro- portion of larvae ingesting microspheres with increasing duration of exposure to hypoxic and anoxic conditions (Figs. 5A-D). Under normoxia (2 1 kPa), the pediveligers were more consistent in their feeding activity compared to the prodissoconch stage larvae. More than 60% were actively feeding and ingesting under normoxia, and the proportion of inactive larvae increased with the duration of exposure and the degree of hypoxia. For example. ^20% were feeding and ingesting after: 1 4 h at 4-5.6 kPa, 7 h at 2.2-3 kPa, 4 h at 0.8-1.7 kPa, and <1 h at 0 kPa (Figs. 5A-D). The rates ofingestion by pediveligers un- der fully air-saturated conditions (controls, time zero) were always above 40 microspheres larva"1 10 min ', whereas under hypoxic conditions (<5.6 kPa) the mean ingestion rate was always <14 microspheres larva"1 10 min"1. However, even after 24 h at a P0, of 0.8-1.7 kPa, 4% of the larvae had ingested an average of 5 micro- spheres larva"1 10 min '. Visual observations suggested that in all conditions, the larvae exhibited some degree of activity that generally involved moving or swimming at or near the bottom of the experimental chamber. After 24 h exposure to the three levels of hypoxia (i.e., 4-5.6, 2.2-3.0, 0.8-1.7 kPa). pediveliger larvae were al- lowed to recover under normoxic conditions in the pres- ence of food before measuring their ingestion rates. A high proportion (>60%) of larvae recovered after 24 h at P0, 4-5.6 kPa and showed signs of feeding, but their in- gestion rates had recovered only partially (22% of the pre-exposure rate). Larvae exposed to the lower P0, con- ditions showed limited recovery, and only to levels of feeding and ingestion comparable to those recorded in the early stages of hypoxia (Figs. 5A-C). -500 -400 _-300 1-200 0) x-100 Crassostrea virginica (300 /jm larvae) Oxygen uptake I I I I I I I I I Heat dissipation 1000 800 600 m 400 °- 200 01 23456 Anoxic-Normoxic Recovery (Hours) Figure 3. Recovery under normoxic conditions of rates of heat dis- sipation (solid line) and oxygen uptake (broken line) by pediveliger lar- vae ofCrassoslrea virginicu following 1 1 hours of anoxia. 162 J. WIDDOWS ET AL. Prodissoconch Larvae Time (h ) Time < h ) 80 A Po 4.5-6.0 kPa 80 B PO 2.2-3.0 kPa 60 8 o o 40 40 ° 0 o 1 O 0 O. 4 . en c 20 w - 1 1 13 ^ 0 6 12 18 24 0 6 12 18 24 03 i 80 03 C PQ 0.8-1.7 kPa 80 - 0 D P0 0 kPa _1 I 60 o 8 QJ o Q- 40 I- 40 0 . 00° 1 4 20 ll n 1 1 1 1 1 I n n ill, , 12 0 - 112 • 4 ^ o 3 Figure 4. Effect of partial pressure of oxygen and duration of exposure on the feeding activity of prodis- soconch larvae of Crassostrea virginica. Feeding activity is expressed in terms of the proportion of larvae feeding (O) and their ingestion rates (number of microspheres ingested per 10 min.; histograms) during hypoxia ( A, 4.5 to 6 kPa; B, 2.2 to 3 kPa. C, 0.8 to 1 .7 kPa) and anoxia (D, 0 kPa). Discussion Metabolic response to anoxia Oyster larvae tolerate anoxia, and this tolerance in- creases with larval development. This suggests that all larval stages have anaerobic metabolic pathways capable of sustaining life for a limited period of time under an- oxic conditions. The increase in anoxia tolerance with larval development appears to be related to an ability to lower their rates of heat dissipation and thus conserve energy expenditure under anoxia. Figure 6 illustrates such a relationship (r = 0.99) in a plot of Qanoxia/Qnormoxia against the reciprocal of anoxia tolerance ( 1/MMT in h). The early larval stages maintain a high rate of heat dissi- pation (34% of Qnormoxia)) at least during the initial 3 h of anoxia, whereas the later stages conserve energy by reducing anoxic rates of heat dissipation (3% of Qnormoxia)- These differences in the level of Qanoxia appear to correlate with the relative anoxic rates of ingestion (1R) by the different larval stages. After 2 h of anoxia, IRanox,a/IRnormox,a was 29% (2.5/8.5, Fig. 4D) for the pro- dissoconch larvae compared to 5% (2.25/42, Fig. 5D) for the pediveliger larvae. Therefore, the higher anoxic met- abolic rates by the prodissoconch compared to the pedi- veliger appear to be coupled to their higher levels of ac- tivity under anoxia (i.e., the observed swimming activity and measured feeding activity). These results provide evidence of differences in the type of anoxibiosis sustained by different developmental stages of the species: 'functional anoxibiosis' in the early larval stages, characterized by high power output at the expense of efficiency, compared with 'environmental an- oxibiosis' in later developmental stages, characterized by low power output and a relatively high efficiency of an- oxic energy conversion (Gnaiger, 1983). If the prodisso- conch larvae use glycogen in anaerobic metabolic path- ways, then the relatively high anoxic rates of heat dissipa- tion represent a very high cost, at least in terms of rate of glycogen use, because of the lower biochemical efficiency or ATP yield per glycosyl unit in anaerobic metabolism (4.71 mol ATP per mol of glycogen for the succinate pathway compared to 37 mol ATP per mol of glycogen for aerobic catabolism; Gnaiger, 1983). However, in con- trast to the situation in adult bivalves, polysaccharides form a relatively small proportion (e.g., 2-3% of organic matter) of the energy store in bivalve larvae ( Holland and Spencer, 1973; Gabbott, 1976; Mann and Gallager, 1985) and in response to nutritional and environmental stress they use lipid (especially neutral lipid) and protein substrates. While lipid represents an efficient form of en- ergy reserve for aerobic catabolism, because of its high energy content (39.5 kJ g '), lipid cannot be used as a substrate for anaerobic metabolism. Bivalve larvae may prove to use protein and amino acids as energy substrates EFFECTS OF ANOXIA ON CR.4SSOSTREA 163 Pediveliger Larvae 80 ' 1|7 A PO2 4 5-6.0 kPa 2i 3 - 80 5f .6 B PQ2 2.2-3.0 kPa 16 1 1 D 60 12 - 15 "S 1 d (B o — n 0 0 _ 40 40 . 8 : o o I CT E 20 •o 0) "- o III 1, 4 0 0 h ta 4 ! 3 o 0 o 06 12 18 24" 18 0 6 12 24"3.5 ra Rec Rec •£ 2 80 nj _j 1 3 C PQ,, 0.8-1.7 kPa 4 ! D Pg2 0 kPa 18 - 16 | CD c 60 O - . 12 - to rti 0) D) °- 40 • - 8 o 1 1 I 3 Q. 20 . 4 ° n III 1 L , 3 0 6 12 18 24" 18 0 6 12 18 24" 3 Time (h) Rec Time (h) Rec Figure 5. Effect of partial pressure of oxygen and duration of exposure on the feeding activity of pedi- veligers ofCrassoslrea virginica. Feeding activity is expressed in terms of the proportion of larvae feeding (O) and their ingestion rates (number of microspheres ingested per 10 min.; histograms) during hypoxia (A, 4.5 to 6 kPa: B. 2.2 to 3 kPa; C. 0.8 to 1.7 kPa), anoxia (D, 0 kPa) and after 3 to 18 h of normoxic recovery. for anaerobic metabolism, similar to sea anemones (An- thozoa), which use protein and accumulate nitrogenous end-products during anoxia (Ellington, 1980, 1982;Na- varro and Ortega, 1 984). The relatively high anoxic rate of heat dissipation by prodissoconch larvae presumably enables the larvae to maintain their observed locomotory activity and to swim away from anoxic conditions (unpub. obs. and V. S. Kennedy, pers. comm.), but at considerable cost to their energy reserves, which are relatively small in the early developmental stages (Gabbott, 1976). These high costs are likely to result in a rapid depletion of energy reserves, thus limiting the anoxia tolerance of early larval stages. This relationship between the anoxic rate relative to the normoxic rate of energy expenditure and anoxia tol- erance is also apparent in interspecific comparisons with adult bivalves. For example, the clam (Miilinia lateralis), has a relatively high Qanoxia (i.e., 97% of Qnormoxia; Shum- way el ai. 1983) and is relatively intolerant of prolonged anoxia (ca. 5 days). The high anoxic energy demands of M. lateralis (confirmed by direct calorimetry; Shumway el ai, 1983) are also related to the maintenance of loco- motor and feeding activities under anoxia, which may be an adaptation to escape periodic burial in unstable, oxygen-deficient sediment. In contrast, the sessile mussel (Mytilus edulis) has a very low Qanox.a (i-e- 4% of Qnormoxia; Widdows, 1987) which enables it to conserve energy and thus tolerate prolonged anoxia (ca. 35 days, Theede^fl/., 1969). Metabolic response to hypoxia Oyster pediveliger larvae and juveniles maintain rates of oxygen consumption down to relatively low P0, val- ues, as indicated by the Pc values and P0, at which the NO, is 50% of the normoxic N0, (Table I). Similar values have been recorded for adult Crassostrea virginica (0.4 g) under comparable conditions of 20°C and 14%oS (Pc of 8 kPa; P0, at 0.5 Qnormmia of 3.4; Shumway and Koehn, 1982). The ability of the prodissoconch larvae to maintain their rates of oxygen uptake down to extremely low levels of P0, appears to be a feature of very small individuals (such as the early larval stages) as a result of the high surface area to volume ratio and the short dis- tance for oxygen diffusion ( Herreid, 1 980). Furthermore, low aerobic demands permit diffusion of oxygen to meet the needs of very small animals, as long as the animals are active and keep the surrounding water stirred. Oyster larvae maintained ciliary activity on the velum down to anoxic conditions, presumably to meet their swimming, respiratory, and feeding requirements. There is no evidence of a major anaerobic component 164 J. WIDDOWS ET AL 0.4 -5 0-3 20.2 a o z 0.1 Crassostrea virginica Prodissoconch larvae Veliconcha larvae Ju Pediveligers veniles (spat) 0.05 1 / Anoxia tolerance (MMT in hours) 0.1 Figure 6. Relationship between anoxic rate/nornioxic rate of heat dissipation and anoxic tolerance ( 1 /median mortality time in hours) of larvae and juveniles of Crassostrea virginica. of total energy metabolism by larvae and juveniles of Crassostrea under hypoxic conditions. Measured oxyca- loric equivalents are only slightly higher than theoretical values for totally aerobic catabolism from 20.33 down to 1.33 kPa. and this increase is not significant (P < 0.05) until 0.67 kPa (Table II). Recovery of pediveliger larvae (300 ^m shell length) after 1 1 hours of anoxia (Fig. 3) shows a similar meta- bolic response to that previously recorded for adult mus- sels, Mytilus edulis (Shick el a/., 1986, 1988; Widdows, 1987). The overshoot in rate of heat dissipation was only 32%, whereas the overshoot in oxygen uptake was >300%, resulting in an experimental oxycaloric equiva- lent of - 1 80 kJ mor ' O: . Experimental studies on adult bivalves have divided the overshoot in oxygen uptake (the 'oxygen debt payment') into two basic components, the metabolic component and the physical reoxygen- ation of water and oxygen stores within the valves of bi- valves during the initial phase of normoxic recovery. This phenomenon is often recorded as a sudden 'oxygen sag' in the oxygen trace and coincides with the opening of the shell valves and the flushing out of deoxygenated water (ca. 15% of oxygen overshoot; Shick el a/.. 1986). Such an 'oxygen sag' was also observed in the oxygen traces produced by the oyster larvae, and indicates that the reoxygenation of water within the shell valves of the larvae may form a significant contribution to the over- shoot in oxygen uptake during the first 30 min. However, even after accounting for the physical reoxygenation in adult M. edulis, the heat equivalent of oxygen uptake during early aerobic recovery exceeds the total heat dissi- pation rate, which results in an experimental oxycaloric equivalent of about -200 kJ mor ' O: (Shick el a/.. 1 986; Widdows, 1987). Values significantly below -440 to -480 kJ mol ' O; indicate the conservation of heat in partially endothermic processes (biosynthesis), such as the restoration of high-energy phosphates, together with catabolic-anabolic coupled glyconeogenesis and the par- tial oxidation of the succinate that is accumulated by adult mussels during anoxia. However, the importance of these processes in the aerobic recovery of oyster larvae is unknown due to the absence of detailed information on the anaerobic metabolic pathways operating in larval stages and the accumulation of metabolic end products during anoxia. Feeding rate in response to hypoxia and anoxia The use of fluorescent microspheres provided an accu- rate method of quantifying ingestion rate, but may un- derestimate filtration rate when particles captured on the velum are rejected and not ingested (Gallager, 1988). In preliminary work using video microscopy, live larvae were observed feeding on the microspheres. These re- cordings showed that the microspheres were handled like algal cells and retained within the digestive system for many hours. They could also be clearly seen revolving in the crystalline style sac, thus indicating that polystyrene microspheres, which had been coated with an algal ex- tract, were not being rapidly passed through the gut. This is in contrast to the observation of Robinson (1981) and Gallager (1988). who reported that Mercenaria merce- naria larvae can pass latex microspheres through their guts as fast as they are ingested. The mean ingestion rates of 55 microspheres larva ' h ' (prodissoconch) and 508 microspheres larva" ' h" ' (pediveliger) recorded for Cras- sostrea virginica. were similar to our measured rates of algal ingestion and the rates for Mercenaria mercenaria (86 Isochrysis cells h"1 for 2-day-old larvae of 100 /^m and 387 cells h ' for 10 day larvae of 234 /urn) measured by Gallager ( 1988) using high-speed video microscopy. The effect of hypoxia on the feeding and ingestion rates of oyster larvae generally reflects the relationship between P0, and rates of heat dissipation by larvae. For example, a high proportion of the prodissoconch larvae were actively feeding and maintaining relatively high in- gestion rates, at least during the initial 6 h at the three levels of hypoxia (Fig. 4A-C) and this corresponded with the maintained rates of heat dissipation down to extreme hypoxic conditions (Table I). The feeding response of the early feeding prodissoconch larvae therefore appears to be an all-or-none response, and this may reflect the low nutrient reserves in these early larval stages (Mann and Gallager, 1985), which necessitates that the larvae feed almost continuously. In contrast, the pediveligers EFFECTS OF ANOXIA ON CR.4SSOSTREA 165 showed a marked reduction in both the proportion of larvae feeding and their ingestion rates at more moderate levels of hypoxia (e.g.. 2.2-3.0 kPa), and it was at these levels that the rate of heat dissipation also declined to 50% of the normoxic rate. Although feeding and inges- tion rates by pediveligers were reduced, swimming activ- ity observed in our study was maintained even after 6 h at 0.8 kPa. thus confirming the reported uncoupling of the swimming and feeding activity of the velum of M. mercenaria larvae (Gallager, 1988). The increased sensitivity of pediveligers to hypoxia may reflect problems of oxygen diffusion associated with increasing body mass and therefore reduced surface area to volume ratio. Under hypoxic conditions the supply of oxygen to pediveligers is insufficient to meet the total metabolic demands of all processes. The 30-50% reduc- tion in oxygen consumption by pediveligers at a P0, of about 2-3 kPa is not compensated by a major contribu- tion from anaerobic metabolism. As a result, there is a reduction in the rates of 'non-essential' (at least in the short-term) and costly processes, such as the ingestion, digestion, and absorption of food and growth. In con- trast, processes with relatively low costs, such as the cili- ary activity of the velum (Silvester and Sleigh, 1 984), are maintained under hypoxia, thus enabling larvae to es- cape and to resume feeding and growth when oxygen is more readily available. In concurrent experiments, V. S. Kennedy (pers. comm.) has demonstrated the impor- tance of upward swimming as a means for larvae to avoid hypoxic conditions. This interpretation of hypoxic re- sponses based on metabolic costs is supported by evi- dence of the partitioning of energy expenditure and the costs associated with different processes in the mussel (Myti/ns edulis). The cost of ciliary activity of the gills represents <3% of the total metabolic energy expendi- ture (TME), the cost of digestion and absorption of food forms is about 17% of the TME, and the cost of growth ranges from 0 to 30% of TME depending on ingestion rate (Widdows and Hawkins, 1989). Consequently, the energy conserved by markedly reducing the processes of digestion and absorption of food and growth, may largely account for the observed reduction in metabolic rate by larvae during hypoxia. Therefore, the present study demonstrates good agree- ment between calorimetric, respirometric, feeding, and behavioral measurements on the larvae and juveniles of C. virginica in response to hypoxia and anoxia. It high- lights the energetic changes associated with a change in behavior, from avoidance of anoxia in the early larval stage to tolerance in the later pediveliger larvae and juve- nile oyster. It suggests that larvae are able to survive short-term (hours) hypoxia associated with low O: in the Chesapeake Bay. However, although the larvae can toler- ate hypoxic conditions, their reduced feeding activity will reduce their growth rate and increase larval develop- ment times. This will expose the larvae for longer periods to predation by larval fishes and gelatinous carnivores, such as ctenophoTes(Mnemiopsis leidyi), hence reducing larval recruitment to the adult oyster population. Acknowledgments This work was supported by grant NA-86-AA-D- SG042 from the National Sea Grant Program, NOAA, to the University of Virginia, grant NA-86-AA-SG006 to the University of Maryland, and the Visiting Scientist Program of the Virginia Institute of Marine Science (J.W.). We are grateful to J. S. Rainer for assistance. Literature Cited Coughlan, J. 1969. The estimation of filtration rate from the clear- ance of suspension. Mar. Biol. 2: 356-358. Ellington, \V. R. 1980. Some aspects of the metabolism of the sea anemone Haliplanella lucia (Venll) during air exposure and hyp- oxia. Mar. Biol. Lett. 1: 255-262. Ellington. \V. R. 1982. Metabolic responses of the sea anemone Bu- nodosonui cavernata (Rose) to declining oxygen tensions and an- oxia. Physio! Zoo/. 55: 240-249. Gabbott, P. A. 1976. Energy metabolism. Pp. 293-355 in Marine Mussels: Their Ecology and Physiology. B. L. Bayne. ed., Cam- bridge University Press. Cambridge. Gallager, S. M. 1988. Visual observations of particle manipulation during feeding in larvae of a bivalve mollusc. In Zooplankton Be- haviour Symposium. 1987. Georgia. Bull Mar. Sci. 43: 344-365. Gnaiger, E. 1983. Heat dissipation and energetic efficiency in animal anoxibiosis: economy contra power. /. Exp. Zool. 228: 47 1 -490. Herreid, C. F. 1980. Hypoxia in invertebrates. Comp. Biochem. Phvs- ZO/.67A: 3 11-320. Holland, D. L., and B. E. Spencer. 1973. Biochemical changes in fed and starved oysters. Oslrea edulis L.. during larval development, metamophosis, and early spat growth. J. Mar. Biol. Asxoc. L>. A" 53: 287-298. Kemp, VV. P., and \V. R. Boynton. 1984. Spatial and temporal cou- pling of nutrient inputs to estuarine primary production: the role of paniculate transport and decomposition. Bull. Mar. Sci. 35: 242- 247. Mackiernan, G. B. 1987. Dissolved oxygen in the Chesapeake Bay: processes and effects. Maryland Sea Grant publication number UM-SG-TS-87-03. 177pp. Malone, T. C, W. M. Kemp, H. W. Ducklow, W. R. Boynton, J. H. Tuttle, and R. B. Jonas. 1986. Lateral variation in the production and fate of phytoplankton in a partially stratified estuary. Mar. Ecoi Prog. Ser. 32: 149-160. Mann, R. 1986. Sampling of bivalve larvae. Pp 107-1 16 in North Pa- cific Workshop on Slock Assessment and .Management of Inverte- brates, G. S. Jamieson and N. Bourne, eds. Canadian Special Publi- cation Fisheries and Aquat Sci. 92: Department of Fisheries and Oceans. Ottawa. Canada. Mann, R., and S. M. Gallager. 1985. Physiological and biochemical energetics of larvae of Teredo navalis L. and Bankiu gouldi (Bartsch) (Bivalvia: Teredinidae). J Exp. Mar. Biol. Ecol 85: 2 1 1- 228. Navarro, E., and M. M. Ortega. 1984. Amino acid accumulation 166 J. WIDDOWS ET AL. from glucose during air exposure and anoxia in the sea anemone Actinia equina(L.). Comp Bwchem. Physiol 78B: 199-202. Newell, R. I. E. 1985. Physiological effects of the MSX parasite Haplosporidiwn nelsom (Haskin. Stauber and Mackin) on the American oyster Crassostrea virgin/en (Gmelin). / She/fish Res. 5: 91-95. Officer, C. B., R. B. Biggs, J. L. Taft, L. E. Cronin, M. A. Tyler, and W. R. Boynton. 1984. Chesapeake Bay anoxia: origin, develop- ment and significance. Science 223: 22-27. Robinson, VV. E. 1981. Quantification of ingestion by Mercenaria mercenaria (L.). veligers feeding on mixed suspensions of inert ma- terial and algae using microspectrofluorimetry. J. Moll. Stud. Suppl 12A: 167-171. Seliger, H. H., J. A. Boggs, and \V. H. Biggley. 1985. Catastrophic anoxia in Chesapeake Bay in 1984. Science 228: 70-73. Shick, J. M., E. Cnaiger, J. Widdows, B. L. Bayne, and A. de Zwaan. 1986. Activity and metabolism in the mussel Mylilus edulis L. during intertidal hypoxia and aerobic recovery. Physiol. Zoo! 59: 627-642. Shick, J. M., J. Widdows, and E. Gnaiger. 1988. Calorimetric studies of behaviour, metabolism and energetics of sessile intertidal ani- mals.. -1m. Zoo/. 28: 161-181. Shumway, S. E., and R. K. Koehn. 1982. Oxygen consumption in the American Oyster Crassostrea virginica. Mar. Ecol. Prog Ser. 9: 59- 68. Shumway, S. E., T. M. Scott, and J. M. Shick. 1983. The effects of anoxia and hydrogen sulphide on survival, activity and metabolic rate in the coot clam, Mulinia lalera/is (Say). J. Exp. Mar. Biol. Ecol 71: 135-146. Silvester, N. R., and M. A. Sleigh. 1984. Hydrodynamic aspects of particle capture by Mytilus. J Mar. Biol. Assoc. U. A' 64: 859-879. Taft, J. L., W. R. Taylor, E. D. Hartwig, and E. D. Loftus. 1980. Seasonal oxygen depletion in Chesapeake Bay. Estuaries 3: 242-247. Theede, H., A. Ponat, K. Hiroki, and C. Schlieper. 1969. Studies on the resistance of marine bottom invertebrates to oxygen-deficiency and hydrogen sulphide. Mar. Biol. 2: 325-337. Widdows, J. 1987. Application of calorimetric methods in ecological studies. Pp. 182-215 in Thermal and Energetic Studies of Cellular Biological Systems, A. M. James ed., Wright, Bristol. Widdows, J., and A. J. Hawkins. 1989. Partitioning of rate of heat dissipation by Mytilus edulis into maintenance, feeding and growth components. Physiol. Zool 62: 764-784. Wood, L., and \V. J. Hargis. 1971. Transport of bivalve larvae in a tidal estuary. Pp. 29-114. Fourth European Marine Biology Sympo- sium, D. J. Crisp, ed., Cambridge University Press, London. de Zwaan, A. 1983. Carbohydrate catabolism in bivalves. Pp 137- 175 in The Mollusca. I'ol I. Metabolic Biochemistry and Molecular Biomechanics. P. W. Hochachka, ed. Academic Press, New York. CONTENTS Annual Report of the Marine Biological Laboratory 1 DEVELOPMENT AND REPRODUCTION Bosch, Isidro Contrasting modes of reproduction in two Antarc- tic asteroids of the genus Porania, with a description of unusual feeding and non-feeding larval types ... Fuller, S. Cynthia, Richard A. Lutz, and Ya-Ping Hu Bilateral asymmetry in the shell morphology and microstructure of early ontogenetic stages of Ano- Parks, Annette L., Brent W. Bisgrove, Gregory A. Wray, and Rudolf A. Raff Direct development in the sea urchin Ph\Ilacanthus parvispinus (Cidaroidea): phylogenetic history and functional modification . ECOLOGY AND EVOLUTION Enzien, Michael, Heather I. McKhann, and Lynn Margulis Ecology and life history of an amoebomastigote. 77 83 96 Paratetramitus jug<>Mi\, from a microbial mat: new evidence for multiple fission 110 Palincsar, Edward E., Warren R.Jones, Joan S. Pa- lincsar, Mary Ann Glogowski, and Joseph L. Mastro Bacterial aggregates within the epidermis of the sea anemone Aiptasia pullida 130 PHYSIOLOGY Elphick, Maurice R., Roland H. Emson, and Mi- chael C. Thorndyke FMRFamide-like immunoreactivity in the nervous system of the starfish Asterias rubens 141 Osses, Luis R., Susan R. Barry, and George J. Au- gustine Protein kinase C activators enhance transmission at the squid giant synapse 146 Widdows, J., R. I. E. Newell, and R. Mann Effects of hypoxia and anoxia on survival, energy metabolism, and feeding of oyster larvae (Crassos- trca virgin ica, Gmelin) 154 Volume 177 THE Number 2 BIOLOGICAL BULLETIN Marine Biological Laboratory LIBRARY NOV 8 1989 Woods Hole, ^ OCTOBER, 1989 Published by the Marine Biological Laboratory THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board GEORGE J. AUGUSTINE, University of Southern California RUSSELL F. DOOLITTLE, University of California at San Diego WILLIAM R. ECK.BERG, Howard University ROBERT D. GOLDMAN, Northwestern University EVERETT PETER GREENBERG, Cornell University JOHN E. HOBBIE, Marine Biological Laboratory LIONEL JAFFE, Marine Biological Laboratory GEORGE M. LANGFORD. University of North Carolina at Chapel Hill Louis LEIBOVITZ, Marine Biological Laboratory RLIDOLF A. RAFF. Indiana University HERBERT SCHUEL, State University of New York at Buffalo VIRGINIA L. SCOFIELD, University of California at Los Angeles School of Medicine KENSAL VAN HOLDE, Oregon State University DONALD P. WOLF. Oregon Regional Primate Center Editor MICHAEL J. GREENBERG. The Whitney Laboratory. University of Florida Managing Editor: PAMELA L. CLAPP. Marine Biological Laboratory OCTOBER, 1989 Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA Marine Biological Laboratory LIBRARY NOV 8 1989 Woods Hole, Mass. THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, MBL Street, Woods Hole, Massachusetts 02543. Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, Woods Hole, Massachusetts 02543. Single numbers, $25.00. Subscription per volume (three issues), $57.50($1 15.00 per year for six issues). Communications relative to manuscripts should be sent to Michael J. Greenberg, Editor-in-Chief, or Pamela L. Clapp, Managing Editor, at the Marine Biological Laboratory, Woods Hole, Massachusetts 02543. POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, Woods Hole, MA 02543. Copyright ifii 1989. by the Marine Biological Laboratory Second-class postage paid at Woods Hole, MA, and additional mailing offices. ISSN 0006-3 185 INSTRUCTIONS TO AUTHORS The Biological Bulletin accepts outstanding original re- search reports of general interest to biologists throughout the world. Papers are usually of intermediate length ( 10-40 manu- script pages). Very short papers (less than 9 manuscript pages including tables, figures, and bibliography) will be published in a separate section entitled "Notes." A limited number of solicited review papers may be accepted after formal review. A paper will usually appear within four months after its accep- tance. The Editorial Board requests that manuscripts conform to the requirements set below; those manuscripts that do not con- form will be returned to authors for correction before review. 1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies of photographs are not acceptable for review purposes.) The original manuscript must be typed in no smaller than 12 pitch, using double spacing (in- cluding figure legends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 8'/2 by 1 1 inches. Please, no right justification. Manuscripts should be proofread carefully and er- rors corrected legibly in black ink. Pages should be numbered consecutively. Margins on all sides should be at least 1 inch (2.5 cm). Manuscripts should conform to the Council oj Biology Editors Style Manual, 4th Edition (Council of Biology Editors, 1 978) and to American spelling. Unusual abbreviations should be kept to a minimum and should be spelled out on first refer- ence as well as defined in a footnote on the title page. Manu- scripts should be divided into the following components: Title page, Abstract (of no more than 200 words). Introduction. Ma- terials and Methods, Results, Discussion, Acknowledgments, Literature Cited, Tables, and Figure Legends. In addition, au- thors should supply a list of words and phrases under which the article should be indexed. 2. Title page. The title page consists of: a condensed title or running head of no more than 35 letters and spaces, the manuscript title, authors' names and appropriate addresses, and footnotes listing present addresses, acknowledgments or contribution numbers, and explanation of unusual abbrevi- ations. 3. Figures. The dimensions of the printed page, 7 by 9 inches, should be kept in mind in preparing figures for publica- tion. We recommend that figures be about 1 '/? times the linear dimensions of the final printing desired, and that the ratio of the largest to the smallest letter or number and of the ihick^st to the thinnest line not exceed 1:1.5. Explanatory matter gener- ally should be included in legends, although axes should always be identified on the illustration itself. Figures should be pre- pared for reproduction as either line cuts or halftones. Figures to be reproduced as line cuts should be unmounted glossy pho- tographic reproductions or drawn in black ink on white paper, good-quality tracing cloth or plastic, or blue-lined coordinate paper. Those to be reproduced as halftones should be mounted on board, with both designating numbers or letters and scale bars affixed directly to the figures. All figures should be num- bered in consecutive order, with no distinction between text and plate figures. The author's name and an arrow indicating orientation should appear on the reverse side of all figures. 4. Tables, footnotes, figure legends, etc. Authors should follow the style in a recent issue of The Biological Bulletin in preparing table headings, figure legends, and the like. Because of the high cost of setting tabular material in type, authors are asked to limit such material as much as possible. Tables, with their headings and footnotes, should be typed on separate sheets, numbered with consecutive Roman numerals, and placed after the Literature Cited. Figure legends should contain enough information to make the figure intelligible separate from the text. Legends should be typed double spaced, with consecutive Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to authors' current ad- dresses, acknowledgments or contribution numbers, and ex- planation of unusual abbreviations. All such footnotes should appear on the title page. Footnotes are not normally permitted in the body of the text. 5. Literature cited. In the text, literature should be cited by the Harvard system, with papers by more than two authors cited as Jones el at, 1980. Personal communications and mate- rial in preparation or in press should be cited in the text only, with author's initials and institutions, unless the material has been formally accepted and a volume number can be supplied. The list of references following the text should be headed Liter- ature Cited, and must be typed double spaced on separate pages, conforming in punctuation and arrangement to the style of recent issues of The Biological Bulletin. Citations should in- clude complete titles and inclusive pagination. Journal abbre- viations should normally follow those of the U. S. A. Standards Institute (USASI), as adopted by BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor differences set out be- low. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS (BIOSIS List of Serials; the most recent issue). Foreign authors, and oth- ers who are accustomed to using THE WORLD LIST OF SCIEN- TIFIC PERIODICALS, may find a booklet published by the Bio- logical Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K.) use- ful, sin~e it sets out the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these differ. CHEMICAL ABSTRACTS publishes quarterly supplements of additional abbreviations. The following points of reference style for THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD LIST) usage: A. Journal abbreviations, and book titles, all underlined (for italics) B. All components of abbreviations with initial capitals (not as European usage in WORLD LIST e.g. J. Cell. Comp. Physiol. NOT/ cell. comp. Physiol.) C. All abbreviated components must be followed by a pe- riod, whole word components must not (i.e. J. Cancer Res.) D. Space between all components (e.g. J. Cell. Comp. Physiol., not J.Cell.Comp.Physiol.) E. Unusual words in journal titles should be spelled out in full, rather than employing new abbreviations invented by the author. For example, use Rit \'isindafjelags Islendinga without abbreviation. F. All single word journal titles in full (e.g. Veliger, Ecol- ogy, Brain). G. The order of abbreviated components should be the same as the word order of the complete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. A few well-known international journals in their pre- ferred forms rather than WORLD LIST or USASI usage (e.g. Na- ture, Science. Evolution NOT Nature. Land.. Science, N.Y.; Evolution. Lancaster, Pa.) 6. Reprints, page proofs, and charges. Authors receive their first 100 reprints (without covers) free of charge. Addi- tional reprints may be ordered at time of publication and nor- mally will be delivered about two to three months after the is- sue date. Authors (or delegates for foreign authors) will receive page proofs of articles shortly before publication. They will be charged the current cost of printers' time for corrections to these (other than corrections of printers' or editors' errors). Other than these charges for authors' alterations. The Biologi- cal Bulletin does not have page charges. CONSISTENCY AND VARIABILITY IN PEPTIDE FAMILIES* Greenberg, Michael J., and Michael C. Thorndyke Consistency and Variability in peptide families: in- troduction 167 Steiner, D. F., S. J. Chan, S. P. Smeekens, G. I. Bell, S. Emdin, and S. Falkmer Evolution of peptide hormones of the islets of Lang- erhans and of mechanisms of proteolytic processing 1 72 Ebberink, R. H. M., A. B. Sin it, and J. van Minnen The insulin family: evolution of structure and func- tion in vertebrates and invertebrates 176 Thorndyke, Michael C., Jennifer H. Riddell, David T. Thwaites, and Rodney Dimaline Vasoactive intestinal polypeptide and its relatives: biochemistry, distribution, and functions 183 Taylor, Ian L. Peptide YY: the ileo-colonic, gastric, and pancreatic inhibitor 187 Vigna, Steven R. Tachykininsand the bombesin-related peptides: re- ceptors and functions 192 Dockray, G. J. Gastrin, cholecystokinin (CCK), and the leukosul- fakinins 195 Price, David A., and Michael J. Greenberg The hunting of the FaRPs: the distribution of FMRFamide-related peptides 198 Kobayashi, Makoto, and Yojiro Muneoka Functions, receptors, and mechanisms of the FMRFamide-related peptides 206 Nagle, Gregg T., Sherry D. Painter, and James E. Blankenship The egg-laying hormone family: precursors, prod- ucts, and functions 210 Goldsworthy, Graham, and William Mordue Adipokinetic hormones: functions and structures 218 Rao, K. Ranga, and John P. Riehm The pigment dispersing hormone family: chemis- try, structure-activity relations, and distribution . . 225 *Reprints of the proceedings of this symposium (pp. 167-225) are available for $9.00 from The Biological Bulletin editorial office. Woods Hole, MA 02543. Reference: Biol. Bull. Ill: 167-171. (October, 1989) Consistency and Variability in Peptide Families: Introduction* MICHAEL J. GREENBERG1 AND MICHAEL C. THORNDYKE2 lThe Whitney Laboratory, University of Florida, St. Augustine, Florida 32086-8623, and 2 Department of Biology, RHBNC, London University, Egham, Surrey, TW20 OEX. U. K. Oxytocin and arginine vasopressin were sequenced and synthesized by du Vigneaud in the early '50s. Since then, and especially in recent years, the number of chem- ically identified peptides has increased enormously. Tak- ing the mammals as an example, and the 1 989 Peninsula Laboratories catalog as our text, we would estimate that there are roughly 55 known regulatory peptides in a stan- dard mammalian species (i.e., consensus of man, rat, ox, and pig). Of course, the number of all mammalian pep- tides would include species variants and would be much larger. In any event, the above estimate must be very conservative; new mammalian peptides are continually being discovered, and many of the novel invertebrate se- quences being found will probably also have mammalian analogs. In the end, the total number of peptides in any mammal — and therefore in any species — may well be closer to 550 than to 55. As the pool of sequences (and sequence-watchers) has increased, the tendency of peptides to occur in families characterized by substantial similarities of structure has become more apparent. For about a decade, reviewers of the field have remarked on these families and have attempted to define them, characterize them, and ex- plain their significance (e.g., Blundell and Humbel, 1980; Niall, 1982; Bloom, 1983; Iverson, 1983; Krieger et al., 1983; Acher, 1984; Vigna, 1986; Greenberg and Price, 1983, 1988; Thorndyke, 1988). But peptide fami- * This symposium, a component of the Second International Con- gress of Comparative Physiology and Biochemistry, was held at the Louisiana State University in Baton Rouge on 1 August 1988. It was sponsored jointly by the Division of Comparative Physiology and Bio- chemistry of the American Society of Zoologists and by the Society for Experimental Biology (United Kingdom). Support for the symposium was provided by the National Science Foundation (DCB-88029 16), the Monsanto Company, and ICI (Agrochemicals) U. K. The organizers and participants are pleased to acknowledge their assistance. lies are similar to human ones in their heterogeneity and their resistance to firm definition and characterization. Distinguishing between members and friends, and dis- cerning the role of the association, are common prob- lems. The structure and functions of peptide families were recently re-examined at a symposium entitled "Consis- tency and Variability in Peptide Families'." The peptide families included in the symposium were meant to re- flect the disparity of these groups; e.g., they vary in size, diversity, and phylum of origin. The participants were also disparate in their interests, ranging from the zoologi- cal to the medical. The substance of their presentations are summarized in the ten reviews that follow, and bear on two primary issues: the synthesis and distribution of peptide families; and the effects and physiological roles of these assemblages. Synthesis The common feature of all secretory peptides is the manner of their synthesis. They are encoded by their genes as a segment of a larger precursor molecule and, after translation, the secretory products are processed out of the precursor by proteolytic enzymes. Steiner et al. (1989) suggest, referring to a model system in yeast, that the proteases and peptidases involved in post-transla- tional processing had already appeared in early eukary- otes. Although the mechanisms have been conserved in general, substantial tissue specificity — e.g., in the cleav- age sites — has evolved (see the reports by Steiner et al., 1989;Dockray, 1989;andNaglee/a/., 1989). As a practi- cal consequence of this divergence, the products of pro- cessing cannot be unerringly predicted from conven- tional processing signals known only from a clone of the gene encoding the precursor. The processed peptides 167 168 M. J. GREENBERG AND M. C. THORNDYKE must be isolated and characterized chemically (Nagle el a/., 1989). The structural similarity characteristic of peptide fam- ilies could arise in two ways. The primary one is through the duplication of the segment of a gene encoding a par- ticular peptide and its processing signals, or the duplica- tion of the entire gene; subsequent point mutation in one or the other of the sister sequences would then yield a novel peptide. This mechanism is evident where very similar sequences are part of the same precursor (e.g., VIP and PHI, substances P and K, the a-, ft-, and 7-bag cell peptides. and FMRFamide and FLRFamide), or similar precursors [e.g.. the PP-RPs, the egg-laying hor- mone (ELH) family ofAplysia, and the adipokinetic hor- mones (AKHs) of locust]. Even where sequence similar- ity is low, the peptides may have similar tertiary struc- tures (e.g., the insulin/IGF family), suggesting diver- gence from a common genetic origin. But peptide families — or new members of existing families — can also occur by chance, particularly when the peptides, or critical sequences, are short. Such exam- ples of convergence are detectable when two modestly similar peptides occur on dissimilar precursors. This is the basis for relegating the tachykinins and bombesin- related peptides to separate families (see Vigna, 1989). Similarly, its distinctive precursor leads Dockray ( 1989) to usher the amphibian peptide caerulein out of the gas- trin/CCK family. As peptides become shorter, the possi- bility that similar sequences are convergent increases. If the precursor organization of a short peptide is not known, convergence may not be detectable (Price and Greenberg, 1989). Just as familial sequences can occur by chance — i.e., without gene duplication — so the FMRFamide precur- sors in molluscs, containing 1 0-28 identical copies of the peptide, exemplify the possibility that replication can oc- cur without the subsequent emergence of a novel pep- tide. This degree of replication, incidentally, is matched by no other peptide precursor, including prometenkeph- alin or the yeast a-mating factor. Distribution Tissues All of our representative peptide families include one or more members that are expressed in nerve. The excep- tion is the sub-group of insulin-related peptides in verte- brates, the long-sought insulin-like molecule in brain having yet to be sequenced. In fact, the tendency of verte- brate neuropeptides to occur, as well, in non-nervous tis- sue— especially the gut and its derivative glands, the heart, and the skin — has been taken as a distinguishing feature of vertebrates; most known regulatory peptides of invertebrates are secreted by neurons [e.g.. the AKH-, pigment dispersing hormone- (PDH-), and FMRF- amide-related peptides in this symposium]. But the dis- covery of molluscan insulin-related peptide in snail gut (see Ebberink el ai. 1989), the occurrence of eledoisin (the first tachykinin) in octopus posterior salivary gland (see Vigna, 1989), and of ELH analogs in Aplysia atrial gland (see Nagle et a/. . 1 989) suggests that this distinction between vertebrates and invertebrates may not be valid. Possibly, the hunt for peptides in invertebrate gut and other non-neural tissues should be intensified. Phyletic distribution Congeners of peptides discovered in one species are routinely identified in other, more or less closely related, species. The set of all such congeners in all species consti- tutes an extended peptide family, and we would expect its taxonomic limits to be at least those of a phylum. We would like to know the extent to which peptide families are restricted in their phyletic distributions. This also bears on peptide evolution; extended peptide families are usually thought of as comprising congeners that have evolved from a common ancestral precursor encoded by an ancestral gene. Thus, widely ranging peptide families would have evolved earlier in metazoan phylogeny than restricted ones. Of course, the ancestral molecules are no longer available, so, as discussed above, the evidence for homology rests on the extent of the structural similarities between the genes, precursors, and peptides (as available) constituting the extant peptide family. The phyletic limits of peptide families are varied. At one end of the range, insulin-like molecules of rather similar structure occur throughout the vertebrates and share structural similarities with the prothoracicotrophic hormone (PTTH, bombyxin) of insects and with a mol- luscan insulin-like peptide (MIP) from growth-regulat- ing neurosecretory cells of the snail Lymnaea stagnalis (Ebberink et al, 1989; Steiner el al, 1989). Ebberink el a/. ( 1989) argue further that, because this superfamily in- cludes peptides sequenced in arthropods, molluscs, and vertebrates, it must have diverged from an ancestral gene present in the earliest metazoans. This hypothesis sug- gests that insulin-like peptides will eventually be found in all animal species. Peptides from four protostomian phyla have se- quences compellingly similar to FMRFamide, and they seem to be homologs. In contrast, the sequence similarity between FMRFamide and certain vertebrate and coelen- terate peptides is probably fortuitous (reviewed by Price and Greenberg, 1989). Therefore, although the family of FMRFamide-related peptides is polyphyletic, it is more narrowly distributed than the insulin-like peptides. If we accept only sequences as evidence, then most vertebrate peptide families appear to be restricted to that PEPTIDE FAMILIES 169 group. The few exceptions, however, tend to inhibit dogma. These are: eledoisin, the molluscan tachykinin (see Vigna, 1989); the molluscan opioid peptides (re- viewed by Greenberg and Price, 1988); and the lobster neurotensin (Kirschenbaum and Carraway, 1986). And although only vertebrate sequences are known from the large VIP/PHI/PHM/glucagon/secretin family. Thorn- dyke et al. (1989) provide evidence of similar peptides in flatworms. On the other hand, the leucosulfakinins are probably not members in cockroaches of the vertebrate gastrin/CCK family (see Dockray, 1989). With the exception of the FMRFamide-related pep- tides, and on current evidence, the invertebrate peptide families are, like the vertebrate ones, limited to a single phylum. An encouraging finding is that congeners of the pigment-dispersing hormones (PDH) of crustaceans have now been demonstrated in insects as well (Rao and Riehm. 1989). Thus, their distribution is parallel to that of the family of peptides related to AKH of insects and red pigment-concentrating hormone (RPCH) of crusta- ceans. In summary, most peptide families seem to have re- stricted ranges; but the restriction may only reflect the technical difficulties of identifying peptides in unusual places. New Peptide Families Most of the peptide families found recently have been detected by the effects of extracts on bioassays. Among the invertebrate assay preparations, the anterior byssus retractor muscle (ABRM) of Mytilus (Hirata et al., \ 987, 1988), the accessory radula closer (ARC) muscle of Aplysia (Cropper et al., 1987; 1988), and the cockroach hindgut (G. M. Holman, references in Goldsworthy and Mordue, 1989) have been especially productive of new peptide families. Some success has also been achieved with relatively unselective chemical assays, genetic tech- niques, and radioimmunoassay (RIA) (see Price and Greenberg, 1989; Elphick^/a/.. 1989). However, the lat- ter two techniques are aimed at chemical structures that are already known. Therefore, although developing new bioassay techniques in strange species is inconvenient, it provides the best hope of identifying the large number of yet unknown peptide families. Functional Significance of Peptide Families Tissue-specific synthesis The co-evolution of families of peptides and their re- ceptors is widely seen as augmenting the number and va- riety of regulatory agents in a species. Where the sibling peptides occur on separate precursors, the genes encod- ing the precursors are expressed in specific tissues and may, therefore, have distinct physiological roles. Even when two or more sibling peptides are produced from a single precursor, tissue-specific processing still permits tissue-specific function. Moreover, there is a growing ros- ter of alternative modes of transcription and processing [considered by Dockray, 1989, and Vigna, 1989; but also seeNewcombrttf/., 1988; and Fisher et al., 1988(ELH); Hekimi et a!., 1989; and Schulz-Aellen et al., 1989 (AKH)]. The high degree of replication of FMRFamide in its precursors remains enigmatic, but the variety of processing signals at the C-terminals of the copies sug- gests that differential processing could be regulating the quantity of peptide released at specific sites, or under par- ticular conditions. A hierarchy of actions Most peptides have a repertoire of fundamental ac- tions, each initiated by binding to one of a set of comple- mentary receptors, and mediated by one of the several common intracellular mechanisms. The many effects of FMRFamide on nerve, muscle, and gland cells are illus- trative (Kobayashi and Muneoka, 1989). But for many peptides, and especially the newly discovered ones, the range of effects is not yet close to being known. An exam- ple is the insect analog of PDH which is assayed on crus- tacean melanophores, not present in insects (see Rao and Riehm, 1989). Complementary actions of peptides at different sites often produce an integrated effect on organs or systems. Dockray ( 1 989) refers to the cluster of effects as "families of actions" pointing in particular to the actions of CCK on the pancreas, gall bladder, and stomach which serve to "regulate the environment of the small intestine." Other families of actions were reported: e.g.. enhanced fluid and protein secretion by intestinal and digestive glands caused by VIP (Thorndyke et al., 1989); inhibi- tion of pancreatic and gastric activity and delay of gastric emptying and intestinal transport to enhance digestion and absorption in small bowel malabsorption, by PYY (Taylor, 1989); and mobilization, transport, and utiliza- tion of metabolic fuel for flight in locust by AKH (Gold- sworthy and Mordue, 1989). At the next highest level of integration, organismal functions are commonly allocated among the peptides in a family. The division of labor among sibling peptides — e.g., CCK and gastrin; PPY and pancreatic polypeptide (PP); and substances P and K — in regulating aspects of vertebrate gut function is set out by Dockray (1989), Taylor (1989), and Vigna (1989), respectively. In other instances of complementarity: the ELH-gene products and atrial gland peptides affect reproductive behavior in Aplysia, in the latter case as secreted pheromones (Nagle et al., 1989); the tetrapeptide and heptapeptide FaRPs 170 M. J. GREENBERG AND M. C. THORNDYKE have distinct effects on muscle tone and posture in pul- monate snails (references in Price and Greenberg, 1989); and FMRFamide and FLRFamide modulate opposed gastropod buccal muscles (Kobayashi and Muneoka, 1989). AKH I and II may have complementary actions on the fat body of the locust (Goldsworthy and Mordue, 1989). Finally, the lack of specific actions by SCPA and SCPB in Aplysia (see Lloyd, 1986) suggests that separate receptors have not yet evolved, or have not yet been found. Specificity of action by sibling peptides is provided by two primary mechanisms. First, the receptors can have different specificities, and this is certainly the case for some of the vertebrate gut peptides (Thorndyke et al., 1989; Vigna, 1989; Dockray, 1989) and for the FaRPs (Payza, 1988; and Kobayashi and Muneoka, 1989). Moreover, the evolution of new receptors has. in some cases, occurred relatively recently (Dockray, 1989; Vigna, 1989). However, sibling peptides with different effects can act at what appears to be the same receptor; e.g., egg-laying induced by atrial gland peptides (Nagle et al.. 1989) and vasoconstriction and inhibition of pancre- atic secretion by PYY and NPY (Taylor, 1989). In such instances, specificity is ensured by differences in the tim- ing and site of release (see especially Taylor, 1989). Vigna (1989) cites two actions of bombesin — acid se- cretion by the stomach, and reduction of body tempera- ture— demonstrable both in mammals and fish, but effected by different mechanisms in the two groups. He suggests, then, that peptide actions, if not their mecha- nisms, are conserved in evolution. The actions of pep- tides certainly appear to be conserved within classes of animals (e.g., vertebrates, gastropods, crustaceans, and insects), but there has been little testing of those taxo- nomic limits. The regulation of growth by the insulin- like peptide of Lymnaea does suggest conservation, but the steroidogenic action of small PTTH-II is less con- vincing. Moreover, important metabolic functions of in- sul-n in insects seem to be carried out by the AKH-like peptides (Goldsworthy and Mordue, 1989). Finally, the chromatophorotrophic hormones of crustaceans, PDH and RPCH, have actions different from those of their ho- mologs in insects (Rao and Riehm, 1989; Goldsworthy and Mordue, 1989). Since PDH and RPCH are antago- nists in crustaceans, will the PDH analogs have actions in insects opposite to those of AKH? Finally, lest we forget, the actions of effector organs are modulated by several neuropeptides from different families, as well as by classical transmitters. This is ap- parent in most of the reports in this symposium, but is pointedly illustrated by the studies of Klaudiusz Weiss and his associates (Cropper et al.. 1987, 1988) showing that a single effector, the accessory radula closer muscle of Aplysia, is regulated by more than ten agents. So, the effect of a single peptide on a tissue, or cell, or even a subcellular fragment, can certainly be designated as its "action," but its "function" or "physiological role" is necessarily fractional or participatory. As for the effector, it can best be seen as a kind of olfactory organ, its physio- logical response dependent upon the particular combi- nation, and relative concentrations, of regulatory agents delivered to its vicinity by neurons or the circulation. The responses of effector organs are mediated by recep- tors, and further study of these proteins should help to elucidate the functional and evolutionary relationships among peptides. Literature Cited Acher, R. 1984. Evolution of neurohormonal peptides: from genetic machinery to functional tailoring. Pp. 181-201 in Evolution and tumour pathology of the neuroendocrine system, S. Falkmer, R. Ha- kanson, and F. Sundler, eds., Elsevier Science Publishers. Amster- dam. Bloom, F. E. 1983. The endorphins: a growing family of pharmaco- logically pertinent peptides. Ann. Rev Pharmacol. Toxicol. 23: 151-170. Blundell, T. L., and R. E. Humbel. 1980. Hormone families: pancre- atic hormones and homologous growth factors. Nature 287: 781- 787. Cropper, E. C., P. E. Lloyd, \V. Reed, R. Tenenbaum, I. Kupferman, and K. R. Weiss. 1987. Multiple neuropeptides in cholinergic mo- tor neurons of Aplysia: evidence for modulation intrinsic to the mo- tor circuit. Pntc. Nail. Acad. Sci.. USA 84: 3486-3490. Cropper, E. C., R. Tenenbaum, M. A. Gawinowicz Kolks, I. Kupfer- mann, and K. R. Weiss. 1987. Myomodulm: a bioactive neuro- peptide present in an identified cholinergic buccal motor neuron of Aplysia. Proc. Nail. Acad. Sci. USA 84: 5483-5486. Cropper, E. C., M. W. Miller, R. Tenenbaum, M. A. G. Kolks, I. Kup- fermann, and K. R. Weiss. 1988. Structure and action of buccalin: a modulatory neuropeptide localized to an identified small cardio- active peptide-containing cholinergic motor neuron of Aplysia cali- fornica. Proc. Nat/. Acad Sci. USA 85: 6177-6181. 'Dockray, G. J. 1989. Gastnn, cholecystokinin (CCK), and the leu- cosulfakinins. Biol Bull. Ill: 195-197. *Ebberink, R. H. M., A. B. Smit, and J. van Minnen. 1989. The insu- lin family: evolution of structure and function in vertebrates and invertebrates. Biol. Bull 111: 176-182. Elphick, M. R., D. A. Price, T. D. Lee, and M. C. Thorndyke. 1989. The SALMFamides: a new family of neuropeptides isolated from an echinoderm. Neurosci. Abst. 15: 1276. Fisher, J. M., W. Sossin, R. Newcomb, and R. H. Scheller. 1988. Multiple neuropeptides derived from a common precursor are differentially packaged and transported. Cell 54: 813-822. 'Goldsworthy, G., and W. Mordue. 1989. Adipokinetic hormones: functions and structures. Biol. Bull 177: 2 1 8-224. Greenberg, M. J., and D. A. Price. 1983. Invertebrate neuropeptides: native and naturalized. Ann Rev. Physiol 45: 271-288. Greenberg, M. J., and D. A. Price. 1988. The phylogenetic and bio- medical significance of extended neuropeptide families. Pp. 85-96 in Biomedical Importance of Marine Organisms. D. G. Fautin, W. Fenical and W. R. Kem, eds. Mem. California Acad. Sci., San Fran- cisco. Hekimi, S., W. Burkhart, M. Mover, E. Fowler, and M. O'Shea. PEPT1DE FAMILIES 171 1989. Dimer structure of a neuropeptide precursor established: consequences for processing. Neuron 2: 1363-1368. Hirata, I . I. Kubota, I. Takabatake, A. Kawahara, N. Shimamoto, and Y. Muneoka. 1987. Catch-relaxing peptide isolated from Mytilus pedal ganglia. Brain Res. 422: 374-376. Hirata, T., 1. Kubota, N. Iwasawa, I. Takabatake, T. Ikeda, and V. Muneoka. 1988. Structures and actions of Mytilus inhibitory pep- tides. Buvhem. Biophys. Res Comm. 152: 1376-1382. Iverson, L. L. 1983. Nonopioid neuropeptides in mammalian CNS. Ann Re\' Phannacol Toxicol. 23: 1-27. Kirschenbaum, S. R., and R. E. Carraway. 1986. Neurotensin-like peptide in an invertebrate. (Homarus americanus). Soc. Neurosci. Ahstr. 13: 236. *Kobayashi, M., and V. Muneoka. 1989. Functions, receptors, and mechanisms of the FMRFamide-related peptides. Biol. Bull. 177: 206-209. Krieger, D. T., M. J. Brownstein, and J. B. Martin. 1983. Introduc- tion. Pp. 1-12 in Brain Peptides, D. T. Krieger. M. J. Brownstein and J. B. Martin, eds. Wiley, New York. Lloyd, P. E. 1986. The small cardioactive peptides: a class of modula- tory neuropeptides in Aplysia. Trends Neurosci. 9: 428-432. *Nagle, G. T., S. D. Painter, and J. E. Blankenship. 1989. The egg- laying hormone family: precursors, products and functions. Biol. Bull. 177:210-217. Newcomb, R., J. M. Fisher, and R. H. Scheller. 1988. Processing of the egg-laying hormone (ELH) precursor in the bag cell neurons of Aplysia. J. Biol. Chem. 263: 12514-12521. Niall, H. D. 1982. The evolution of peptide hormones. Ann. Rev. Physiol. 44:615-624. *Price, D. A., and M. J. Greenberg. 1989. The hunting of the FaRPs. Biol. Bull 111: 198-205. *Rao, K. R., and J. P. Riehm. 1989. The pigment-dispersing hormone family: chemistry, structure-activity relations, and distribution. Biol. Bull. 177:225-229. Schulz-Aellen, M.-F., E. Roulel, J. Fischer-Lougheed, and M. O'Shea. 1989. Synthesis of a homodimer neurohormone precursor of locust adipokmetic hormone studied by in vitro translation and cDNA cloning. Neuron 2: 1 369- 1 373. *Steiner, D. F., S. J. Chan, S. P. Smeekens, G. I. Bell, S. Emdin, and S. Falkmer. 1989. Evolution of peptide hormones of the islets of Langerhans and of mechanisms of proteolytic processing. Biol. Bull. Ill: 172-175. Taylor, 1. 1989. Peptide YY: the ileo-colonic. gastric, and pancreatic inhibitor. Biol. Bull. 177: 187-191. Thorndyke, M. C. 1988. Molecular diversity and conformity of neu- rohormonal peptides: clues to an adaptive role in evolution. Biol. J. Lmnean Soc. 34: 249-267. •Thorndyke, M. C., J. H. Riddell, D. T. Thwaites, and R. Dimaline. 1989. Vasoactive intestinal polypeptide and its relatives: bio- chemistry, distribution, and functions. Biol. Bull 111: 183-186. Vigna, S. R. 1 986. Evolution of hormone and receptor diversity: cho- lecystokinin and gastrin. Am. Zoo/. 26: 1033-1040. *Vigna, S. R. 1989. Tachykinins and the bombesin-related peptides: receptors and functions. Biol Bull 111: 192-194. * Reviews by the participants in this symposium. Reference: Bwl Bull. Ill: 172-175. (October, 1989) Evolution of Peptide Hormones of the Islets of Langerhans and of Mechanisms of Proteolytic Processing D. F. STEINER1, S. J. CHAN1, S. P. SMEEKENS1, G. I. BELL1, S. EMDIN2, AND S. FALKMER1 1 Howard Hughes Medical Institute and the Department of Biochemistry and Molecular Biology, the University of Chicago, Chicago, Illinois; 'Department of Surgery, the University ofUmed, Umed, Sweden: and* Department of Pathology, Karolinska Institute Hospital. Stockholm, Sweden The islets of Langerhans in vertebrates are the sources of four major peptide hormones — insulin, glucagon, pancreatic polypeptide, and somatostatin. In develop- ment, the islets probably arise from ductal precursor cells within pancreatic rudiments that grow out from the mid- gut region of the embryo. Various lines of evidence sup- port the view that the islets are of endodermal origin, de- spite indications that islet cells seem to share some con- stituents and properties with nerve cells (Falkmer el ai, 1 984; Steiner, 1 984). In evolution the islets also appear to have arisen from the gut; e.g., inamphioxus, theexocrine and endocrine pancreatic cells are within the gut, while in the cyclostomes the insulin-producing cells and soma- tostatin-producing D cells have moved out from the in- testinal epithelium to form a small tissue mass — the islet organ — that lies near the entrance of the bile duct ( Falk- mer el al.. 1984). Studies on the biosynthesis of insulin in the hagnsh (Myxine glutinosa) have demonstrated the existence of a precursor that is closely similar in structure to the pre- proinsulins of higher forms (Steiner, 1984). The peptide begins with a typical 24 amino acid hydrophobic signal sequence, which is followed by the B chain, C-peptide, and A chain, making up the proinsulin molecule. The processing of the signal peptide occurs very early in bio- synthesis, during or shortly after the transfer of the na- scent preproinsulin chain into the cysternae of the rough endoplasmic reticulum (Steiner, 1984; Steiner et al., 1986). Hagnsh proinsulin has a connecting segment, or C-peptide, that is similar in length to that of human and other mammalian C-peptides. Because mutations have a rapid rate of fixation in this part of the molecule (about 1 5 times faster than in the insulin A and B chains), there is no sequence similarity with the C-peptides of higher vertebrates (Steiner, 1 984). Nonetheless, the typical pairs of basic residues (Lys-Arg) lie at either end of the C-pep- tide, linking it to the B and A chains in hagnsh proinsu- lin. The processing of proinsulin occurs in the early (pro)secretory vesicles derived from the trans Golgi cys- ternae and, although the time course is considerably slower in the hagfish, the proteolytic mechanism itself appears to be fundamentally similar. Hagfish insulin differs at about 40% of positions from higher vertebrate insulins, which is to be expected since the cyclostomes diverged from the vertebrate line ap- proximately 500 M years ago. Despite these substitu- tions, the folding of the peptide chain in crystals of hag- fish insulin is almost superimposable on that derived from x-ray analyses of porcine insulin crystals (Steiner, 1984). Clearly, therefore, in the evolution of proteins, it is the "fold" (tertiary structure) that is more highly con- served than the amino acid sequence (primary structure). Precursors similar in general structural organization to that of proinsulin have been identified for the other three principal islet hormones ( Fig. 1 ), as well as for most other small peptide hormones, growth factors, and neuropep- tides. In the majority of prohormones, proteolytic matu- ration occurs at pairs of basic residues, usually Lys-Arg or Arg-Arg. Occasional cleavages also occur at monoba- sic sites, such as Pro-Arg, Arg-Pro, Lys-Ser, etc., as well as rarely at non-tryptic sites such as Leu-Ala. Evidence suggests that a number of different proteases may be in- 172 ISLET PEPTIDES 173 PROPANCREATIC POLYPEPTIDE PROSOMATOSTATIN PROGLUCAGON Figure 1. Schematic comparison of the primary structures of the major islet prohormones produced in the four islet cell suhpopulations; e.g.. proinsulin (/i cell), propancreatic polypeptide (PP cell), proso- matostatin (D cell), proglucagon (a cell). All four precursors are synthe- sized initially with an N-terminal hydrophobic signal peptide (not shown). The solid lines denote the identified product peptides. Sites of processing, at dibasic or single basic residues, are indicated by the letters K. (lysine) or R (arginine); these residues are removed during the con- version process. G denotes glycine which, when it occurs at the C-termi- nal, serves as a substrate for amidation. Note: in islets of Langerhans. proglucagon is processed only at the three KR sequences to generate glucagon (heavy line) and a large C-termmal fragment ( major progluca- gon fragment) containing glucagon-like peptides 1 and 2 (GLI-1 and GLI-2); but in intestinal proglucagon-expressing cells, proglucagon is processed to liberate GLI- 1 and -2 and glicentin, a larger glucagon-con- taming N-terminal peptide. Product peptides are usually stored and secreted together, whether biologically active or inactive. Abbrevi- ations: PP, pancreatic peptide; IP, icosapeptide; CHO, potential N-gly- cosylation sites. volved in various endocrine tissues, most having cleav- age specificities similar to trypsin, but, of course, with much greater site selectivity. Perhaps the best model system for studying these en- zymes is yeast, where direct identification of several con- verting proteases has been achieved by applying genetic approaches. The yeast mating pheromone, a factor, is derived from a multicopy prohormone that is similar in its organization to many of the mammalian prohor- mones (Fuller el al, 1988). In yeast, pro « factor is pro- cessed by the Kex2 proteases at Lys-Arg sequences, to liberate several copies of an extended form of the « factor peptide which is then trimmed from the N-terminus by an amino dipeptidase, and from the C-terminus by a car- boxypeptidase B-like enzyme ( Kex 1 ) to yield the mature a factor peptide. Analysis of the gene encoding the Kex2 protease revealed a large open reading frame encoding an 814 amino acid protein (Fuller el al., 1988; Mizuno el al. 1988). Within this sequence, lying nearer the N- terminus, is a region that is homologous to the serine protease subtilisin. and near the C-terminus is a 2 1 resi- due hydrophobic region that may serve as a transmem- brane domain, indicating that the protease is membrane- associated. The Kex2 product, when expressed and puri- fied, has the expected proteolytic activity and is stimu- lated by calcium ions. When its gene is expressed in mammalian cell lines, Kex2 correctly cleaves neuroen- docrine precursors (Thomas el al, 1988). No other proteases of comparable structure to Kex2 have yet been isolated. Recently, however, Davidson el al. described two calcium stimulated proteases in rat islet tumor secretory granules (Davidson el al. 1988). Each of these selectively cleaves only one of the two sites, Arg- Arg or Lys-Arg. respectively, in rat proinsulin. This find- ing implies that mammalian enzymes somewhat similar to Kex2 may exist and, furthermore, that these en- zymes may be selective for particular pairs of basic resi- dues. A related observation of interest is that, although the insulin precursors described thus far in the most primitive vertebrates (e.g., cyclostomes and teleosts) have only Lys-Arg at cleavage sites, mammalian pre- cursors frequently contain other combinations, such as Arg-Arg. They occur, for example, in site I of pro- insulin (B chain-C-peptide function), and in pro- glucagon, proopiomelanocortin (POMC), and many others. Therefore, we speculate that the proteolytic converting enzymes have diverged into a superfamily comprising a range of slightly differing specificities to provide more precise control of tissue specific processing of precursors. Classic examples are the differential processing of POMC in the middle- vs. the anterior pituitary lobes, and that of proglucagon in the pancreatic « cells vs. the intestinal glicentin-producing cells (Fig. 1 ). The relative contribu- tions of different levels of processing enzymes, as op- posed to changes in prosecretory granule pH and cal- cium content, will be important areas for future research when the various processing enzymes can be identified and measured more accurately. In contrast to the endoproteases, which in higher or- ganisms still elude detection, a carboxypeptidase B-like processing enzyme has been identified in brain (Fricker el al, 1986) and islet (Davidson and Hutton, 1987) tis- sues. Its amino acid sequence has been deduced by means of molecular cloning, and the enzyme has been well characterized biochemically (Fricker el al. 1986). Now known as carboxypeptidase H (Skidgel, 1988), it is a homologue of the pancreatic carboxypeptidases A and B and is a soluble constituent of secretory granules, un- like the yeast Kexl carboxypeptidase, which is mem- brane-bound and homologous in part to the yeast vacuo- lar carboxypeptidase Y (Fuller el al. 1988). These differences between the yeast and higher vertebrate pro- cessing carboxypeptidases indicate that evolution has in- troduced significant changes, and suggests that some care should be exercised in extrapolating results from yeast to higher vertebrate processing systems. The insulin gene provides an interesting example of the distribution and diversification of a fundamental 174 D. F. STEINER ET AL. protein structure in the course of evolution. This gene exists in a single copy in most vertebrates and, as men- tioned earlier, encodes a highly conserved preproinsulin molecule (Steiner el a/.. 1985). Indeed, the same high de- gree of structural conservation is again evident when the insulin genes from a number of vertebrates are com- pared. From hagfish, through bony fishes, birds and mammals, this gene has retained two introns in the same relative positions, although these vary considerably in length (Steiner el al, 1985). Moreover, the human insu- lin gene, which resides on the short arm of chromosome 1 1 (in man), is flanked by tyrosine hydroxylase on the 5' side (O'Malley and Rotwein, 1988), and by the insulin like growth factor II (IGF II) gene on its 3' side (Bell et al., 1985). However, both of these flanking genes are reg- ulated differently in terms of their tissue specificity and transcriptional control. The insulin gene is expressed only in the (3 cells of the pancreas in most vertebrates that have been examined and is regulated by glucose and cyclic AMP (Steiner et al.. 1986). The tyrosine hydroxy- lase gene is transiently expressed early in the develop- ment of the /3 cells but is repressed at about the time that insulin gene transcription begins (Alpertf/ al., 1988). On the other hand, IGF II, while being a member of the insu- lin superfamily and sharing many of the structural fea- tures and activities of insulin, is expressed predomi- nantly in the fetus and in multiple tissues (Van Wyk et al.. 1984). Other members of the insulin superfamily include IGF I, relaxin, the insect prothoracicotrophic hormone, PTTH (Nagasawa et al., 1986), and the recently discov- ered molluscan insulin-like peptide, MIP (Smit et al., 1988; Ebberink et al.. 1989). These peptides are related to insulin in that they all conserve the characteristic terti- ary "fold." However, it is unlikely that either of the two invertebrate hormones will bind to vertebrate insulin re- ceptors, as they lack important features for binding, such as the conserved residues A21 Asn, B23 Gly, B24 Phe, B25 Phe, and B26 Tyr. Both the Lymnaea and Bombyx hormones lack these and other important groupings sug- gesting that they bind to different receptors. On the other hand, insulin receptors show great evolutionary conser- vation of their binding characteristics throughout the vertebrates (Muggeo et al., 1979). The IGF I receptor is structurally homologous to the insulin receptor (Ullrich et al., 1986), and a homologous receptor that preferen- tially binds insulin has also been found in Drosophila (Nishida et al., 1986; Fernandez-Almonacid and Rosen, 1987). Thus, we suspect that insulin-like molecules more similar to vertebrate insulin must exist in some inverte- brates. According to this view, the insulin-like peptides identified thus far in invertebrates, as well as relaxin and any other insulin-related peptides that may ultimately turn up in vertebrates, are likely the products of genes that branched away from the main insulin gene line very early. What seem to be missing at present are molecular links between these more divergent insulin-like peptides and the vertebrate insulin/IGF families with their dis- tinctive receptor binding features (Steiner and Chan, 1988). The new genetic tools should permit us to analyze the gene lineages of peptide hormones and their precur- sors much more definitively than has hitherto been pos- sible. Literature Cited Alpert, S., D. Hanahan, and G. Teitelman. 1988. Hybrid insulin genes reveal a developmental lineage for pancreatic endocrine cells and imply a relationship with neurons. Cell 53: 295-308. Bell, G. I., D. S. Gerhard, N. M. Fong, R. Sanchez-Pescador, and L. B. Rail. 1985. Isolation of the human insulin-like growth factor genes: the insulin-like growth factor II and insulin genes are contig- uous. Proc. Nail. Acad. Sci. USA 82: 6450-6454. Davidson, H. W., and J. C. Hutton. 1987. The insulin-secretory-gran- ule carboxypeptidase H. Biochem. J. 245: 575-582. Davidson, H. W., C. J. Rhodes, and J. C. Hutton. 1988. Intraor- ganellar calcium and pH control proinsulin cleavage in the pancre- atic (1 cell via two distinct site-specific endopeptidases. Nalure 355: 93-96. Ebberink, R. H. M., A. B. Smit, and J. van Minnen. 1989. The insulin family: evolution of structure and function in vertebrates and inver- tebrates. Biol. Bull 111: 176-182. Falkmer, S., M. El-Salhy, and M. Titlbach. 1984. Evolution of the neuroendocrine system in vertebrates — A review with particular reference to the phylogeny and postnatal maturation of the islet pa- renchyma. Chapter 4, pp. 59-87 in Evolution and Tumour Pathol- ogy oj the Neuroendocrine System, S. Falkmer, R. Hakanson, and F. Sundler, eds., Elsevier, North Holland. Fernandez-Almonacid, R., and O. M. Rosen. 1987. Structure and li- gand specificity of the Drosophila melanogaster insulin receptor. Molecular Cell ''Biol. 7:271 8-2727. Fricker, L. D., C. J. Evans, F. S. Esch, and E. Herbert. 1986. Cloning and sequence analysis of cDNA for bovine carboxypeptidase E. Na- ture 323: 46\ -464. Fuller, R. S., R. E. Sterne, and J. Thorner. 1988. Enzymes required for yeast prohormone processing. Ann. Rev. Physiol 50: 345-362. Mizuno, K., T. Nakamura, T. Ohshima, S. Tanaka, and H. IVlatsuo. 1988. Yeast Ke.\2 gene encodes an endopeptidase homologous to subtilisin-likesenne proteases. Biochem. Biophys. Res Comm 156: 246-254. Muggeo, M., B. H. Ginsberg, J. Roth, D. M. Neville, Jr., P. De Meyts, and C. R. Kahn. 1979. The insulin receptor in vertebrates is func- tionally more conserved during evolution than insulin itself. Endo- crinology 104: 1 393- 1 402. Nagasawa, H., and II. Kataoka, A. Isogai, S. Tamura, A. Suzuki, A. Mizoguchi, Y. Fujiwara, A. Suzuki, S. Y. Takahashi, and H. Ishi- zaki. 1986. Amino acid sequence of a prothoracicotropic hor- mone of the silk worm Bombyx mori. Proc. Null. Acad. Sci. USA 83: 5840-5843. Nishida, Y., M. Hata, Y. Nishizuka, W. J. Rutter, and Y. Ebina. 1 986. Cloning of a Drosophila cDN A encoding a polypeptide sim- ilar to the human insulin receptor precursor. Biochem. Biophys. Res Comm 141:474-481. O'Malley, K. L., and P. Rotwein. 1988. Human tyrosine hydroxylase ISLET PEPTIDES 175 and insulin genes are contiguous on chromosome I 1 . Nucleic Acid Res. 16: 4437-4446. Skidgel, R. A. 1988. Basic carboxypeptidases: regulators ol" peptide hormone activity. Trends in Pharmacol. Sci. 9: 299-304. Smit, A. B., E. Vreugdenhil, R. H. M. Ebberink, W. P. M. Geraerts, J. Klootsijk, and J. Joosse. 1988. Growth-controlling molluscan neurons produce the precursor of an insulin-related peptide. Nature 331:535-538. Steiner, D. F. 1984. The biosynthesis of insulin: genetic, evolutionary and pathophysiologic aspects. Pp. 1 9 1 -228 in The Harvey Lectures. Series 78, Academic Press. New York. Steiner, D. F., and S. J. Chan. 1988. Perspective: an overview of insu- lin evolution. Hormone Metabci Res. 20: 443-444. Steiner, D. F., S. J. Chan, J. M. Welsh, D. Nielsen, J. Michael, H. S. Tager, and A. H. Rubenstein. 1986. Models of peptide biosynthe- sis— the molecular and cellular basis of insulin production. Clin. Invest. Med. 9: 328-336. Steiner, D. F., S. J. Chan, J. M. Welsh, and S. C. M. Kwok. 1985. Structure and evolution of the insulin gene. Ann. Rev. Genet. 19: 463-484. Thomas, G., B. A. Thorne, L. Thomas, R. G. Allen, D. E. Hruby, R. Fuller, and J. Thorner. 1988. Yeast KEX2 endopeptidase cor- rectly cleaves a neuroendocrine prohormone in mammalian cells. Science 241: 226-230. Ullrich, A., A. Gray, A. W. Tarn, T. Yang-Feng, M. Tsubokawa, C. Collins, W. Henzel, T. Le Bon, S. Kathuria, E. Chen, S. Jacobs, U. Francke, J. Ramachandran, and II. Fujita-Yamaguchi. 1986. In- sulin-like growth factor I receptor primary structure: comparison with insulin receptor suggests structural determinants that define functional specificity. EMBOJ 5: 2503-2512. Van Wyk, J. J. 1984. The somatomedins: biological actions and physi- ologic control mechanisms. Pp. 8 1 - 1 25 in Hormonal Proteins and Peptides. Academic Press, NY. Reference: Bial. Bull 177: 176-182. (October, 1989) The Insulin Family: Evolution of Structure and Function in Vertebrates and Invertebrates R. H. M. EBBERINK, A. B. SMIT. AND J. VAN MINNEN Biological Laboratory, Vrije Universiteit, de Boelelaan 1087, NL-1081 H\' Amsterdam. The Netherlands Abstract. Insulin and related peptides are key hor- monal integrators of growth and metabolism in verte- brates. Recently, the amino acid and DNA sequences of insulin-related peptides in invertebrates have become available. The discovery of such peptides in insects and molluscs substantiates the evidence for an early origin and widespread evolution of the insulin superfamily. In the silkworm Bombyx (Insecta) the prothoracico- tropic hormones (bombyxins 1, II, and III; previously called PTTH) are produced in the brain and may stimu- late synthesis and release of ecdysone; thus they play a central role in insect development. In the freshwater snail Lymnaea (Mollusca), a growth stimulating hor- mone (molluscan insulin-related peptide; MIP) is pro- duced in the brain, and two other insulin-related pep- tides are produced in the digestive system. The MIPs are involved in body and shell growth and energy metabo- lism. The finding that bombyxin and MIP are involved in the control of growth fits with ideas being developed in the vertebrate field that the role of insulin is not con- fined to glucose metabolism, but is also related to growth. Introduction The structure of the insulin molecule has been highly conserved during vertebrate evolution (Chance el a/.. 1968; Blundell et at.. 1972; Cutfield et al, 1979, 1986; Chan et al.. 1981; Bajaj et al., 1983; Emdin et al.. 1985; Le Roith et al., 1987; Pollock et al.. 1987). At the mo- ment, the primary structures of insulins from over 40 vertebrate species are known. In addition, the preproin- sulin genes and cDNA sequences from over 12 species have been determined (Lomedica et al.. 1979; Hahn et al.. 1983;Steinert>/<7/., 1985). Therefore, for vertebrates, the structures of preproinsulins and the insulin-like growth factors could be integrated to produce an evolu- tionary picture of this hormone superfamily. We will not discuss the evolution of vertebrate insulins; this has al- ready been done several times (e.g., Steiner et al., 1985). Rather, we will consider the varied evidence, obtained during the last few years, that modern invertebrates, and in particular the insects and the molluscs, also contain insulin-related peptides. Furthermore, we will inquire about the nature of the insulin molecule and compare its functions in insects and molluscs with those in verte- brates. When considering the evolutionary aspects of insulins in the animal kingdom, we should keep in mind that the various phyla have had a polyphyletic origin; i.e.. four major groups — the chordates and vertebrates, the echi- noderms and tentaculates, the coelenterates, and the molluscs, worms, and arthropods — are now considered to have evolved independently of each other (Fig. 1). An important consequence of this polyphyletic origin is that if insulin occurs in the two main branches of the phyloge- netic tree, then it must have already been present in the Archaemetazoa. Evidence for Insulins in Invertebrates Because invertebrates as well as vertebrates rely upon the same organic molecules for metabolism, both groups should, in theory, possess insulin. The experimental evi- dence in support of this notion comes from two different approaches: immunocytochemistry and biochemistry. With immunocytochemistry, rapid strides have been made in the identification of invertebrate cells and tis- sues that are reactive to anti-insulin. Most of the observa- tions have been carried out with antisera raised to mam- malian insulin, and positive results have been obtained 176 THE INSULIN FAMILY 177 PROTOSTOMIA DEUTEROSTOMIA Osteichihyes Chondnchlhyes Cycloslomala Archaecoelomala Tenia- i • Echmo- culaia [ * dermata Table I Identification of insulin-like peptides in invertebrates by immunocytochemistry Figure 1. A phylogenetic tree, showing the polyphyletic origin of the various phyla in the animal kingdom. (Modified and extended after Karlson, 1983). in a range of different species, primarily insects and mol- luscs (Table I). In molluscs, immunoreactivity occurs not only in neuronal tissue, but also in the epithelia of the gut and hepatopancreas. Of course, there are problems and pitfalls in immuno- cytochemistry. The epitope for anti-insulin deduced from structure-activity analyses, is formed by the region including residues 8, 9, and 10 of the A-chain, and resi- dues 2, 3, and 4 of the B-chain of insulin. The ability of invertebrate tissues to bind anti-mammalian insulin is surprising because non-mammalian insulins, insulin- like growth factors, and relaxin are variable in this region and, therefore, do not bind antibodies to mammalian in- sulin. However, the neuroendocrine light green cells in the cerebral ganglia of the central nervous system of the freshwater snail have been identified as anti-porcine in- sulin immunopositive cells (Fig. 2). Indeed, these cells produce an insulin-related peptide with a different epi- tope region (see below). The second approach in the identification of insulin- related peptides is biochemistry: extraction, purification, and chemical characterization. Several early reports of insulin-like substances in invertebrates relied upon rather simple or even crude tissue extraction procedures followed by heterologous bioassays (Table II). Later on, RIA and purification studies were performed. Studies on the blowfly, Calliphora vomitoria. by Thorpe and Duve Insecta Calliphora \omiloria Median neuro- Duve and Thorpe, 1979 secretory cells Bombyx mori Median neuro- \\i\etal.. 1980 secretory cells Locust migratoria Median neuro- Orchard and Loughton. secretory cells 1980 Manduca sexta Median neuro- E\-Sa\hy et al. 1984 secretory cells Eristalis aeneus Pars intercerebralis El-Salhytf a/.. 1980 Apis milliter a Brain Maierera/.. 1981 Mollusca Lymnaea stagnalis Small cells in Schotrta/.. 1981 cerebral ganglia Light green cells van Minnen., 1987 Anodonta cygnea Midgut Plisetskaya. 1978 Unio pictorum Midgut Plisetskaya, 1978 Mylilus edit/is Hepatopancreas FritchetaL, 1976 Tunicata Steyla clava Endocrine cells Bevis and Thorndyke, esophagus 1978 * For references see literature cited in Joosse and Geraerts ( 1 983) and Thorpe and Duve (1984). (1984) resulted in the purification and amino acid com- position of anti-insulin immunoreactive material, al- though no amino acid sequence analysis was done. Our recent studies on the snail Lymnaea stagnalis have re- sulted in the structural analysis of an insulin-related mol- ecule from the CNS and the purification of two insulin- like substances from the midgut. We have cleaved and Figure 2. Transverse section through the cerebral ganglia of Lym- naea stagnalis. The light green cells ( LGC) in the cerebral ganglia (CG) and the canopy cell (CC) in the lateral lobe are labeled by anti-porcine insulin. DB, dorsal body; Com, commissure. 178 R. H. M. EBBERINK. ET AL. Table II Biochemical characterization of insulin-like peptides in invertebrates Insecta Apis mellifera Extract/bioassay Patel, 1964 Drosophila melanogaster Extract/bioassay Meness and Ortiz, 1975 Manduca sexta Exract/RIA/ Kramer et al., 1977 bioassay Drosophila melanogaster Heamolymf/RIA Seecofand Dewhurst. 1974 Manduca sexta GPC/RIA Tagerrta/.. 1975, 1976 Manduca sexta Ammo acid Kramer, 1984 composition Calliphora vomitoria GPC/IEG/HPLC Duvet'/ al.. 1979, 1982 Amino acid Thorpe and Duve, composition 1984 Drosophila melanogaster GPC/RIA Le Roith el al., 1981 Bombix mori HPLCetc. and Nagasawa rt a/., sequence 1984, 1986 Crustacea Homarus americanus Mollusca RIA/bioassay Sanders. 1983 Mya arenana extract/bioassay Collip, 1923 Buccmum imdatum extract/bioassay Davidson, 1971 Pectiim maximus extract/bioassay Davidson, 1971 Ostrea edulis extract/bioassay Martinez et al.. 1973 Unio pectorum IEC/RIA Plisetskaya, 1978 Prophysaon foliolatum heamolymph/RIA Plisetskaya and Deymp, 1987 Lymnaea stagnalis GPC/HPLC gut- Hemminga, 1984 insulin Ebberinkand Joosse, 1985 Lymnaea stagnalis HPLC brain-insulin Ebbennk et al.. 1987 Lymnaea stagnalis cDNA brain insulin Smite/ at.. 1988 Tunicata Pyura pachydermatina HPLC/R1A Galloway and Cuttield, 1988 * For references see literature cited in Joosse and Geraerts ( 1 983) and Thorpe and Duve (1984). separated the A and B chains of the midgut insulin-like substance (Fig. 3), and have determined the amino acid composition of each chain (Table III), but efforts to se- quence this material using a protein sequencer have failed since both chains have a blocked N-terminal. Structure and Function of Insulin-Related Peptides in Invertebrates The first amino acid sequence information about an insulin-related structure in invertebrates came from the pioneer work of Nagasawa and his colleagues (Nagasawa et al., 1984, 1986) on the prothoracicotrophic hormone (PTTH, now called bombyxin) of the silkworm Bombyx mori. Bombyxin is produced by the median neurosecre- tory cells of the pars intercerebrale of the brain, and con- trols the secretion of ecdysone from the prothoracic glands during metamorphosis (Ishizaki et al., 1987). Bombyxin was not previously suspected of having any relationship to insulin. The similarity emerged only after 25 years, during which Nagasawa et al. purified the pep- tide from several million heads of Bombyx using a 14- step procedure. Bombyxin II consists of two non-identical chains: the A-chain of 20 residues, and the B-chain of 28 residues (Fig. 4). Besides bombyxin II, two other peptides have been purified. Only the N-terminals of the A-chains of bombyxins I and III have been sequenced, and both have an 80% homology with bombyxin II. Four different forms of bombyxin-II have been published (Fig. 4). Insulin-related peptides of Lymnaea are not only pro- duced in the gut, but also in the neuroendocrine light green cells (LGCs). There are about 200 LGCs located in two pairs of clusters in the cerebral ganglia of the central nervous system of this snail (Fig. 2). The LGC are in- volved in body and shell growth (Geraerts, 1976; Joosse and Geraerts, 1983; Ebberink and Joosse, 1985). The effects on shell formation include: ( 1 ) calcium and bicar- 0.25 | 10 20 30 40 50 60 70 time (min) Figure 3. Reverse phase liquid chromatography of purified gut in- sulin of Lymnaea stagnalis after reduction and carboxymethylation. A mixture of two insulin-related peptides was reduced, and the A and B chains were separated on a Bakerbond wide pore C18 column with a gradient of acetonitrile (5-45% in 70 min) in 7.5 mA/ trifluoroacetic acid. The peaks at 5 1 min are the intact insulins. THE INSULIN FAMILY 179 Table III Amino acid composition of the A- and B-cham of gut insulin of Lymnaea and comparison with brain insulin (MIP) <)/' Lymnaea and human insulin A-chain B-chain Lymnaea Lymnaea gut insulin Lymnaea Human gut insulin Lymnaea Human I II brain insulin insulin I 11 brain insulin insulin Asx 2 I 1 2 3 3 6 I Glx 2 I 4 4 2 2 2 3 Cys -4 \ 5 4 ~2 ~2 3 2 His — — 3 2 1 2 Ser 2 1 2 2 3 5 1 Arg 1 1 — 1 1 3 1 Gly 1 1 1 2 3 2 3 Thr 1 3 1 — — — 2 Ala 2 — 5 6 5 1 Tyr 1 1 2 1 1 — 2 Val 1 1 1 2 3 3 3 Phe — — — 2 3 He 1 1 2 2 2 1 Leu 2 : ! 2 2 4 4 2 4 Lys i 1 — 2 2 — 1 Met nd ni 1 1 — nd nd 1 — Pro nd IK 1 2 — nd nd 2 1 Tip nd nc 1 — — nd nd — — bonate incorporation into the shell; (2) formation of the periostracum (the proteinaceous component); and (3) the maintenance of high concentrations of calcium- binding protein in the cells of the mantle edge. The effects on the soft body parts are: ( 1 ) stimulation of the ornithine decarboxylase activity; (2) mobilization of gly- cogen stores; (3) regulation of the blood glucose concen- tration (Fig. 5); and (4) neurite outgrowth. The first evidence that the LGC may contain an insu- lin-related peptide came from immunoytochemical data (Fig. 2). To prove that the anti-insulin immunoreactive material is a hormone, the LGCs together with the me- dian lip nerve (the neurohemal area of the LGC), were incubated (/'/; vitro) with and without 4-aminopyridine (Fig. 6). After the addition of 4-aminopyridine, the LGC show a strong increase in the number of action poten- tials, and they release immunoreactive insulin which reaches a maximum level within one hour. In the ab- sence of 4-aminopyridine, only a small amount of im- munoreactive insulin was released. A-chaln. 15 10 15 20 Gly-Ile-Val-Asp-Glu-Cys-Cys-Leu-Arg-Pro-Cys-Ser-Val-Asp-Val-Leu-Leu-Ser-Tyr-Cys. B-chain. 15 10 15 20 Glp-Gln-Pro-Gln-Ala-Val-His-Thr-Tyr-CysH31y-Arg-His-Leu-Ala-Arg-Thr-Leu-Ala-Asp- 21 25 28 Leu-Cys-Trp-Glu-Ala-Gly-Val-Asp N-terminal region of B-chain of four different forms of Bombyxin-II Glp-Gln-Pro-Gln-Al»-val Glp-Gln-Pro-Gln-Gly-Val Glp Gln-Ala-Val Glp Gln-Gly-Val Figure 4. Amino acid sequence of the A and B chains of bombyxin-II (PTTH-I1) of Bombyxmori. 180 R. H. M. EBBERINK ET AL. •a 3 30- 20 - I control brain insulin gut insulin bovine insulin Lymnaea Lymnaea Figure 5. The effect of different insulins on the hemolymph glucose concentration of Lymnaea stagnalis. Purified insulins from the light green cells and gut. as well as commercial bovine insulin, were tested. The amount of brain insulin injected in each snail is about the amount stored in one animal (about 2 pmol); for gut insulin, it is the amount stored in about 0.2 animal (about 0.5 pmol); and for bovine insulin. 200 pmol. The blood volume is about I ml. (n = 4). The primary structure of the insulin-related peptide was not obtained via peptide chemistry, but via the nu- cleotide sequence of an LGC specific cDNA clone (Smit et al, 1988). A differential hybridization technique was used to isolate cerebral LGC cDNA from a central ner- ity of LGC in vil DB LGC * Com • CG DB LGC Figure 7. //; .S-//H hybridization with a 35S-labelled cDNA-probe in sections of the cerebral ganglia. Only hybridization of mRNA in the light green cells (LGC) and canopy cell (not shown) was observed. DB, dorsal body; CG, cerebral ganglia; Com, commissure. vous system-specific library ofLymnaea cloned in Xgt 1 0. Therefore, replica filters of 20,000 clones were screened with a positive cDNA probe synthesized from messenger RNA of the LGC, and with a negative cDNA probe pro- duced from other parts of the cerebral ganglia and the hepatopancreas. The LGC specificity of the clones was tested by in situ hybridization using histological sections of the central nervous system (Fig. 7). The LGCs in the cerebral ganglia, and the canopy cell in each lateral lobe (not shown), were the only cell types to express this clone. The nucleotide sequence revealed a single open read- ing frame encoding a protein with characteristics of pre- proinsulin (Fig. 8) (Smit et ai. 1988). Thus, an A and B chain, together with a C peptide equivalent and a puta- tive signal sequence, are present. We called this peptide molluscan insulin-related peptide (MIP). Comparison of Invertebrate and Vertebrate Insulins Overall, the amino acid sequences of MIP, bombyxin, and human insulin, are not very similar (Fig. 9). The se- 1.0! 0.8 Figure 6. The effect of 4-aminopyridine on the electrical activity of the light green cells (LGC) (top panel). The release of anti-porcine immunoreactive material from the light green cells (bottom panel). Ce- rebral ganglia (CG) with the median lip nerves were incubated as de- scribed previously, with and without 4-aminopyridine (Ebberink et ai, 1987). Figure 8. Amino acid sequence of prepro molluscan insulin-related peptide ( preproM I P). Residues are designated by their one-letter abbre- viations. The putative proteolytic processing sites are indicated (lines between some residues). THE INSULIN FAMILY 181 A chain PTTH-II Insulin MIP-1 10 20 ©®(Y)© ^^ B-cham PTTH-II Insulin MIP-I ©©CMXSXy)®©^^ (yXS)©(iM3(eX§>(s)(^^ ©(H)®®^)^^ Figure 9. Comparison of the amino acid sequences of the A and B chains of bombyxin (PTTH) of Bombyx mori, human insulin, and molluscan insulin-related peptide (MIP) of Lymnaca stagnalis. The amino acids are identified bv their one-letter abbreviations. quence similarity in the A chain is about 50% between bombyxin and human insulin, and 40% between MIP and human insulin. In the B chain, the similarity is 25% and 20%., respectively. An amino acid sequence compar- ison of MIP and bombyxin with known insulins of all vertebrate species permits some interesting conclusions. All of these peptides have cysteines present at position A6,A7,A11,A20,B7, and B19. They have glycine at Al (except MIP) and at B8, but glycine at B20 is only present in the vertebrate insulins. Most of the hydrophobic resi- dues at the hydrophobic core of the globular structure of insulin are conserved in all insulins. A three-dimensional model of bombyxin and MIP has been constructed with interactive computer graphics and energy minimization techniques (Blundell el al, 1972); homology with por- cine insulin (the structure of which has been determined by X-ray analysis) was assumed. The model shows that both bombyxin and MIP form neither the dimers nor the hexamers characteristic of mammalian insulins. The region formed by residues B9-B19 and B22-B26 is in- volved in the binding of insulin to its receptor (Pullen el al., 1976), and the phenylalanine in position B25 is particularly essential (Blundell and Wood, 1975). Since this region, including B25, differs from that of mamma- lian insulins, it has been suggested that MIP and bom- byxin cannot bind to the vertebrate insulin receptor. However, gut insulin ofLynmaea binds very well to insulin receptors of rat fat cells (Ebberink and Joosse, 1985). Conclusions It is important that the structures of molluscan and insect insulins have been found at last, more than 30 years after the discovery of the first vertebrate insulin structure. The structure of the insulin molecule is largely deter- mined by a characteristic arrangement of certain resi- dues in its precursor. That similar arrangements of amino acids seen in the various insulins would have arisen independently in different branches of the phylo- genetic tree (Fig. 1 ) is extremely improbable. We cannot deduce the evolutionary pathway for insulin on the basis of nucleotide differences, since these arise predomi- nantly from neutral mutations. Rather, the finding of in- sulins in two branches of the phylogenetic tree confirms the model of Blundell and Wood (1975), according to which the evolution of insulins is determined mainly by adaptive processes. The model depends critically on the relationship between such factors as the effects of se- quence changes on the three-dimensional structure of the peptide, and the role of various parts of this structure in the conversion of the proinsulin to the active form, the storage of insulin, its transport to the site of action, and its interaction with the receptor. According to this hy- pothesis, the primary and three dimensional structures conserved in vertebrate insulins must also be conserved in the related peptides of insects and molluscs. Since MIP, the vertebrate insulins and possibly bombyxin are involved in growth, it is important to discover whether the insulin receptors of invertebrates are homologous with those of vertebrates. According to currently accepted theory, the origin of insulin is to be found in the nervous systems of early multicellular organisms (Pearse, 1967;Pictett>/a/., 1976; Alpert ?/«/., 1988). Indeed, the localization of an inverte- brate insulin within specific neurons of the brain would seem to support such a notion. Until now, the only insu- lin sequences available were from the central nervous systems of invertebrates and from the pancreas of verte- brates— data that do not test the theory. But we have shown that insulins are present in the gut of molluscs, as well as in the brain. This finding suggests that, contrary to dogma, the insulins might have originated in the brain or in the gut, and possibly also in other tissues of early metazoans. Further, these considerations could imply that the brain of vertebrates also produces insulins. How- ever, at present, the question of the synthesis of insulin in the vertebrate brain is still arguable (Baskin el al., 1987; LeRoitht'/tf/., 1987). 182 R. H. M. EBBERINK ET AL Literature Cited Alpert, S., D. Hanahan, and G. Teitelman. 1988. Hybrid insulin genes reveal a developmental lineage for pancreatic endocrine cells and imply a relationship with neurons. CV//53: 295-308. Bajaj, M., T. L. Blundell, J. E. Pitts, S. P. Wood, and M. A. Tatnell. 1983. Dogfish insulin. Eur. J Biochem. 135: 535-542. Baskin, D. G., D. P. Figlewisz, S. C. Woods, D. Porte, and D. M. Dorsa. 1987. Insulin in the brain. Ann Rev Physiol 49: 335-347. Blundell, T. L., J. F. Cutfield, S. M. Oilfield, E. J. Dodson, G. G. Hodgkin, and D. A. Mercola. 1972. Three-dimensional atomic structure of insulin and its relationship to activity. Diabetes 21, suppl 2: 492-505. Blundell, T. L., and S. P. Wood. 1975. Is the evolution of insulin Dar- winian or due to selectively neutral mutation? Nature 257: 197-203. Chan, S. J., S. O. Emdin, S. C. M. Kwok, J. M. Kramer, S. Falkmar, and D. F. Steiner. 1981. Messenger RNA sequence and primary structure of preproinsulin in a primitive vertebrate, the Atlantic hagfish. / Biol. Chem. 256: 7595-7602. Chance, R. E., R. M. Ellis, and W. W. Bromer. 1968. Porcine proinsu- lin: Characterization and aminoacid sequence. Science 161: 165-167. Cutfield, J. F., S. M. Cutfield, E. J. Dodson, G. G. Dodson, S. F. Emdin, and C. D. Reynolds. 1 979. Structure and biological activity of hag- fish insulin. J. Mol. Biol 132: 85-100. Cutfield, J. F.. S. M. Cutfield, A. Carne, S. O. Emdin, and S. Falkmar. 1986. The isolation, purification and amino acid sequence of in- sulin from the teleost fish Coitus corpius (daddy sculpin). Eur. J Biofhem. 158: 117-123. Ebberink, R. H. M., and J. Joosse. 1985. Molecular properties of vari- ous snail peptidesfrom brain and gut. Peptides 6, suppl 3: 45 1-457. Ebberink, R. H. M., H. van Loenhout, J. van Beek, K. de Wilde, and J. van Minnen. 1987. Characterization of peptides isolated from growth-controlling neuroendocnne cells ofLymnaea stagnalis with immunoreactivity to anti-insulin. Pp. 224-227 in Neurobiology, Molluscan Models. H. H. Boer. W. P. M. Geraerts. and J. Joosse, eds. North Holland Publishing Company, Amsterdam. Emdin, S. O., D. F. Steiner, S. J. Chan, and S. Falkmar. 1985. Hagfish insulin: evolution of insulin. Pp. 134 in Evolution- ary Biology of Primitive Fishes, R. E. Foreman, A. Gorban. J. M. Dodd, and R. Olsson, eds. Plenum, New York. Galloway, S. M., and J. F. Cutfield. 1988. Insulin-like material from the digestive tract of the tunicate Pyura pachydermatina (sea tulip). Gen Comp. Endocrinol. 69: 106-1 13. Geraerts, W. P. M. 1976. Control of growth by the neurosecretory hormone of the light green cells of the freshwater snail Lymnaea stagnalis. Gen. Comp. Endocrinol. 29: 6 1 -7 1 . Hahn, V., J. Winkler, T. A. Rapoport, D. H. Liebscher, Ch. Coutelle, and S. Rosenthal. 1983. Carp preproinsulin cDNA sequence and evolution of genes. Nucleic Acid Res, 11:4541-4552. Ishizaki, H., A. Suzuki, and Y. Suzuki. 1987. Prothoracicotropic hor- mone and functionally related peptides of the Bombyx mori: an overview of our studies. Pp. 55-56 in Proc. Jpn. Soc. Comp. Endo- crinol 2. E. Ohnishini. Y., Nagahama., H., Ishizaki, eds. Nagoya University Corporation. Joosse, J., and W. P. M. Geraerts. 1983. Endocrinology. Pp. 317- 406 in The Mollusca. Vol. 4. Part 1. Physiology. A. S. M. Saludin and K. M. Wilbur, eds. Academic Press, New York. LeRoith, D., W. L. Lowe, and C. T. Roberts. 1987. Evolution of insulin and insulin receptors. Pp. 1 56 in Insulin. Insulin-Like Growth Factors, and their Receptors in the Central Nervous System. M. K. Raizada. M. I. Philips and D. LeRoith, eds. Plenum Press, New York. Lomedico, P., N. Rosenthal, A. Erstratiadis, W. Gilbert, R. Kolodner, and R. Tizard. 1979. The structure and evolution of the two non- allelic rat preproinsulin genes. Cell 18: 545-558. Nagasawa, H., H. Kataoka, A. Isogai, S. Suzuki, A. Suzuki, H. Ishi- zaki, A. Mizoguchi, Y. Fujiwara, and At. Susuki. 1984. Ammo- terminal amino acid sequence of the silkworm prothoracicotropic hormone: homology with insulin. Science 226: 1344-1345. Nagasawa, H., H. Kataoka, A. Isogai, S. Suzuki, A. Suzuki, H. Mizo- guchi, Y. Fujiwara, At. Susuki, S. Y. Takahashi, and H. Ishizaki. 1986. Amino acid sequence of the prothoracicotropic hormone of the silkworm Bomby\mon. Proc. Nat. Acad. Sci. USA 83: 5840-5843. Pearse, A. G. E. 1967. Peptides in brain and intestine. Nature 262: 92-94. Pictet, R. L., L. B. Rail, P. Phelps, and W . J. Rutter. 1976. The neu- ral crest and the origin of the insulin-producing and other gastroin- testinal hormone-producing cells. Science 191: 191-192. Pollock, H. G., J. R. Kimmel, J. W. Hamilton, J. B. Rouse, K. E. Ebner, V. Lance, and A. B. Rawitch. 1987. Isolation and struc- tures of alligator gar (Lepisosteit.s spatula} insulin and pancreatic polypeptide. Gen Comp. Endocrtnol. 67: 375-382. Pullen, R. A., G. Lindsay, W. P. W ood, I. J. Tickle, and T. L. Blundell. 1976. Receptor-binding region of insulin. Nature 259: 369-373. Smit, A. B., E. Vreugdenhil, R. H. M. Ebberink, W. P. M. Geraerts, K. Klootwijk, and J. Joosse. 1988. Growth-controlling molluscan neurons produce the precursor of an insulin-related peptide. Nature 331:535-538. Steiner, D. F., S. J. Chan, J. M. Welsh, and S. C. M. Kwok. 1985. Structure and evolution of the insulin gene. Ann Rev Genet. 19: 463-484. Thorpe, A., and H. Duve. 1984. Insulin- and glucagon-like peptides in insects and molluscs. Mol. Physio/. 5: 235-260. Reference: Biol. Bull. Ill: 183-186. (October. 1989) Vasoactive Intestinal Polypeptide and its Relatives: Biochemistry, Distribution, and Functions MICHAEL C. THORNDYKE1, JENNIFER H. RIDDELL1, DAVID T. THWAITES2, AND RODNEY DIMALINE2 'Department of Biology. RHBNC, London University, Egham, Surrey, TW20OEX, U. A'., and 2 Physiological Laboratory, University of Liverpool, Liverpool, L69 3BX U. K. Introduction Vasoactive intestinal polypeptide (VIP) was first iso- lated from the small intestine of the pig. the purification being monitored with a bioassay based on the vasoactive properties of the material (Said and Mutt, 1972). The new peptide had been anticipated, for in their pioneering experiments leading to the first discovery of the hormone secretin. Bayliss and Starling (1902) had noted the pres- ence, in intestinal extracts, of a factor with vasodepressor activity. This may be the first recorded evidence of VIP activity. Structures Since its isolation as an octacosapeptide, VIP has been recognized as having a highly conserved sequence, not only in mammals, but throughout the vertebrates (Di- maline, 1989). The peptides from pig, man, dog, cow, and rat are identical, while the VIPs from guinea pig and chicken have only four substitutions, and dogfish and cod have five (Dimaline et a/., 1987; Thwaites el ai, 1988) (Fig. 1). Elucidation of VIP sequences also imme- diately revealed their structural similarities with secretin and other hormones, such as glucagon and glucose-de- pendent insulinotropic peptide (GIP). The subsequent characterization of growth hormone releasing factor (GRF) and a series of peptides isolated from reptilian venoms (helodermin and the helospectins) — all of which share sequence relationships with VIP — makes this as- semblage one of the largest and most diverse structurally related peptide families known. Recent structure-activity studies show that fish VIPs are equipotent to their mammalian counterparts in bind- ing to VIP receptors on the guinea pig acinar cell or aci- nar cell membrane (Dimaline et ai. 1987; Dimaline et al. . 1 988 ). This suggests that, while the C-terminal substi- tutions plainly influence antibody recognition sites (Di- maline et al., 1986), they have little effect on biological activity. Therefore, the biologically active sites of VIP have been highly conserved during evolution. In 1981 a peptide was isolated from pig intestine by Tatemoto and Mutt ( 1 98 1 ) and given the notation PHI, based on its N- and C-terminal amino acids (histidine and isoleucine, respectively). The similarity between PHI and VIP was immediately noticed (Fig. 1), but not until the gene-sequence for the human VIP precursor had been determined was it apparent that both VIP and PHI are encoded within the same precursor (Itoh et al., 1983). This finding strongly supports the importance of internal duplication of bioactive regions in neurohormo- nal peptide gene evolution. Localization In mammals, VIP is distributed widely throughout the central and peripheral nervous systems, and is especially concentrated in the innervation of the gastrointestinal tract. In birds, too, it is probably exclusively a neurally secreted peptide, with the rare accounts of an endocrine location being due (as in mammals) to non-specific cross-reactivity with N-terminally directed antisera. The localization of VIP in the remaining vertebrates has al- ways been controversial. Thus, although there are re- ports of both neuronal and endocrine localizations throughout the lower vertebrate groups, other studies de- scribe the sites as being of either one sort or the other (Falkmer et al., 1980; Holmgren and Nilsson, 1983; El- 183 184 M C. THORNDYKE ET AL. 5 10 15 20 25 COD HSDAVFTDNYSRFRKQMAAKKYLNSVLA* DOGFISH ------------I-----V---I--L--* CHICKEN ------------------V--------T* PIG ----------T-L-----V------I-N* pPHI -A-VG--SDF--LLG-LS-----E-LI* Figure 1. Amino acid sequences of some members of the VIP fam- ily of peptides. The full sequence is shown for cod VIP, and the hyphens indicate an identical residue in the other peptides. p = porcine; * = amide. Sahly, 1984). In part, these discrepancies may be due to genuine species variation, but our studies on the charac- terization of elasmobranch and teleost VIPs suggest that they may result from the use of inappropriate antisera and conditions of fixation (Dimaline and Thorndyke, 1986: Dimaline el al., 1987). However, the duplication within the precursor in mammals suggests that the differences in localization might also reflect differential expression or processing of the precursors. Further stud- ies of the gene sequences and precursors in lower verte- brates may provide answers to such problems. Actions VIP in vertebrates has two broad effects: (i) smooth muscle relaxation, and (ii) stimulation of gastrointestinal fluid and electrolyte secretion. These are fundamental activities and clearly provide for a range of actions, the extent of which is dependent upon the number of sites at which control is exerted. Indeed, VIP illustrates as well as any other peptide the idea that the same molecule may have a variety of roles, each defined according to the pre- cision of delivery and action. These two effects of VIP— secretion and smooth mus- cle relaxation — also illustrate the complementarity of such actions; thus, enhanced secretion can derive from a direct effect on epithelial secretory cells, while an aug- mented local blood flow (vasodilation) provides the met- abolic resources for the cellular secretory activity. Ele- gant experiments have demonstrated such a dual effect of VIP in the gut of the cat; /.<:>., a stimulation of intestinal and colonic blood flow accompanied by an increase in net fluid transport (Eklund el al., 1979). Similarly in sali- vary glands, VIP dilates the submandibular arteries (thereby increasing blood flow to the glands) and, at the same time, directly enhances cellular fluid and protein secretion (Reid and Heywood, 1988). Secretin, the founder member of the VIP family, con- trols pancreatic fluid and electrolyte (bicarbonate) secre- tion in mammals. Indeed, this effect served in the origi- nal identification of secretin by Bayliss and Starling (1902). As might have been expected from the similarity in their sequences, VIP is an effective agonist for pancre- atic secretion in many mammals although, apart from pig, it is considerably less potent than secretin. This ac- tivity relationship is reversed in birds; i.e., VIP is a potent stimulant of avian pancratic secretion, whereas secretin is only a weak one (Dimaline and Dockray, 1979). The proven secretory effects of VIP in mammals have provided the basis for investigating the role of this pep- tide in non-mammalian species. In fishes, much atten- tion has been focussed on water and electrolyte secretion in the gut, and VIP has clearly been implicated in the regulation of these parameters in teleosts (Foskett el al, 1982). However, contention has surrounded the actions of VIP in elasmobranchs. Much has been made of the pro- posed role of VIP as a stimulator of rectal gland activity, an attractive idea because it foreshadows the function of the peptide in enhancing the secretory activity of the mammalian gut. As noted above, however, glandular se- cretion comprises two components: ion transport at the epithelial secretory cell, and vascularization of the gland. In the spiny dogfish Sqitahts, volume expansion stimu- lates both flow and chloride secretion from the duct of the rectal gland (Solomon el al., 1984). However, the sole evidence favoring VIP as a direct agonist of epithelial cell chloride secretion in this species appears to be the ability of somatostatin to inhibit the phenomenon (Solomon et al., 1984). VIP certainly induces vasodilation in Squahis, and a pilot experiment to test partially purified dogfish VIP (C. Woods, T. J. Shuttleworth, and M. C. Thorndyke. un- pub. obs.) showed a potent stimulation of rectal gland flow in vivo (Fig. 2). At the same time, studies on rectal gland slices from Scyliorhinus and Raja show that VIP lacks a direct effect on the secretory epithelium; rather a second candidate peptide — rectin — appears to be a po- tent secretagogue (Shuttleworth and Thorndyke, 1984). Perhaps we are seeing here a differential dual control of rectal gland secretion: by way of the vascular supply (VIP); and by direct action on the secretory epithelium (rectin). Confirmation of the dual control hypothesis awaits the results of tests with rectin on rectal gland slices from Squalus and in vivo experiments on Scyliorhinus. This is important, for we need to rule out differences in response due to differences among dogfish species. In this respect, recent work on the duck salt gland shows that VIP has a potent effect on the blood supply, but that this same peptide also directly stimulates the epithelial cells (Gerstberger, 1988). The vasodilatory effects of VIP are a reflection of its potency in smooth muscle relaxation. Indeed, one of the best recorded effects of VIP in mammals is the relaxation of the smooth muscle of the gut, which is thought by VIP AND RELATED PEPTIDES 185 3- 2- 0J 75ulE VIPI10 M) »925;H 500M NaCl 1 ml 500 mM NaCl I 2 3 Hme h) Figure 2. Effect of elasmobranch VIP on rectal gland flow in the spiny dogfish Squalus acanthias. The rectal gland duct is cannulated to allow collection of secreted fluid. NaCl (1 ml of 500 m.\/) is applied through the cannulated rectal gland artery at the first arrow; the test sample is applied at the second arrow. Flow rate is measured by collec- tion of rectal gland fluid through an automated drop counter. many to be responsible for descending relaxation in the peristaltic reflex. The roles of VIP in the gut are most often manifest when the fine control of the peptide mal- functions. For example, human pathological conditions, such as the severe Verner-Morrison syndrome (watery diarrhea), Crohn's disease, or even chronic constipation, reflect either increased (Verner-Morrison, Crohn's) or decreased (constipation) VIP levels. Invertebrates Until recently, VIP has not been widely sought after in invertebrates. Immunoreactive VIP has been reported only occasionally from worms, insects, and molluscs where (as has been the case with many immunochemi- cally demonstrated vertebrate peptides) it is restricted to the nervous system (see Thorndyke and Goldsworthy, 1988, for reviews). Recent work from our own labora- tories has gone some way toward redressing this lack of information, and has also thrown light on an unexpected and perhaps novel example of the inherent sophistica- tion of adaptive evolution of host-parasite interrelation- ships. The digenean platyhelminth Echinostoma liei spends the mature phase of its life cycle in the small intes- tine of the mouse. We have discovered a subpopulation of tegumental cells in the flatworm which clearly elabo- rates an immunochemically VIP-like molecule (Thorn- dyke and Whitfield, 1987). The sequence of this peptide is presently unknown, although its immunological prop- erties suggest a resemblance to the N-terminus of mam- malian VIP. Since the tegumental cells generate the ex- tracellular surface coat of the worm, there exists the fasci- nating possibility that the parasite may be manipulating host neuropeptide levels to its own advantage. Thus, heavy infections of E. liei are associated with local in- flammation (vasodilation) and mucosal bloating (in- creased fluid secretion). These are exactly the responses that a local elevation of VIP should produce, and they are also profitable responses for E. liei which ingests mu- cus and fluid from the gut lumen. So here is a demonstra- tion of adaptive evolution, wherein the regulatory prop- erties of a peptide family transcends the phyletic divide to promote and encourage a parasitic association. Acknowledgments Much of our work has received support from the Sci- ence and Engineering Research Council (U. K.), the Nuffield Foundation, and the Royal Society (U. K.). Literature Cited Bayliss, W. M., and E. H. Starling. 1902. The mechanism of pancre- atic secretion J. Physiol. 28: 325-353. Dimaline, R., and G. J. Dockray. 1979. Potent stimulation of the avian exocine pancreas by porcine and chicken vasoactive intestinal peptides. 7. Physio!. 294: 153-163. Dimaline, R., and M. C. Thorndyke. 1986. Purification and charac- terization of VIP from two species of dogfish. Peptides 7: suppl. 1: 21-25. Dimaline, R., J. Young, D. T. Thwaites, C. M. Lee, T. J. Shuttleworth, and M. C. Thorndyke. 1987. A novel vasoactive intestinal peptide (VIP) from elasmobranch intestine has full affinity for mammalian pancreatic VIP receptors. Biochim. Biophys Ada 930: 97- 1 00. Dimaline, R., M. C. Thorndyke, and J. Young. 1986. Isolation and partial sequence of elasmobranch VIP. Reg Peptides 14: 1-10. Dimaline, R., J. Young, D. T. Thwaites, C. M. Lee, and M. C. Thorn- dyke. 1988. Amino acid sequence of a biologically active vaso- active intestinal peptide from the elasmobranch Scyliorhinus canic- ula. Ann. N. Y Acad. Sci. 527: 62 1-623. Dimaline, R. 1989. Vasoactive intestinal polypeptide. Pp. 150-173 in Comparative Physiology oj Regulatory Peptides, S. Holmgren, ed. Chapman and Hall. Publishers. Beckenham, England. Eklund, S., M. Jodel, O. Lundgren, and A. Sjoqvist. 1979. Effects of vasoactive intestinal polypeptide on blood flow, motility and fluid transport in the gastrointestinal tract of the cat. Ada Physiol. Scand. 109:461-468. El-Sahly, M. 1984. Immunocytochemical investigation ofthegastro- entero-pancreatic (GEP) neurohormonal peptides in the pancreas and gastrointestinal tract of the dogfish Squalus acanthias. Htsto- chemistrySO: 193-205. I alkiiH-r, S., J. Fahrenkrug, J. Alumets, R. Hakanson, and F. Sundler. 1980. Vasoactive intestinal polypeptide (VIP) in epithelial cells of the gut mucosa of an elasmobranchious cartilaginous fish, the Ray. Endocrinol Jpn. Suppl. 1: 31-35. Foskett, J. K., G. M. Hubbard, T. E. Macken, and H. A. Bern. 1982. Effects of epinephrine, glucagon and vasoactive intestinal polypeptide on chloride secretion by teleost opercular membrane. J. Com/). Physiol. 146: 27-34. Gerstberger, R. 1988. Functional vasoactive intestinal polypeptide (VlP)-system in salt glands of the Pekin duck. Cell Tissue Res 252: 39-48. 186 M C. THORNDYKE ET AL. Holmgren, S., and S. Nilsson. 1983. Bombesin-, gaslnn/CCK-, 5-hy- droxytryptamine-. neurotensin-, somatostatin-, and VIP-Iike im- munoreactivity and catecholamine fluorescence in the gut of the elasmobranch. Squalus acanlhias. Cell Tissue Res. 234: 595-618. Itoh, N., K-I. Obata, N. Yanaihara, and H. Okamoto. 1983. Human preprovasoactive intestinal polypeptide contains a novel PHI-27- hke peptide, PHM-27. Nature 304: 547-549. Reid. A. M., and L. H. Heywood. 1988. A companson of the effects of vasoactive intestinal polypeptide on secretion from the submaxil- lary gland of the sheep and pig. Reg Pepttdes 20: 2 1 1-22 1 . Said, S. I., and V. Mutt. 1972. Isolation from porcine intestinal wall of a vasoactive octacosapeptide related to secretin and to glucagon. Eur. J Biochem. 28: 199-204. Shuttleworth, T. J., and M. C. Thorndyke. 1984. An endogenous peptide stimulates secretory activity in the elasmobranch rectal gland. Science 225: 3 1 9-32 1 . Solomon, R., M. Taylor, J. S. Stoff, P. Silva, and F. H. Epstein. 1984. //; vivo effect of volume expansion on rectal gland function. I. Humoral factors. Am J Physio! . 246: R63-66. Tatemoto, K., and V. Mutt. 1981. Isolation and Characterisation of the intestinal peptide procine PHI (PHI-27), a new member of the glucagon-secretin family. Proc. Nail. Acad. Sci. USA 78: 6603- 6607. Thorndyke, M. C., and P. J. \\hirfield. 1987. Vasoactive intestinal polypeptide-like immunoreactive tegumental cells in the digenean helminth. Echinosloma liei: possible role in host-parasite interac- tions. (Jen. Comp Endocrinol. 68: 202-207. Thorndyke, M. C. and G. J. Goldsworthy. 1988. Neurohormones in Invertebrates. Cambridge University Press, Cambridge, England. 318pp. Thwaites, D. T., J. Young, M. C. Thorndyke, and R. Dimaline. 1988. Isolation and characterisation of two teleost VIP's. Regula- tory Peptides 22: 436. Reference: Biol .Bull IT!: 187-191. (October, 1989) Peptide YY: The Ileo-Colonic, Gastric, and Pancreatic Inhibitor IAN L. TAYLOR Department of Gastroenterology. Duke University Medical Center, Box 39 13. Durham, NC 27710 Introduction Peptide YY (PYY) was initially isolated from an ex- tract of 400 kg of porcine duodenum using a novel chem- ical assay that recognizes peptides with amidated car- boxyl termini (Tatemoto and Mutt, 1981; Tatemoto, 1 982). This latter structural modification is a characteris- tic of many biologically active brain-gut peptides. Amino acid sequencing of the final purified product demon- strated that peptide YY is a 36 amino acid residue pep- tide that exhibits structural homology (Floyd et al., 1976; Tatemoto et al.. 1982) to two other brain-gut peptides: pancreatic polypeptide (PP) and neuropeptide Y (NPY). These three peptides constitute the third family of struc- turally related peptides to be isolated from the brain-gut axis. (This family is discussed further in this issue by Price and Greenberg.) Molecular Biology The cDNA encoding the PYY precursor has been identified and isolated by screening a rat intestinal Xgtl 1 cDNA library with an antiserum raised against porcine PYY (Leiter et al. . 1987). The nucleotide sequence of the cDN A encoded a 98-residue precursor molecule (11,121 daltons) that contained within it a sequence identical to that of porcine PYY. The rat PYY sequence was pre- ceded within the pre-prohormone by a hydrophobic sig- nal sequence and followed by a carboxyl terminal exten- sion of 3 1 amino acids. A classic cleavage and amidation sequence (Gly-Lys-Arg) joined the carboxyl terminal extension to the carboxyl terminal tyrosine residue in PYY. Abbreviations: PYY— Peptide YY; NPY— Neuropeptide YY; PP- Pancreatic Polypeptide. Distribution PYY has been localized by immunocytochemical techniques to a distinct population of cells in the mucosa of the distal small intestine and colon (El-Salhy et al.. 1 983; Aponte et al., 1 985 ). The majority of the PYY cells in the ileum and colon are typically endocrine in form (El-Salhy et al.. 1983; Aponte et al. 1985). These cells are of the "open-ended type" with apical projections that reach into and sample the contents of the gut lumen; the greatest density of secretory granules occur in the base of the cell in close proximity to the underlying blood ves- sels. In a subpopulation of these cells, PYY has been found to co-localize with enteroglucagon, demonstrating that these cells are capable of expressing both genes. An- other cell type exhibits cytoplasmic processes that ema- nate from the basal region and extend for as long as 25- 30 ^/(Lundberg et al, 1982). These basal processes are typically seen in paracrine cells; i.e.. cells that release their chemical messenger locally to exert effects on adja- cent cells. Thus, PYY may act as both a paracrine and endocrine messenger. There is yet a third population of PYY immunoreactive cells that has been identified in the pancreas by immunohistochemical techniques, ra- dioimmunoassay, and the use of molecular probes (Leiter et al.. 1987). The significance of this cell popula- tion remains to be determined. Similar distribution profiles of PYY immunoreactiv- ity have been observed in rat, dog, and man; specific ra- dioimmunoassay were used to measure PYY in mucosal extracts (Lundberge/ al, 1 982; Chen et al, 1984; Taylor, 1985; Adrian et al, 1985a). Highest concentrations of PYY are observed in the distal gut, particularly in the muscosa of the ileum and colon. Release The initial studies (Lundberg et al, 1982; Chen et al, 1984) in rat and man failed to demonstrate release of 187 188 I. L. TAYLOR ~ 700- f 600- •S- 500- £ 400- £ 300- £ 200 ^ 100- 0 2 400i Hi Q rT 300H 200- PYY T 100- ct u 60 90 120 PP B B 30 60 90 120 TIME (min) Figure 1. Comparison of the Peptide YY (PYY) and pancreatic polypeptide ( PP) responses to a 15% liver extract meal in dogs. PYY into the circulation after ingestion of a meal. These observations were held to make an endocrine role for PYY unlikely. However, subsequent studies in dog (Tay- lor, 1985) and man (Adrian et ai. 1985a) have demon- strated release of PYY into the circulation, supporting a true hormonal role for PYY. These apparently conflict- ing findings may reflect greater sensitivity of the radioim- munoassays employed in the latter two studies. The PYY response to meals is distinctly different from that of its sister peptide, PP. The PP response is charac- terized (Floyd et a!.. 1 976; Taylor elal., 1978) by an early peak approximately 15-30 min after ingestion of the meal (Fig. 1 ), which reflects the importance of the ce- phalic-vagal and gastric phases of PP release. In contrast, PYY release after a meal is not as rapid, nor does it ex- hibit the same vagal-cholinergic dependence that charac- terizes the release of its sister peptide (Taylor et ai, 1978; Taylor, 1985). Although significant increases in PYY levels can be observed 30 min after the ingestion of a meal, the peak response does not occur for several hours. As shown in Figure 1, PYY concentrations are still in- creasing in the dog 4 h after the ingestion of a meal. The most potent stimulant of PYY release observed to date is the presence of fat in the intestinal lumen ( Aponte etui, 1985; Pappas t>?c//., 1986a; Adrian etai, 1986). As anticipated, PYY release is markedly enhanced in pa- tients with steatorrhea due to the presence of malab- sorbed fat in the distal gut (Adrian et ai, 1986). As an example, patients with short bowel syndrome have markedly enhanced PYY release, and the response to in- gested meals occurs much more rapidly than in normal subjects. In contrast, patients with gastrointestinal dis- ease unassociated with rapid intestinal transit or malab- sorption — e.g., diverticular disease — do not exhibit en- hanced PYY release (Adrian et ai. 1986). Biological Actions PYY and neuropeptide Y are both capable of inducing prolonged vasoconstriction when infused into peripheral blood vessels (Lundberg et ai, 1982; Edvinnson, 1985). NPY has been localized to the sympathetic nerves that innervate the smooth muscle of arteries throughout the body (Allen and Bloom, 1986), and PYY probably acts through a generic PYY-NPY receptor on vascular smooth muscle. When PYY is infused into the mesen- teric circulation, the resulting vasoconstriction is associ- ated with an inhibition of colonic motility that lasts for an hour after the peptide infusion has ceased (Lundberg etai. 1982). PYY shares with the other members of its peptide fam- ily the ability to inhibit pancreatic exocrine secretion (Tatemoto, 1982; Pappas et ai, 1985a, b). This inhibi- tory action is demonstrable whether pancreatic exocrine secretion is stimulated by a meal, or by exogenous secre- tagogues such as secretin and cholecystokinin. Al- though PYY inhibits pancreatic secretion in dog, cat, and rat (Tatemoto, 1982; Pappas et ai, 1985a, b; Louie et ai, 1985), it has no effect on pancreatic secretion in man (Adrian et ai. 1985c). This observation may reflect a true species difference, or the lower doses used in man compared to the other species studied. PYY inhibits prostaglandin-induced secretion of fluid and electrolytes from the small intestine (Saria and Beu- bler, 1985), and disturbs the distal propagation ofthein- terdigestive myoelectric complex (Lundberg et ai, 1982; Al-Saftarrtrt/.. 1985). In keeping with this latter observa- tion, treatment of animals with PYY delays transport of a labeled meal through the small intestine (Al-Saffar et ai. 1985). PYY also inhibits gastric emptying, delaying the entry of nutrients into the small intestine (Allen et ai. 1984; Pappas et ai. 1986a). As the peptide is also a potent inhibitor of gastric acid secretion in both man (Adrian et ai, 1985c) and dog (Pappas et ai, 1986b), it is a good candidate enterogastrone; i.e., a hormonal in- hibitor of gastric function. PEPTIDE YY 189 An examination of the mechanisms by which PYY in- hibits gastric secretion has given a clue to the cellular site of action of PYY (Pappas el al, 1986b). Although PYY is a potent inhibitor of meal-stimulated acid secretion in the dog. it exerts no inhibitory effect on the acid secretory response to cholinomimetics such as carbachol. In addi- tion it is, at best, a weak inhibitor of acid secretion stimu- lated by pentagastrin and histamine. As PYY does not block the action of the known hormonal and neurotrans- mitter secretogogues at the level of the parietal cell, we must assume that its effects on acid secretion are indirect. In contrast to its lack of potency in blocking the effects of external secretogogues, PYY is a very potent inhibitor of cephalic phase acid secretion, causing 95% inhibition of acid secretion induced by sham feeding in dogs. These results suggest that PYY inhibits gastric function by in- hibiting vagal tone on the stomach; as such it appears to function as an endocrine neuromodulator. Physiological Significance of PYY Over a century ago, Ewald and Boas ( 1 886) noted that the addition of olive oil to a test meal of gruel inhibited acid secretion and delayed gastric emptying in man. One of Pavlov's students went on to demonstrate that this in- hibition of acid secretion could only be demonstrated if fat was allowed to pass beyond the stomach (Gregory, 1962). In 1930, Kosaka and Lim (1933) isolated an in- hibitor of gastric secretion from the mucosa of both the small and large intestine, which they termed "enterogas- trone." Despite Kosaka and Lim's initial observation (Kosaka and Lim, 1933) that colonic extracts were as po- tent as jejunal extracts, the search for enterogastrone largely centered on the small intestine. Interest in the dis- tal bowel as the site of release of the gastric inhibitor (col- ogastrone) was rekindled by the demonstration (Seal and Debas, 1980) that perfusion of the distal bowel with a variety of nutrients — e.g.. fat and glucose — inhibited gastric secretion (Seal and Debas, 1 980; Jian el al., 1 98 1 ; Kihlelal., 1981). PYY must be considered a good candidate enterogas- trone or cologastrone since it is found in highest concen- trations in the mucosa of the ileum and colon where it is localized to endocrine-type cells. It is also a potent inhib- itor of both gastric secretion and gastric emptying. The recent demonstration that PYY is released into the circu- lation in amounts sufficient to inhibit gastric function af- ter the ingestion of fatty meals (Pappas el al, 1986b) es- tablishes the peptide's credentials as an enterogastrone. PYY may also play a pathophysiological role in dis- ease states associated with malabsorption or maldiges- tion of food. For example patients with the post-vagot- omy dumping syndrome — i.e.. those who have rapid gastric emptying after gastric surgery — exhibit markedly THE I LEAL BRAKE Decreased Pancreatic Secretion Decreased Intestinal Transit Figure 2. Schematic representation of the "ileal brake" and the po- tential role (arrows) of peptide YY in delaying gastric emptying, inhib- iting pancreatic secretion and slowing intestinal transit. enhanced PYY responses to glucose meals (Adrian el al., 1 985b). The enhanced release of PYY would be expected to delay gastric emptying. In so doing, it would help re- lieve some of the symptoms caused by the rapid and un- controlled entry of hyperosmolar solutions into the small intestine. The term "ileal brake" has been coined (MacFarlane el al., 1983) to describe a phenomenon whereby perfu- sion of the ileum and colon with fat slows intestinal tran- sit and delays gastric emptying in man. Undefined hor- monal signals originating from the ileum and colon were proposed by the original authors to mediate this re- sponse. The ileal brake is engaged when unabsorbed nu- trients pass beyond the absorptive surface of the intestine and delays the rate of delivery of unprocessed food into an already over-burdened small intestine. PYY release is markedly enhanced in diseases associated with malab- sorption of food (Adrian a al., 1986). In such states, the combination of delayed gastric emptying and delayed in- testinal transit resulting from enhanced PYY release would allow increased time for nutrient digestion. In ad- dition, the increased nutrient-mucosal contact time would lead to increased efficiency of nutrient absorption. PYY deserves consideration as a potential mediator of this phenomenon based on its distribution in the gut, its enhanced release in the face of malabsorption, and its unique spectrum of biological actions (Fig. 2). PYY may also explain another biological phenome- non initially described by two different research groups working quite independently. Sarles and his co-workers (Hagee/tf/.. 1974; Sarles el al.. 1979) described the exis- tence of "an anti-cholecystokinin hormone" in extracts of the ileocolonic mucosa. Harper and his colleagues (Harper et al. , \ 979a, b) described an inhibitor of pancre- atic exocrine secretion, pancreotone, in the mucosa of the ileum and colon. In their initial report, Harper and coworkers (Harper et ai. 1979a, b) emphasized the simi- 190 I. L. TAYLOR larities between the inhibition of pancreatic secretion in- duced by pancreotone and that observed after infusion of PYY's sister peptide, PP. As PP could not be demon- strated in colonic mucosal extracts, the authors suggested that pancreotone might act by stimulating PP release from its cell of origin in the pancreas. The discovery of large concentrations of PYY in the mucosa of the ileum and colon precludes the necessity for postulating the ex- istence of a PP secretogogue in the colonic mucosa. The potent inhibition by PYY of pancreatic secretion in rat, cat, and dog, and its occurrence in high concentrations in the ileum and colon, support the hypothesis that PYY is at least a component of both Harper's pancreotone and Sarle's anti-CCK hormone. Conclusions In summary, PYY may be of disproportionate interest to the gut endocrinologist because of the lessons it teaches about the neuro-endocrine control of gut func- tion. First, these studies re-emphasize the importance of the ileum and colon as endocrine organs. Second, they suggest a role for the distal gut in the modulation of up- per gut function. Thus, the distal gut monitors the effi- ciency of nutrient digestion and absorption in the upper intestine, and releases peptides that influence upper gut function in a way that will lead to increased efficiency of nutrient use. Finally, the recent demonstration that PYY acts as an endocrine neuromodulator calls into question the historic criteria used to define hormonal actions. Thus, a biological action that persists after denervation of the target organ has, in the past, been held to be hor- monally mediated. Conversely, an action that is abol- ished by denervation of the target organ has been held to be neurally mediated. Studies with PYY demonstrate that the actions of some gastrointestinal peptides are so intimately connected with the actions of the autonomic nervous system, that they cannot be so easily differenti- ated. These findings suggest that the conceptional dichot- omy between hormonal and neural control of gut func- tion is an artificial one. Acknowledgments The author wishes to acknowledge support from the Veterans Administration and the NIH (DK38126), and from the Sarah W. Stedman Center for Nutritional Studies. Literature Cited Adrian, T. E., G-L. Ferri, A. J. Bacarese-llamilton, H. S. Fuessl, J. M. Polak, and S. R. Bloom. 1985a. Human distribution and release of PYY, a putative new gut hormone. Gastroenterology&9: 1070-1077. Adrian, T. E., R. G. Long, H. S. Fuessl, and S. R. Bloom. 19856. Plasma peptide YY (PYY) in dumping syndrome. Dig Dis.Sct.3Q: 1145-1148. Adrian, T. E., A. P. Savage, A. J. Bacarese-Hamilton, K. Wolfe, H. S. Besterman, and S. R. Bloom. 1986. Peptide YY abnormalities in gastrointestinal diseases. GastroenterologyW: 379-384. Adrian, T. E., A. P. Savage, G. R. Sagor, J. M. Allen, A. J. Bacarese- Hamilton, K. Tatemoto, J. M. Polak, and S. R. Bloom. 1985c. Effect of peptide YY on gastric, pancreatic, and biliary function in humans. Gastroenterology&9: 494-499. Allen, Y. S., and S. R. Bloom. 1986. Neuropeptide Y: a putative neu- rotransmitter. Neurochem. Int. 8: 1-8. Allen, J. M., M. L. Fitzpatrick, J. C. Yeats, K. Darcy, T. E. Adrian, and S. R. Bloom. 1984. Effects of peptide YY and neuropeptide Y on gastric emptying in man. Digestion 30: 255-262. AI-SafTar, A., P. M. Hellstrom, and G. Nylander. 1985. Correlation between peptide YY-mduced myoelectric activity and transit of small-intestinal contents in rats. Scand. J. Gastroenterol. 20: 577- 582. Aponte, G., A. S. Fink, J. H. Meyer, K. Tatemoto, and I. L. Taylor. 1985. Regional distribution and release of peptide YY with fatty acids of different chain length. Am. J. Physiol. 249: G745-G750. Chen, M-H., A. Balasubramanian, R. F. Murphy, K. Tabata, J. E. Fischer, I-W. Chen, and S. N. Joffe. 1984. Sensitive radioimmu- noassay for measurement of circulating peptide YY. Gastroenterol- ogy 87: 1332-1338. Edvinsson, L. 1985. Characterization of the contractile effect of neu- ropeptide Y in feline cerebral arteries. Ada Physiol. Scand. 1 25: 33- 41. El-Salhy, M., E. Wilander, L. Juntti-Berggren, and L. Grimelius. 1983. The distribution and ontogeny of polypeptide YY (PYY)- and pancreatic polypeptide (PP)-immunoreactive cells in the gas- trointestinal tract of rat. Histochemistrylty, 53-60. Ewald, C. O., and J. Boas. 1886. Bentrage zur Physiologic und Patho- logie Derverlaung. Virchovis Arch. 104:271-305. Floyd, J. C., S. S. Fajans, and S. Pek. 1976. Regulation in healthy subjects of the secretion of human pancreatic polypeptide, a newly recognized pancreatic islet polypeptide. Trans. Assoc. Am. Physi- cians 98: 146-158. Gregory, R. A. 1962. Inhibition of gastric secretion. Pp. 103-133 in Secretory Mechanisms of the Gastrointestinal Tract. R. A. Greg- ory, ed., Arnold, London. Hage, G., O. Tiscornia, G. Palasciano, and H. Sarles. 1974. In- hibition of pancreatic exocrine secretion by intra-colonic oleic acid infusion in the dog. Biomedicine2\: 263-267. Harper, A. A., A. J. C. Hood, J. Mushens, and J. R. Smy. 1979a. Inhibition of external pancreatic secretion by intracolonic and intraileal infusions in the cat. J. Physiol. Land. 292: 445-454. Harper, A. A., A. J. C. Hood, J. Mushens, and J. R. Smy. 1979b. Pancreotone. an inhibitor of pancreatic secretion in ex- tracts of ileal and colonic mucosa. / Physiol. Land. 292: 455-467. Jian, R., H. S. Besterman, D. L. Sarson, C. Aymes, J. Hostein, S. R. Bloom, and J. C. Rambaud. 1981. Colonic inhibition of gastric secretion in man. Dig. Dis. Sci 26: 195-201. Kihl, B., A. Rokaeus, S. Rosell, and L. Olbe. 1981. Fat inhibition of gastric acid secretion in man and plasma concentrations of neuro- tensin-like immunoreactivity. Scand J. Gastroenterol. 16: 5 1 3-526. Kosaka, T. and R. S. K. Lim. 1933. On the mechanism of the inhibi- tion of gastnc motility by fat. An inhibitory agent from the intesti- nal mucosa. Chin. J Physiol. 7: 5-12. Leiter, A. B., A. Toder, H. J. Wolfe, 1. L. Taylor, S. Cooperman, G. Mandel, and R. H. Goodman. 1987. Peptide YY: structure of the precursor and expression in exocnne pancreas. / Bio/. Chem 262: 12,984-12,988. Louie, D. S., J. A. Williams, and C. Owyang. 1985. Action of pancre- atic polypeptide on rat pancreatic secretion: in vivo and in vitro. Am. J. Phvsiol. 249: G489-G495. PEPTIDE YY 191 Lundberg, J. M., K. Tatemoto, L. Terenius, P. M. Hellstrom, V. Mutt, T. Hokfelt and B. Hamberger. 1982. Localization of peptide YY (PYY) in gastrointestinal endocrine cells and effects on intestinal blood ttow and motility. Proc. Nail. Acad. Sa. USA 79: 447 1-4475. MacFarlane, A., R. Kinsman, N. W. Read, and S. R. Bloom. 1983. The ileal brake: ileal fat slows small bowel transit and gastric emptying in man. GUI 24: A47 1 -A472. ( Abs. ). Pappas, T. N., H. T. Debas, Y. Goto and I. L. Taylor. 1985a. Peptide YY inhibits meal-stimulated pancreatic and gastric secretion. Am. J Physio/. 248:G118-G123. Pappas, T. N., H. T. Debas, and I. L. Taylor. 1985b. Peptide YY: metabolism and effect on pancreatic secretion in dog. Gastroenter- o/ogy 89: 1387-1392. Pappas, T. N., H. T. Debas, A. M. Chang, and I. L. Taylor. 1986a. Peptide YY release by fatty acids is sufficient to inhibit gastric emptying in dogs. Gastroenterology91: 1386-1389. Pappas, T. N., H. T. Debas, and 1. L. Taylor. I986b. Enterogastrone- like effect of peptide YY is vagally mediated in the dog. /. Clin. Im-esl. 77: 49-53. Saria, A., and E. Beubler. 1985. Neuropeptide Y (NPY) and peptide YY (PYY) inhibit prostaglandin E:-induced intestinal fluid and electrolyte secretion in the rat jejunum in vivo. Eur J Pharmacol. 119:47-52. Sarles, H., G. llage, R. Laugier, P. Demol, and D. Bataille. 1979. Present status of the anticholecystokimn hormone. Diges- tion 19: 73-76. Seal, A. M., and H. T. Debas. 1980. Colonic inhibition of gastric acid secretion in the dog. Gaslroenlerology79: 823-826. Tatemoto, K. 1982. Isolation and characterization of peptide YY (PYY), a candidate gut hormone that inhibits pancreatic exocrine secretion. Proc. Nail. Acad. Sci. USA 79: 25 1 4-25 1 8. Tatemoto, K., M. Carlquist and V. Mutt. 1982. Neuropeptide Y— a novel brain peptide with structural similarities to peptide YY and pancreatic polypeptide. Nature 296: 659-660. Tatemoto, K., and V. Mutt. 1981. Isolation of two novel candidate hormones using a chemical method for finding naturally occurring polypeptides. Nature 285: 4 1 7-4 1 8. Taylor, I. L. 1985. Distribution and release of peptide YY in dog measured by specific radioimmunoassay. Gastmenterologv 88: 731-737. Taylor, I. L., M. Feldman, C. T. Richardson and J. H. Walsh. 1978. Gastric and cephalic stimulation of human pancreatic poly- peptide release. GastroenteroIogylS: 432-437. Reference: Biol. Bull, 111: 192-194. (October, 1989) Tachykinins and the Bombesin-Related Peptides: Receptors and Functions STEVEN R. VIGNA Department of Cell Biology, Duke University Medical Center, Box 3709. Durham, North Carolina 27710 Two vertebrate peptide families — the tachykinins and bombesin-related peptides — share many similarities of sequence and distribution. The amino acid sequences of the peptides in these families are similar in the carboxy- terminal region, and this is the molecular center of bio- logical activity in both groups. All tachykinins share the C-terminal sequence -Phe-X-Gly-Leu-Met-NH:, and many (but not all) bombesin-related peptides have -Leu- Met-NH: at the carboxy terminus (Table I). Further- more, in both families, oxidation of the methionine resi- due at the C-terminus reduces or abolishes the biological activity of the peptides (Vigna el ai, 1988). Peptides from both families are expressed in nerves and gut endo- crine cells of vertebrates; representatives from each group are also expressed in exocrine glands located in the skins of certain frogs. The similarity of these observa- tions suggests that the tachykinins and bombesin-related peptide families may overlap in their functions or may even share a common origin in evolution. The mRNAs encoding the precursor proteins for sev- eral mammalian members of these peptide families have recently been characterized. The gastrin-releasing pep- tides (GRPs) of human (Spindel el at., 1984) and rat (Lebacq-Verheyden et ai, 1988) are bombesin-like and their precursors are very similar to each other. However, these precursors are both quite different from the «- and /j-preprotachykinins (Nawa et ai, 1983) which arise from the alternate splicing of the mRNA transcribed from a single gene. Thus the mammalian tachykinins and bombesin-related peptides appear to be encoded by genes that are unrelated. The precursor /3-preprotachykinin is particularly in- teresting because it contains one copy each of two ta- chykinins, substance P and substance K. This suggests that, in the neurons that express |tf-preprotachykinin. both substance P and substance K are processed out of the precursor. Therefore, although both peptides may be released, their production of a functional response will be determined by the presence or absence of the appropriate postsynaptic receptor. To begin evaluating the functions of and the interac- tions among the bombesin-related peptides and tachyki- nins in mammals, we studied the localization and prop- erties of the receptors for various peptide members of these two families. We have also initiated comparative studies of these receptors in nonmammalian vertebrates to analyze the putative evolutionary relationship be- tween these molecular families. Our studies have fo- cussed on the expression of receptors in the gastrointesti- nal tract because the pharmacological actions of both the bombesin-related peptides and the tachykinins have been extensively studied in the guts of mammals and nonmammals. We have shown, by quantitative autoradiography of saturable radioligand binding, that receptors specific for substance P and for substance K are expressed in the ca- nine gastrointestinal tract (Mantyh el ai. 1988b). More- over, these two receptors exhibited different patterns of anatomical localization. Therefore, separate receptors for substance P and substance K are expressed and differentially distributed in the gastrointestinal tract. No evidence for expression of receptors for the third mam- malian tachykinin, neuromedin K, was observed, even though we were able to demonstrate neuromedin K re- ceptors in the canine central nervous system. Similar studies of bombesin receptors in the canine gastrointestinal tract revealed a pattern of binding sites specific for the radioligand i:5I-(Tyr4)-bombesin. The distribution was clearly distinct from that of the sub- stance P and substance K receptors (Vigna et ai, 1987). 192 TACHYKIN1NS AND BOMBESIN-LIKE PEPTIDES Table 1 Amino acid sequences of bombesin- and tachykinin-related peptides 193 1 5 10 15 20 25 GRP Human VPLPAGGGT VLTKMYPRGNH WA V G H L MNH, Pio A V S V A NH rig r\ IN OT r-. A V G Q Y\ NH A--QP--SP A I S i Nni NH C_ hicken p. fi L. 1 A v P N n s F p * * o NH Bombesin rt — v c i» v^ — or r — o — — pQ Q R L - - Q - — IN FIT - NH: Substance P - PKPQQFFG- - NH, Substance K (Neurokinin A) HKTDSFVG- - NH: Neuromedin K (Neurokinin B) DMHDFFVG- - NH, Scyliorhinirr AKFDKFYG- - NH_, Eledoisin' pQPSKD-FI G- - NH: Residues similar to those in human gastrin releasing peptide (GRP), a bombesin-related peptide, are indicated by (-). 1 Dogfish (Scyliorhinus) GRP; * = position of two presumed deletions (Conlon el ai. 1987). : A dogfish tachykinin (Conlon etal, 1986). ' A molluscan tachykinin (Erspamer and Anastasi, 1962). Further evidence for the distinct nature of these three neuropeptide receptors came from the demonstration of increased expression of substance P receptors, but not substance K or bombesin receptors, in inflammatory bowel disease in people (Mantyh el at., 1988a). The distribution of binding sites for iodinated bombe- sin was also examined in the stomach of the bony fish, Scorpaeichthys marmoratus. High affinity, saturable bombesin binding sites were localized on the circular and longitudinal layers of gastric smooth muscle in both the antral and oxyntic regions of the stomach (Vigna and Thorndyke, 1989); no receptors were seen in the antral mucosa, which was the richest source ofbombesin bind- ing in the dog stomach. These observations demonstrate that the simple techniques used to localize and character- ize receptors in mammals are applicable to nonmamma- lian species and, furthermore, that putative bombesin re- ceptors have different distributions in mammalian and nonmammalian stomachs. Recent studies of the structure and actions of bombe- sin-related peptides in nonmammalian vertebrates re- veal a remarkable degree of conservation. Shark GRP is reported to be nearly or completely identical to mamma- lian GRP at the biologically active carboxy-terminus of the peptide (Conlon el ai, 1987) (Table I). Of the general categories of action of bombesins in mammals — includ- ing the stimulation of release of gastrointestinal hor- mones, contraction of gut muscle, digestive enzyme se- cretion, and decrease of body temperature — most have now been demonstrated in response to bombesin admin- istration in one or more nonmammalian vertebrate spe- cies. Two examples of such effects deserve comment be- cause they illustrate a phenomenon that may be of par- ticular importance in understanding the evolution of this peptide family. First, bombesin causes acid secretion in mammals by stimulating the release of the gastric hor- mone, gastrin, which in turn directly initiates gastric acid secretion. Bombesin also stimulates acid secretion from the bony fish stomach (Holstein and Humphrey, 1980), but this must occur by a different mechanism, as yet un- known, because fish do not have gastrin in their stom- achs. The second comparison concerns the reduction, by bombesin, of body temperature in rats and fish. In rats, this action is caused by an unknown physiological mech- anism (Brown and Vale, 1979). In carp, although bom- besin has the same overall effect, it is accomplished by a behavioral mechanism (Kavaliers and Hawkins, 198 1 ). In both cases described above, the actions ofbombesin have been conserved in evolution, but the mechanisms by which bombesin carries out these actions have changed. Future analysis of other aspects of this fascinat- ing pattern of evolution will contribute much to our un- derstanding of the importance of these families of regula- tory peptides to biological success. Literature Cited Brown, M., and W. Vale. 1979. Bombesin — a putative mammalian neurogastrointestinal peptide. Trends Neurosci. 2:95-97. Conlon, J. M., C. F. Deacon, L. O'Toole, and L. Thim. 1986. Scyliorhimn I and II: two novel tachykinins from dogfish gut. f-'EBS Lett. 200: 111-116. Conlon, J. M., I. \V. Henderson, and L. Thim. 1987. Gastrin-releas- ing peptide from the intestine of the elasmobranch fish. Scyliorhi- nus canicula (common dogfish). Gen. Comp. Endocrinol. 68: 415- 420. Erspamer, V., and A. Anastasi. 1962. Structure and pharmacological actions of eledoisin, the active endecapeptide of the posterior sali- vary gland ofElcdone. Experientia 181: 58-59. 194 S. R. VIGNA Holstein, B., and C. S. Humphrey. 1980. Stimulation of gastric acid secretion and suppression of VIP-like immunoreactivity by bombe- sin in the Atlantic codfish, Gadus morhua. Ada Phy.tiol. Scand. 109:217-223. Kavaliers, M., and M. F. Hawkins. 1981. Bombesin alters behavioral thermoregulation in fish. LifeSci. 28: 1361-1364. Lebacq-Verheyden, A.-M., G. Krystal, O. Sartor, J. Way, and J. F. Battey. 1988. The rat prepro-gastrin releasing peptide is tran- scribed from two initiation sites in the brain. Mol. Endocritml. 2: 556-563. Mantyh, C. R., T. S. Gates, R. P. Zimmerman, M. L. Welton, E. P. Passaro, Jr., S. R. Vigna, J. E. Maggio, L. Kruger, and P. W. Man- tyh. 1988a. Receptor binding sites for substance P, but not sub- stance K. or neuromedin K, are expressed in high concentrations by arterioles, venule. and lymph nodules in surgical specimens ob- tained from patients with ulcerative colitis and Crohn disease. Proc. Nail. Acad. Sci. USA 85: 3235-3239. Mantyh, P. W., C. R. Mantyh, T. S. Gates, S. R. Vigna, and J. Maggio. 1988b. Receptor binding sites for substance P and substance K. in the canine gastrointestinal tract and their possible role in inflam- matory bowel disease. Neurnsdence 25: 8 1 7-837. Nawa, H., T. Hirose, H. Takashima, S. Inayama, and S. Nakanishi. 1983. Nucleotide sequences of cloned cDNAs for two types of bo- vine brain substance P precursor. Nature 306: 32-36. Spindel, E. R., \V. VV. Chin, J. Price, L. H. Rees, G. M. Besser, and J. F. Habener. 1984. Cloning and characterization of cDNAs en- coding human gastnn-releasing peptide. Proc. Nail. Acad. Sci. USA 81:5699-5703. Vigna, S. R., C. R. Mantyh, A. S. Giraud, J. H. Walsh, A. H. Soil, and P. W . Mantyh. 1987. Localization of specific binding sites for bombesin in the canine gastrointestinal tract. Gasiroenlerology93: 1287-1295. Vigna, S. R., A. S. Giraud, J. R. Reeve, Jr., and J. H. Walsh. 1988. Biological activity of oxidized and reduced iodinated bombesins. Peptides9: 923-926. Vigna, S. R., and M. C. Thorndyke. 1989. Bombesin and related pep- tides. Pp. 34-60 in Comparative Physiology of Regulatory Peptides, S. Holmgren, ed. Chapman and Hall, London. Reference: Bioi Bull 177: 195-197. (October, 1989) Gastrin, Cholecystokinin (CCK), and the Leukosulfakinins G. J. DOCKRAY Physiological Laboratory. University of Liverpool, Brownlow Hill, Liverpool L69 3BX, United Kingdom Introduction The gastrin-CCK family is operationally defined as consisting of peptides with the common C-terminal tet- rapeptide amide structure: Trp-Met-Asp-Phe-NH;. This is also the minimal fragment with biological activity. Therefore, the distinctive patterns of activity of gastrin or CCK are determined largely by other regions of the molecule (Dockray, 1989). In addition to the two mammalian peptides, there is an amphibian member of the group, caerulein, and an avian peptide that Rod Dimaline has characterized, which we call chicken gastrin (Dimaline el al., 1986) (Fig. 1). In the invertebrates, the leucosulfakinins have been isolated by Nachman et al. ( 1 986), and share several residues with the vertebrate gastrins, notably the se- quence Tyr(SO3H)-Gly-X,-Met-X2-Phe-NH2 (Fig. 1). The leucosulfakinins are also similar to the FMRFamide family in terminating in Met-Arg-Phe-NH2 (Fig. 1; see also Price and Greenberg, 1989). Whether they are phy- logenetically related to gastrin and CCK is hard to say; there is no evidence of common sites of action, and the leucosulfakinin precursor is not yet known. I tend to the view that this is an example of convergent evolution. Precursors The genes encoding gastrin, CCK, and caerulein have been cloned and sequenced by a number of groups. Pre- progastrin and preproCCK are organized on rather sim- ilar lines. The chain length is about 100 residues includ- ing the signal sequence, and there is a single copy of the major active molecular form. Away from the main active sites, however, the primary structures of the two precur- sors are not at all similar. In the case of procaerulein, Kriel's group has found a variety of different possible precursors, one of which has at least two copies of caerulein and is probably quite large (Richter^a/., 1986; Vlasak et al., 1987). Again, outside of the pentapeptide amide, there are virtually no regions of the caerulein precursor common to any in either gas- trin or CCK. More important, the organization of the caerulein precursor is plainly different from that of gas- trin and CCK. On this latter basis, one might reasonably question whether there is any phylogenetic relationship between caerulein and the other members of the family (Dockray, 1989). Processing Peptide precursors are converted by a variety of steps to the major active products. In the case of progastrin there are only two main cleavages, but in proCCK there are several. Moreover, different cells expressing these genes show different cleavage patterns. One interesting point is that, within this family, the cleavage sites are quite variable. In progastrin there is cleavage at pairs of basic residues, in proCCK mainly at single basic residues, and in procaerulein and avian gastrin there is evidence of cleavage at dibasic residues followed by dipeptidyl aminopeptidase activity. However, other types of post-translational processing mechanisms are better conserved. The mechanisms of tyrosine sulfation follow the usual rules for sulfotransfer- ase activity, namely an acid residue in the immediate up- stream portion. However, avian gastrin is not typical in this regard (see below). On present evidence, the mechanisms of C-terminal amidation are the same as those found in other peptides, but there is an interesting feature associated with the rele- vant part of the gastrin and CCK precursors. We have recently isolated the C-terminal flanking peptides of hu- 195 196 G, J. DOCKRAY Human gastrin ! pGlu-Gly-Pro-Trp-Leu-Glu-Glu-Glu-Glu-Glu-Ala-Tyr-Gly-Trp-Met-Asp-Phe-NH, * CCK8 Asp-Tyr-Met-Gly-Trp-Met-Asp-Phe-NH, * Chicken gastrin -Gly-Ala-Val-Glu-Ala-Leu-His-Asp-His-Pne-Tyr-Pro-Asp-Trp-Met-Asp-Phe-NH, Caerulein pGlu-Gln-Asp-Tyr-Thr-Gly-Trp-Met-Asp-Phe-NH, * Leucosulfakinin I Glu-Gln-Phe-Glu-Asp-Tyr-Gly-His-Met-Arg-Phe-NH, * Leucosulfakinin II pGlu-Ser-Asp-Asp-Tyr-Gly-His-Met-Arg-Phe-NH, FMRFanide Phe-Met-Arg-Phe-HH, Figure 1. Amino acid sequences of human gastrin I (G17), chole- cystokinin octapeptide (CCK.8), chicken gastrin (20-36), caerulein (amphibian), the leucosulfakinins (cockroach), and FMRFamidel mol- luscs). Alignment is from the C-terminal. "Indicates sulfated tyrosine residues. pGlu = pyroglutamic acid. man and hog progastrin, and have found two forms of each of these peptides. These forms were poorly sepa- rated by reverse phase HPLC, but were well separated on ion exchange chromatography. Using the latter method, then, we were able to demonstrate that alkaline phospha- tase digestion converts one form of the peptide to the other. On sequencing, it was found that one form was phosphorylated on the N-terminal serine, and the other was not. The phosphorylation site (i.e., the serine resi- due) seems to be a part of two consecutive consensus se- quences: one (-Arg-Arg-Ser-) recognized by cyclic AMP- dependent protein kinase, and the other (-Ser-Ala-Glu- Glu-) by the casein kinase group of kinases. Interestingly, similar sequences also occur in proCCK, and at several locations in proFMRFamide. Phosphorylation may, in some way, identify, tag, or label this region as important for subsequent processing to the C-terminal amide (Dockrayetal., 1989). Receptors Throughout the vertebrates, the CCK gene is ex- pressed both in gut endocrine cells and central neurones, particularly in forebrain regions and in the hypothala- mus. An important question arising from this pattern of distribution is whether there is any functional link be- tween brain and gut, and if not, how is specificity of ac- tion maintained in the two systems. This, in turn, is part of the larger question of how specificity is, in any case, maintained between gastrin and CCK. The results of re- ceptor binding assays indicate that there are three types of receptors in mammals and birds: the so-called central and peripheral CCK receptors, and the gastrin receptor. We have a good antagonist for at least one of these — L- 364,7 1 8 — which is selective for peripheral CCK recep- tors. Steven Vigna's work suggests that, in lower verte- brates, there is a single receptor in brain and gut. Whether there is also a gastrin receptor is difficult to say, particularly because there is no general agreement on whether gastrin actually exists below the reptiles (Vigna etal.. 1986). The crucial feature that determines specificity of ac- tion of CCK at mammalian receptors is the position and sulfation of the tyrosine. The characteristic CCK activity depends on a sulfated tyrosine at position 7 from the C- terminus. In lower vertebrates, a sulfated tyrosine is im- portant, but the position is less important. The selectivity explains why mammalian gastrin, with the tyrosine at position 6 from the C-terminus, is virtu- ally inactive at CCK receptors. But there is a problem with chicken gastrin. This peptide has a tyrosine, not in the gastrin position, but in the CCK position. The natu- rally occurring material is nevertheless a gastrin because it stimulates acid secretion but does not stimulate pan- creatic secretion, as Rod Dimaline has shown (Dimaline etal., 1986). Recent studies comparing the sulfated and unsulfated forms of chicken gastrin confirm this pattern — both are virtually inactive on turkey pancreatic secretion and gall bladder contraction (Rod Dimaline and Caroline Lee, pers. comm.). One possibility appears to be that the chicken has found an interesting way of determining specificity for gastrin over CCK receptors. In particular, the substitution of a proline immediately adjacent to the sulfated tyrosine in chicken gastrin (Fig. 1) may intro- duce a conformational constraint that has the same effect as shifting the sulfate group relative to the C-terminus. It should be possible, now, to test this hypothesis by synthe- sis of a range of analogs. Actions and Roles Finally, I would like to consider families of actions, rather than families of peptides. Until recently, we knew CCK could act at a number of sites, but we didn't know whether these actions were physiologically important. Now we can study this question because we have a spe- cific antagonist selective for the peripheral type of CCK receptor. Applied to the pancreas, L364.718 inhibits CCK-, but not bombesin-evoked amylase release, and it has the same inhibiting effect on the gall bladder. CCK also de- lays gastric emptying, probably by relaxing the body of the stomach. A variety of different liquid test meals in- hibit gastric emptying in the conscious gastric fistula rat. The action of protein-rich meals in delaying gastric emp- tying is inhibited by the CCK-antagonist, but the action of other meals is not (Green el a/.. 1988). We can start, then, to arrange the different actions of CCK into an integrated picture in which CCK plays a role in regulating the environment of the small intestine. Thus, on the one hand. CCK stimulates the pancreas and THE GASTRIN-CCK FAMILY 197 gall bladder directly to control delivery of pancreatic en- zymes and bile salt to the duodenum, and on the other hand, it acts to delay gastric emptying and food intake, and so balances the delivery of nutrient substrate with that of the enzymes and bile salt required for its digestion and assimilation. Some or all of these actions are found throughout the vertebrates. The appearance of gastrin in the higher vertebrates extended the role of this group of peptides to control of the stomach and may have been a later development. The tools are at hand to examine this scheme at all levels, from the cellular and molecular to the whole ani- mal. It should now be possible to trace the phylogeny of this family through its integrative roles in the gut. Acknowledgments My work is supported by grants from the Medical Re- search Council, the Science and Engineering Research Council, and the Agriculture and Fisheries Research Council of the United Kingdom. I am grateful to Rod Dimaline, Mike Thorndyke, and Mike Greenberg for helpful discussions. Literature Cited Dimaline, R., J. Young, and H. Gregory. 1986. Isolation from chicken antrum. and primary amino acid sequence of a novel 36- residue peptide of the gastrin/CCK family. FEES Leu 205: 318- 322. Dockray, G. J. 1989. The comparative neuroendocrinology of gut peptides. In The Neuroendocrinology' of the Gut, G. Makhlouf, ed. Handbook of Physiology. Section on the Gastrointestinal System, American Physiological Society, in press. Dockray, G. J., R. Dimaline. S. Pauwels, and A. Varro. 1989. Gastrin and CCK-related peptides. Pp. 244-284 in Peptide Hormones as Prohormones. J. Martinez, ed. Ellis Horwood, Chichester. Green, T., R. Dimaline, S. Peikin, and G. J. Dockray. 1988. Action of the cholecystokmin antagonist L-364,7 1 8 on gastric emptying in the rat. Ant J. Physiol. 225: G685-689. Nachman, R. J., G. M. Holman, B. J. Cook, W. F. Haddon, and N. Ling. 1986. Leucosulfakinin II, a blocked sulphated insect neuro- peptide with homology to cholecystokinin and gastrin. Biochem. Biophys. Res. Comm. 140: 357-364. Price, David A., and Michael J. Greenberg. 1989. The hunting of the FaRPs: the distribution of FMRFamide-related peptides. Bio/ Bull. 177:000-000. Richter, K., R. Egger, and G. Kreil. 1986. Sequence of preprocaeru- lein cDNAs cloned from skin ofXenopus laevis. A small family of precursors containing one, three, or four copies of the final product. J Biol. Chem 261: 3676-3680. Vigna, S., M. C. Thorndyke, and J. A. Williams. 1986. Evidence fora common evolutionary origin of brain and pancreas cholecystokinin receptors. Proc. Natl. Acad. Sci. USA 83: 4355-4359. Vlasak, R., O. Wiborg, K. Richter, S. Burgschwaiger, J. Vuust, and G. Kreil. 1987. Conserved exon-intron organization in two different caerulein precursor genes ofXenopus laevis. Additional detection of an exon potentially coding for a new peptide. Eur. J. Biochem. 169: 53-58. Reference: Biol Bull. 177: 198-205. (October, 1989) The Hunting of the FaRPs: The Distribution of FMRFamide-Related Peptides DAVID A. PRICE AND MICHAEL J. GREENBERG The Whitney Laboratory of the University of Florida, 9505 Ocean Shore Blvd., St. Augustine, Florida 32086 For the FaRPs are peculiar peptides, that won't Be caught in a commonplace way. Do all that you know, and try all that you don't: Not a chance must be wasted to-day!1 Introduction The family of F MRFamide-related />eptides (FaRPs) is a large assemblage of neuropeptides found throughout the Metazoa( Priced a/., 1987a; Greenberg et ai. 1988; Greenberg and Price, 1988). The first FaRP completely characterized— the tetrapeptide amide Phe-Met-Arg- Phe-NH2 (i.e.. FMRFamide- itself) was isolated from a clam by virtue of its cardioexcitatory effect (Price and Greenberg, 1977); but we will use the term FaRP for any peptide that can be found by looking with an assay for FMRFamide, and these assays need not involve cardio- excitation or any other biological effect. A few molluscan FaRPs have been found using bioassay, but most, espe- cially in other phyla, have been found with immunoas- says. New FaRPs are constantly being added to the fam- ily rolls as the number of taxa investigated, and the detail with which each is studied, increases. Therefore, the gen- eralizations made here, and based on only a few scattered ' With apologies to Lewis Carroll. : The uppercase letters in the name "FMRFamide" are the approved one-letter abbreviations of the amino acids constituting the peptide. Thus, "FMRFamide" conveys both the sequence of the peptide and its blocked C-terminal. FMRFamide can also be pronounced, and people often ask us how to do it. The answer is: "Anyway you want." We have heard, for example: fmerf'-amid, fuh-merf'-amid, fer-merf'-amid. fer- maf-amid, fer-mahf'-amid, fah-mahf'-amid. and ef-em-ahr-ef-ay'- mide. All of these pronunciations, and any others that anyone can pro- duce, are totally, and therefore equally, correct. In summary, the mean- ing of FMRFamide resides unambiguously and eternally in the written characters, and is independent of linguistic variability. data, will undoubtedly prove incorrect in the long term, but still provide a framework for further studies. Our objectives in this review are to enumerate the large, heterogeneous set ofFMRFamide-related peptides (FaRPs) and to classify them into more obviously related subsets. We begin by describing the molluscan peptides and our assumptions about their relatedness. We go on to identify (more or less in chronological order) the non- molluscan peptides that have been thought of as mem- bers of the FaRP family, and consider the characters that seem to relate them. From this broader perspective, we then attempt to divide the FaRPs into those that show clear evidence of homology from those that do not. Since FMRFamide is a mere tetrapeptide, we cannot establish homology by examining the peptide sequences alone; they are simply too short. Fortunately, several genes for FaRP precursors have been cloned and sequenced, and we will rely heavily on these data for demonstrating likely homologies. Hunting the Molluscan FaRPs By the time FMRFamide was sequenced there was al- ready chromatographic evidence that FMRFamide-like biological activity is present throughout the Mollusca (Agarwal et ai, 1972). We now believe that FMRFamide itself accounts for the majority of this activity in most species, with FLRFamide as a ubiquitous and always mi- nor component (Price, 1986, 1987). FLRFamide is typi- cally present at 10-20% the level of FMRFamide, except in opisthobranchs (or at least Aplysia) where it is difficult to even detect. Two close congeners of FLRFamide (se- quences in Table I) have been found as minor compo- nents in Octopus vulgaris by Martin and Voigt (1987). FMRFamide is not, however, the major immunoreac- tive FaRP in all molluscs. In one group— the pulmonate snails and slugs — heptapeptides of the form XDPFLR- 198 THE HUNTING OF THE FARPs 199 Table I The FMRFamide-relatedpeptides (FaRPs) of molluscs The tetrapeptides and two relatives from Octopus Phe-Met-Arg-Phe-NH, FMRFamidc Phe-Leu-Arg-Phe-NH: FLRFamide Thr-Phe-Leu-Arg-Phe-NH: TFLRFamide Ala-Phe-Leu-Arg-Phe-NH: AFLRFamide The pulmonati' heptapeptides pGlu-Asp-Pro-Phe-Leu-Arg-Phe-NH: pQDPFLRFamide Ser-Asp-Pro-Phe-Leu-Arg-Phe-NH2 SDPFLRFamide Gly-Asp-Pro-Phe-Leu-Arg-Phe-NH: GDPFLRFamide Asn-Asp-Pro-Phe-Leu-Arg-Phe-NH2 N DPFLRFamide Right-hand column: the amino acids are represented by their one- letter abbreviations; pQ and pGlu are pyroglutamic acid. Famide constitute a significant fraction of the FMRFam- ide-like activity. In collaboration with others (Price et ai, 1985;Ebberink£>/a/., 1987; Priced al, 1987b), we have identified four such heptapeptides with X = Gly (G), Ser (S), pGlu (pQ), or Asn (N) (Table I). The FMRFamide precursor gene ofAplysia has been sequenced (Schaefer et ai, 1985; Taussig and Scheller, 1986). Most of it is composed of repeated stretches of 15 or 16 amino acids, each segment containing one FMRF- amide, its associated processing signals, and an acidic se- quence which may only be a spacer. All together, the gene encodes 28 copies of FMRFamide, one copy of FLRFamide, and one copy of an apparently related pep- tide ending in -Tyr-Leu-Arg-Phe-amide. Though this last peptide has never been identified in extracts, it is ex- pected to be a minor FaRP in Aplysia. Since many of the repeated segments of the Aplysia gene are more than 90% identical at the nucleotide level, we suggested that they arose relatively recently (geologi- cally speaking), and that back-extrapolation would lead to a more general molluscan precursor (Price et al.. 1987b). Such a hypothetical, general molluscan precur- sor would have only about nine copies of FMRFamide, one of FLRFamide, and one corresponding to the as yet undetected Aplysia peptide: -Tyr-Leu-Arg-Phe-NH2. Such a precursor would come close to accounting for the FMRFamide/FLRFamide ratio observed in most mol- luscs, and, of course, assumes that all molluscs have a FMRFamide precursor homologous to, but less highly duplicated than, that ofAplysia. We further hypothesized that the pulmonate hepta- peptides are encoded on a separate gene, perhaps corre- sponding to the extreme N-terminal part of the ancestral precursor (Price et al., 1987b). This hypothesis was based, not so much on the Aplysia gene (which is not really relevant since Aplysia does not have the heptapep- tides), as on the disjoint tissue distributions of FMRFam- ide and the heptapeptides (Lehman and Price, 1987), and on their distinct biological effects on some muscles (Lehman and Greenberg, 1987) in Helix. Likewise, this hypothesis assumes that the heptapeptide gene is homol- ogous to the Aplysia FMRFamide gene. A recent report has given some support to parts of this hypothesis. A cDNA clone containing the entire precur- sor encoding region has been isolated and sequenced from the pulmonate Lymnaea (Linacre et al., 1989). This precursor encodes nine copies of FMRFamide and two of FLRFamide, so it is only about half the size of that from Aplysia and is also less highly replicated. Nev- ertheless, the genes from the two species are clearly ho- mologous. The Lymnaea precursor also encodes neither of the heptapeptides found in this snail by Ebberink et al. (1987). These findings support our concept of the ori- gin of the Aplysia precursor, justifies our supposition that the FLRFamide/FMRFamide ratio in tissue extracts re- flects the relative number of copies of the two peptides in the precursor, and bolsters the idea that the heptapep- tides are encoded on a separate gene, or genes. An unusual molluscan FaRP Though many neurons in Aplysia abdominal ganglion stain with antisera against FMRFamide (Brown et al., \ 985), not all of them express the FMRFamide precursor gene described above. For example, neurone L5 was found to express a small precursor, dissimilar in se- quence to that of FMRFamide, but containing one copy of a potentially secreted peptide ending: -Gln-Gly-Arg- Phe-NH2 (Shyamala et al, 1986). Because the substitu- tion of Gly for Met in FMRFamide is associated with a large decrease in biological activity (Greenberg et al., 1988), the Aplysia L5 peptide is probably not functional at FMRFamide receptors. Though the tetrapeptides and heptapeptides have dis- tinct effects, they are usually biologically cross-reactive, and we have always assumed that they are homologous. We argue below that the L5 peptide is not homologous to FMRFamide and, in fact, use it as an outlier in our attempts to identify those FaRPs that are most likely to be homologous. Hunting the Non-Molluscan FaRPs Soon after FMRFamide was sequenced, antisera were raised against it and were used in immunocytochemistry to reveal FMRFamide-like immunoreactivity through- out the animal kingdom (reviewed in Greenberg et al., 1985). Even the first paper to appear (Boer et al., 1980) reported FMRFamide-like immunoreactivity in arthro- pods and vertebrates, as well as molluscs. Vertebrates Immunocytochemical studies of mammalian (rat) brain ( Weber et al.. 1981; also Williams, 1983;Dockray 200 Sequences of some non-molluscan FaRP.i D. A. PRICE AND M. J. GREENBERG Table II Common name Invertebrate peptides antho-RFamide pol-RFamide antho-RWamide antho-RWamide II leucomyosuppressin schistoFLRFamide leucosulfakinin LSKll Source Sequence t 'ertebrate peptides gamma,-MSH ox YVMGHFRWDRFa BPP ox APLEPQYPGDDATPEQMAQYAAELRRYINMLTRPRYa APP chicken GPSQPTYPGDDAPVEDLIRFYDNLQQYLNVVTRHRYa NPY Pig YPSKPDNPGEDAPAEDLARYYSALRHYINLITRQRYa PYY Pig YPAKPEAPGEDASPEELSRYYASLRHYLNLVTRQRYa PP salmon YPPKPENPGEDAPPEELAKYYTALRHYINLITRQRYa PP alligator TPLQPKYPGDGAPVEDLIQFYNDLQQYLNVVTRPRFa A18Fa ox AGEGLSSPFWSLAAPQRFa F8Fa ox FLFQPQRFa chicken LPLRFa sea anemone (Anthopleura) hydrozoan jellyfish (Polyorchis) sea anemone sea anemone fruitfly (Drosophila) roundworm (Ascaris) American lobster (Homarus) Amencan lobster cockroach (Leucophaea) locust (Schistocerca) cockroach cockroach pQGRFa pQLLGGRFa pQSLRWa pQGLRWa OPKQDFMRFa KNEFIRFa SDRNFLRFa TNRNFLRFa pQDVDHVFLRFa PDVDHVFLRFa pQSDDYGHMRFa EQFEDYGHMRFa References and further information about the peptides in the text. Abbreviations: MSH, melanophore stimulating hormone; BPP, bovine pancre- atic polypeptide; APP, avian PP; NPY, neuropeptide tyrosine; PYY, peptide with N- and C-terminal tyrosines; LSK II, leucosulfakinin II. The one-letter abbreviations of the amino acids are used in the sequences; E, Glu; H, His; I, He; K, Lys; Q, Gin; pQ, pyroglutamic acid; V, Val; W, Trp; Y, Tyr; a, amide; others in Table I. and Williams, 1 983) showed extensive anti-FMRFamide staining, but most exceptionally in the cortex where few peptides are localized. Gamma,-MSH. One mammalian peptide — gammar MSH — had been predicted to end in RFamide based on the sequence of the bovine pro-opio-melanocortin (Na- kanishi et ai, 1979) gene (sequence in Table II). Thus, it might cross-react with FMRFamide antisera, but being restricted mainly to the pituitary, seemed unlikely to ac- count for the cortical staining. Moreover, rats (and other rodents) do not produce the amidated peptide because their precursors lack the appropriate processing signals. Later, Ali-Rachedi et al. (1983) showed that some FMRFamide and gammarMSH antisera do indeed stain the same pituitary corticotrophs in those mamma- lian species having the amidated form of the peptide. But gammarMSH can only account for a small part of the total FMRFamide-like staining in the brain. Pancreatic polypeptide. The pancreatic polypeptides are also vaguely similar to FMRFamide. The first mem- bers of the family sequenced (from pancreas, of course; Lin and Chance, 1974;Kimmel etal, 1 975) both ended - Arg-Tyr-NH:. Though the C-terminal pentapeptide of bovine pancreatic polypeptide (BPP) lacked FMRFam- ide-like biological activity (Price and Greenberg. 1980), FMRFamide antisera stain vertebrate pancreas and gut, so it was argued that the PPs might be FMRFamide ana- logs. Antisera to avian PP, or to the C-terminal hexapeptide of BPP also showed a pattern of staining in mammalian brain (Lundberg et ai, 1980) similar to that revealed by anti-FMRFamide. Were the PP antisera revealing FaRPs or were the FMRFamide antisera revealing PPs? The issue was resolved by Tatemoto, who had devel- oped a chromatographic method for identifying peptides with C-terminal amides and was applying it to porcine brain and gut. He isolated and sequenced two peptides that are clearly related to the pancreatic polypeptides (Tatemoto, 1982a, b; sequences in Table II). These pep- tides— NPY from brain and PYY from intestine — share the C-terminal sequence Arg-Tyr-amide with the pancre- atic members of the family. When antisera to NPY were raised, they stained the same brain structures as the APP antiserum (Lundberg el al., 1984; Moored al., 1984), so the story seemed to be that the pancreatic polypeptide- like (and possibly the FMRFamide-like) immunoreac- tivities of brain were due to NPY. THE HUNTING OF THE FARPs 201 However, though some FMRFamide antisera stain the same brain structures as antisera to NPY, others do not (e.g., Chronwall el a/.. 1 984; Triepel and Grimmelikhuij- zen, 1984). So NPY could not account for all of the stain- ing. When alligator PP was shown to end in Arg-Phe- NH:, rather than Arg-Tyr-NH2 (Lance et a/., 1984), it revived interest in the relation between the FaRPs and the PPs. Such a relationship would also have seemed stronger if NPY had been found using antisera to FMRF- amide rather than by hunting for peptide amides. Chicken and beef. The first peptide completely charac- terized by following its FMRFamide-like cross-reactivity was LPLRFamide from chicken brain (Dockray el ai, 1983; sequence in Table II). LPLRFamide seems to be unrelated to the pancreatic polypeptide family, is similar in size to FMRFamide, and has a tripeptide amide se- quence in common with the natural FMRFamide conge- ner FLRFamide. It even has detectable, though low, FMRFamide-like biological activity (3/10,000 of FMRFamide on the clam heart and the radula protrac- tor). Two peptides were then isolated by Yang et al. (1985) from bovine brain (sequences in Table II). Both of these peptides have a glutamine where the chicken peptide has a leucine, so they are more NPY-like than the chicken peptide, but they still may well be its bovine homologs. By 1985, therefore, the hunt for vertebrate FaRPs was winding down. Three new peptides had been discovered and characterized, and they seemed to represent a new peptide family. As for the relationship between the pan- creatic polypeptide family and FMRFamide. it did not seem to be a very close one. Indeed, recent work has shown that the PP of fish is very NPY-like and has the usual C-terminal: Arg-Tyr-NH2 (Kimmel et al., 1986; se- quence in Table II). So the Arg-Phe-NH: terminal of alli- gator PP is not primitive, but rather a reptilian specializa- tion convergent to FMRFamide. Non-molluscan invertebrates Coelenterates. The first coelenterate FaRP was iso- lated from an anthozoan and named antho-RFamide (pQGRF-NH:; Grimmelikhuijzen and Graff, 1986; see Table II). It is a tetrapeptide, like FMRFamide, but actu- ally has more sequence in common with the L5 peptide of Aplysia (see Greenberg and Price, 1988). A similar, but longer peptide (pQLLGGRFamide, Grimmelikhuij- zen et al., 1988), has been isolated from another class (Hydrozoa) of coelenterates. It appears to be simply the hydrozoan version of anthoRFamide. More recently, a pair of peptides that are more FMRF- amide-like have been isolated from a sea anemone. These two peptides, pQSLRWamide and pQGLRWam- ide (antho-RWa and antho-RWa II, see Table II; Graff and Grimmelikhuijzen, 1 988a, b), have a Trp-NH: at the C-terminal, rather than a Phe-NFL. But thev were dis- covered by their activity in an RIA for FMRFamide, and our structure-activity studies indicate that FMRFamide analogs with a Trp substituted for either Phe residue are effective in molluscan bioassays. They both also have the leucine residue in common with FLRFamide, so their FMRFamide-like biological potency would exceed that of the coelenterate FaRPs terminating in -GRFamide. The relationship between these two types of peptide, both coelenterate FaRPs, remains unclear. Arthropods. A bevy of FaRPs have been flushed from a disparate group of arthropod species. Two related pep- tides from the American lobster (Trimmer et al., 1987), one from the fruitfly, Drosophila (Nambu et al., 1988), and one from a locust (schistoFLRFamide; Robb et al., 1989) have been isolated by following their FMRFam- ide-like immunoreactivity and sequenced (Table II). Still, all three species had several additional peaks. Some of the additional fruitfly peaks probably represent pep- tides predictable from the structure of the precursor gene (Nambu et al., 1988; Schneider and Taghert, 1988). Three additional FaRPs — leucomyosuppressin (Hoi- man et ai, 1986) and leucosulfakinins I and II (Nach- man et ai, 1986a, b; Table II) — were isolated from cock- roaches solely on the basis of their biological actions. Leucomyosuppressin is a single amino acid variant of schistoFLRFamide (Robb et ai, 1989), and the leuco- sulfakinins, in addition to being FaRPs, show similarities to the gastrin/CCK family, a matter discussed by G. J. Dockray in this volume (see also Greenberg and Price, 1988). Since none of the several FaRPs predictable from the fruitfly precursor are very similar to the cockroach or lo- cust peptides. more fly FaRPs may remain to be discov- ered. Thus, there are probably distinct sub-families of FaRPs in insects, just as there are in molluscs. Nematodes. Neurons in the free-living nematode Caenorhabditis elegans were immunochemically stained with anti-FMRFamide sera (Li and Chalfie, 1986). Shortly thereafter, a FaRP, KNEFIRFamide, was de- tected by RIA in Ascaris suutn and sequenced (Cowden et ai . 1 989; Table II). This was the first report of a natural FaRP with an isoleucine replacing the methionine resi- due. Our structure-activity relations (SAR) studies indi- cate that this change would produce an analog with only moderately lowered FMRFamide-like biological activity on the clam heart. The Thrill of the Hunt The hunting of FaRPs has led to the successful identi- fication of peptides in diverse groups of animals from coelenterates to mammals, and in some of these groups they are among the first peptides completely character- ized. How to account for this success? First, all described 202 D. A. PRICE AND M. J. GREENBERG antisera to FMRFamide (and virtually all bioassays sen- sitive to this peptide) seem to discriminate very well the amidated peptide from unamidated or C-terminally ex- tended forms. The presence of a C-terminal amide is one unmistakable mark of a peptide that has gone through the secretory pathway, and all known peptide amides seem to be involved in intercellular communication. Second, and beyond amidation. an RFamide is gener- ally regarded as the minimum structural requirement for immunoreactivity with even the least selective anti- FMRFamide sera. But the stringency of this requirement for most antisera in wide use has not been evaluated. We have noted above that at least some anti-FM RFamide sera can stain NPY-containing neurons (-Arg-Tyr-NH;) in mammalian brain, and the -Arg-Trp-NH; peptides in coelenterate neurons. Even the requirement for a penul- timate Arg is not universal (Dockray, 1985). In sum- mary, FMRFamide antisera are relatively unselective de- tectors, so the number of potentially detectable peptides in a species is large. Thus, the catch is rather good, if var- ied (Table II). Subgroups of FaRPs Come listen, my men, while I tell you again The clear unmistakable marks By which you may know, wheresoever you go, The warranted genuine FaRPs.1 The FaRPs collected in Table II are a heterogeneous assemblage, ranging in size from 4 to 36 amino acids, with a penultimate Arg and a C-terminal amide as the only common structural features. Thus, there appears to be no natural limit to what someone might consider to be a FaRP (e.g., Morris et a/., 1982). However, within the FaRPs, certain peptides constitute structurally re- lated, and possibly homologous, subgroups. We are par- ticularly interested in denning the subgroup of peptides homologous to FMRFamide. Homologous FaRPs Since homology implies inheritance of common fea- tures from common ancestors, detection of homology is best achieved by comparing the genes encoding the pre- cursors. Since few FaRP genes or precursors are known, we are left, in most cases, with examining the similarity between the peptides themselves. Sequence similarity. Unfortunately, the reliability of sequence similarity as a clue to homology declines with the length of the peptides being compared. Short stretches of amino acid sequence occur frequently by chance; Price (1983) calculated that any given sequence of four amino acids will occur at least once in the pro- teins of any species, but the odds go down dramatically as the sequences become longer. Thus, matches of six amino acids in a row (without gaps) are fairly unique and indicate a likely homology. The expression of FMRFam- ide actually requires a sequence of at least seven amino acids: Phe, Met, Arg, and Phe (the body of the peptide); Gly (to form the amide); and at least one basic amino acid (e.g., Arg) — but more usually a dipeptide (Lys-Arg, Lys-Ser, or Arg-Ser) — on each end as a cleavage signal. On the above argument, we assume a match of six resi- dues (including processing signals) as the minimum re- quirement for homology. Then, for example, the Dro- sophila peptide DPKQDFM RFamide, which matches FMRFamide in six of seven residues, is a likely homolog. And so are leucomyosuppressin and schistoFLRFam- ide, the two lobster peptides, and the pulmonate hepta- peptides; all of them have six residues in a row matching FLRFamide (compare sequences in Table I and II). The nematode peptide KNEFIRFamide, has only four resi- dues in a row matching FMRFamide or FLRFamide. Nevertheless, it is very similar to DPKQDFM RFamide from the fruittly; four of seven residues are identical, and the other three are conservative substitutions. So we con- clude that the Ascaris and Drosophila peptides are ho- mologous to each other and that the worm peptide is a homologous FaRP. In contrast, the leucosulfakinins (which have the C- terminus-DYGHMRF-NH:) have a match of only five amino acids, and the tyrosine residue is sulfated (Nach- man et a!., 1986a, b). Thus, this peptide is an unlikely homolog of FMRFamide. Similarly, we would exclude LPLRFamide, the rest of the vertebrate FaRPs, and the coelenterate peptides. On the basis of sequence, then, we have selected a set of homologous FaRPs that is characterized by the C-ter- minal structure FXRFamide (where X = M, L, or I). Precursors. For the FaRPs that appear to be homolo- gous on the basis of sequence, precursor genes have only been completely sequenced from Aplysia and Drosoph- ila. Inspection of these precursors reveals several gross similarities, especially the highly repetitive nature of each (Namburtfl/.. 1988; Schneider and Taghert, 1988). This repetitiveness can be illustrated by diagonal plots of each precursor upon itself (Fig. 1). The strong lines parallel to the main diagonal in these plots reflect the internal similarities within the precursors (Fig. la, b). In contrast, a diagonal plot comparing the Aplysia and Drosophila precursors (Fig. Ic) shows no such prominent diagonal lines, but only a series of very short diagonal dashes indi- cating the matching FMRFamide sequences within the two precursors. The longest such match (and also the longest stretch of exact amino acid sequence match) be- tween the precursors is the heptapeptide FMRFGRS, whereas each precursor has 5 tandem repeats of the exact same 1 5 (Aplysia) or 1 1 (Drosophila) amino acids. Thus, the repeats within each precursor are more similar to THE HUNTING OF THE FARPs 203 Aplysla FMRFa Precursor I Drosophila FMRFa Precursor M O M Aplysia FMRFa Precursor I Figure 1. Diagonal plots (dot matrix plots, see Doolittle, 1986, for more explanation) comparing the Aplysia and Drosophila FaRP pre- cursor sequences to themselves (top, middle) and to each other (bot- each other than they are to those of the other precursor. How could such a relationship develop? The simplest explanation is that each precursor origi- nated with an independent partial duplication of an an- cestral precursor. Alternatively, as suggested by Nambu el at. (1988), their most recent common ancestor could have been highly duplicated and one or both of its daugh- ters suffered partial gene conversion to regularize their repeats. These possibilities are not unconnected, and both would be favored if the FMRFamide gene con- tained a so-called "hot spot"; i.e.. a region of high recom- bination frequency. Of the FaRPs that do not appear to be homologous on the basis of peptide sequence, precursors are known for the pancreatic polypeptide-like peptides (Minth et ai, 1984; Leiter et ai. 1985), gammai-MSH (Nakanishi et ai. 1979), and the Aplysia L5 peptide (Shyamala et ai. 1986). The precursor encoded by the L5 gene is exem- plary: it is not detectably repetitive (Fig. 2a), and con- tains only a single stretch of sequence [RFGKR; the ter- minal dipeptide ( RF ) plus its amidation and cleavage sig- nals] in common with either the Aplysia or Drosophila FMRFamide precursor (Fig. 2b, c). So a comparison of the genes supports our conclusion from the peptide se- quences that this precursor is not homologous to that of FMRFamide. Conclusions The subset of FaRPs terminating in FXRFamide are probably homologs. But some members of this group may not be homologous to FMRFamide at all, while other FaRPs that lack the FXRFamide sequence will prove to be homologs once their precursors can be scruti- nized. Some FaRPs that are not in the FXRFamide sub- set seem to fall into distinct families of peptide homologs. But lacking the requisite genetic data, we cannot yet reli- ably define these groups. The plethora of metazoan FaRPs — peptides recog- nized by assays for FMRFamide — may reflect idiosyn- cracies of the eukaryotic secretory pathway. In particu- lar, the presence of a C-terminal amide is a universal re- quirement for a FaRP, but not all amino acids are equally suitable substrates for the amidation system; non-polar and aromatic amino acids seem to be espe- cially favorable. Likewise, the frequent occurrence of ar- tom). The computer compares successive tetrapeptide sequences and draws a dot wherever four in a row from each protein match. The self comparisons show a complete diagonal line of self identity, and parallel lines showing internal duplications. A match of more than four amino acids in a row results in a diagonal line (down and to the right) of n-3 dots where n is the number of identical ammo acids. 204 D. A. PRICE AND M J. GREENBERG Aplysia FMRFa Precursor Drosophila FMRFa Precursor M — - Q in . -> o TO e 11 < Q. IT) . -> O OJ S? II < Q- Aplysia L5 Precursor M— -.L Figure 2. Diagonal plots comparing the Aplysia L5 gene product to itself and to the FaRP precursors of Figure 1 . ginine in the penultimate position may reflect character- istics of the trypsin-like processing system. The FXRFamides are found in nematodes, molluscs, annelids, and arthropods. These and related phyla con- stitute the Protostomia, a group of invertebrates that are supposedly related because they share certain develop- mental characters. Of course, the leukosulfakinins and the L5 peptide show us again that every protostomian FaRP detected is not necessarily homologous to FMRF- amide. Finally, if the protostomian FXRFamides are indeed homologs, why have they been conserved? We may spec- ulate that, since most of these peptides appear to modu- late muscle contractility, they may be involved in con- trolling some feature unique to protostomian muscle. Acknowledgments The original work reported here was supported by NIH grant HL-28440 and NSF grant DCB-86 16356. This is contribution #285 from the Tallahassee, Sop- choppy and Gulf Coast Marine Biological Association. Literature Cited Agarwal, R. A., P. J. B. Ligon, and M. J.Greenberg. 1972. The distri- bution ofcardioactive agents among molluscan species and tissues. Comp. Gen. Phannacol. 3: 249-260. Ali-Rachedi, A., G. L. Ferri, I. Varndell, S. Van Noorden, L. P. C. Schot, N. Ling, S. R. Bloom, and J. M. Polak. 1983. Immunocy- tochemical evidence for the presence of gammarMSH-like immu- noreactivity in pituitary corticotrophs and ACTH-producing tu- mours. Neuroendacrinol, 37: 427-433. Boer, H. H., L. P. C. Schot, J. A. Veenstra, and D. Reichelt. 1980. Immunocvtochemical identification of neural elements in the central nervous system of a snail, some insects, a fish and a mamma] with an antiserum to the molluscan cardio-excitatory tet- rapeptide FMRF-amide. Cell Tis. Res 213: 21-27. Brown, R. O., D. Gussman, A. I. Basbaum, and E. Mayeri. 1985. Identification of Aplysia neurons containing immunoreac- tive FMRFamide. Neuropeptides 6: 5 17-526. Chronwall, B. M., J. A. Olschovvka, and T. L. O'Donohue. 1984. Histochemical localization of FMRFamide-like immuno- reactivity in the rat brain. Peptides 5: 569-584. Cowden, C., A. O. W. Stretton, and R. E. Davis. 1989. Al, a se- quenced bioactive peptide isolated from the nematode Ascaris siiinn. Neuron 2: 1465-1473. Dockray, G. J. 1985. Characterization of FMRFamide-likeimmuno- reactivity in rat spinal cord by region-specific antibodies in radioim- munoassay and HPLC. J Neurochem. 45: 152-158. Dockray, G. J., and R. G. Williams. 1983. FMRFamide-like immu- noreactivity in rat brain: development of a radioimmunoassay and its application in studies of distribution and chromatographic prop- erties. Brain Res 266: 295-303. Dockray, G. J., J. R. Reeve, Jr., J. Shively, R. J. Gayton, and C. S. Barnard. 1983. A novel active pentapeptide from chicken brain identified by antibodies to FMRFamide. Nature 305: 328-330. Doolittle, R. F. 1986. OJVRFs andORFs: A Primer on How to Ana- lyze Derived Ammo Acid Sequences. University Science Books. Mill Valley, California. Fbberink, R. H. M., D. A. Price, H. van Loenhout, K. E. Doble, J. P. Riehm, W. P.M. Geraerts, and M. J.Greenberg. 1987. The brain of Lymnaea contains a family of FMRFamide-like peptides. Pep- tidesS: 515-522. Graff, D., and C. J. P. Grimmelikhuijzen. 1988a. Isolation of /0/., 1987a,b, 1988). There- fore, rather than redescribing all of the disparate data, we have concentrated on results from a limited number of representative preparations, including heart, somatic muscle, and nerve. We ask, finally, whether any generali- ties are applicable to the actions of FMRFamide and its relatives. Heart of Rapana thomasiana FMRFamide is cardioexcitatory in many molluscs, but in some species it also has inhibitory effects (Painter and Greenberg, 1982; other references in Kobayashi, 1987). In the prosobranch Rapana, both serotonin and FMRFamide enhanced the amplitude and frequency of heart beat, with FMRFamide having the lower threshold and producing the greater enhancement ( Kawakami and Kobayashi, 1984). The excitatory effects of serotonin were blocked by methysergide, a potent antagonist of se- rotonin receptors in molluscan hearts. On the contrary. the effects of FMRFamide were not affected by methy- sergide, showing that the receptors for serotonin and for FMRFamide are different. The amplitude of heart beat was also augmented when the right or /eft cardiac «erves, RCN 3a, RCN 4, or LCN 1 were electrically stimulated. The excitatory effects of nerve stimulation were not affected by the application of methysergide for 60 min or more. Thus, serotonin is probably not involved in the neurally induced excitation of this heart. Recently, we have found, by staining with an antise- rum to FMRFamide, that immunoreactive cell bodies and fibers are distributed throughout the visceral ganglia, and that such fibers also occur in the atrium (Kobayashi, 1987). Therefore, although the mechanism of action of FMRFamide is still not yet clarified, this peptide may well play a physiological role as a cardioactive agent in Rapana. We have studied the structure-activity relations (SAR) of FMRFamide on the isolated Rapana heart (Kobaya- shi and Muneoka, 1986), and have shown that: ( 1 ) The C-terminal RFamide is critical for activity; potency is markedly diminished by substitution with D- amino acids and is abolished upon removal of the amide. (2) The N-terminal phenylalanine and the methio- nine could be replaced by other residues, but a total length of at least four residues is important for activity. (3) N-terminal elongation may have little effect. (4) FMRFamide was the most potent of 14 peptides tested. These features are common, but not inevitable, char- acteristics of FMRFamide-receptor interactions. In par- ticular, N-terminal elongation enhances the action on some pulmonate preparations, and in such cases, FMRFamide may be a relatively weak agonist (see be- 206 ACTIONS OF FMRFAMlDE-LIKE PEPTIDES 207 low). However, in general, the contribution to potency of the various residues in each position along a peptide needs to be examined in much greater detail. Heart of Achatina fulica The mode of action of FMRFamide on this pulmonate heart was different from that observed in Rapana. So far, we have identified nine neurons in the central nervous system of Achatina that are involved in regulating the heart beat (Furukawa and Kobayashi, 1987a, b). One gi- ant neuron, designated the 'Periodically Oscillating Neu- ron' (PON), was the most potent heart excitor. Both the cardioexcitation produced by PON, and the excitatory effects of serotonin application, were depressed by meth- ysergide, suggesting that the transmitter between the PON neuron and the heart is serotonin. The effects of several FaRPs were tested on the atrium of Achatina (Hori ct a!., unpub.). The preparation (com- prising the atrium with most of the ventricle cut away, the intestinal nerve, and the central ganglia) was isolated in a bath to which the peptides were added. Of the pep- tides studied, only FMRFamide showed conspicuous ex- citatory effects. Although the threshold for the direct effects of FMRFamide on the heart was quite high (i.e., 10 - M or more), lower concentrations of FMRFamide enhanced the cardioexcitatory actions of both PON stimulation and serotonin application. The sites of ac- tion of FMRFamide (i.e., pre- or post-synaptic) have not been established, but in any case, these effects were in contrast to those of SCPB which depressed the cardioex- citatory actions. Therefore, as one of its roles in Acha- tina, FMRFamide appears to be modulating the excit- atory action of a transmitter to the heart. Neither the FMRFamide- nor the SCP-containing neurons in Acha- tina have been identified. Buccal Muscles of Rapana The reciprocating movement of the radular rasp dur- ing feeding is produced in this prosobranch by the alter- nating contraction and relaxation of two pairs of oppos- ing buccal muscles, the radula protractors and retractors. These muscles are innervated by the radula nerves which arise in the buccal ganglia (Furukawa and Kobayashi, 1985). In these radula muscles, the FaRPs seem to mod- ulate the release of transmitters by acting presynaptically (Yanagawat'/fl/.. 1988). FMRFamide enhanced contractions of the radula pro- tractor that were elicited by short pulses of electrical field stimulation (probably affecting nerve elements in the muscle), but the peptide had no such effect on the oppos- ing muscle, the radula retractor. In contrast to FMRF- amide, its close analog FLRFamide enhanced the con- traction of the retractor but had no enhancing effect on the protractor. When neuromuscular transmission was blocked by application of 80 mM Mg++, contraction of the protractor elicited by stimulation with long pulses (i.e., direct muscle stimulation) was not enhanced by FMRFamide. Similarly, contraction of the retractor caused by muscle stimulation was not enhanced by FLRFamide (Yanagawa tV al, 1988). Previously, we showed that the principal excitatory transmitter in the radula protractor is ACh, whereas that in the retractor is glutamate. Moreover, we know that serotonin acts to excite the protractor and to inhibit the retractor (Kobayashi and Muneoka, 1980;Muneokaand Kobayashi, 1980). Now we present the hypothesis that FMRFamide and FLRFamide act on presynaptic sites in the protractor and retractor, respectively, to enhance their contractions, possibly by increasing the release of transmitter. Anterior Byssus Retractor Muscle (ABRM) of Mytilus edulis FMRFamide also appears to have a presynaptic action on the ABRM of a bivalve mollusc (Muneoka and Ma- tsuura, 1985). The ABRM of Mytilus can be set into a prolonged contracture by acetylcholine (ACh), and this catch tension is relaxed by serotonin released from the relaxing nerve; the serotonergic relaxation is blocked by mersalyl (references in Muneoka and Matsuura, 1985). FMRFamide at low concentrations (10~8-10~7 M) also relaxes ACh-induced catch tension. Moreover, this relaxation of the ABRM, like that of serotonin, was also blocked by mersalyl. However, when the muscle wasde- nervated, ACh-induced catch tension was not relaxed by FMRFamide, although serotonin still relaxed it. These results are consistent with the notion that FMRFamide is acting on the relaxation inducing neurons in the mus- cle to release serotonin from their terminals. FMRFamide showed various actions on the Mytilus ABRM, and catch relaxation is only one of them. The peptide also enhanced the contraction elicited by electri- cal stimulation of the muscle, or by application of ACh to it (Muneoka and Matsuura, 1985). The threshold for these effects was about 10~9 M. At higher concentrations (more than 10~7 M), FMRFamide caused its own con- traction. These actions of FMRFamide are probably postsynaptic. The structure-activity relations of FMRFamide for contraction of the ABRM were different from those for relaxation (Muneoka and Saitoh, 1986). Although the C- terminal RFamide was, as usual, critical for contraction, effective relaxation could still be produced when D-Arg and D-Phe (or other residues) were substituted for the C- terminal Arg and Phe, respectively. Even the C-terminal amide was not essential for relaxation. These results sug- 208 M. KOBAYASHI AND Y. MUNEOKA A2 B1 C1 Figure 1. Duration of the action potential of the "Periodically Oscillating Neuron' (PON) was increased by a burst of impulses in a cerebral neuron (d-LCDN) (A), and by an application of 5-HT (5 x 1CT6 M) (B), but was decreased by FMRFamide (10 " Al) (C). In (A) and (B), PON was hyperpolarized to -50 mV (dotted line in Al ) and driven to fire by a depolarizing current injection. In (C), spontaneous firings were recorded. Arrows in A 1 , B 1 , and C 1 indicate selected spikes which are displayed at expanded time scale in A2. B2, and C2. (A and B; from Furukawa and Kobayashi, 1988). gest that the Mylihts ABRM has at least two pharmaco- logically distinct classes of receptors which can be acti- vated by FMRFamide. Central Neurons and Synapses Recently, the mechanisms underlying the actions of the FaRPs at central synapses have been intensively in- vestigated (CottrelU'/ a/.. 1 984; Colombaioni ctal.. 1985; Ruben el al.. 1986; Berladetti el at.. 1987; Brezina el al. 1987; Thompson and Ruben, 1988). In sensory neurons ofAplysia, the inhibitory responses to FMRFamide ap- pear to be mediated by lipoxygenase metabolites of ara- chidonic acid, which open S-type K+-channels (Piomelli et al., 1987). This action of FMRFamide is in contrast to that of serotonin which closes these S-K+-channels; i.e.. the effect of serotonin is mediated by cAMP by way of a different guanine nucleotide-binding protein than that coupled to the FMRFamide receptor (Volterra and Sie- gelbaum, 1988). Similar results have also been obtained in the heart ex- citatory neuron, PON, of Achatina. FMRFamide in- creased background 1C -conductance (i.e.. it increased K+-current through S-channels), and it also decreased in- ward Ca++-current and, in turn, reduced Ca++-depen- dent K+-current. Moreover, these actions of FMRFam- ide were, again, opposite to those of serotonin. As a re- sult, the duration of the action potential of a PON was decreased by FMRFamide, but was increased by seroto- nin (Fig. 1 ). Serotonin is considered to act as an excitatory neuro- transmitter at the synapse between a pair of command neurons in the cerebral ganglia and the PON (Furukawa and Kobayashi, 1988). Now we have found that FMRF- amide is acting counter to serotonin at the same synapse (Hori et al.. unpub.), although the FMRFamide contain- ing neuron has yet to be identified. The actions of the tetra- and heptapeptide FaRPs have been examined recently on identified neurons of a pul- monate snail. Helix aspersa (Cottrell and Davies, 1987). The tetrapeptides were more potent than the heptapep- tides at producing a slow increase in potassium conduc- tance (gK). In addition, the tetrapeptides produced an increase in gNa, a conductance change not seen at all in response to the heptapeptides. On the other hand, the heptapeptides produced a fast increase in gK which was not observed when tetrapeptides were applied. The exclusive actions of the tetrapeptides and hepta- peptides on different ionic currents strongly suggest that multiple receptors are present. Recently, a receptor was demonstrated in Helix heart and brain that is specific for the tetrapeptides; i.e.. the heptapeptides were very in- effective at displacing radioligand bound to membranes from these tissues (Payza, 1987). Summary We have surveyed the functions, receptors, and mech- anisms of the FMRFamide-related peptides by focussing primarily on preparations we have studied. Even these few examples clearly illustrate the versatility of the FaRPs: acting as neurotransmitters, they can directly ex- cite or inhibit target cells; or they can potentiate or op- pose the actions of a variety of other neuroefTector mole- cules. In the end, there are no general characteristics that can be assigned to the effects of FMRFamide or its ana- logs. Rather, the results show that the FaRPs exhibit multiple actions on various tissues, reflecting the struc- ACTIONS OF FMRFAMiDE-LIKE PEPT1DES 209 tural variation, not only of the peptides, but also of their receptors. Acknowledgments We are grateful to Michael J. Greenberg for valuable discussions and for reviewing the manuscript. This re- search was supported in part by Grants-in-Aid (Nos. 62540549 and 63540575) from the Ministry of Educa- tion, Science, and Culture, Japan. Literature Cited Belardetti, F., E. R. Kandel, and S. A. Siegelbaum. 1987. Neuronal inhibition by the peptide FMRFamide involves opening of S K* channels Nature 325: 153-156. Brezina, V., R. Eckert, and C. Erxleben. 1987. Modulation of potas- sium conductances by an endogenous neuropeptide in neurones of Aplysia californica. J Physiol. 382: 267-290. Bulloch, A. G. M., D. A. Price, A. D. Murphy, T. D. Lee, and H. N. Bowes. 1988. FMRFamide peptides in Helisoma: identification and physiological actions at a peripheral synapse. J. Neurosci. 8: 3459-3469. Colombaioni, L., D. Paupardin-Tritsch, P. P. Vidal, and H. M. Gersch- enfeld. 1985. The neuropeptide FMRF-amide decreases both the Ca;* conductance and a cyclic 3',5'-adenosine monophosphate-de- pendent K' conductance in identified molluscan neurons. J. Neu- rosci. 5:2533-2538. Cottrell, G. A., and G. S. Bewick. 1989. Novel peripheral neurotrans- mitters in invertebrates. Pharmac. Ther. 41: 41 1-442. Cottrell, G. A., and N. W. Davies. 1987. Multiple receptor sites for a molluscan peptide (FMRFamide) and related peptides of Helix. J Physiol. 382:51-68. Cottrell, G. A., N. W . Davies, and K. A. Green. 1984. Multiple actions of a molluscan cardioexcitatory neuropeptide and related peptides on identified Helix neurons. / Physiol. 356: 3 1 5-333. Cropper, E. C., P. E. Lloyd, VV. Reed, R. Tenenbaum, I. kupfermann, and K. R. Weiss. 1987a. Multiple neuropeptides in cholmergic motor neurons ofAplysia: evidence for modulation intrinsic to the motor circuit. Proc. Nail. Acad Set. USA 84: 3486-3490. Cropper, E. C., R. Tenenbaum, M. A. G. kolks, I. Kupfermann, and K. R. Weiss. 1987b. Myomodulin: a bioactive neuropeptide pres- ent in an identified cholinergic buccal motor neuron of Aplvsia. Proc. Nad. Acad. Sci. USA 84: 5483-5486. Cropper, E. C., M. W. Miller, R. Tenenbaum, M. A. G. Kolks, I. Kup- fermann, and K. R. Weiss. 1988. Structure and action of buccalin: a modulatory neuropeptide localized to an identified small cardio- active peptide-containing cholinergic motor neuron ofAplysia cali- fornica. Proc. Nail. Acad. Sci. USA 85: 6 1 77-6 1 8 1 . Eurukawa, Y., and M. Kobayashi. 1985. Neural mechanisms under- lying the feeding movements of a mollusc. Rapana thomasiana. Comp. Biochem. Physiol. 81A: 779-786. Furukawa, Y., and M. Kobayashi. 1987a. Neural control of heart beat in the African giant snail. Achatina fulica Ferussac. 1. Identification of the heart regulatory neurones. J. Exp. Biol. 129: 279-293. Furukawa, Y., and M. Kobayashi. 1987b. Neural control of heart beat in the African giant snail. Achatina fulica Ferussac. II. Interconnec- tions among the heart regulatory neurones. J. Exp. Biol. 129: 295- 307. Furukawa, Y., and M. Kobayashi. 1988. Modulation of ionic currents by synaptic action and 5-HT application in the identified heart ex- citatory neurone of the African giant snail. Achatina fulica Ferussac. J. Exp. Biol. 137:319-339. Greenberg, M. J., S. D. Painter, K. E. Doble, G. T. Nagle, D. A. Price, and H. K. Lehman. 1983. The molluscan neurosecretory peptide FMRFamide: comparative pharmacology and relationship to the enkephalins. Fed. Proc. 42: 82-86. Kawakami, H., and M. Kobayashi. 1984. Pharmacological approach to the analysis of regulation of molluscan heart activity. Zoo/. Sci. 1:389-397. Kobayashi, M. 1987. Innervation and control of the heart of a gastro- pod, Rapana. Experiential: 981-986. Kobayashi, M., and Y. Muneoka. 1980. Modulatory actions of octo- pamine and serotonin on the contraction of buccal muscles in Ra- pana thomasiana — I. Enhancement of contraction in radula pro- tractor. Comp. Biochem Physiol. 65C: 73-79. Kobayashi, M., and Y. Muneoka. 1986. Structural requirements for FMRFamide-like activity on the heart of the prosobranch Rapana thomasiana. Comp. Biochem. Physiol. 84C: 349-352. Lehman, H. K., and D. A. Price. 1987. Localization of FMRFamide- like peptides in the snail Helix aspersa. J. Exp. Biol. 131: 37-53. Lloyd, P. E., M. Frankfurt, P. Stevens, I. Kupfermann, and K. R. Weiss. 1987. Biochemical and immunocytological localization of the neuropeptides FMRFamide. SCPA. SCPB. to neurons involved in the regulation of feeding in. -tplysia. J. Neurosci. 7: 1 123-1 132. Muneoka, Y., and M. Kobayashi. 1980. Modulatory actions of octo- pamine and serotonin on the contraction of buccal muscles in Ra- pana thomasiana — II. Inhibition of contraction in radula retractor. Comp. Biochem. Physiol. 65C: 81-86. Muneoka, Y., and M. Matsuura. 1985. Effects of the molluscan neu- ropeptide FMRFamide and the related opioid peptide YGGFMRF- amide on .\fytilus muscle. Comp Biochem. Physiol. 81C: 61-70. Muneoka, Y., and H. Saitoh. 1986. Pharmacology of FMRFamide in Mytilus catch muscle. Comp. Biochem. Physiol. 85C: 207-214. Painter, S. D., and M. J. Greenberg 1982. A survey of the responses of bivalve hearts to the molluscan neuropeptide FMRFamide and to 5-hydroxytryptamine. Biol. Bull. 162: 31 1-332. Payza, K. 1987. FMRFamide receptors in Helix aspersa. Peptides 8: 1065-1074. Piomelli, D., A. Volterra, N. Dale, S. A. Siegelbaum, E. R. Kandel, J. H. Schwartz, and F. Belardetti. 1987. Lipoxygenase metabo- lites of arachidonic acid as second messengers for presynaptic inhi- bition ofAplysia sensory cells. Nature 328: 38-43. Raffa, R. B. 1988. The action of FMRFamide (Phe-Met-Arg-Phe- NHi) and related peptides on mammals. Peptides 9: 9 1 5-922. Ruben, P., J. W. Johnson, and S. Thompson. 1986. Analysis of FMRF-amide effects on Aplysia bursting neurons. / Neurosci. 6: 252-259. Sossin, W. S., M. D. Kirk, and R. H. Scheller. 1987. Peptidergic modulation of neuronal circuitry controlling feeding in Aplysia. J. Neurosci. 7:671-681. Thompson, S., and P. Ruben. 1988. Inward rectification in response to FMRFamide in Aplysia neuron L2: summation with transient K current. J. Neurosci. 8: 3200-3207. Volterra, A. and S. A. Siegelbaum. 1988. Role of two different gua- nine nucleotide-binding proteins in the antagonistic modulation of the S-type K4 channel by cAMP and arachidonic acid metabolites in Aplysia sensory neurons. Proc. Natl. Acad. Sci. USA 85: 7810- 7814. Weiss, K. R., P. E. Lloyd, E. C. Cropper, M. Frankfurt, and I. Kupfer- mann. 1986. FMRF-amide is present in the arc muscle ofAplysia and depresses its contractions. Soc. Neurosci. Abstr. 12: 947. Yanagawa, M., M. Fujiwara, I. Takabatake, Y. Muneoka, and M. Ko- bayashi. 1988. Potentiating effects of some invertebrate neuro- peptides on twitch contraction of the radula muscles of a mollusc; Rapana thomasiana. Comp. Biochem Physiol. 90C: 13-11 . Reference: Bint. Bull 177: 210-217. (October, 1989) The Egg-Laying Hormone Family: Precursors, Products, and Functions GREGG T. NAGLE, SHERRY D. PAINTER, AND JAMES E. BLANKENSHIP The Marine Biomedical Institute and Department of Anatomy and Neurosciences, The University of Texas Medical Branch, Galveston, Texas 77550 Abstract. The marine mollusc Aplysia produces an egg-laying hormone (ELH), which induces ovulation and acts on central neurons to effect egg-laying behavior. ELH is synthesized in the neuroendocrine bag cells; it is encoded by the ELH gene, one of a small family of genes, each of which is expressed in a tissue-specific manner. We review what is known about post-translational pro- cessing of the ELH precursor, and report the isolation and chemical characterization of t-bag-cell peptide, the seventh peptide product of the ELH precursor to be iden- tified to date. Amino acid compositional and sequence analyses demonstrated that the primary structure of the 19-residue peptide is: NH;-Ser-Val-Leu-Thr-Pro-Ser- Leu-Ser-Ser-Leu-Gly-Glu-Ser-Leu-Glu-Ser-Gly-Ile-Ser- COOH. Several other ELH-related genes are expressed in the atrial gland, an exocrine organ secreting into the oviduct of Aplysia. We review post-translational process- ing of these ELH-related precursors, and compare the events to those in the neuroendocrine bag cells. Finally, we compare the sequences of six ELH-related peptides from Aplysia with one ELH-related peptide (caudodor- sal cell hormone) from Lymnaea to gain insight into the structure-activity relations of ELH at the ovotestis re- ceptor. Address reprint requests to Dr. Gregg T. Nagle, The Marine Biomed- ical Institute, 200 University Boulevard, University of Texas Medical Branch, Galveston, TX 77550. Abbreviations: A-NTP, postsignal sequence NH;-terminal peptide encoded by the A-related gene; BCP, bag-cell peptide; B-NTP, postsig- nal sequence NH;-terminal peptide encoded by the B-related gene; CDCH. caudodorsal cell hormone; ELH, egg-laying hormone; AP, bag- cell acidic peptide; A-AP, an acidic peptide encoded by the A gene; A- ELH, an ELH-related peptide encoded by the A gene; HPLC, high- performance liquid chromatography; HRBP, histidme-rich basic pep- tide; PTH, phenylthiohydantom; TFA, trifluoroacetic acid. Introduction The control of egg-laying behavior in the marine mol- lusc Aplysia has been particularly amenable to a multi- disciplinary approach aimed at clarifying the cellular and molecular bases of neuroendocrine function. The neuro- secretory bag cells are part of the final common pathway leading to egg deposition in these animals. Egg laying is initiated when the bag cells, which are located in the ab- dominal ganglion (Coggeshall, 1967;Frazierrttf/., 1967) and are normally electrically silent, begin to fire in a pro- longed and synchronous "afterdischarge" that may last 30 min or longer (Kupfermann and Kandel, 1970; Du- dek and Blankenship, 1977a, b; Pinsker and Dudek, 1977). Several peptides, including the egg-laying hor- mone (ELH) (Table I), are released during this activity (Stuart et a/., 1980); they may act as classical neurohor- mones on peripheral targets, such as the ovotestis, or as non-synaptic neurotransmitters within the abdominal ganglion. The peptides released by the bag cells are encoded by a single gene, known as the ELH gene, which directs the synthesis of a 37-kDa polyprotein precursor (Scheller et a/., 1 983). A schematic diagram of the precursor (prepro- ELH), as modified by recent peptide studies, is presented in Figure 1. Processing signals in the predicted amino acid sequence of preproELH suggest that it is post-trans- lationally processed to generate nine or more peptide products in addition to the signal sequence. But is this processing scheme actually followed? Seven of the nine peptides have now been identified in bag-cell extracts or releasates (Table II), and each of them, with the exception of alpha- and delta-bag-cell peptides («-BCP, 5-BCP), has corresponded to a pre- dicted product of the precursor. The sequence of 5-BCP 210 MOLLUSCAN EGG-LAYING HORMONES 211 is unusual in two respects. First, it contains the only diba- sic sequence in the precursor that does not appear to be cleaved during post-translational processing. Second, it is liberated from the processing intermediate (and from rt-BCP, which occurs next to it on the precursor) by hy- drolysis at a single arginyl residue, the only monobasic site to be cleaved during processing. Of course, this cleav- age might have been predicted, since the region sur- rounding this arginyl residue has sequence characteris- tics that are often observed in association with cleaved monobasic sites (Benoit et a/., 1987). Nonetheless, as a result of these unexpected processing events, 5-BCP is 39 rather than 7 residues in length, and «-BCP is 9 rather than 40 residues in length. These observations empha- size the importance of chemically identifying peptide products — i.e., of validating the processing steps that are merely predicted from precursor sequences which are, in turn, predicted from nucleotide sequence analyses of genomic or cDNA clones. The physiological functions of most of the bag-cell products remain unclear, but their chemical properties (e.g., molecular weight, NH2-andCOOH-terminal mod- ifications) provide important clues as to their stability in biological fluids and thus to the kinds of function(s) that each could serve. Alpha-BCP. for example, probably could not act as a classical neurohormone because of its small size and lack of NH2- or COOH-terminal modifi- cation (Mayeri and Rothman, 1985; Rothman et a/., 1987); it is, in fact, rapidly degraded upon release into the extracellular space. On the other hand, «-BCP could, and presumably does, act as a non-synaptic neurotrans- mitter within the abdominal ganglion (Rothman et ai. 1983; Mayeri et ai. 1985; Brown and Mayeri, 1986; Sigvardttva/.. 1986). In this report, we describe the chemical characteriza- tion of epsilon-bag-cell peptide (e-BCP), the seventh pep- tide product of the ELH gene to be identified. We also review the chemical characteristics of the ELH peptide family, since this peptide is the most frequently analyzed product of ELH-related genes and it has a well-defined physiological function (or activity) — i.e., the release of mature oocytes from the ovotestis into the ducts of the reproductive tract. Materials and Methods Bag-cell clusters and the proximal 1 cm of the pleuro- visceral connectives were removed from A. californica and immediately stored at — 70°C until they were used. Clusters from 50 animals were heated for 3 min at 1 00°C and homogenized for 1 min (Brinkmann Polytron: set- ting 4; 4°C) in 15 ml of 1 M acetic acid containing 20 mM HC1. The extract was centrifuged (48,000 X g) for 20 min at 4°C, and the supernatant chromatographed at 4°C on a Sephadex G-50 superfine column (2.5 cm X 50 cm), which had previously been calibrated with molecu- lar weight standards. Fractions were pooled based on ab- sorbance at 280 nm, filtered (0.2 nm pore size), and the filtrate applied to a Supelcosil Cl 8 reversed-phase HPLC column without prior Sep-Pak purification or lyophiliza- tion. The column was washed until the absorbance at 220 nm returned to baseline, and was then eluted at a flow rate of 1 .0 ml/min with a gradient of Solvent A (0.1% TFA) and Solvent B (acetonitrile containing 0.1% TFA). One-minute fractions ( 1 .0 ml) were pooled based on absorbance, and were subjected to amino acid analy- sis and automated amino acid sequence analysis. Sam- ples were hydrolyzed with 5.7 N HC1 in vacua at 107°C for 24 h, and amino acid compositional analyses were carried out on a Beckman 6300 analyzer. The primary structure of the peptide was determined by micro- sequence analysis using an Applied Biosystems Model 475 A Protein/Peptide Sequencer with an on-line 120A microbore PTH analyzer and a Model 900 data proces- sor. The repetitive yield was 86.5%. Results An acid extract of heat-treated bag cells was initially chromatographed on a Sephadex G-50 column. The 0-10-kDa peptides in fraction A, a region of low absor- bance at 280 nm (not shown), were filtered and fraction- ated by C18 reversed-phase HPLC using shallow gradi- ent conditions to optimize peptide separation. The elu- ate was monitored at 220 nm (Fig. 2). Several fractions occurred as relatively broad peaks, primarily due to the shallow gradient conditions employed; however, the rel- atively low HPLC flow rate and the large amounts of sample loaded onto the column probably also contrib- uted to peak broadening. Fraction AI, which eluted from 159 through 162 min (Fig. 2), had the following amino acid composition: Thr (0.9), Ser (6.6), Glu (2.0), Pro (0.7), Gly ( 1 .8), Val (0.8), He (1.0), Leu (3.8). A comparison of this composition with the reported nucleotide sequence analysis of the ELH gene (Scheller et ai. 1983) suggested that fraction AI was a 19-residue peptide from the ELH precursor. Fraction AI was subjected to quantitative micro- sequence analysis (10.0 nmol; 20 cycles), and the result- ing amino acid sequence (Table III) was identical to that predicted for residues 156 through 174 of proELH (Scheller eta/.. 1983). Following convention, this 19-res- idue peptide was termed e-BCP. Assuming that the COOH terminus of e-BCP was not amidated (since the last residue was not followed by Gly in the precursor), the calculated Mr is 1863. Approximately 2.6 j*g (1.39 212 G. T. NAGLE ET AL. Table I Comparison of the primary structures of bag cell, arid gland and caudodorsal cell peptides3 ELH-related peptides 1 10 12 13 14 15 16 17 Aplysia califarnica ELH h He Ser lie Asn Gin Asp Leu Lys Ala He Thr Asp Met Leu Leu Thr Glu Aplysia brasi/iana ELHC He Ser He Asn Gin Asp Leu Lys Ala lie Thr Asp Met Leu Leu Thr Glu Aplysia californica [Gln:3,Ala:7]A-ELHd lie Ser He Asn Gin Asp Leu Lys Ala He Thr Asp Met Leu Leu Thr Glu Aplysia californica [Ala"]A-ELHd He Ser lie Asn Gin Asp Leu Lys Ala He Thr Asp Met Leu Leu Thr Glu Aplvsia californica A-ELH" He Ser He Asn Gin Asp Leu Lys Ala He Thr Asp Met Leu Leu Thr Glu Aplysia pami/a ELHe He Ser He Asn Gin Asp Leu Lys Ala lie Ala Asp Met Leu He Val Glu Lymnaea stagnalis CDCH' Leu Ser He Thr Asn Asp Leu Arg Ala He Ala Asp Ser Tyr Leu Tyr Asp a Boxed residues indicate positions where the peptides differ from A californica ELH. All of the ELH-related peptides are presumed to be amidated based on molecular genetic (Mahon el al. 1985; Nambu and Scheller. 1986; Scheller el ai. 1983; Shyamala el at.. 1986) and peptide studies (Chiuc/ al.. 1979; Ebbennk etal.. 1985). h Determined by Chiu el al. (1979). c Determined by Nagle et al. ( 1988b). J Determined by Nagle et al. ( 1986) and Rothman et al. (1986). c Predicted from nucleotide sequence analysis of an. 4. pamda bag cell genomic ELH clone (Nambu and Scheller, 1986). ' Determined by Ebberink et al. (1985). nmol) of f-BCP was recovered from each pair of bag-cell clusters. The physiological function of this peptide is not known. The prominent fraction which eluted from 1 1 5 through 124 min(Fig. 2) was also examined. Aminoacid compositional and sequence analyses demonstrated that this fraction corresponded to the myoactive 43-residue histidine-rich basic peptide (HRBP) of neurons R3-R14 (Nagle et ai, 1989). A wealth of data, both anatomical (Frazier et al., 1967) and biochemical (Newcomb and Scheller, 1987), indicates that the R3-R14 neurons have axon terminals in the neurohemal region surrounding the bag cells. Discussion Chemical characterization of the peptide products pre- dicted from the ELH gene expressed in the bag cells pro- vides definitive information about post-translational processing in these model neuroendocrine cells. Of the seven peptides characterized to date, five are processed as predicted, in response to signals that have been empiri- cally determined from similar studies in other systems. However, two peptides comprising contiguous segments of preproELH, are not processed as initially predicted: a dibasic site in the middle of the 5-BCP sequence is not cleaved, while a monobasic Arg separating 5-BCP from «-BCP is. Examination of the characteristics of these sites in greater detail, and their comparison to qualita- tively similar sites in the precursor that are processed differently, may provide insights that will allow increas- ingly accurate predictions of post-translational process- ing events to be made in the future. Moreover, since the bag-cell ELH gene is only one of a small family of struc- turally related genes that are expressed in a tissue-specific manner in Aplysia (Scheller et al., 1983), peptide charac- terization studies may help to distinguish between gen- eral and tissue-specific processing events. Comparisons of specific peptide sequences (e.g., the ELH-related pep- tides) from different tissues and species may be equally useful for preliminary structure-activity analyses of pep- tide action. The atrial gland, an exocrine organ secreting into the oviduct of Aplysia (Arch etal., 1980; Beard et al., 1982; Painter et al., 1985), expresses several ELH-family genes (Scheller et ai. 1983; Mahon et ai. 1985). The peptide products of this gland pharmacologically elicit egg depo- sition when injected into a receptive animal (Arch et ai, 1978), but have no known physiological function inside the organism. Recent experiments suggest that the se- creted peptides may be deposited onto the egg cordon as it is transported through the oviduct, and may mediate the sexual and social behaviors often associated with egg laying and egg cordons ( Painter et ai. 1989). MOLLUSCAN EGG-LAYING HORMONES 213 Table I (Continued) 18 19 23 24 25 26 28 29 30 31 32 33 34 35 Gin He Arg Glu Arg Gin Arg Tyr Leu Ala Asp Leu Arg Gin Arg Leu Leu Glu Lys-NH2 Gin He Arg Glu Arg Gin Arg Tyr Leu Ala Asp Leu Arg Gin Arg Leu Leu Glu Lys-NH, Gin He Gin Ala Arg Gin Arg Cys Leu Ala Ala Leu Arg Gin Arg Leu Leu Asp Leu-NH: Gin He Gin Ala Arg Arg Arg Cys Leu Ala Ala Gin He Gin Ala Arg Arg Arg Cys Leu Asp Ala Gin Lys Gin Glu Arg Glu Lys Tyr Leu Ala Asp Leu Arg Gin Arg Leu Leu Asp Leu-NH2 Leu Arg Gin Arg Leu Leu Asp Leu-NH, Leu Arg Gin Arg Leu Leu Asn Lys-NH, Gin His Trp Leu Arg Glu Arg Gin Glu Glu Asn Leu Arg Arg Arg Phe Leu Glu Leu-NH: Each of the ELH-family genes expressed in the atrial gland encodes a large polyprotein precursor. Two of the precursors, preproA and preproB, are diagrammed in Figure 1 ; the diagrams are based on nucleotide sequence analyses of genomic and cDNA clones (Scheller et ai. 1983; Mahon et a/., 1985) and have been modified by peptide sequence analyses (Heller et al, 1980; Nagle et ai. 1986, 1988c; Rothman et al.. 1986). The products that have been isolated and chemically characterized are summarized in Table IV. preproELH Signal preproB preproA Signal A-NTP Figure 1. Schematic diagram of Aplysia calijornica preproELH, preproB, and preproA as predicted from nucleotide sequence analyses of genomic and cDNA clones (Scheller et al.. 1983; Mahon el ai. 1985) and modified by peptide studies (Nagle et al.. 1986, 1988c; Rothman et al.. 1986). Signal peptides are represented by hatched boxes. Peptides that have been identified in extracts or releasates are indicated by black boxes and have been labeled. Known or predicted mono-, di-, tri-, and tetrabasic cleavage sites, as well as the Gly-Lys-Arg signal for COOH-terminal amidation, are shown. Abbreviations are denned in a footnote on the first page of this paper. 214 G. T. NAGLE ET AL Table II Pepl ides derived from Aplysia californica bag-cell preproELH" Peptide Number Means of of Source identification residues Function or pharmacological action a-BCPb Extracts Sequence 7, 8, or 9 Inhibits LUQ cells Inhibits/excites bag cells 0-BCPe Releasates Comigration Excites L 1 , R 1 , and bag with Standard cells in abdominal ganglion VBCPC Releasates Comigration 5 Excites bag cells with Standard 5-BCP" Extracts Sequence 39 Stimulates Ca flux into mitochondria of albumen gland secretory cells e-BCP" Extracts Sequence 19 Not known ELH< Extracts Sequence 36 Induces egg release from gonad. excites R 1 5 and LLQ cells in abdominal ganglion and BI6 in buccal ganglion AP' Extracts Sequence 27 Not known * A schematic diagram of preproELH is presented in Figure I Abbreviations are denned in a footnote on the first page of this paper. "Rothmantvu/-. 1983. ' Rothman el al . 1985. dNaglert«/. I988a. 'Ch'metal , 1979. rSchellerrta/., 1983. The atrial gland ELH-related peptides are 36-residue peptides that are identical to bag-cell ELH at residues 1- 19 and at six of eight COOH-terminal residues (Table I); they are approximately equipotent to bag-cell ELH in eliciting egg deposition. Nevertheless, the atrial gland peptides differ from their homolog in the bag cells in sev- eral important characteristics. First, there is a potential tribasic cleavage site in A-ELH and [Ala:7]A-ELH which is missing from bag-cell ELH (Fig. 1; Table I). Second, each of the three atrial gland peptides is disulfide-bonded to an 1 8-residue acidic peptide, A-AP, that is located ad- jacent to it in the precursor. This linkage, through Cys25 of the ELH-related molecules, may sterically inhibit cleavage at the tribasic sequence (Arg"-Arg:3-Arg24) and explain why the site is not used during post-translational processing. The function of this heterodimeric complex is not known, but it is approximately equipotent to bag- cell ELH in eliciting egg deposition. Finally, a proportion of the A-AP/A-ELH and A-AP/[Ala:7]A-ELH com- plexes are further processed in the atrial gland by a renin- like enzyme, with cleavage occurring at the Leu'4-Leu15 and Leu"-Leu34 bonds of the ELH-related sequences. The Leu" -Leu '- bond in A-AP is not cleaved, however. 30 20 10 0 40 j o 30 « < 20 250 Time (mm) Figure 2. Reversed-phase HPLC purification of 0-lO-kDa peptides from Aplvsia californica bag cells. An extract of the bag cells was ini- tially fractionated by Sephadex G-50 column chromatography to gen- erate fraction A, which contained the 0-lO-kDa peptides (not shown). Fraction A was filtered and then subfractionated by C 1 8 reversed-phase HPLC using two linear gradients of 0. 1 % TFA and acetonitrile contain- ing 0. 1 % TFA (0-22% in 96 min, 22-44% in 3 10 min) to generate frac- tion A I . A single elution profile has been divided into two panels. Ab- breviations are defined in a footnote on the first page of this paper. perhaps due to differences in secondary structure that have been predicted to occur in these regions by Chou- Fasman analysis («-helix in the ELH-related peptides, but /5-sheet in A-AP) (Nagle et a!.. 1986). Cleavage of the Table III A utomated sequence analysis of fraction A I J Edman cycle Residue (pmol) Al 0" — (10000) 1 Ser (2819) 2 Val (3836) 3 Leu (5036) 4 Thr(3540) 5 Pro (4459) 6 Ser (2317) 7 Leu (3424) 8 Ser (1890) 9 Ser (1433) 10 Leu (1960) 11 Gly(1383) 12 Glu(1148) 13 Ser (632) 14 Leu (1029) 15 Glu(643) 16 Ser (342) 17 Gly (434) 18 He (358) 19 Ser (127) " Results obtained with gas-phase microsequencer. Phenylthiohy- dantoin ( PTH ) amino acids were quantitated by HPLC. h Initial amount of peptide (pmol) applied to sequencer. MOLLUSCAN EGG-LAYING HORMONES Table IV Peptides derived from the ELH-family genes expressed in the Aplysia californica a/rial gland'' 215 Peptide Method of identification Number of residues Pharmacological action A-NTP" Sequence 13 Not known B-NTPh Sequence 13 Ac Sequence 34 Bag-cell activation Bc Sequence 34 A-ELHdc Sequence 36 Egg release from [Ala:7]A-ELHde Sequence 36 ovotestis [Gln;-\ Ala-7]A-ELH"'e Sequence 36 A-ELH-U-14)' Composition 14 Not known A-ELH-( 15-36)" Sequence 22 A-ELH-(l-33)b Composition 33 [Ala:7]A-ELH-( 15-36)" Sequence 22 [Ala27]A-ELH-(l-33)b Composition 33 [Gin23, Ahr7]A-ELH-(16-36)h Sequence 21 A-APde Sequence 18 Not known aA schematic diagram of two of the polyprotein precursors, preproA and preproB, is presented in Figure 1. Abbreviations are denned in a footnote on the first page of this paper. Peptides were isolated from tissue extracts. "Naglerta/.. 1988c. c Heller el a/.. 1980. d Nagle tVtf/. 1986. e Rothman el al . 1986. 'Rothmanrtci/.. 1984. Leu'4-Leu15 bond of the ELH-related peptides abolishes egg-laying activity. Since these atrial gland peptides do not induce egg deposition in vivo, however, it is not clear whether this processing step represents an activation or inactivation with regard to their actual function. It is im- portant to note that the Leu'4-Leu15 and Leu"-Leu34 bonds of bag-cell ELH are not cleaved during processing, even though each is predicted to occur in an a-helical segment of the molecule. The renin-related proteolysis thus appears to be a tissue-specific processing event, and one that would not be predicted to occur based on nucle- otide sequence analyses of genomic or cDNA clones. In spite of the differences in their sequences, ELH and the atrial gland ELH-related peptides are predicted to have the same secondary structure: two regions of strong rt-helical potential (residues 6-21 and 26-36) separated by a 0-bend (residues 22-25) (Nagle el al., 1986). These observations suggest that ELH may be a U-shaped pep- tide, and that egg-laying activity in Aplysia may be corre- lated with conservation of the NH:- and COOH-termi- nal regions of the molecule. Two additional ELH-related sequences have been reported to date — the A. brasiliana bag-cell ELH, determined by direct chemical character- ization of the isolated peptide (Nagle el al.. 1988b), and the A. pamila bag-cell ELH, predicted from nucleotide sequence analyses of genomic clones (Nambu and Scheller, 1986). Both are 36 amino acids in length (Table I). The sequence of A. brasiliana ELH is identical to that of A. californica bag-cell ELH and so does not provide any further information about regions of the peptide im- portant for receptor recognition and activation. The A. pannila ELH, in contrast, is only 78% identical to A. cali- fornica bag-cell ELH, and it displays the same pattern of residue conservation observed with the A. californica peptides: it is identical to all five sequenced peptides at 13 of 14 NH:-terminal positions and at 6 of 8 COOH- terminal positions, but differs significantly from them in the intervening region (Table I). More detailed structure-activity information has been obtained recently using synthetic ELH and analogs to in- duce egg deposition (Strumwasser el al., 1987). These studies confirm that peptide chain-length is important for egg-laying activity. Removal of the NH2-terminal He, or extension of the COOH terminus by one residue (ELH-Gly37), results in a loss of egg-laying activity. In contrast, ELH-( 1-34) and ELH-( 1-35) are at least mod- erately active, indicating that the COOH-terminal amide is not essential for biological activity, and that the identi- ties of the amino acids at positions 35 and 36 may be relatively unimportant. This conclusion is consistent with the comparative peptide studies outlined above, since substitutions at positions 35 and 36 of the ELH- 216 G. T. NAGLE ET AL. related peptides did not significantly decrease biological activity relative ioA. californica bag-cell ELH (Table I). Other positional requirements are less clear, however. The amino acids critical for receptor recognition and ac- tivation are probably concentrated at positions 1-10 and 29-34, since all six Aplysia ELH-related sequences are identical at these positions. If we extend the comparison to include the caudodorsal cell hormone (CDCH) of the freshwater pulmonate Lymnaea stagnalis (Table I; Eb- berink et a!.. 1985), an interesting pattern emerges. (CDCH is a 36-residue peptide secreted by the neuroen- docrine caudodorsal cells during a burst of activity com- parable to a bag-cell afterdischarge; it induces ovulation and is homologous to ELH.) Only 13 amino acid resi- dues are conserved in all seven peptides, and 10 of the 1 3 (77%) occur in the regions encompassing residues 1-10 and 29-34. The number increases to 11 of 13 if we in- clude positions 1 1 and 12 in the comparison, and four of them are charged (Asp6, Asp12, Arg30, Arg3J). These observations strengthen the notion that the regions en- compassing residues 1-12 and 29-34 may be important for receptor recognition and activation, and suggest that these positions should be modified in ELH analogs for future structure-activity studies. Acknowledgments This investigation was supported by NSF BNS 85 1 7575 (GTN) and BBS 87 1 1 368, and by NIH NS 22079 (SDP), NS 23169 (JEB), and NS 1 1255. Literature Cited Arch, S., J. Lupatkin, T. Smock, and M. Beard. 1980. Evidence for an exocnne function of the Aplysia atnal gland. J. Comp. Physiol. 141:131-137. Arch, S., T. Smock, R. Gurvis, and C. McCarthy. 1978. Atrial gland induction of the egg laying response in Aplysia californica. J Comp. Physiol. 128:67-70. Beard, M., L. Millecchia, C. Masuoka, and S. Arch. 1982. Ultra- structure of secretion in the atrial gland of a mollusc (Aplysia). Tis- sue Cell 14:297-308. Benoit, R., N. Ling, and F. Esch. 1987. A new prosomatostatin-de- rived peptide reveals a pattern for prohormone cleavage at monoba- sic sites. Science 238: 1 1 26- 1 1 29. Brown, R. O., and E. Mayeri. 1986. Evidence for excitatory auto- transmission in Aplysia bag cell neurons mediated by alpha-, beta-, and gamma-bag cell peptides. Soc. Neiirosci. Ahstr. 12: 946. Chiu, A. Y., M. W. Hunkapiller, E. Heller, D. K. Stuart, L. E. Hood, and F. Strumwasser. 1979. Purification and primary structure of the neuropeptide egg-laying hormone of Aplysia californica. Proc. Nati Acad. Sci. USA 76: 6656-6660. Coggeshall, R. E. 1967. A light and electron microscope study of the abdominal ganglion of Aplysia californica. J. Neurophvsioi 30: 1263-1287. Dudek, F. E., and J. E. Blankenship. 1977a. Neuroendocrine cells of Aplysia brasiliana. I. Bag cell action potentials and afterdischarge. J. Neurophysiol 40: 1 30 1 - 1 3 1 1 . Dudek, F. E., and J. E. Blankenship. 1977b. Neuroendocrine cells of Aplysia brasiliana. II. Bag cell prepotentials and potentiation. J. Neurophysiol 40: 1312-1324. Ebberink, R. H. M., H. van Loenhout, W. P. M. Geraerts, and J. Joosse. 1985. Purification and amino acid sequence of the ovula- tion neurohormone of Lymnaea stagnalis. Proc. Nail. Acad. Sci. USA 82: 7767-7771. Frazier, W. T., E. R. Kandel, I. Kupfermann, R. Waziri, and R. E. Cog- geshall. 1967. Morphological and functional properties of identi- fied neurons in the abdominal ganglion of Aplysia californica. J Neurophysiol. 30: 1288-1351. Heller, E., L. K. Kaczmarek, M. W. Hunkapiller, L. E. Hood, and F. Strumwasser. 1980. Purification and primary structure of two neuroactive peptides that cause bag cell afterdischarge and egg-lay- ing in Aplysia. Proc. Nail. Acad. Sci. USA 77: 2328-2332. Kupfermann, I., and E. R. kandel. 1970. Electrophysiological proper- ties and functional interconnections of two symmetrical neurose- cretory clusters (bag cells) in abdominal ganglion of Aplysia. J. Neu- rophysiol. 33: 865-876. Mahon, A. C., J. R. Nambu, R. Taussig, M. Shyamala, A. Roach, and R. H. Scheller. 1985. Structure and expression of the egg-laying hormone gene family in Aplysia. J. Neiirosci. 5: 1 872- 1 880. Mayeri, E., and B. S. Rothman. 1985. Neuropeptides and the control of egg-laying behavior in Aplysia. Pp. 285-301 in Model Neural Networks and Behavior. A. I. Selverston, ed. Plenum, New York. Mayeri, E., B. S. Rothman, P. H. Brownell, W. D. Branton, and L. Padgett. 1985. Non-synaptic characteristics of neurotransmission mediated by egg-laying hormone in the abdominal ganglion of Aplysia. J Neiirosci. 5: 2060-2077. Nagle, G. T., M. de Jong-Brink, S. D. Painter, J. E. Blankenship, and A. Kurosky. 1988a. Purification and structure of the delta and ep- silon bag cell peptides of Aplysia. Soc. Neiirosci. Abstr. 14: 1 76. Nagle, G. T., S. L. Knock, S. D. Painter, J. E. Blankenship, R. R. Fritz, and A. Kurosky. 1989. \.AplysiacaIifornicaneuTonsR3-Rl4: pri- mary structure of the myoactive histidine-nch basic peptide and peptide I. Peptides (in press). Nagle, G. T., S. D. Painter, J. E. Blankenship, J. V. A. Choate and A. Kurosky. 1988b. The bag cell egg-laying hormones of Aplysia brasiliana and Aplysia californica are identical. Peptides 9: 867- 872. Nagle, G. T., S. D. Painter, J. E. Blankenship, J. D. Dixon and A. Kurosky. 1986. Evidence for the expression of three genes encod- ing homologous atrial gland peptides that cause egg laying in Aplysia. J. Bioi Client 261: 7853-7859. Nagle, G. T., S. D. Painter, J. E. Blankenship and A. Kurosky. 1988c. Proteolytic processing of egg-laying hormone-related pre- cursors in Aplysia: identification of peptide regions critical for bio- logical activity. J. Bioi Client 263: 9223-9237. Nambu, J. R., and R. H. Scheller. 1986. Egg-laying hormone genes of Aplysia: evolution ol the ELH gene family. / Neiirosci. 6: 2026- 2036. Newcomb, R., and R. H. Scheller. 1987. Proteolytic processing of the Aplysia egg-laying hormone and R3-R14 neuropeptide precursors. J. Neiirosci. 1: 854-863. Painter, S. D., A. R. Gustavson, V. K. Kalman, G. T. Nagle, and J. E. Blankenship. 1989. Induction of copulatory behavior in Aplysia: atrial gland factors mimic the excitatory effects of freshly deposited egg cordons. Behav. New. Bioi 51: 222-236. Painter, S. D., V. K. Kalman, G. T. Nagle, R. A. Zuckerman, and J. E. Blankenship. 1985. The anatomy and functional morphology of the large hermaphroditic duct of three species of Aplysia. with spe- cial reference to the atrial gland. J. Morphoi 18: 167-194. MOLLUSCAN EGG-LAYING HORMONES 217 Pinsker, H. M., and F. E. Dudek. 1977. Bag cell control of egg laying in freely behaving Aplysia. Science 197: 490-493. Rothman, B. S., D. H. Hawke, R. O. Brown, T. D. Lee, A. A. Dehghan, J. E. Snivel)1, and E. Mayeri. 1986. Isolation and primary struc- ture of the califins, three biologically active egg-laying hormone-like peptides from the atrial gland of Aplysia californica. J. Bio/. Chem. 261: 1616-1623. Rothman, B. S., E. Mayeri, R. O. Brown, P.-M. Yuan, and J. E. Shively. 1983. Primary structure and neuronal effects of «-bag cell peptide, a second candidate neurotransmitter encoded by a sin- gle gene in bag cell neurons of Aplysia. Proc. Natl. Acad. Sci. USA 80: 5753-5757. Rothman, B. S., G. A. Phares, and T. A. Groves. 1987. Inactivation of alpha-bag cell peptide in the abdominal ganglion of Aplysia. Soc. Neurosci. Abslr. 13: 39. Rothman, B. S., J. Shively, D. Hawkes, R. O. Brown, and E. Mayeri. 1984. Two neuroactive peptides from a common precursor in Aplysia atrial gland. Trans. Am. Soc. Neurochem. 15: 1 19. Rothman, B. S., K. A. Sigvardt, and E. Mayeri. 1985. Co-release of five peptides, ELH, AP, «-, /}-, and T-BCP, derived from a common precursor protein of the bag cells of Aplysia. Soc. Neurosci. Abstr. 11:482. Scheller, R. H., J. F. Jackson, L. B. McAllister, B. S. Rothman, E. Mayeri, and R. Axel. 1983. A single gene encodes multiple neuro- peptides mediating a stereotyped behavior. Cell 32: 7-22. Shyamala, M., J. R. Nambu, and R. H. Scheller. 1986. Expression of the egg-laying hormone gene family in the head ganglia ofAplvsia. Brain Res 371:49-57. Sigvardt, K. A., B. S. Rothman, R. O. Brown and E. Mayeri. 1986. The bag cells of Aplysia as a multitransmitter system: iden- tification of alpha bag cell peptide as a second neurotransmitter. J. Neurosci. 6:803-813. Strumwasser, F., D. L. Schiller, and S. B. H. Kent. 1987. Synthetic neuropeptide egg-laying hormone (ELH) of Aplysia californica in- duces normal egg-laying: structure-activity studies. Soc. Neurosci. Abstr. 13: 38. Stuart, D. K., A. Y. Chiu, and F. Strumwasser. 1980. Neurosecretion of egg-laying hormone and other peptides from electrically active bag cell neurons of Aplysia. J Neurophysiol. 43: 488-498. Reference: Bio/. Bull. 177: 218-224. (October, 1989) Adipokinetic Hormones: Functions and Structures GRAHAM GOLDSWORTHY1* AND WILLIAM MORDUE2 1 Department of Biology, Birkbeck College, University of London, England, and -Department of Zoology, University of Aberdeen, Scot/and Introduction The AKH/RPCH family of arthropod neuropeptides comprises at least eleven (see Table I) members at the time of writing. Many of these peptides were identified initially by their pharmacological activities in causing hyperlipemia in locusts and/or hypertrehalosemia in cockroaches, but in this account we will restrict our dis- cussion to physiological functions of the endogenous peptides of locusts, and attempt to relate differences in biological activity with the variations in amino acid se- quences seen in peptides from other insects. Functions of Adipokinetic Hormones in Locusts In locusts, two adipokinetic hormones are synthesized in the glandular lobes of the corpora cardiaca (see He- kimi and O'Shea, 1987): the decapeptide, AKH-I, and either one (depending upon species) of two octapeptides, AKH-IIL or AKH-IIS (see Goldsworthy and Wheeler, 1989). These peptides are released during flight, and the release may be under octopaminergic control (Orchard, 1987; but see Konings et ai. 1988) from secretomotor centers in the brain (Rademakkers, 1977). The decapep- tide and the octapeptides can each stimulate the release of diacylglycerols from the fat body into the hemolymph, and thus provide fuel for migratory flight. These neuro- peptides also bring about a re-organization of circulating hemolymph lipoproteins and activate fat body glycogen phosphorylase. In all these activities, AKH-I is more po- tent than AKH-II (Goldsworthy et ai, 1986a, b). How- ever, although both peptides are thought to act on the fat body via adenylate cyclase, AKH-II increases fat body cAMP levels to a greater extent than AKH-I (Goldswor- thy et ai, 1986a), suggesting that they may act at difter- * Address for correspondence: Department of Biology. Birkbeck Col- lege, Malet Street, London WCI E 7HX, U. R. ent receptors and that AKH-II may have other, as yet undiscovered, actions. Actions on the fat body Adipokinetic peptides stimulate adenylate cyclase ac- tivity in the fat body. This activation, by analogy to that of triacylglycerol lipase in vertebrates, is generally as- sumed to involve a conventional protein kinase cascade (see Goldsworthy, 1983; Beenakkers et al., 1985; Or- chard, 1987). Ca2+, acting as another second messenger, may also be involved because the lipid-mobilizing action of AKH on the fat body is dependent on Ca2+ in vitro. Activation of the triacylglycerol lipase is thought to hy- drolyze triacylglycerols to monoacylglycerols, which are subsequently re-acylated by a monoacylglycerol transfer- ase to form sterospecific sn-1.2 diacylglycerols, and are released from the fat body (see Beenakkers et al., 1985) as part of a special carrier lipoprotein (see below). Fuel transport to the flight muscles Neutral lipids are essentially insoluble in water and, as in vertebrate blood, their transport in hemolymph in- volves carrier lipoproteins; these change in composition during flight (Mayer and Candy, 1967) or after injection of AKH, and have been studied extensively by several groups (see Goldsworthy, 1983; Beenakkers et al., 1985; Wheeler, 1989). Adipokinetic hormones cause a re- grouping of hemolymph proteins and lipoproteins with a consequent increase in the lipid-carrying capacity of the hemolymph. Thus, a new 'activated' lipoprotein complex, or A" (see Fig. 1 ), forms in the hemolymph. Compared with the lipoprotein present at rest, Aye/low, A1 particles are of larger diameter (Wheeler et al., 1984a), they bind large quantities of other hemolymph proteins called CL-apoproteins (Wheeler and Goldswor- thy, 1983a, b) to them, and they carry up to 18 times 218 THE AK.H FAMILY 219 Figure 1. Diagrammatic scheme of the major actions of AK.H in locusts. Diacylglycerols(DGL), made available by the action AK.H in increasing triacylglycerol lipase in the fat body, are loaded onto lipoprotein Ayf//mi'(Ay), which forms lipoprotein A* which reversibly binds C, -proteins. At the flight muscle, diacyl- glycerols associated with lipoprotein A* are hydrolyzed, and the fatty acids (FFA) are used to support flight muscle metabolism. Lipid unloading leads to the liberation of free CL-proteins. The lipoproteins thus form a re-usable 'shuttle,' carrying energy to the flight muscles. more neutral lipid (see Goldsworthy, 1983). The forma- tion of A+ lipoprotein does not require protein synthesis de novo and is reversible: the lipoproteins act as re-usable shuttles, carrying diacylglycerols from the fat body to the flight muscles. In the first author's laboratory, two Ci_- apoproteins are recognized in the hemolymph of Lo- custa (Goldsworthy et ai. 1985), each of which bind to A+( Wheeler and Goldsworthy, 1986; Wheeler, 1989). In other laboratories, only a single locust apoprotein has been described (see Beenakkers et al, 1985). Indirect actions of AKH at the flight muscles The unloading of diacylglycerol from A+ lipoprotein particles at the flight muscle is also regulated by AKH, but in this case the control is indirect. A lipoprotein li- pase, membrane-bound in the flight muscles, hydrolyses lipids of lipoprotein A+ at over four times the rate of those ofAyellow. and enzyme activity against Aye/low is reduced by about 90% in the presence of lipoprotein A+ (Wheeler et al., 1984b; 1986; Van Heusden et al., 1986). The CL-apoproteins inhibit flight muscle lipoprotein li- pase in a competitive manner at low concentrations, but this becomes mixed at higher concentrations of the apo- proteins; both CL-I and CL-II apoproteins are effective (Wheeler e/fl/., 1986; Wheeler and Goldsworthy, 1986). This is the basis of the mechanism by which AKH con- trols lipoprotein lipase indirectly during rest and flight (Wheeler and Goldsworthy, 1985). That is, when the free (i.e., not lipoprotein-bound) Q -apoprotein concentra- tion in the hemolymph is high, as it is in resting locusts, lipase activity is inhibited; but during flight, or when AKH is injected into resting locusts, CL-apoprotein inhi- bition of the lipase is removed by the decrease in the con- centration of free CL-apoproteins as they bind to lipopro- tein A+. This regulation of lipoprotein lipase by the hemo- lymph concentration of free CL-apoprotein ensures rapid uptake of lipid by the flight muscles when A+ is present in high concentrations. But more important perhaps, to prevent the flight muscles from 'stalling,' the inhibition of trehalose oxidation by products of lipid oxidation oc- curs only when lipoprotein A"* (the preferred substrate) is present in sufficient quantities to sustain flight. Histochemical staining for lipoprotein lipase in ultra- thin sections oflocust flight muscle confirms that the en- zyme is membrane-bound, but shows also that activity is restricted to the T-tubules; therefore, the enzyme has ready access to hemolymph-borne lipoproteins. Inhibi- tion of enzyme activity in vitro by free CL-apoproteins, and activation after injection of AKH into donor locusts, can also be demonstrated by histochemical staining of 220 G. GOLDSWORTHY AND W. MORDUE membrane preparations from flight muscles (Wheeler, 1989). Direct actions of A KH at the flight muscles Rates of fuel use in locusts flying under laboratory conditions can be assessed from changes in the concen- trations of hemolymph metabolites; such data provide substantial circumstantial evidence that, at least during the first moments of flight, trehalose and diacylglycerols compete as fuels for the flight muscles (see Goldsworthy, 1983). Fuel use by the flight muscles has been studied directly in half-thorax preparations of Locusta (Robinson and Goldsworthy, 1977). Trehalose use decreases to about 50% on addition of substrates containing hemolymph li- poproteins. This inhibition of substrate use is competi- tive, but in the presence of extracts of the corpora cardi- aca containing AKH, or purified natural AKH-I, it be- comes non-competitive. In addition, in these half thorax preparations AKH increases the rate of oxidation of lip- ids contained in lipoprotein preparations from AKH-in- jected locusts (see Goldsworthy, 1983). Inactivation and excretion oj AKH The half life of AKH-I in the hemolymph of locusts is about 30 min (Cheeseman and Goldsworthy, 1979) but, while the hemolymph of some insects contains enzymes which degrade neuropeptides like AKH, locust hemo- lymph does not apparently contain AKH-degrading en- zymes; the blocked peptides would presumably have some protection against non-specific peptidases. Work in Aberdeen (Mordue and Stone, 1978; Siegert and Mordue, 1987) shows that the Malpighian tubules of lo- custs are a potent source of degradative enzymes active against AKH-I, but we cannot be sure that inactivation by the tubules represents the major mechanism by which the AKH signal is removed: the CNS contains similar enzymes to those in the Malpighian tubules. Siegert and Mordue (1987) have shown that AKH-I can be hydrolyzed by an endopeptidase from the Mal- pighian tubules. When AKH-I adopts a B-bend confor- mation (see below), the Asn at residue 7 could perhaps be relatively more susceptible to enzymatic attack. Anal- ysis of the breakdown products obtained from incubat- ing AKH-I with an homogenate of Malpighian tubules did not identify Asn, but three breakdown products pres- ent in small amounts could not be identified fully, and Asn may be contained in these. Three major breakdown products were identified, however: pGlu-Leu-Asn-Phe- Thr-Pro; Trp-Gly-Thr-NH;; and tryptophan. Siegert and Mordue (1987) suggest that at least three proteolytic en- zymes are present in the tubules: an endopeptidase cleav- ing at position 6-7 or 7-8; a carboxypeptidase that at- tacks the pGlu . . . Pro fragment at its (now) free C-ter- minus; and an aminopeptidase (during incubation the fragment Trp-Gly-Thr-NH: disappears; certainly a po- tent leucine aminopeptidase is present in high activity in tubules). The carboxypeptidase differs from carboxypep- tidase A, because the latter did not attack the pGlu . . . Pro fragment. Certainly a carboxypeptidase and an ami- nopeptidase are present, but they are not responsible for the initial cleavage of AKH-I. There are no data for the breakdown of other AKH-related peptides in locusts. We assume that AKH enters the Malpighian tubules passively and is then degraded. Molecules with larger molecular weights than AKH, such as inulin, readily per- meate the tubule membrane. The relatively low molecu- lar mass of AKH-I ( 1 158), in combination with its lipo- philic nature may mean that AKH readily enters the Malpighian tubule cells. Adipokinetic Structure-Activity Requirements Studying a series of synthetic peptide analogs of AKH- I, Stone el al. ( 1978) concluded that, for biological activ- ity, peptides must be blocked by L-pyroglutamatic acid (cyclized L-glutamic acid) at the N-terminal. and by ami- dation at the C-terminal; they must be between 8 and 10 amino acids long; and substitutions in the area of resi- dues 6 to 8, which were assumed to interfere with a pro- posed 5-turn around residue 6 (proline), reduced biolog- ical activity. Recently the biological activities of natu- rally occurring analogs of AKH found in different insects have been examined (Goldsworthy et al., 1986a, b; Wheeler et al.. 1988). While sufficiently high doses of most peptides can elicit full (maximum possible) re- sponses, others elicit only attenuated adipokinetic (see Goldsworthy and Wheeler, 1 986b) or hypertrehalosemic (Gade, 1986; Wheeler et al., 1988) responses. The sequences of naturally occurring members of the AKH family known presently are shown in Table I. In natural peptides of more than 8 residues, residue 10 is threonine and residue 9 is glycine. Residue 8 is trypto- phan in all peptides. Residue 7 is either asparagine or glycine, except in Afanduca-AK.il where it is serine. Residue 6 is proline in most of the natural peptides, the exceptions being the AKH-II's of locusts and of a grasshopper (Romalea), where it is alanine in Locusta, and threonine in Schistocerca and Romalea, and the AKH ofManduca in which it is serine. Residue 5 is either threonine or serine. In all peptides, residue 4 is phenylal- anine. Residue 3 is either asparagine or threonine. Resi- due 2 is either leucine or valine. Residue 1 is always pyro- glutamate. The potencies of drugs and hormones can be repre- sented by their ED50 values — the doses needed to bring about a half-maximal response (see Table I). In the first THE AK.H FAMILY Table I The relative potencies of members of the AKH /RPCH family of arthropod neuropeptides in causing hyperlipaemia in Locusta and. where indicated, in Schistocerca 221 Neuropeptide Amino acid residue 1234 5 6789 10 ED50 pmol EDmax °, pmol i. Maximum response CD evidence ofB-bend Locust pGlu-Leu-Asn-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH; 1 3 100 Yes AK.H-I (assayed in Schistocerca) 6 20 100 Carausius-H1¥-l\ pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH: 2 8 100 Yes Synthetic Leu-Thr-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH, Not active Yes Synthetic pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-Gly Not active Yes Manduca ^/Heliothis 4 AKH pGlu-Leu-Thr-Phe-Thr-Ser-Ser-Trp-Gly-NH: 10 40 45 No (20 40 100) Crustacean 5 pGlu-Leu-Asn-Phe-Ser-Pro-Gly-Trp-NH: 4.6 20 100 Yes RPCH (assayed in Schistocerca) 15 40 100 Locusta b pGlu-Leu-Asn-Phe-Ser-Ala-G!y-Trp-NH: 2 3 60 No AKH-II (assayed in Schistocerca) 20 30 55 Schistocerca 6 pGlu-Leu-Asn-Phe-Ser-Thr-Gly-Trp-NH: 2 5 95-100 No AKH-II (assayed in Schistocerca) 12 25 60 Periplanela1 M-I pGlu-Val-Asn-Phe-Ser-Pro-Asn-Trp-NH: 5 30 95-100 Yes Periplaneta^ M-II pGlu-Leu-Thr-Phe-Thr-Pro-Asn-Trp-NH, 5 20 95-100 Yes Neuphoeta "/ pGlu-Val-Asn-Phe-Ser-Pro-Gly-Trp-Gly-Thr-NH, (4-5 20 100) Yes Blabents HTH Romalea- 1 " pGlu-Val-Asn-Phe-Thr-Pro-Asn-Trp-Gly-Thr-NH, (2 10 95-100) Yes Romalea-ir/Grylhts™ pGlu-Val-Asn-Phe-Ser-Ala-Gly-Trp-NH, (4-5 20 100) No 1 Stone el al. (1978): : Gade and Rinehart ( 1987a); ' Ziegler el ai (1985): 4 Jaffe et al. (1986); 5 Fernlund and Josefsson (1972); " Gade et al. (1984; 1986); Siegert el ai ( 1985); 7 O'Shea et al. (1984); " Hayes and Keeley (1986); Gade and Rinehart (1986): ' Gade et al . (1988); '° Gade and Rinehart (1987b). The responses are calculated as a percentage of the maximum response obtained to a crude extract of Locusta corpora cardiaca (0.02 pair equivalents/locust) tested on the same batch of locusts on the same day (from Goldsworthy et al., 1986b, and unpub. obs.). In our hands, with our strains of locusts, Schistocerca appears very much less sensitive than Locusta to those peptides tested. The final column shows whether CD spectros- copy indicates the presence of a fi-bend (A. Drake, G. J. Goldsworthy, C. Wheeler, and G. Gade, unpub. obs.). Assay data in brackets from G. Gadefpers. comm.). author's colony of Locusta, AKH-I has an ED50 of 1 pmol, and at doses above 3 pmol gives 'full' adipokinetic activity (EDmax = 3 pmol). This represents the maximum rate of lipid mobilization possible in the locusts at the time of assay, and can be used as a comparator for other peptides. Carausiiis peptide (HTF-II) has only a single residue change of threonine instead of asparagine at posi- tion 3 and, although this doubles the ED50 from 1 to 2 pmol, 'full' adipokinetic activity is still shown, but the EDmax increases so doses in excess of 8 pmol are required for a 'full' response. From the data in Table I it seems that the various amino acid substitutions in the naturally occurring pep- tides RPCH, M-I, and M-II have little overall effect on hyperlipemic activity. However, these octapeptides all have 'weak' activity compared with AKH-I, with EDM> values around 5 pmol; they show 'full' adipokinetic ac- tivity, but their EDmax values are between 20 and 30 pmol. Octapeptides lacking proline are particularly interest- ing because of the prediction of a 5-turn at proline in AKH-I. Such predictions assess the statistical probability of there being a bend, calculated from the particular se- quence of amino acids present and known secondary structures in proteins (Chou and Fasman, 1974). It is problematic whether such predictions apply to short pep- tides. But when the calculations are applied to the AKH family, with the exception of those members lacking pro- line at residue 6, all are predicted with high probability to have a turn around residue 6. Using circular dichroism (CD) spectroscopy, we have shown at Birkbeck (A. Drake, G. J. Goldsworthy C. H. Wheeler and G. Gade, unpub. obs.) that, in aqueous so- lution at room temperature, none of the naturally occur- ring neuropeptides of this family has a CD spectrum characteristic of a 5-turn. However, on addition of SDS micelles or liposomes (which act as mimics of biological membranes in this system ), the CD spectra of all peptides containing proline change to ones characteristic of a type-I 5-turn (Goldsworthy and Wheeler, 1989). None 222 G. GOLDSWORTHY AND W. MORDUE of the naturally occurring locust or grasshopper octapep- tides, AKH-IIL, AKH-IIS, Ro-II, ortheManduca-AKH, show such changes in their CD spectra; they have no identifiable CD spectra with or without liposomes or SDS micelles (Table I). Those members of the family that show a 5-turn con- formation, can show 'full' hyperlipemic and hypertreha- losemic activity, when tested at sufficiently high doses. However, when tested intraspecifically, the octapeptides which we have tested, and for which the CD analysis shows no evidence of a 5-turn, all elicit attenuated hy- perlipemic and hypertrehalosemic responses. Unfortu- nately, however, the situation is not completely clear- cut: while AKH-IIL gives an attenuated response, even at doses greater than 50 pmol, whether tested in Locitsta or in Sc/iisiocerca, AKH-IIS gives an attenuated re- sponse when tested intraspecifically (in Schistocerca), but gives a full adipokinetic response in Locusta (Golds- worthy and Wheeler, 1986b). Apparently, contrary to our earlier conclusion (Goldsworthy et ai, 1986b), the presence of proline or of a B-ium is not essential for full adipokinetic activity. Apparently, AKH receptors on the fat body of Locusta can discriminate between the locust AKH-II octapep- tides, whereas those of Schistocerca cannot. When in- jected intraspecifically, the two locust AKH-II octapep- tides do not show 'full' activity; they give truncated re- sponses. AKH-IIL is potent in comparison with AKH-I (ED50 about 2 pmol), but only gives about 60% of the maximum response in Locusta even at doses up to 50 pmol. AKH-IIS1 is less potent in Schistocerca (ED50 = 1 2 pmol) and gives a truncated response. On the other hand, AKH-IIS is very potent in Locusta (ED50 = 2 pmol) and gives a 'full' adipokinetic response. Manditca (and Helio- this)-AKH is relatively inactive in Locusta, having an ED50 of 1 0 pmol and giving a truncated response even at doses up to 75 pmol. At Birkbeck, we have recently undertaken molecular modelling of members of the AKH family. A model of AKH-I has been generated using computer graphics. A type-I turn was built into the 10 residue AKH-I neuro- hormone using HYDRA on an Evans & Sutherland PS300 graphics system. The dihedral angles of the se- quence were set to B-conformation except that the cen- tral residues of the proposed turn were set to the ideal angles for a type-I 5-turn. Slight, by eye, adjustments produced a hydrophobic cluster of phenylalanine, tryp- tophan, and leucine on one side of the peptide. If, as the CD data suggests, the 5-turn is involved in receptor bind- ing conformation, then this hydrophobic face of the pep- tide may be a likely candidate for interaction with the receptor. Additional main-chain stabilizing hydrogen bonds were also built in (Leu2 main chain oxygen to Gly9 main chain NH; Thr5 main chain NH to Trp8 main chain oxygen; Thr5 main chain oxygen to Trp8 main chain NH; and Asn7 main chain NH to Thr5 side chain oxygen), and the structure was energy minimized with GROMOS. Our model differs in detail from the ear- lier one proposed by Mordue and his colleagues (Mordue and Morgan, 1985). Molecular dynamics suggested that the phenylalanine may flip to give a tighter hydrophobic cluster when interacting with the tryptophan ring. This hydrophobic cluster on one face of the molecule requires further investigation. Evolutionary Considerations The locust adipokinetic hormones and structurally similar peptides from other insects and Crustacea (Table I) are commonly referred to as a family of peptides. Un- derlying these references are suggestions that close evolu- tionary relationships exist between different arthropod groups. It will be necessary to analyze closely both mo- lecular evolution and links between and within the crus- tacea and the insects. It is apparent from the known structures that a number of core residues are strictly con- served in adipokinetic and related peptides: thus residues 1 , 4, and 8 are constant; and in nona- and deca- peptides Gly is residue 9; and in the three known decapeptides residue 10isThr-NH:. Even this limited degree of conservation gives a strong indication of family groups, and analysis of the variabil- ity in amino acid residues in other positions adds sub- stantially to these indications. It is essential, when deal- ing with molecular evolution of peptides, to give consid- eration to the codons that specify particular amino acids. The structure-activity relationships discussed above are integral to, and a manifestation of, the evolutionary pro- cess. We have already discussed the variations in sequence in members of the AKH/RPCH family. The precise co- dons used for each amino acid in the various insects are not yet known, but routes for single point mutations to change from one amino acid residue to another can be worked out. For example, with respect to Leu or Val at position two, there are three possible routes. The vari- ability at residue 5 is limited and involves only Thr and Ser, and for these single point mutations are also pos- sible. Proline is most common at residue 6, and is present in all the decapeptides; but as we have discussed above, ei- ther Thr, Ser, or Ala are present at this position in some of the natural peptides of this family. The four possible codon sequences for each amino acid residue (Pro: Thr: Ser: Ala) have identical second and third nucleotides. In consequence, single point mutations in the first nucleo- tide of each codon could allow mutations during evolu- tion. One concept of peptide-families necessitates a pri- THE AK.H FAMILY 223 mary peptide from which others have evolved. However, at present, we cannot predict with certainty which of the four residues at position 6 could have been present in an ancestral peptide. The first peptides to be characterized fully in this family were RPCH and AKH-I; the codon sequences for Gly and Asn, the residues present at posi- tion 7 in these two peptides respectively, indicate that single step mutations could not occur. Other peptides of the family, which were sequenced subsequently, also contain either Asn or Gly at residue 7 (see Table I); one possible interpretation of these findings was that two separate families existed. However, a candi- date link peptide exists in the adipokinetic hormone iso- lated from lepidoptera which has Ser at residue 7. Theo- retical single steps are possible in either direction be- tween peptides containing either Asn and Gly via the peptide found in Manduca (Ziegler el a/., 1985) or Heli- othis (Jaffe et al.. 1986). It must be emphasized that in our approach so far we have been concerned only with theoretical problems. Other factors have to be taken into consideration; for example, some arthropod biologists would, on a priori grounds, consider insects and crusta- ceans to belong to separate phyla, and therefore any commonality between RPCH and AKH-peptides would stem from a peptide present in pro-arthropod stock or simply be another example of convergence in evolution. Moreover, the highly specialized lepidoptera are most unlikely candidates to provide evolutionary linkage be- tween either Crustacea and insects on the one hand or between different insect groups on the other. Neverthe- less, in terms of molecular evolution within the insects, we can perhaps talk of a family of peptides, but it is un- fortunate that for the largest animal classes we are at- tempting to construct relationships on data from so few species and from very few insect orders. There is an ur- gent need for data on the composition of precursor mole- cules. The homologies present in the AKH family may reflect both modifications from a single ancestral pep- tide, and their processing from a common precursor. Are the AKH/RPCH peptides related to the leucoki- nins? The latter are part of a second insect peptide fam- ily, but have all been identified in a single species of cock- roach, Leucophaea maderae, by Holman and his col- leagues (1984, 1986, 1987). They have some sequence affinities with AKH and RPCH, but comparisons are not straight-forward. In the leucokinins for which structures are published, residues 4 (Phe); 6 (Ser); 7 (Trp); and 8 (Gly-NH2) are strictly conserved. If these are considered to be analogous to residue 1-8 in the octapeptides from the AKH series, then it is possible to consider possibili- ties of links between the leucokinins and peptides of the AKH/RPCH family. The possibilities for mutations are slight at residue 1 (using glutamine as the residue to be cyclized to form pyroglutamate) and at residues 7 and 8. However, rather than using Phe at residue 4 as the refer- ence point, the N-terminal Trp-Gly-NH: may be a more rational cross-reference point; in this situation there are many possible linkage points between the adipokinetic and leucokinin families. However, only when more is known of the structure of the precursor molecules con- cerned with the synthesis of insect peptide hormones, will a clearer evolutionary picture emerge. Literature Cited Beenakkers, A. M. Th., D. J. Van Der Horst, and W. J. A. Van Mar- rewijk. 1985. Biochemical processes directed to flight muscle me- tabolism. Pp. 451-486 in Comprehensive Insect Physiology, Bio- chemistry and Pharmacology, vol. 10. G. A. Kerkut and L. I. Gil- bert, eds. Pergamon Press, Oxford. Cheeseman, P., and G. J. Goldsworthy. 1979. The release of adipoki- netic hormone during flight and starvation in Locusta. Gen. Comp. Endocrinoi 37: 35-43. Chou, P. V., and G. D. Fasman. 1974. Prediction of protein confor- mation. Biochemistry 13: 222-245. Fernlund, P., and L. Josefsson. 1972. Crustacean color-change hor- mone: amino acid sequence and chemical synthesis. Science 177: 173-175. Gade, G. 1986. Relative hypertrehalosaemic activities of naturally occurring neuropeptides from the AKH/RPCH family. Z Nattir- forttVi. 41:315-320. Gade, G., and K. L. Rinehart. 1986. Amino acid sequence of a hyper- trehalosaemic neuropeptide from the corpus cardiacum of the cock- roach. Nauphoela cinerea. Biochem. Biophys. Res. Comm. 141: 774-781. Gade, G., and K. L. Rinehart. 1987a. Primary structure of the hyper- trehalosaemic factor II from the corpus cardiacum of the Indian stick insect. Cannisius morosus, determined by fast atom bombard- ment mass spectrometry. Bial. Chem. Hoppe-Seyler)68: 67-75. Gade, G., and K. L. Rinehart. 1987b. Primary sequence analysis by fast atom bombardment mass spectrometry of a peptide with ad- ipokinetic activity from the corpora cardiaca of the cricket Gryllus bimaculatus. Biochem. Biophys. Res. Comm 149: 908-914. Gade, G., G. J. Goldsworthy, G. Kegel, and R. Keller. 1984. Single step purification of locust adipokinetic hormones I and II by re- versed phase high-performance liquid chromatography, and amino acid composition of the hormone II. Hoppe-Seyler's Z. Physiol. Chem 365:391-398. Gade, G., G. J. Goldsworthy, M. H. Schaffer, J. C. Cook, and K. L. Rinehart. 1985. Sequence analyses of adipokinetic hormones II from corpora cardiaca of Schistocerca nitons, Schislocerca gregaria. and Locusta migraloria by fast atom bombardment mass spectrom- etry. Biochem Biophys. Res. Comm. 134: 723-730. Gade, G., C. Hilbich, K. Beyreuther, and K. L. Rinehart. 1988. Se- quence analyses of two neuropeptides of the AKH/RPCH-family from the lubber grasshopper. Romalea microptera. Peptides 9: 68 1 - 688. Goldsworthy, G. J. 1983. The endocrine control of flight metabolism in locusts. Pp. 149-204 in Advances in Insect Physiology. Vol. 17, M. J. Berridge, J. E. Treherne and V. B. Wigglesworth, eds. Aca- demic Press, New York. Goldsworthy, G. J., and C. H. Wheeler. 1986a. The endocrine con- trol of flight metabolism in locusts. Pp. 49-59 in Insect Flight Dis- persal and Migration, W. Danthanarayana, ed. Springer- Verlag, Heidelberg. Goldsworthy, G. J., and C. H. Wheeler. 1986b. Structure/activity re- lationships in the adipokinetic hormone/red pigment concentrating 224 G. GOLDSWORTHY AND W. MORDUE hormone family. Pp. 1 83- 1 86 in Insect Neurochemistry and Neuro- plivsiologv 1986, A. B. Borkovec and D. B. Gelman, eds. Humana Press, New Jersey. Goldsworthy, G. J., and C. H. Wheeler. 1989. Physiological and structural aspects of adipokinetic hormone function in locusts. Pes- ticide Chem. 25: S5-95. Goldsworthy, G. J., K. Mallison, and C. H. Wheeler. 1 986a. The rela- tive potencies of two known locust adipokinetic hormones. / Insect Physiol. 32:9S-W\. Goldsworthy, G. J., K. Mallison, C. H. Wheeler, and G. Gade. 1986b. Relative adipokinetic activities of members of the AKH/ RPCH family. J. Insect Physiol. 32: 433-438. Goldsworthy, G. J., A. R. Jutsum, and N. L. Robinson. 1979. Sub- strate utilisation and flight speed during tethered flight in the locust. / Insect Physiol. 25: 183-185. Goldsworthy, G. J., C. M. Miles, and C. H. Wheeler. 1985. Lipo- protein transformations during adipokinetic hormone action. Phys- iol. Eniomol. 10: 151-164. Hayes, T. K., and L. L. Keeley. 1986. Isolation and structure of the hypertrehalosaemic hormone from Blabcms discoidalis cock- roaches. Pp. 195-198 in Insect Neurochemistry and Neurophysiol- ogy 1986. A. B. Borkovec and D. B. Gelman. eds. Humana Press. New Jersey. Hekimi, S., and M. O'Shea. 1987. Identification and purification of two precursors of the insect neuropeptide adipokinetic hormone. / Nenrosci. 1: 2773-2784. Molman, G. M., B. J. Cook, and R. M. Wagner. 1984. Isolation and partial characterization of five myotropic peptides present in head extracts of the cockroach Leticophaea maderae. Comp. Biochem Physiol. 11C: 1-5. Holman, G. M., B. J. Cook, and R. J. Nachman. 1986. Isolation pri- mary structure and synthesis of two neuropeptides from Leuco- phaea maderae: members of a new family of cephalomyotropins. Comp. Biochem. Physiol. 84C: 271-276. Holman, G. M., B. J. Cook, and R. J. Nachman. 1987. Isolation, pri- mary structure and synthesis of leucokinins VII and VIII: the final members of this new family of cephalomyotropins isolated from head extracts of Leitcophaea maderae Comp. Biochem. Physiol. 88C31-34. Jaffe, H., A. K. Raina, C. T. Riley, B. A. Fraser, G. M. Holman, R. M. Wagner, R. L. Ridgway, and D. K. Hayes. 1986. Isolation and primary structure of a peptide from the corpora cardiaca ofHeli- othis :ea with adipokinetic activity. Biochem Biophys. Res. Comm. 135:622-628. Konings, P. N. M., H. G. B. Vullings, M. Geffard, R. M. Buijs, J.H. B. Diederen, and W. F. Jansen. 1988. Immunocytochemical demonstration of octopamine-immunoreactive cells in the nervous system of Locusta migratona and Schistocerca gregaria. Cell Tissue Res 251:371-379. Mayer, R. J., and D. J. Candy. 1967. Changes in haemolymph lipo- proteins during locust flight. Nature 2 1 5: 987. Mordue, W.. and P. J. Morgan. 1985. Chemistry of peptide hor- mones. Pp. 153-183 in Comprehensive Insect Physiology, Bio- chemistry and Pharmacology. Vol 7, G. A. Kerkut and L. I. Gilbert, eds. Pergamon Press, Oxford. Mordue, W., and J. V. Stone. 1978. Structure and metabolism of ad- ipokinetic hormone. Pp. 487-490 in Comparative Endocrinology. P. J. Gaillard and H. H. Boer, eds. Elsevier, Amsterdam. Mwangi, R. W., and G. J. Goldsworthy. 1977. Diglyceride-transport- ing lipoproteins in Locusta. J. Comp. Physiol. B. 114: 1 77- 1 90. O'Shea, M., J. Witten, and M. Schaffer. 1984. Isolation and charac- terization of two myoactive neuropeptides: further evidence of an invertebrate peptide family. Neuroscienc e 4: 52 1-529. Orchard, I. 1987. Adipokinetic hormones — an update. / Insect Physiol. 33:451-463. Rademakkers, L. H. P. M. 1977. Identification of a secretomotor centre in the brain of Locus/a migratoria, controlling the secretory activity of the adipokinetic hormone producing cells of the corpus cardiacum. Cell Tiss. Res. 184: 381-395. Robinson, N. L., and G. J. Goldsworthy. 1977. Adipokinetic hor- mone and the regulation of carbohydrate and lipid metabolism in a working flight muscle preparation. J. Insect Physiol. 23: 9-16. Siegert, K. J., and W. Mordue. 1987. Breakdown of locust adipoki- netic hormone I by Malpighian tubules of Schistocerca gregaria. Insect Biochem. 17: 705-710. Siegert, K., P. J. Morgan, and W. Mordue. 1985. Primary structures of locust adipokinetic hormones II. Biol Chem. Hoppe-Seyle 366: 723-727. Stone, J. V., W. Mordue, C. E. Broomfield, and P. M. Hardy. 1978. Structure-activity relationships for the lipid-mobilizing ac- tion of adipokinetic hormone. Synthesis and activity of a series of hormone analogues. Eitr. J Biochem. 89: 195-202. Van Heusden, M. C, D. J. Van Der Horst, J. M. Van Doom, J. Wes, and A. M. Th. Beenakkers. 1986. Lipoprotein lipase activity in the flight muscle of Locusta migratoria and its specificity for haemolymph lipoproteins. Insect Biochem 16: 5 1 7-523. Wheeler, C. H. 1989. Transport of fuels to the flight muscles. Pp. 273-303 in Insect Flight. G. J. Goldsworthy and C. H. Wheeler, eds. CRC Press, Boca Raton, Florida. Wheeler, C. H., K. M. Boothby, and G. J. Goldsworthy. 1986. CL- proteins and the regulation of lipoprotein lipase activity in locust flight muscle. Biol. Chem. Hoppe-Seyler 36T. 1 127-1 133. Wheeler, C. H., G. Gade, and G. J. Goldsworthy. 1988. Humoral functions of insect neuropeptides. Pp. 141-157 in Neurohormones o/ Invertebrates. M. C. Thorndyke and G. J. Goldsworthy, eds. Cambridge University Press, Cambridge. Wheeler, C. H., and G. J. Goldsworthy. 1983a. Qualitative and quan- titative changes in Locusta haemolymph proteins and lipoproteins during ageing and adipokinetic hormone action. / Insect Physiol. 29: 339-354. Wheeler, C. H. and G. J. Goldsworthy. 1983b. Protein-hpoprotein interactions in the haemolymph of Locusta during the action of adipokinetic hormone: the role of C, -proteins. J Insect Physio/. 29: 349-354. Wheeler, C. H., and G. J. Goldsworthy. 1985. Lipid transport to the flight muscles in Locusta. Pp. 126-135 in Insect Locomotion. M. Gewecke. ed. Paul Parey Press. Berlin. Wheeler, C. H., and G. J. Goldsworthy. 1986. Lipoprotein/apopro- tein interactions during adipokinetic hormone action in Locusta. Pp. 187-190 in Insect Neurochemistry and Neurophysiology 1986. A. B. Borkovec and D. B. Gelman, eds. Humana Press, New Jersey. Wheeler, C. H., J. E. Mundy, and G. J. Goldsworthy. 1984a. Locust haemolymph lipoproteins visualised in the electron microscope. / Comp. Physiol. 154: 281-286. Wheeler, C. H., D. J. Van Der Horst, and A. M. Th. Beenakkers. 1984b. Lipolytic activity in the flight muscles ofLocusta migrato- ria measured with haemolymph lipoproteins as substrates. Insect Biochem. 14:261-266. /iegler, R., K. Eckart, H. Schwarz, and R. Keller. 1985. Amino acid sequence of Manduca sexla adipokinetic hormone elucidated by combined fast atom bombardment ( FAB)/tandem mass spectrome- try. Biochem. Biophys. Res. Cominun. 133: 337-342. Reference: Biol. Bull 177: 225-229. (October. 1989) The Pigment-Dispersing Hormone Family: Chemistry, Structure-Activity Relations, and Distribution K. RANGA RAO AND JOHN P. RIEHM Department of Biology, The University oj West Florida, Pensacola, Florida, 32514-5751 Abstract. This report summarizes recent work on the pigment-dispersing hormone (PDH) family, a set of re- lated neuropeptides common to arthropods. The pri- mary structures are known for the major form of PDH in several crustacean species (Pandahts borealis, Uca pugilator, Cancer magister. Penaeus aztecus, Procam- barus clarkii) and for related pigment-dispersing factors from two insects (Acheta domesticus, Romalea micro- ptera). In this peptide family, the amino acid chain length (18 residues), termini (N-terminal Asn, C-termi- nal Ala-NH:), and at least 50% of the sequence are con- served. Synthetic analogs have been used to analyze the structure-activity relations of PDH, leading to: an evalu- ation of the role of specific residues; a tentative identifi- cation of the message sequence; and the preparation of stable and superpotent analogs including tyrosinated an- alogs for radioiodination. An enzyme-linked immuno- sorbant assay (ELISA) has been developed for fi-PDH. Antisera raised against a-PDH and /i-PDH were used to determine the distribution of PDH. This distribution and other evidence indicate that, besides its role in hu- moral regulation of the pigmentary system, PDH may serve extra-pigmentary functions. The functions of the PDH-related peptides in insects are unknown. Introduction Crustaceans display reversible color changes and eye pigment movements. The color changes result from dis- persion or concentration (aggregation) of pigment gran- ules within epithelial chromatophores. The somewhat less conspicuous eye pigment movements, associated with light or dark adaptation, may be restricted to retinu- lar cells (photoreceptor cells), or may also involve extra- retinular ommatidial pigment cells. Whereas pigment movements within retinular cells are induced mainly by a direct action of light, the extra-retinular ommatidial pigment cells as well as epithelial chromatophores are controlled by neurosecretory hormones. The total num- ber of pigmentary-effector hormones present in any given species is unknown, but they are separable into two sets having mutually antagonistic actions. The hormones causing chromatophoral pigment concentration and ommatidial dark adaptation belong to one set, and they are distinct from the hormones eliciting chromatophoral pigment dispersion and ommatidial light adaptation (see reviews: Rao, 1985; Rao and Riehm, 1988a, b). Among hormones in the first set, primary structure is known for only the red pigment-concentrating hormone (RPCH) isolated from eyestalks of the shrimp Pandalus borealis (Fernlund and Josefsson, 1972). This crustacean RPCH, an octapeptide ( relative to 0-PDH). The antiserum showed considerable affinity for insect PDFs ( 1 3-21%-) and Penaens PDH (75%), which are very sim- ilar in sequence to /i-PDH. The markedly lower affinity (0.4%^ ) noted with Pwcambanis PDH was reminiscent of the low potency as a pigment disperser and could be due to its single substitution — Glu17 for Asp17. When «-PDH 228 K. R. RAO AND J. P. RIEHM analogs containing ^-PDH-related substitutions were tested, the analogs containing Asn16 or Asp17 reacted bet- ter than those with Glu\ Leu4, Leu", or Lys1 \ The anti- serum failed to recognize a C-terminally truncated ana- log of 0-PDH (1-1 3-NH2). These findings suggest that the antiserum needs several of the residues closer to the C- terminus in /3-PDH for recognition. The specificity of the antiserum raised against «-PDH (Schueler et al., 1986) was not reported. Extra-Pigmentary Functions The RPCH/AKH family members show considerable sequence homology, but serve distinct functions in different arthropods: chromatophoral pigment concen- tration in Crustacea; hyperglycemia, hypertrehalosemia, hyperlipemia, and cardioacceleration in insects (see Goldsworthy and Mordue, 1989). Since insects lack a chromatophoral system, the PDH functions (yet unde- termined) are most likely to be extra-pigmentary in these animals. Immunocytochemical distribution points to a role as neuromodulator or neurotransmitter for PDH. Recent work indicates that RPCH and PDH may also serve extra-pigmentary functions in Crustacea. RPCH has strong excitatory effects on the stomatogastric gan- glion (Nusbaum and Marder, 1988). RPCH and /3-PDH have stimulatory and inhibitory effects, respectively, on the secretion of methyl farnesoate by crustacean man- dibular organs (Laufer el ai, 1987), and thus seem to have a role in the regulation of reproduction. Perspectives Now that the major PDH-like peptides in several spe- cies have been sequenced, and their structure-activity re- lations, immunoreactivity, and tissue distributions have been determined, a new arthropodan peptide family has emerged. However, because this is a very new family, most aspects of its distribution, functions, and evolution remain to be explored. The first major problem is to define the limits of the PDH family and the variability within it. To those ends, the amino acid sequences of the unknown multiple forms of PDH in selected species must be determined, and the genetic basis of those sequences will also have to be established. Moreover, the PDH-related peptides — and the genes encoding them — should be sought in a wider selection of arthropods and crustaceans, as well as in other phyletic groups. Second, although the pigmentary effects of the PDH family will continue to be an object of study, the extra- pigmentary effects will have to be investigated. Sites of action can be located by immunocytochemistry, and functions may be identified by correlating immunoreac- tive hormone liters with various physiological states. Fi- nally, PDH receptors and receptor mechanisms must be characterized. Acknowledgments This investigation was supported by Grant DCB- 871 1403 from the National Science Foundation. The au- thors are thankful to Ms. Carol Hatcher for assistance in preparing the manuscript. Literature Cited Bonomelli, S. L., K. R. Rao, and J. P. Riehm. 1988. Development and application of an ELISA for crustacean /3-PDH. Am. Zool. 28: 117A. Dircksen, H., C. A. Zahnow, G. Gaus, R. Keller, J. P. Riehm, and K. R. Rao. 1987. The infrastructure of nerve endings containing pigment-dispersing hormone (PDH) in crustacean sinus glands: identification by an antiserum against synthetic PDH. Cell Tissue Res 250: 377-387. Fernlund, P. 1976. Structure of a light-adapting hormone from the shnmp Pandalus borealis. Biochim. Biophys. Ada 439: 17-25. Fernlund, P., and L. Josefsson. 1972. Crustacean color change hor- mone: amino acid sequence and chemical synthesis. Science 111: 173-175. Goldsworthy, G. J., and \V. Mordue. 1989. Functions and structure- activity relations of the AKH/RPCH-like peptides. Biol. Bull. Ill: 218-224. Homberg, U., T. G. Kingan, and J. G. Hildebrand. 1987. Gastrin/ CCK-like peptides in the brain of the tobacco hawkmoth Manduca se.\ta. Soc. Neurosci. Ahstr. 13: 225. Kleinholz, L. H. 1975. Purified hormones from the crustacean eye- stalks and their physiological specificity. Nature 258: 256-257. Kleinholz, L. H., K. R. Rao, J. P. Riehm, G. E. Tarr, L. Johnson, and S. Norton. 1986. Isolation and sequence analysis of pigment-dis- persing hormone from eyestalks of the crab Cancer magister. Biol. Bull. 170: 135-143. Laufer, H., E. Homola, and M. Landau. 1987. Control of methyl farnesoate in crustacean mandibular organs. Am. Zool. 21: 69A. Mangerich, S., and R. Keller. 1988. Localization of pigment-dispers- ing hormone (PDH) immunoreactivity in the central nervous sys- tem ofCarcinus maenas and Orconectes limosus (Crustacea), with reference to FMRFamide immunoreactivity in O limosus. Cell Tissue Res. 253: 199-208. Mangerich, S., R. Keller, H. Dircksen, K. R. Rao, and J. P. Riehm. 1987. Immunocytochemical localization of pigment-dispersing hormone (PDH) and its coexistence with FMRFamide immunore- activity in the eyestalks of the decapod crustaceans Carcinus mae- nassmd Orconectes limosus. Cell Tissue Res 250: 365-375. McCallum, M. L., K. R. Rao, J. P. Riehm, C. J. Mohrherr, and VV. T. Morgan. 1988. Isolation of a 0-PDH analog from the cray- fish, Procambarus darkii. Am Zool. 28: 1 17A. Nusbaum, M. P., and E. Marder. 1988. A neural role for a crustacean red pigment concentrating hormone-like peptide: neuromodula- tion of the pylonc rhythm in the crab Cancer borealis. J. Exp. Biol. 135: 165-181. Phillips, J.M.,C. A. Zahnow, and K. R. Rao. 1987. An immunocyto- chemical study of the eyestalk of Penaeus a-iecus utilizing antisera for synthetic fi-PDH and FMRFamide. Am. Zool. 27: 69A. Phillips, J. M., K. R. Rao, J. P. Riehm, and W. T. Morgan. 1988. Isolation and characterization of a pigment-dispersing hor- PIGMENT-DISPERSING HORMONE FAMILY 229 mone from the shrimp Penaeus a:tecus. Soc. Neurosci. Abslr. 14: 534. Rao, K. R. 1985. Pigmentary effectors. Pp. 395-462 in The Biology of Crustacea. \'ol, 9, D. E. Bliss, and L. H. Mantel, eds. Academic- Press, Orlando. Rao, K. R., and J. P. Riehm. 1988a. Chemistry of crustacean chro- matophorotropins. Pp. 407-422 in Advances in Pigment Cell Re- search, J. T. Bagnara. ed. Alan R. Liss, New York. Rao, K. R., and J. P. Riehm. 1988b. Pigment-dispersing hormones: a novel family of neuropeptides from arthropods. Peptides9, Suppl. 1: 153-159. Rao, K. R., J. P. Riehm, C. A. Zahnow, L. H. Kleinholz, G. E. Tarr, L. Johnson, S. Norton, M. Landau, O. J. Semmes, R. M. Sattelberg, W. H. Jorenby, and M. F. I lint/. 1985. Characterization of a pig- ment-dispersing hormone in eyestalks of the fiddler crab L'ca pugi- lator. Proc. Natl. Acad. Sci. USA 82: 5319-5322. Rao, K. R., C. J. Mohrherr, J. P. Riehm, C. A. Zahnow, S. Norton, L. Johnson, and G. E. Tarr. 1987. Primary structure of an analog of crustacean pigment-dispersing hormone from the lubber grasshop- per Romalea microptera. J Biol. Chem. 262: 2672-2675. Schueler, P. A., A. J. Madsen, VV. S. Herman, and R. Elde. 1986. Immunohistochemical mapping of distal retinal pigment hormone in the crayfish central nervous system. Soc. Neurosci. Ah- sir- 12:242. Zahnow, C. A. 1987. Synthesis and bioassay of N-terminal deletion peptides and certain "core" analogs of a crustacean pigment-dis- persing hormone. Masters Thesis, The University of West Florida, Pensacola. Zahnow, C. A., K. R. Rao, C. J. Mohrherr, and J. P. Riehm. 1987. Immunocytochemistry of neuropeptides in the cephalic neuroendocrine system of the lubber grasshopper. Romalea mi- croptera. Soc. Neurosci. Abstr. 13: 993. Reference: Biol. Bull 177: 230-236. (October. 1989) Modified Sperm Ultrastructure in Four Species of Soft- Bodied Echinoids (Echinodermata: Echinothuriidae) From the Bathyal Zone of the Deep Sea KEVIN J. ECKELBARGER, CRAIG M. YOUNG, AND J. LANE CAMERON Division of Marine Sciences, Harbor Branch Oceanographic Institution, 5600 Old Dixie Highway, Fort Pierce. Florida 34946 Abstract. Sperm of the bathyal echinothuriid echi- noids Phormosoma placenta, Sperosoma antillense, Ara- eosoma fenestratitm, and A. belli are similar to those of other echinoids, but have several unique morphological features involving the acrosomal vesicle, the nucleus, and the middlepiece. The acrosomal vesicle shows re- gional staining differences including a densely staining central region surrounded by an electron-opaque com- ponent. Sperm nuclei are highly elongated and abruptly taper posteriorly. With the exception of one species, the nuclei lack a distinct centriolar fossa. Intracellular drop- lets resembling lipid extend from the extreme posterior region of the middlepiece to form a collar around the proximal portion of the axoneme. The presence of lipid- like bodies in the middlepiece suggest that the sperm are long-lived and therefore require additional energy stores not found in most metazoan sperm. These findings are compared with a similar study of sperm ultrastructure in three shallow-water echinothuriid species, and their potential significance is discussed in relation to the pres- ent knowledge of echinothuriid reproductive biology. Introduction Echinothuriid echinoids have soft, flexible tests atypi- cal of most sea urchins. The majority of species are con- fined to the deep sea but some littoral species are re- ported from the Indo-Pacific region (Amemiya el a!.. 1980). Our knowledge of the reproductive biology of members of this family has been limited mostly to gen- eral observations of a few deep-sea species (Mortensen, 1927; Ahfeld, 1977). Tyler and Gage (1984) have exam- ined gametogenesis in five echinothuriids from the Received 15 March 1989; accepted 27 June 1989. Rockall Trough in the Northeast Atlantic Ocean using light microscopic histology. They suggest that the domi- nant reproductive strategy is the year-round production of large eggs undergoing benthic direct development. However, other authors have reported that echinothuriid eggs float (Amemiya and Tsuchiya, 1979; Young and Cameron, 1987; Cameron et ai, 1988) suggesting that the larvae are probably pelagic lecithotrophs. Echinoderms are among the most abundant and di- verse macrofaunal invertebrates in the deep sea (Pawson, 1982; Billet and Hansen, 1982), yet little is known about gamete and gonad ultrastructure in species from this habitat. The only published account deals with the ultra- structure of the highly aberrant sperm of the deep-sea concentricycloid Xyloplax (Healy et ai, 1988). General sperm morphology is known for nearly 70 shallow-water echinoids, although fewer than 10 species have been the subject of ultrastructural studies. These combined inves- tigations show that echinoid sperm are consistently con- servative in morphology and exhibit little interspecific structural variation (see Summers et ai, 1975; Chia and Bickell, 1983, for review). The present paper describes ultrastructural features of the mature sperm of the bathyal echinothuriids Phormo- soma placenta. Sperosoma antillense. Araeosoma fenes- tnitiim. and A. belli collected in the Bahama Islands by manned submersibles at depths ranging from 600 to 900 meters. Although their sperm are similar to those of other echinoids, significant ultrastructural differences are noted in the acrosome, nucleus, and middlepiece. Sperm morphology is influenced by the environment into which they are released prior to fertilization (Fran- zen, 1956, 1970; Afzelius, 1977), so structural alterations in these sperm suggest that unique selective pressures may be present in the deep sea which are absent from shallow water habitats. 230 BATHYAL ECHINOID SPERM ULTRASTRUCTURE 231 These results are compared with an earlier ultrastruc- tural study of sperm morphology in three species of shal- low-water echinothuriids from Japanese waters (Amem- iya et a!., 1980). The potential significance of these findings are discussed in relation to echinothuriid repro- ductive biology. Materials and Methods Live, sexually mature specimens of Phormosoma pla- centa (Fig. 1), Sperosoma antillense, Araeosoma fenes- tratwn (Fig. 2) and A. belli were collected during April, May, and October, 1986, at depths ranging from 600 to 900 m in the Bahamas. The seawater temperature at the various collection sites ranged from 7 to 10°C. Collection sites included the west side of San Salvador Is. (24°02'N, 74°32.4'W), the north side of New Providence Is. (25°03.1TSf, 77°31.4'W), Southwest Reef (24°53.41'N, 77°33.14'W), and the east side of Andros Is. (24°51.4'N, 77°50.62'W). Specimens were obtained using specialized collecting equipment on Johnson-Sea-Link submers- ibles. In most instances, live animals were dissected on board ship shortly after collection, and testes were re- moved for immediate fixation. Occasionally, however, specimens were maintained on board in cooled aquaria where they remained active and healthy until later use. Whole testes were fixed for transmission electron micros- copy (TEM); several methods, including those of Eckel- barger (1979) and Smiley (1988), were used, although none produced satisfactory results. Acceptable results were finally obtained with a modification of the method published by Bickell et al. (1980). Tissue was fixed by immersion for 1 h in a primary fixative consisting of 2.5% glutaraldehyde in filtered seawater at room temper- ature, followed by a 1 5-min wash in 2.5% NaHCO,, and postfixation for 1 h in 2% OsO4 in 1.25% NaHCO3 at room temperature. Tissues were rapidly dehydrated in ascending concentrations of ethanol, transferred through two changes of propylene oxide, and embedded in Epon. Thin sections were cut with a diamond knife and stained with alcoholic uranyl acetate and aqueous lead citrate for 10 min each, then examined with a Zeiss EM9S-2 trans- mission electron microscope. Live active sperm released from dissected testes were collected and fixed for scanning electron microscopy (SEM). Sperm were fixed using the same procedure as for TEM, followed by dehydration through an ascending ethanol series to 50%, air-dried on cover slips attached to aluminum stubs, and sputter-coated with gold-palla- dium. Sperm were then photographed using a Novascan 30 SEM with an accelerating voltage of 1 5 kV. Results The mature spermatozoa of all four species are similar in general morphology but differ in head length (aero- some, nucleus, and middlepiece) and relative dimen- sions. They are bullet-shaped with a small, terminal aero- some, a conical nucleus, and short middlepiece (Figs. 5- 12). The sperm head length of the four species range from the longest, Phormosoma placenta (12 X 1.5 to the shortest, Araeosoma fenestratum (8.5 X 1.0 The sperm heads of A. belli and Sperosoma antillense each measure 9.0 X 1.25 ^m. All four species have large, yolky eggs, ranging from the largest, Araeosoma fenestra- tum ( 1 290 nm ), to Sperosoma antillense ( 1 060 ^m ), A ra- eosoma belli (965 /urn), and the smallest, Phormosoma placenta (890 ^m). The acrosome is positioned just anterior to the nucleus and consists of a subspherical acrosomal vesicle with a flattened or slightly concave basal surface (Figs. 3, 4). The acrosomal membrane and the sperm plasmalemma are not well defined due to poor fixation. However, the posterior-lateral acrosomal membrane is thicker and more electron dense than the apical portion. The acroso- mal vesicle contains two types of granular material of differing electron densities. The central portion contains a more densely staining vase-shaped region surrounded laterally by an electron-opaque component. A deep sub- acrosomal fossa is present just posterior to the acrosomal vesicle. Although the acrosomal region of all four species are similar, the subacrosomal fossa of Sperosoma antil- lense (Fig. 4) is narrower and deeper than that of the other species, typified by A. belli (Fig. 3). The subacroso- mal fossa contains granular periacrosomal material ex- tending around the basolateral sides of the acrosomal vesicles and running posteriorly between the anterolat- eral margin of the nucleus and sperm plasmalemma. The nucleus contains highly condensed chromatin and gradually tapers anteriorly in all four species. Nuclear vacuoles are common in all sperm examined (Fig. 15). In the sperm of Phormosoma placenta, the ex- treme posterior region of the nucleus narrows abruptly to form a neck-like extension that extends into the mid- dle piece (Figs. 9, 13). This posterior extension tapers more gradually in the sperm of Sperosoma antillense (Figs. 10, \4), Araeosoma feneslrattim (Figs. 11, 15), and A. belli (Figs. 12, 16). The posterior surface of this exten- sion is slightly concave and the axoneme-bearing distal centriole is closely associated with it ( Figs. 1 3- 1 6). A dis- tinct centriolar fossa is absent except for a shallow one in P. placenta (Fig. 13). The proximal centriole is posi- tioned laterally to the distal centriole and at a slight angle to the long axis of the sperm (Figs. 15, 16, 18). The distal centriole possesses a centriolar satellite consisting of nine radiating arms (Fig. 19). A single, circular mitochon- drion surrounds the posterior nuclear extension (Fig. 17). In the sperm of all four species, intracellular droplets morphologically resembling lipid, extend from the ex- treme posterior region of the middlepiece (Figs. 13-16). They are consistently round in P. placenta (Fig. 1 3), but 232 K. J. ECKELBARGER ET AL Figure 1 . Adult Phurmosoma placenta photographed in silit at a depth of 700 m in Bahamian waters. Arrow indicates one of several epidermal sacs of unknown function which project from the surface of the animal. Figure 2. Adult AruciuntHii Mil photographed in xiln at a depth of 750 m in Bahamian waters. BATHYAL ECHINOID SPERM ULTRASTRUCTURE 233 irregular in shape in the other species (Figs. 14-16). In cross section, a collar of lipid-like droplets surrounds the basal region of the axoneme (Fig. 20). The axoneme is of the 9 + 2 pattern. Discussion Recent investigations of sperm morphology from deep-sea echinoderms have revealed a variety of struc- tural modifications that differ markedly from the sperm of shallow water species. These include a high incidence of elongated sperm heads in bathyal echinoids and holo- thuroids (Eckelbarger et a/., in press), the discovery of dimorphic sperm in the abyssal echinoid Phrissocystis nntltispina (Eckelbarger el al, 1989), and the presence of highly aberrant sperm in the abyssal concentricycloid Xyloplax turnerae (Healy et al, 1988). The present paper describes additional modifications of male gametes from four species of bathyal echinoids from Bahamian waters. The sperm acrosomes of echinoids are morphologi- cally conservative and have been described ultrastructur- ally for a number of species (see Chia and Bickell, 1983). The acrosomal vesicle typically contains a homoge- neously distributed particulate material of medium elec- tron density except for its basal region where denser ma- terial is deposited along the inner acrosomal membrane. The sperm acrosomes of all four echinothuriid species we examined resemble those of other echinoids with re- spect to position and superficial morphology, but show unique regional staining differences within the acroso- mal vesicle. In other echinoderm classes, the contents of the acrosomal vesicle are often homogeneous, although densely staining, centrally positioned acrosomal gran- ules have been reported in the holothuroids Citcumaria lubrica, C. miniata, and Leptosynapta clarki (Atwood and Chia, 1974), and the crinoid, Antedon petasus (Af- zelius, 1977). Therefore, the echinothuriid acrosomal vesicle represents a morphological variant different from any observed in other echinoderm sperm. We do not be- lieve this structural variation results from a fixation arti- fact because no differences were observed in sperm pre- pared for electron microscopy using several fixation methods. The functional significance of this novel acro- somal morphology is unknown, but it may reflect re- gional differences in enzyme distribution such as that ob- served in the cortical granules of sea urchin eggs ( Alliegro andSchuel, 1988). Echinothuriid sperm nuclei have a shape unique to echinoids. Echinoid sperm nuclei typically are wider posteriorly than anteriorly and have a relatively deep centriolar fossa into which the distal centriole and its as- sociated axoneme are inserted (Chia and Bickell, 1983). Echinothuriid sperm nuclei abruptly taper posteriorly and a centriolar fossa is virtually absent in Sperosoma. Araeosoma belli, and A. fenestratum, and weakly devel- oped in Phorornosa placenta. The presence of a deep cen- triolar fossa in echinoderm sperm has been suggested as a means of strengthening the connection between the tail and the elongated nucleus (Chia and Bickell, 1983). The position of the mitochondrion relative to the nucleus varies slightly in echinoderm sperm, but generally sur- rounds the extreme posterior end of the nucleus (Chia and Bickell, 1983). In some echinoids, such as Paracen- trotus lividus (Anderson, 1968), Strongylocentrotus purpuratm (Longo and Anderson, 1969), and Arbacia lixula and Echinometra lucimter (Cruz-Landim and Beig, 1976), the mitochondrion is positioned posterior to the nucleus and forms a collar around the proximal portion of the axoneme. In the four echinothuriid spe- cies, the sperm mitochondrion wraps around the poste- rior portion of the nucleus, starting where the nucleus abruptly tapers. The modified sperm of the small brood- ing holothuroid Cucumaria lubrica is strikingly similar to the echinothuriid sperm with respect to nuclear shape and mitochondria! position (Atwood and Chia, 1974), although a well-developed centriolar fossa is present in C. lubrica. The lipid-like droplets we observed in the middlepiece of the echinothuriid sperm closely resemble intracellular triglyceride deposits that are common in a variety of so- matic cells (Bloom and Fawcett, 1968; Alberts et al., 1983). To our knowledge, there are no other reports of lipid deposits in metazoan sperm aside from a few echi- noderm species. Intracellular deposits that resemble lipid on morphological grounds, have been reported posterior to the middlepiece mitochondrion in the echinoids Echi- norachniits parma (Summers and Hylander, 1974; Sum- mers # al., 1975), Arbacia pitnctulata (Bernstein, 1962; Longo and Anderson, 1969), Echinocardiumflavescens Figures 3, 4. Anterior region of mature sperm of Araeosoma bell: (Fig. 3) and Sperosoma antillense (Fig. 4) showing acrosomal vesicle with two regions of differing electron densities. Arrow indicates dense staining posterior acrosomal membrane; *, subacrosomal fossa; N, nucleus. Figures 5-8. Scanning electron micrographs of mature sperm of Phormosoma placenta (Fig. 5), Spero- soma antillense (fig. 6), Araeosoma fenestratum (Fig. 7), and A belli ' (Fig. 8). Figures 9-12. Transmission electron micrographs of longitudinal sections through mature sperm of Phormosoma placenta (Fig. 9), Sperosoma antillense (Fig. 10), Araeosoma fenestratum (Fig. 1 1), and A belli (Fig. 12). The sperm acrosomes in Figures 9, 1 1, and 12 are not clearly indicated due to the slightly oblique angle of the sections. A, acrosome; N, nucleus; M. mitochondrion; L, lipid-like deposit. 234 K. J. ECKELBARGER ET AL. M. 17 Figures 13-16. Longitudinal sections through the posterior region of the mature sperm of Phormo- si/ina placenta (Fig. 1 3 ), Sperosoma antillense ( Fig. 1 4 ). A raeosomafeneslratum ( Fig. 1 5 ), and A belli ( Fig. 16). Horizontal lines in Fig. 13 indicate level of cross-sections through sperm of Plwrmosoma placenta shown in Figures 17-20. N. nucleus; M, mitochondrion; CF, centriolar fossa; DC, distal centriole; PC, proximal centnole; L, lipid-hke deposit. Figure 17. Cross section through posterior nucleus (N) and surrounding mitochondrion (M) of sperm of P. placenta BATHYAL ECHINOID SPERM ULTRASTRUCTURE 235 and Brissopsis lyrifera (Afzelius and Mohri, 1966), Ha- palosoma gemmiferum and Aracosoma owstoni ( Amem- iya et a!.. 1 980), and in the holothuroid Cucumaria mini- ata (Fontaine and Lambert, 1976). However, qualitative comparisons of micrographs from the above studies show that, with the exception of E. parma (Summers and Hylander, 1974) and A. owstoni (Amemiya el al. 1980), these deposits are minor in comparison to those observed in the present study. The role of lipid deposits has not been determined in any of these echinoderm sperm. Echinoderm spermatozoa are dependent on the me- tabolism of mitochondria! phospholipids during swim- ming (Rothschild and Cleland, 1952), or in some in- stances, glycogen stores (Anderson and Personne, 1975). In experiments with the sperm of the echinoid Brissopsis lyrifera. Afzelius and Mohri (1966) reported that pro- longed swimming caused a gradual disappearance of mi- tochondrial cristae. They hypothesized that the sperm consume these structural phospholipids as an energy source. However, no change was observed in the lipid- like inclusions after 6 h of swimming. Triglycerides of fatty acids are used as an energy reserve, and they are commonly associated with mitochondria (Alberts et al.. 1983). The presence of lipid deposits in intimate associa- tion with mitochondria in some echinoderm sperm sug- gests that the cells are long-lived or must expend energy at a rate not commonly demanded of other metazoan sperm. We have not observed spawning in any of the echinothuriids we examined, and we know nothing of their fertilization biology or the potential life span of nat- urally released sperm. With the occasional exception of Phormosoma placenta, adult echinothuriids generally occur at very low densities in the Bahamas. The exten- sive lipid stores in these sperm could provide an ex- tended window of opportunity for fertilization when males and females are widely separated. Some cidarid ur- chins improve fertilization success by aggregating during the breeding season (Young, unpub. data). However, echinothuriids appear not to breed seasonally (Tyler and Gage, 1984) nor to aggregate for spawning. The present study of bathyal echinothuriid sperm morphology provides an interesting comparison to an earlier investigation of sperm ultrastructure in the three shallow-water echinothuriids Asthenosoma ijimai. Araeosoma owstoni. and Hapalosoma gemmiferum (Amemiya et al.. 1980). The latter authors described sperm morphologies very similar to those in the present study, including regional staining differences within the acrosomal vesicles, a long, tapering nucleus that abruptly narrows posteriorly, and "follicular bodies" in the mid- dlepiece of//, gemmiferum and A owstoni sperm, which they viewed as analagous to the lipid-like bodies reported from other echinoid sperm. However, they reported an "electron opaque rod" within the acrosomal vesicle. In addition, the lipid-like droplets appear to be smaller than those we described, and they do not extend from the pos- terior region of the middlepiece as they do in the deep- sea echinothuriid sperm. No information was presented regarding the process of natural spawning or sperm lon- gevity. The most striking difference noted between our two studies is the shorter head lengths in the sperm of the shallow-water species. Sperm head length was estimated to be 7 ^m for A. ijimai. 6 ^m for .-1. owstoni. and 4 /urn for H. gemmiferum. In contrast, we measured sperm head lengths of 1 2 /urn for Phormosoma placenta, 8.5 ^m forA.fenestratum, and 9 ^m for A. belli and Sperosoma antillense. A high incidence of elongated sperm heads was noted recently in a survey of bathyal echinoids(Eck- elbarger et al.. in press). Ultrastructural studies of the gametes of deep-sea or- ganisms have been rare, with only three studies of gamete morphology and sperm development in the vestimintif- eran Riftia pachyptila (Gardiner and Jones, 1985; Gary et al.. 1989), and spermiogenesis in the abyssal echino- derm Xyloplax turnerae (Healy et al.. 1988) having been published recently. This is unfortunate because deep-sea habitats offer a unique laboratory for the study of gamete evolution in an environment substantially different from that of shelf waters (see Wilson and Hessler, 1987). In- deed, recent observations of unique gonadal, gamete, and larval developmental features in bathyal and abyssal echinoderms have demonstrated that deep-sea echino- derms have undergone evolutionary changes in their re- productive biology unlike that of their shallow water rel- atives (Eckelbarger et al., in press, 1989; Young et al., in press). This suggests that the deep sea is a region ripe for studies dealing with the reproductive evolution of ma- rine invertebrates. Acknowledgments We thank Pamela Blades-Eckelbarger and Pat Linley for specimen preparation for scanning and transmission electron microscopy and John Miller for assistance with some specimen identification and systematics. The manuscript benefited from discussions with Dr. F. S. Figure 18. Cross section through centrioles (C) and surrounding mitochondrion (M) in middlepiece of P. placenta sperm. Figure 19. Cross section through centriolar satellite showing nine radiating arms (arrows) in middle- piece of sperm of P. placenta. Figure 20. Cross section through middlepiece of sperm of P. placenta showing ring of lipid-like bodies (L) surrounding proximal portion of axoneme (A). 236 K. J. ECKELBARGER ET AL Chia. We thank the HBOI ship and submersible crews of the R/V Seward Johnson and R/V Edwin Link for their assistance. Brian Bingham, Dr. Paul Tyler, and several others assisted at sea. The research was supported in part by N.S.F. grant OCE-877922. This is contribution num- ber 7 19 of Harbor Branch Oceanographic Institution. Literature Cited Afzelius, B. A. 1977. Spermatozoa and spermatids of the crinoid Antedon petasus, with a note on primitive spermatozoa from deu- terostome animals. J. Ultrastmct. Res. 59: 272-28 1 . Afzelius, B. A., and H. Mohri. 1966. Mitochondria respiring without exogenous substrate — a study of aged sea urchin spermatozoa. Exp Cell Res. 42: 10-11. Ahfeld, T. E. 1977. A disparate seasonal study of reproduction of eight deep sea macroinvertebrate species from the Northwestern At- lantic Ocean. Ph.D. dissertation. Florida State Univ.. 105 pp. Alberts, B., D. Bray, J. Lewis, M. Raff, K. Roberts, and J. D. Watson. 1983. Molecular Biology of the Cell. Garland Publishing, Inc., New York. 1146pp. Alliegro, M.C., and H.Schuel. 1988. Immunocytochemical localiza- tion of the 35-kDa sea urchin egg trypsin-like protease and its effects upon the egg surface. Dev Bio/. 1 25: 168-1 80. Amemiya, S., and T. Tsuchiya. 1979. Development of the echino- thurid sea urchin Asthenosoma ijimai. Mar. Biol 52: 93-96. Amemiya, S., T. Suyemitsu, and I. Uemura. 1980. Morphological ob- servations on the spermatozoa of echinothurid sea urchins. Dev. Growth Diff. 22: 327-335. Anderson, VV. A. 1968. Cytochemistry of sea urchin gametes. I. In- tramitochondnal localization of glycogen, glucose-6-phosphatase and adenosine triphosphate activity in spermatozoa of Paracenlro- tus lividus. J. Ultrastmct. Res. 24: 398-441 . Anderson, VV. A., and P. Personne. 1975. The form and function of spermatozoa: a comparative view. Pp. 3-14 in The Functional Anatomy of the Spermatozoon, B. A. Afzelius, ed. Pergamon Press. Oxford. Atwood, D. G., and F.-S. Chia. 1974. Fine structure of an unusual spermatozoon of a brooding sea cucumber, Ciienmaria luhrica. Can.J. Zoo/. 52:519-523. Bernstein, M. H. 1962. Normal and reactive morphology of sea ur- chin spermatozoa. Exp. Cell Res. 27: 197-209. Bickell, L. R., F.-S. Chia, and B. J. Crawford. 1980. A fine structural study of the testicular wall and spermatogenesis in the crinoid, Flo- rometraserratissima(Ech\noderma\a). J Morphol. 166: 109-126. Billet, D. S. M., and B. Hansen. 1982. Abyssal aggregations of Kolga hyalina Danielssen and Koran (Echinodermata: Holothuroidea) in the northeast Atlantic Ocean: a preliminary report. Deep Sea Res. 29:799-818. Bloom, W., and D. W. Fawcett. 1968. A Textbook of Histology. W. B. SaundersCo., Philadelphia. 858 pp. Cameron, J. L., F. S. McEuen, and C. M. Young. 1988. Floating leci- thotrophic eggs from the bathyal echinothuriid sea urchin Araeo- soma feneslratum. Pp. 177-180 in Echmoderm Biology. R. D. Burke, P. V. Mladenov, P. Lambert, and R. L. Parsley, eds. Bal- kema Press, Rotterdam. Cary, A. C, H. Felbeck, and M. D. Holland. 1989. Observations on the reproductive biology of the hydrothermal vent tube worm Riftia pachyptila. Mar. Ecol. Prog Ser. 52: 89-94. Chia, F.-S. and L. R. Bickell. 1983. Echinodermata. Pp. 545-620 in Reproductive Biology of Invertebrates, Vol. II: Spermatogenesis and Sperm Function, K. G., and R. G. Adiyodi, eds. John Wiley & Sons, New York. Cruz-Landim, C. and D. Bcig. 1976. Spermiogenesis in the sea ur- chins Arbacia lixula and Eclunometra lucunter (Echinodermata). Cytologia 41: 331-344. Eckelbarger, K. J. 1979. Ultrastructural evidence for both autosynthetic and heterosynthetic yolk formation in the oocytes of an annelid ( Phragmatopoma lapidosa: Polychaeta). Tissue Cell 1 1 : 425-443. Eckelbarger, K. J., C. M. Young, and J. L. Cameron. In press. Modified sperm in echinoderms from the bathyal and abyssal zones of the deep sea. International Symposium Series: 23rd European Marine Biology Symposium. P. A. Tyler and J. Ryland, eds. Olsen & Olsen, Denmark. Eckelbarger, K. J., C. M. Young, and J. L. Cameron. 1989. Ultra- structure and development of dimorphic sperm in the abyssal echjnoid Phrissocrstis multispina (Echinodermata: Echinoidea): implications for deep sea reproductive biology. Biol. Bull. 176: 257-27 1 . Fontaine, A. R. and P. Lambert. 1976. The fine structure of the sperm of a holothurian and an ophiuroid. J Morphol. 148: 209-226. Franzen, A. 1956. On spermiogenesis, morphology of the spermato- zoan, and biology of fertilization among invertebrates. Zoo/. Bidr. Uppsala 30: 355-482. Franzen, A. 1970. Phylogenetic aspects of the morphology of sperma- tozoa and spermiogenesis. Pp. 29-46 in Comparative Spermatol- ogy. B. Bacetti, ed. Academic Press, New York. Gardiner, S. L., and M. L. Jones. 1985. Ultrastructure of spermio- genesis in the Vestimentiferan tube worm Riftia pachyptila (Pogo- nophora: Obturata). Trans. Am. Microsc. Soc. 104: 19-44. Healy, J. M., F. W. E. Rowe, and D. T. Anderson. 1988. Spermatozoa and spermiogenesis in \'ylopla\ (Class Concentricycloidea): a new type of spermatozoon in the Echinodermata. Zoo/. Scripta 17: 297- 310. Longo, F. J., and E. Anderson. 1969. Sperm differentiation in the sea urchins Arbacia punctulata and Strongylocentrotus purpuralus. J. Vltrastrucl. Rex 27: 486-509. Mortensen, T. 1927. Handbook of the Echinoderms of the British Isles. W. Backhuys, ed. Oxford Univ. Press. 47 1 pp. Pawson, D. L. 1982. Deep-sea echinoderms in the Tongue of the Ocean, Bahama Islands: a survey using the Research Submersible \LVlN.Aum. Mits. Mem. 16: 129-145. Rothschild, L., and K. W. Cleland. 1952. The physiology of sea ur- chin spermatozoa. The nature and location of endogenous sub- strate. J. Exp. Biol. 29: 66-7 1 . Smiley, S. 1988. The dynamics of oogenesis in Stichopus californicus (Echinodermata: Holothuroidea), and an explanation of its annual ovarian cycle. Biol Bull. 175: 79-93. Summers, R. G. and B. L. H> lander. 1974. An ultrastructural analysis of early fertilization in the sand dollar, Echinorachnius parma. Cell Tissue Res 150:343-368. Summers, R. G., B. L. Hylander, L. H. Colwin, and A. L. Colwin. 1975. The functional anatomy of the echinoderm spermatozoon and its interaction with the egg at fertilization. Am. Zoo/. 15: 523-55 1 . Tyler, P. A. and J. D. Gage. 1984. The reproductive biology of echi- nothuriid and cidarid sea urchins from the deep sea (Rockall Trough. Northeast Atlantic Ocean). Afar Biol. 80: 63-74. Wilson, G. D. F. and R. R. Messier. 1987. Speciation in the deep sea. Ann. Re^•. Ecol. Syst. 18: 185-207. Young, C. M., and J. L. Cameron. 1987. Laboratory and in situ floa- tation rates of lecithotrophic eggs from the bathyal echinoid Phor- mosoma placenta. Deep-Sea Res. 34: 1629-1639. Young, C. M., J. L. Cameron and K. J. Eckelbarger. In press. Extended prefeeding period in planktotrophic larvae of the bathyal echinoid Aspidodiadema jacobyi Agassiz. J. Mar Biol. Assoc. U. K Reference: Bioi Bull. Ill: 237-246. (October, 1989) Growth and Energy Imbalance During the Development of a Lecithotrophic Molluscan Larva (Haliotis rufescens) WILLIAM B. JAECKLE1 AND DONAL T. MANAHAN2 Department of Biological Sciences, University of Southern California. Los Angeles, California 90089-0371 Abstract. Larvae of the gastropod Haliotis rufescens are classified as "nonfeeding" because they cannot cap- ture paniculate foods. However, for only 1 out of 5 inde- pendent cultures was a net decrease observed in dry or- ganic weight during the complete period of larval devel- opment (5 to 7 days). In fact, there were net increases in dry organic weight from the oocyte (day 0) to the newly formed veliger larva (2-day-old). These weight increases during early development could be explained by in- creases in the amounts of specific biochemical compo- nents of the larvae, relative to oocytes. The metabolic rates of larvae were measured (oxygen consumption) and used to compare (i) the required energy for development with (ii) the energy supplied from the catabolism of bio- chemical reserves. This analysis revealed that the cost of development for larvae could not be explained by the rates of use of the energy stores initially present in the oocyte. Larvae, from two independent cultures, could only supply 25% or 7 1% of their energy requirements by the use of internal reserves. Larvae of H. rufescens can- not use paniculate foods and, thus, this energy resource cannot be invoked. Estimates of the contribution that dissolved organic material in seawater could provide to larvae, showed that this pool of exogenous material could supply the missing energy. It is suggested that "nonfeeding" larvae can feed, but that their only avail- able nutrients are in a dissolved form. Received 15 March 1989; accepted 31 July 1989. ' Present address: Harbor Branch Oceanographic Institution, 5600 Old Dixie Highway, Fort Pierce, Florida 34946. : To whom reprint requests should be directed. Introduction The traditional separation of marine planktonic larvae into three trophic groupings (planktotrophy, facultative planktotrophy, and lecithotrophy) is based upon an abil- ity, or lack of the ability, to concentrate and capture par- ticulate foods from seawater (Thorson, 1946; Chia, 1974). However, this classification scheme does not con- sider an energy resource available to all soft-bodied in- vertebrate larvae, namely dissolved organic material (DOM) in seawater. By ignoring this resource, an im- plicit assumption has been made that planktotrophic (feeding) larvae are energetically dependent on the envi- ronment for nutrients, whereas lecithotrophic larvae are not. It follows then, that lecithotrophic (nonfeeding) lar- vae, using an energy source of fixed content (yolk), should continually decrease in organic weight during de- velopment. Growth, denned here as an increase in dry organic weight, can only occur in these nonfeeding forms following the development of distinct juvenile or adult feeding structures. Changes in the biochemical composition have been well studied for embryos and larvae of marine inverte- brates that produce planktotrophic larvae. There is gen- eral agreement in the literature that prefeeding embryos use internal energy stores (carbohydrate, lipid, and pro- tein) to supply the energy requirements of early develop- ment (e.g.. Cognetti, 1982). Following the development of feeding structures, larvae deprived of paniculate foods are also assumed to rely on an energy source of fixed con- tent (yolk or accumulated reserves). This assumption is supported by the observed decreases in the biochemical components of starving larvae. For example, Millar and Scott (1967) reported that larvae of the bivalve Ostrea 237 238 W. B. JAECKLE AND D. T. MANAHAN edulis decreased both in dry organic weight and in all measured biochemical components during periods of starvation. Catabolism of the neutral lipid fraction pro- vided the greatest proportion of liberated energy fol- lowed, in order, by protein and carbohydrate. Holland and Spencer (1973) also examined the changes in bio- chemical composition of starved O. edulis larvae. During starvation, the decrease in dry organic weight was ex- plained by net decreases in lipid and protein, and again, the greatest amount of energy was made available through the catabolism of lipid. This pattern of energy use (lipid > protein > carbohydrate) was also seen during the cyprid stage of development (nonfeeding) in the bar- nacle Bahunis balanoides (Lucas et al, 1 979). Mann and Gallager (1984, 1985) reported that the dry organic weight of larvae of the bivalves Martesia cuneiformis. Te- redo navalis, and Bankia gouldi also decreased during starvation. For these larvae, the protein fraction, rather than the lipid fraction, served as the primary energy source during periods of nutrient deprivation. Studies have also been conducted examining net changes in or- ganic weight and biochemical composition of echino- derms with nonfeeding development (Lawrence et al., 1984; McClintock and Pearse, 1986). These authors have reported that, for a total of five species, there was little net change in the organic weight or energy content between eggs and juveniles. Lawrence et al. (1984) sug- gested that their results could be explained by the fact that either (i) the metabolic rates were too low to cause measurable depletion of energy reserves or (ii) the ani- mals were gaining sufficient energy from dissolved or- ganic material in seawater to offset the metabolic de- mands of development. The only possible source of nutrients from the envi- ronment available to nonfeeding developmental stages of marine invertebrates would be in the form of DOM. Total carbon concentrations of DOM in coastal waters are at least ten times greater than those of organic carbon associated with particles (Parsons, 1975; Williams, 1975; MacKinnon, 1981; Sugimura and Suzuki, 1988). De- spite the fact that only a small percentage (ca. 10%) of the total DOM has been chemically characterized (Williams, 1975). it is clear that biologically important compounds such as monosaccharides, amino acids, fatty acids, and nucleic acids are present in dissolved form in coastal ma- rine environments (Testerman, 1972; Mopper el al.. 1 980; Parkes and Taylor, 1983;Carlucci etal.. 1984;De- Flaunetal.. 1987). There is evidence that planktotrophic larvae can take up specific compounds from the total pool of DOM (Reish and Stephens, 1969; Manahan, 1983; Manahan et al.. 1983; Davis and Stephens, 1984). However, the ability of nonfeeding larvae to transport specific fractions of DOM has been less extensively studied. Recently, it has been shown that nonfeeding trochophore and veliger larvae of the gastropod Haliotis rufescens are able to take up dissolved amino acids from seawater, and that trans- ported alanine is rapidly used in specific anabolic and catabolic pathways in these veliger larvae (Jaeckle and Manahan. 1989). In the present study, larvae of the gastropod Haliotis rufescens were cultured in natural seawater from the fer- tilized oocyte until they were competent to settle. The changes in dry organic weight, biochemical composition, and metabolism (oxygen consumption) were measured throughout the complete development of these larvae. An energy budget was constructed to assess the contribu- tion of endogenous reserves to the energy demands of these larvae. This comparison revealed that larvae of//. rufescens are not in energy balance, suggesting that these "nonfeeding" larvae are obtaining energy from their en- vironment, presumably in the form of dissolved organic material. Materials and Methods Culturing methods Fertilized oocytes of Haliotis rufescens were obtained from the "Ab Lab" (Port Hueneme, California). The oo- cytes, embryos, and larvae were cultured in seawater at concentrations of approximately 5/ml at 16 to 17°C in 200-1 culture vessels. It is known that the routine proce- dures (e.g., sand filtering) used in marine laboratories to remove particles also affect the organic chemistry of sea- water (Manahan and Stephens, 1983). To maintain the organic composition of seawater as close as possible to in situ conditions, all seawater used for larval culturing was passed only through a 0.2-^m (pore size, Nuclepore) polycarbonate filter. Assays for changes in the organic chemistry of seawater, caused by filtration procedures, were performed with high-performance liquid chroma- tography (procedures described elsewhere: Manahan, 1989). Filtered seawater samples were taken and ana- lyzed for the concentrations of individual dissolved free amino acids. The total concentration of amino acids in the filtered seawater was always within the range re- ported for coastal marine environments (i.e., lOnA/to 1 nAf: Williams, 1975). Fresh seawater was collected daily, and used as soon as it had thermally equilibrated to the culture temperature. For each larval culture, following the formation of the definitive larval shell (day 2), the water was changed on a daily basis. At each sampling period, the larvae were gently sieved onto a 80-^m poly- ester mesh, and samples were retained for analysis of weight and biochemical composition. The culture vessel was then cleaned by a brief swabbing with a 5% sodium ENERGY IMBALANCE IN NONFEEDING LARVAE hypochloride solution (bleach) followed by sequential washes of hot water, deionized water, and then filtered seawater. Collection ofoocytes and larvae Samples of either oocytes, prior to fertilization, or lar- vae were placed in a graduated cylinder. The concentra- tion of individuals was determined by counting several (usually 3 to 5) aliquots of the suspension until the co- efficient of variation of the mean was less than 6%. Then, 9 samples of the suspension were removed and each sam- ple placed in a separate 1 5-ml centrifuge tube. Following centrifugation, the pellet was washed with three times the sample volume using ammonium formate (3.4% w/v, isotonic with seawater). This washing procedure was then repeated twice. Replacement of seawater by ammo- nium formate is an important procedural step because ammonium formate is a volatile salt and there is no re- sidual inorganic residue following drying. However, the ammonium formate solution must be passed through a 0.2-^m (pore size) filter immediately before use to avoid the carry-over of paniculate material that will bias the measured weights. Six of the nine samples were placed in preashed (12 h at 500°C), preweighed aluminum boats. frozen, and retained for determinations of weights. The remaining samples were placed in 1.5-ml centrifuge tubes, frozen, and held at -20°C for biochemical anal- ysis. Weight determinations Samples retained for determinations of organic weight were placed in an 80°C drying oven, and dried to con- stant weight (total dry weight). To determine the ash weight, organic material in each sample was combusted in a muffle furnace for 4 h at 450°C. Ashed samples were weighed, and 1-h reashing cycles repeated if the samples gained weight by hydration. Completely ashed materials do not adsorb water (Gnaiger and Bitterlich, 1 984). This protocol avoids the inaccuracies associated with decom- position of CaCO, due to prolonged heating (Paine, 1 97 1 ). All weight measurements were made with a Cahn Model 29 electrobalance (accurate to 0.1 Mg). The amount of dry organic weight could then be calculated as the difference between the total dry weight and the ash weight. These values (dry organic and ash weight) were divided by the number ofoocytes, or larvae, in each sam- ple and the data expressed as ng material per individual. Biochemical composition To remove any residual fluid, samples used for deter- mination of biochemical composition were lyophilized 239 for at least 8 h at a pressure of 0.2 mm Hg. The lyophi- lized samples were then sonicated in 1 ml of glass dis- tilled water by ultrasonic disruption (Sonics and Materi- als Brand, Model VC 40 fitted with a microprobe). The homogenates were centrifuged for 10 min at 12,200 X g and then sonicated a second time. Care was taken throughout the sonication procedure not to heat the samples. To determine whether the observed changes in the dry organic weight of the oocytes and larvae were ac- counted for by corresponding changes in biochemical composition, the amount of carbohydrate, lipid, and protein was measured. The biochemical composition of the oocytes and larvae was determined using the frac- tionation scheme devised by Holland and Gabbott (1971), as modified by Mann and Gallager (1985), with two additional alterations: (i) homogenates were ex- tracted in trichloroacetic acid (TCA) for 20 min at - 1 0°C and (ii) the total protein was assayed using Coomassie Brilliant Blue G-250 (BioRad Laboratories) as a colori- metric reagent. The TCA-insoluble pellet was dissolved in 500 Ml of 1.0 MNaOH by heating at 60°C for 30 min. The alkaline protein solution was then acidified with 300 Ml of 1.67 M HC1, and 200 Ml of the concentrated Coo- massie Blue dye solution was then added. The absor- bance of the samples (at 595 nm) was determined after 10 min, and no later than 60 min, following addition of the dye solution. Rates of larval respiration The metabolic rates of larvae, of different ages, were measured as the rates of oxygen consumption. Larvae were placed in a conical analyzer cup (Curtin Matheson; 2 ml total volume, precalibrated to 345 n\) with filtered, then autoclaved, seawater. A Clark-type oxygen elec- trode (Model E5057, Radiometer Copenhagen) was placed into the cup, and excess air and seawater were dis- charged through a small purge hole melted into the cup. The purge hole was sealed by the membrane o-ring. The electrode was connected to a blood-gas analyzer (Model PHM 73; Radiometer Copenhagen) and, following a 5- min equilibration period, the change in the partial pres- sure of oxygen (mm Hg) was monitored for 20 to 30 min. During all experiments, the respiration chamber and the electrode were immersed in a 17°C water bath (Model RDL 20; Precision Instruments; ±0.02°C). Prior to the experiments with larvae, the oxygen consumption rate of the electrode itself was determined under the same conditions used for experiments with larvae. At the end of each experiment, the larvae were removed from the respiration chamber and counted (50 to 1 50 larvae). The rate of change in the partial pressure of oxygen was cor- rected for any self-consumption by the electrode, and 240 W. B. JAECKLE AND D. T. MANAHAN Table I Changes in the amount oj dry organic' weight between oocytes (day 0) and ve/iger lan'ae (day 5,6 or 7), sampled al the end oj the larval life span (competent to settle), during the complete development o/"Haliotis rufescens Culture 1 Culture 2 Culture 3 Culture 4 Culture 5 Age Age Age Age Age (day) (day) (day) (day) (day) 0 1.36 ±0.02 0 1.41 ±0.06 0 1.65 ±0.03 0 1.19 + 0.04 0 1.39 ±0.05 7 1 . 1 7 ± 0.05 6 1.42 ±0.08 6 1.67 ±0.03 6 1.20 + 0.02 5 1.59 + 0.09 (-14.0%) Percent difference between beginning and end of each culture (+0.7%) (+1.2%) (+0.8%) (+14.4%) Cultures 1 -5 refer to batches of larvae reared from gametes obtained from five separate spawings. All data are presented as the mean dry organic weight per individual (jjg ± 1 standard error of the mean). The percent difference, given at the bottom of the table, represents the net change in do- organic weight for each culture. calculated as mm Hg larva 'h '. This depletion rate was converted to moles of oxygen consumed by calibrating the electrode relative to the amount of oxygen in isother- mal seawater, as determined by the Winkler titration method ( Parsons et al,, 1984). Results Changes in weight during the lan'al life span Figure 1 shows the changes in dry organic weight, and ash weight, from an unfertilized oocyte (day 0) to a larva competent to settle (day 7). In this culture (Culture 1). and all others studied (see Table I, Cultures 1 to 5), there was a continual linear increase in ash weight during lar- val development. There was a statistically significant net increase in dry organic weight from the oocyte (day 0) to the newly formed veliger larva (day 2), as can be seen in Figure 1 (Variance ratio, VR = 25.0, F00,n[i.4] = 22.9). Similar increases during this period of development (day 0 to day 2) were seen in four other cultures. For Culture 1 (Fig. 1 ), there was a statistically significant net decrease from the oocyte, at 1.36 ± 0.02 /*g, to the 7-day-old veli- ger at 1.17 + 0.05 Mg(VR = 18.3,Fo.ol[i.8]= 12.2). Of the 5 cultures, reared in an identical manner (Table I), this culture (Culture 1) was the only one that showed a net decrease in dry organic weight during the larval life span. For Cultures 2, 3, and 4 there was no statistically signifi- cant change in organic weight between the oocyte and the last larval stage sampled. Larvae from Culture 5 had a net increase in organic weight, from 1.39 ±0.05 jig(day 0)to 1.59±0.09Mg(day 5). Changes in biochemical composition and energy content The changes in the biochemical composition of larvae, for the entire lifespan of two independent cultures (Cul- tures 1 and 5), are presented in Figure 2. The correspond- ing weight and energy values, for each total lipid and pro- tein fraction, are given in Table II. The amount of total carbohydrate was always below the level of detection (0. 1 /ug per sample using a glucose standard) and, thus, a value for carbohydrate could not be included in calculations of energy budgets. The increase in dry organic weight dur- ing the first two days of development of larvae from Cul- ture 1 (see Fig. 1 ) was explained by the combined in- creases of both the total lipid and protein fractions (Table II, Culture 1 , Day 0 to 2). For Culture 5, there was also a significant increase in the dry organic weight from the oocyte (1.39 ± 0.05 ^g) to the 2-day-old veliger larva (1.81 ± 0.09 Mg)- Again, this initial growth could be ac- 0123456 Age (days) Figure 1. Change in the dry organic weight and ash weight (mean + 1 standard error of the mean) during the complete development of Haliotis rufescens, from the unfertilized oocyte (day 0) to a larva fully competent to settle (day 7). Diagrams of larval shape are redrawn from Leighton(1972). ENERGY IMBALANCE IN NONFEEDING LARVAE 241 Culture 1 o> Culture 5 245 Age (days) Figure 2. Changes in the biochemical composition during the com- plete development of Haliotis rufescens, from the unfertilized oocyte (day 0) to a larva fully competent to settle. Larvae in Culture 1 took 7 days to reach competence; those in Culture 5 took 5 days. For each sample the dry' organic weight is represented by the sum of the weights of protein, total lipid, and the uncharacterized fraction (remainder). The cross-hatched portions of the bars represent protein; the solid por- tion represents total lipid; the open portions represent the remainder fraction. counted for by the net changes in lipid and protein. There was a slight decrease in protein of —0.06 ^g (day 0 to 2), and a large increase in total lipid of +0.45 /ig during the same time period (Table II, Culture 5, Day 0 to 2). Following the formation of the veliger larva (day 2), the changes in protein for larvae from both Cultures 1 and 5 were qualitatively similar. There was a 65% decrease (0.30 ^g) in protein content of larvae from Culture 1 (Day 2 to 7) and a 32% decrease in the amount of protein in larvae from Culture 5 (Day 2 to 5). For lipid, the net changes were similar between Cul- tures 1 and 5 (day 2 to 5; day 2 to 7. respectively), follow- ing the net increases in total lipid observed during the period of day 0 to day 2 for both cultures (Table II). Lar- vae from Culture 1 had a decrease of 0. 1 3 jug of lipid (21%) from day 2 to day 7. Those in Culture 5 also had a decrease in lipid content of 0.34 ^g (26%) from day 2 to day 5. The energy equivalents of the changes in weight for lipid and protein were calculated using the combustion enthalpy values for each fraction (lipid = 39.5 kJ/g, pro- tein = 24.0 kJ/g; Gnaiger, 1983). As observed by other workers (e.g., Holland and Gabbott, 1971), the sum of the weights of the measured biochemical components per individual (Fig. 2, filled bars), did not equal the dry organic weight. The "remainder" fraction (see Fig. 2, open bars), is denned as the difference between the dry organic weight and the sum of the measured biochemical components. To account for the energy represented by this material, a value of 27.0 kJ/g was given for the en- thalpy of combustion of this uncharacterized fraction. This value is equal to the average of the combustion en- thalpies of carbohydrate (17.5 kJ/g), lipid. and protein. In the absence of biochemical information on what com- prises the remainder fraction, an average value for en- thalpy of combustion was taken to be representative of the low molecular weight compounds that probably make up the majority of the remainder fraction. In Cul- ture 1, the energy equivalent of the uncharacterized ma- terial (see "remainder," Table II) decreased from 14.85 mJ, at day 0, to 1 3.77 mJ by day 7. Larvae in Culture 5 had 5.40 mJ, at day 0, and increased to 1 1.61 mJ per larva by day 5. The metabolic rates of larvae The rates of oxygen consumption by veliger larvae of Haliotis rufescens are presented in Figure 3. The data presented in this figure are based upon the results of 47 independent determinations, obtained with larvae from 4 different cultures. The inset in Figure 3 shows the mean respiratory rate per larva, at each day of development examined. The average respiratory rate ranged from a low of 58 ± 7.2, for a 2-day-old larva, to a high of 93 ± 1 3 pmol O2 larva"' h~' for a 3-day-old larva. Discussion Larvae of Haliotis rufescens are presumed to be non- feeding because they cannot capture particulate foods. Thus, a net loss of the organic weight contained in the oocyte would be expected during larval development. However, for four out of five cultures of larvae studied, there was a net increase in dry organic weight during early development from an oocyte (day 0) to a newly formed veliger larva at day 2 (e.g.. Fig. 1, day 0 to 2). These increases in dry organic weight were explained by changes in the amounts of protein and lipid during the same time period (Table II). Similarly, in only one out 242 W. B. JAECKLE AND D. T. MANAHAN Table II Changes in drv organic weight and energy content during the complete larval development of Haliotis ruf'escens Culture 1 Weight of each biochemical fraction Age (day) 0 4 Dry organic weight Mean ± 1 SE 1. 36 ±0.02 Protein Mean ± 1 SE 0.29 + 0.00 Total lipid Mean±lSE 0.52 ±0.02 1.61 ±0.09 0.46 + 0.01 0.63 ± 0.03 1.28 ±0.01 0.36 + 0.01 0.56 ±0.07 1.34 + 0.04 0.34 ±0.01 0.36 + 0.05 Energy content of each biochemical fraction Protein (mj) Total lipid(mJ) Remainder (mj) 6.96 20.54 14.85 11.04 24.89 14.04 8.64 22.12 9.72 Culture 5 8.16 14.22 17.29 1 .27 ± 0.03 0.37 ± 0.03 0.41 ±0.02 16.20 13.23 1.17+0.05 0.16 + 0.01 0.50 + 0.03 3.84 19.75 13.77 Weight of each biochemical fraction Age (day) 0 Dry organic weight Mean + 1 SE Protein Mean± 1 SE 0.31 ±0.00 Total lipid Mean±lSE 0.88 ± 0.09 1.81 ±0.09 0.25 + 0.01 1.33 ±0.05 1.61 +0.08 0.21 ±0.01 0.93 ± 0.04 1.59 ±0.09 0.17 + 0.00 0.99 ± 0.05 Energy content of each biochemical fraction Protein (mj) Total lipid (mJ) Remainder (mj) 7.44 34.76 5.40 6.00 52.54 6.21 5.04 6.74 12.69 4.08 39.11 11.61 Larvae in Cultures I and 5 were reared from gametes obtained from separate spawnings. The weights of each protein and total lipid fractions were converted to an equivalent energy by multiplying each weight fraction by its value for enthalpy of combustion (total lipid = 39.5 kJ/g; protein = 24.0 kJ/g). The "remainder" fraction was calculated as the difference between the dry organic weight and the sum of the measured biochemical fractions. The energy value for this uncharactenzed fraction was taken to be the average (27.0 kJ/g) of the combustion enthalpies for carbohydrate ( 1 7.5 kJ/g). lipid, and protein. All weight values are presented as the mean (^g) ± 1 standard error of the mean. of five independent cultures of larvae was there a statisti- cally significant decrease in dry organic weight from the oocyte to a competent larva (Culture 1, Fig. 1 and Table I). For the other four cultures (Table I), there was either no significant net change (Cultures 2, 3 and 4), or an in- crease in dry organic weight (Culture 5). The net changes in dry organic weight, from the oocyte to a larva that is competent to settle, ranged from a de- crease of 14% (Culture 1 ) to an increase of 14% (Culture 5) (Table I). An energy budget was constructed for each of these cultures because they represented the maximum range of weight changes observed. The amount of energy required by larvae in Cultures 1 and 5 was calculated from their respiratory rates (Fig. 3). The total rate of oxy- gen consumed by larvae was calculated for their entire life span. Where no measurements were made for a par- ticular day, a value was estimated based on the rate mea- sured for the nearest day. The energy equivalent to the rate of oxygen consumption was then compared to the available energy released by the catabolism of internal biochemical reserves (Table II). This comparison is given in Table III. For both cultures, the energy available from the net changes in lipid, protein, and the "remainder" fractions were always insufficient to account for the en- ergy requirements of larval development. For Culture 1, there was a net decrease in all biochemical components ENERGY IMBALANCE IN NONFEEDING LARVAE 243 . D) 200- 180- 160- 120- 100- 80- i 100 i ^ I m « 20 •5 1 - - ~~ •:•:•] 2 3456 Aoe (Oays) D [&> CO cr cP O* 60- f i * 40- " 20- n - » 0123456 Age (days) Figure 3. Metabolic rates of veliger larvae of Haliotis rufescens. Data are presented from four cultures, obtained from independent spawnings. The respiratory rates of larvae from Culture I are depicted as solid rectangles; larvae from Culture 5 are shown as solid circles. Other symbols represent larvae whose rates of respiration were not dis- cussed in the text. The inset shows the mean respiratory rate ( + 1 stan- dard error of the mean (averaged for all larvae for each day. and the sum of the equivalent energies ( -4.99 mJ ) equals the amount of energy available to the larva from its re- serves. In this case, 71% of the metabolic needs (+6.99 mJ) could be accounted for by the energy available from internal reserves (i.e., 4.99/6.99 = 0.71, Table III). For Culture 5, there was a net increase in weight during de- velopment for both the lipid and remainder fractions (Table II). Thus, these increases represent an energy re- quirement for the larva. In these larvae, the amount of required energy was (i) the sum of the energy equivalent of the total amount of oxygen consumed over 5 days (+2.93 mJ) and (ii) the energy equivalent to the net in- creases in the biochemical components (+10.56 mJ). The amount of energy available from the net decrease in protein (-3.36 mJ) represented only 25% of the total energy required during the 5 days of development (i.e., 3.36/1 3.49 = 0.25, Table III). In general, larvae of Haliotis rufescens do not show a decrease in dry organic weight and energy content (Fig. 1, Tables I, II) during development. This finding is sur- prising in view of the fact that these larvae cannot use paniculate foods and were cultured in natural seawater that had been passed only through a 0.2-^m (pore size) filter. Yet our results support previous findings on the net biochemical and energy changes that occurred in inver- tebrates unable to feed during development. When the energy content was compared between eggs, and later de- velopmental stages, for six species of echinoderms col- lected from temperate and antarctic environments (Turner and Rutherford, 1976; Lawrence et a/.. 1984; McClintock and Pearse, 1986), the results showed little net change in energy content between eggs and juveniles. Also, in a study of the energetics of embryonic develop- ment in the teleost Sehastes schlege/i, Boehlert et al. (1986) reported that the energy provided by the catabo- lism of endogenous reserves represented only 55% of the metabolic demand (oxygen consumption). In this latter case, the source of exogenous energy was presumed to be provided as dissolved organic compounds in ovarian fluid. A possible alternative explanation for the observed im- balance between the rate of use of energy reserves and the measured metabolic rates of larvae, is that the latter were affected by some experimental artifact and were too high during our experiments. The average respiratory rate of all larvae of Haliotis rufescens measured (from Fig. 3) was 84 pmol O: larva"1 h~'. The dry organic weight, averaged for all veliger larvae (Cultures 1 and 5, Table II) was 1.5 ^g- Thus, the average metabolic rate was 56 /nmol O2 g~' h ' at a temperature of 17°C. This value is at the lower end of the metabolic rates of marine invertebrate larvae (see Crisp, 1976) and is lower than that reported for a feeding larvae of the gastropod Ilya- nassa obselotus ( 165 jtmol Oi g~' h~', recalculated from Pechenik, 1980, assuming a Q,0 of 2). Hence, the possi- bility that our reported values are artificially high is not viable. However, these analyses do suggest that within the Gastropoda, lecithotrophic larvae may have a lower metabolic rate than planktotrophic veligers. This differ- ence may be due to the larger size of the velum of plank- totrophic larvae when compared to lecithotrophic forms (Fretter and Graham, 1962). This suggestion is also sup- ported by the work of Erickson (1984), who showed that differences in the respiratory rates of veliger larvae of the gastropod Strombm gigas corresponded with the extent of velar lobe expansion. The calculated values given in Table III show that the energy reserves, initially present in the oocyte, are not being used by these larvae to supply their metabolic needs. However, this does not mean that there is in- sufficient energy in the oocyte to meet the metabolic de- mands of larval development. The energy content per oocyte (day 0) from Culture 1 was 42.35 mJ and 47.60 mJ for Culture 5. The amount of energy equivalent to the total number of moles of oxygen consumed (Fig. 3) during the complete development of larvae from these two cultures was 6.99 mJ and 2.93 mJ per larva for Cul- tures 1 and 5, respectively (Table III). Therefore, the amount of energy initially present in the oocyte exceeded the metabolic demand by 6-fold (i.e., 42.35/6.99 = 6) for Culture 1 and 16-fold (i.e., 47.60/2.93 = 16) for Culture 5. Even though there is more than enough energy con- tained in an oocyte to meet the metabolic demand, the 244 W. B. JAECKLE AND D. T. MANAHAN Table HI An energy budget for the complete lan-al development of Haliotis rufescens Symbols used: "+" = required energy, and "— ' = available energy. Culture 1 Culture 5 Available energy (decreases in biochemical content) Protein (mJ) Lipid (mJ) Remainder (mJ) Sum of available energy -3.12 -0.79 -1.08 -4.99 Available energy (decreases in biochemical content) Protein (mJ) Lipid (mJ) Remainder (mJ) Sum of available energy -3.36 +4.35 +6.21 -3.36 Required energy (metabolic rate over 7 days) 13.27nmolO;/larva = (increases in biochemical content) None Sum of required energy +6.99 mJ 0.00 mJ +6.99 Required energy (metabolic rate over 5 days) 5.56 nmol O/2 /larva = (increases in biochemical content) Lipid and remainder Sum of required energy +2.93 mJ + 10.56 mJ + 13.49mJ Energy balance (available/required) [(-4.99)/(+6.99)] = 71%* Energy balance (available/ required) 13.49)] = 25%* 'Contribution from energy stores to metabolic demand Larvae in Cultures 1 and 5 were reared from gametes obtained from separate spawnings. The net change in total energy per individual, over the course of development, was taken from the data given in Table II (Culture I, day 0-7; Culture 5, day 0-5). The metabolic rates of the larvae for each culture were calculated from the data provided in Figure 3. To relate the energy demand of development to the energy made available from the catabolism of internal stores, the total amount of oxygen consumed was converted to an energy equivalent. The energy reserve having the largest net decrease throughout development was protein and, therefore, the oxyenthalpic equivalent of this reserve was used (protein 527 kJ/mol O,,fromGnaiger, 1983). conclusions drawn from the energy budgets given in Ta- ble III are that larvae do not fully use this source of en- ergy. Similar conclusions have been made by Lawrence el al. (1984) who reported that there was little change in energy content, between the egg and the juvenile, for the echinoid Abatus cordatm and the asteroid Anasterias perreiri. These authors suggested that the function of the large energy content of these eggs, 1 5.5 J/egg and 39.6 J/ egg, respectively, was to produce a large juvenile rather than supply a large amount of energy for development. Also, Emlet et al. (1987) reported that, in general, juve- nile asteroids that develop from nonfeeding larvae are larger than juveniles produced from feeding larvae. Thus, for development of Haliotis rufescens, the adap- tive significance of the large amount of energy initially present in the oocyte may be related to juvenile survivor- ship, rather than energy metabolism during larval devel- opment. If larvae of Haliotis rufescens are not using their inter- nal reserves to fully meet the energy demands of develop- ment, there must be an input of material from the envi- ronment. Veliger larvae of Haliotis spp. cannot ingest paniculate foods and, therefore, the most likely source of nutrients for these larvae would be in a dissolved form. Jaeckle and Manahan (1989) showed that veliger larvae can take up dissolved free amino acids from seawater, and that the rate of amino acid uptake, from a concentra- tion of 1 .6 nM, was sufficient to supply 55% of the meta- bolic demand of the larvae. Larvae from Culture 5 had the higher energy imbal- ance, where only 25% of the metabolic cost of develop- ment could be accounted for by the use of energy stores. Could the rate of DOM uptake account for the missing energy observed for these larvae? The seawater used to culture the larvae was natural and, thus, approximated in situ concentrations of DOM. The calculations given below suggest that DOM could supply this missing en- ergy. The average rate of amino acid uptake from a con- centration of 1.6 juA/ (Jaeckle and Manahan, 1989) was 5.9 pmol amino acid larva"1 h~'. Based on the average molecular weight of the amino acids used ( 140 g/mol), this equaled 826 pg larva"' h~', or 99 ng larva"1 (5- daysr1. Giving DOM a value for combustion enthalpy based on the average of carbohydrate, protein, and lipid (27.0 kJ/g, see Table II), 99 ng of DOM would equal 2.7 mJ/larva over a 5-day period. From Table III, it can be seen that an additional 10.13 mJ (13.49-3.36 = 10.13) was needed to balance the required energy. Thus, dis- ENERGY IMBALANCE IN NONFEEDING LARVAE 245 solved amino acids could contribute 27% of the missing energy (2.7/10.13 = 0.27). Dissolved amino acids repre- sent less than 1% of the total pool of DOM in seawater (Williams, 1975). Thus, if only 4% of the total pool of DOM was used by larvae, at rates similar to those ob- served for the amino acid fraction, then 100% of the missing energy would be supplied from DOM. The re- quirement that 4% of the DOM has to be used is not an unreasonable estimate, given the wide variety of biologi- cally available organic compounds that make up the pool of DOM in seawater. The increase in both organic weight and specific bio- chemical components in these larvae suggests the ques- tion: are lecithotrophic invertebrate larvae really "non- feeding"? Larvae of Haliotis rufescens can take up and metabolize dissolved free amino acids from seawater (Jaeckle and Manahan, 1989) and increase in organic weight (this study). This suggests that these "nonfeeding" larvae are feeding. However, the source of exogenous food exists in a dissolved form. Acknowledgments We would like to thank John McMullen (Ab Lab) for generously providing fertilized oocytes of Haliotis rufes- cens. This research was supported by a grant from NOAA, Office of Sea Grant (U.S.C. grant R/RD-27) and, in part, by a grant from the National Science Foundation (OCE-86-0889). Literature Cited Boehlert, G. W., M. Kusakari, M. Shimizu, and J. Vamada. 1986. Energetics during embryonic development of kurosoi, Seb- asies schlegeli Hilgedorf. / Exp. Mar. Bio/. Ecol. 101 : 239-256. Carlucci, A. F., D. B. Craven, and S. M. Henrichs. 1984. Diel produc- tion and microheterotrophic utilization of dissolved free amino acids in waters off southern California. Appl. Environ. Microbiol. 48: 165-170. Chia, F. S. 1974. Classification and adaptive significance of develop- mental patterns in marine invertebrates. Thalass. Jugoslav. 10: 121-130. Cognetti, G. 1982. Nutrition of embryos. Pp. 469-477 in Echinoderm Nutrition. M. Jangoux and J. M. Lawrence, eds. A. A. Balkema, Rotterdam. Crisp, D. J. 1976. The role of the pelagic larva. Pp. 145-155 in Per- spectives in Experimental Biology, Vol. 1 , P. Spencer Davis, ed. Per- gamon Press, Oxford and New York. Davis, J. P., and G. C. Stephens. 1984. Uptake of free amino acids by bactena-free larvae of the sand dollar Dendraster excentricus. Am. J. Physiol. 247: R733-R739. DeFlaun, M. F., J. H. Paul, and W. H. Jeffrey. 1987. Distribution and molecular weight of dissolved DNA in subtropical and estua- rine and oceanic environments. Mar. Ecol. Prog. Ser. 36: 65-75. Emlet, R. E., L. R. McEdward, and R. R. Strathmann. 1987. Echinoderm larval ecology viewed from the egg. Pp. 55- 136 in Echinoderm Studies. Vol. 2, M. Jangoux and J. M. Law- rence, eds. A. A. Balkema, Rotterdam. Erickson, J. T. 1984. Gradient-diver respirometry applied to free- swimming larvae of the queen conch Strombus gigas Linne. J Shellfish Res. 4:5-15. Fretter, V., and A. Graham. 1962. British Prosobranch Molluscs: The Functional Anatomy and Ecology. The Ray Society. London. 755 pp. Gnaiger, E. 1983. Calculation of energetic and biochemical equiva- lents of respiratory oxygen consumption. Pp. 337-345 in Polaro- graphic Oxygen Sensors E. Gnaiger and H. Forstner, eds. Springer- Verlag, Berlin. Gnaiger, E., and G. Bitterlich. 1984. Proximate biochemical compo- sition and caloric concept calculated from elemental CHN analysis: a stoichiometric concept. Oecologia 62: 289-298. Holland, D. L., and P. A. Gabbott. 1971. A microanalytical scheme for the determination of protein, carbohydrate, lipid and RNA lev- els in marine invertebrate larvae. J. Mar. Biol. Assoc. U.K. 51: 659- 668. Holland, D. L., and B. E. Spencer. 1973. Biochemical changes in fed and starved oysters, Ostrea edulia L. during larval development, metamorphosis and early spat growth. /. Mar. Biol. Assoc. U.K 53: 287-298. Jaeckle, W. B., and D. T. Manahan. 1989. Feeding by a "nonfeed- ing" larva: uptake of dissolved amino acids from seawater by leci- thotrophic larvae of the gastropod Haliotis rufescens Swainson. Mar. Biol (in press). Lawrence, J. M., J. B. McClintock, and A. Guille. 1984. Organic level and caloric content of eggs of brooding asteroids and an echi- noid (Echinodermata) from Kerguelen (South Indian Ocean). Int. J Invert. Repro. Dev. 7: 249-257. Leighton, D. L. 1972. Laboratory observations on the early growth of the abalone Haliotis sorensoni. and the effects of temperature on larval development and setting success. Fish. Bull. 70: 373-38 1 . Lucas, M. I., G. Walker, D. L. Holland, and D. J. Crisp. 1979. An energy budget for the free-swimming and metamorphosing larvae ofBalanus balanoides (Crustacea: Cirripedia). Mar. Biol. 55: 221- 229. McClintock, J. B., and J. S. Pearse. 1986. Organic and energetic con- tent of eggs and juveniles of antarctic echinoids and asteroids with lecithotrophic development. Comp. Biochem. Physiol. 85A: 341- 345. MacKinnon. M. A. 1981. The measurement of organic carbon in sea- water. Pp. 4 1 5-444 in Marine Organic Chemistry: Evolution, Com- position, Interactions and Chemistry of Organic Matter in Seawa- ter. E. K. Duurmsa and R. Dawson, eds. Elsevier Publ. Co.. Amster- dam. Manahan, D. T. 1983. The uptake and metabolism of dissolved amino acids by bivalve larvae. Biol. Bull. 164: 236-250. Manahan, D. T. 1989. Amino acid fluxes to and from seawater in axenic veliger larvae of a bivalve (Crassostrea gigas). Afar. Ecol. Prog. Ser. 53: 247-255. Manahan, D. T., J. P. Davis, and G. C. Stephens. 1983. Bacteria- free sea urchin larvae: selective uptake of neutral amino acids from seawater. Science 220: 204-206. Manahan, D. T., and G. C. Stephens. 1983. The use of high-perfor- mance liquid chromatography to measure dissolve organic com- pounds in bivalve aquaculture systems. Aquaculture 32: 339-346. Mann, R., and S. J. Gallager. 1984. Physiology of the wood boring bivalve mollusk, Martesia cuneiformis Say. Biol Bull. 166: 167- 177. Mann, R., and S. Gallager. 1985. Physiological and biochemical en- ergetics of larvae of Teredo navalis L. and Bankia gouldi (Bartsch) (Bivalvia: Teredimdae). ./. Exp. Mar Biol. Ecol. 85: 21 1-228. 246 W. B. JAECKLE AND D. T. MANAHAN Millar, R. H., and J. M. Scott. 1967. The larva of the oyster Oslrea edulix during starvation. / Mar. Bio/. Assac. U. K. 47: 475-484. Mopper, K., R. Dawson, G. Liebezeit, and V. Ittekkot. 1980. The monosaccharide spectra of natural waters. Mar. Chem. 10: 55-66. Paine, R. T. 1971. The measurement and application of the calorie to ecological problems. Ann. Rev. Syxl. Ecol. 2: 145-164. Parkes, R., and J. Taylor. 1983. Analysis of volatile acids by ion ex- clusion chromatography. with special reference to marine pore wa- ter. Afar Biol. 77: 113-118. Parsons, T. 1975. Paniculate organic carbon in the sea. Pp. 365-386 in Chemical Oceanography. Vol. 2, P. J. Riley and G. Skirrow, eds. Academic Press, London. Parsons, T. R., Y. Maita, and C. A. Lalli. 1984. A Manual of Chemi- cal and Biological Methods for Seawater Analysis. Pergamon Press, Oxford. 173pp. Pechenik, J. A. 1980. Growth and energy balance during the larval lives of three prosobranch gastropods. J Exp. Mar. Biol Ecol 44: 1-28. Reish, D. J., and G. C. Stephens. 1969. Uptake of organic material by aquatic invertebrates. V. The influence of age on the uptake of glycine-HC by the polychaete Neanlhes arenaceodentata. Mar. Biol. 3: 352-355. ' Sugimura, V., and Y. Suzuki. 1988. A high-temperature catalytic oxi- dation method for the determination of non-volatile dissolved or- ganic carbon in seawater by direct injection of a liquid sample. Mar. Chem. 24: 105-131. Testerman, J. K. 1972. Accumulation of free fatty acids from sea wa- ter by marine invertebrates. Biol. Bull. 142: 160-177. Thorson, G. 1946. Reproduction and larval development of Danish marine bottom invertebrates with special reference to the plank- tonic larvae of the Sound (Oresund). Medd. fiomm. Damn. Fisker- iog. Havunders. Ser. Plankton 4: 1-523. Turner, R. L., and J. C. Rutherford. 1976. Organic, inorganic and caloric composition of eggs, pentaculae, and adults of the brooding sea cucumber Cucumaria curala Cowles (Echinodermata: Holo- thuroidea). J. Exp. Mar. Biol. Ecol. 24: 49-60. Williams, P. J. LeB. 1975. Biological and chemical aspects of dis- solved organic material in seawater. Pp. 307-364 in Chemical Oceanography, Vol. 2, P. J. Riley and G. Skirrow, eds. Academic Press, London. Reference: Bioi Bull 111: 247-253. (October, 1989) Cryopreservation of Spermatophores and Seminal Plasma of the Edible Crab Scylla serrata C. JEYALECTUMIE AND T. SUBRAMONIAM Unit of Invertebrate Reproduction. Department of Zoology. University of Madras, Guindy Campus. Madras 600 025. India Abstract. This paper describes the development of a suitable biotechnology to cryopreserve the spermato- phores of the edible crab Scylla serrata in a viable condi- tion. Three temperatures (-4°C, -79°C. and -196°C) were chosen to preserve the spermatophores and seminal plasma, collected from the middle region of vas deferens of mature male crabs, for 30 days. Sperm viability was determined by the eosin-nigrosin dye exclusion method, as applied to entire spermatophores. Of the three temper- atures tested, the maximum percentage of sperm viabil- ity was obtained at -196°C, whereas it significantly de- creased at — 4°C. Biochemical alterations of the major substrates such as proteins, carbohydrates, and lipids, as well as the enzyme Lactate Dehydrogenase (LDH) oc- curred only at — 4°C, reflecting their use in the metabolic activities of the spermatozoa contained within the sper- matophores. At -4°C, the TCA-soluble total free sugars increased in correspondence with a decline in the bound sugars, suggesting that the latter may be used rapidly dur- ing sperm storage. Our data also suggest that, at -79°C and - 196°C, the frozen spermatozoa retain viability but do not exhibit metabolic activity. To investigate the role of cryoprotectants in prevent- ing damage to the sperm cells/spermatophores during Cryopreservation, four cryoprotectants, glycerol, di- methyl sulfoxide (DMSO), trehalose, and DMSO + tre- halose combination, were tested. Using the phosphate buffer as the standard diluent, glycerol gave the best re- sult. When used alone, trehalose gave only a low sperm survival, but in combination with DMSO. it gave an in- creased viability that equalled the result with glycerol. We recommend that glycerol is the best cryoprotectant inasmuch as the biochemical alterations in the glycerol Received 8 December 1988; accepted 3 1 July 1989. DMSO— Dimethyl sulfoxide. medium is less, compared to that of DMSO + trehalose. When used alone, DMSO may be more toxic to the sperm cells as it gave a low viability value even at - 1 96°C and -79°C. Introduction Cryopreservation of gametes is a common method practiced in conjunction with artificial insemination in mammals (Leverage el a/., 1972). Many attempts have also been made in this respect to cryopreserve spermato- zoa of the teleost fishes (Horton and Ott, 1976). Al- though decapod crustaceans have become important in aquaculture in recent years, Cryopreservation techniques have not been employed for in vitro fertilization in prawns, lobsters, or crabs. However, using electroejacu- lation techniques, extraction of intact spermatophores and their attachment onto females have been accom- plished in several decapods like shrimps and lobsters (Chow el a/.. 1985; Ishida el a/.. 1986). On the other hand, no work has been done on the artificial insemina- tion or in vitro fertilization of crabs. Brachyuran crabs produce simple spermatophores that are carried in the fluid medium of seminal plasma. Therefore, any attempt to cryopreserve male gametes will include both the sper- matophores and the seminal plasma. Previous studies on the seminal plasma of crustaceans such as crabs ( Jeyalec- tumie and Subramoniam, 1987) and cirripede (Barnes, 1962) indicated a similarity, in terms of providing ener- gy-yielding substrates, with the semen of mammals. The objective of the present study was to determine a suitable temperature and an extender to cryopreserve the sper- matophores as well as seminal plasma of the crab, Scylla serrata. Fluctuations in the biochemical composition of the spermatophores and seminal plasma during cryo- preservation at selected subzero temperature conditions 247 248 C. JEYALECTUMIE AND T. SUBRAMONIAM were also followed. LDH enzyme showed significant fluctuation during spermatophore storage in the crab, Paratelphusa hydrodromous (Jeyalectumie and Subra- moniam, 1987) and hence, in the present study, fluctua- tions in the LDH activity of seminal plasma of Scy/la serrata during cryopreservation were also determined. Materials and Methods Collection of sample Seminal plasma containing vesiculate spermatophores was collected in a petri dish after puncturing the thin wall of the mid vas deferens. The samples collected were used for the cryopreservation studies. Initial cooling, dilution, and freezing The diluent for spermatophore preservation was pre- pared, according to Behlmer and Brown ( 1 984), immedi- ately before use by mixing 25 ml of 0.4 M NaCl/0. 1 M glycine, 4 ml of 0.028 M NaH:PO4/0.072 A/Na2HPO4, and 5 ml ofglycerol. Apart from this, three other diluents were prepared by substituting glycerol with trehalose (0.25 M), DMSO (5%), and a combination of DMSO (5%) and trehalose (0.25 A/). These are the cryoprotec- tants commonly used for various biological systems (Lovelock and Polge, 1954; Hughes, 1973; Asahina and Takahashi. 1978; Zell et a!., 1979; Behlmer and Brown, 1 984; Chow etal. 1985; Stephens, 1986; Anchordogy et ai, 1988). Analytical grade glycerol, trehalose, and DMSO were purchased from Sarabai Chemicals (India), BDH (England), and SD's (India), respectively. The freshly collected sample was diluted to 20% sper- matophore suspension by mixing four volumes of dilu- ent to one volume of seminal plasma. Similarly, the sper- matophoric suspensions in four different diluents were prepared separately. After dilution, the seminal plasma was aspirated into 0.5-ml semen storage straws having one end sealed. After filling, the open side of the straw was sealed with polyvinyl alcohol. Immediately after this process, all the straws containing the sample were kept for equilibration at 4°C for 16 h. Prior to freezing at - 196°C in liquid nitrogen, the straws were kept exposed to the liquid nitrogen vapor for an hour and then im- mersed completely in the liquid nitrogen. Similarly, for -79°C, the straws were exposed to gaseous Co: for an hour, and then placed directly onto the dry ice. For — 4°C, the straws were kept inside the freezer. All the above samples were stored for 30 days. Fresh, unfrozen semen, with appropriate diluents, were used as a control. Our attempts to freeze the seminal plasma without adding the diluent in the three temperature conditions did not yield fruitful results. The 'dry' seminal plasma becomes coagulated after its transfer to the subzero con- dition. A similar problem was also mentioned by Asah- ina and Takahashi (1978) and Yoo et al. (1987), who found that removal of cryoprotectant led to agglutina- tion of the spermatozoa of both sea urchins and salmon. Storing semen in diluent without adding any cryoprotec- tant resulted in the clumping of the spermatophores after thawing, thus precluding the preparation of good smear for the viability study. Moreover, in the previous studies on cryopreservation of male gametes in fishes and inver- tebrates such as horseshoe crabs, molluscs, and sea ur- chins, only the pre-frozen semen has been used as the control (Asahina and Takahashi, 1978; Zell et al., 1979; Withler, 1982; Behlmer and Brown, 1984; Kurokura et al., 1986). Thawing Thawing was accomplished by immersing the straws containing frozen seminal plasma in tap water (room temperature). The thawed seminal plasma was collected from the straws into clean test tubes by cutting open both the sealed ends. Evaluation of viability ofcryopreserved spermatophores To evaluate sperm viability inside the spermatophore, the eosin-nigrosin staining method of Zaneveld and Po- lakoski (1977) was used. A smear was prepared by mix- ing one drop of thawed seminal plasma with one drop of 0.5% eosin and two drops of 10% nigrosin. These slides, after being air-dried, were examined using a brightfield microscope. Care was taken to complete the observation within 2 min of smear preparation. Dead sperm cells in- side the spermatophores appeared pink, whereas live cells were unstained against a red background of nigro- sin. For each observation, about 300-400 spermato- phores were counted in a given square area. The validity of eosin-nigrosin staining method was tested using the sperm cells, intentionally killed by exposing them to room temperature for 30 min. Biochemical analysis Protein was estimated by the method of Lowry et al. (1951). The protein in 0.2 ml of diluted seminal plasma was precipitated using 10% TCA. Using bovine serum albumin as a standard, the absorbancy was recorded at 500 nm. Total free sugars (TFS) were estimated by the method of Roe (1955). Protein-bound sugars (PBS) and glycogen were estimated following the method of Caroll et al. ( 1956). After hydrolyzing the sample to release the bound sugars in 1 N H:SO4 at 95°C, the same was treated with anthrone reagent. Glycogen was precipitated by eth- anol from the supernatant obtained during protein pre- cipitation and was dissolved in 0.5 ml distilled water and CRYOPRESERVATION OF CRAB SPERMATOPHORES Table I Cryopreservalian experiment on seminal plasma of ~Scylla serrata with different cryoprolectants: evaluation ofspermatophore viability (percentage) (mean ± S.E.) 249 Glycerol DMSO Trehalose DMSO-trehalose Temperature P.F 5th day 30th day P.F 5th day 30th day P.F 5th day 30th day P.F 5th day 30th day -196°C X 98.27 97.46 95.29" 96.96 94.80 88.99a 98.23 95.69 88.67a 98.39 97.08 94.25" S.E. ±1.217 ±1.377 ±1.366 ±0.494 ±0.425 ±0.852 ±0.503 ±0.563 ±0.676 ±0.892 ±0.881 ±0.012 -79°C X 96.38 95.57 93.18" 96.98 93.40 88.76a 97.75 94.98 87.16a 97.56 96.14 92.98" S.E. ±0.858 ±0.646 ±1.005 ±0.645 ±1.022 ±0.942 ±0.694 ±0.592 ±0.489 ±1.155 ±1.200 ±1.146 -4°C X 98.27 93.27 80.37" 96.96 88.19 66.75a 98.23 86.17 66.44a 98.39 92.00 78.62a S.E. ±1.217 ±0.636 ±1.081 ±0.494 ±0.965 ±1.176 ±0.503 ±1.301 ±1.602 ±0.892 ±0.811 ±0.905 P.F. = Prefreeze. No. of observation (n) = 6. a: P < 0.001. n: not significant. then treated with anthrone reagent. The color developed for PBS and glycogen was recorded as for TFS. Total lipids were estimated using sulphophospho- van- illin method (Barnes and Blackstock, 1973), after ex- tracting the lipid in chloroform:methanol mixture by the method of Folch et al. (1957). The above results were expressed in mg/ml seminal plasma. LDH enzyme activity was measured according to Yo- shida and Freese (1975), which is based on the lactate formed from pyruvate, by the oxidation of NADH. The units of enzyme per ml reaction mixture contained in a silica cell with 1 cm light path were calculated from the rate of absorbancy change at 340 nm, by using the milli- molar extinction coefficient of 6.22, and expressed in units/mg protein. One unit of enzymatic activity repre- sents the conversion of 1 jumole of NADH per minute. The enzyme protein was also assayed following Lowry etal.(l95l). All results were tested for significance using analysis of variance (ANOVA) and least significant difference (LSD)(SnedecorandCochran, 1967; Winer, 1971; Zar, 1974). Arcsin transformation for proportions were used. Results liability ofcryopreserved spermalophores Results of sperm (spermatophore) viability at - 1 96°C, -79°C, and -4°C are presented in Table I. The viability test was made with every 5-day interval, but the results are given only for the 5th and 30th days for brevity. The change in sperm viability was also not markedly different between the 10th and 30th day. The viability of sperm in 30 days of storage was fairly high, but it showed a range from 95% (at -196°C) to 80% (at -4°C). This high per- centage of viability was obtained in samples containing glycerol as the cryoprotectant. However, in DMSO, the viability was reduced to 89-67%. A similar trend was ob- served when trehalose was used as the cryoprotectant. Conversely, when DMSO + trehalose combination was used, a higher percentage of viability (between 94% and 79%) was obtained. Of the three temperatures tested, sperm viability was poor at -4°C, whereas at -196°C, sperm viability was maximum. Biochemical analysis The results are shown in Table II. Statistically, there was a significant difference between the three tempera- tures (P < 0.001 ); the LSD revealed that at - 196°C, the decrease of the biochemical components was less than at -79°C and — 4°C. For protein, total free sugars, glycogen, and LDH, there was no significant difference between -196°C and -79°C, whereas lipid and protein-bound sugars showed a significant difference between these two temperatures. However, at -4°C, the decrease of PBS is highly significant (P < 0.00 1 ); whereas glycogen and lipid do not show any significant decrease except in DMSO for glycogen (P < 0.005) and in trehalose for lipid (P < 0.00 1 ) at this temperature. Protein, glycogen, lipid, and LDH did not show any significant difference between glycerol and DMSO + tre- halose media; whereas PBS significantly decreased in DMSO + trehalose medium when compared to glycerol. Interestingly, the TFS showed a significant increase in glycerol and DMSO media at -4°C. At this temperature, TFS increased from a control (pre-freeze) value of 6.563 mg to 8.346 mg/ml in glycerol, and from 4.268 mg to 5.563 mg/ml in DMSO on the 30th day. That this in- crease may be due to a release of free sugars from the bound sugars was shown from a corresponding decrease of their value from 10.257 mg of the control to 6.707 mg/ ml in glycerol and 7.646 mg to 3.215 mg/ml in DMSO on the 30th day (Table II). However, in media containing trehalose, there was very little difference in value of TFS between the control and the preserved seminal plasma. 250 C. JEYALECTUMIE AND T. SUBRAMONIAM Table II Fluctuation in biochemical components of cryopreserved seminal plasma o/'Scylla serrata at three different temperatures and with different cryoproleclanlx (mean ± S.E.I Glycerol DMSO Trehalose DMSO-trehalose Components P.F 5th day 30th day P.F 5th day 30th day P.F 5th day 30th day P.F 5th day 30th day Protein (mg/ml) -196°C X 312.348 306.375 300.913" 214.863 204.152 180.390" 292.985 276.670 248.985" 298.713 292.437 281.920" S.E. ±12.877 ±11.761 ±13.934 ±11.754 ±11.745 ±13.383 + 11.496 ±17.370 ±8.982 ±14.137 ±13.318 + 12.468 -79°C X 227.928 221.695 218.715" 294.230 265.078 244.890" 372.875 343.360 315.848" 395.783 386.075 379.650" S.E. ±12.295 ±12.649 ±12.771 ±1 1.167 ±12.647 ±11.694 ±10.033 ±12.696 ±10.396 ±10.534 ±11.209 +9.185 -4°C X 312.348 277.997 224.350d 214.863 191.912 115.188a 292.985 261.523 160.512a 298.713 274.333 194.753a S.E. ±12.877 ±20.041 ±10.961 ±11.754 ±7.162 ±5.919 ±11.496 ±14.748 ±8.243 ±14.137 ± 1 1 .046 +8.346 LDH (Units/mg protein) -196°C X 91.118 86.047 85.990" 68.372 64.095 60.215" 95.193 90.752 80.705" 99.552 96.568 93.228" S.E. ±2.176 ±2.517 ±3.068 ±2.339 ±2.853 + 3.611 +3.989 ±4.202 ±3.638 ±4.731 ±4.231 ±4.543 -79°C X 89.547 86.715 84.687" 78.670 73.755 67.662" 94.992 89.483 78.870" 91.553 88.512 85.957" S.E. ±1.789 ±1.802 ±1.915 ±3.788 ±3.753 ±3.302 ±3.549 ±3.290 ±3.597 +2.000 ±2.163 ±1.697 -4°C X 91.118 79.555 57.227a 68.372 52.543 32.980a 95.193 76.357 41.267" 99.552 84.340 60.583a S.E. ±2.176 ±4.507 ±5.677 ±2.339 ±3.789 ±3.677 ±3.989 +2.689 + 1.674 ±4.731 ±2.967 +5.056 Total free sugars (mg/ml) -196°C X 6.563 6.337 5.966" 4.268 4.069 3.588" 446.672 439.036 423.371" 409.324 404.885 397.177" S.E. ±0.284 0.276 ±0.289 ±0.238 ±0.241 +0.243 + 14.069 ±14.284 ±12.259 ±6.869 ±7.899 +6.954 -79'C X 5.110 4.919 4.628" 3.801 3.633 3.229" 442.540 430.845 411.262" 403.074 387.748 385.254" S.E. ±0.226 ±0.211 ±0.222 ±0.156 ±0.181 ±0.169 ±13.451 ±15.676 ±13.393 ±11.485 + 12.478 + 12.769 -4°C X 6.563 7.096 8.346a 4.268 4.542 5.563' 446.672 448.649 450.935" 409.324 410.327 412.085" S.E. ±0.284 ±0.224 ±0.115 ±0.238 ±0.189 ±0.154 ±14.069 ±16.434 ±20.782 ±6.869 +6.410 +7.397 Protein bound sugars (mg/ml) -196°C X 10.257 9.992 9.569" 7.646 7.350 6.726" 15.294 14.688 13.828" 15.012 14.433 13.953" S.E. ±0.328 ±0.323 ±0.331 ±0.434 ±0.382 +0.443 +0.507 +0.440 ±0.425 ±0.609 ±0.583 ±0.587 -79°C X 10.193 9.868 9.434" 6.913 6.666 6.043" 14.669 13.801 12.896" 12.075 11.606 10.654" S.E. ±0.424 ±0.421 ±0.394 ±0.301 ±0.312 ±0.338 ±0.366 ±0.492 ±0.429 ±0.316 ±0.307 ±0.454 -4°C X 10.257 9.062 6.707a 7.646 6.303 3.2 15a 15.294 12.141 7.4403 15.012 12.588 8.0 13a S.E. ±0.328 ±0.304 ±0.223 ±0.434 ±0.414 ±0.423 +0.507 +0.485 +0.474 ±0.609 ±0.562 ±0.490 Glycogen (mg/ml) -196°C X 0.478 0.465 0.436" 0.579 0.533 0.496" 1.310 1.248 1.135" 2.177 2.120 1.933" S.E. ±0.021 ±0.024 ±0.020 ±0.019 ±0.018 ±0.020 ±0.222 ±0.233 +0.238 +0.219 ±0.223 ±0.235 -79'C X 0.458 0.435 0.411" 0.649 0.612 0.548" 2.265 2.148 1.894" 2.175 2.048 1.853" S.E. ±0.021 ±0.017 +0.012 +0029 +0032 +0028 +0 ">00 +0225 +0232 +0 164 ±0.162 ±0.160 -4°C X 0.478 0.452 0.381" 0.579 0.515 0.415" 1.310 1.205 0.945" 2.177 1.945 1.611" S.E. ±0.021 ±0.027 ±0.021 ±0.019 ±0.023 ±0.022 ±0.222 ±0.232 ±0.189 +0.219 +0.220 +0.242 Lipid (mg/ml) -196°C X 11.270 11.062 10.642" 8.161 7.859 7.082" 9.811 9.425 8.570" 10.070 9.883 9.303" S.E. ±0.472 ±0.477 ±0.461 ±0.514 ±0.468 ±0.459 +0.407 ±0.362 ±0.344 +0.509 +0.478 ±0.524 -79°C X 11.152 10.836 10.248" 8.494 7.849 6.985" 8.388 7.801 6.886" 11.925 11.614 10.833" S.E. ±0.538 ±0.529 ±0.513 ±0.484 ±0.558 ±0.517 ±0.590 ±0.568 ±0.613 ±0.458 ±0.434 ±0.414 -4°C X 11.270 10.676 9.502" 8.161 7.385 5.946" 9.811 9.046 6.742J 10.070 9.419 8.040" S.E. ±0.472 ±0.453 ±0.467 ±0.514 +0.611 ±0.526 ±0.407 ±0.407 ±0.481 ±0.509 ±0.532 +0.513 P.F. = Prefreeze; No. of observation ( n) = 6; a: P < 0.00 1 ; b: P < 0.005; d: P < 0.05; n: not significant. CRYOPRESERVATION OF CRAB SPERMATOPHORES 251 Table HI Conditions oj cryoprese rraiion of male gametes of Arthropoda SI. no. Species Cryoproteclant Preservation Percentage of Method of testing Temperature period survival viability References Decapod Crustacea 1 . Sicyonia ingentis 2. S. ingentis 3. Macrobrachium rosenbergii 4. Scvlla serrata Insecta 5. Apis mellifera Arachnida 6. Limulus polyphemus DMSO + Trehalose Trehalose, Sucrose. Proline. Glycerol, DMSO Glycerol Glycerol. DMSO, Trehalose, DMSO + Trehalose Glycerol, Seminal vesicle fluid, Spermathecal fluid Glvcerol -196T 2 months 60-70% Acrosome reaction Anchordogy el al. 1987 -196°C 1 month 56% Acrosome reaction Anchordogy et al. 1988 -196X" -I96T -79"C -4°C -79°C -74°C 3 1 days 30 days 53% Fertility 95% (Glycerol, Eosin dye exclusion DMSO + Trehalose) 89% (DMSO. Trehalose) Chowrtfl/.. 1985 Present study 16 days 50 days 50% Motility Eosin dve exclusion Sawada and Chang, 1964 Behlmerand Brown, 1984 There was also no interaction between temperature and media for components other than the TFS and PBS. Discussion The present results indicate that there is a definite in- fluence of freezing at different subzero temperatures on sperm viability and metabolism. In Scylla serrata, the viability of sperm varies between 95%-67% from -196°C to — 4°C. This is in accordance with Clegg and Pickett (1966), who suggested that there is no significant decrease in fertility of bull semen stored in liquid nitro- gen, in contrast to the deterioration that occurs in semen stored in dry ice. Cryopreservation of sperm or sperma- tophore has also been carried out in selected arthropod groups (Table III). According to Behlmer and Brown (1984), 64% of the post-thawed spermatozoa of Limulus at -74°C showed dye exclusion against 88% of control spermatozoa. Hughes (1973) also reported a notable de- crease in viability during cryopreservation of oyster sperm. Our experiments with freshly collected spermato- phores as well as spermatophores killed intentionally, further support the utility of the dye exclusion method in determining sperm viability. Intentionally killed sperm gave a 100% dye accumulation. It is of interest to note in this context that Bishop and Walton (1968) showed an increased permeability of cell membranes at the time of cell death. Damage caused by physical change in the membrane may be responsible for the reduced sperm viability as in- dicated by the work on RBC (Lovelock, 1954). During freezing and thawing, cell damage is due to the destruc- tive action of the concentrated salt solution to which the cells are exposed when water is removed as ice. Several cryoprotectants prevent the damage to cell membrane caused during cryopreservation. Polge et al. (1949) first showed that the spermatozoa could be frozen and thawed without losing motility if glycerol was included in their suspending medium. In glycerol, the electrolyte concentration at temperatures below the freezing point is sufficiently reduced (Lovelock and Polge, 1954). In this study, glycerol and DMSO + trehalose gave high sperm survival. At — 4°C, DMSO and trehalose are not efficient cryoprotectants, whereas, when DMSO is combined with trehalose. sperm viability increased sig- nificantly. Successful preservation of spermatophores of shrimp in glycerol at -196°C has also been reported (Chow et a I., 1985). Similar results were obtained by Ste- phens (1986) in chicken spermatozoa where the percent- age of cryoinjuries to the spermatozoa was less in glycerol and DMSO + glycerol media; when DMSO was used in the place of glycerol, for Limulus spermatozoa, post- thawing survival of spermatozoa was nil (Behlmer and Brown, 1984). Cryopreserved spermatozoa of Crassostrea virginica in DMSO fertilized 11% of eggs exposed to them 252 C. JEYALECTUMIE AND T. SUBRAMONIAM (Hughes, 1973), whereas combining DMSO with glycine and NaHCO increased the fertility to 96% (Zell et a/., 1979). Asahina and Takahashi (1978) showed that DMSO exhibited protection similar to that of ethylene glycol against freezing injury, whereas at room tempera- ture DMSO was toxic. In S. serrata, trehalose, when used alone, is not a good cryoprotectant; however, when com- bined with DMSO, the viability of sperm is significantly increased. In the ridge back shrimp Sicyonia ingentis, Anchordogy et at. (1987, 1988) also found a low sperm viability when trehalose was used as the cryoprotectant. However, when the DMSO was combined with treha- lose, these authors found increased viability. Interest- ingly, in this shrimp, the free spermatozoa stored very well with DMSO either alone or in combination with other cryoprotectants such as trehalose, sucrose, proline, and glycerol (Anchordogy et al., 1988). Obviously, the chief concern regarding spermatozoa that have undergone cryopreservation is whether they are of a quality, in terms of viability and vitality, compa- rable to that of fresh spermatozoa. Although fertilization is the ultimate criterion by which the quality of post-pre- served spermatozoa can be assessed, it is also presently an impractical one. It is difficult to obtain ova, as the exact period of oviposition of these crabs is unknown, which, therefore, precludes routine fertility tests. Thus the quality of cryopreserved spermatozoa was assessed in lieu of fertilization, as has been determined for Salnio salar spermatozoa by Yoo et al. ( 1 987), by estimating the biochemical fluctuations in the preserved spermatozoa. Biochemical alterations, if any, in the spermatophores of crustaceans that have been cryopreserved have not been previously reported. However, the protein compo- nent in the ovine seminal plasma undergoes qualitative changes during cold storage (Garner and Ehlers, 1971). Yoo et al. (1987) reported the loss of protein from the cryopreserved spermatozoa into the outer seminal plasma in salmon, due to leakage through the cell mem- brane into the outer medium. In bovine semen, Pickett and Komarek (1964) also showed leakage of lipids into the seminal plasma from the cryopreserved spermato- zoa. It is not known whether any such change occurs in the protein or lipid contents relative to the sperm cells and seminal plasma of 5. serrata during cryopreserva- tion. In the present study, there were practical difficulties in completely isolating the spermatophores from the seminal plasma after cryopreservation; even after re- peated centrifugation of seminal plasma diluted with the diluents, the granular substances of the seminal plasma adhered to the spermatophores, thus making it difficult to isolate the spermatophores. It should also be noted that arthropod spermatozoa are sensitive to centrifuga- tion (Behlmer and Brown, 1984) and hence repeated centrifugation may cause increased leakage of organic substrates and enzymes from the sperm cells to the me- dium, as has also been reported by Barnes and Black- stock ( 1 974) for cirripede semen. Biochemical results of the present study reveal an in- teresting pattern of substrate use during cryopreserva- tion. At — 4°C, the total carbohydrates showed a signifi- cant decline, although protein also showed a decline. In S. serrata. carbohydrates formed the main substrate used during sperm maintenance within the spermatheca of the mated female (Jeyalectumie, 1 989). The reduction in the carbohydrate level, when preserved at -4°C, further indicates that the metabolic activity of sperm may con- tinue using carbohydrate substrates. The activity of the LDH also declined at -4°C. This may be due to the death of sperm cells, which accounts for 1 8%. According to Ja- cobs et al. (1986), the dissociation and loss of LDH activ- ity occur at low temperatures; by 45 days of storage, hu- man serum retained 74%, 53%, and 87% of initial activ- ity when stored at 25°C, 4°C, and -20°C, respectively. Our preliminary observations on glycosidases activity in the seminal plasma and spermatophores of this crab us- ing different substrates such as /}-D-glucopyranoside, 0- D-galactopyranoside, /3-D-mannopyranoside, and a-N- Acetyl galactosaminide indicate high enzyme activity (Yu-Teh Li, T. Subramoniam, and C. Jeyalectumie, un- pub. obs.). All these may indicate that sperm storage in a metabolically active condition require active substrate use by spermatozoa contained in the spermatophores. In conclusion, our results suggest that glycerol pro- vided the best protection for the seminal plasma and maintained a high percentage of sperm viability. DMSO is toxic at room temperature; the decreased value of sper- matophore viability at -4°C when DMSO was used as the cryoprotectant, may also support this view. Acknowledgments This work was supported by the Department of Sci- ence and Technology, Government of India, New Delhi (Grant No. 22(8p-8)/STP-II). Literature Cited Anchordogy, T., J. H. Crowe, W. H. Clark Jr., and F. J. Griffin. 1987. Cryopreservation of sperm from the penaeid shrimp Sicyo- nia ingentis. Abstract from 1 8th Annual meeting of the World Aquaculture Society, Guayaquil, Ecuador, Jan. 1987. Anchordogy, T., J. H. Crowe, F. J. Griffin, and W. H. Clark Jr. 1988. Cryopreservation of sperm from the marine shrimp Sicyo- nia ingentis. Cryobiology 25: 238-243. Asahina, E., and T. Takahashi. 1978. Freezing tolerance in embryos and spermatozoa of the sea urchin. Cryobiology 15: 122-127. Barnes, H. 1962. The composition of the seminal plasma of Balanus balanus. J Exp. Biol. 39: 345-35 1 . Barnes, H., and J. Blackslock. 1973. Estimation of lipids in marine animals and tissue: detailed investigation of the sulphophospho- CRYOPRESERVATION OF CRAB SPERMATOPHORES 253 vanillin method for 'total' lipids. / Exp. Mar. Biol. Bail. 12: 103- 118. Barnes, H., and J. Blackstock. 1974. Biochemical composition ot'the seminal plasma of the cirripede Balanus halanus(L) with particular respect to free amino acids and proteins. J. Exp. Afar. Biol. Ecol. 16:47-85. Behlmer, S. D., and G. Brown. 1984. Viability of cryopreserved sper- matozoa of the horseshoe crab Limulus polyphemus. Int. J Invert. Reprod. Dev 7: 193-199. Bishop, M. W. H., and A. Walton. 1968. Spermatogenesis and the structure of mammalian spermatozoa. Pp. 1-101 in Marshall's Physiology of Reproduction, Vol. I. Part II, A. S. Parkes, ed. Long- mans, London. Caroll, W. V., R. W. Longley, and J. H. Roe. 1956. The determina- tion of glycogen in the liver and muscle by the use of anthrone re- agent. J. Biol. Chem. 220: 583-593. Chow, S., Y. Taki, and Y. Ogasawara. 1985. Cryopreservation of spermatophore of the fresh water shrimp, Macrobrachium rosen- bergii. Biol. Bull 168: 47 1-475. Clegg, E. D., and B. W. Pickett. 1966. Effect of storage at - 196°C on fertility. A. I. Digest Vol. XIV, No. 3. Folch, J., M. Lees, and G. H. Bloane-Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 266: 497-509. Garner, D. L., and M. H. Ehlers. 1971. Effect of storage at 5°C on the Disc electrophoretic patterns of ovine and bovine seminal proteins. ./ Reprod. Pert. 27: 43-52. Morton, H. F., and A. G. Ott. 1976. Cryopreservation offish sperma- tozoa and ova. J. Fish. Res. Board Can. 33: 995-1000. Hughes, J. B. 1973. An examination of eggs challenged with cryopre- served spermatozoa of the American oyster, Crassostrea virginica. Cryobiohgy 10: 342-344. Ishida, T., P. Talbot, and M. Kooda-Cisco. 1986. Technique for the long term storage of lobster (Hotnarus) spermatophores. Gamete Res 14: 183-195. Jacobs, E., P. J. Hissin, W. Propper, L. Mayer, and L. Sarkozi. 1986. Stability of lactate dehydrogenase at different storage tem- peratures. Clin. Biochem. 19: 183-188. Jeyalectumie, C., and T. Subramoniam. 1987. Biochemical composi- tion of seminal secretions with special reference to LDH activity in the reproductive tissues of the field crab, Paratelphusa hydro- dromons (Herbst). Exp. Biol 46: 23 1 -236. Jeyalectumie, C. 1989. Biochemical investigations on the reproduc- tive tissues and Cryopreservation of seminal secretions of a brachy- uran crab Scylla serrata (Forskal) (Decapoda: Portunidae). PhD Thesis, University of Madras, India. Kurokura, H., H. Kumai, and M. Nakamura. 1986. Hybridization be- tween female red sea bream (Pagriis major) and male crimson sea bream (Evynnisjaponica) by means of sperm Cryopreservation. Pp. 1 13-1 16 in The First Asian Fisheries Forum, J. L. Macleon. L. B. Dizon, and L. V. Hosillos, eds. Asian Fisheries Society, Manila, Philippines. Leverage, W. E., D. A. Valeric, A. P. Schultz, E. Kingsbury, and C. Dorey. 1972. Comparative study on the freeze preservation of spermatozoa of primate, bovine and human. Lab. Anim. Sci. 22: 882-889. Lovelock, J. E. 1954. Physical instability and thermal shock in red cells. Nature 173: 659-66 1 . Lovelock, J. E., and C. Polge. 1954. The immobilization of sperma- tozoa by freezing and thawing and the protective action of glycerol. Biochem. 7.58:618-622. Lowry, D. H., N. J. Rosebrough, A. L. Fair, and R. J. Randall. 1951. Protein measurement with Fohn phenol reagent. J. Biol. Chem. 193:265-275. Pickett, B. W., and R. J. Komarek. 1964. Evidence for loss of lipid from bovine spermatozoa due to freezing. J. Dairr Sci. 47(8): 905- 908. Polge, C., A. U. Smith, and A. S. Parkes. 1949. Revival of spermato- zoa after vitrification and dehydration at low temperatures. Nature 164(4172): 666. Roe, J. H. 1955. The determination of sugar in blood and spinal fluid with anthrone reagent. J Biol. Chem. 212: 335-343. Sawada, Y., and M. C. Chang. 1964. Tolerance of honey bee sperm to deep freezing. / Earn. Entonwl. 57(6): 891-892. Snedecor, G. \V., and VV. G. Cochran. 1 967. Statistical Methods. Ox- ford and IBH Publishing Co. 593 pp. Stephens, A. J. A. 1986. A comparison of diluents for cryopreserva- tion of rooster semen. PhD Thesis, Tamil Nadu Agricultural Uni- versity, Coimbatore, India. Winer. 1 97 1 . Statistical Principles in Experimental Design. McGraw Hill, Inc. Withler, F. C. 1982. Cryopreservation of spermatozoa of some fresh- water fishes cultured in South and Southeast Asia. Aquaculture 26: 395-398. Yoo, B. Y., M. A. Ryan, and A. J. Wiggs. 1987. Loss of protein from spermatozoa of Atlantic salmon (Salmo salar L.) because of Cryo- preservation. Can. J Zoo/. 65: 9-13. Yoshida, A., and E. Freese. 1975. Lactate dehydrogenase from Bacil- lus subtilis. Pp. 304 in Methods in Enzymology. Vol. XLI, W. A. Wood. ed. Academic Press. New York. Zaneveld, L. J. D., and K. L. Polakoski. 1977. Collection and physi- cal examination of the ejaculate. Pp. 147-172 in Techniques of Hu- man Andrology, E. S. E. Hafez, ed. Elsevier/North-Holland Bio- medical Press. Zar, J. H. 1974. Biostatistical Analysis. Prentice-Hall. Inc. Engle- wood Cliffs, N.J. 620 pp. Zell, S. R., M. H. Bamford, and H. Hidu. 1979. Cryopreservation of spermatozoa of the American oyster, Crassostrea virginica Gmelin. Cryobiohgy 16: 448-460. Reference: Bwl. Bull. Ill: 254-276. (October. 1989) On the Early Development of the Vestimentiferan Tube Worm Ridgeia sp. and Observations on the Nervous System and Trophosome of Ridgeia sp. and Riftia pachyptila MEREDITH L. JONES' AND STEPHEN L. GARDINER2 1 Department of Invertebrate Zoology. National Museum of Natural History, Smithsonian Institution, Washington. D.C. 20560 and2 Department of Biology, Bryn Mawr College, Brvn Mawr, Pennsylvania 19010 Abstract. Stages in the development of the vestimentif- eran Ridgeia sp., based on lengths of preserved specimens processed for scanning electron microscopy, were exam- ined. Features of the nervous system and trophosome of Riftia pachyptila were studied by light, scanning-, and transmission-electron microscopy. Development proceeds from a trochophore-type larva with an anterior prototro- chal ciliary ring and a posterior assemblage of transient lar- val setae, through intermediate stages, some of which lack endosymbiotic bacteria but all of which display additional transient features such as larval branchial filaments, a ven- tral medial process, and digestive tract, to a young juvenile stage that possesses endosymbiotic bacteria and exhibits the morphology characteristic of adult vestimentiferans. Lar- val branchial filaments are resorbed in later developmental stages and replaced by paired rows of ciliated branchial fil- aments with pinnules. The larval gut is divisible into fore- gut, midgut, and hindgut regions based on cytological fea- tures of the epithelium. The establishment of the symbiotic association in the midgut region is confirmed. The develop- ment of the gut and the establishment of the endosymbiotic association appear to be correlated with the timing of settle- ment by the young juveniles. Aspects of the development of the nervous system include the appearance of the brain near the base of the ventral medial process followed by the development of a nerve cord in the epithelium of the body wall throughojt the length of the juvenile. The nerve cord includes one or two giant axons, except in its most posterior region. The trochophore larva likely serves as a dispersal stage in the life history of vestimentiferans. The trocho- phore larva in the early development of vestimentiferans strengthens the assertion that Vestimentifera and Annelida are closely related. Introduction The first vestimentiferan species, Lamellibrachia bar- hami, was described by Webb (1969) from a cold-water site in the eastern Pacific. Among other features, the ab- sence of a gut led Webb to place this species in the phy- lum Pogonophora. In their description of a second vesti- mentiferan species, L. luymesi, van derLand and N0rrev- ang (1975) introduced the term "trophosome" for a peculiar tissue that occupies the trunk region. In a later report, van der Land and N0rrevang( 1977) examined in greater detail the organization of the "parenchymatous" trophosomal tissue and stated that it consists of numer- ous small lobules. In each lobule, a peripheral layer of pigment cells surrounds a central region of basophilic cells filled with numerous vacuoles. Van der Land and N0rrevang suggested that the trophosome functions as a liver and is also a source of nutrients for developing spermatozoa. In the same year, Webb (1977) described and illustrated a "spongy tissue" associated with the male reproductive system of L. barhami, but he did not suggest a function for this tissue nor did he refer to it as trophosome. In his description of a third vestimentiferan species, Riftia pachyptila, Jones ( 198 la) noted the pres- ence of trophosomal tissue in the trunk region. In a sub- sequent report, Jones ( 1 98 1 b) described the trophosome as consisting of many lobules, each with a central blood vessel that gives rise to numerous capillaries that ramify throughout the tissue. In their description of a third spe- cies of Lamellibrachia, L. victori, Mane-Garzon and Montero ( 1 986) commented on the lobular nature of the trophosome and its vascularization. Further, they stated, apparently on the basis of light microscopy, that there 254 EARLY DEVELOPMENT OF R1DGEIA 255 are ". . . numerous accumulations of symbiotic micro- organisms which are not bacteria but spores or algae" (p. 18). These observations have not been confirmed by other investigations. Based on transmission electron microscopic (TEM) observations and analysis of lipopolysaccharide, Cava- naugh (1980) suggested that the bulk of the trophosome of R. pachyptila is occupied by bacteria that potentially serve as chemoautotrophic symbionts. In the first study of trophosomal tissue to include TEM micrographs, Cavanaugh et al. ( 1 98 1 ) confirmed the presence of bacte- ria in the trophosome of/?, pachyptila. Subsequent stud- ies have confirmed the presence of endosymbiotic bacte- ria in R. pachyptila (Bosch and Grasse, 1984a, b; Hand, 1987), Escarpia spicata (Felbeck el al., 1981; Cava- naugh, 1983a, b; reported as an unnamed new species), Lamellibraehia barhami (Felbeck et al.. 1981), Ridgeia piscesae, Ridgeia phaeophiale, and two additional unde- scribed species of vestimentiferans (de Burgh, 1 986), Oa- sisia alvinaeand Ridgeia sp. (the latter two species from personal TEM observations by SLG). These and other studies have also provided insight into the nature of the relationship between the bacteria and their hosts (for re- cent references, see Felbeck and Childress, 1988). One question that arises with respect to this association is how it is established in new individuals. Jones (1985b, 1987, 1988b) and Jones and Gardiner (1988) noted the presence of a so-called ventral medial process (=siphon; Southward, 1988b) at the base of the branchial plume of juvenile R. pachyptila. Ridgeia sp., and Oasisia alvinae. The ciliated aperture at the distal end of the process leads to a duct that passes through the brain, through the vestimentum, and communicates with the established trophosome. They suggested that the aperture and duct are the means of entry of bacteria into the trophosome. Jones (1987) suggested that free-living bacteria are collected at random by the ciliated aperture of the process and transported to the trophosome via the duct, and if such bacteria include sulfide-oxidizing bacte- ria and others necessary for the worm, the worm sur- vives; if they do not, the worm ultimately dies. Jones and Gardiner (1988) stated that the ciliated aperture at the distal end of the ventral medial process is the mouth. In addition, they described the gut and anus of a complete digestive tract in Ridgeia, as well as the fine structure of the foregut ofRiftia and of cell junctions of its trophoso- mal bacteriocytes. Southward ( 1 988a. b) also reported on the digestive tract in early developmental stages of Rid- geia, suggested an early phase of ciliary feeding, and dis- cussed the relationship of the Vestimentifera and Pogo- nophora and the relationship of these to the Annelida. Preliminary results of additional studies of early develop- mental stages of Ridgeia were presented by Jones ( 1988a) and form the basis of expanded results presented in this report, which includes the first description of a trochophore larva in the development of vestimentiferans. Additional new information is provided on the development of the ventral process and digestive system, the nervous system, larval and opisthosomal setae, branchial filaments, ventral ciliated field, and vestimentum. Observations on certain as- pects of the development of the trophosome are presented. When appropriate, results from Southward (1988b) are compared with our findings. Materials and Methods Larval and juvenile specimens of Ridgeia sp. were col- lected at "Axial Seamount," Juan de Fuca Ridge (Alvin Dive 1413, 45°56'N; 130°01'W, 18 July 1984, 1546 m depth, and.-1/vm Dive 1924, 45°55'N; 130°02'W, 30 Sep- tember 1987, 1540 m depth). Specimens of Riftia pa- chyptila were collected on the Galapagos Rift at the "Rose Garden" hydrothermal vent site (Alvin Dive 889, 00°48.7'N; 86°12.7nW, 14 February 1979, 2458 m depth) and a juvenile Riftia was collected at the "Garden of Eden" vent site (Alvin Dive 993, 00047N; 86°08'W, 10 December 1979, 2518 m depth). Additional specimens of Riftia pachyptila were obtained at the "Clam Acres" site on the EPR at 21° N (Alvin Dive 1225, 20°50'N; 109 WW, 9 May 1982, 2618 m depth). For all speci- mens, except those from "Clam Acres" and the 1 987 col- lections from "Axial Seamount," after initial fixation in 10% formalin (buffered with CaCOi, pH 6.95) in seawa- ter, the specimens were transferred to 70% ethanol. For light microscopy, specimens were dehydrated and embedded in a mixture of epon-araldite, using propylene oxide as the infiltration solvent; semi-thin sections ( 1.5 or 2.0 ^m) were cut on a Sorvall MT-2 ultramicrotome and were stained with Masson's triple stain. For scanning electron microscopy (SEM), specimens were placed in Ruthenium Red (as a mordant for OsO4) for one hour, post-fixed in 1% OsO4, on ice, for one hour, dehydrated through a graded ethanol series, and then critical-point or freeze dried. Specimens were sputter-coated with gold- palladium (about 1 5-nm thick) and examined in either a Cambridge 100 Stereoscan or a Hitachi S-570 scanning electron microscope. The collections from "Clam Acres" were intended for examination by transmission electron microscopy; samples of adult trophosome were fixed aboard ship at room temperature in 3.5% glutaraldehyde in 0.1 A/ phosphate buffer (pH 7.3) containing 10% su- crose and a trace of CaCN . Tissues were refrigerated and stored in glutaraldehyde fixative until their use for this study. The recent collections from "Axial Seamount" (1987) yielded Ridgeia larvae/juveniles for light micros- copy and TEM examination from a clump of tubes ofR. piscesae, bulk-fixed onboard ship in approximately 4% glutaraldehyde in 0. 1 M cacodylate buffer (pH 7.2). Tis- 256 M. L. JONES ET AL sues for TEM were ("Clam Acres") post-fixed in phos- phate-buffered OsO4 for 1 h at 4°C and dehydrated in a standard ethanol series, or ("Axial Seamount") post- fixed in cacodylate-buffered OsO4 for 2 h at room tem- perature, pre-embedded in agar, and dehydrated in a graded series of ethanol, with 2,2-dimethyoxypropane after 70% ethanol. Trophosome was embedded in a mix- ture of epon-araldite and larvae/juveniles in agar blocks were embedded in Spurr's resin; propylene oxide was the infiltration solvent in both procedures. Thin-sections were cut on a Sorvall MT-2 ultramicrotome, stained with aqueous uranyl acetate and lead citrate, and examined in a JEOL 1 DOS. JEM 1 200EX. or a Zeiss EM9S-2 trans- mission electron microscope. Measurements of length of the 38 specimens of Rid- geia examined by SEM are not "total length" (Table I). Due to the variability of contraction of branchial fila- ments of preserved specimens, as well as the lack of branchial filaments in early stages, lengths were mea- sured from the prototrochal band of larvae and the base of branchial filaments of juveniles to the posterior ex- tremity. This length, of specimens processed for SEM, is the basis for the ranking of specimens and does not necessarily reflect a true estimate of actual or relative age of the specimens. Where appropriate, rank numbers (#X) are included in the text and figure legends to iden- tify individual specimens listed in Table I. Those illus- trated specimens not ranked are noted as "nr. #X" to indicate their length relative to ranked specimens. The lengths of sectioned specimens were adjusted to take into account shrinkage due to processing for SEM. A compar- ison of measurements before and after critical-point dry- ing indicated a shortening of body length by about 7.5% during processing; this factor was applied to lengths of sectioned material to make them comparable to SEM specimens. This shrinkage was variable (the 7.5%, above, is an average of shrinkage of specimens ranging from 4% to 10%), and may explain apparent inconsistencies in some of the results below. Where a comparison with the observations of Southward (1988b) was appropriate, measurements, from the base of branchial filaments to the posterior end, were based on her SEM micrograph and camera lucida drawings, using her indicated scales and correcting for shrinkage, and were converted to "nr. #'s" as follows: Fig. 3A,— 0.727 mm— nr. #27 Fig. 4, left— 0. 1 19 mm— nr. #6 Fig. 4, right— 0.108 mm— nr. #5 Fig. 5 A, spec. RJ— 0.142 mm— nr. #9 Fig. 5C, spec. RB— 1.304 mm— nr. #32 Fig. 5D, spec. RC— 1 .6 1 3 mm— nr. #33 Two species of Ridgeia are known from Axial Sea- mount and, although the 38 larvae and juveniles exam- ined by SEM were sorted from clumps of Ridgeia pisce- sae tubes, it is not possible to determine their specific identity. On the basis of overall body shape, specimens with rank numbers 6,7, 17, and 1 9 may be different from the rest of the series; rank numbers 15 and 22 have an unusually thick ventral process; these characteristics may be artifacts of fixation or processing. The Vestimentiferan Body Plan and the Systematic Relationships of the Vestimentifera The Vestimentifera presently contains ten species whose anatomy and morphology have been described to varying degrees (see Jones, 1985c; Mane-Garzon and Montero, 1986). Although differences exist pertaining to certain aspects of the anatomy and morphology of these ten species, they, nevertheless, share a common body plan as adults. The following description, based largely on observations from Jones ( 1 985c), is intended to famil- iarize the reader with this vestimentiferan body plan. It is our desire that subsequent sections of this report, which describe certain aspects of the development of Ridgeia sp. and the anatomy of Riftia pachyptila, can be exam- ined with the characteristic adult body plan of the Vesti- mentifera in mind. The body of juvenile and adult vestimentiferans is di- visible externally into four distinct regions. From ante- rior to posterior ends, these regions are ( 1 ) the obturacu- lar region, (2) the vestimentum, (3) the trunk, and (4) the opisthosome (Fig. 1 ). The obturacular region (Fig. 1, OB (comprises the cen- tral obturaculum (Fig. 1, OBT), which supports and bears the respiratory plume, and the plume itself; the lat- ter consists of many branchial filaments ( Fig. 1 , BF ) that are provided with pinnules (lobes that increase the respi- ratory surface) and are fused as left and right series of branchial lamellae. In the case of R. pachyptila, the branchial lamellae extend perpendicularly from the ob- turaculum and are free for most of their length (class Ax- onobranchia). In all other vestimentiferans described to date (class Basibranchia). the branchial lamellae extend anteriorly from the base of the obturaculum and are fused for most of their length. One or two openings of the excretory organ are situated on the dorsal surface near the base of the obturaculum. The vestimentum (Fig. 1, VS) is provided ventrally with a conspicuous pear- or teardrop-shaped ciliated field (Fig. 1, VC), which is bounded by the two parts of the separated ventral nerve cord. Plaques (Fig. 9, PL) and the openings of the so-called pyriform glands are also evi- dent on the ventral and lateral surfaces; the latter secrete tube material at and near the open end of the tube. Dor- sally, the vestimentum bears a pair of genital apertures with paired ciliated grooves extending anteriorly in the EARLY DEVELOPMENT OF R1DGEIA 257 Table I Rankings (RK) of specimens o/ Ridgeia sp. observed by scanning electron microscopy (SEM). arranged by length (L - distance from prototroch or base of branchial filaments to posterior end), and notations of presence of larval setae (LS), number of branchial filaments (BF). condition of ventral ciliated field (I 'CF), number of rows ofopislhosomal setae (OS), stale qfvestimentum (V 'ST), obturaculum (OBT) and anus (AN) and presence or absence ofendosymbiotic bacteria (BA) RK L(mm) LS BF VCF OS VST OBT AN BA 1 0.058 P 0 L 0 L L 7 7 2 0.075 P 0 L 0 L L 7 7 3 0.087 P 2 S ? L L 9 ? 4 0.100 P 3 S 1 L L 7 7 *G 0.104(0.112) P 2 S 0 L L L L 5 0.110 9 2 S ? L L 7 9 6 0.111 P 2 L 0 L L 7 ? 7 0.140 P 2 S 1 L L ?P 7 8 0.141 P 3 S 1 L L 9 ? 9 0.143 P 2 S 1 L L P 7 *F 0.148(0.160) P 2 S 1 L L L L 10 0.150 P 3 S 1 L L P 9 1 1 0.155 P 3 S 1 L L 7 7 12 0.160 P 5 S 1 L L 7 7 13 0.162 P 2 S 1 L L P 7 *D 0.163(0.176) 9 2 S 1 L L L L 14 0.169 P 2 S 1 L L 7 9 *H 0.176(0.190) P 4 S 1 L L P P *B 0.178(0.192) 9 4 S 1 L L P L 15 0.180 P 5 S 1 L L 7 7 16 0.180 P 4 S 2 D L 7 7 *E 0.182(0.197) P 2(?+) 7 1 L L L L 17 0.184 P 2 S 1 L L L 9 18 0.190 P 4 S 1 L L 7 7 19 0.214 P 2 7 1 L L 7 7 20 0.220 7 5 S 2 D L 7 7 *CB7 0.247(0.267) 9 5 ?s 1 L L 7 P 0.257 P ?4 S 1 L L L 9 *C 0.272(0.294) 7 4 S 1 L L P L 21 0.280 P 2 S 2 D L 7 9 *A 0.285(0.308) 7 6 ?s 1 L L P P 22 0.300 P 5 S 2 L L 7 7 23 0.400 P 4 7 2 P L 9 7 *CB1 0.400(0.432) 9 8 7 3 P L L P 24 0.440 P M F 1 P L 9 9 25 0.540 P 8 F 2 P L 7 7 26 0.690 L M F 4 P 7 7 7 27 0.750 P M F 4 P 9 9 9 28 .010 P M F 7 P P 9 7 29 .060 L M F 7 P 9 7 7 30 .120 ? M F 8 P 9 ?L 7 31 .210 P M F 5 P P 9 9 32 .340 P M F 9 P 9 7 7 33 .560 P M F 8 P P 7 9 34 .740 P M F 7 P P L 7 35 2.000 L M F 4 P 9 7 9 36 2.190 L M F 7 P P 7 7 37 2.230 L M F 11 P P L 9 38 4.040 L M F 22 P P L 7 Lengths of sectioned specimens, marked with "*" and an identifying letter, are adjusted to allow for lack of shrinkage due to SEM processing and actual lengths follow, in parentheses. D: developing; F: fused; L: lacking: M: many, not countable accurately; P: present; S: segmented: ?: not known, unobservable. 258 M. L. JONES ET AL. Figure 1. Ru/witi sp. Rank #36. SEM. Ventral view, juvenile. Arrow, posterior margin of vestimen- tum. Scale bar, 500 ^m. Figure 2. Rank #1, SEM. Trochophore larva. Scale bar. 15 Mm Figure 3. Rank #3, SEM. Right dorsolateral view of larva. Scale bar, 30 ^m. Figure 4. Rank #3, SEM. Anterior view of larva of Fig. 3. Scale bar, 10 ^m. Figure 5. Rank #3, SEM. Ventral view of larva of Fig. 3. Scale bar, 30 ^m. Figure 6. Rank #5, SEM. Right lateral view, ventral process of larva. Scale bar, 20 ^m. Figure 7. Rank #14, SEM. Anterior view, anterior region of larva, developing ventral process; tip of branchial filament damaged. Scale bar, 2 EARLY DEVELOPMENT OF R1DGEIA 259 case of males. Internally, the bulk of the vestimentum consists of muscles and connective tissue and these act to keep the vestimentum at the open end of the tube and maintain the position of the branchial plume of the obtu- racular region outside the tube; the brain is situated ven- trally near the anterior margin and the excretory organ is posterior to the brain. The trunk region (Fig. 1 , TK) occupies the greatest rel- ative portion of the body in all vestimentiferans so far described (up to 80% of total body length in the largest specimen of R. pachyptila; Jones, 198 la). The surface of the trunk bears the trace of the united nerve cord in the ventral midline (Fig. 1, VN) and the papillae of pyriform gland openings; secretions from the latter thicken the tube wall. Internally, the trunk contains vascular ele- ments, the trophosome, which houses endosymbiotic bacteria, and the gonad. A digestive tract has not been observed in this region in adult specimens. The opisthosome (Fig. 1 , OP) is the only region of the body that is multisegmented in the adult. Externally, the continued trace of the nerve cord is visible in the ventral midline. Anterior segments are provided with transverse rows of opisthosomal setae ( Fig. 1 , OS) that act as a hold- fast to the inner surface of the tube when the worm with- draws; a variable number of posterior segments lack se- tae but are indicated externally by furrows that mark the positions of internal septa. Internally, the coelomic space of each segment is paired due to the presence of a median mesentery that extends throughout the length of the opis- thosome. The systematic relationship of the Vestimentifera with other higher Bilateria is still unsettled. This lack of agree- ment among investigators centers in large part on the im- portance of the segmentation pattern (and the arrange- ment of coelomic spaces) and the manner of segment for- mation in the Vestimentifera relative to other segmented groups. Webb (1969) cited the long trunk and opistho- some (=metasoma) of Lame/librachia barhami as dis- tinctive pogonophoran features and placed that species in a new class and order in the phylum Pogonophora. Van der Land and Narrevang (1975, 1977) did not at- tribute special phylogenetic importance to the regiona- tion of the body of Lamcllibrachia and considered the Vestimentifera, as well as the Pogonophora, as separate classes in the phylum Annelida. In his original descrip- tion of Riftia pachyptila, Jones (198 la) considered the body regionation as indicating a close relationship with the pogonophorans. He retained the Pogonophora at the level of phylum and placed the Vestimentifera in it as a new subphylum, the Obturata. Jones ( 1 985a) contrasted the arrangement of the apparent segments of the vesti- mentiferans and pogonophorans and the development of segments in the opisthosome. He suggested that these two groups may not be as closely related as previously thought and that their close relationship with the Annel- ida required reexamination. Based on previous observa- tions and additional new information, Jones (1985b, c) separated the Vestimentifera and Pogonophora at the level of phyla. Citing, in particular, an apparent lack of difference in the development of segments in the opistho- somes of the vestimentiferans and pogonophorans. Southward ( 1 988b) suggested that the two groups should be considered as subclasses in the class Pogonophora. In addition, she suggested that the outcome of future devel- opmental studies would be instrumental in determining if the class Pogonophora should be placed in the phylum Annelida or phylum Brachiata. Results Trochophore The trochophore larva of Ridgeia is provided with a prototrochal ring of cilia (Fig. 2, PRO) and lacks a neuro- troch, metatroch, and an apical tuft on the pretrochal area (Fig. 2, PRE); that part of the posterior region that is not obscured by mounting adhesive does not reveal a telotroch (Fig. 2). Examination of the prototroch at higher magnification does not reveal the presence of compound cilia; this may be due to a less than optimal fixation of the original sample or may reflect the true state of the prototrochal cilia. There are no apparent mouth or anal openings. That the specimen is a vesti- mentiferan trochophore is confirmed by the presence of larval setae typical of later larval and juvenile stages (see Discussion, below). I 'enl ral process and gut At about the time that the first pair of larval branchial filaments develop (Figs. 3, 4, LBF) (see Branchial fila- ments, below), a lip-like protrusion (Fig. 5, LL) arisesjust posterior to the prototroch (Figs. 3-5, PRO). This, in ad- Figure 8. Rank #15, SEM. Left lateral view, ventral processor larva. Scale bar, 30 ^m- Figure 9. Rank #25, SEM. Right lateral view, ventral process of juvenile. Scale bar, 30 ^m. BF, branchial filament(s); CR. ciliary row; DBF, developing branchial filament; DP. displaced pinnules; LBF, larval branchial filaments; LL, lower lip; MO, mouth; OB. obturacular region; OBT, obturaculum; OP, opisthosome; OS. opisthosomal setae; PL, vestimental plaque; PN, pinnule; PRE, pretrochal region; PRO, prototroch; TK. trunk; LIL, upper lip; VC, ventral ciliated field; VN, ventral nerve cord; VP. ventral pro- cess; VS, vestimentum; X, damage during processing. 260 M. L. JONES ET AL •^' ' 'i^M^^m EARLY DEVELOPMENT OF R1DGE1A 261 dition to the first appearance of the ciliated field (Fig. 5, VC), establishes the ventral surface of the developing larva. As in the case of the trochophore, close examina- tion of the prototroch failed to reveal the presence of compound cilia. In time, the posttrochal protrusion de- velops cilia and becomes the lower (= posterior) lip of the mouth opening (Figs. 6, 7, LL). The mouth is formed between the lower lip and the ventral portion of the pro- totroch; the latter becomes the upper (=anterior) lip of the mouth (Figs. 6, 7, UL). There is a differential growth such that the prototroch is displaced ventrally and the ventral process, with its terminal mouth, is formed (Figs. 6, 7, MO). The residuum of the prototroch is distributed laterally along the length of the process (Fig. 8, PRO) and persists for some time as the process elongates (Fig. 9, rank #25. PRO). The entire pretrochal region of the trochophore appears to be restricted to the upper ( =ante- rior) surface of the ventral process (Figs. 6-9, PRE). Examination of sectioned specimens by light micros- copy and whole specimens by SEM (see Table I) reveals details of several aspects of the early development of the gut ofRidgeia. First, the mouth is formed early in devel- opment (G-nr. #4) but is not connected to the foregut ( F- nr. #10, D-nr. #13, E-nr. #16) until later in development (H-nr. #15. CB7-nr. #20, with endosymbionts; C-nr. #21, lacking endosymbionts). Second, the anal opening is established after the mouth is open, but the stage at which the anus appears, and a complete digestive system can be established, is variable. Third, the appearance of the anus seems to precede the establishment of the bacte- rial association in young juvenile stages (note specimens B-nr. # 1 5 and C-nr. #2 1 in Table I). Fourth, the bacterial association is established in the midgut region of Rid- geia, but the time at which this association occurs is vari- able (in particular, note specimens H- and B-nr. #15, C- and A-nr. #2 1 ). Finally, the closure of an anus in later juvenile stages, e.g., #34, #37, and #38. is correlated with the probability that the symbiotic association with bacte- ria has been established (see Discussion, below). Throughout its length, the gut is lined by an epithe- lium of multiciliated cells whose cilia nearly obscure the lumen (Figs. 10, 13, 16). The presence of accessory cen- trioles and a system of rootlets associated with basal bod- ies of the cilia have not been confirmed. When viewed by TEM, a foregut, midgut, and hindgut are distinguish- able in Ridgeia, based on cytological features of the epi- thelial cells. The foregut epithelium is characterized by the pres- ence of numerous electron-dense secretory granules up to 500 nm in diameter in the apical region of the cells (Figs. 10, 11). Mitochondria are scattered in the cyto- plasm beneath the area occupied by the granules. Bas- ally, the cells contain extensive profiles of rough endo- plasmic reticulum (Fig. 12, RER) and numerous Golgi complexes that are actively releasing vesicles. The fore- gut epithelium rests on a blood sinus (Figs. 10, 12, BS), and a layer of peritoneal cells is situated between this si- nus and the body wall (Fig. 10, P). Prior to the establishment of the symbiotic associa- tion, the cytology of the epithelial cells of the midgut is rather unremarkable. Apically, the cells contain numer- ous mitochondria, scattered Golgi complexes, and a few electron-dense granules (Fig. 13). Nuclei of the cells are situated basally (Fig. 14) and RER is not extensively de- veloped. A thin layer of extracellular matrix separates the epithelial cells from a layer of peritoneal cells. A blood sinus was not observed in our specimens. When viewed by light microscopy, the hindgut ap- pears transparent when compared with other regions of the gut (Figs. 21, 45, 49, 50). This observation is ac- counted for in TEM preparations in that the hindgut epi- thelial cells contain large vacuoles that are filled with a slightly granular, electron-translucent substance, which may represent mucus or unused yolk material (Fig. 15, Figure 10. Riiigeia sp. Unranked juvenile, TEM. Cross-section of foregut epithelium. Note secretory granules in region of cells adjacent to gut lumen. Scale bar. 4 ^m. Figure 11. Unranked juvenile. TEM. Enlargement of apical region of foregut epithelium, showing numerous secretory granules. Scale bar, 1 .5 fim. Figure 12. Unranked juvenile. TEM. Enlargement of basal region of foregut epithelium, showing rough endoplasmic reticulum. Scale bar. 1 .5 //m. Figure 13. Unranked juvenile. TEM. Apical region of midgut epithelium prior to establishment of bacterial association. Scale bar. 2 ^m. Figure 14. Unranked juvenile. TEM. Cross-section of midgut epithelial cell prior to establishment of bacterial association. Scale bar. 2 ^m. Figure 15. Unranked juvenile. TEM. Hindgut epithelium. Scale bar. 3 ^m. Figure 16. Unranked juvenile, TEM. Tangential section of epithelium surrounding anal opening. Scale bar. 1.5 ^m. Figure 17. Rank A-nr. #21, frontal section, epon, 2.0 nm. Midgut of juvenile. Scale bar, 30 /jm. Figure 18. Unranked juvenile. TEM. Cross-section of midgut epithelium after establishment of bacte- rial association. Scale bar, 2 ^m. BA, bacteria; BS, blood sinus; CO, coelom of trunk; GL, lumen of gut; M. mitochondria; MU, mucus; P, peritoneum; RER. rough endoplasmic reticulum. 262 M. L. JONES ET AL. 21 - (KQ EARLY DEVELOPMENT OF R1DGE1A 263 MU). Cytoplasm is restricted to a thin layer around the periphery of the cells, adjacent to the membranes. Scat- tered mitochondria are present in the apical region of the cells, but nuclei and other cellular organelles have not been observed. Bacteria and other materials are present in the lumen of the hindgut. An associated layer of peri- toneal cells appears to be absent. In the region of the anus, the cytology of the epithelial cells resembles that of the midgut (Fig. 16). Bacterial association and trophosome The bacterial association is established in the midgut region of Ridgeia. During early stages of the association, the lumen of the gut is open, and the bacteria appear to be scattered throughout the epithelial cells (Figs. 1 7, BA, GL). However, when viewed by TEM, the bacteria (Fig. 1 8, BA) are seen to occupy the basal area of the epithelial cells. Bacteria are coccoid in shape, up to 2.5 ^m in diam- eter, and appear to be housed in separate vacuoles. Nu- clei of the epithelial cells and mitochondria are situated in the cytoplasm surrounding the bacteria (Figs. 18, M; 20, NU). The presence of different sizes of bacteria or of the digestion of bacteria by midgut cells, as noted by Southward (1988b), were not confirmed in our speci- mens. At least one instance of a bacterium undergoing binary fission has been observed (pers. obs., SLG). Mito- chondria, RER, and, occasionally, a Golgi complex, are observed in the apical region of the cells (Fig. 18). As development proceeds, the lumen of the midgut disappears, and the trophosome develops through an elaboration of the original epithelial lining of the midgut. In adults of Riftia pachyptila, and presumably all vesti- mentiferans, the trophosome consists of numerous elon- gated lobules that appear circular or somewhat elliptical when viewed in cross-section (Fig. 19). Each lobule is provided with a central axial blood vessel from which extend numerous capillaries 1 .0-3.6 /um in diameter that connect with blood vessels on the outer surface of the lobule (Fig. 19, arrows; for additional details on the vas- cular system of the trophosome, see Jones, 1988b). The cytology of the specialized cells that house the bacteria in the trophosome (bacteriocytes) differs from that of the original midgut epithelium mainly in the absence of most cellular organelles. Only the nucleus and a few scat- tered mitochondria have been observed in the cytoplasm surrounding the bacteria (Fig. 20). Analysis of TEM mi- crographs indicates that bacteria occupy at least 40% of the area of a bacteriocyte, in section, and that bacterio- cytes account for at least 41% to 53% of the total area of the trophosome, in section. Nervous system In the development of the ventral process, accumula- tions of presumed nervous tissue are present at the base of the process (Figs. 21-23, PB), just below its dorsal (=anterior) surface. These arise between the outer layer of epithelial cells and the wall of the foregut. There are suggestions of continuity of this nervous tissue, lateral to the foregut and to just internal to the ventral surface; this growth around the foregut has yet to be confirmed at these stages. The presumptive brain, within the ventral process, has been observed in stages D-nr. #13, H-nr. # 1 5 and C-nr. #21 (Figs. 21-23, PB), but not in presumed younger stages (G-nr. #4 and F-nr. #10). Later, the brain comes to be situated in the vestimentum, internal to the ventral process, posterior to the branchial filaments, with the foregut traversing it (Fig. 24, BR, FG). In larval and young juvenile stages (up to C-nr. #2 1 ), a differentiated ventral nerve cord has not been observed in the epithelium of the body wall. In later juvenile stages and adults, a single nerve cord exits the brain on its ven- tral surface, just internal to the cuticle and epithelium of the body wall (see Jones, 198 la, 1985a, for additional descriptions of the nervous system of adult vestimentif- erans). In addition to other nervous tissue, the nerve cord, here, contains a pair of giant axons. In juveniles of Ridgeia sp. and Oasisia alvinae, and, presumably, all vestimentiferans, the perikarya of these giant axons are situated adjacent to each other in the dorsal region of the brain. In contrast to other cells in the brain, the cyto- Figure 19. Riftia pachyptila USNM No. 59958, adult, transverse section, paraffin, 5 A"TI. Cross-section of lobule of trophosome. Arrows, surface blood vessels. Scale bar, 50 ^m. Figure 20. Riftia pachyptila, adult, TEM. Bacteriocyte in trophosome. Scale bar, 3 nm. Figure 21. Ridgeia sp. Rank D-nr. #13. sagittal section, epon. 1.5 nm. Anterior of larva. Scale bar. 30 Mm. Figure 22. Rank H-nr. #15, sagittal section, epon, 1.5 ^m. Anterior of larva. Arrows, ventral ciliated field. Scale bar, 30 /im. Figure 23. Rank C-nr. #21. sagittal section, epon, 1.5 ^m. Anterior of larva. Scale bar, 30 ^m. Figure 24. RankCB7-nr. #20, frontal section, epon, 2.0 ^m. Anterior of larva. Scale bar, 30 nm. Figure 25. Rank CBI-nr. #23. transverse section, epon, 2.0 nm. Ventral ciliated field and adjacent paired ventral nerve cords. Large arrows, nerve cords: small arrows, extent of ventral ciliated field. Scale bar, 30 ^m. BA, bacteria; BF, branchial filament; BR. brain; BV, axial blood vessel; CA, capillary. FG, foregut; HG, hindgut; LBF, larval branchial filament; MG, midgut; NU, nucleus; PB. presumed brain; PG, pyriform glands; VP. ventral process; VV. ventral vessel. 264 M. L. JONES ET AL Figure 26. Ridgeia sp. Unranked juvenile, TEM. Cross-section of vestimentum in region of ventral ciliated field, showing portion of giant axon. Scale bar, 5 nm. Figure 27. Unranked juvenile, TEM. Cross-section of nerve cord in vestimentum, showing giant axon, neurites and glial cell bodies of myelin sheath. Scale bar, 5 ^m. Figure 28. Unranked juvenile, TEM. Enlargement of myelin sheath of giant axon. Note varying thick- ness of lamellae of sheath. Scale bar, 1 . EARLY DEVELOPMENT OF RIDGEIA 265 plasm of the perikarya of the giant axons stains lightly in TEM preparations. Numerous mitochondria and dense- cored vesicles are visible in these cells. A single giant axon exits each perikaryon and extends ventrally through the neuropile of the brain. Microtubules, mito- chondria, and dense-cored vesicles are present in the gi- ant axons in this region. Upon reaching the ventral ciliated field, the nerve cord diverges and carries one giant axon in each branch (Figs. 25, large arrows; 26, 27, GA). The cytoplasm of the giant axons in this region stains lightly in TEM preparations and contains scattered vesicles and mitochondria (Figs. 27, 29). Microtubules, however, have not been observed here. In the region of the vestimentum, each giant axon is surrounded by a myelin sheath of irregularly spaced lamellae (Figs. 26, 27, MS) derived from glial cells whose cell bodies are mostly clustered in the ventrolateral re- gion of each nerve cord (Fig. 27, GC). Occasionally, a glial cell body is observed in the outermost lamellae of the sheaths. In one juvenile, the thickness of the sheaths varied from 0.5 ^m to 2.5 j*m, and each sheath consisted of about 12 lamellae, although it was difficult to deter- mine this number precisely because the lamellae fre- quently branch and fold back on themselves (Fig. 28). By contrast, the number of lamellae surrounding the giant axon in the adult stage of Riftia pachyptila exceeds 50 (pers. obs., SLG). The layer of cytoplasm between mem- branes of the lamellae varies in thickness from as little as 50 nm to as much as 2 ^m. Mitochondria, endoplasmic reticulum, and numerous vesicles are present in the cyto- plasm of the thicker lamellae. The nerve cords fuse at the posterior margin of the ventral ciliated field, and a single nerve cord extends through the trunk and opisthosome. In the trunk, the nerve cord contains one giant axon whose organization is similar to that of the giant axons in the vestimentum. The nerve cord in the opisthosome lacks a giant axon (Jones, 198 la, 1985a). Further development of larvae and juveniles If the 38 specimens examined by SEM are arranged from trochophore (Fig. 2) to established juvenile (Fig. 1 ), lengths from 58 ^m to 4.04 mm, it is possible to deter- mine the progressive development of a number of mor- phological characters (Table I; Figs. 29-52). Larval setae. In the single trochophore observed, three larval hooks and one capillary seta are visible in one mi- crograph (Fig. 29. small arrows, CS), and one other hook and two other capillary setae can be seen in micrographs taken at right angles to the first; about one-half of the circumference of the trochophore, at the level of the lar- val setae, is obscured by mounting adhesive. We suggest that this trochophore may bear up to eight larval hooks and six to eight capillary setae. In later stages, where ob- served, there are two pairs of larval hooks, situated later- ally, with a single, capillary seta between each pair (Fig. 30, small arrows, CS; Jones and Gardiner, 1 988, Fig. 7E). Each hook bears a single cluster of three to five denticles, palmately arranged around a central one, all pointing an- teriorly (Figs. 29-31, LH/small arrows). In longitudinal sections, it has been confirmed that the larval setae are situated in the trunk wall and are well-separated from the opisthosome and the opisthosomal setae (Fig. 50, LH). Larval setae persist for a longer or shorter time, appar- ently depending on the amount of abrasion in life or the amount of handling that the specimens receive during sorting and processing; capillary setae are lost even more readily. Figure 53 shows the presence or absence of larval setae among the ranked specimens and indicates that they are present on an individual as long as 1.74 mm. Branchial filaments. The first (larval) branchial fila- ments develop as a pair on the dorsal surface of the larva, arising just posterior to the prototroch (Figs. 3, 4, 33, 34, LBF), and a second pair may develop just posterior to the first. These pairs of larval branchial filaments do not develop pinnules but bear two longitudinal rows of cilia, one dorsomedial and the other ventrolateral; both pairs are considered to be larval structures because they appear to be resorbed as development proceeds. In the youngest stage that was sectioned (G-nr. #4) there is a suggestion that the coelomic cavity of the ventral process (Fig. 32, CVP), at some time in development, communicates with the cavity of the larval branchial filaments (Fig. 32, CBF); the coelomic cavity of the ventral process was not Figure 29. Rank # 1 , SEM. Larval setae on posterior of trochophore larva. Large arrow, anterior: small arrows, larval hooks. Scale bar, 5 ^m. Figure 30. Rank nr. #2 1 , SEM. Setae on posterior of larva. Large arrow, anterior; small arrows, larval hooks. Scale bar, 10 ^m. Figure 31. Rank #15, SEM. Setae on posterior of larva. Arrow, anterior. Scale bar, 10 Mm. Figure 32. Rank G-nr. #4, sagittal section, epon. 1.5 /arn. Base of branchial filament and ventral pro- cess. Small arrow, mouth/foregut; large arrow, ventral. Scale bar. 10 ^m. Figure 33. Rank #10, SEM. Dorsal view of anterior part of larva. Scale bar, 10 ^m. Figure 34. Rank #10, SEM. Left lateral view of larva. Scale bar, 50 Mm. BF, branchial filament; CBF. coelom of branchial filament; CR, ciliary row(s); CS. capillary seta; CVP, coelom of ventral process; GA, giant axon; GC, glial cell bodies; LBF, larval branchial filament; LH. larval hook; MO, mouth; MS, myelin sheath; NE. neurites; OS, opisthosomal setae; PN, pinnule(s); VC, ventral ciliated field; VP, ventral process. 266 M L. JONES ET AL. EARLY DEVELOPMENT OF RIDGEIA 267 observed in sections of other, older, larvae. Subsequent juvenile branchial filaments develop after formation of the ventral process and mouth and, initially, are pro- vided with two rows of pinnules on their dorsal surface and ciliary tracts on their dorsomedial and ventrolateral surfaces (Figs. 33, 34, PN, CR). The second, as well as subsequent, juvenile branchial filaments initially de- velop with the same disposition of pinnules and ciliary tracts as the first filaments (Figs. 8,51, PN, CR). At about the time of appearance of the second pair of juvenile branchial filaments, the first pair of branchial filaments appears to undergo a reorientation of pinnules, which be- come more lateral (Fig. 8, PN), and of ventrolateral cili- ary tracts, which become more ventral (Fig. 43, CR). The reorientation continues until, at a point where four pairs of branchial filaments are present in the first branchial lamella, pinnules are situated on the outer faces of fila- ments (Fig. 35, large arrows) and ciliary tracts are situ- ated adjacent to the rows of pinnules, ventral or lateral, and on the surface of the filaments, directly opposite each other (Fig. 35, small arrows). This disposition of rows of cilia is continued throughout the life of the worm (Figs. 36, 37, small arrows). In adults, two rows of pin- nules are on the outer face of the filament, one row is central and the other is displaced to the outer dorsal area (Figs. 36, 37, CP, DP). It may also be noted that each filament carries two bundles of nerves just below the ven- tral-facing surface, that the pair of branchial blood ves- sels is in line with the axis of the lamella, that the cuticle, generally rather thick, is quite thin over the ciliated cells and over the surface of the pinnules (Figs. 36, 39, NB, BV, CU), and that right branchial filaments are mirror images of left filaments (Figs. 35-37). Branchial filaments are added in paired rows (Figs. 35, 38) that are the forerunners of the branchial lamellae of older specimens. The oldest filaments of a lamellar series are ventral and medial, adjacent and lateral to the ventral process or dorsal to it (Figs. 9, DBF; 35, Rl, LI; 38, Ra, La), and younger filaments are added laterally and dor- sally (Figs. 35, R4, L4; 38, Rb, Ld). Subsequent lamellae develop in the same manner, outside the previous lamel- lae (Figs. 38, 46). Relative to later-developing branchial filaments, larval filaments appear to lack well-developed longitudinal muscles, as well as pinnules (Figs. 39, 40). In earlier stages (#'s 3-9), larval branchial filaments are the primary respiratory surface for developing Ridgeia and are so served by simple vascular loops (Fig. 41). Later, respiratory surfaces are enhanced by the develop- ment of elongated filaments bearing pinnules (Fig. 40). Figure 54 discloses that, after they appear, there is a mod- est increase in the number of branchial filaments, to about eight, until a length of about 0.5 mm, after which the number of filaments is great enough to preclude their being counted. Ventral ciliated field. There is no neurotroch on the trochophore. Up to nine tufts of cilia, linearly arranged and, perhaps, representing a late-developing neurotroch (from #3 on), are present on the mid-ventral surface of Figure 35. Ridgeia sp. CBl-nr. #23, transverse section, epon. 2.0 pm. Cross-section, four pairs of branchial filaments. LI (oldest) — L4 (youngest), left filaments; Rl (oldest) — R4( youngest), right filaments; large arrows, general location of pinnules; small arrows, location of ciliary rows. Scale bar, 50 ^m. Figure 36. Ridgeia piscesae, USNM No. 98106, adult, transverse section, paraffin, 5 pm. Cross-section, left branchial filament. Large/thick arrow, direction to longitudinal axis of worm; long/thin arrows, direction of effective stroke of ciliary rows, Riftiapachyptila; short arrows, location of ciliary rows. Scale bar, 30 jim. Figure 37. Ridgeia piscesae, USNM No. 98106, adult, transverse section, paraffin, 5 ^m. Cross-sec- tion, right branchial filament. Large/thick arrow, direction to longitudinal axis of worm; long/thin arrows, direction of effective stroke of ciliary rows, Riftia pachypt ila ; short arrows, location of ciliary rows. Scale bar, 30 jim. Figure 38. Ridgeia sp. CB 1-nr. #23, transverse section, epon, 2.0 jim. Cross-section, base of branchial plume, two pairs of branchial lamellae (proximal to Fig. 35). La (oldest)— Ld (youngest) filaments of second left lamella; Ra (older) — Rb (younger) filaments of second right lamella. Scale bar, 50 ^m. Figure 39. Rank CB7-nr. #20, frontal section, epon, 2.0 nm. Base of larval branchial filaments. Scale bar, 30 ^m. Figure 40. Rank CB7-nr. #20, frontal section, epon, 2.0 ^m. Base of branchial filaments. Scale bar, 30 Mm. Figure 41. Rank F-nr. #10, longitudinal section, epon, 1.5 ^m. Larval branchial filament, showing vascular loop. Scale bar, 10 ^m. Figure 42. Rank #9, SEM. Ventral view showing posterior position of segmented ventral ciliated field. Scale bar, 30 jim. Figure 43. Rank #20, SEM. Ventral view, showing segmented ventral ciliated field (arrows); ventral process damaged. Scale bar, 30 ^m. AN, anus; AVM, anterior vestimental margin; BF, branchial fila- ment(s); BV, blood vessel(s); CP, central pinnule; CR, ciliary row; CU, cuticle; CVP, cilia of ventral process; D, dorsal; DP, dorsal pinnule; L, left; LBF, larval branchial filaments; LH, larval hook; LL, lower lip; NB, nerve bundles; OS, opisthosomal seta; PL, vestimental plaque; PN, pinnule; R, right; V, ventral; VC, ventral ciliated field; VP, ventral process. 268 M. L. JONES ET AL Figure 44. Ridgeia sp. Rank #20, SEM. Right lateral view, showing segmented ventral ciliated field (arrows); ventral process damaged. Scale bar, 30 j/m. Figure 45. Rank D-nr. #13, sagittal section, epon, 1.5 pm. Segmented ventral ciliated field (arrows). Scale bar, 30 nm. Figure 46. Rank #33, SEM. Ventral view, showing fused ventral ciliated field. Scale bar, 30 ^m. EARLY DEVELOPMENT OF R1DGE1A 269 the posterior one-half to one-third of the vestimentum- trunk of larvae and young juveniles (Fig. 42, VC); these tufts ultimately fuse to form the pear-shaped ventral cili- ated field of adults (Fig. 1. VC). Although the tufts pres- ent the appearance of being segmentally arranged (Figs. 43, 44, arrows), sagittal sections through the tufts and the body wall do not reveal any indication of internal segmentation (Fig. 45, arrows). In early stages, the tufts overlie the trunk cavity and are displaced anteriorly by differential growth of the posterior trunk region. By the time the vestimentum is formed, the ciliated tufts, later- ally expanded, are fused to form the ventral ciliated field (Fig. 46, VC). Figure 55 reveals that the linearly seg- mented ventral ciliated field persists in specimens of up to about 0.5 mm in length and the pear-shaped field of adults is present in juveniles of greater length. Opisthosomal setae. These are confined to the anterior opisthosomal segments. As in adults, they bear two groups of denticles, the larger group projecting anteriorly and the smaller, posteriorly (Figs. 30, 31, OS; see also Jones, 1985c, Figs. 15, 48, 52). The first opisthosomal setae of juveniles may bear a superficial resemblance to larval setae and have a single set of denticles (Fig. 47, IOS). Opisthosomal setae develop in transverse rows in the four quadrants of the body (Fig. 48, OS), and those of each side meet laterally; setal rows do not appear to meet dorsally or ventrally. Semi-thin sections of the pos- terior end of an early larva show that, prior to the appear- ance of the first row of opisthosomal setae, the opistho- some is undeveloped (Fig. 49, G-nr. #4). At the time of development of the first opisthosomal setae, internal seg- mentation has commenced (Fig. 50, F-nr. #10). Figure 56 shows that, until juveniles are about 1.0 mm long, there are usually up to four rows of opisthosomal setae, and it is only when a length of about 2.0 mm has been attained that the number of rows of setae begins to in- crease rapidly; there is a quite apparent variability in the development of setal rows (see Discussion, below). Vestimentum. Development of the vestimentum is visible externally by the delineation of a pair of dorsolat- eral longitudinal ridges just posterior to the branchial filaments (Fig. 51, DVS). Further development leads to an extension of the posterior ends of these ridges, later- ally and ventrally, to the posterior edge of the ventral cili- ated field (Fig. 46). The presence of plaques also indicates the establishment of the vestimentum (Figs. 9, 43, 44, 5 1 , PL). Much later (#36, >2.00 mm), a flange-like posterior vestimental margin, free of the trunk, is developed (Fig. 1, arrow). Figure 57 indicates that the vestimentum is established when juveniles are 0.44 mm in length and it may be developing from a juvenile-length of 0. 1 8 mm. Obtiiraciilum. The developing obturaculum has not been observed among these larval and juvenile Ridgeia. Most probably its development is similar to that ofRiftia (Fig. 52, OBT) where paired, medially fused, thin lobes, each about the diameter of a branchial filament, develop on the mid-line, dorsal to all branchial filaments. Figure 58 shows that the obturaculum of Ridgeia is not present up to a juvenile length of about 0.5 mm; its development may be initiated during the next 0.5 mm of growth. In the case of lengths greater than 1 .0 mm, the obturaculum is present to support the developing branchial filaments. Discussion Trochophore The discovery of a trochophore larval stage in the life history of a vestimentiferan provides an explanation of the geographical distribution of certain species of these worms. Vent fields may occur at 100-200 km intervals (Crane, 1985). Hydrothermal vents, to which most vesti- mentiferan species appear to be restricted, are discrete environments that, in the same general hydrothermal field, may be as close as 3 km to one another (on the Galapagos Rift: Rose Garden site to Mussel Bed site, about 10 km; Mussel Bed to Garden of Eden site, about 3 km) (J. F. Grassle, 1986) or as far apart as the Galapa- gos Rift and 13°N on the East Pacific Rise (EPR) (about Figure 47. Rank #17, SEM. Lateral view, trunk/opisthosomal region showing profile of initial opistho- somal seta. Arrow, anterior. Scale bar, 10 pm. Figure 48. Rank nr. #21, SEM. Posterior view of opisthosome, showing disposition of four sectors of opisthosomal setae. Arrow, ventral. Scale bar, 30 tim. Figure 49. Rank G-nr. #4, sagittal section, epon, 1.5 /im. Section through undeveloped opisthosome. Scale bar, lO^m. Figure 50. Rank F-nr. #10, sagittal section, epon, 1.5 ^m. Section through developing opisthosome. Scale bar, 10 ^m. Figure 51. Rank #16, SEM. Right lateral view, showing developing vestimentum. Scale bar, 50fim. Figure 52. Riftia pachyptila, SEM. Dorsal view, showing developing obturaculum. Scale bar, 100/im. BF, branchial filament(s); CR, ciliary row; DVS. developing vestimentum; HG, hindgut; IOS, initial opis- thosomal setae; LBF, larval branchial filament; LH, larval hook(s); MG, midgut; OB, obturacular region; OBT, obturaculum; OP, opisthosome or opisthosomal region; OS, opisthosomal setae, first row, OS2, opisthosomal setae, second row; PL, vestimental plaque; PN, pinnule; TK., trunk; VC, ventral ciliated field; VP, ventral process; VS. vestimentum. 270 M. L. JONES ET AL 2500 km farther northwest) and 2 1 °N on the EPR (about 1 100 km yet farther northwest), with Rijiia pachyptila being common to all of these localities. The trochophore represents a life stage that can be transported by sea-bot- tom currents to maintain genetic continuity between widespread populations of a single species. This particular specimen of Ridgeia (Fig. 2) was washed from among adult tubes and, apparently having settled, had not yet developed to the point where it was permanently in place, i.e., in a tube affixed to a solid sub- stratum. The development of larval setae (Fig. 29) sug- gests that this trochophore had developed sufficiently so that it was nearly ready to secrete a tube that would allow the proper functioning of the larval hooks in maintaining the larva within its tube. If, as is thought, vestimentiferans fall into the Gastro- neuralia, as used by Nielsen (1987), compound cilia are to be expected in the prototroch of Ridgeia. That they are not present may well be due to the incidental cavalier treatment received during fixation on shipboard, when the first thought was to preserve the whole sample, pri- marily adults. Nielsen (1987, p. 206) specifically states that a special, gentle preservation is necessary to main- tain the integrity of compound cilia. Although Southward (1988b) had no specimens smaller than nr. #5 and #6, she compared her smallest with the patterns of ciliation noted by Nielsen ( 1 987) and suggested that the vestimentiferan larva belongs to the "trochophore type." We confirm her suggestion. We were able to examine only one specimen that was young enough to be identified as a trochophore larva. Although this specimen lacks openings into the digestive system, as well as an apical ciliary tuft, neurotroch, meta- troch, and telotroch, the complete ring of cilia situated near the apical end, which is the characteristic position and appearance of a prototroch, confirms the larva's identity as a trochophore. The presence of larval hooks suggests that it is a late trochophore and confirms that it is a larval vestimentiferan. The absence of digestive openings is possibly correlated with the young age of the specimen, whereas the absence of ciliary bands, other than the prototroch, could be related to age of the speci- men or could reflect developmental features specific to vestimentiferans. The presence of a larva of the trocho- phore type in the early development of vestimentiferans provides additional evidence for the belief that vesti men- tiferans should be placed along the evolutionary line that includes annelids, molluscs and other smaller groups of protostomes. Ventral process and gut On the basis of our observations, we suggest that the prototroch and pretrochal region of the trochophore are intimately involved in the formation of the ventral pro- cess. The gut appears to develop at about the same time, but the anal opening may occur early or late in develop- ment, as reflected by length (Table I). The pathway of the foregut through the brain to the trophosome has been established (Jones and Gardiner, 1988; Southward, 1988b; Jones, 1988b); the ventral pro- cess has been reported to be present in juveniles of Rijiia up to 15.5 mm in overall length but is not present in young adults of about 1 1 1 mm overall length (Jones and Gardiner, 1988). In the latter, and in still older speci- mens, the track of the foregut, now apparently closed and functionless, can be traced through the brain and vesti- mental musculature to the trophosome (pers. obs., MLJ). TEM examination of cross-sections of the cilia of the lower lip indicates that the effective stroke of these cilia is inward (pers. obs., SLG), thus justifying our conten- tion that the mouth, indeed, might be employed in pick- ing up free-living bacteria and/or other particulate food material. Jones and Gardiner (1988) noted that cells of the foregut are multiciliated. Southward (1988b) sug- gested that, prior to the installation of symbiotic bacteria, Ridgeia larvae are ciliary feeders; she noted, in a small specimen (nr. #9), that bacteria were in the wall of the gut, ". . . apparently undergoing digestion . . ." and that rod-shaped bacteria were present in the opisthoso- mal hindgut. These observations suggest at least three possibilities concerning the intracellular bacteria: ( 1 ) if all bacteria are endosymbionts, then the digestion of some by the worm indicates the early use of the endo- symbionts as a food source; (2) if all bacteria are not en- dosymbionts, then the digestive process has been initi- ated for some but not for others and would confirm an early phase of ciliary feeding by the larvae; and (3) if the undigested bacteria are endosymbionts and the digested bacteria are not endosymbionts, then a period of overlap occurs between the phase of ciliary feeding and the full establishment of the symbiotic association. There are indications that the mouth/foregut, even though morphologically developed, may not be open un- til H-nr. #15, based on the presence of endosymbionts. This is supported by the fact that the midgut wall, in younger, sectioned individuals, is devoid of bacterial symbionts (Table I, Figs. 21, 45; Jones and Gardiner, 1988, Fig. 7C). Likewise, the anus, although present at #7 and #9, does not appear to open to the exterior until C-nr. #2 1 , based on light microscopy. In a small individ- ual (G-nr. #4) we have observed a presumptive mouth and a gut, but the anus is totally lacking. Southward (1988b) stated that, based on sections, a mouth, complete gut, and anus were present in a two- filament-stage larva (RJ, nr. #9); no trophosome was, as yet, present (see Delayed settlement, below). E I h- o z UJ LACKING o PRESENT • NOT KNOWN ? EARLY DEVELOPMENT OF R1DGEIA 5 10 20 RANK 30 53 0 . 0° LLJ 40 MANY M 54 **** 4M 00232223233522542425 25 10 20 RANK 30 40 271 LACKING O SEGMENTED o FUSED * 4h NOT KNOWN ? *-* O E O 55 * - ?* 0 1 - NOT KNOWN ? 007, 2' 10 20 RANK 30 40 0 10 20 RANK 56 22. 30 40 Figure 53. Ridgeia sp. Presence of larval setae among SEM specimens, ranked by length. Figure 54. Numbers of branchial filaments among SEM specimens, ranked by length. Figure 55. State of ventral ciliated field among SEM specimens, ranked by length. Figure 56. Number of rows, opisthosomal setae, among SEM specimens, ranked by length. 272 M. L. JONES ET AL — i — — i — 57 LACKING o o ~~ ' ' 1 1 58 LACKING o DEVELOPING * PRESENT • 4 PRESENT • 4 NOT KNOWN ? ~ 3 E - !3 - I X cB z J 2 r— o z LU •* 1 .•** 1 •7? °( Qooooooooooooooaooo**0 °. o oo 00000000000(X100000000 10 20 30 4( RANK 10 20 30 4C RANK Figure 57. Ridgvia sp. Condition of vestimentum among SEM specimens, ranked by length. Figure 58. Presence of obturaculum among SEM specimens, ranked by length. Bacterial endosymbionts Although we have not, as yet, reconstructed serial thin-sections and counted total numbers of bacteria in bacteriocytes, we have observed by TEM as many as 35 bacteria in a single section of a bacteriocyte. In this case, the bacteria occupied approximately 40% of the total area. Jones ( 1988b) elucidated details of the vascular sys- tem of the trophosome and observed that, in one in- stance, no bacterium was separated from a capillary by more than three bacteria. A chemoautotrophic mode of nutrition has been suggested for the bacteria in the tro- phosome of vestimentiferans (for extensive literature, see Jones and Bright, 1985, and Felbeck and Childress, 1988). In previous accounts (Jones, 1988b; Jones and Gardi- ner, 1988), we suggested that the association between bacteria and the vestimentiferans Riftia pachyptila, Oa- sisia alvinae, and Ridgeia sp. begins with the ingestion of free-living bacteria, from surfaces available to the larvae/ juveniles or from the waters surrounding the vent com- munities, into the transient digestive system of the worm. Once inside the gut, the appropriate bacterial strain, i.e.. sulfide-oxidizing. is phagocytized by midgut epithelial cells to establish the symbiotic association. Pre- sumably other bacteria, non-essential to the vestimentif- eran, not prospective endosymbionts, may be digested and used as a food source until the endosymbiotic associ- ation is established and the gut regresses. Jannasch and Wirsen (1979) reported the presence of at least 200 different strains of free-living bacteria in vent waters and noted that most strains oxidized hydrogen sulhde or thiosulfate as their energy source. Oxidation of sulfide has been suggested as an energy source also for the endosymbiotic bacteria in R. pachyptila (see Arp et a/.. 1985. for references). Distel et a/. (1988) examined the 1 6S rRNA sequences of endosymbiotic bacteria from several invertebrate hosts, including R. pachyptila. Their data indicate that, in all cases, the host contains a single type of endosymbiont and hosts from different geo- graphic localities possess similar types of endosymbionts. Based on these data, they concluded that the mechanism of selection of potential endosymbionts from the sur- rounding environment or the mechanism of transmittal of endosymbionts to new individuals must be specific. They suggested three possible mechanisms: ( 1 ) the sym- biont is passed to the egg during spawning, (2) the symbi- ont is selected by the host during early development, and (3) the symbiont uses a host-specific "infection" (their quotation marks) mechanism. At present, passing of endosymbionts to eggs during EARLY DEVELOPMENT OF RIDGEIA 273 spawning appears unlikely. Our TEM observations of eggs of R. pachyptila taken from just inside the aperture of the oviduct have not revealed the presence of endo- symbionts in the egg cytoplasm or nuclei (pers. obs., SLG). TEM observations of freshly spawned eggs of R. pachyptila also have failed to reveal the presence of endo- symbionts (Gary et al., 1989). The method of establish- ment of the symbiotic association outlined by Jones ( 1 988b) and Jones and Gardiner ( 1 988) conforms to the second mechanism of Distel el al. (1988). However, this method also implies that midgut cells of larval/juvenile vestimentiferans are capable of distinguishing between potential endosymbiontsand other, non-essential, bacte- ria; this method of recognition has not yet been deter- mined. We suggest the following possible sequence of events in the establishment of the symbiotic association: ( 1 ) en- gulfment of bacteria by the ventral process and their pas- sage into the gut; (2) growth of one specific strain of bac- teria close to midgut epithelial cells; and (3) phagocytosis of colonies of bacteria by midgut cells. Nen'ous system In his original description of Lamellibrachia barhami, Webb (1969) described the presence of a pair of tubes in the nerve cords in the vestimental region of that species. He indicated that these tubes ". . . enter the substance of the brain as very fine tubules" and that the nerve cord of the trunk contains a single tube throughout its length. Van der Land and Narrevang (1977) presented de- tailed descriptions of the organization and histology of fluid-filled "neurular tubes" in the nervous system of La- mellibrachia Inymesi. They demonstrated that the ar- rangement of the tubes is similar to that described by Webb for L. barhami, except that a single tube extends only a short distance into the trunk region; the remaining nerve cord of the trunk lacks a neurular tube. They stated that, in the area of the brain, the tubes give off one or two side branches, which could correspond with Webb's observation. Because the opisthosome was not present on the specimens examined by Webb (see Jones, 198 la, p. 1310, 1985c, pp. 122-123) and by van der Land and N0rrevang, they could not comment on the organization of the nerve cord in that region. The wall of the tubes was described by van der Land and Nerrevang as "dense" and consisting of concentric fibers or lamellae in which nuclei are embedded. Following the terminology of van der Land and Norrev- ang ( 1977), Jones (198 la) described the organization of "neurular tubes" in Riftia pachyptila. In the vestimen- tum and trunk, the organization of the tubes reflects that of L. barhami. Because the opisthosome was present in Jones' specimens, he added the observation that the nerve cord in that region lacks a neurular tube. Among invertebrates (excluding the "invertebrate chordate" groups), neural tubes of varying construction have been reported in the nervous systems of some Nem- ertea, Sipuncula, Echiura, and Hemichordata, whereas giant axons, which may present the morphological ap- pearance of neural tubes in the light microscope, have been reported in some Nemertea, Phoronida, Sipuncula, Polychaeta, Oligochaeta. Cephalopoda, and Crustacea (for additional details and references, see Bullock and Horridge, 1965). Our TEM examination of the nerve cord of Ridgeia confirms the presence of paired giant ax- ons in the paired nerve cords of the vestimentum and a single giant axon in the nerve cord of the trunk. We sug- gest that the nerve cords of Lamellibrachia barhami and L. litymesi be re-examined by TEM to confirm the pres- ence of "tubes" or "neurular tubes," as reported by Webb (1969) and van der Land and Narrevang (1977), respectively, or giant axons, as suggested by our study of Ridgeia. The description of the wall of the neurular tube in L. luymesi by van der Land and N0rrevang (1977) is consistent with the morphological appearance of the myelin sheath around the giant axons of Ridgeia and R pachyptila, when viewed by TEM. This observation sug- gests that giant axons are present in L. luymesi. In the case of Riftia pachyptila, the nerve cord of the trunk has been examined by TEM (pers. obs., SLG) and the so- called "neurular tube" reported by Jones ( 198 la) is a gi- ant axon surrounded by a myelin sheath. Several alternatives may be suggested to explain the presence of two giant axons in the vestimentum and only a single axon in the trunk of Ridgeia and, presumably, all vestimentiferans. The single giant axon in the trunk may represent ( 1 ) the continuation of one of the giant axons that originate in the brain, (2) a third giant axon whose cell body is situated in the nerve cord near the border between the vestimentum and trunk, or (3) the product of the fusion of the two giant axons that origi- nate in the brain. Fusion of giant axons has been reported in the nervous systems of other invertebrate groups, e.g., cephalopods (Bullock and Horridge, 1965). In their de- scription of the "neurular tubes" in L. luymesi. van der Land and Norrevang (1977) included a reconstruction that shows the fusion of the tubes of the vestimentum to form the single tube of the trunk (near the posterior mar- gin of the ventral ciliated field). Our preliminary exami- nation of serial semi-thin sections (1.5 /urn) of Ridgeia appears to support van der Land and N0rrevang's obser- vation effusion. However, additional light microscopic and TEM observations are required to confirm the exact nature of the organization of the giant axons in Ridgeia. Giant axons are particularly well-developed among tube-dwelling animals, e.g., the phoronids and the sabel- lid and serpulid polychaetes (Bullock and Horridge, 1965) and are frequently implicated as an important 274 M. L. JONES ET AL component in the "startle response" of those animals, e.g.. the polychaete Myxicola (Nicol, 1948). A with- drawal response has been observed in situ in the case of Riftia pachyptila (pers. obs., MLJ). We suggest that the giant axons in vestimentiferan nervous systems have an important role in this response. In that regard, the myelin sheath observed around the giant axons of Ridgeia and R. pachyptila would enhance the speed of conduction of a nerve impulse in a manner analogous to the myelin sheath of vertebrate nerves (for additional discussion and references, see Jamieson, 1981). Further development of larvae and juveniles Of the characters of adult morphology followed in this study, the ventral ciliated field, vestimentum, and obtur- aculum are all established at a length (from prototroch or base of branchial filaments to posterior end) of 1.0 mm. There is a consistency of development of the ventral ciliated field from its "segmented" state to the fused field of the adult (Table I). Likewise, once the obturaculum appears during growth, it is always present, although it may be obscured by overlying branchial filaments. The vestimentum begins differentiation at a length of about 0.18 mm and is present, then, after a length of about 0.4 mm. There is a certain amount of variation in the initia- tion of vestimental development between specimens of 0. 1 8 and 0.4 mm in length (Table I). Larval setae, having fulfilled their apparent function in providing an early holdfast for the developing larva, are of variable occur- rence as they are worn, discarded, or extracted. The number of rows of opisthosomal setae remains approxi- mately constant at one or two, until a length of about 0.5 mm, and then new rows are added with considerable variation among specimens (see Delayed settlement, be- low). Branchial filaments develop in much the same manner, except that the variability in addition occurs earlier in development, at a length of about 0. 16 mm. Delayed settlement Of the speculations concerning the dispersion of the larvae of bivalved molluscs of the hydrothermal vents, there is a consensus that the mussel, Bathymodiolus ther- mophilux, has a pelagic (J. P. Grassle, 1985) or plankto- trophic (Lutz?/ a/., 1980; Berg, 1985; Turners a!., 1985; Lutz, 1988) stage that may be epipelagic (Berg, 1985) or demersal (Lutzetal., 1980; Turners al, 1985). The pos- sibility of delayed settlement of such planktotrophic lar- vae has been suggested by Lutz et al. (1980). Waren and Bouchet (1989) have concluded that gastropods of the hydrothermal vents have a lecithotrophic development and can delay settlement until the proper habitat is en- countered. Van Dover et al. (1985) have suggested a planktotrophic stage for the crab Bythograea thennydron and the shrimp Alvinocaris lusca and mention the possi- bility of delayed settlement for these species. Southward ( 1988b) proposed that, in the case of Rid- geia sp., there is early ciliary feeding in developing larvae or juveniles but did not indicate whether this would take place in a pelagic or a demersal phase. Gary et al. (1989) observed that spawned eggs of Riftia pachyptila have a positive buoyancy due, in large part, to a considerable amount of cytoplasmic lipids. They suggest a pelagic de- velopment in Riftia, either planktotrophic, based on the small size of eggs, or lecithotrophic, based on the lipid content of the eggs (about half of the cytoplasmic vol- ume). They further suggest that, as development pro- ceeds, buoyancy decreases and that, ultimately, the de- veloping Riftia return to the sea floor and settle in re- sponse to cues noted by Lutz et al. (1980), in the case of Bathymodiolus. The preceding observations bear on the following dis- cussion. Based upon the nine specimens sectioned (Table I), there is no apparent consistency in the presence of an anus or of internal bacteria, during development, as re- flected by length. Specimens G-nr. #4 and E-nr. #16 lack an anus and bacteria; H-nr. # 1 5 and A-nr. #2 1 have both an anus and bacteria; B-nr. #15 and C-nr. #21 have an anus but lack bacteria. These inconsistencies might have a reasonable explanation. It may be assumed that a developing embryo of Rid- geia will be positively buoyant (Gary et al., 1989). As a mouthless trochophore among the plankton, lecitho- trophic development can proceed only so far and further development would be postponed until a vent site is en- countered (Waren and Bouchet, 1989). If there is further development in the plankton, resulting in the formation of the ventral process and the mouth, a planktotrophic phase would be initiated (Berg, 1985; Turner etal., 1985; Lutz, 1988), ciliary feeding would be possible (South- ward, 1988b), and a later anal opening for the gut would establish an efficient, one-way digestive tract; along with this, there might be a reduction of larval buoyancy (Gary et al., 1989). Upon finding a hydrothermal vent, using cues like those suggested by Lutz et al. (1980), and set- tling there, the developing larva would be able to acquire sulfide-oxidizing bacteria for its trophosome. Because the trophosome will become internally isolated, there would now be no necessity for a hindgut or an anus; these could be resorbed and, over time, the ventral process, the mouth, and the foregut could be resorbed or become atrophied. Under this scheme the larval branchial fila- ments would serve as relatively simple respiratory sur- faces for an early larva still using its original supply of yolk or a later form, now feeding by cilia. The later branchial filaments, with their better-developed rows of cilia and the augmented respiratory surface afforded by EARLY DEVELOPMENT OF R1DGE1A 275 rows of pinnules, would serve as efficient structures for taking up, among other things, sulfide (for the endosym- biotic bacteria) and oxygen (for both the bacteria and the worm), both of which are crucial for the survival of the worm. If the developing larva does not quickly en- counter a hydrothermal vent, then it can survive by maintaining ciliary feeding, grows more than it might otherwise and, when finally arrived at a vent site, pro- ceeds with the acquisition of bacteria and the closing of the anal opening at a more advanced stage of develop- ment and at a length greater than its early-arriving neigh- bors. The following is suggested, in summary (Table I): 1. G-nr. #4, F-nr. #10, D-nr. #13, E-nr. #16, at the time of preservation, were still using yolk for nutrition or were feeding by cilia (no anus, no bacteria); 2. H-nr. #15 and A-nr. #21 had obtained endosymbi- otic bacteria and the anus could atrophy (with anus, with bacteria); 3. B-nr. #15 and C-nr. #21 were still ciliary feeders, newly arrived at the vent site, not yet with endosymbi- onts (with anus, no bacteria). Applying these summary statements, it might be that a larger size than expected, as in the case of D-nr. #13, A-nr. #2 1 , and C-nr. #2 1 (relative to G-nr. #4, H-nr. # 1 5, and B-nr. #15. respectively), suggests a delayed settle- ment. In addition, E-nr. #16, apart from size, has charac- ters that might place it near #7 or #8 (Table I); D-nr. #13 (Fig. 21) and C-nr. #21 (Fig. 23) show about the same relative development of the ventral process; and #3 1 and #35 (Table I) have fewer rows of opisthosomal setae than might be expected; perhaps E-nr. #16, C-nr. #21, #31 and #35 all had a delayed settlement, as well. Further, #5 (Fig. 6) and #14 (Fig. 7) have about the same develop- ment of the ventral process, whereas #15 (Fig. 8), al- though of about the same length as # 1 4, has a much bet- ter developed ventral process and its branchial filaments appear to be further along in development; this suggests that #14 might have had a delayed settlement. Of South- ward's specimens RB (nr. #32) and RC (nr. #33), both about the same length (Southward, 1988b), RC agrees best with the ranked specimens near #33, as regards de- velopment of characters, and RB would fit better at about #25; perhaps RB had a delayed settlement. Finally, Southward's specimen RJ (nr. #9), with a mouth, gut (without endosymbionts), and anus, would appear to be further along in development than our F-nr. #10 and D- nr. #13, in both of which the anus has not yet appeared. Acknowledgments We thank Cheryl F. Bright, National Museum of Nat- ural History, for her technical assistance and extended discussions throughout this investigation. We thank J. C. Harshbarger, Registry of Tumors of Lower Animals, for the use of his histology laboratory, W. R. Brown, S. G. Braden, and B. E. Kahn, Scanning Electron Microscope Laboratory, for the operation of their SEM instruments; both of these units are in the National Museum of Natu- ral History. For material used in this study, we thank J. F. Grassle, Woods Hole Oceanographic Institution, Chief Scientist for the expedition to the Galapagos Rift, De- cember 1979; A. Malahoff, University of Hawaii, Chief Scientist for the expedition to the Juan de Fuca Ridge. July 1984; K. L. Smith, Scripps Institution of Oceanog- raphy, Chief Scientist for the Oasis Expedition to the EPR at 2 1° North; and A. E. DeBevoise, Hatfield Marine Science Center, Oregon State University, who collected and preserved specimens during the expedition to the Juan de Fuca Ridge, September 1987, S. R. Hammond. Pacific Marine Environmental Laboratory, NOAA, Newport, OR, Chief Scientist. We thank M. H. Pettibone and K. Fauchald, National Museum of Natural History, for their reviews of the manuscript. This is Contribution No. 82 of the Galapagos Rift Biology Expedition, sup- ported by the National Science Foundation, and a con- tribution of the Oasis Expedition. Literature Cited Arp, A. J., J. J.Childress, andC. R. Fisher Jr. 1985. Blood gas trans- port in Riftia pachyptila. Pp. 289-300 in The Hydrolhermal I'ents of the Eastern Pacific: An Overview. M. L. Jones, ed. Bull Biol. Soc. Wash.. No. 6. Berg, C. J., Jr. 1985. Reproductive strategies of mollusks from abys- sal hydrothermal vent communities. Pp. 185-197 in The Hydro- lhermal I 'ents of the Eastern Pacific: An Overview. M. L. Jones, ed. Bull. Bioi Soc- Wash.. No. 6. Bosch, C., and P.-P. Grasse. 1984a. Cycle partiel des bacteries chimioautotrophes symbiotiques et leur rapports avec les bacterio- cytes chez Riftia pachyptila Jones (Pogonophore Vestimentifere). I. Le trophosome et les bacteriocytes. C. R Acad. Sci. Paris. Ser. III. 299:371-376. Bosch, C., and P.-P. Grasse. 1984b. Cycle partiel des bacteries chimioautotrophes symbiotiques et leur rapports avec les bacterio- cytes chez Riftia pachyptila Jones (Pogonophore Vestimentifere). II. L'evolution des bacteries symbiotiques et des bacteriocytes. C. R Acad. Sci. Paris. Ser. Ill, 299: 413-419. Bullock, T. H. and G. A. Horridge. 1965. Structure and Function in the Nervous Systems of Invertebrates. Vols. I and II. W. H. Freeman and Co.. San Francisco. xv+ and viii+, 1719 pp. de Burgh, M. E. 1986. Evidence for a physiological gradient in the vestimentiferan trophosome: size-frequency analysis of bacterial populations and trophosome chemistry. Can. J. Zool. 64: 1095- 1103. Gary, S. C., H. Felbeck, and N. D. Holland. 1989. Observations on the reproductive biology of the hydrothermal vent tube worm Riftia pachyptila. Mar. Biol. Prog. Ser. 52: 89-94. Cavanaugh, C. M. 1980. Symbiosis of chemoautotrophic bacteria and invertebrates. Biol. Bull 159: 457 (Abstr.). Cavanaugh, C. M. 1983a. Symbiotic chemoautotrophic bacteria in marine invertebrates from sulphide-rich habitats. Nature 302: 58- 61. Cavanaugh, C. M. 1983b. Chemoautotrophic bacteria in marine in- 276 M. L. JONES ET AL vertebrates from sulftde-nch habitats: a new symbiosis. Pp. 699- 708 in Endocytobiology II: In/racellular Space as Oligogenetic Eco- system, H. E. A. Schenk and W. Schwemmler, eds. Walter de Gruyter. New York. Cavanaugh, C. M., S. L. Gardiner, M. L. Jones, H. W. Jannasch, and J. B. Waterbury. 1981 . Prokaryotic cells in the hydrothermal vent tube worm Rijiia pachyptila Jones: possible chemoautolrophic symbionts. Science 213: 340-342. Crane, K. 1985. The distribution of geothermal fields along the Mid- Ocean Ridge. Pp. 3-18 in The Hydrothermal I'ents of the Eastern Pacific: An Overview. M. L. Jones, ed. Bull. Biol. Soc. Wash.. No. 6. Distel, D. L., D. J. Lane, G. J. Olsen, S. J. Giovannoni, B. Pace, N. R. Pace, D. A. Stahl, and H. Felbeck. 1988. Sulfur-oxidizing bacte- rial endosymbionts: analysis of phylogeny and specificity by 16S rRNA sequences. J. BaaerioL 170: 2506-25 10. Felbeck, H., and J. J. Childress. 1988. Rijiui pachyptila: a highly in- tegrated symbiosis. Pp. 131-138 in Actes du Collogue Hydrother- malisme, Biologie el Ecologie, L. Laubier, ed. Oceanol. Ada. Spe- cial Vol. No. 8. Felbeck, H..J.J. Childress, and G.N.Somero. 1981. Calvin-Benson cycle and sulphide oxidation enzymes in animals from sulphide- nch habitats. Nature 293: 291-293. Grassle, J. F. 1986. The ecology of deep-sea hydrothermal vent com- munities. Pp. 301-362 in Adv. Afar. Biol.. vol. 23, J. H. S. Baxter and A. J. Southward, eds. Academic Press. London. Grassle, J. P. 1985. Genetic differentiation in populations of hydro- thermal vent mussels (Bathymodiolus thermophilus [sic]) from the Galapagos Rift and 1 3°N on the East Pacific Rise. Pp. 429-442 in The Hydrothermal I 'enls oj the Eastern Pacific: An Overview, M. L. Jones, ed. Bull. Biol. Soc. Wash.. No. 6. Hand, S. C. 1987. Trophosome ultrastructure and the characteriza- tion of isolated bacteriocytes from invertebrate-sulfur bacteria sym- bioses. Biol. Bull. 173: 260-276. Jamieson, B. G. M. 1981 . The i 'Itrastructureofthe Oligochaeta. Aca- demic Press, NY. xv + 462 pp. Jannasch, H. \V., and C. O. Wirsen. 1979. Chemosynthetic primary production at east Pacific sea floor spreading centers. BioSciencc 29: 592-598. Jones, M. L. 1981a. Rijiia pachypiila, new genus, new species, the vestimentiferan worm from the Galapagos Rift geothermal vents (Pogonophora). Proc: Biol. Soc. Wa-.li. 93: 1295-1313. Jones, M. L. 1981 b. Rijiia pachyptila Jones: observations on the ves- timentiferan worm from the Galapagos Rift. Science 213: 333-336. Jones, M. L. 1985a. Vestimentiferan pogonophores: their biology and affinities. Pp. 327-342 in The Origins and Relationships of Lower Invertebrates. S. C. Morris, J. D. George. R. Gibson and H. M. Platt. eds. The Systematics Association, Special Volume No. 28. Clarendon Press. Oxford. Jones, M. L. 1985b. Vestimentifera, their biology and systematic and evolutionary paterns [sic]. Program for the Colloque "Les sources hydrothermales de la ride du Pacifique Oriental: Biologie et ecolo- gie." IFREMER, Paris, 4-7 November 1985 (Abstr). Jones, M. L. 1985c. On the Vestimentifera, new phylum: six new species, and other taxa, from the hydrothermal vents and elsewhere. Pp. 1 1 7- 1 58 in The Hydrolhemial I 'cuts of the Eastern Pacific: An Overview. M. L. Jones, ed. Bull. Biol. Soc. Wash.. No. 6. Jones, M. L. 1987. Recent observations on the Vestimentifera. Pp. 2 I -22 in Abstracts of the 1? 3rd Annual Meeting of the AAAS Jones, M. L. I988a. The Vestimentifera of the Pacific and elsewhere. Program for the Third International Symposium on Marine Bioge- ography & Evolution in the Pacific, Western Society of Naturalists/ University of Hong Kong, Hong Kong, 26 June-3 July 1988, p. 13 (Abstr.). Jones, M. L. 1988b. The Vestimentifera, their biology, systematic and evolutionary patterns. Pp. 69-82 in Actes du Colloque Hvdro- thermahsme, Biologie et Ecologie. L. Laubier. ed. Oceanol. Ada, Special Vol. No. 8. Jones, M. L., and C. F. Bright. 1985. Bibliography of hydrothermal vents and related areas, their biotas, ecological parameters and an- cillary data. Pp. 495-538 in The Hydrothermal Vents of the Eastern Pacific: An Overview, M. L. Jones, ed. Bull. Biol Soc. Wash.. No. 6. Jones, M. L., and S. L. Gardiner. 1988. Evidence for a transient di- gestive tract in Vestimentifera. Proc. Biol. Soc. Wash. 101: 423- 433. van der Land, J. L., and A. Nerrevang. 1975. The systematic position of Lamellibrachia [sic] (Annelida. Vestimentifera). Pp. 86-101 in The Phylogeny and Systematic Position of Pogonophora, A. Nerrev- ang, ed. Z. Zoo/. Sysl. Evohttionsforsch.. Sonderheft. van der Land, J. L., and A. Narrevang. 1977. Structure and relation- ships of Lamellibrachia (Annelida, Vestimentifera). A'. Dan I'ide- nsk Selsk. Biol. Skr. 21(3): 1-102. Lutz, R. A. 1988. Dispersal of organisms at deep-sea hydrothermal vents: a review. Pp. 23-29 in Acles du Colloque Hydro/herma/isme. Biologie el Ecologie, L. Laubier, ed. Oceanol Ada. Special Vol. No. 8. Lutz, R. A., D. Jablonski, D. C. Rhoads, and R. D. Turner. 1980. Larval dispersal of a deep-sea hydrothermal vent bivalve from the Galapagos Rift. Mar. Biol. 57: 127-133. Mane-Garzon, F., and R. Montero. 1986. Sobre una nueva forma de verme tubicola — Lamellibrachia victori [sic] n. sp. (Vestimentif- era)— Proposicion de un nuevo phylum: Mesoneurophora. Rev. Biot. Uruguay [for 1985?] 8(1): 1-28. Nicol, J. A. C. 19-18. The giant nerve-fibres in the central nervous system of Myxicola (Polychaeta, Sabellidae). Q. J. Microsc. Sci.. Scr 2.S9: 1-45. Nielsen, C. 1987. Structure and function of metazoan ciliary bands and their phylogenetic significance. Ada Zool. 68: 205-262. Southward, E. 1988a. Ultrastructure of the developmental stages of the symbiosis between bacteria and vestimentiferan Pogonophora. Program for the Fifth Deep-Sea Biology Symposium. IFREMER, Brest. France, 26 June-1 July 1988 [not paginated] (Abstr.). Southward, E. 1988b. Development of the gut and segmentation of newly settled stages of Ridgeia (Vestimentifera): implications for relationship between Vestimentifera and Pogonophora. / Afar. Biol.Assoc. r K 68: 465-487. Turner, R. D., Lutz, R. A., and D. Jablonski. 1985. Modes of mollus- can larval development at deep-sea hydrothermal vents. Pp. 167- 1 84 in The Hydrothermal I 'ent.i of the Eastern Pacific: An Overview, M. L. Jones, ed. Bull. Biol. Soc. Wash.. No. 6. Van Dover, C. L., J. R. Factor, A. B. Williams, and C. J. Berg Jr. 1985. Reproductive patterns of decapod crustaceans from hydro- thermal vents. Pp. 223-227 in The Hydrothermal I 'ents of the East- ern Pacific An Oven'iew. M. L. Jones, ed. Bull. Biol. Soc. Wash.. No. 6. Waren, A. and P. Bouchet. 1989. New gastropods from East Pacific hydrothermal vents. Zool. Scr. 18: 67-102. Webb, M. 1969. Lamellibrachia barhami. gen. nov., sp. nov. (Pogo- nophora). from the northeast Pacific. Bull. Mar. Sci. 19: 18-47. Webb, M. 1977. Studies on Lamellibrachia barhami (Pogonophora). II. The reproductive organs. Zool. Jahrb. Abt. Anal. Ontog. Tiere 97:455-481. Reference: Biol Bull, 177: 277-286. (October. 1989) Variation in Growth Rate and Reproduction of the Bryozoan Bugula neritina MICHAEL J. KEOUGH Department of Zoology. University of Melbourne. Parkville. \ 'ictoria 3052 Australia Abstract. Colonies of the arborescent cheilostome bryozoan Bugula neritina vary dramatically in their growth rate even when in apparently identical microhab- itats. Comparison of growth rates of juveniles derived from four parent colonies at each of two sites showed only weak effects of parental colony on juvenile growth. These effects accounted for at most only 5.4% of total variation in growth. Variation in growth, and hence age at first reproduction, is interpreted as a plastic response of colonies to fine-scale environmental variation. Bryozoans from seagrass meadows mature at a smaller size than those colonies from nearby rocky reefs ( 1 200, vs. 3500 zooids at first reproduction, respectively). When juveniles from both of these habitats were grown in a common garden, there was, again, no variation among parental groups, but a highly significant effect of origin of juveniles. Juveniles matured at a size similar to that seen in their parental population, indicating that genetic or very early maternal effects influence timing of repro- duction. A post hoc test of the effect of onset of reproduction on colony growth showed no reduction in growth rate. Instead, colonies that reproduced grew faster than sim- ilar aged and sized colonies that did not reproduce. Introduction A growing body of empirical evidence suggests that, for modular organisms, many demographic variables de- pend more strongly on size than on age. Three of the most important such variables are mortality rates, timing of onset of reproduction, and reproductive output. Mor- tality rates are often size-dependent in two ways; first, the probability of a colony being eaten or overgrown may decrease with increasing size (e.g., Reiswig, 1974; Wer- Received 15 March 1989; accepted 31 July 1989. nerandCaswell, 1977; Gross, 1981; Antonovics and Pri- mack, 1982; Russ, 1982; Sebens, 1982; Hughes and Jackson, 1985; Hughes and Connell, 1987). Second, small colonies may be killed completely when attacked, while a similar attack on a larger colony may only cause damage ("partial mortality"), from which the colony subsequently recovers (Bak et al.. 1981; Palumbi and Jackson. 1982). Size, rather than age, may determine the onset of reproduction for many clonal animals and plants (Inouye and Taylor, 1980; van Duyl et a/.. 1981; Gross, 1981; Wahle, 1983; Augspurger, 1985; Keough, 1 986; but see Harvell and Grosberg, 1 988). For a number of clonal organisms, size is correlated positively with re- productive output (Hayward, 1973; Inouye and Taylor, 1980; van Duyl et al.. 1981; Nakauchi, 1982; Sebens, 1983; Wahle, 1983; Augspurger, 1985), which in turn is thought to be an important component of relative fit- ness. A negative correlation between size and mortality and a positive correlation between size and reproductive out- put both favor rapid initial growth of juveniles. However, many clonal organisms show extensive variation in growth rate. Harper ( 1977; 1985) reviews data for terres- trial plants, and Hughes and Jackson (1985), Jackson and Winston (1982), and Keough (1986) provide exam- ples of such variation in clonal marine animals. For most marine organisms, the causes of this variation have not been examined in enough detail to estimate relative con- tributions of phenotypic responses to fine-scale environ- mental variation and genetic or maternal factors. For many plants, however, considerable variation in mor- phology and growth can be attributed to phenotypic plasticity in response to small-scale environmental varia- tion (Silander, 1985). In animals, both genetic and envi- ronmental influences on growth rate are reported com- monly (e.g., Travis et al.. 1987). The cosmopolitan bryozoan Bugula neritina appears 277 278 M. J. KEOUGH 20 Figure 5 10 15 Colony Size Onset of reproduction in Bugula neritina. The figure shows the percentage of colonies reproductive as a function of size for: Solid triangles: colonies in seagrass in the northeastern Gulf of Mexico. The curve is a composite from the sites and experiments reported else- where (Keough, 1986, Keough and Chernoff, 1987). n > 1000. Solid squares: colonies living on rock surfaces at Panama City. Florida. Feb- ruary 1986, n = 1 54. Open squares: colonies living on rock surfaces at Santa Catalina Island, California. Data from Keough (in prep.). n > 500. to fit these emergent generalizations for clonal organ- isms. It is an arborescent species, in which a single colony represents the genet. Neither regrowth from dispersed fragments nor fusion between neighboring colonies has ever been observed. Fertilization is internal and larvae are brooded singly by maternal zooids in hyperstomial ovicells. The relatedness between larvae within a single colony is unclear, barring somatic mutation, the mater- nal zooids are genetically identical, but the number of paternal genotypes is unknown. Larvae from a single col- ony are at least half-sibs and at most full sibs. The onset of reproductive activity is more closely associated with size than with age, and survivorship increases with some function of size or age (Keough, 1986; Keough and Cher- noff, 1987). In this paper, I re-analyze earlier data to show effects of both size and age on survivorship. When colonies are small, they suffer complete mortality, while larger colonies are frequently damaged, rather than killed outright (Keough, unpub. ob.). Growth rates vary extensively within cohorts, even when colonies are grow- ing in what appears to be a relatively homogeneous habi- tat— the distal tips of artificial seagrass leaves (Keough, 1986). Similar variation occurs among cohort members attached to rock surfaces (Keough, 1989). In addition to variation within populations, size at first reproduction also varies among populations. Bugula neritina colonies from seagrass meadows in the north- eastern Gulf of Mexico grow rapidly in spring and fall, reproduce while small, and generally are short-lived, probably because their substratum is ephemeral (Keough, 1986; Keough and Chernoff, 1987). In con- trast, colonies living on rock faces in southern California grow large, do not reproduce until they are large, and live considerably longer than one year (Keough, 1989). At Panama City, in the northeastern Gulf of Mexico, colo- nies live on large permanent structures, including artifi- cial reefs, natural limestone outcroppings, and rock jet- ties. In this habitat, the demography of B. neritina ap- pears closer to that seen in southern California (Fig. 1): they grow large, and appear not to reproduce while small. What are the causes of this demographic variation? Here I describe experiments to determine causes of varia- tion in growth rate and size of reproduction. I partition within-cohort variation in post-settlement growth rate of juveniles into variation occurring among and within ma- ternal colonies, with the aim of detecting genetic compo- nents of growth rate. I also examine the basis of geo- graphic variation in size at first reproduction. Materials and Methods To distinguish between effects of age and size on survi- vorship, I re-analyzed data from Bugula neritina colo- nies in seagrass meadows in Florida. In these experi- ments, cohorts of juveniles were established in the labo- ratory, then transplanted to the field and monitored (see Keough, 1986 for details). At any point in time, then, the experimental population comprised colonies from ear- lier cohorts that had grown rapidly (large, old), similar, slower-growing colonies (small, old), and small, young colonies from later cohorts. For colonies living in sea- grass meadows, survivorship appears dependent on both size and age (Fig. 2). oo 3 40 - 20 - Age (weeks) Figure 2. Effects of age and size on survivorship of Bugula neritina colonies. Histograms show probabilities of survival for juveniles of different sizes from a single experimental cohort, and juveniles of the same size, but of different ages. Data were taken from Keough (1986). BRYOZOAN GROWTH AND REPRODUCTION 279 • Reproductive I I Not reproductive 0123456789 Colony Size (No. bifurcations) Figure 3. Reproduction as a function of size in Bugula neritina col- onies under experimental conditions. Data are shown for experimental juveniles from Panama City after 5 weeks, and for natural recruits de- rived from Alligator Point after the same period of time. I 'aria! ion in growth At the time of the first experiment (25 March 1986), there were no mature Bugula neritina colonies in sea- grass meadows or at Alligator Point marina, which had been the source population for other demographic work (Keough, 1986; Keough and Chernoff, 1987), and where colonies mature at the same size as those from seagrass meadows (Keough, unpub. obs.). Therefore, I collected colonies from a sunken barge approximately 4 km off Panama City, at a depth of 20 m. Parental colonies were separated from each other by >10 m, to reduce the chance that they were closely related. After 2 days in the dark, the colonies were exposed to bright light and the four colonies that released the most larvae were used in the experiment. Four or five larvae from a given colony were pipetted individually into a sterile plastic petri dish (50-mm diam- eter) containing seawater, and allowed to settle and metamorphose. Juveniles close to the edges of plates were removed. Without removing it from the water, each dish then had a 4-mm hole drilled into its center, and was bolted to a 60 X 20 cm piece of clear plexiglass. Each plexiglass sheet held 16 dishes in two rows of 8, with 4 dishes from each parental colony. The dishes were not arranged randomly, but in four 2x2 arrays, each array containing one dish from each parental colony. Two of the arrays were at ends of the plexiglass plates, and the positions of the dishes were further constrained so each parent appeared once at the corner of a plate and three times away from the corners. This design ensured that any micro-environmental variation could not bias the outcome of the experiment. The plexiglass plates were then transported in seawater to a study site in the Thalassia meadow at Lanark, 1 km from the F.S.U. marine laboratory (see Keough and Chernoff, 1 987, for a description of this site). There, they were bolted to a l-m: PVC frame, the legs of which had been driven into the sediment until the frame was ap- proximately 1 5 cm above, and parallel to, the substra- tum. All fasteners used in the experiment were stainless steel. The petri dishes faced downwards, with the plexi- glass above them, because Young and Chia (1984) have shown that sedimentation can be a major source of mor- tality for some newly settled sessile animals, and my aim was to maximize the number of juveniles available for growth measurements. At weekly intervals I returned, unbolted the plates, and transported them back to the laboratory, where I measured each juvenile. The plates were kept in running seawater, from which they were never removed for more than a few seconds. Bugula neritina colonies in north Florida grow with regular bifurcations. New pairs of zooids are produced (budded) at the distal tip of each branch, with usually four pairs of zooids (budding events) between each bifur- cation point (see Keough and Chernoff, 1987, their Fig. 2, for more details). Unless the colony has been dam- aged, each branch is approximately the same size, and the colony can be viewed as a series of distal buddings, with "waves" of bifurcations at regular intervals. I mea- sured size by counting first the number of bifurcation waves and then the number of zooid pairs at the growing tip of each branch, i.e., the number of budding events since the last bifurcation. This measurement can be used to estimate either the number of zooids or to compute the number of times the colony has budded. Thus, a col- ony of size 3.2 (bifurcated 3 times, with 2 further zooid pairs after the last bifurcation point) has budded 14 (i.e., 3 bifurcations X 4 zooids/bifurcation + 2 zooids) times, and has 88 zooids. When measuring each colony I noted any missing branches, pale colored zooids, partial foul- ing, and the presence or absence of embryos. After the third week, some of the juveniles were ap- proaching a size at which they grew close to their neigh- bors, so I reduced their numbers to 1 or 2 per dish. In the first experiment I used 5 such plates, with a total of 247 juveniles, distributed approximately equally among the 4 parents (57, 73, 57 and 60 juveniles). Three weeks into the experiment, I obtained fertile col- onies from Alligator Point and did a second experiment, using four parental colonies and four plexiglass sheets. The protocol for this experiment was identical to that of the first one, with colonies retained in collecting buckets for a period similar to travel time from Panama City to the laboratory, and the experiment began on 15 April 280 M. J. K.EOUGH 1986. There were 43, 51, 39, and 51 juveniles from the four parents. Analysis The plexiglass plates were a logistical convenience and were almost touching each other. They were not replaced in the same arrangement each week. A preliminary anal- ysis of variance for the first week's growth, using mean growth of all juveniles in a dish as the dependent vari- able, showed neither a significant effect of plates nor a parent x plate interaction. For these reasons, I ignored plates in the analysis, and had simply a nested design, with juveniles within petri dishes within parents. The ju- veniles were small and well-separated from each other, so I did not expect juveniles to interfere with each other so as to make their growth non-independent. Any corre- lation among juveniles within a dish should reflect com- mon responses to microhabitat conditions within dishes. This assumption does not affect the test of the main hy- pothesis about parental colonies, however, because growth rates of individual juveniles within dishes are not used to test the effect of parents (which is tested using variation among dishes). I used the number of budding events as the measure of growth. Two analyses are possible from this experi- mental design. First, the weekly growth increment of each juvenile can be used as the dependent variable in a series of separate analyses for each week. This experi- ment is a simple nested design (juveniles within dishes, dishes within parent colonies), and has the drawback that statistical tests based on each week may not be indepen- dent, if juvenile growth over one period influences subse- quent growth. This analysis cannot detect juveniles that grow consistently faster or slower, in successive weeks. The alternative analysis is to use only those juveniles that survived undamaged and were not culled, in a repeated measures analysis, with weekly growth increment as the dependent variable. The design is complicated, with nested factors (dishes) and repeated factors (weeks), and has two deficiencies; first, no information is gained from colonies that may have survived for most of the time, but were culled or damaged late in the experiment. The power of the analysis of very early growth is reduced by the decreased sample sizes. Second, repeated-measures analyses of variance have restrictive assumptions about the structure of covariance matrices (Winer, 1971 ), and the levels of the repeated factor (weeks) are always ap- plied in a fixed, rather than random, order. Here, I present the results of both sets of analyses, but their relative strengths and weaknesses must be consid- ered. Table I .\ihilysis of weekly growth increments for Bugula neritina colonies from Panama City Week after settlement Variable 1 3 Mean squares Among parent colonies 9.7 0.5 26.3 104.4 30.7 Dishes within colonies 2.9 7.4 26.5 78.2 57.5 "Siblings" within dishes 1.4 7.5 25.0 42.8 51.6 Degrees oj freedom Parent colonies 33333 Dishes — actual (quasi) 71(64) 71(58) 68(57) 67(57)61 Juveniles 172 163 154 113 67 F-statistics Colonies 3.2* O.lns 1.0ns 1.2ns 0.5ns Dishes2.0*** 1.0ns l.lns 1.8* l.lns Variance components Colonies 5.4 0.0 0.0 0.8 0.0 Dishes within colonies 22.4 0.0 1.8 24.2 5.5 Siblings within dishes 72.2 102.3 98.2 75.0 96.1 The dependent variable was the number of budding events in the week concerned. The analysis was a 2-level nested analysis of variance with unequal samples sizes, using quasi F-ratios to test the main effect of parental colonies. The table shows F-ratios, mean squares, degrees of freedom, and estimated variance components. The degrees of free- dom are also shown for the composite denominators used to test the main effect of parental colonies, ns, P> 0.05; *0.05 > P> 0.01; **0.01 > P > 0.00 l:***P< 0.001. Size at first reproduction At the beginning of the experiment there was natural recruitment onto the petri dishes and the plexiglass plates. These larvae were probably produced by the pop- ulation established at the Lanark site during a previous experiment using Alligator Point parents (Keough and Chernoff. 1987). To make a comparison of the size at first reproduction between colonies from Panama City and those occurring within seagrass meadows, I recorded the positions of some of these early recruits. Up through the fourth week I recorded the sizes of any Bugula colo- nies that were reproductive. In the fifth week I identified each natural recruit that had bifurcated more than four times and measured its exact size and reproductive con- dition. Results The growth rates of progeny from different Panama City colonies varied little. Only in the first week after set- tlement was there significant heterogeneity among pa- rental colonies (Table I), and even then, only 5.4% of the total variation in growth rates was accounted for by parent colonies. The repeated measures analysis of vari- BRYOZOAN GROWTH AND REPRODUCTION 281 Table II Repeated measures analysis of variance using weekly growth increments oj all Bugula colonies that survived undamaged to the end of the experiment Panama City Alligator Point Source of variation DF MS F DF MS F 3 3.20 1.34ns 11.85 4.92** Among colonies Siblings within colonies 80 2.39 145 2.41 Amongweeks 4 107.94 20.11*** 1 159.25 51.84*** Colonies x weeks 12 4.36 0.81ns 3 9.62 3.13*** Weeks x siblings 320 5.37 145 3.07 Data were unbalanced, and sums of squares were computed by un- weighted means (Winer 1971). ns, P > 0.05; **P < 0.0 1 ; ***/" < 0.00 1 . ance using only those juveniles that survived undamaged for 5 weeks showed no significant heterogeneity among parents, but highly significant variation with time (Table II). Inspection of mean weekly growth increments showed that the means did not differ by more than 1.5 buddings for the first few weeks (Table III). The clearest trend was a decline in the mean growth through the ex- periment, accompanied by a substantial increase in the variance (Table III). For the juveniles from Alligator Point, the repeated measures analysis showed a highly significant effect of parent colony on growth rates (Table II), but individual weekly analyses showed only a single significant result, with parental colonies explaining a maximum of 7.4% of growth variation (Table IV). Inspection of means showed that the differences among colonies were slight in the first week after settlement and larger in the second week, although variances increased during the second week (Table I II). In both experiments, variation among dishes of juve- niles from the same parent colony accounted for 1 5-20% of growth variation (Tables I, II, IV). Mortality of the juveniles in this protected environ- ment was usually <5% per week (Table V). Almost 80% of the Panama City juveniles survived for the 5 weeks of the experiment. There was no significant variation in post-settlement mortality among parental groups for the Panama City experiment, but slight variation among Al- ligator Point colonies (Table V). This latter result was caused by a single colony, the progeny of which suffered 23% mortality. The other three colonies did not differ significantly in the mortality rate of their juveniles (G = 3.62,df=2,P>0.10). Panama City colonies had embryos by the fourth week, and many (46% of 1 54 colonies) were reproductive after the fifth week. The proportion of reproductive colo- nies increased with colony size (Fig. 2), and the mean size at first reproduction was 7.3 bifurcations (31.3 bud- dings, SD = 3.27, n = 92). Natural recruits matured at a smaller size, with a mean of 6.1 bifurcations (i.e., 25 buddings, SD = 3.18, n = 58). Although most colonies were fouled by late May, there were a few unfouled Alli- gator Point colonies that began to reproduce. These colo- nies matured at a size similar to the naturally recruited juveniles (mean size at 1st reproduction 5.2, 22.3 bud- dings, SD = 2.6, n = 7). The natural recruits and Alliga- tor Point juveniles did not differ significantly from each other, but both differed from the Panama City juveniles (single factor analysis of variance with unequal sample sizes, using size at first reproduction as the dependent variable, F2.,54 = 82.7, P « 0.001; SNK procedure indi- cated that natural and Alligator Point means differed from Panama City at P = 0.00 1 ). Colonies became fouled by algae during the later stages of both experiments, and large colonies often lost branches. Eight of the 2 1 Panama City juveniles that had begun to reproduce by the end of the fourth week had all reproductive structures removed during the fifth week, leaving them as small, asymmetric colonies. There was no major reduction in colony growth rate associated with the onset of reproduction. No colony re- produced within the first three weeks, so I compared the growth rates of colonies that eventually reproduced to the rates of colonies that failed to reproduce. I used the growth rate for the first three weeks as a measure of pre- reproductive growth rate, and weeks 4-5 to estimate "post-reproductive" growth rates. I pooled data across parent colonies, since that factor accounted for little vari- ation in growth rate. For each colony I used the mean growth rate for the first three weeks and the mean growth rate for weeks 4-5, and I analyzed the data using a re- peated measures analysis of variance with two factors: presence or absence of reproduction, and growth period (pre- and post-reproduction). Only undamaged colonies were used in the analysis. Surprisingly, colonies that re- produced showed slightly increased growth at the onset of reproduction, while those that did not reproduce had a diminished growth rate in weeks 4 and 5 (Table VI). Size at first reproduction was not significantly hetero- geneous among parental groups for the Panama City ju- veniles (single factor analysis of variance using undam- aged colonies, F487 = 2.216, P = 0.074). The number of reproductive colonies from the Alligator Point experi- ment was too small for such an analysis. Size at first re- production was independent of growth rate for the Pan- ama City juveniles: using mean growth rate for the first two weeks and size at first reproduction, R2 = 0.007. The only significant correlation was between growth rate and age at first reproduction (R2 = 0.07, r = -0.239, n = 92, P = 0.02), and even that relationship was weak. These 282 M. J. KEOUGH Table III Mean weekly growth inc rements for Bugula colonies from Panama City and Alligator Point Colony Weeks 1 2 3 4 5 Panama City 1 X 8.3(1.5) 5.7(3.1) 3.6(3.9) 4.0(7.5) 3.8(7.1) n 57 53 48 39 26 3 X 8.3(1.4) 5.7(2.2) 3.0(5.0) 2.3(8.9) 2.7(7.5) n 73 72 71 59 41 4 X 7.5(1.1) 5.9(2.1) 2.2(6.5) 3.2(7.5) 1.9(7.9) n 57 57 55 43 30 6 X 7.7(1.4) 5.7(3.4) 3.7(4.2) 5.8(4.8) 4.0(6.9) n 60 56 52 43 35 Alligator Point 3 X 7.0(1.7) 8.2(3.5) n 30 40 4 X 6.2(1.1) 6.4(5.0) n 51 48 5 X 6.4(1.6) 4.4(7.1) n 39 35 6 X 6.5(1.3) 5.7(4.5) n 51 47 Dataware means of the number ot budding events, with standard deviations in parentheses. All undamaged colonies were included in each week, regardless of their subsequent fates. relationships are as expected when size, rather than age, is important. Discussion The two demographic variables described in this study showed contrasting patterns of variation. Growth rates were very plastic, varying primarily among juveniles from the same parent, with only a minor component of variation that could be ascribed to parental colony, and little apparent variation between colonies from the two collection sites. In contrast, size at first reproduction var- ied strikingly among the progeny of colonies from two different sites, but showed relatively little variation within populations. The experiments were designed to partition variation in juvenile growth into that occurring among and within "sibships." It is unclear whether these juveniles are closer to half or full sibs, because of uncertainty about the method of fertilization in most bryozoans; for Bugula neritina. electrophoretic attempts to resolve paternity have been thwarted by low levels of variability and few useable allozymes (Keough, unpub. obs.). Variation among sibships reflects mostly maternal (ge- netic + maternal environmental), and possibly paternal genetic, effects, while variation within sibships should re- flect phenotypic plasticity, genotype-environment inter- actions and paternal effects. In the "common garden" environment used in this study, there was little evidence of any maternal component, and, therefore, little evi- dence of additive genetic variation in growth rate. The largest amount of growth variation attributable to mater- nal colonies in any one week was 7.2%. Even when there were visible differences in mean growth rates of juveniles in a particular week, the rank order of sibships was not Table IV Analyses of weekly growth for Bugula nentina juveniles from Alligator Point Source of variation DF MS Var. Comp Growth—Week 1 Among colonies 3 5.6 2.9 Dishes within colonies 58(54) 2.8 2.0ns 17.3 "Siblings" within dishes 122 1.7 1.6* 79.9 Growth— Week 2 Among colonies 3 95.3 5.1 Dishes within colonies 58(51) 35.2 2.6ns 19.8 "Siblings" within dishes 108 20.5 1.7* 72.2 The dependent variable was the number of buddings, and the analy- sis was a 2-level nested analysis of variance with unequal samples sizes, using quasi F-ratios to test the main effect. The table shows F-ratios. mean squares, degrees of freedom and variance component analysis. The degrees of freedom are also shown in parentheses for each compos- ite denominator used to test the effect of colonies, ns, P > 0.05; *0.05 BRYOZOAN GROWTH AND REPRODUCTION 283 Table V (I 'eekly mortality of juveniles from different Panama City and Alligator Point parents Colony # 1 3 Total Panama City 1 10.0(57) 3.5(57) 3.6(55) 15.2(46) 2.3(43) 16.0(50) 3 0.0(73) 1.4(73) 6.9(72) 12.5(64) 4.2(48) 19.0(62) 4 0.0(57) 1.8(57) 12.5(56) 9.5(42) 5.4(37) 25.8(49) 6 0.0(60) 5.0(60) 8.8(57) 4.2(48) 2.2(46) 28.6(53) Pooled 0.0(247) 2.8(247) 6.7 (240) 8.8(217) 3.4(174) 21.7(214) Alligator Pt 3 2.3(43) 4 1 1.8(51) 5 23.1 (39) 6 9.8(51) Pooled 11.4(184) All juveniles were pooled, regardless of dishes. Mortalities are shown as percentages, with sample size in parentheses. G-test to compare mortality among colonies: Panama City. G = 4.37. df = 3, P> 0.1: Alligator Pt, G = 9.0, df = 3, P<0.05. maintained in subsequent weeks, so no groups of juve- niles grew consistently faster than others. Usually about 20% of variation in growth was attributable to dishes, suggesting that there may have been slight differences in microhabitat among dishes. However, most of the varia- tion occurred among related juveniles within single petri dishes. The causes of this latter variation are unclear. In the presence of apparently strong selection for rapid growth rates, it is not surprising to find little remaining variation that could be attributed to maternal genetic effects (Fisher 1958). However, responses to selection on single traits, such as growth rate, may depend on selec- tion of other traits (Via and Lande, 1 985 ), and small neg- ative genetic correlations between traits may retard ap- proaches to selective equilibria. There was no noticeable difference in growth rate be- tween juveniles from the two sites, although both cohorts showed variation through time. I have previously shown considerable seasonal variation in juvenile growth of co- horts from the Alligator Point population (Keough, 1986). Juveniles suffered very low mortality over five weeks. The cumulative mortality over this whole period was less than most weekly mortality rates for juveniles trans- planted onto distal sections of artificial seagrass leaves at the same site (Keough, 1986; Keough and Chernoff, 1987). All of these juveniles were handled in the same way. The petri dish experiments demonstrate at least that the mortality of juveniles on seagrass leaves is not a han- dling effect, but represents the action of crawling preda- tors, sedimentation, or abrasion of leaves, because these processes were prevented from occurring on the petri dishes. Algal growth may kill juvenile ascidians (Young and Chia, 1984), but algae seem unimportant in my study, because the petri dishes had markedly higher standing crops of algae than do seagrass mimics or natu- ral Thalassia leaves. Growth rates of juveniles attached to artificial seagrass blades are more variable than those of juveniles in petri dishes (cf. Keough, 1986; Keough and Chernoff, 1987). On seagrasses, by the time the first juveniles begin to re- produce, others may not have grown since settlement. The greater variation in growth may be due to an envi- ronmental difference between the inverted petri dishes and the distal 10 cm of plastic seagrass mimics, or, more precisely, that the seagrass mimics, and by implication, seagrass leaves, are a more variable environment than the petri dishes. The two sets of experiments were done in different years, and food supply may have been more variable in 1985, when artificial seagrasses were used, than in 1986 in the petri dishes. However, similar results from another seagrass site in other seasons and in Cali- fornia suggest that juveniles on substrata mimicing natu- ral ones generally show extensive within-cohort varia- tion in growth (Keough, 1986; Keough and Chernoff, 1987; Keough unpub. obs.). Why do juveniles on the same part of a seagrass blade have such disparate growth rates? One plausible reason may be that the hydrody- namic "neighborhood" of a leaf is likely to vary both temporally and spatially, depending on local weather and growth and loss of surrounding blades. A single bryozoan probably experiences a range of conditions, so plasticity of growth may be more likely than specializa- tion. Theoretical considerations of the evolution of plas- ticity focus on variation between, rather than within, en- vironments (Via, 1987; Via and Lande, 1985), with ge- notypes dispersing between environments. It may be that for many sessile animals, dispersal between habitats, such as Alligator Point and Panama City, is rare, but that there is substantial small-scale variation within environ- 284 M. J. K.EOUGH Table VI Growth of Bugula_/nvi'H//« before and after the unset of reproduction Source of variation DF MS Reproductive group 1 13.15 10.64** Juveniles within reprod. gps 85 1.23 Before/after onset of reprod. 1 0.47 0.26 ns Reprod. group x onset 1 16.07 8.92** Onset x juveniles 85 1.80 Wks 1-3 Wks 4-5 n Reproduced 6.32 6.88 62 No reproduction 6.39 5.60 25 The analysis was a repeated measures analysis of variance, compar- ing the growth of juveniles that did and did not reproduce. The depen- dent variable was the average weekly growth (no. of buddings) of juve- niles; the repeated measure was the growth of juveniles before (weeks 1-3) and after (weeks 4-5) the onset of reproduction. Only undamaged colonies were used in the analysis. Data were unbalanced, and sums of squares were computed by unweighted means ( Winer, 1971). The fig- ures beneath the analysis of variance table are treatment means, ns, P > 0.05; **/> < 0.0 1 ; ***/> < 0.00 1 . ments. This possibility seems worthy of further investiga- tion. The two cohorts differed in their mean size at first re- production, with the juveniles from the Panama City rocky reef maturing at a larger size than those from Alli- gator Point (31 vs. 25 budding events). The number of zooids increases exponentially with the number of bud- dings, so this apparently small difference is equivalent to Alligator Point colonies maturing after producing around 1200 zooids, while those from Panama City did not reproduce until they had >3500 zooids. The growth rates of the two cohorts were similar (Alligator Point col- onies actually grew slightly faster), so Panama City colo- nies reproduced at least a week later than those from Alli- gator Point. A difference in onset of reproduction is consistent with other life history differences between the two popula- tions; on seagrasses, Bitgula colonies are necessarily short-lived because of the ephemerality of their substra- tum. They grow rapidly, and never get very large. The largest colonies have branched 1 2 times, but these colo- nies are rare (Figs. 4, 8 in Keough and Chernoff, 1987). In contrast to seagrasses, rocky reefs and jetties are more permanent habitats. Colonies near Panama City and at Santa Catalina Island are more long-lived: newly settled colonies have survived at Santa Catalina Island for > 1 year (Keough, unpub. obs.), by which time they were still considerably smaller than the largest colonies observed at that location. Colonies on rocky substrata also grow much larger than colonies on seagrasses, branching at least twice as many times, resulting in at least three or- ders of magnitude more zooids. Because these animals are clonal and most zooids bear embryos, later reproduc- tion results in more embryos per colony. Thus, rocky reef Biigula have an early commitment to asexual (clonal) propagation of zooids, with later sexual, dispersive prop- agation. Natural recruitment allowed me to test for any effect of transplantation by comparing the experimental cohorts and the natural juveniles drawn from parent colonies on seagrass. This comparison showed that transplanting had no substantial effect on size at first reproduction. In both experimental cohorts, the juveniles matured at approximately the same size as did colonies in their parental populations, suggesting a maternal component to size at first reproduction. It is impossible to determine whether this is a genetic or environmental effect, since Bugula larvae develop completely within maternal ovi- cells. For some solitary species, egg size can have a strong influence on larval morphology and development (Si- nervo and McEdward, 1988), but there are no compara- ble data for clonal species. It is not clear how maternal environmental effects on 400-^m larvae might influence size at first reproduction, since by this time even colonies from seagrass meadows have > 1200 asexually produced zooids. I suggest that a substantial genetic component to size at first reproduction is more likely, but there are no data to resolve this question. Reproduction did not result in any reduction in growth rate; rather, those colonies that began to repro- duce actually grew faster after the onset of reproduction, while those not reproducing had a reduced growth rate during the same period. The latter observation is not sur- prising; failure to reproduce may be a result of insuffi- cient food early in life or disease, which may subse- quently cause reduced growth. The experiment provides no evidence for a major cost associated with reproduc- tion. When a colony reproduces, >50% of zooids may bear spherical, 400-m diameter embryos, whose tissue masses probably exceed those of the maternal zooids. Production of larvae seems such a substantial invest- ment of resources that it is perhaps surprising that the allocation of considerable resources for the production of embryos had no effect on the absolute amount of re- sources devoted to clonal growth. Although growth variation and its causes have been examined in detail for many terrestrial and freshwater organisms (Berven and Gill, 1983; Travis, 1983), such studies are less common in marine environments. There exist relatively few documentations of the extent of growth variation that is not associated with obvious envi- ronmental gradients (e.g.. Koehn el al.. 1980; Levinton, 1983; Levinton and Monahan, 1983), or interactions with similar organisms (Peterson, 1982;Wethey, 1982), and the causes of this variation are obscure. In this case. BRYOZOAN GROWTH AND REPRODUCTION 285 do not interact strongly with other sessile organ- isms because of the abundance of free space on seagrass leaves (Chernoff, 1985), and although some growth vari- ation can be explained by basal-distal gradients in physi- cal conditions (see Luckenbach, 1984; Eckman, 1987, for a discussion of flow effects), there remains a substan- tial amount of growth variation that presumably is a re- sponse to fine-scale environmental heterogeneity. I sug- gest that the variability of their environment on very small scales reduces the likelihood of within-population genetic differentiation of growth rate. On larger, between-population scales, variation in re- productive variables is also well documented in other habitats, and is often associated with strong selection. Some marine species may switch between planktotrophy and lecithotrophy over parts of their range (Eyster, 1979) or within populations (Levin, 1984), or individuals may have the capacity to switch between iteroparity and semelparity in response to food availability (McKillup and Butler, 1979). In the present case, longevity of the substratum provides strong selection for rapid reproduc- tion on seagrasses. Not only is the substratum short-lived relative to rocks, but there is considerable variation in life-expectancy among seagrass tips, depending on her- bivory, leafage, and storms. The short planktonic period of larvae allows population differentiation over relatively small scales, since most rocky reefs in the northern Gulf of Mexico are in water deep enough for seagrasses to be rare. Thus, larvae are only likely to encounter one kind of substratum, and there should be little genetic ex- change between seagrass and rocky reef populations. McKillup and Butler (1979) suggested that species with long planktonic larval periods might have flexible repro- ductive patterns. Very restricted dispersal is widespread among subtidal sessile animals and some plants (Thor- son, 1950, for review), and population differentiation of reproductive patterns, rather than flexibility, may be more common in these organisms. Acknowledgments I am grateful to Jon Schmidt for his enthusiastic assis- tance in the field, and A. Butler and A. Davis for their comments on the manuscript. This work was supported by National Science Foundation grant OCE-8400404. This is a publication from the Florida State University Marine Laboratory. Literature Cited Antonovics, J., and R. B. Primack. 1982. Experimental ecological ge- netics in Plamago. VI. The demography of seedling transplants of P lanceolata. J Ecol. 70: 55-75. Augspurger, C. K. 1985. Demography and life history variation of Puya dasylirioides, a long-lived rosette in tropical subalpine bogs. CW«M45:341-352. Bak, R. P., J. Sybesma, and F. C. van Duyl. 1981. The ecology of the tropical compound ascidian Trididemnum solidum. II. Abundance, growth and survival. Mar. Ecol. Prog. Ser. 6: 43-52. Berven, K. A., and D. E. Gill. 1983. Interpreting geographic variation in life-history traits. Am. Zool. 23: 85-97. Chernoff, II. 1985. The distribution and abundance of epifauna on the seagrass ThaUissia lestudinum. Thesis, Florida State University. van Duyl, F. C, R. P. M. Bak, and J. Sybesma. 1981. The ecology of the tropical compound ascidian Trididemnum solidum. I. Repro- ductive strategy and larval behaviour. Mar. Ecol. Prog. Ser. 6: 35- 42. Eckman, J. E. 1987. The role of hydrodynamics in recruitment, growth, and survival of Argopcclcn irradians (L.) and Anomia sim- plex (D'Orbig,ny)wilhin eelgrass meadows. / Exp. Mar. Biol. Ecol. 106: 165-192. Eyster, L. S. 1979. Reproduction and developmental variability in the opisthobranch Te nellia pallida. Mar. Biol. 51: 133-140. Fisher, R. A. 1958. The Genetical Theory of Natural Selection. Do- ver, England. Gross, K. L. 1981. Predictions of fate from rosette size in four "bien- nial" plant species: \ 'erbascum thapsus. Oenolhera biennis. Dauciis carota. and Tragopogon dubius. Oecologia 48: 209-2 1 3. Harper, J. H. 1977. Population Biology of Plants. Academic Press, New York, New York, USA. Harper, J. L. 1985. Modules, branches, and the capture of resources. Pp. 1-34 in Population Biology and Evolution of Clonal Organisms. J. B. C. Jackson. L. W. Buss and R. E. Cook, eds., Yale University Press, Connecticut. Harvell, C. D., and R. K. Grosberg. 1988. The timing of sexual matu- rity in clonal animals. Ecology 69: 1855-1 864. Hayward, P. J. 1973. Preliminary observations on settlement and growth in populations of Alcyonidium hirsutum (Fleming). Pp. 107-1 13 in Living and Fossil Bryozoa. G. P. Larwood and B. R. Rosen, eds. Academic Press, New York. Hughes, T. P., and J. H. Connell. 1987. Population dynamics based on size or age? A reef coral analysis. Am. Nat. 129:818-829. Hughes, T. P., and J. B. C. Jackson. 1985. Population dynamics and life histories of foliaceous corals. Ecol. Monogr. 55: 141-166. Inouye, D. W., and O. R. Taylor. 1980. Variation in generation time in Frasera speciosa (Gentianaceae), a long-lived perennial mono- carp. Oecologia 47: 1 7 1 - 1 74. Jackson, J. B. C. 1985. Distribution and ecology of clonal and aclo- nal benthic invertebrates. Pp. 297-356 in Population Biology and Evolution of Clonal Organisms. J. B. C. Jackson, L. W. Buss, and R. E. Cook, eds. Yale University Press, CT. Jackson, J. B. C, and J. E. Winston. 1982. Ecology of cryptic coral reef communities. I. Distribution and abundance of major groups of encrusting organisms. ./. Exp. Mar. Biol. Ecol. 57: 135-148. Keough, M. J. 1986. The distribution of the bryozoan on seagrass blades: settlement growth and mortality. Ecology 67: 846-857. Keough, M. J. 1989. Dispersal of the bryozoan Bugula neritina and effects of adults on newly settled juveniles. Mar. Ecol. Prog. Ser. (in press). Keough, M. J., and H. Chernoff. 1987. Dispersal and population vari- ation in the bryozoan. Ecology 68: 1 99-2 10. Koehn, R. K., R. J. E. Newell, and F. Immermann. 1980. Main- tenance of an aminopeptidase allele frequency cline by natural se- lection. Proc. Nat. Acad. Sci. USA 77: 5385-5389. Levin, L. A. 1984. Multiple patterns of development in Streblospio benedicti Webster (Spionidae) from three coasts of North America. Biol. Bull. 166:494-508. Levinton, J. S. 1983. The latitudinal compensation hypothesis: growth data and a model of latitudinal growth differentiation based 286 M. J. KEOUGH upon energy budgets. I. Interspecific comparison of Ophryotrocha (Polychaeta: Dorveillidae). Bwl Bull. 165: 686-698. Levinton, J. S., and R. K. Monahan. 1983. The latitudinal compensa- tion hypothesis: growth data and a model of latitudinal growth differentiation based upon energy budgets. II. Intraspecific compar- ison between subspecies of Opliryotroclia pncrili.i (Polychaeta: Dor- veillidae). Biol. Bull. 165: 699-707. Luckenbach, M. W. 1984. Sediment movements around animal tubes: the roles of hydrodynamics processes and biological activity. Limnol. Oceunogr. 31: 779-787. McKillup. S. C., and A. J. Butler. 1979. Modification of egg produc- tion and packaging in response to food availability by Nassarins pauperaius. Oecologia 43: 221-231. Nakauchi, M. 1982. Asexual development of ascidians: its biological significance, diversity, and morphogenesis. Am. Zoo/. 22: 753-763. I'.ilumlji, S. R., and J. B. C. Jackson. 1982. Ecology of cryptic coral reef communities. II. Recovery from small disturbance events by encrusting Bryozoa: the influence of "host" species and lesion size. J. Exp. Mar. Biol. Ecol. 64: 103-1 1 5. Peterson, C. H. 1982. The importance of predation and intra- and interspecific competition in the population biology of two infaunal suspension-feeding bivalves. Protothaca slam/nea and Chione 1111- Jatella. Ecol. Monogr. 52: 437-475. Reiswig, H. M. 1974. Water transport, respiration and energetics of three tropical marine sponges. J. Exp Mar Biol. Ecol. 14:231 -249. Russ, G. R. 1982. Overgrowth in a marine epifaunal community: competitive hierarchies and competitive networks. Oecologia 53: 12-19. Sebens, K. P. 1982. Competition for space: growth rate, reproductive output, and escape in size. Am. Nat. 120: 189-197. Sebens, K. P. 1983. The larval and juvenile ecology of the temperate octocoral Alcyonium siderium Verrill. II. Fecundity, survival, and juvenile growth. J. Exp. Mar. Biol. Ecol. 72: 263-285. Silander, J. A. 1985. Microevolution in clonal plants. Pp. 107-152 in Population Biology and Evolution of Clonal Organisms, J. B. C. Jackson, L. W. Buss and R. E. Cook, eds., Yale University Press, Connecticut. Sinervo, B.. and L. R. McEdward. 1988. Developmental conse- quences of an evolutionary change in egg size: an experimental test. Evolution 42: 885-899. Thorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. 25: 1-45. Travis, J. 1983. Variation in development patterns of larval anurans in temporary ponds. I. Persistent variation within a Hyla gratiosa population. Evolution 37: 496-5 1 2. Travis, J., S. B. Emerson, and M. S. Blouin. 1987. A quantitative- genetic analysis of larval life-history traits in Hyla crucifer. Evolu- tion 41: 145-156. Via, S. 1987. Genetic constraints on the evolution of phenotypic plas- ticity. Pp. 47-7 1 in Genetic Constraints on Adaptive Evolution, V. Loeschcke, ed.. Springer- Verlag, Berlin. Via, S., and R. Lande. 1985. Genotype-environment interaction and the evolution of phenotypic plasticity. Evolution 39: 505-522. VVahle, C. M. 1983. The role of age, size and injury in sexual repro- duction among Jamaican gorgonians. Am Zoo/. 23: 96 1 . Werner, P. A., and H. Caswell. 1977. Population growth rates and age versus stage-distribution models for teasel (Dipsacus sylveslris Huds.). Ecology 58: 1 103-1 111. VVethey, D. S. 1983. Intrapopulation variation in growth of sessile organisms: natural populations of the intertidal barnacle Ba/anus balamndes. Oikos40: 14-23. Winer.B.J. 1971. Statistical Principles in Experimental Design Mc- Graw-Hill, Kogakusha. Tokyo. Young, C. M., and F.-S. Chia. 1984. Microhabitat-associated vari- ability in survival and growth of subtidal solitary ascidians during the first 21 days after settlement. Mar. Biol. 81:61-68. Reference: Biol. Bull. 177: 287-294. (October. 1989) Population Genetics of the Common Squid Loligopealei LeSueur, 1821, from Cape Cod to Cape Hatteras RONALD L. GARTHWAITE1, CARL J. BERG JR.2. AND JUNE HARRIGAN3 Marine Biological Laboratory, Woods Hole, Massachusetts 02543 Abstract. Collections of Loligo pealei LeSueur, 1 82 1 from the Atlantic seaboard between Cape Hatteras and Georges Bank differ significantly in both allele and geno- type frequencies at the phosphoglucomutase locus. Lol- igo pealei collected off the coast of Virginia are distinct at this locus from all other areas surveyed. Loligo pealei collected from Georges Bank also differ significantly at this locus from those collected inshore off Cape Cod. These results suggest that L. pealei along the Atlantic sea- board consists of several distinct populations. A compar- ison of allele frequencies at nine allozyme loci among L. pealei, Loligo plei Blainville, 1 823, and Lolliguncula brevis (Blainville, 1823) reveals that L. pealei differs completely from L. plei at five loci, L. pealei from L. brevis at six loci, and L. plei from L. brevis at three loci. Introduction Data collected over the past one hundred years still present a confused picture of the stock4 characteristics of the common squid Loligo pealei LeSueur, 1821. The existence of two major discrete "groups" and a third "mixed group" has been demonstrated repeatedly in size frequency data from squid caught in the region extend- ing from Georges Bank to Cape Hatteras. This size struc- ture has been explained by invoking different sub-species (Verrill, 1882), brood stocks (Summers, 1971), or alter- Received 3 October 1988; accepted 31 July 1989. 1 Present address: Institute of Marine Sciences. University of Califor- nia. Santa Cruz. CA 95064. 2 Present address: Florida Department of Natural Resources, Marine Research Institute. 13365 Overseas Highway, Marathon. FL 33050. 5 Present address: Department of Ocean Engineering. University of Hawaii, Honolulu, HI 96822. 4 In this paper we use the term "stock" as defined by Ihssen ct ai (1981): "An intraspecific group of randomly mating individuals with temporal or spatial integrity." nate generations (Mesnil, 1977). In addition, there is some uncertainty as to how many species this taxon rep- resents throughout its range (Cohen, 1976; Summers, 1983). In reviewing the assumptions needed for statisti- cal analysis of population structure based on size fre- quency distributions. Summers ( 1 983) points out that L. pealei is not homogeneous either throughout its latitudi- nal range, across the continental shelf, through time at a single station, or even between successive tows. All of these data suggest that there are genetically discrete pop- ulations which are isolated by seasonal or geographic spawning differences. The single fishery "stock" or "pop- ulation" of L. pealei on the eastern seaboard (e.g., Lange and Sissenwine, 1980, 1983) probably consists of several stocks or populations. To date, biochemical genetic data have not been ap- plied to the problem of stock structure in Loligo pealei. In this study we used allozyme data to clarify the popula- tion or stock structure of this species along the northeast- ern coast of the United States. In the process, we also present a biochemical genetic comparison of L. pealei with the morphologically similar Loligo plei Blainville, 1823, and with Lolliguncula brevis (Blainville, 1823). Materials and Methods Squid were collected by trawl nets from research ves- sels of the Marine Biological Laboratory and the North- east Fisheries Center of the United States National Ma- rine Fisheries Service (NMFS). Collection data for the trawl stations are listed in Table I. The location of these stations is illustrated in Figure 1. For purposes of analy- sis, mainly to increase sample sizes, these trawl samples were grouped into six regions [North Carolina, Virginia, Delaware, Woods Hole, Cape Cod, and Georges Bank (Table I)], which we subsequently treated as six separate samples. With the exception of station 45, stations were 287 288 Collection data lor squid samples used in this study R. L. GARTHWAITE ET AL. Table I Region NMFS station Date Depth (m) Approximate #of Loligo pea/ei surveyed Other species present Latitude Longitude North Carolina 34 2 OCT 85 36-43 35°41' 75"06' 32 L plei 35 2OCT85 58-63 35°44' 74°56: 24 Virginia 45 3 OCT 85 16-15 36°28' 75°48' 16 53 4 OCT 85 25-24 37°25' 75°25' 54 L pic, 56 4 OCT 85 14 37°45' 75°27' 9 L brevis 57 4 OCT 85 19 37°54' 75" 11' 10 58 4 OCT 85 23-25 37°57' 75°02' 20 Delaware 83 6 OCT 85 192-197 38°32' 73° 16' 75 85 6 OCT 85 64-67 38°52' 73°21' 40 Woods Hole OCT 84 10-20 4I°30' 70°30' 110 MAY 85 10-20 41°30' 70°30' 120 OCT 85 10-20 41°30' 70°30' 100 Cape Cod 371 7 NOV 85 29-33 41*57' 69°56' 48 372 7 NOV 85 32-34 41°59' 69°57' 137 Georges Bank 198 2 1 OCT 85 99 40°42' 67° 15' 199 NMFS = National Marine Fisheries Service. See Figure 1 for station locations. grouped on the basis of physical proximity. Station 45, which lies approximately half way between the North Carolina stations and the other Virginia stations (Fig. 1 ), was grouped with the Virginia stations because it repre- sents a similar, near-shore, shallow-water habitat (Table I). The three samples collected off Woods Hole represent three different year classes (1983, 1984,and 1985). Squid collected in October of 1984 and 1985 were juveniles that hatched during the preceding summers, while the sample from May 1985 consisted of breeding adults most likely hatched in 1983. Soon after collection, a piece of mantle tissue was re- moved from each animal and frozen in liquid nitrogen for later electrophoretic analysis. The remainder of each squid was then frozen at -20°C. These specimens were later thawed and their mantle lengths measured to the nearest mm using a meter ruler. Tissue samples for electrophoresis were returned to the laboratory and stored at -70°C until analyzed. Before electrophoretic analysis, these tissue samples were soni- cated on ice in approximately equal weight to volume of 0.05 A/Tris-HCl pH 7.5 and centrifuged for 15 min in a clinical centrifuge at 2°C. Horizontal starch gel electro- phoresis (Sigma starch) using filter paper wicks (What- man #2) and buffer systems two and five of Selander et al. ( 1 97 1 ) was used to determine allele frequencies in the collections at 19 biochemical loci. Buffer system two was used to survey the collections for aminopeptidase (AP), leucine aminopeptidase- 1 (LAP-1), leucine aminopepti- dase-2 (LAP-2), malic enzyme (ME), phosphoglucomu- tase (PGM), phosphoglucose isomerase-1 (PGI-1 ), phos- phoglucose isomerase-2 (PGI-2), superoxide dismutase (SOD), and xanthine dehydrogenase (XDH). Buffer sys- tem five was used to survey the collections for «-glycero- phosphate dehydrogenase («-GPDH), alcohol dehydro- genase (ADH). esterase (EST), glutamate oxaloacetate transaminase-1 (GOT-1). glutamate oxaloacetate trans- aminase-2 (GOT-2), isocitrate dehydrogenase (IDH), malate dehydrogenase- 1 (MDH-1), malate dehydroge- nase-2 (MDH-2), nothing dehydrogenase (NDH), and sorbitol dehydrogenase- 1 (SDH-1). Initially, 40 speci- mens of Loligo pealei from Woods Hole were surveyed for all 19 loci. Nine of these loci (PGM, ME, MDH-1, MDH-2, IDH, GOT-1, GOT-2, PGI-1, and PGI-2) were chosen for use in the larger survey. Enzyme stain recipes were taken from Shaw and Prasad ( 1 970) and Ahmad et al. (1977) with minor modification. Differences in size frequency distributions among trawl samples and among individuals possessing various alleles within trawl samples were tested for significance using F-tests on variances and t-tests on means. Allele frequency differences among stations and regions were tested for significance using the tables of Mainland et al. ( 1956). Genotype frequency differences among stations and regions were tested for significance using G tests on contingency tables (Sokal and Rohlf, 1969). During our survey, we encountered one species of squid that was morphologically distinct from Loligo POPULATION GENETICS OF LOLIGO PEALE1 289 Figure 1. The location of the National Marine Fisheries Service trawl stations from which squid used in this study were collected. See Table I for collection data. WH = Woods Hole, MA. pealei and another that was electrophoretically distinct from L. pealei. We suspected that these squid were Lolli- guncula brevis and Loligo p/ei, respectively. Loligo plei is morphologically very similar to L. pealei, and the two species are not easily separated on morphological grounds (Cohen, 1976; Whitaker, 1980; Summers, 1983; Hanlon, 1988). Since the ranges of L. pealei and L. p/ei overlap in our study area (Whitaker, 1980), it is impor- tant that we demonstrate that we can consistently iden- tify L. p/ei electrophoretically and that we have not acci- dentally included specimens of this species in our study of L. pealei. To confirm the identities of these species in our collections we obtained known specimens of L. plei and L. brevis from Dr. Roger Hanlon of Galveston, Texas. These Texas specimens were compared directly, on the same gels, to our specimens of L. pealei, L. p/ei, and L. brevis for the nine loci listed above. Results Population genetics o/Loligo pealei Significant variation in squid size distributions existed among trawl samples both within and among regions. Out of a total of 55 pair-wise comparisons among the 1 1 trawl samples taken in October 1985, all but 5 were significantly different in variance or mean size (P < .05). Banding patterns for all loci surveyed conformed to Mendelian expectations (Harris and Hopkinson, 1976). Heterozygotes for PGM were found in both males and females, indicating that this locus is not sex linked. Al- leles, allele frequencies, and sample sizes for the collec- tions of Loligo pealei from all six regions are listed in Table II. Phosphoglucomutase genotype frequencies for all regions are given in Table III. On the whole, Loligo pealei possessed very low levels of genetic variation. Of the 19 loci surveyed, variant al- leles were found only in PGM, ME, MDH-2, IDH, GOT- 1, and PGI-2. However, of these six loci, the frequencies of the variant alleles were 1% or less in all but PGM for which the maximum frequency of variant alleles in any region surveyed was only 7.5% (Georges Bank, Table II). Combining the data over all collections, only 5% of all loci surveyed in L. pealei were polymorphic at the 1% level and the average heterozygosity per individual was only 0.6%. Given the low levels of genetic variation present in the collections of Loligo pealei surveyed, it is unlikely that differences in allele frequencies among the collections will be significant. Only PGM is polymorphic enough to reasonably give significant results with these sample sizes. No significant differences in allele or genotype fre- quencies were found among stations within regions for any locus. Among the three Woods Hole collections, which represent squid hatched in three different years (1983, 1984,and 1985) no significant differences in allele or genotype frequencies were found at any locus. Individuals possessing allelic variants generally did not differ significantly in size from other squid from the same trawl; the exceptions being squid possessing PGM allele 1.18 from Georges Bank, which had a significantly larger mean length than squid which did not possess this allele (P< .05), and squid possessing ME allele 1 .05 at Georges Bank and PGM allele 1.18 at Woods Hole (May 1985), which both had significantly larger variances than the re- maining squid in each sample (P < .05). 290 R. L. GARTHWAITE ET AL. Table II Allelc frequencies, and sample sizes in number ofalleles surveyed (in parentheses) for collections of Loligo pea\e\ from the northwest Atlantic. Alleles are expressed us migration rale relative to the most common allele Locus Allele North Carolina Virginia Delaware Woods Hole Woods Hole total Cape Cod Georges Bank OCT84 MAY 85 OCT 85 PGM (N) (112) (218) (230) (220) (240) (200) (660) (370) (398) 1.18 .054 — .030 .036 .021 .030 .029 .022 .030 1 .00 .928 .972 .948 .946 .962 .950 .953 .967 .925 .78 .018 .028 .022 .018 .017 .020 .018 .011 .045 ME (N) (112) (146) (164) (220) (240) (200) (660) (370) (398) 1.05 — — — — — — — — .008 1.00 1 .000 1.000 1.000 1.000 1 .000 1 .000 1.000 1.000 .992 MDH-2 (N) (112) (120) (92) (220) (240) (200) (660) (370) (398) 1.00 1.000 1.000 1.000 1.000 1.000 1.000 1.000 .997 1.000 .64 — — — — — — — .003 — IDH (N) (112) (100) (102) (220) (240) (200) (660) (370) (398) 1.17 — — — — .004 — .002 .005 — 1.00 1 .000 1 .000 1.000 1.000 .996 .990 .995 .992 1.000 .88 — — — — — .010 .003 — — .69 — — — — — — — .003 — GOT-1 (N) (112) (136) (150) (160) (240) (200) (600) (370) (398) 1.16 — — — — — — — — .003 1.00 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 .992 .79 — — — — — — — — .005 PGI-2 (N) (112) (126) (122) (220) (240) (200) (660) (370) (398) 1.00 1.000 1.000 1.000 1.000 .996 1.000 .998 1.000 1.000 .68 — — — — .004 — .002 — — MDH-I (N) (112) (120) (92) (220) (240) (200) (660) (370) (398) 1.00 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 GOT-2 (N) (112) (74) (150) (160) (240) (200) (600) (370) (398) 1.00 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 PGI-1 (N) (112) (126) (122) (220) (240) (200) (660) (370) (398) 1.00 1 .000 1 .000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 Among all pair-wise comparisons of squid collections (with the three Woods Hole collections combined and treated as a single sample), the Virginia sample stands out as being significantly different from the North Caro- lina sample (P < .05) and the Woods Hole sample (P < .05) in the frequency of PGM allele 1.18. In PGM ge- Table III Phosphoglucomutase genotype frequencies for collections of Loligo pea\ei from the northwest Atlantic Region Sample size PGM genotype 1.18/1.00 I.18/.78 1.00/1.00 1.00/.78 J8/.78 North Carolina 56 .107 .857 .036 _ Virginia 109 — — .945 .055 — Delaware 115 .061 .896 .043 — Woods Hole OCT 1984 110 .064 .009 .900 .027 — MAY 1985 120 .042 .925 .033 — OCT 1985 100 .060 .900 .040 — Total 330 .055 .003 .909 .033 — Cape Cod 185 .043 .935 .022 — Georges Bank 199 .060 — .855 .080 .005 POPULATION GENETICS OF LOLIGO PEALE1 291 notype frequencies, the Virginia sample is significantly different from the samples from all other regions (P < .05). In addition, the Georges Bank sample is signifi- cantly different from its nearest neighbor. Cape Cod, in PGM genotype frequencies (P < .05). All other pair-wise comparisons of regions for PGM and all other loci gave nonsignificant results for both allele and genotype fre- quencies. Taxonomy Alleles, allele frequencies, and sample sizes for the comparison of allele frequencies in Lo/igo pealei. Loligo plei. and Lolliguncula brevis are listed in Table IV. Our specimens of L. plei and L. brevis were electrophoreti- cally identical to the Texas specimens of these species for all loci surveyed. Sample sizes for Loligo plei and Lolliguncula brevis were small (eight individuals apiece) and alleles with fre- quencies lower than approximately 6% were not likely to have been detected. However, all loci surveyed, with the exception of IDH, PGI-2, and MDH-1, showed fixed al- lelic differences among species such that individuals of all three species can be identified easily and unambigu- ously solely on the basis of the alleles they possess at these diagnostic loci. Loligo pealei differs completely from L. plei at five loci, L. pealei from L. brevis at six loci, and L. plei from L. brevis at three loci (Table IV). Nei's genetic distances (Nei, 1972) for pair-wise comparisons of the three species are: .8 1 for L. pealei vs. L. plei, 1 . 10 for L. pealei vs. L. brevis. and .40 for L. plei vs. L. brevis. Discussion Generally, biochemical genetic data, of the kind re- ported here, are much more appropriate and effective in determining population or stock structure than are size frequency data (Ihssen et al, 198 1 ). Size and growth rate are both affected by numerous environmental factors such as temperature and food quality or availability (Hixon, 1983; Ihssen el al., 1981). Thus individuals from the same population or stock can differ significantly in these measures if they experience different environ- ments. This is particularly a problem with Loligo pealei, which appears to have a rather extended spawning and hatching period (Summers, 1971; Lange and Sissenwine, 1980, 1983). Biochemical genetic data are typically not directly affected by the environment and thus serve as a permanent and direct measure of genetic relatedness among collections or populations (Ihssen et al.. 1981; Avise, 1974; Ayala, 1983). This technique has, therefore, shifted from an optional to a primary position among methods used in studies of population or stock structure ( Ihssen et al.. 1981). The low level of genetic variation found here in Loligo pealei limits the effectiveness of this technique in deter- mining population structure in this species since only one locus (PGM) was polymorphic enough to detect differentiation among collections with reasonable sam- ple sizes. Nevertheless, for this locus, significant differ- entiation was detected among collections. The pattern- ing of this differentiation suggests that L. pealei in the northwest Atlantic is comprised of at least three popula- tions. Genetically, the Virginia sample is distinct from samples from all other regions included in this study and the Georges Bank sample is distinct from its nearest neighbor. Cape Cod. However, given that L. pealei are migratory and highly mobile (Lange and Sissenwine, 1983), it is unlikely that the spatial arrangement of these populations is constant. Previous studies on the population or stock structure of Loligo pealei. based on size frequency distributions, have postulated two groups of squid which may or may not be genetically isolated (Verrill, 1882; Summers, 1971;Mesnil, 1977; Lange and Sissenwine. 1980, 1983). We can find little genetic evidence for this structure. In no trawl sample was there an obvious bimodality in size frequency distributions. Nor did we find significant gene frequency differences among the three Woods Hole col- lections. While we found exceptional amounts of varia- tion among trawl samples in size frequency distributions, there is no overall correlation between mean size and PGM gene frequency (r = .02, P < .05). The fact that the Virginia trawl samples (which are genetically coher- ent and, as a whole, distinct from all other regions) are heterogeneous in size frequency (P < .05) indicates that size frequency is not a good indicator of population structure in L. pealei. It is possible that size frequency differences among samples are due to differences in envi- ronmental factors experienced by different groups of squid or differences in sex ratio among groups [male L. pealei tend to grow larger than females (Mesnil, 1977)]. Whether size frequency differences are due to environ- mental differences, sex ratio differences, or differences in spawning time, they do not seem to be associated with significant restrictions in gene flow. The differences in size distributions between individu- als possessing different alleles that was found in several trawl samples is interesting, but the small number of al- lelic variants obtained in any one sample makes analysis difficult. These differences in size distributions may be a result of selective forces acting on alleles or may indicate that our trawl samples were composed of a mixture of squid populations which differ in both size and allele fre- quency. In this latter case, this data would be evidence for the population structure postulated by previous workers (Verrill, 1882; Summers, 1971; Mesnil, 1977). The degree of interpopulational genetic differentiation reported here is almost certainly a minimum value for 292 R. L. GARTHWAITE ET AL. Table IV Allele frequencies at nine loci in three species of squid Locus Allele Loligo pealei (1704) Loligo plei (16) Lolliguncula brevis (16) PGM 1.18 .026 _ 1.00 .950 — — .78 .024 — — .60 — 1.000 1.000 ME 1.05 .002 — — 1.00 .998 — — .95 — 1.000 1.000 MDH-2 1.00 .999 1.000 — .90 — — 1.000 .64 .001 — — IDH 1.17 .002 — — 1.00 .996 1.000 1.000 .88 .001 — — .69 .001 — — GOT-1 1.16 .001 — — 1.00 .998 — — .88 — — 1.000 .85 — 1.000 — .79 .001 — — PGI-2 1.00 .999 1.000 1.000 .68 .001 — — MDH-1 1.00 1.000 1.000 1 .000 GOT-2 1.00 1.000 — — .86 — — 1.000 * — 1.000 — PGI-1 1.00 1.000 — — .74 — 1 .000 1.000 Minimum sample sizes (in number of alleles surveyed) are in paren- theses. Alleles are expressed as migration rate relative to the most com- mon Loligo pealei al\e\e. * = no activity. Loligo pealei. All of the collections of L. pealei surveyed in this study came from a limited biogeographic range extending north from Cape Hatteras to Cape Cod. The range of L. pealei extends south as far as the Gulf of Ven- ezuela (Cohen, 1976; Summer, 1983) and it may be ex- pected that comparisons of populations north of Cape Hatteras with those south of Cape Hatteras or from the Gulf of Mexico may reveal additional genetic differenti- ation since Cape Hatteras and Florida form natural boundaries for many other species (Briggs, 1974). The low levels of intrapopulational genetic variation in Loligo pealei found in this study seem to be a general characteristic of squid. Similar low levels of genetic varia- tion have been reported for Loligo opalescens Berry, 1911 (Christofferson et a/., 1978; Augustyn and Grant, 1988), Loligo vulgaris Lamarck, 1798 (Augustyn and Grant, 1988), and Illex illecebrosus (LeSueur, 1821) (Romero and Amaratunga, 1981). In addition, even though our data are limited both in number of loci (9) and number of individuals (8) surveyed, Loligo plei and Lolliguncula brevis also appear to fit this pattern since allelic variants for the loci surveyed here, if present at all, would likely be in frequencies of 6% or less. Table V presents a comparison of measures of genetic variation based on data taken from the literature for several squid species and for invertebrates in general. While the lack of genetic variation in squid may limit the use of this kind of data in studies of population struc- ture, it makes it that much more useful in taxonomic studies since banding patterns are simpler and intraspe- cific variation is minimized. In this study, we found bio- chemical data to be an excellent taxonomic tool. While Loligo pealei and Loligo plei are morphologically very similar and difficult to differentiate on morphological grounds (Cohen, 1976; Whitaker, 1980; Summer, 1983; Hanlon, 1988), they are quite distinct biochemically (differing completely at five out of nine loci) and easily distinguishable from one another. In addition, both spe- cies are easily distinguishable electrophoretically from Lolliguncula brevis. That L. plei appears to be more closely related to L. brevis than L. pealei is unexpected and interesting but this result may be an artifact of the small number of loci surveyed. Several other studies of squid population structure and taxonomy have been performed using biochemical ge- netic data. Two of these studies (Ally and Keck, 1978; Christofferson et at., 1978) are concerned with Loligo opalescens along the California coast, and both suggest a separate southern population on the basis of biochemical data. This population structure for L. opalescens is sub- stantiated by data on spawning peaks (Fields, 1965) and morphological indices (Kashiwada and Recksiek, 1978). A third study concerns Illex illecebrosus off the eastern coast of Canada (Romero and Amaratunga, 1981). In this study, no significant differences were detected among collections, but this is not surprising considering the limited geographical range over which the collections were taken and the fact that sample sizes were small. Smith et at. (1981) studied Nototodarus sloani Gray, 1849, in New Zealand. On the basis of mainly electro- phoretic evidence. Smith et al. found that what was pre- viously thought to be eight sub-populations of A', sloani instead consisted of two species that are largely non-over- lapping in their distributions. This conclusion concern- ing N. sloani was substantiated by data on morphology and parasite load. Finally, Augustyn and Grant (1988), in their morphological and biochemical study of African squid, discovered that what had previously been consid- ered two separate species on morphological grounds (Loligo vulgaris and Loligo reynaudii d'Orbigny, 1845) actually consisted of two subspecies of L. vulgaris as demonstrated by biochemical data. POPULATION GENETICS OF LOL1GO PEALEI Table V 293 Genetic variability in squid #of % Individuals Polymorphic # of Loci surveyed loci Average Average surveyed per locus (1% level) Ho Ne Reference Illex illecebrosus 11 10-156 9 .005 1.01 Romero and Amaratunga. 1981 Loligo opalescens 30 45 17 .037 1.06 Augustyn and Grant. 1988 Loligo pealei 19 40-994 5 .006 1.01 This paper Loligo plei 9 8 0 .000 1.00 This paper Loligo vulgaris 30 44 23 .030 1.05 Augustyn and Grant, reynaudii 1988 Loligo vulgaris 30 15 7 .011 1.01 Augustyn and Grant. vulgaris 1988 Lollignncu/a brevis 9 8 0 .000 1.00 This paper Average for other 38 .100 1.11 Nevoe/a/.. 1984 invertebrates Values for Loligo pealei and Illex illecebmsus are for combined data over several collections. Ho = observed heterozygosity, Ne = effective number of alleles. Thus, on the whole, biochemical genetic studies have proven to be very useful in the study of squid stock and population structure, furnishing data on the number of species present (Smith el a/.. 1981: Augustyn and Grant, 1988) and the spatial distribution of breeding units within these species (Ally and Keck, 1 978; Christofferson el ai, 1978; this paper). Obviously, this sort of informa- tion should be of central importance to fisheries manage- ment because, to use a fishery stock effectively, the num- ber of species present, their spatial distributions, and their population structure must be considered. Acknowledgments This project was made possible through a collabora- tion of scientists and staff of the United States National Marine Fisheries Service, Northeast Fisheries Center and the Marine Biological Laboratory, both in Woods Hole, Massachusetts. We thank Y. P. Wang for technical assis- tance. Dr. R. T. Hanlon kindly supplied specimens through DHHS Grant RR01024. This project was sup- ported in part by a grant from the Charles Ulrick and Josephine Bay Foundation to Dr. Garthwaite. Literature Cited Ahmad, M., D. O. F. Skibinski, and J. A. Beardmore. 1977. An esti- mate of the amount of genetic variation in the common mussel Mytilusedulis. Bwchem. Genet. 15: 833-846. Ally, J. R. R., and S. A. Keck. 1978. A biochemical-genetic popula- tion structure study of market squid, Loligo opalescens, along the California coast. Calif. Depl. Fish Game Fish Bull. 169: 1 1 3- 1 2 1 . Augustyn, C. J., and W.S. Grant. 1988. Biochemical and morpholog- ical systematics of Loligo vulgaris vulgaris Lamarck and Loligo vul- garis reynaudii D'Orbigny NOV. COMB. (Cephalopoda: Myop- sida). Malacologia 29: 2 1 5-233. Avise, J. C. 1974. Systematic value of electrophoretic data. Sysl. Zool. 23: 465-48 1 . Ayala, F.J. 1983. Enzymes as taxonomic characters. Pp. 3-26inPro- lein Polymorphism Adaptive and Taxonomic Significance. G. S. Oxford and D. Rollinson, eds. Academic Press, New York. Berry, S. 1911. Notes on some cephalopods in the collection of the University ofCalifornia. Univ. Calif Publ. Zool. 8: 301-310. Blainville, H. 1823. Memoiresur lesespecesdu genre calmar(Lo//go. Lamarck)./. Physique Chimie d'Histi >ire Naturelle 96: 1 16-135. Briggs, J. C. 1974. Marine Zoogeography. McGraw Hill, New York. 475 pp. Christofferson, J. P., A. Foss, W. E. Lambert, and B. Welge. 1978. An electrophoretic study of select proteins from the market squid, Loligo opalescens Berry. Calif. Depl Fish Game Fish Bull 169: 123-133. Cohen, A. C. 1976. The systematics and distribution of Loligo (Ceph- alopoda, Myopsida) in the western North Atlantic, with descrip- tions of two new species. Malacologia 15: 299-367. Fields, W. G. 1965. The structure, development, food relations, re- production, and life history of the squid, Loligo opalescens Berry. Calif. Depl. Fish Game Fish Bull. 131: 1-108. Gray, J. E. 1849. Catalogue of the mollusca in the collection of the British Museum. Pt. 1. Cephalopoda Antepedia. British Museum, London. Hanlon, R. T. 1988. Behavioral and body patterning characters useful in taxonomy and field identification of cephalopods. Malacologia 29: 247-264. Harris, H., and D. A. Hopkinson. 1976. Handbook ofEn:yme Elec- trophoresis in Human Genetics. American Elsevier, New York. Hixon, R. F. 1983. Loligo opalescens. Pp. 95-1 \4inCephalopodLife Cycles. Vol. 1, P. R. Boyle, ed. Academic Press, New York. Ihssen, P. E., H. E. Booke, J. M. Casselman, J. M. McGlade, N. R. Payne, and F. M. Utter. 1981. Stock identification: materials and methods. Can J Fish. Aauat. Sci. 38: 1838-1855. Kashiwada, J., and C. W. Recksiek. 1978. Possible morphological 294 R. L. GARTHWAITE ET AL. indicators of population structure in the market squid. Loligo opalescens. Calif. Depl. Fish Game Fish Bull 169: 99- 1 1 1 . Lamarck, C. 1 798. Extrait d'un memoire sur le genre de la Seche. du Calmaretdu Poulpe. vulgairement nommes, polypes de mer. Bull Sociele Philomatique Paris 2: 129-131. Lange, A. M. T., and M. P. Sissenwine. 1980. Biological considera- tions relevant to the management of squid (Loligo pealei and Illc\ illecebrosus) ofthe northwest Atlantic. Mar. Fish. Rev. 42(7-8): 23- 38. Lange, A. M. T., and M. P. Sissenwine. 1983. Squid resources of the northwest Atlantic. FAO Fish, Tech. Pap 231: 21-54. LeSueur, C. A. 1821. Descriptions of several new species of cuttle- fish. / Acad. Nat. Sci. Philadelphia 2: 86-101. Mainland, D., L. Herrera, and M. I. Sutcliffe. 1956. Statistical Tables for use with Binomial Samples — Contingency Tests. Confidence Limits, and Sample Size Estimates. Department of Medical Statis- tics, New York University College of Medicine, New York. 83 pp. Mesnil, B. 1977. Growth and life cycle of squid, Loligo pealei and Ille\ illecebrosus. from the northwest Atlantic. ICNAF Res Doc. 76/VI/65: 55-69. Nei, M. 1972. Genetic distance between populations. Am. Nat. 106: 283-292. Nevo, E., A. Beiles, and R. Ben-Shlomo. 1984. The evolutionary sig- nificance of genetic diversity: ecological, demographic and life his- tory correlates. Pp. 13-213 in Evolutionary Dynamics of Genetic Diversity, G. S. Mani, ed. Springer- Verlag, New York. d'Orbigny, A. 1845. Mollusques vivants et fossiles. I. Paris. Romero, M. C. L., and T. Amaratunga. 1981. Preliminary results of biochemical-genetic population structure study of the squid lllex illecebrosus. Third annual meeting of the Northwest Atlantic Fish- eries Organization. NAFOSCR Doc. 81/IX/103. Serial No. N405. Selander, R. K., M. H. Smith, S. Y. Yang, W. E. Johnson, and J. B. Gentry. 1971. IV. Biochemical polymorphism and systematics in the genus Peromyscits. I. Variation in the old-field mouse (Peromys- cus polionotus). Univ. Texas Puhl. 7103:49-90. Shaw, C. R., and R. Prasad. 1970. Starch gel electrophoresis of en- zymes— a compilation of recipes. Biochem. Genet. 4: 297-320. Smith, P. J., P. E. Roberts, and R. J. Hurst. 1981. Evidence for two species of arrow squid in the New Zealand fishery. New Zealand J. Mar Freshwater Res. 15: 247-253. Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman & Co., San Francisco. 776 pp. Summers, W. C. 1971. Age and growth of Loligo pealei, a population study of the common Atlantic Coast squid. Biol. Bull. 141: 189- 201." Summers, W. C. 1983. Loligo pealei. Pp. 1 1 5- 1 42 in Cephalopod life Cycles. Vol. 1 , P. R. Boyle, ed. Academic Press, New York. Verrill, A. E. 1882. II. Report on the cephalopods of northeastern coast of America. Pp. 211-455 in Report to the Commissioner. 1879. Part VII. U. S. Commission of Fish and Fisheries. VV hitaker, J. D. 1980. Squid catches resulting from trawl surveys off the southeastern United States. Mar. Fish. Rev 42(7-8): 39-43. Reference: Biol. Bull 177: 295-302. (October. 1989) Factors Controlling Attachment of Bryozoan Larvae: A Comparison of Bacterial Films and Unfilmed Surfaces J. S. MAKI1, D. RITTSCHOF2, A. R. SCHMIDT2, A. G. SNYDER1, AND R. MITCHELL1 ^Laboratory of Microbial Ecology, Division oj Applied Sciences. Harvard University, Cambridge. Massachusetts 02138. and2 Duke University Marine Laboratory, Fivers Island. Beaufort, North Carolina 28516 Abstract. The effects of individual species of marine bacteria on the attachment of larvae of the bryozoan Bu- gula neritina were examined in the laboratory. Bacteria, grown to mid-exponential phase and allowed to adsorb to polystyrene petri dishes, attached at densities of 106- 107 cells cirT2. Bryozoan attachment assays (30 min) were used to compare the effects of adsorbed cells of three species of bacteria with unnlmed surfaces. Larvae permanently attached, at high percentages (65-94%), to unnlmed polystyrene, hydrophobic (i.e., low wettability, low surface energy) control surfaces. This activity agrees with reports in the literature. Films of individual species of bacteria can influence bryozoan attachment. Three separate strains of the bacterial species Deleya marina inhibited attachment, but two other species of marine bacteria did not. Measurements indicated that all five bacteria tested differed in their cell-surface hydrophobic- ity, but that their films were similar in that they were all highly wettable (i.e., high surface free energy). Our data indicate that factors in addition to substratum surface energy determine attachment of bryozoan larvae espe- cially when bacterial films are present. Bacterial extracel- lular materials may be involved. Introduction Bryozoans are economically important marine fouling organisms (Ryland, 1976; Soule and Soule, 1977; Wool- lacott, 1984). The substratum preferences exhibited by their larvae range from the very specific to the very gen- eral (Ryland, 1974). Most bryozoans require a substra- tum that provides firm support for attachment, and many also prefer surfaces that have a smooth or glossy Received 16 September 1988; accepted 31 July 1989. finish (Ryland, 1974, 1976). One critical characteristic in larval preference in settlement appears to be the surface free energy of the substratum (Mihm el ai. 1981; Wool- lacott, 1984). A surface with a low surface free energy has a low wettability and is hydrophobic, while a surface with a high surface free energy has a high wettability and is hydrophilic. Many bryozoan species prefer to settle and metamorphose on low surface free energy, hydrophobic substrata (Eiben, 1976; Loeb, 1977; Mihm et ai. 1981; RittschofandCostlow, 1987a. 1987b, 1988). Although a bacterial film is not essential for larval at- tachment (Ryland, 1976), microbial films on substrata influence the attachment of larvae from a number of different species of bryozoa (Miller et ai. 1948; Wisely, 1958; Crisp and Ryland, 1960; Ryland, 1974; Mihm et al, 1981; Brancato and Woollacott, 1982). Mihm et al. ( 198 1 ) demonstrated that bacterial films on polystyrene (a low surface energy substratum) caused the surface to become more wettable (higher surface free energy) and also decreased the attachment of larvae of the bryozoan Bugnla neritina. However, Mihm et al. (1981) pointed out that the reduction in larval settlement could not be attributed to the change in surface free energy alone and suggested that Bugnla larvae respond to two sensory stimuli: one for surface free energy, the second to some aspect of the bacterial-organic film. The latter stimulus could override the response to surface free energy alone (Mihm et al.. 1981). The response of bryozoan larvae to natural products that inhibit larval attachment may function by the same mechanism (Rittschof et ai. 1988). It has been demonstrated that films, each composed of individual species of bacteria, can cause different attach- ment responses by spirorbid polychaete larvae (Kirch- man et al., 1982), barnacle cypris larvae (Maki et al., 1988), and macroalgal swarmers (Thomas and Allsopp, 295 296 J. S. MAKI ET AL. 1983). Perhaps the species composition of the bacterial films may also influence attachment by bryozoan larvae. We examined the following questions: ( 1 ) could films of individual species of bacteria elicit different attachment responses from bryozoan larvae? (2) would aging the bac- terial films change the larval attachment response? (3) could larval attachment be correlated with surface free energy measurements using either bacterial cell-surface hydrophobicity or bacterial film wettability? We report here the results of laboratory experiments in which we tested the attachment responses of Bugula neritina lar- vae to bacteria irreversibly attached (Marshall el al., 1971) to polystyrene surfaces. Our data indicate that the attachment of bryozoan larvae varied for both films of individual bacteria and films of different ages, and that larval responses were not correlated with measurements of either the bacterial cell-surface hydrophobicity or the bacterial film wettability. Materials and Methods All seawater used in larval attachment experiments was passed through a septic 100,000 Dalton ultrafiltra- tion system (Millipore) and subsequently passed through two sterile filters (0.2 ^m pore size, Millipore), placed one on top of the other. This seawater is referred to as filtered seawater (FSW). Bacteria Five cultures of marine bacteria were used in the larval attachment experiments. Four of the pure cultures were obtained from the American Type Culture Collection, Rockville, Maryland: Deleya marina 25374, D. marina 27129, D. marina 35142, and Vibrio vnlnifiats 27562. The fifth bacterium was isolate DLS1, a gram negative, polarly flagellated, oxidase positive, fermentative, rod- shaped bacterium that was isolated from the estuarine waters around the Duke University Marine Laboratory using Marine Agar 2216 (Difco, Detroit, Michigan). To preserve the bacteria and their respective surface charac- teristics, cultures upon receipt or isolation were frozen in vials with glycerol from which new stock cultures were periodically established. Preparation of dishes with attached bacteria Bacteria were allowed to attach to polystyrene petri dishes (Falcon 1006, 50 X 9 mm) following the methods outlined by Fletcher ( 1977). Cultures were grown to ex- ponential phase at 26°C in Marine Broth 2216 (Difco, Detroit, Michigan) and harvested by centrifugation. Bac- teria were washed, centrifuged, and resuspended in FSW ( 1 09 cells • mr ' ). Petri dishes were filled with 7 ml of the bacterial suspension and incubated at 26°C for 2.5-3.0 h. Dishes were rinsed by dipping them 10X in 500 ml of FSW. Bacteria still attached to the dishes were consid- ered irreversibly attached (Marshall el al., 1971). The dishes were then filled with 5 ml of FSW. Experiments were performed using either these dishes (for conve- nience termed Day 1 ) or with dishes in which the bacteria were aged in situ for 1 to 5 days. Aging of the attached bacteria in the dishes was accomplished through the fol- lowing manner: every day after filling the Day 1 dish with FSW. it was emptied, and fresh FSW was added. Aged dishes were rinsed as above and refilled with 5 ml of FSW immediately prior to attachment assays. Following the larval attachment experiments, two dishes of each bacterial treatment and control were fixed with formaldehyde (final concentration 1 to 2%, v/v) for quantification of attached bacteria using acridine orange and epifluorescence microscopy (Daley and Hobbie, 1975). These dishes did not receive any larvae. At least 300 bacteria were counted per dish and the number of cells expressed as bacteria per cnr. Aged films were streaked on Marine Agar 2216 to check for contamina- tion. Larvae and attachment experiments The larval bryozoan attachment assay was that pre- viously described by Rittschof et al. (1988). Bugula neri- tina colonies were collected from the Duke University Marine Laboratory floating dock and from pilings of the north end of the Atlantic Beach, North Carolina, bridge. In the laboratory, colonies were maintained at 25° ± 3°C in the dark in aerated seawater and fed cultured diatoms (Skeletonema cost at um Greville) at 100,000 cells ml ' day"'. B. neritina larvae were released in the morning in response to exposure to artificial light. Larvae (20-80) were collected in a 250-450 ^1 volume of seawater and pipetted into one of two dishes of each treatment (i.e., bacterial films and controls) that already contained 5 ml of FSW. Repeated transfers of larvae resulted in a maxi- mum of 160 larvae in any one dish. Assays were for 30 min at 22° ± 2°C. Timing of the assay began with the final larval addition and was terminated by the addition of a drop of formalin. Larvae adhering to the substratum and having no visible cilia (due to involution of the co- rona during metamorphosis, Zimmer and Woollacott, 1977; Woollacott and Zimmer, 1978) were counted as attached, while those that either were not adherent or had visible cilia were counted as not attached (90X mag- nification). Colonies used as a source of larvae were re- placed when larval attachment to polystyrene fell below 50% in 30 min. Experiments examined the effect of axenic films of bacteria and bacterial film age on the attachment of lar- vae. The first set of experiments used bacterial films of BRYOZOAN LARVAE AND BACTERIAL FILMS 297 different ages composed of either Deleya marina (ATCC 25374), Vibrio vulnificus (ATCC 27562), or isolate DLS 1 . The second set of experiments used bacterial films composed of either D. marina ATCC 25374, ATCC 27 129, or ATCC 35 142. Controls were FSW with no ad- ditions (polystyrene). Frequencies of attached and not-attached individuals pooled from the two dishes were compared between treatments by G statistic (corrected for continuity) gener- ated from a contingency analysis (Zar, 1984). The null hypothesis for the contingency analysis was that there was no significant difference between the control (poly- styrene) and any one treatment. The family and individ- ual level of significance in each group of comparisons was determined using Bonferroni's method for multiple comparisons (Seber, 1977). Surface free energy measurements: bacterial cell-surface hydrophobicity and film wettability Because the surface free energy has been shown to be important in the attachment of bryozoan larvae to a sub- stratum, we determined both the cell-surface hydropho- bicity of the bacteria in solution, using the adhesion to hydrocarbon method (Rosenberg el a/.. 1980) and the wettability of films of attached bacteria derived from the same cultures, using measurements of air bubble contact angles (Fletcher and Marshall, 1982; Dillon etal., 1989). Bacteria were grown to mid-exponential phase in Marine Broth 2216 (Difco, Detroit, Michigan) and harvested by centrifugation. Bacteria were washed once and resus- pended in FSW or an artificial seawater. Nine Salt So- lution (NSS) (Little et al, 1986) to approximately 10g cells mr'. To measure cell-surface hydrophobicity by adhesion to hydrocarbons (Rosenberg et al.. 1980), hexadecane (0.08, 0. 16, 0.32, and 0.64 ml) was added to triplicate test tubes containing 4 ml of the bacterial solution (A400 = 1.3-1.5) and vortexed for 2 min. The phases were al- lowed to separate for 15 min and the absorbance (A400) of the aqueous phase was measured spectrophotometri- cally. The results of the adhesion to hexadecane experi- ments are presented as the percent absorbance ( A400 ) left in the aqueous phase (bacteria with a high surface free energy and a hydrophilic cell surface would have a value of 100%). Air bubble contact angle determinations (Fletcher and Marshall, 1982; Dillon et al, 1989) were used as a mea- sure of the wettability of unfilmed and filmed surfaces. Coupons (approximately 1 cm X 2 cm) of the polysty- rene petri dishes (Falcon 1006) were placed in larger petri dishes (100 X 15 mm, Falcon 1029) and bacteria were allowed to attach to the coupons as for the petri dishes above. After attachment, coupons were retrieved with sterile forceps and rinsed as above and placed in another large petri dish containing FSW or NSS. For bubble con- tact angle measurements, the coupons were placed in a stage at the top of a chamber containing FSW or NSS. An air bubble, injected from a syringe (0.25 mm ID), was allowed to rise 6-7 mm to rest against the test surface. The average diameter of the bubble was 2.0 mm. Contact angles were measured directly using a Vernier micro- scope with a goniometer eyepiece. Results represent the mean of at least ten observations. For air bubbles where the air came in contact with the surface, errors were within 2° unless recorded otherwise; for air bubbles that did not make contact with the surface, indicating a high surface free energy, a value of <15° was recorded (Fletcher and Marshall, 1982; Dillon et al., 1989). Cou- pons were then fixed with formaldehyde for quantifica- tion of attached bacteria as above. Comparisons of air bubble contact angle measurements on bacterial films were made to parallel measurements on muffled glass (500°C for 4 h) and polystyrene. Bugula neritina larvae have known attachment responses to these last two sur- faces using the above attachment assay (Rittschof and Costlow, 1987a, b; 1988). Results Experiments were designed to examine the attach- ment of Bugula neritina larvae to bacterial films in the laboratory and to examine the larval attachment in rela- tionship to estimates of surface free energies. These fac- tors were hypothesized to be involved in the larval at- tachment response. Bacterial densities Bacteria adhered in densities of 10"- 107 attached bac- teria per cm2 both to the polystyrene petri dishes (Tables I, II) and to polystyrene coupons (Table III). Films with lower densities of bacteria ( 106 per cm2) were not con- fluent but visually appeared randomly distributed rather than patchily. Films with higher densities of bacteria were confluent. Attached bacteria were undetected on the control polystyrene dishes indicating that the filtra- tion of the seawater was effective in removing bacteria. The densities of attached cells on the aged dishes were lower than on Day 1 dishes, suggesting that some bacte- ria may have desorbed from the surface or lysed. Exami- nation of the agar plates inoculated with bacteria from the aged dishes revealed only one colony type. Larval attachment The percentage of Bugula neritina larvae that attached to polystyrene control dishes in 30 min ranged between 66% and 93% (e.g.. Tables I, II). Bacterial films com- 298 J. S. MAKJ ET AL Table I Bugula nentina larval attachment: data from experiments using films of different ages composed of three different species of bacteria, Deleya marina. Vibrio vulnificus, and isolate DLS1 attached to polystyrene petri dishes Treatment No. of bacteria" (xl07)cirT:(+SD) Total no. of larvae h % larvae attached G statistic' vs polystyrene GNo. P Polystyrene nd 73 91.8 D. marina Day 1 5.10(0.18) 54 7.4 99.46 <0.001 Day 2 3.52(0.30) 55 40.0 39.12 <0.001 Day 4 0.40 (0.06) 95 60.0 21.91 <0.001 I', vulnijicits Day 1 4.59(0.49) 67 79.1 3.65 NS Day 2 2.60(0.24) 263 97.7 3.58 NS Day 4 1.47(0.14) 179 97.2 2.25 NS Isolate DLS1 Day 1 1.73(0.28) 74 81.1 2.78 NS Day 2 0.86(0.04) 94 87.2 0.48 NS Day 4 0.53(0.04) 140 96.4 0.37 NS J Mean number of attached bacteria cm 2 from counts of two dishes using epifluorescence microscopy after staining with 0.0 1 % (final concentra- tion, w/v)acridine orange. Bacteria were grown to mid-exponential phase in Marine Broth 2216(Difco, Detroit, MI) at 26°C, harvested by centnfu- gation, washed, and resuspended to 109 cells ml '. Dishes were exposed to bacterial solution for 2.5 to 3.0 h before being rinsed and used for experiments. Day 1 dishes were prepared the same day as the experiment, while Day 2 and 4 dishes were prepared 2 and 4 days prior to the experiment, respectively. FSW in these dishes was replaced daily after their preparation, nd = none detected. h Total number of larvae in two dishes. c Using Bonferroni's method of multiple comparisons, the family level of significance in the experiment was a = 0.05 with an individual signifi- cance level of a/9 = 0.0056 with 1 df where 9 is the number of comparisons. NS = not significant. posed of either Deleya marina, Mbrio vulnificus. or iso- late DLS1, and aged in situ for different lengths of time were tested for their effect upon attachment of B. neritina larvae. Films of D. marina of all ages significantly inhib- ited larval attachment compared to the polystyrene con- trols (Table I). Films of all ages composed of V. vulnificus and isolate DLS1 did not significantly inhibit larval at- tachment when compared to polystyrene controls (Table I). Similarly, larval attachment to films composed of D. marina was significantly lower than attachment to films composed of the other two bacteria on all days (38.427 < G < 1 10.676, P < 0.001, 1 df). Larval attachment on films composed of I '. vulnificus or isolate DLS 1 was only different on films aged for 2 days (G = 1 1 .826, P < 0.00 1 , 1 df). For all three bacteria, the larval attachment to Day 4 films was significantly higher than attachment to Day 1 films (P< 0.00 1,G statistic with 1 df). In an experiment using films aged up to 5 days, larval attachment on D. marina films was again significantly inhibited when compared to the polystyrene control (control = 93.0% out of 2 14 total larvae attached, D. ma- rina = 3.5-10.3% out of 58-86 total larvae attached, P < 0.001, G statistic with 1 df, family level of significance of 0.05, individual level of significance of 0.0056). Films of V. vulnificus (larval attachment = 82-98% of 70-108 total larvae) were not inhibitory (P > 0.05, G statistic with 1 df). Films composed of isolate DLS1 aged 3 days inhibited larval attachment (45.8% out of 59 total larvae attached, P < 0.00 1 , G statistic with 1 df ). Larval attach- ment on all other DLS1 films was similar to attachment to the control. Comparisons of larval attachment be- tween films of different bacteria showed that attachment to films composed of/), marina was lower than on films of the other two bacteria on all days (28.514 < G < 1 7 1 .368, P < 0.00 1 , 1 df ) and that attachment on iso- late DLS1 was only different from that on V. vulnificus on films aged 3 days (G = 58.393, P< 0.001, 1 df). Lar- val attachment to older and younger films of D. marina was not significantly different (P > 0.05) while attach- ment to older and younger films of V. vulnificus and DLS1 were (P < 0.005 and P < 0.024, respectively, G statistic with 1 df). The data indicate that films of indi- vidual species of bacteria can affect the attachment of B. neritina larvae. Although there were statistical differ- ences in larval attachment to older and younger films of the same species (with one exception, 3-day-old films of DLS1), in general, films of inhibitory species remained inhibitory and films of non-inhibitory species remained non-inhibitory compared to the controls. Experiments were conducted with three separate cultures of bacteria, all classified as Deleya marina, to determine if phenotyp- ically similar bacteria could elicit different attachment responses from bryozoan larvae. All three cultures of D. marina significantly inhibited larval attachment com- BRYOZOAN LARVAE AND BACTERIAL FILMS Table II Bugula neritina larval attachment: data from experiments using films of three strains oj the marine bacterium. Deleya marina, attached to polystyrene petri dishes 299 Treatment No. of bacteria" (X107)cm :(+SD) Total no. of larvae h larvae attached G statistic1 vs polystyrene GNo. P A. Polystyrene nd 83 84.3 D. marina ATCC 25374 2.86(0.43) 66 4.5 105.42 0.05, 1 df). Surface free energy measurements: bacterial cell-surface hydrophobicity and wettability of films The test to determine cell-surface hydrophobicity by adhesion to hexadecane showed that the three cultures of D. marina in solution were more hydrophobic (i.e.. had a lower surface free energy) than the other two bacte- ria (Fig. 1). However, air bubble contact angle measure- ments on films of the attached bacteria were all <15° even after aging the bacterial films. The air bubble did not come in contact with the surface and all films had a high surface free energy (Table III). Air bubble contact angles on polystyrene controls were 90° (low surface free energy) while those on glass baked at 500°C were also < 15°. Discussion Bryozoan larvae have well-developed mechanisms for determining a suitable substratum, and these may be species specific (Ryland, 1976; Woollacott, 1984). The process begins with larvae gliding or crawling on a sur- face and testing it with cilia (Woollacott and Zimmer, 1978), and is often followed by a temporary attachment that employs an acid mucopolysaccharide adhesive (Loeb and Walker, 1977). Permanent attachment in- volves the eversion of the metasomal sac (Woollacott and Zimmer, 1978), which releases proteins that to- gether with acid mucopolysaccharide provide the perma- nent adhesive (Loeb and Walker, 1977). Previous investigations have illustrated that natural films of microorganisms can inhibit (Crisp and Ryland, 1960;Mihme/a/.. 1 98 1) or facilitate (Miller et a/., 1948; Wisely, 1958; Ryland, 1974;Mihmtfa/., 1981;Brancato and Woollacott, 1982) the attachment of bryozoan lar- vae. Mihm et al. (1981) demonstrated that the presence of microbial films could make an unattractive substra- tum (e.g.. glass) attractive, and an attractive substratum (e.g., polystyrene) unattractive. Our data demonstrate that, on a suitable polystyrene substratum, films com- posed of some bacteria significantly inhibit attachment of Bugula neritina larvae when compared to unfilmed controls. Other bacteria did not inhibit larval attach- ment. These data suggest that the species composition of a film may be important in the larval attachment re- sponse. Larval attachment to aged films of bacteria was generally higher than to freshly prepared films. The aging of the films resulted in a decrease in the density of at- tached bacteria (Table I) suggesting that bacterial density may be one important factor in the larval response to the film. However, in our experiments films composed of bacteria that were inhibitory to larval attachment re- mained inhibitory regardless of the age of the film. In contrast, films composed of bacteria that did not inhibit larval attachment remained uninhibatory in comparison with unfilmed polystyrene controls (Table I). Our experi- 300 J. S. MAKJ ET AL. .08.16 .32 .64 .08.16 .32 .64 .08.16 .32 .64 HEXADECANE (ML) Figure 1. Affinity of mid-exponential phase bacterial cells to hexadecane as a function of hexadecane volume. Results are from three separate batch cultures and are expressed as percentage of the initial absor- bance (A4,Ki) remaining in the aqueous phase as a function of hexadecane volume. A. I'ibrio vulnificus ATCC 27562. B. Isolate DLS1. C. Delcya manna ATCC 25374. D. D. manna ATCC 27129. E. D manna 35142. Bars = standard deviation. mental data indicate that the bacterial densities of 106 attached cells cm"2 were detectable by B. ncritina larvae. Because the surface free energy of the substratum is such an important factor in the attachment of bryozoan larvae, with larvae attaching in greater numbers to low surface energy, low wettability, hydrophobic surfaces, we used tests to measure both the cell-surface hydrophobic- ity and film wettability of the bacteria to determine if any correlations could be made between these measurements and larval attachment. The results of our cell-surface hy- drophobicity experiments using the adhesion to hexa- decane tests indicated that the three cultures of Deleya marina were the most hydrophobic (had the lowest sur- face free energy) of the five bacteria (Fig. 1 ). If cell-surface hydrophobicity of the bacteria was the dominant factor favoring bryozoan attachment to surfaces coated with bacteria, the larvae should have attached in greater num- bers to the more hydrophobic (lowest surface free energy) bacteria (i.e.. the cultures of D. marina). However, films of D. marina were inhibitory to larval attachment when compared to both unfilmed polystyrene and films com- posed of I '. vulnificus or isolate DLS 1 . Therefore, it ap- pears that cell-surface hydrophobicity is not the domi- nant factor controlling the attachment of bryozoan lar- vae to surfaces possessing a bacterial film. The measurements of the wettability of the bacterial film to estimate the surface free energy of the substratum may be more indicative of the surface sensed by a settling larva. Our data show that, regardless of the differences in the cell-surface hydrophobicity determinations of the bacteria, the films of all five bacteria had a similar high wettability and surface free energy (i.e.. they were hydro- philic) (Table III). Differences between cell-surface and colonial/film hydrophobicity have been previously re- ported for fish skin bacterial isolates and other bacteria (Sar, 1 987; Sar and Rosenberg, 1987). The use of bubble contact angles permits the quantification of the wettabil- Table III H 'ettahility measurements of bacterial films on polystyrene coupons using air bubble contact angles Bubble contact No. ofbacteriab Bacterium Film age angle" (xl07)cm 2(+SD) Deleya marina ATCC 25374 Day -Day 5 <15° 2.24(1.37)-1.29(0.95)c ATCC 27 129 Day Palae monetes pugio > Rhithropanopeus harrisii > Penaeus aztecus > Callinectes sapidus (121 Torr = LC50). Callinectes sapidus and E. depressus die during 1 2-h exposure to anoxia and their heat dissipation rates (quantified by microcalorimetry) are depressed in seawater at 25% air saturation (normoxia) to only 32 and 47% of their metabolic rate at normoxia. In contrast, starved Crassostrea virginica and Thais haemastorna are anoxia tolerant; their metabolic rates are depressed un- der anoxia to 75% and 9% of the normoxic rate. Hypoxia tolerance is greater at 20°C than at 30°C for Penaeus az- tecus and Crassostrea virginica, but no temperature effect on tolerance exists for Callinectes sapidus. Hyp- oxia tolerance varies inversely with salinity for Penaeus aztecus at 20° and 30°C and for Callinectes sapidus at 30°C, but it varies directly with salinity at 20°C in Calli- nectes sapidus. Greater depression of metabolic rate oc- curs in molluscs during anoxia exposure (and is corre- lated with greater hypoxia tolerance) than occurs in Cal- linectes sapidus and Penaeus aztecus, which are not anoxia tolerant. Heavy mortality probably occurs in young Callinectes sapidus and Penaeus aztecus and in Received 10 November 1988: accepted 31 July 1989. ' Present address: Department of Zoology, Iowa State University. Ames, Iowa 50011. : Permanent address: Cyclobios. Institut fur Zoologie. Abeilung Zoo- physiologie. Universitat Innsbrook, A-6020 Innsbrook, Austria. 3 Present address: Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831. stages of the life history when the organisms are incapa- ble of avoiding hypoxic water masses. Introduction Mass mortality of marine and estuarine benthic com- munities due to hypoxia has been widely reported (San- tos and Simon, 1980a, b; Harper et a/.. 1981; Officer et ai, 1984; Rabalais el al.. 1985). The occurrence of hyp- oxic bottom waters offthe Louisiana coast is a common, recurrent, virtually annual phenomenon, locally known as "dead water" (Bedinger et al., 198 1 ; Turner and Allen, 1982;Boesch, 1983; Renaud, 1985; 1986a: and Rabalais et al.. 1986a, b). Hypoxic water masses may persist for weeks. Reports suggest that fish and crustaceans avoid hypoxic waters (Pavela et al. , 1983). Juveniles of two spe- cies of shrimp, Penaeus aztecus and Penaeus setiferus, are capable of detecting hypoxic waters and initiating a pattern of avoidance behavior (Renaud, 1986b). Others have suggested that crustacean mortality may be taxon specific, and that hypoxia heavily affects the more sus- ceptible forms (Garlo et ai, 1979). The dissolved oxygen concentrations of offshore bottom waters have been ob- served to be positively correlated with the combined catches of Penaeus aztecus and Penaeus setiferus, and with fish biomass (Renaud, 1986a). The tolerance and the physiological and behavioral re- sponses of benthic and demersal invertebrates to long- term (weeks) exposure to hypoxic water is poorly under- stood. Many studies of tolerance, physiology, and bio- chemistry are carried out only for hours to a few days. During environmental anoxia, metabolism in bivalves is reduced to a relatively greater extent than in crustaceans, suggesting that bivalves should tolerate long-term anoxia 303 304 W. B. STICKLE ET AL better than crustaceans (Gade, 1983; Gnaiger 1983a, 1987). The objectives of this study are: ( 1 ) to compare the hypoxia tolerance of five species of decapod crustaceans with that of the oyster, Crassostrea virginica; (2) to com- pare the effects of temperature and salinity on the hyp- oxia tolerance ofCal/inecles sapidus, Crassostrea virgin- ica. and Penaens aztecus since these species spend all or part of their lives in the estuarine environment; and (3) to correlate the hypoxia tolerance of each species [as well as that of Thais haemastoma (see Kapper and Stickle, 1987)], with the degree of depression of metabolic rate during exposure to hypoxia and anoxia. Materials and Methods Collection and maintenance Specimens of all five species of crustaceans used in this study, and Thais haemastoma were collected in the vi- cinity of Grand Isle, LA. Crustaceans studied include the blue crab (Callinectes sapidus), the brown shrimp (Pen- aens aztecus), the xanthid crabs (Eurypanopeus depres- sus and Rluthropanopeus harrisii), and the grass shrimp (Palaemonetes pitgio). Crassostrea virginica was pur- chased at dockside in the same area. Almost all of the specimens were obtained in May or June, a time when the water temperature increased from 2 1 to 30°C. Ameri- can oysters (C. virginica used in the 10°C experiments) were collected in late October. Specimens were returned to the laboratory in Baton Rouge (LA) where they were adapted stepwise to the experimental water temperatures and salinities. Water temperature was maintained at the desired value (±0.5°C) by placing the experimental aquaria in a constant temperature water table. Experi- mental salinity (±0.5%oS) was maintained by determin- ing the salinity of Instant Ocean™ artificial seawater ( ASW) with a refractometer, and adding either deionized water or an ASW made to 40%oS. Details of the hypoxia bioassay system are given in Kapper and Stickle (1987). Briefly, each experimental chamber consisted of an aquarium (38 1), containing an undergravel filter overlaid with oyster chips. Seawater was pumped through the aquarium at a rate sufficient to ensure the water was completely exchanged several times per day. Bottled nitrogen, oxygen, and carbon dioxide were mixed with Matheson gas mixers to produce a de- fined mixture of desired P0, and pH. This air mixture passed through the undergravel filters at target oxygen tensions of 107, 53, 15,andOTorr. Ambient air was used to drive the undergravel filters of the control tanks ( 142- 157 Torr). P0, was always within 10-15% of the target value at the three higher levels, the 1 5 Torr tanks were within ±5 Torr, and the P0, of the 0% air saturation tanks was usually in the range of 3-8 Torr. Each aquarium was covered with Plexiglas and the water level was main- tained by a constant-level siphon that drained into a fil- tration unit that received water from all five experimen- tal chambers in a bioassay series at the same temperature and salinity. Experimental conditions were checked in each chamber daily by measuring P0,, pH, and ammo- nium concentration (Solorzano, 1969). The pH varied between 7.6 and 8.1 and ammonium levels were consis- tently below 25 (iM. The range of sizes and the number of individuals used at each P0, of a bioassay series varied among species: Crassostrea virginica. 30-50 mm long, an average of 20 oysters per P0,; Callinectes sapidus, 28-54 mm carapace width, an average of 8 crabs per P0,; Eurypanopeus de- pressus, 9-16 mm carapace width, with 25 crabs per P0,; Palaemonetes pugio. 16-27 mm total length with 25 shrimp per P0l ; Penaeus aztecus, 2 1 -32 mm total length, with 20 shrimp per P0,; and Rluthropanopeus harrisii, 6-11 mm carapace width, with 1 5 crabs per P0, . Animals were selected for each temperature series so as to minimize size differences within and among the temperature treatment. None of the crustaceans was ob- served to molt during the bioassay experiments. Survival at each P0, was determined daily for 28 days for each bioassay series. LC5n values — the P0, at which 50 percent of the organisms were dead on each day — were calculated by the SAS Probit procedure (SAS Insti- tute, 1982), or by the Spearman-Karber technique (Hamilton et al., 1977) if mortalities in at least two of the five P0,s were not between 0 and 100%. Percent mortality in the control P0, tanks varied from 0 to 10% for Crassos- trea virginica, from 0 to 29%> for Callinectes sapidus; and from 0 to 50% for Penaeus aztecus. Control tank mortal- ity, after 28 days exposure at 30°C and 20%oS, was 20% for Eurypanopeus depressus, 52% for Palaemonetes pugio, and 40%. for Rluthropanopeus harissii. Abbott's correction was used to correct control tank mortalities. Significant differences in LC50 values among species, temperatures, and days were determined by non-overlap of the 95%. fiducial limits. LT50 values are the elapsed days of exposure to anoxia until 50% of the experimental animals died. Thus LT50 values are a measure of anoxia tolerance only, whereas LCjo values measure the degree of hypoxia tolerance. Metabolic rate determinations Rates of heat dissipation were determined by perfu- sion (open-flow) microcalorimetry using a system de- scribed by Gnaiger ( 1 983a, b). Rate functions were calcu- lated as joules -g dry wt~'-h~' and were determined at 25°C and 10%«S for Callinectes sapidus, Penaeus az- tecus, and E. depressus. Rate functions were determined at 25°C and 20%oS for Crassostrea virginica and T. HYPOXIA TOLERANCE AND METABOLISM 305 haemastoma. The size of experimental animals was lim- ited by the size of the perfusion chamber (3.5 cm1; inner diameter = 1 1 mm, inner chamber height = 53 mm). Flow rate through the perfusion chamber was 20 ml • h ' . This method of determining metabolic rates has the ad- vantage that the sum of metabolism due to aerobic and anaerobic processes can be measured (Gnaiger, 1983a, b). The metabolic rate of each individual was determined over consecutive periods of perfusion with normoxic wa- ter, anoxic water, or 25% normoxic water ( = 39 Torr). The P0, of the outflow water was >80% of the inflow un- der normoxia, and <0.5 Torr under anoxia, as measured with a Cyclobios Twin-Flow respirometer connected to the perfusion calorimeter (Gnaiger, I983b). Hourly rates were determined from rates integrated for each minute of the hour. Steady state rates were calculated from the average of the last six hours of exposure to each experi- mental condition. Differences in steady state rates of heat dissipation among P0, treatments were determined by one-way analyses of variance (ANOVA), and specific differences among treatment means were determined by the Students /-test (SAS Inst., Inc., 1982). Results Hypoxia tolerance The long term hypoxia tolerance, at 30°C and 10%oS, of the five species of crustaceans and the oyster can gen- erally be divided into two groups. Callinectes sapidus and Penaeits aztecits were very sensitive to hypoxia with 28-day LC50 values of 1 2 1 and 1 23 Torr (79. 1 and 80.4% air saturation), respectively (Table I). The remaining spe- cies all had 28-day LC50 values lower than 60 Torr, and their hypoxia tolerance decreased in the following order: Eiirypanopeus depressus > Palaemonetes pugio > Rhi- thwpanopeits harrisii > Crassostrea virginica. Similar results are obtained whether the LC50 values are calcu- lated as Torr or percent saturation, but the variation in oxygen solubility with temperature and salinity causes the LC50 values calculated in terms of oxygen content (PPM) to deviate significantly from values calculated in Torr or percent saturation. Species differences also appeared in the rate of mortal- ity of the five species of crustaceans, and the oyster, ex- posed to defined levels of hypoxic or anoxic seawater. The LCjo values for Callinectes sapidus and Penaeus az- tecus increased rapidly as a function of time of exposure; most mortalities occurred within two days exposure (Fig. 1 ). In contrast, LCso values for Rhithropanopeus harrisii and Eiirypanopeus depressus increased slowly with dura- tion of exposure suggesting that these species are more tolerant and also more variable in their sensitivity to hyp- oxia (Fig. 2). LC5o values for Palaemonetes pugio in- creased rapidly to near the 28-day value of 46 Torr (30% saturation) on the second day of exposure with little mor- tality occurring thereafter ( Fig. 2), whereas mortality oc- curred on the seventh day for Crassostrea virginica (Fig. 1 ). The rate of mortality varied directly with temperature in Callinectes sapidus, Penaeus aztecus, and Crassostrea virginica (Fig. 1 ). No mortality occurred in oysters ex- posed to anoxia (3-8 Torr) at 10°C for 28 days (Table I). The 28-day LC50 of the crustacean species appears to be associated with differences in their natural habitats and activity levels (Fig. 3). That is, the xanthid crabs Eii- rypanopeus depressus and Rhithropanopeus harrisii, usually associated with oyster reefs, and the grass shrimp Palaemonetes pugio are significantly more tolerant to hypoxia than the potentially nektonic Callinectes sapi- dus and Penaeus aztecus. The molluscan species tested were more tolerant of hypoxia than the crustaceans (Fig. 1, 2, 3). The average 28-day LC50 for the molluscs, at 30°C and 10%oS, was 37 Torr compared with 78 Torr for the crustaceans. Furthermore, the 7-day LC50 at 30°C and 10%oS averaged 59% of the 28-day value for the crus- taceans (range 32-89%) compared with 29%. for the mol- luscs (range 0-57%), indicating that crustaceans die more rapidly. Although the oysters were very tolerant of hypoxia at 10 and 20°C, they were sensitive at 30°C. But they had spawned just before the 30°C experiment was conducted, which might have increased their hypoxia sensitivity. Metabolic rate The rate of heat dissipation was depressed in the spe- cies of crustaceans exposed to hypoxia and in Crassos- trea virginica and Thais haemastoma exposed to anoxia, as shown by analysis of variance (ANOVA); but heat dis- sipation was considerably more depressed in Thais haemastoma than in the other species (Table II). There was no significant reduction in the metabolic rate of T. haemastoma exposed to hypoxia (ANOVA: Table II). Two each of Callinectes sapidus, E. depressus, and Pa- laemonetes pugio exposed to anoxia for 12 h in the perfu- sion microcalorimetry system died during the experi- ment. Callinectes sapidus, E. depressus, Palaemonetes pugio, and T. haemastoma were therefore treated with hypoxic water at 25% air saturation (39 Torr; Figs. 4, 5). Three Palaemonetes pugio died upon exposure to 25% normoxic seawater. When Callinectes sapidus and E. depressus were ex- posed for consecutive 12-h periods to normoxic water, hypoxic water (25% air saturation = 39 Torr) and nor- moxic water, the heat dissipation rates of both species declined markedly upon exposure to hypoxic water (Fig. 4). However, the posthypoxia metabolic rate of Calli- nectes sapidus in normoxic water only returned to 75% of its pre-exposure normoxic rate, while the posthypoxia 306 W. B. STICKLE ET AL Table I Twenty-eight-day LC,,, values for several species of crustaceans and molluscs LC5(,a Species T(°C) S(%o) Torr % SAT PPM LT50b Crustaceans Callinectes sapidus 20 10 74 ± 19 47.7 4.08 <1 20 124 ± 0 79.9 6.44 <1 30 123 ± 0 79.3 6.03 <1 30 10 121 ± 0 79.1 5.63 <1 20 119± 0 77.8 5.23 <1 30 11 1 ± 0 72.5 4.61 <1 Eurypanopeus depressus 30 10 38 ± 6 24.8 1.76 1 Palaemoneies pugio 30 10 46 ± 6 30.1 2.14 1 Penaeus a:lecus 20 10 105 ± 12 67.7 5.79 <1 20 92 ± 14 59.3 4.78 <1 30 93+ 15 59.9 4.55 <1 30 10 123 ± 0 80.4 5.72 <1 20 122 ± 0 79.7 5.36 <1 30 1 1 5 ± 0 75.1 4.77 <1 Rhil hropanopeus harnsn 30 10 57 ± 18 37.3 2.65 <1 Molluscs Crassostrea virginica 10 10 <0 <0 <0 >28 20 <0 <0 <0 >28 30 <0 <0 <0 >28 20 10 27 ± 8 17.4 1.49 20 20 16± 4 10.3 0.83 18 30 30 ± 5 19.3 1.47 20 30 10 59 + 9 38.6 2.75 8 20 78+ 18 51.0 3.43 4 30 120 ± 8 78.4 4.98 3 Thais haemastoma* 10 10 15 10 1.01 20 20 8 5 0.50 27 30 9 6 0.54 20 20 10 20 13 1.10 18 20 12 8 0.62 19 30 29 19 1.43 20 30 10 13 9 0.61 >28 20 22 14 1.17 10 30 19 12 0.79 15 J LQo P0, causing 50% mortality after 28 days of exposure expressed in: Torr (x ± 95? (% SAT); content, mgO:/l (PPM). b LTS(>: days of exposure to anoxia causing 50% mortality. * Data from Kapper and Stickle (1987). confidence limits, where possible: % air saturation metabolic rate of E. depressus in normoxic water re- turned to 101% of its pre-exposure metabolic rate. The rate of heat dissipation of the T. haemastoma ex- posed to 25% air saturated water (39 Torr) was depressed significantly less relative to the rate under normoxia, than that of the decapods Callinectes sapidus and E. de- pressus (Fig. 5, Table II). Thus, Thais is a significantly better metabolic regulator than either species of crusta- cean. Furthermore, the metabolic rates of the two T. haemastoma exposed to various combinations of nor- moxia, hypoxia (39 Torr), and anoxia adjust rapidly to anoxia and exhibit an oxygen debt upon return to nor- moxic water (Fig. 5). Changes in the metabolic rates of starved Crassostrea virginica (Fig. 6A), as well as in four of the six T. haema- stoma (Fig. 6C) provided oysters ad libitum in the lab prior to their use (Fig. 6C), were examined. The heat dis- sipation rates of these animals did not return to the pre- exposure normoxic rate during the 12-h post-exposure period in normoxic seawater. The heat dissipation rate of C. virginica increased dramatically upon initial expo- sure to normoxic seawater, then declined to a steady state HYPOXIA TOLERANCE AND METABOLISM 307 Salinity I2C 100 80- 60 40 20 0 120 100 o 80 5 60 5 40 jo 20 -i 0 120 100 80 60 40 20 0 IO%oS 20%oS 30%oS 0 10 20 300 10 20 300 DAY OF EXPOSURE 10 20 30 Figure 1. LG,, values or the oxygen tension causing 50% mortality (TorrOi ± 95% fiducial limits), as a function of exposure time of Calli- nectes sapidus. Penaeus azlecus, and Crassostrea virginica- Values were obtained over 28 days, at 10. 20, and 30%oS, and at 20 (D) and 30°C (D). level that was unchanged during 12 h of exposure to an- oxic water. Upon the return to normoxic seawater, the heat dissipation rates of the oysters increased dramati- cally for the first three hours, and then declined to the initial normoxic steady state level. In contrast, the heat dissipation rate of the two apparently fed T. haemastoma exhibited an oxygen debt upon reexposure to normoxic water after 1 2 h of anoxia and then returned to the initial normoxic metabolic rate (Fig. 6B). 140 _ 120 O t 100 £ O 80 " 60 40 20 R puqio R. harrisn E. depressus 10 20 30 0 10 20 30 0 DAY OF EXPOSURE 10 20 30 Figure 2. LCs,, values (Torr O: ± 95% fiducial limits) as a function of exposure time ofPalaemonetespugio, Rhithropanopeus harrisii, and Eurypanopeus depressus: 28 days at 30°C and 10%oS. UJ o o: o P aztecus R horrisu T haemastoma 20 40 60 80 PERCENT SATURATION 100 Figure 3. LC50 values, expressed in percent saturation, for seven species after 7 days (cross hatched portion of bars) and 28 days (total bar lengths) of exposure at 30°C and 10%oS. As expected, the magnitude of metabolic rate depres- sion of these crustaceans and molluscs is directly related to their LT5(I — their mortality upon exposure to anoxia. Callinectes sapidus, E. depressus, and Pa/aemonetes pugio have LT50 values of one day or less at 20 and 30°C (Table I). Two individuals from each of these species were examined for evidence of metabolic rate depres- sion: none was found, and all the specimens died during the 12-h exposure to anoxia. The metabolic rate of Cras- sostrea virginica exposed to anoxic water was 75% of the normoxic rate, and their LT50 was 1 8 and 4 days at 20°C and 30°C, respectively. The metabolic rate of Thais haemastoma under anoxia was reduced to 9% of their rates in normoxic water (Table II) and their LT50 was 19 and 10 days at 20 and 30°C and 20%oS, respectively (Ta- ble I). No relationship exists between the degree of metabolic rate depression upon exposure to anoxia, and the 28-day LC50 values, which are indicative of hypoxia tolerance. Metabolic rate depression as a function of hypoxia ap- pears to be inversely correlated with hypoxia tolerance. Metabolic rates during exposure to 25% air saturation are reduced to 74% of the normoxic rate for T. haema- stoma, 32% for Callinectes sapidus, and 47% for E. de- pressus (Table II). Evidence of a classical oxygen debt exists in two T. haemastoma upon return to normoxic seawater after ex- posure to anoxic water (Fig. 5, 6B). This oxygen debt is of short duration in the two oyster drills shown in Figure 5. However, four T. haemastoma did not exhibit an oxy- gen debt upon exposure to normoxic seawater after 12 h to anoxic water (Fig. 6C). Discussion The five species of crustaceans and two species of mol- luscs studied differ in their tolerance to chronic hypoxia, as well as in their sensitivity to acute exposure to anoxia. Among the crustaceans, hypoxia tolerance in each spe- cies appears to be closely correlated with activity level 308 W. B. STICKLE ET AL. Table II Steady state rate of/teat dissipation (joules • g dry wt ' • h ') of four species of molluscs and crustaceans under normoxic (100% air saturation), hvpoxic (25% air saturation), and anoxic (<5% air saturation) conditions Species Normoxia Hypoxia Anoxia Crustaceans Cattinectes sapidus 5 10 18.47± 0.31 5.91 ±0.25* 32 N.D. Eurypanopeus depressus 5 10 8.70 ± 0.27 4.06 ±0.46* 47 N.D. Molluscs Crassostrea virginica 3 20 3.16± 0.51 N.D. 2. 38 ±0.47* 75 Thais haemastoma 6 20 8.76± 0.99 N.D. 0.78 + 0.04* 9 2 20 19.60 ± 10.08 14.90 + 8.55 76 N.D. All metabolic rates were determined at 25°C. N.D. = no data. All crabs died upon exposure to anoxic water. % = Percent of normoxic rate. = significantly different (P < 0.05) from the normoxic rate. and metabolic rate (Fig. 3). Eurypanopeus depressus and Rhithropanopeus harrisii are associated with oyster reefs; Palaemonetes pugio is associated with salt marsh vegeta- tion; and juvenile Callinectes sapidus and Penaeus az- tecus are active swimmers, migrating between estuaries and offshore waters during their life cycles. A two-fold difference also exists in the metabolic rates of Eurypano- peus depressus and Callinectes sapidus exposed to nor- moxic seawater (Table II). Sensitivity to hypoxia has also been measured as a function of mortality in anoxic seawater, represented by LT50 values. The resistance of marine invertebrates to oxygen deficiency is correlated with the natural habitats of the species (Fig. 3; Theede et a/., 1969; Theede 1973). In this study, mortality occurred more rapidly in the crustaceans exposed to hypoxia than in the molluscs, when all of the temperature-salinity combinations tested were considered. All of the crustacean species exposed to anoxia had LT50 values of one day or less, whereas the LT50 values of Crassostrea virginica ranged from greater than 28 days at 10°C, to three days at 30"C and 30%oS. Moreover, the LT50 of Thais haemastoma ranged from greater than 28 days at 30°C and 10%oS, to 10 days at 30°C and 20%oS. The LT50 value is not a very sensitive indicator of hyp- oxia tolerance, because only animals exposed to anoxia can be included in the calculation of the parameter. In contrast, the determination of a LC50 value requires data about survival as well as mortality, from several P0, treat- ments, so several degrees of hypoxia are represented. There is no correlation between the LT50 and the 28- day LCsoPo,. If, for the species studied, the LC50 values for short term exposure are expressed as a function of the 28-day LC50, the phylogenetic differences between crustaceans and molluscs are clearly highlighted. Thus, day 2 and day 7 LC.so values represent 0 and 21% of the 28-day values for the two species of molluscs at 30°C and 10%oS, whereas they represent 32 and 60% of the 28-day value for the crustaceans. Other environmental factors have an antagonistic effect upon the tolerance of estuarine invertebrates to hypoxia and anoxia. Sensitivity to hypoxia increases with temperature in Crassostrea virginica and Penaeus aztecus, as well as in T. haemastoma (Kapper and Stickle, 1987), and probably results from an increased metabolic rate at elevated temperature. The survival time of oysters, experimentally buried to simulate natu- ral sedimentation events, varied inversely with tempera- ture, from more than 5 weeks at <5°C, to 4 days at tem- peratures >25°C (Dunnington, 1968). Prolonged expo- sure of oysters to fresh water and low salinities (<5%oS) has caused heavy mortality because they remained closed and could not feed and maintain an aerobic meta- bolic rate (Andrews, 1982). Oysters died, presumably be- cause of anoxic conditions produced by dredging which resulted in an oxygen demand of spoil bank sediments and modification of the local hydrographic regime (HoeseandAncelot, 1987). Hypoxia tolerance varied inversely with salinity for Penaeus aztecus at 20 and 30°C and for Callinectes sapi- dus at 30°C, but varied directly with salinity at 20°C for Callinectes (Table I). The inverse relationship between tolerance and salinity noted for Callinectes sapidus and Penaeus aztecus is the opposite of that expected on the basis of theoretical osmoregulatory costs; but activity patterns associated with feeding may override the ener- getic costs of osmoregulation. Juvenile blue crabs and shrimp sometimes use the most brackish regions of estu- aries as a nursery ground so the inverse relationship be- tween salinity and hypoxia tolerance is correlated with the distribution of life history stages of these species. Behavioral avoidance activities by juvenile (65 to 101 mm total length) penaeid shrimp may temporarily allow them to escape oxygen deficient water, i.e., below 2.0 ppm (29% air saturation or 45 Torr at 22°C and 22%oS) HYPOXIA TOLERANCE AND METABOLISM 309 30 25 20 15 ~ 10 100 50 15 20 25 (Hour) Figure 4. Hourly heat dissipation rates of 5 Callinectes sapidus (A) and 5 Eurypanopeus depressus (B) at 25°C and 10%»S. Rates ex- pressed as J-g dry wr'-h"' ± S.E. (mean -^- and standard error limits — • — ) for crabs exposed to consecutive 1 2 h periods of norm- oxia, hypoxia (25% air saturation), and normoxia. Dashed lines: per- cent air saturation values for ambient seawater. for Penaeus aztecus and 1.5 ppm (22% air saturation or 34 Torr) for Penaeus setiferus ( Renaud, 1 986b). Juvenile shrimp may be able to alter their migration patterns to move around patches of hypoxic water (Rabalais et al., 1986a, b; Renaud, 1986a), provided the patches are spa- tially and temporally isolated. The values for metabolic rate depression observed in this study fall within the range of values reported in the literature and emphasize the need to consider this com- ponent of physiological adaptation in relation to the life cycle niche occupied by each species. It is not surprising that juvenile blue crabs, which are active swimmers, are intolerant of anoxia, and that their metabolic rate at 25% air saturation is depressed only to 32% of their normoxic metabolic rate. Although Eurypanopeus depressus is also intolerant of anoxia and exhibits metabolic rate depres- sion to 47% of its normoxic rate upon exposure to 25% air saturated seawater, its brackish water oyster reef habi- tat is not exposed to hypoxic water masses of the same duration as those that develop offshore (Rabalais et al., 1985). These crabs are exposed to diurnal variations in oxygen tension. In contrast, the copepod Cyclops abys- sorum. which lives in alpine ponds that become hypoxic in the winter, exhibits metabolic rate depression to 1 7% of the normoxic rate upon exposure to anoxia at 6°C (Gnaiger, 1981). The metabolic rate depression of molluscs exposed to anoxic seawater is also highly variable among species, ranging from 9% of the normoxic rate in T. haemastoma (Table II), 1 1%. in Mytilus edit/is (Famme et a/., 1981; Shick et al., 1983), 75% in Crassostrea virginica to only 97% in Mulinia lateralis (Shumway et al., 1983). Sessile and infaunal bivalves generally show a strong resistance to anoxia due in part to a reduction in activity and hence energy use (Shick et al., 1986). In this study, Crassostrea virginica (2 to 19 mg dry weight) were starved for 35 to 65 days prior to the experiment. In bivalves, their oxygen consumption rate is directly coupled with nitration activ- ity associated with feeding (see discussion by Bayne et al., 1976). Starved Crassostrea virginica probably exhib- ited a reduced nitration activity and heat dissipation rate. In Mytilus edulis, the increase in oxygen consumption of starved mussels offered food is almost instantaneous (Widdows, 1973). The increased heat dissipation rate of oysters immediately after perfusion with normoxic sea- water (Fig. 6A) may therefore represent "testing" of an altered ambient environment after which the active rate of the oysters was reduced to the standard, nonfeeding rate. T. haemastoma is exposed to diurnal and seasonal periods of anoxia, both in the water column, and when it burrows into the anoxic zone of sediments for a large portion of the winter and intermittantly in the summer (Kapper and Stickle, 1987). The two metabolic patterns shown by T. haemastoma _ o s 100 100 50 5 10 15 20 25 30 36 40 45 60 (Hour) Figure 5. Hourly heat dissipation rates (at 25°C and 30%oS) of two (A and B) Thais haemastoma (solid lines — | — ) exposed to various com- binations of normoxia (100% air saturation), hypoxia (25%- air satura- tion), and anoxia. Dashed lines: percent air saturation values for ambi- ent seawater. 310 W. B. STICKLE ET AL in response to normoxic seawater after 12-h exposure to anoxic seawater (Fig. 6B, C), are probably related to the feeding history of the snails. Feeding rate is the primary bioenergetic component to become variable in gradients of environmental factors, and certain individuals cease feeding altogether under stressful conditions (Stickle, 1985). Small oyster drills, such as those used in this study, are particularly sensitive to the selection of opti- mum-sized bivalve prey, because prey size can limit the ingestion rate, and hence the energy budget of the preda- tor (Garton, 1986). T. haemastoma prefers a number of different prey items (Butler, 1953), the importance of which may vary with the size of the snail. Oyster drills also exhibit a large specific dynamic action effect, ele- vated metabolic rates associated with digestion of food, in normoxic seawater, anoxic seawater, and when ex- posed to the air (Stickle el a/., 1986). All of these factors probably contributed to the variability in individual met- abolic rates which resulted in two apparent patterns of response in the recovery of oyster drills from 12 h of an- oxia. The metabolic rate depression of T. haemastoma ex- posed to anoxia (Table II) suggests a switch to the rela- tively more efficient succinate and propionate pathways in the molluscs, compared with the well developed, but less efficient, classical glycolysis system in the crusta- ceans(Gade, 1983; Gnaiger, 1983a, 1987; and deZwaan and Thillart, 1985). During initial exposure to environ- mental anaerobiosis, the biochemically estimated ATP turnover rate may drop to about 10% of aerobic resting rates in crustaceans, and the reductions may be even larger in molluscs (deZwaan and Thillart, 1985). During the initial exposure to anoxia, when aspartate is still the precursor of succinate in the molluscs, the rate is three to five times higher than the subsequent anaerobic steady state and is fueled by both phosphagen and ATP hydro- lysis (deZwaan and Thillart. 1985; Rapper and Stickle, 1987). When the steady state is reached, the glycolytic flux is reduced and channeled towards malate, whereas the phosphagen pool is somewhat depleted relative to normoxia levels (deZwaan and Thillart, 1985; Kapper and Stickle. 1987). No evidence of oxygen debt was observed with Calli- nectes sapidus (Fig. 4A), E. depressus (Fig. 4B), starved Crassostrea virginica (Fig. 6A), or specimens of T. haemastoma whose metabolic rates did not even recover to the initial normoxic rate after 1 2-h exposure to anoxic seawater (Fig. 6C). In contrast, the T. haemastoma indi- viduals that did recover to the initial normoxic rate after 12-h exposure to anoxic seawater, exhibited an oxygen debt upon their postanoxic exposure to normoxic seawa- ter (Fig. 6B). Two basic processes occur during recovery from envi- ronmental anoxia: ( 1 ) recharging of the phosphagen 10 15 20 25 30 35 (Hour) Figure 6. Hourly heat dissipation rates of three Crassostrea virgin- ica (A), two Thais haemastoma which recovered from exposure to an- oxia at 25°C and 20%»S (B), and four T. haemastoma which did not recover from exposure to anoxia (C). Rates expressed as: J -g dry wt ' • h ' ± S.E. (solid lines for the mean ^? : standard error limits by solid lines). The animals were exposed to consecutive 12-h periods of nor- moxia, anoxia, and normoxia at 25°C and 20%»S. Dashed lines: percent air saturation values for ambient seawater. pool; and (2) the disposal of end products by excretion, oxidation, or conversion back to anaerobic substrates (Ellington. 1983). Oxygen debts are regular phenomena in free-living invertebrates and may be attributed, at the molecular level to the increased energy demands for dis- posal of end products and recharging of the phosphagen and ATP pools (Herreid, 1980). Patterns of oxygen debt in invertebrates tend to be highly variable from species to species, and may reflect differences in the degree of reduction of energy metabolism, hence end-product ac- cumulation under anoxic conditions (Herreid, 1980) and the duration of exposure to anoxia. Upon exposure to anoxia for 24 h, Thais haemastoma, the species that exhibited the greatest metabolic rate depression in our study (Table II), showed a return of the adenylate energy charge to the pre-exposure level within 6 h. But the argi- nine phosphate concentration returned to only about half of its pre-exposure value 24 h after the oyster drills were returned to normoxic water (Kapper and Stickle, 1987). Lack of an oxygen debt in some T. haemastoma (and possibly Crassostrea virginica), whose postanoxia metabolic rate did not recover to the initial rate after 12- h exposure to anoxia, may be due to a reduced metabolic HYPOXIA TOLERANCE AND METABOLISM 311 rate after long term starvation (Fig. 6C compared with 6B) coupled with a slow replenishment of the phospha- gen pool and, perhaps, washout of anaerobic end prod- ucts during exposure to anoxic water. In conclusion, significant interspecies variability exists in the 28-day LC50 values for the five species of crusta- ceans studied, ranging from 38 Torr in Eurypanopeits de- pressus, to 121 Torr in Callinectes sapidm. In addition, species differences exist in the rate of mortality of the five species of crustaceans, and in the oyster, when exposed to defined levels of hypoxic or anoxic seawater. LC50 val- ues for Callinectes sapidus. Penaeus aztecus, and Palae- monetes pugio increase rapidly during the first two days of exposure, in contrast to those ofRhithropanopeus har- risii and Eurypanopeus depressits which increased slowly over the exposure period. Sensitivity to hypoxia increased with temperature in Callinectes sapidus, Cras- sostrea virginica. and Penaeus aztecus. with salinity effects being less significant. Natural habitat, activity level, and seasonal differences appear to exist in the mor- tality rate of these five species of crustaceans. Both Calli- nectes sapidus and Eurypanopeus depressus died during 12 hours exposure to anoxia with little decline in the metabolic rate, and their metabolic rate in 25% air satu- rated seawater is reduced to only 32-40% of their meta- bolic rate under normoxia. In contrast, both Crassostrea virginica and Thais haemastoma are tolerant of 25% air saturated seawater, with the rate of Thais haemastoma being 76% of that under normoxic conditions. Metabolic rate depression occurs in both species under anoxic sea- water, to 75% of the normoxic rate in Crassostrea virgin- ica, and 9% in Thais haemastoma. Acknowledgments The authors acknowledge the Coastal Fisheries Insti- tute at LSU and the Petroleum Refiners Environmental Council of Louisiana for partially funding this research. E. G. was supported by FWF Austria grants J001 1 and J0187B, and by the LSU Visiting Investigator Program. Special thanks go to LKB-Thermometrics for loaning us a prototype Thermal Activity Monitor perfusion system. We are grateful to Nancy Rabalais of LUMCON for re- viewing drafts of this manuscript. Literature Cited Andrews, J. D. 1982. Anaerobic mortalities of oysters in Virginia caused by low salinities. / Shellfish- Res. 2: 127-1 32. Bayne, B. L., R. J. Thompson, and J. Widdows. 1976. Physiology I. Pp. 121-206 in Marine Mussels: Their Ecology and Physiology. B. L. Bayne. ed. Cambridge University Press, Cambridge. Bedinger, C. A., R. E. Childers, J. W. Cooper, K. T. Kimball, and A. Kwok. 1981 . Pollution fate and effect studies. Pp. 1 -53 in Ecologi- cal Investigations of Petroleum Production Platforms in the Central Gulf of Mexico, Vol. 1, Pt. 1, C. A. Bedinger, ed. Final Report to the Bureau of Land Management. New Orleans, LA, Contract No. A5551-CT9-17. Boesch, D. F. 1983. Implications of oxygen depletion on the conti- nental shelf of the northern Gulf of Mexico. Coastal Ocean Pollut. Assess. News. 2: 25-28. Butler, P. A. 1953. The southern oyster drill. Proc. Nail. Shellfish Assoc. 53:67-75. Dunnington, E. A., Jr. 1968. Survival time of oysters at various tem- peratures. Proc. Nail. Shellfish. Assoc. 58: 101-103. Ellington, VV. R. 1983. The recovery from anaerobic metabolism in invertebrates./ Exp. Zool. 228: 431-444. Famme, P., J. Knudson, and E. S. Hansen. 1981. The effect of oxygen on aerobic-anaerobic metabolism of the marine bivalve, Mvlilus edulis L. Mar. Biol. Lett. 2: 345-35 1 . Cade, G. 1983. Energy metabolism of arthropods and molluscs dur- ing environmental and functional anaerobiosis. /. Exp. Zool. 228: 415-429. Carlo, E. V., C. B. Milstein, and A. E. John. 1979. Impact of hypoxic conditions in the vicinity of Little Egg Inlet. New Jersey in summer 1976. Esiuar. Coast. Mar. Sci. 8: 421-432. Carton, D. W. 1986. Effect of prey size on the energy budget of a predatory gastropod. Thais haemastoma canaliculata (Gray). J. Exp. Mar. Biol. Ecol. 98: 21-33. Cnaiger, E. 1981. Pharmacological application of animal calorime- try. Thermochim. Ada 49: 75-85. Gnaiger, E. 1983a. Heat dissipation and energetic efficiency in ani- mal anoxibiosis: economy contra power. J. Exp. Zool. 228: 471- 490. Gnaiger, E. 1983b. The twin-flow microrespirometer and simulta- neous calorimetry. Pp. 134-166 in Polarographic Oxygen Sensors, Aquatic and Physiological Applications. E. Gnaiger and H. Forst- ner. eds. Springer, Berlin. Gnaiger, E. 1987. Optimum efficiencies of energy transformation in anoxic metabolism. The strategies of power and economy. Pp. 7- 36 in Evolutionary Physiological Ecology. P. Calow, ed. Cambridge University Press, Cambridge. Hamilton, M. A., R. C. Russo, and R. V. Thurston. 1977. Trimmed Spearman-Karber method for estimating median lethal concentra- tions in toxicity bioassays. Environ. Sci. Techno!. 11:71 4-7 1 9. Harper, D. E., L. D. McKinney, R. R. Salzer, and R. J. Case. 1981. The occurrence of hypoxic bottom water off the upper Texas coast and its effect on the benthic biota. Contrib. Mar Sci. 24: 53-79. Herreid, C. F. II. 1980. Review: hypoxia in invertebrates. Comp. Bio- chem. Phyxiol. 67A: 31 1-320. Hoese, H. D., and R. Ancelet. 1987. Anoxia induced mortality of oys- ters. Crassostrea virginica, associated with a spoil bank bisecting a bay. / Shellfish Res. 6: 41-44. Kapper, M. A., and W. B. Stickle. 1987. Metabolic responses of the estuarine gastropod Thais haemas/oma to hypoxia. Physiol. Zool. 60: 159-173. Officer, C. B., R. B. Biggs, J. L. Taft, L. E. Cronin, M. A. Tyler, and W. R. Boynton. 1984. Chesapeake Bay anoxia: origin, develop- ment and significance. Science 113: 22-27. Pavela, J. S., J. L. Ross, and M. E. Chittenden. 1983. Sharp reduc- tions in abundance of fishes and benthic macroin vertebrates in the Gulf of Mexico off Texas associated with hypoxia. Northeast Gulf Sci. 6: 167- 1 72. Rabalais, N. N., M. J. Dagg, and D. F. Boesch. 1985. Nationwide review of oxygen depletion and eutrophication in estuarine and coastal waters: Gulf of Mexico (Alabama. Mississippi. Louisiana and Texas). Final Rep. NOAA, National Ocean Service, Ocean As- sessment Division, Rockville, Maryland. 60 pp. Rabalais, N. N., R. E. Turner, W. J. Wiseman, Jr., and D. F. Boesch. 312 W. B. STICKLE ET AL. 1986a. Hydrographic, biological, and nutrient characteristics of the water column in the Mississippi River Delta Bight, June, 1985 to December, 1985. Data Report No. 2, Louisiana Universities Ma- rine Consortium, Chauvin, LA 70344. 167 pp. Rabalais, N. N., R. E. Turner, W. J. Wiseman, Jr. and D. F. Boesch. I986b. Hydrographic, biological and nutrient characteristics of the water column on the Louisiana Shelf, July and September, 1985. Data Report No. 3, Louisiana Universities Marine Consor- tium, Chauvin, LA 70344. 1 50 pp. Renaud, M. L. 1985. Annotated bibliography on hypoxia and its effects on marine life, with emphasis on the Gulf of Mexico. NOAA Technical Report NMFS 21, U.S. Dept. of Commerce, Washing- ton, DC. 9 pp. Renaud, M. L. 1986a. Hypoxia in Louisiana coastal waters during 1983: implications for fisheries. Fish. Bull 84: 19-26. Renaud, M. L. 1986b. Detecting and avoiding oxygen deficient sea water by brown shrimp, Penaeus aztecus (Ives), and white shrimp Penaeus seliferus (Linnaeus). / Exp. Mar. Biol. Ecol. 98: 283-292. Santos, S. L., and J. L. Simon, 1980a. Marine soft-bottom commu- nity establishment following annual defaunation: larval or adult re- cruitment? Mar. Ecol. Prog. Ser. 2: 233-24 1 . Santos, S. L., and J. L. Simon. 1980b. Response of soft-bottom ben- thos to annual catastrophic disturbance in a south Florida estuary. Mar Ecol. Prog. Ser. 3: 347-355. SAS Institute, Inc. 1982. SAS User's Guide: Statistics, 1982 Edition. SAS Institute, Inc.. Cary, NC. 584 pp. Shick, J. M., A. deZwaan, and A. M. J. DeBont. 1983. Anoxic meta- bolic rate in the mussel Mytilus edulis L. estimated by simultaneous direct calonmetry and biochemical analysis. Phvsiol. Zoo/. 56: 56- 63. Shick, J. M., E. Gnaiger, J. W'iddows, B. L. Bayne, and A. deZwaan. 1986. Activity and metabolism in the mussel Mytilus edulis L. during intertidal hypoxia and aerobic recovery. Phvsiol. Zoo/. 59: 627-642. Shumway, S. E., T. M. Scott, and J. M. Shick. 1983. The effects of anoxia and hydrogen sulphide on survival, activity and metabolic rate in the coot clam, Mulinia laleralis (Say). / Exp. Mar. Biol. Ecol. 71: 135-146. Solorzano, L. 1969. Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr 14: 799-801. Stickle, W. B. 1985. Effects of environmental factor gradients on scope for growth in several species of carnivorous invertebrates. Pp. 60 1-6 16 in /Vor. 18lh European Mar. Biol. Symp.. 1983, J. S. Gray and M. E. Christiansen, eds. Wiley, New York. Stickle, W. B., L. L. Liu, and E. Gnaiger. 1986. Metabolic rate varia- tion in Thais haemasloma as a function of exposure to anoxia, and air and during its feeding cycle. Am. Zoo/. 26: 47A. Theede, H., A. Ponat, K. Hiroki, and C. Schlieper. 1969. Studies on the resistance of marine bottom invertebrates to oxygen deficiency and hydrogen sulphide. Mar. Biol. 2: 325-337. Theede, H. 1973. Comparative studies on the influence of oxygen de- ficiency and hydrogen sulphide on marine bottom invertebrates. Neth. J. Sea Res. 1: 244-252. Turner, R. E., and R. L. Allen. 1982. Bottom water oxygen concentra- tion in the Mississippi River Delta Bight. Contr. Mar. Sci. 25: 161- 172. Widdows, J. 1973. Effect of temperature and food on the heart beat, ventilation rate and oxygen uptake ofMvtilus edulis. Afar. Biol 20: 269-276. deZwann, A. andG. V. D. Thillart. 1985. Low and high power output modes of anaerobic metabolism: invertebrate and vertebrate strate- gies. Pp. 166-192 in Circulation. Respiration and Metabolism: Cur- rent Comparative Approaches, R. Gilles, ed. Spring- Verlag Berlin Heidelberg. Reference: Biol. Bull. 177: 313-330. (October, 1989) Abstracts of Papers Presented at the General Scientific Meetings of the Marine Biological Laboratory August 14-16, 1989 Abstracts are arranged alphabetically by first author within the following categories: cell motility and cytoskel- eton: gametes and developmental biology; general physi- ology: neurobiology; pathobiology and environmental studies; and sensory biology. Author and subject refer- ences will be found in the regular volume index in the December issue. Cell Motility and Cytoskeleton Calcium-sensitive ATP-reactivated models o/Ciona in- testinalis branchial basket cilia. DWIGHT E. BERGLES AND SIDNEY L. TAMM (Boston University Marine Program). The cilia lining the stigmata of the branchial basket ofdona inlesli- na/is generate the water currents necessary for feeding and gas ex- change. The metachronal beating of these cilia is often interrupted by sudden stigmata-wide arrests, in which the cilia lie flat against the stig- matal opening. These arrests are always followed by a transient inactive state in which the cilia rise to an upright position before metachronal beating resumes (Takahashi et al. 1973, / Fac. Sci. U. Tokyo 13: 123- 137). To investigate the ionic and molecular control of these motor responses, we have developed permeabilized ATP-reactivated models of dona cilia. Cell models were produced by extracting small pieces of branchial basket in 0.1%saponin, 10 mA/EGTA. 30 mA/ PIPES. 150mA/KCl, 10 mA/ MgCl;, pH 7 (ES) for 4-7 min at room temperature. TEM showed that this treatment disrupted or completely removed the ciliary membranes. Transfer of extracted tissue to reactivation solution (RS: 10 mA/ EGTA. 30 mA/ PIPES. 150 mA/ K.C1, 10 mA/ MgCU. 2 mA/ ATP. 1 mA/ DTT, pH 8.0) caused the cilia to beat rapidly and without metachrony for 15-30 min. Calcium (20 nM-\ mA/) in EGTA-free RS induces the cilia to as- sume an inactivated posture (stigmata closed), but not the arrested state (stigmata open). Calcium-dependent mactivation is reversed, and the cilia resume beating in the presence of 100 nM trifluoperazine (TFP) or calmidazolium, indicating that ciliary inactivation is regulated by a calcium-calmodulin pathway. The absence of a full laydown arrest in our ciliary models may be due to a loss or modification of critical con- trol factors during extraction. Preliminary results using cAMP, phos- phatases, and phosphatase inhibitors suggest that ciliary responses are mediated by protein phosphorylation levels. Supported by NIH Grant GM 27903 to S.L.T. Studies of membrane and cytoskeletal structures by elec- troporation using a radio-frequency electric field. DON- ALD C. CHANG AND QIANG ZHENG (Baylor College of Medicine, Houston, TX 77030). High intensity pulsed electric fields can be used to transiently per- meabilize cell membranes. This method, called "electroporation," is employed to introduce exogenous molecules into cells. Recently, we developed a new method of electroporation, in which a radio-frequency (RF) electric field, rather than a direct current (DC) field, permeabilizes cells. This new method improves cell viability and provides higher poration efficiency. In the first part of this study, we used the RF pora- tion method to introduce rhodamme-labelled phalloidin into an at- tached cell line (CV-1) in order to study the microfilament system within the living cell. The CV-1 cells were grown on coverslips. In the presence of phalloidin, cells in the center region of the coverslip were exposed to pulses of RF electric field (0.8 kV/cm, 1 00 kHz, 1 ms wide). The cells were washed and then incubated in normal culture medium at 37°C for 30 min before being observed under a fluorescence micro- scope. Brilliant rhodamin-labelled stress fibers were found in nearly all cells exposed to the RF field, but none were found in the control cells. From morphological observations, we estimated that over 60% of the cells loaded with rhodamine were viable. In the second study, we used a rapid-freezing electron microscopy technique to examine the change of membrane structure following the RF electroporation process. Human red blood cells were used in this study because their membranes had already been well characterized. Following the exposure to the electric field, volcano-shaped, pore-like structures were found in the membranes of the porated cells. Their openings were approximately 20 nm in diameter in the first few milli- seconds after the pulsation. In the next 50 ms, their openings expanded and some of them were as large as 1 20 nm in diameter. Such expansion suggests that the electropores were shaped, not only by the electrical breakdown of the membrane, but also by secondary effects involving movement of water, ions, and other molecules. Supported in part by a grant from the Texas Advanced Technology Program. Quantitative motion analysis of vesicle movement in Y-l adrenocortical cells and the use of fluorescent probes to identify the organelles. GEORGE M. LANGFORD (Uni- versity of North Carolina, Chapel Hill), SANDRA A. MURRAY, BROCKTON HEFFLIN, AND KATHLEEN J. PENNY. 313 314 ABSTRACTS FROM MBL GENERAL MEETINGS Video microscopy was used to study the motion of vesicular organ- elles in V-l adrenocortical tumor cells. The organelle movements were recorded in real time and subjected to quantitative motion analysis. Y-l cells form a typical epithelial monolayer in culture, with broad extensions of very shallow depth ( 1-2 ^m). In the extensions, vesicles can be observed moving away from (anterograde), toward (retrograde), and lateral to the nucleus. The movements observed are presumed to be microtubule-dependent, as has been shown for organelle movement in axons. When stimulated with dibutyryl cAMP (1 mA/) for 8-24 hours, these cells showed a 3-6 fold increase in steroid production. The motion of several organelles (lipid granules) in unstimulated cells were quantitatively analyzed. Their movement can be classified as inter- rupted motion II according to the preliminary scheme of Weiss el al. ( 1 986, Cell Alotil. Cytoskel. 6: 128). This classification refers to organ- elles that show pauses and reversals, and is typical of organelle move- ment in cultured cells. The instantaneous velocity (0.33-s intervals) was determined for three lipid granules. The fastest moving organelle had a maximum velocity of 2.0 ^m/s. The velocities of the organelles fluc- tuated, and the average velocities for the three organelles ranged be- tween 0.4 and 1.1 ^m/s. Each organelle showed periods of movement followed by pauses. During a pause, the organelle remained in place as though tethered to the microtubule and resumed moving at maximum rates as though a force was being constantly applied to free it from an obstacle in its path. Rhodamine 123 and DIOC6 were used to identify the mitochondria and endoplasmic reticulum in these cells. Future studies are designed to determine changes in the movement pattern upon stimulation with ACTH and dibutyryl cAMP. Supported by NSF grants DCB 88 1 8279 to G.M.L. and RII-8402666 to S.A.M. B.H. was supported by a fellowship from the American Society for Cell Biology; K.J.P. was supported by a fellowship from the American Physiological Society. Paddle cilia occur as artifacts in veliger larvae o/"Spisula solidissima and Lyrodus pedicellatus. GRAHAM SHORT AND SIDNEY TAMM (Boston University Ma- rine Program). Cilia with paddle-shaped tips (paddle cilia or discocilia) have been described by TEM and SEM in a variety of marine invertebrates, most recently in the pretrochal ciliary bands of Spisula solidissima veligers (Campos et ai 1989, Biol. Bull. 175: 343-348). We have investigated whether such modified cilia are genuine structures or artifacts of osmo- larity or fixation conditions. Living veligers ofSpisula were observed in normal seawater by high- speed video microscopy (DIC and phase-contrast) synchronized with a strobe flash. No paddle cilia were present. The SEM fixative of Campos et al. ( 1 989), consisting of 2.5% gluteraldehyde. 0. 1 M Na cacodylate, pH 7.2, has an osmolarity of 404 mosmol; it induced paddle cilia in our Spisula veligers, as determined by light microscopy and SEM. Addition of 0.29 Al NaCl to this fixative (pH 7. 1 ) to make it isosmotic with MBL seawater (920 mOsmols) produced no paddle cilia. Similarly, an isos- motic fixative (pH 6.3) containing 2.5% gluteraldehyde, 0.13 Al NaCl, and 50% seawater did not induce paddle cilia. The same fixation conditions applied to shipworm larvae (Lyro- dus pedicellatus) gave the same results as described in Spisula. When living Spisula and Lrmdus were placed in diluted seawater (45%; 420 mOsmols), the distal ciliary membrane vesiculated, as determined by light microscopy. This hypotonic swelling is reversible; the cilia regain their normal appearance when the larvae are returned to 100% sea- water. We conclude that paddle cilia in Spisula larvae, and probably in other invertebrates as well, are artifacts caused by swelling of the distal ciliary membrane in hypotonic medium. Various hydromechanical and chemosensory functions attributed to paddle cilia by previous au- thors (Matera et al. 1982, Cell Tiss. Res. 222: 25-40; Stebbing et al.. 1972, Mar Biol. Assoc. UK 52: 443-448) must therefore be abandoned. Supported by the Woods Hole Marine Science Consortium and NIH Grant GM 27903 to S.L.T. Cell fusion induced by a radio-frequency electric field. Qi- ANG ZHENG AND DONALD C. CHANG (Baylor College of Medicine, Houston, TX 77030). Cell fusion is a very important and useful technique for hybndoma production and somatic hybridization. High intensity pulsed electric fields can be used to induce cell fusion; this is called "electrofusion." Using a new method of electrofusion developed in this laboratory, we applied a radio-frequency (RF) electric field to induce fusion of plant protoplasts and mammalian cultured cells. In the first study, proto- plasts enzymatically digested from cabbage leaves were fused using three pulses of an RF field ( 1 kV/cm. 60 kHz. 0.2 ms). Within a few minutes after pulsation, over 70% of protoplasts fused, and microscopic observation showed very little damage to the protoplasts by the RF field. In the second study, attached mammalian cells (CV-1, COS-M6, Pam, 3T3) were fused by the RF method. We observed a high frequency of fusion in all of these cells. For example, the fusion frequency of at- tached CV-1 cells could reach 80% even when they were treated using a single RF pulse (1.0 kV/cm, 100 kHz, 0.2 ms); in this case, almost all cells were alive. To understand the mechanisms of cell fusion, we investigated the role of the cytoskeleton in reorganizing the cellular structures during the fusion process. The cytoskeletal changes of at- tached CV-1 cells were examined by fluorescence microscopy using rhodamine-phalloidin and anti-tubulin antibodies labelled with FITC. Within a few minutes after the electric pulsation, the membranes fused. During the next 20-30 min, some of the stress fibers gradually disap- peared; F actin began to condense near the nucleus or at the cell periph- ery, while microtubules condensed between nuclei within the fusing cells. If the fusion involved only 2 or 3 neighboring cells, the fusion process was almost completed after about 90 min. Apparently normal stress fibers reappeared at this time. Supported in part by Texas Advanced Technology Program. Gametes and Developmental Biology Binding of 5-hydroxytryptamine to isolated plasma membranes of Spisula gametes. A. H. BANDIVDEKAR (The Population Council), S. J. SEGAL, AND S. S. KOIDE. Serotonin (5-hydroxytryptamine. 5-HT) added in vitro to a suspen- sion ofSpisula oocyles induced germinal vesicle breakdown (GVBD). The neurotransmitter also stimulated, in vitro, the motihty of cold-im- mobilized Spisula sperm. In the present study, the plasma membranes were prepared from Spisu/a oocytes and sperm, and the binding of ['H]5-HT to the membranes was determined. 5-HT analogs were used to displace the bound [' H]5-HT to determine the types of receptors present on oocytes and sperm membranes. Plasma membranes of oocytes were prepared by suspending isolated washed Spisula oocytes in 50 mA/Tns-HCl buffer, pH 7.4, containing 1 mAf EDTA, 0.001% sodium azide. The membranes were homoge- nized with 2-4 light strokes in a glass-teflon homogenizer, and sedi- mented by centrifugation. Spisu/a sperm membranes were prepared by a nitrogen cavitation method, at a pressure of 1 500 psi for 30 min. The treated sperm were then centrifuged at 10,000 X g for 30 min at 4°C. and the supernatant centrifuged at 100,000 x g for 2 h at 4°C to sediment the membranes. The radioligand binding assay system contained 1 00 ng of membrane GAMETES AND DEVELOPMENTAL BIOLOGY 315 protein, and 100 ^1 of ['H]5-HT (40 pmoles); labelled ligand was dis- placed with 5-HT and its analogs. The Kd of [ H]5-HT binding to Spisula oocyte and sperm mem- branes was 17.5 nA/and 2.7 nA/, respectively. The maximum binding capacity was 7.9 pmoles/mg and 1 1.25 pmoles/mg, respectively. The order of decreasing potency in the displacement of ['HJ5-HT binding to Spisu/u oocyte membranes by 5-HT agonists was: 5-HT > 5-CT > 8- OH-DPAT > 2-methyl-5-HT > alpha-methyl-5-HT, and that by 5-HT antagonists was ICS-205-930 > mianserin > methysergide > BMY- 7378 > ketanserin > quipazine. The order of decreasing potency in the displacement of ['H]5-HT bound to sperm plasma membranes by 5- HT agonists was 2-methyl-5-HT > 8-OH-DPAT > 5-HT > 5-CT > alpha-methyl-5-HT, and that by antagonists was ICS-205-930 > BMY-7378 > mianserin > methysergide. The present findings demonstrate that plasma membranes ofSpixula oocytes and sperm possess 5-HT,A and 5-HT, receptors. Oocyte mem- brane may also contain 5-HT: receptor sites. This study was supported by a grant from The Rockefeller Founda- tion. The 5-HT analogs were gifts from Glaxo, Sandoz, CIBA-Geigy, Roussel UCLAF, Farmatalia, Lilly, and Bristol-Myers. Early cleavage and the role oj the macromeres in the de- velopment ofthepolydadflatworm Hoploplana. BAR- BARA C. BOYER AND GWENDOLYN A. WALLACE (Union College). Blastomere deletion experiments were used to investigate the role of the macromeres at the eight-cell stage (first quartet) in the development of the polyclad turbellarian Hoploplana inquilina. In particular, eye de- velopment, general morphology, and determination of the embryonic axes of symmetry were examined after deletion of macromeres 1A or 1C, IB or ID, two adjacent macromeres, three macromeres, and all four macromeres. Normal Miiller's larvae resulted in 31% of cases in which 1A or 1C was deleted, and in 23% after removal of IB or ID. Macromere dele- tions usually did not lead to loss of eyes (no more than one quarter of any experimental category included one-eyed or eyeless larvae), but often did result in formation of supernumerary eyes. The occurrence of larvae with three or more eyes ranged from one quarter of experi- ments in which 1A or 1C was deleted, to 77% after deletion of all four macromeres. Normal morphology was compatible with loss of no more than one macromere, while deletion ol three and four macromeres re- sulted in almost all of the larvae exhibiting the "swollen syndrome," characterized by spherical shape, abnormal tissues, and fluid accumula- tion. Following deletion of I A or I C, 92% of the larvae exhibited bilat- eral symmetry, although none did when three or four macromeres were killed. These results suggest the presence of an inhibitor of eye forma- tion in the vegetal region that is more likely to be localized in the 1 B or 1 D cell than in 1 A or 1 C. Results also indicate that normal morphology requires the presence of at least three macromeres and provide evidence that macromere-micromere interactions are involved in the determina- tion of embryonic symmetry, in which the B or D quadrant is more likely to become dorsal than A or C. Scanning electron microscopic studies of four and eight-cell embryos suggest that the surface of one of the macromeres is smoother (less blebbed) than that of the other three, which may be associated with localization of morphogenetic determinants. This work was supported by NSF grant DCB-88 17760 and a grant from Earthwatch. Involvement of Ca2+ channels in 5-hydroxytryptamine- induced oocyte maturation in Spisula. A. L. KADAM (The Population Council), P. A. KADAM, S. J. SEGAL, ANDS. S. KOIDE. Serotonin (5-hydroxytryptamine, 5-HT) induced in vitro maturation of Spisula oocytes. The present study was carried out to determine whether extracellular calcium is essential for 5-HT induction of matu- ration, and whether oocytes possess calcium channels regulated by the neurotransmitter. 5-HT induced germinal vesicle breakdown (GVBD) in Spisula oo- cytes suspended in ASW, but not in Ca24-free ASW. When Ca2' is added to Ca:+-free ASW to concentrations of 5, 10, 20, 30, and 50% (100% equivalent to 9.27 mA/), the percent of GVBD induced with 5- HT at a concentration of 5 nM was 2, 30.64, 86, and 96%-, respectively. The calcium channel antagonists verapamil, nitrendipme. nifedipine, nimodipine, and Cd2\ at a concentration of 50 pM. blocked 5-HT- induced maturation by 89, 18, 16, 5, and 1%, respectively. The 1,4- dihydropyridine agonist. BAY K8644. at a concentration of 1 0 ^A/. did not induce GVBD in Spisula oocytes. The capacity of 5-HT to stimulate 45Ca: " uptake by Spisula oocytes was determined. To a suspension of oocytes in Ca:+-free ASW, 45Ca2+ (0.8 ^Ci/^mole) and test substances were added. The reaction mixture was incubated at 20-22°C for 10 min; the reaction was stopped by add- ing 5 mA/ KCI buffer containing 3 mA/ EGTA, and the mixture was filtered through a GF/C filter. Radioactivity on the filter was measured using a liquid scintillation counter. 5-HT at concentrations of 0.5, 1 , 2, and 5 pM stimulated 45Ca2+ up- take by Spisula oocytes. The uptake values were 0, 5.2, 17, and 24.8 nmoles/mg protein, respectively, showing a dose response. The time course of 45Ca2* uptake at 20-22°C, showed a lag period of 2 min fol- lowed by a dramatic increase at 5 and 10 min post-treatment. Vera- pamil, at a concentration of 10 ftM, inhibited 5-HT-stimulated Ca2+ uptake. The receptor-selective 5-HT agonists, alpha-methyl-5-HT (5- HT:) and 8-OH-DPAT (5-HTIA), at a concentration of 5 nM. stimu- lated Ca;< uptake by Spisula oocytes. Mianserin (5-HT,, 5-HT:), at a concentration of 5 /iA/, blocked 5-HT-stimulated Ca2* uptake. In conclusion, extracellular Ca2t is required for 5-HT induction of Spisula oocyte maturation. 5-HT acts by opening receptor-regulated calcium channels of Spisula oocytes. BAY K 8644 was a gift of Miles, Inc. The study was supported by a grant from the Rockefeller Foundation. 5-hydroxytryptamine receptor types on Spisula gametes. P. A. KADAM (The Population Council), A. L. KA- DAM, S. J. SEGAL, AND S. S. KOIDE. Receptors for serotonin (5-hydroxytryptamine, 5-HT) are classified into various types and subtypes. To establish the biologically functional 5-HT receptor types in Spisula gametes, selective agonists and antago- nists for the various receptor types were tested for their capacity to in- fluence Spisula oocyte maturation and sperm motility. Oocyte maturation was assayed by examining the oocytes under a light microscope for dissolution of the germinal vesicle (GVBD); sperm motility was assessed by stimulation of cold-immobilized Spisula sperm. The drugs were tested at final concentration ranging from 1 to The order of decreasing potency of the 5-HT agonists to induce GVBD in Spisula oocyte was: 5-HT = alpha-methyl-5-HT = 8-OH- DPAT [8-hydroxy-2-(di-n-propylamino)tetralin]; both 2-methyl-5-HT and 5-CT (5-carboxyamidotryptamine) were inactive. The order of de- creasing potency in stimulating sperm motility was: 5-HT = alpha- methyl-5-HT = 8-OH-DPAT > 2-methyl-5-HT > 5-CT. Phenylbigua- nide and PAPP (LY-165, 163) were inactive on both gametes. The or- der of decreasing potency of 5-HT antagonists in blocking 5-HT induced oocyte maturation was: mianserin > ketanserin > s-(-)propan- olol > GR3832F > methysergide maleate. To block 5-HT stimulation of Spisula sperm motility. the order of decreasing potency of 5-HT was: mianserin > ICS-250-930 > GR3832F > ketanserin. 316 ABSTRACTS FROM MBL GENERAL MEETINGS These results indicate that Spisulu oocytes possess biologically func- tional 5-HT,A and 5-HT; receptor sites, and sperm contain 5-HT,A. 5-HT; and 5-HT, receptor sites. The presence of 5-HT,A and 5-HT: receptor sites on the oocyte membrane was validated by the finding that mianserin (5-HT, and 5-HT:)and BMY-7378 (5-HT, A) blocked 8-OH- DPAT-(5-HT,A) induced oocyte maturation, and ketansenn (5-HT:) and mianserin blocked alpha-methyl-5-HT-(5-HT:) induced matura- tion. In conclusion. Spisula gametes possess 5-HT receptors of mixed or multiple types. The 5-HT analogs were gifts from Glaxo, Sandoz, Ciba-Geigy, Rousse UCLAF, Farmatalia. Lilly, and Bristol-Myers. The study was supported by a grant from the Rockefeller Foundation. A fluorescent study of sensory neurons in normal and re- generating squid embryos. BARBARA E. MACLAY AND RACHEL D. FINK (Mount Holyoke College). We found two fluorescent lipophilic dyes to be vital markers specific for the ciliated sensory neurons of embryonic squid. The cationic mem- brane probes DiIC,s and R 1316 were used to study the appearance, distribution, and regeneration of these neurons in embryos of Loligo pea/ei. Manually dechorionated embryos and hatched larvae were soaked in a 20 Mg/ml solution of dye for 3 mm, rinsed in seawater, and viewed with epifluorescence. We followed neurogenesis in organ primordia such as tentacles and fins. After 2-5 days of outgrowth and differentiation of these structures, fluorescent staining revealed popula- tions of peripheral neurons and an extensive network of axonal projec- tions. All stained neurons had nonmotile cilia; those on the cells of the tentacles were numerous and long, whereas those in the fins were few and short. The presence of cilia on these cells may have facilitated up- take of the dyes, explaining their selectivity. By hatching, ciliated neu- rons were seen on most regions of the larva, including the epithelial lines on the head, along the mantle edge and surface, and on the siphon. We established that larval squid completely regenerate fins 5-7 days following surgical removal. Soon after the appearance of any regener- ant tissue, fully differentiated sensory neurons, complete with nonmo- tile cilia, can be stained. This differs from normal fin development where no staining of neurons can be seen until 3-4 days after first out- growth. Possible models to explain this rapid ennervation of regener- ated tissue include migration of pre-existing neurons from neighboring tissue, dedifferentiation and redifferentiation of cells at the wound site, and the presence of a stem cell population that differentiates into cili- ated sensory neurons. This work was supported by Steps, R. D. Allen, and S. W. Kuffler Fellowships of the MBL, an MHC Howard Hughes Summer Research Fellowship, and a William and Flora Hewlett Foundation Grant of Re- search Corporation to R.D.F. Afetalloproteinases of sea urchin embryo and sponge: de- tection by gelatin-substrate polyacrylamide gel electro- phoresis. JAMES P. QUIGLEY (SUNY, Stony Brook) AND PETER B. ARMSTRONG. Tissue remodeling is a conspicuous feature of invasion, inflamma- tion, wound repair, and embryonic and larval morphogenesis. Protein- ases have been implicated in the degradation and turnover of extracel- lular matrix macromolecules presumed necessary for remodeling. Of particular interest are the enzymes responsible for the degradation of collagen, the principal structural element of most extracellular matri- ces. Two morphogenetic systems have been investigated for the pres- ence of collagenases: the developing Arbucia embryo, and reaggregating Microciona tissue. In the first system, processes of cellular ingression, cell migration across basal lamellae, and invagination are prominent. In the second system, dissociated cells reaggregate, then sort out to re- constitute normal tissue morphology. Collagenases were identified by subjecting tissue extracts and conditioned seawater to SDS-polyacryl- amide gel electrophoresis (non-reducing conditions) on gels containing gelatin. Enzymatic activity was detected as bands of clearance of the gelatin from the polyacrylamide gel matrix following removal of the SDS. incubation in buffers containing calcium (room temperature, 36- 48 h.), and staining with Coomassie blue. Arbacia embryos showed three prominent gelatinases that migrated with approx. molecular masses of 65, 45. and 40 kDa. Gastrula and prism stage embryos re- leased gelatinases of approx. 70 and 80 kDa into the seawater. The ac- tivity of the 65 and 40 kDa tissue gelatinases increased, and that of the 45 kDa tissue gelatinase decreased, during development. Reaggregating Microciona tissue released several gelatinases into the seawater; one of them, of approx. 85 kDa, was the most prominent. All of the Arbacia enzymes and the 85 kDa Microciona enzyme were EDTA-sensitive and PMSF-insensitive. Some of the low molecular mass Microciona gelati- nases were EDTA-insensitive and PMSF-sensitive. Noneofthe.-frton'a enzymes had detectable casemolytic activity. Supported by NIH grant No. GM35 1 85. Comparative aspects of gossypol action. SHELDON SEGAL (The Rockefeller Foundation) AND HIROSHI UENO. Gossypol (gp) has both anti-sperm and anti-virus activity in vitro (Polsky et ai 1989, Contraception 39: 579-587). Consequently, gp may have an application in humans as an active ingredient of a vaginal pro- tective cream (VPC). Factors influencing the use of gp in this manner were studied. The biological assay employed was the effect on motility and fertilizing capacity of Arbacia sperm (Segal el al. 1985, Bio/ Bull 169: 543-544). Biological activities of gp solutions in filtered seawater, prepared from dilutions of gp in two vehicles, ethanol (roh) and cyclo- dextran (cd), were compared. Gp/roh concentrations of 25 \tM or higher are 100% effective in destroying the fertilizing capacity of Ar- bacia sperm within 3 min. Sperm treated with 50 i*M or 25 pM gp/cd retain about 30% fertilizing capacity. The anti-oxidant glutathione (gt) is proposed as an additive to a VPC in order to extend its shelf-life under non-refngerated, non-lightproof conditions. In the Arbacia sperm test, gt at a concentration of 200 \iM has no effect on sperm motility. nor does it reduce the activity of solutions of gp/roh. The minimal effective dose for 100% inhibition of motility is higher for human sperm than for Arbacia sperm. Twenty-five pM gp/roh is required to inactivate fully a normal human sperm sample within 15 min. a time at which control samples retain their motility. Immobilization of human sperm within 3 min is achieved with a gp/roh concentration of 200 pM. Gp/roh re- tains full spermicidal activity when exposed to ethylene oxide under conditions used for gas sterilization. These observations suggest that a VPC containing gp/roh at a concentration of 200 nM or higher, incor- porating the anti-oxidant gt, and prepared in an ointment base vanish- ing cream that can be sterilized with ethylene oxide, may be an effective spermicidal/virucidal modality. Interphase paniculate tiibulin revisited. KATHY A. Su- PRENANT (University of Kansas, Lawrence). I have repeated the experiments of Weisenberg (J Cell Biol. 1972, 54: 266-278) in order to identify the source of the paniculate and sedi- mentable pool of tubulin described in surf clam (Spisula solidissima) oocytes. Spisula oocytes were homogenized at 2 PC in 10 volumes of 1 ,U hexylene glycol, 0.01 M potassium phosphate at pH 6.2(HGl)and centnfuged at 5000 rpm (JA-20) for 30 min at 4°C through a cushion of 10% (w/v) sucrose in HG1. Tubulin was extracted from the pellet with 2 volumes of ice-cold 0.1 M KC1, 0.01 M potassium phosphate, and 0.2 m.UGTP at pH 7.0, and the sample was clarified by centnfuga- GAMETES AND DEVELOPMENTAL BIOLOGY 317 tion at 1 0,000 rpm for 1 0 min. The tubulin in the soluble and sediment- able fractions was analyzed by a quantitative immunoblot with a monoclonal antibody against «-tubulin (DM- la). Approximately 6- 10% of the total tubulin in the oocyle sedimented at low g forces under these extraction conditions. The "tubulin-containing structure." de- scribed by Weisenberg as a 10-20 pm granular sphere, was identified by phase and differential interference contrast microscopy as the nucle- olus, but is not the source of the paniculate tubulin. The sedimentable tubulin fraction comprised short microtubules (5-10 /im) associated with membranes, and an amorphous granular material, as well as an aggregated and unidentified form of tubulin. The amount of paniculate tubulin was determined during the first meiotic cell cycle following par- thenogenetic activation with K.CI. Extracts were prepared at 3-min in- tervals and analyzed for soluble and paniculate tubulin by the immu- noblot assay. The total amount of paniculate tubulin decreased by 30% during germinal vesicle breakdown and formation of the first meiotic apparatus. This decline in the sedimentable tubulin fraction dunng the first meiotic cell cycle is presently inexplicable. Nicotinamide suppresses Arbacia punctulata develop- ment. WALTER TROLL (New York University Medical Center) AND GERALYN CORCORAN. Nicotinamide is one of a group of compounds that inhibits the for- mation of poly(ADP)ribose (PADPR), a polymer formed by dividing cells. The role of PADPR in cell division and differentiation can be studied by observing the effects of inhibitors of its formation. In this study, the effect of three PADPR inhibitors, nicotinamide, benzamide, and 3-aminobenzamide, on Arbacia pttnctulata development was in- vestigated. We noted, when these PADPR inhibitors were added imme- diately after fertilization of Arbacia. that differentiation to plutei was blocked 2 days after fertilization, and that normal division proceeded to gastrula over the first 20 h. The complete blocking of a differentiation step suggests that a specific piece of information on DNA has been de- leted and is responsible for this differentiation. Specific deletion of on- cogenes in NIH-3T3 cells has been noted on addition of nicotinamide and other PADPR inhibitors (Nakayasu el al. 1988, Proc. Natl. Acad. Set. USA 85: 9066-9070). The blocking of differentiation mimics the deletion of genetic information provided by an oncogene and may serve as an assay to test other substances that may be capable of interfer- ing with cancer development. Supported by NIEHS Center Grant ES 00260 and Superfund Grant 1 P42ES 048995. Binding ofgossypol and its analog to sperm proteins from Arbacia, Chaetoptems, fl«200 kDa, is only partially solubilized with NP- 40 (which leaves part of the membrane skeleton intact), but is almost fully solubilized with octyl glucoside, indicating that it is an a.xoneme- associated transmembrane linkage similar to that reported in Chla- mydomonas moewusn flagella (Bloodgood 1988, / Cell Sci. 89: 521- 531). Two other proteins of lower molecular weight ( 1 40 and 44 kDa) behave similarly. These biotinylated proteins represent a small fraction of the ciliary membrane proteins, suggesting that they are either derived from a subpopulation of cilia, or correspond to transmembrane ele- ments that do not occur along the whole length of the axoneme. Mem- brane tubulin subunits of cilia labeled in vivo, or labeled after isolation, are not biotinylated, even though labeling of the axoneme in the latter case indicates that the reagent can enter the penaxonemal space. There- fore, these tubulin subunits must be integral to the membrane, rather than peripheral proteins or proteins derived from the periaxonemal matrix, and they are not exposed at the membrane surface. Supported by USPHS Grant CM 20,644. Near-UV effects on the thymidine incorporation into dog- fish lens. SEYMOUR ZIGMAN, KRIS LOWE, AND NANCY S. RAFFERTY (University of Rochester School of Medicine & Dentistry, Rochester, New York 14642). The question of how near-UV radiation in the environment can damage the ocular lens so as to positively influence opacities has been considered from the standpoint of protein anomalies, but not from the standpoint that DNA may be damaged. The hypothesis that DNA is susceptible to such UV energy in the most anterior region, the lens epithelial cells, was tested by this work. Fresh dogfish (Mustelus canis) eyes were dissected to remove the cornea, and put into Ringer's me- dium so that the lens epithelium was facing upward toward a UV lamp emitting maximally at 365 nm with an intensity of 5 mW/cm: for up to 22 h. Control eyes were treated and incubated similarly, but without UV-exposure. Observations of the presence or absence of visible opaci- ties were made, as were histological observations, using lenses fixed in formalin:glutaraldehyde. The incorporation of 'H-thymidine into DNA was also measured. At 5 h of incubation, no changes in the above- stated parameters were observed. At 22 h, mild opalescence was noted in the anterior superficial cortical region only of UV-exposed. but not control, lenses. Both pyknotic and swollen epithelial cells were ob- served. In numerous experiments, the UV-exposed epithelia incorpo- rated significantly greater amounts of thymidine than controls. H:O; production did not occur. This finding applied, whether the radiola- beled DNA was isolated using Qiagen resins, or if it was TCA precipi- tated and extensively washed. Qiagen resin tubes and agarose A50 M columns were employed to obtain DNA essentially devoid of RNA and protein. No molecular weight change was observed. Two possible ex- planations for these observations are proposed: ( 1 ) Swollen and pyk- notic epithelial cells may result from osmotic insult due to UV-induced inhibition of NaK-ATPase (as observed in mammalian lens epithe- lium). (2) Thymidine incorporation is enhanced as the result of the repair of single strand breaks in DNA (as observed in tissue culture mammalian lens epithelial cells exposed to H;O:). Support: N.E.I. (Ey00459); Research to Prevent Blindness, Inc.; Mullie and Pledger Funds (University of Rochester). Neurobiology Functional and structural consequences of activation of protein kinase C (PKC) and injection ofG-protein sub- strates ofPKC in Hermissenda neurons. D. L. ALKON, C. COLLIN, I. LEDERHENDLER, R. ETCHEBERIGAR- RAY, P. HUDDIE, M. SAICAKJBARA, S. REDLICH, E. YAMOAH, A. PAPAGEORGE, T. NELSON (Lab. Molec. Cell. Neurobiol., NINDS-NIH, Bethesda, MD). In both Hermissenda and rabbit hippocampus neurons, PKC trans- location accompanies and is probably responsible for long-lasting re- duction of current flow through la and Ic K* channels (Alkon 1989, Sci. Am. 7: 42-50). Persistent memory-specific changes of Hermis- senda neuronal branches were also closely correlated with memory ac- quisition and K* current reduction. Finally, a 21 kDa G-protein (CP20). a substrate for PKC in Hermissenda Type B cells, underwent learning-specific changes in phosphorylation (Neary el al. 1981. Na- ture 293: 658-660; Nelson el al. 1989, Bioessays 10: 75-79). Here we show that, within 20 min of exposure to phorbol ester (PDBLI) in the presence oflight. Type B soma area increased 10. 1 ± 3.4% (n = 8), and projections ("blebs") consistently appeared on the soma surface, but not during exposure to inactive 4a-PDBLI with light (n = 6). or PDBU in darkness (n = 9) (ANOVA. P < .001). Type B cells injected with Ni4+/lysine showed reduced volume of terminal branches (P < .001, n = 9) after the PDBU/light treatment, but not after the control treat- NEUROBIOLOGY 321 ments. Iontophoresis of CP20 potently reduced la by 40 ± 6% and Ic by 53 ± 9%, P < .001, n = 7), but heat-inactivated CP20 (n = 8) or vehicle injections (n = 8) did not. Iontophoresis ofv-ras reduced la by 50 ± 18% and Icby 72 ± 26%. P< .001, n = 7, whereas c-ras injections increased the same K* currents by 43 ± 9% and 63 ± 12%. P < .001 . n = 7. Vehicle injections were without effect. Thus, the G-protein sub- strates of PK.C, when phosphorylated, may trigger molecular cascades with far-reaching structural and functional consequences in learning, development, and carcinogenesis. What is the origin of photoreceptor noise'1 ROBERT B. BARLOW AND EHUD KAPLAN (Marine Biological Lab- oratory). Photoreceptors are noisy in the dark. In the absence of light, both vertebrate and invertebrate photoreceptors produce discrete waves (quantum bumps) similar to those evoked by photon absorptions. The dependence of the rate of spontaneous bumps on temperature has been taken as evidence that they are produced by thermal isomerizations of rhodopsin. However, photoreceptor noise in the Limiilus lateral eye can also be modulated by a circadian clock located in the brain. At night, efferent signals from the clock reduce the rate of spontaneous bumps without affecting those triggered by light. How can neural activity change the effects of thermal energy without influencing those initiated by light? We investigated this question by measuring the effect of temperature on spontaneous activity of retinu- lar and eccentric (second order) cells in the Limn/us eye. Action poten- tials were recorded extracellularly from eccentric cells both day and night ;/; situ, whereas quantum bumps were recorded intracellularly from retinular cells in the excized eye preparation, isolated from the circadian clock. We calculated the activation energies for eliciting spon- taneous events from the Arrhenius relationship (log event rate vs. in- verse absolute temperature). We found that the average activation energy for eliciting action po- tentials from eccentric cells was 26.3 ± 7.8 kcal/mol (n = 10) during the day and 27.9 ± 6.5 (n = 12) kcal/mol at night. The average activa- tion energy for eliciting quantum bumps from retinular cells in vitro was 26.5 ± 7.5 kcal/mol (n = 8). Our results indicate that the circadian clock does not influence the energy required to elicit spontaneous events from the Linmlus retina. Isomerization of rhodopsin by light requires energies >45 kcal/mol, which is more than twice that required for the thermal production of quantum bumps by photoreceptors and action potentials by eccentric cells. Therefore, the spontaneous events cannot be caused by thermal isomenzation of rhodopsin. Supported by NSF grant BNS-8709059 and NIH grants EY-00667 and EY-4888. G-Proteins modulate calcium currents in Paramecium and Helix neurons. JUAN BERNAL (University of Con- necticut, Farmington, CT) AND BARBARA EHRLICH. Previously we reported that the calcium-dependent swimming be- havior and the calcium action potential in Paramecium are modulated by G-proteins (Mcllveen el al 1987, Biol. Bull. 173: 445; Bernal and Ehrhch 1988, Biol. Bull. 175: 314). To test the hypothesis that both of these changes are due to modification of calcium channels, we mea- sured the calcium currents in Paramecium calkinsi. In addition, the effects of G-proteins in Paramecium and in Helix aspersa neurons were compared. Although G-proteins modified the calcium currents in both cell types, we were surprised to find that the modifications were in op- posite directions. To isolate the calcium currents, the cells were super- fused with sodium-free solution containing potassium channel blockers (125 mAI tetraethylammonium chloride, 5 mAI 4-aminopyridine. 5 mAt 3,4-diaminopyndine and 10 mAI CsCI). The cells were studied using a two-microelectrode voltage clamp. The cell was held at -40 mV (Paramecium) or -50 mV (Helix) and depolarizing command pulses were applied to elicit the calcium currents. The compounds of interest were injected into the cell by pressure, with fast green as a dye indicator to ensure that the drug entered the cell. The effects of the injected compounds were studied only in those cells in which the hold- ing and leakage currents were unaltered by the injection. We found that GTP-yS, an analogue of GTP which binds to and activates G proteins, enhanced the magnitude of the calcium current in Paramecium by 20- 90% (mean = 40%). GTP^S reduced the calcium current in Helix neu- rons by the same amount. GDP0S, which binds to and inactivates G- protems. had the opposite effect of GTP-yS in Paramecium as well as in Helix. These results demonstrate that "T-type" calcium channels in Paramecium may be activated, whereas "L-type" calcium channels in Helix neurons may be inhibited by G-proteins. J.B. is a Fellow of the American Heart Association. Connecticut Affiliate. B.E.E. is a PEW Scholar in the Biomedical Sciences. Preliminary molecular structure of FTX and synthesis of analogs that block ICa in the squid giant synapse. B. CHERKSEY, R. LLINAS, M. SUGIMORI, AND J.-W. LIN (Dept. Physiology and Biophysics, NYU Medical Cen- ter, New York, NY 10016). FTX, a specific P channel blocker, is one of many channel blocking factors contained in the venom of the American funnel web spider. We have previously reported (Cherksey el al. 1988, Biol. Bull. 175: 304; Llinas el al. 1989, PNAS 86: 1686) on the use of FTX to construct an affinity gel for the isolation and characterization of P-type Ca++ chan- nels from squid optic lobe and mammalian CNS. Purification and structural analysis of FTX have been performed. FTX could not be adequately purified by reverse phase HPLC using acetomtnle: water gradients. FPLC on Superose indicated that FTX was of low molecular weight (200-400 Da), but did not effect an adequate purification. Anion exchange methods were ineffective. However, cat- ion exchange on Mono S permitted a high level of purification of FTX, with elution of the active factor at approximately 0.8 A/ NaCl. Purified FTX exhibited a sharp UV absorption at 220 nm. No ring (aromatic) structure was detected. The absorption at 220 nm showed a pro- nounced shift with acidification suggesting that FTX possesses a titrat- able amine group. FT-1R (Fourier transform infrared spectroscopy) in- dicated the presence of C-C, C-N, N-H, C-H. and the absence of C=O, absorptions. These results ruled out the possibility that FTX is a small peptide and suggest that it is a polyamine. The known polyamine gluta- mate channel blockers (which contain ring structures) are ineffective as presynaptic blockers. On the basis of these results, model compounds were constructed with the general structure of arginine-polyamine: H,N These compounds exhibited the selectivity of FTX, but not its potency. Compounds of the structure arginine-polyamine-arginine were in- effective as blockers. Thus, the free terminal amine is critical for effi- cacy, perhaps being the moiety that actually enters the pore of the chan- nel. The arginyl group, perhaps via its strong charge, may act to secure the toxin in the channel. Therefore, the difference in potency between FTX and the model compounds may be due to the negative charge on 322 ABSTRACTS FROM MBL GENERAL MEETINGS the carbonyl of the latter, which, on the basis of FT-IR spectra, is absent from FTX. Arterial perfusion of FMRFamide-related peptides po- tentiate transmission at the giant synapse of the squid. G. A. COTTRELL, E. STANLEY, M. SUGIMORI, J-W. LIN, AND R. LLINAS (Department of Physiology and Biophysics, NYU Medical Center, New York, NY 10016). Peptides of the FMRFamide (Phe-Met-Arg-Phe-NH:) family (Price el al. 1987, Zoo/. Sci. 4: 395) have several different actions on mollus- can neurones, and the naturally occurring N-terminally extended forms can have different actions to the tetrapeptides on specified snail cells, suggesting the presence of more than one type of receptor (Cottrell and Davies 1987. J. Physiol. 382: 51). Last year we observed that FLRFamide (L = Leu), micro-injected within the squid stellate gan- glion, increases the rate of rise and amplitude of the EPSP and EPSC at the giant synapse in the absence of any observable effect on either the pre-synaptic spike or resting post-synaptic current (Cottrell el al. 19&9,J. Physiol. 412: 64P). We have now obtained a similar response by arterial perfusion of the ganglion with solutions containing FLRFamide, using the method of Stanley and Adelman ( 1984, Biol. Bull 167: 467), and also by passing FLRFamide solutions directly into the stellate ganglion artery. The threshold for potentiation is less than 10 pM. SDPFLRFamide (S = Ser, D = Asp, P = Pro) has a similar effect, but we have been unable to observe any response with either Leu-enkephalin or the cephalopod peptide eledoisin. With both FLRFamide and SDPFLRFamide, poten- tiation of the EPSP usually wanes on exposure to the peptide for more than a few minutes. Another general feature is that the effect is more pronounced as the synapse is fatigued to a steady state by repeated high frequency stimulation. Potentiation with FLRFamide under these con- ditions can be greater than 2.5-fold. We must now determine which of the FMRFamide-related peptides occur in Loligo pealeii, and establish their modes of action in potentiating transmission at the giant synapse. Calcium currents recorded in cells of anterior pituitary slices using the patch clamp technique. SUSAN A. DE- RIEMER, MEYER B. JACKSON, AND ARTHUR KON- NERTH (Max-Planck-Institut fur biophys. Chemie, Gdttingen, FRG). The patch-clamp technique was applied to analyze Ca* '-currents in cells found in slices of the rat anterior pituitary. Pituitary glands were removed from 15-30 day old rats, and ultrathm (60-100 ^m) slices were cut on a vibratome. Single cells were visualized with an upright microscope equipped with a long-distance water immersion objective. In the most commonly observed cell type (80-90% of all cells), even at very negative holding potentials (-120 mV). only non-inactivating Ca"-currents. similar to the high-voltage activated or L-type current, could be evoked. Substitution ofCa1 * by Ba** increased the amplitude and shifted the peak current from 20 to 10 mV. Addition of 20 nM Cd" blocked the current, while 100^,1/Ni'* was ineffective. The dihy- dropyridine agonist BayK.8644 increased the current, while nimodi- pine and nifedipine blocked it partially. A partial block was also seen with u)-conotoxin (50 itM). A second class of larger cells ( 10-20%) was observed with both the non-inactivating and an additional, transient Ca++-current which was activated at holding potentials more negative than -80 mV (Carbone and Lux 1984, Biophys. J. 46: 413-418; Arm- strong and Matteson 1985, Science 227: 65-67; Nowycky et al. 1985, Nature 316: 440-443). Immunocytochemistry, using antibodies to growth hormone (GH). prolactin (Prl). and lutenizing hormone (LH), showed that the primary cell type present in animals of this age contains growth hormone. Both prolactin and lutenizing hormone positive cells were distributed much more sparsely, but of the two, the Prl cells were larger. These results confirm and extend data obtained on cells in primary culture (De- Riemer and Sakmann 1986, £.\/> Brain. Res. Ser. 14: 139-154.) Use of this version of the patch clamp method, in combination with immunocytochemistry in situ, extends the possibilities for understand- ing the significance of functional differences in different cell types in regulating secretion and how the local environment of cells within the pituitary affects their responses. Supported by a fellowship from the MBL, and grants from the March of Dimes and NSF (S.A.D.); DFG (SFB 236) (A.K.). Properties of detached nerve terminals from skate electric organ: a combined biochemical, morphological, and physiological study. M. DOWDALL, G. PAPPAS, AND M. KRIEBEL (SUNY, Health Science Center, Syra- cuse, NY 13210). The electric organs of the skate (Raja erimicea) dissociate into func- tional electrocytes when incubated with 1 % ( w/v) collagenase. Dissocia- bility of tissue is time and temperature dependent, and electrocytes with varying degrees of innervation can be produced according to con- ditions. At room temperature (26°C) most electrocytes are denervated after 6 h of treatment. These have normal resting potentials (-50 mV) and show a normal ultrastructure. After 2-4 h of treatment, innerva- tion is present, and normal miniature end-plate potentials (MEPPs) and end-plate potentials (EPPs) can be recorded. Electron microscopy shows loosely adherent nerve terminals and detached terminals that can be concentrated by differential centrifugation. The Schwann cells migrate over newly exposed nerve terminal surfaces to encapsulate free terminals. Incubation at 6°C retards dissociation by 40-fold, although the time characteristics of the MEPPs and EPPs remain normal during dissociation of normal-appearing nerve terminals from Schwann cells and electrocytes. After 4 days of collagenase treatment at 6°C, electro- cytes were washed several times, and isolated nerve terminals appeared in subsequent saline suspensions while MEPP frequencies fell and the denervated surface increased. Biochemical measurements, with cho- line acetyltransferase activity as a diagnostic nerve terminal marker and acetylcholinesterase activity as an auxiliary marker, show that termi- nals became detached from electrocytes after 3-4 days at 6°C. This pro- cedure produces isolated endings up to 5 p.m in diameter, which are recognizable, with Nomarski optics, by their size and included station- ary particles (mitochondria). Moreover, these nerve terminals are readily distinguished from nucleated Schwann cells. Supported by NSF 19694 and NIH NS25683. I 'erv high resolution and dynamic stereo images of neu- rons. SHINYA INOUE (Marine Biological Laboratory), TED INOUE, ROBERT A. KNUDSON, AND RUDOLF OL- DEN BOURG. We have devised a method for obtaining very high resolution stereo- scopic images. A stack of 81 serial optical sections of Golgi-stained mouse neurons were recorded in 0.5-^m steps using a 100/1.35 NA Plan Apo lens (condenser NA = 1. 1 ). In the Image-I image processing computer, each image in the stack was compressed and sheared later- ally by appropriate amounts in order to generate rotated "image cubes." The angles of rotation were chosen to provide the parallax needed for stereo imaging. To retain a clear sharp view of the large number of optical sections making up each cube projected into a single plane, a sharpening convolution was applied to each section, and then NEUROBIOLOGY 323 the minimum of each pixel gray value in adjoining sections was calcu- lated. Images for the left and right eyes thus generated were placed at half height with the left image above the right in the video frame. To view in stereo, the left and right images were re-expanded vertically and projected as left- and right-circularly polarized images alternating at 1 20 Hz with a StereoGraphics projector. Complementary polarizing glasses worn by the viewers provide each eye with the left or right stereo image 60 times a second, giving nse to flicker-free stereoscopic images of an approximately 40-^m cube. We demonstrated very high resolu- tion, detailed arrangements of neuronal spines on the dendrites. Rotat- ing stereo views of the neuron at moderately high resolution (40/0.95 NA) were also demonstrated. Supported by grants NIH R37 GM 3 1 6 1 7 and NSF DCB 85 1 8672. Biphasic modulation of calcium-dependent potassium current in pituitary tumor cells examined with the per- forated patch clamp technique. RICHARD H. KRAMER (Columbia University) AND EDWIN S. LEVITAN. The pituitary tumor GH3 cell line is a model system for studying the actions of secretogogues. such as thyrotropin releasing hormone (TRH), which stimulate phosphatidylinositol hydrolysis and the for- mation of the intracellular messengers inositol trisphosphate (IP,) and diacylglycerol. However, the electrophysiological effects of TRH have been difficult to study with the whole cell patch clamp method because the hormone response rapidly "washes out" of cells. To alleviate this problem, we have used the perforated patch configuration; the iono- phore nystatin, contained in the patch pipette, is inserted into the mem- brane patch, providing a low-resistance pathway into the cell without causing washout of intracellular constituents. Using this configuration, we have elicited stable biphasic TRH responses (hyperpolarization fol- lowed by hyperexcitability). Under voltage clamp, the responses are characterized by an initial large increase in the outward current elicited during depolarizations, followed by a suppression of the outward cur- rent. The outward current, and both phases of the TRH response, are blocked by exposing the cell to the membrane-permeable calcium che- lator BAPTA-AM, or by addition of charybdotoxin plus apamin, sug- gesting that modulation of Ca-dependent K current accounts for both phases of the TRH response. The increase in Ca-dependent K. current is thought to be due to IP,-gated Ca release from intracellular organ- elles. We suggest that the subsequent decrease in this current is indirect, due to TRH-induced inactivation of voltage-gated calcium channels (which we have observed) and consequent reduction of the calcium transient during depolarizations. In support of this hypothesis, we find that nimodipine. a dihydropyridine Ca channel blocker, selectively eliminates the suppression phase of calcium-dependent K current mod- ulation. Ontogeny ofserotonergic neurons in Hermissenda: a pre- liminary study. EBENEZER YAMOAH, ALAN M. Kuzi- RIAN (Marine Biological Laboratory), AND CATHER- INE TAMSE. The nudibranch mollusc Hermissenda crassicornis is now in culture. As part of the effort to establish this invertebrate as a model system for neurobiological research, we are beginning an ontogenetic study of its nervous system. Applying function to morphology, we first undertook an immunocytological study to identify the developing serotonergic neurons in Hermissenda. We have observed that serotonin plays a ma- jor role in regulating feeding in adults, but one of our concerns in rear- ing this nudibranch is to maximize nutrition at all stages of develop- ment. Therefore, we have planned to investigate the role of these cells in feeding, and also to use them as markers documenting the developing nervous system in larvae and juveniles. Their putative role in metamor- phic induction will also be examined. Larvae raised from eggs that had been laid in the laboratory by field collected animals, were sampled on larval days: 1,7, 14, 2 1,28, and 35. We fixed the larvae in 4% paraformaldehyde and processed them for immunocytochemistry using a goat antibody to serotonin (INCSTAR); this was followed by rabbit anti-goat biotinylated secondary antibody and labeling with strep-avidin conjugated FITC (Vector Labs). Juve- niles (metamorphic day-6), and adult central nervous systems (CNS) were treated similarly. Observations and photo documentations were done using epifluorescence microscopy. Areas of anti-serotonin labeling were present in larvae at hatching (larval day- 1); moreover, during development, these areas became con- solidated into easily recognizable cells (day-7. ! 4) and finally into defi- nite ganglia (day-21, 28. 35). Specifically, in the early larval stages, the staining occurred in the velar region and in association with the velar and larval retractor muscles. A large serotonergic cell occurred in the metapodium (day-28) and, by day-21, several positively staining cells appeared in the vicinity of the newly formed eyes. Byday-35 (metamor- phically competent larvae), there were clearly identifiable ganglia ven- tral to each eye (putative cerebropleural ganglia). Juvenile Hermis- senda showed characteristic and numerous, scattered serotonergic sen- sory cells (including some proximal axons) in the outer epithelium, including the oral tentacles, rhinophores, cerata, and tail. Staining for serotonin-containing cells in the adult CNS was only minimally suc- cessful, although we used previously proven methods and had obtained positive results with the larvae. The same minimal staining was ob- tained in both field-collected and lab-reared adults. The findings are suggestive of a seasonality associated with the levels of serotonin in adult Hermissenda, and further study of this possibility is being under- taken. This research was aided in part by the microscopic facilities spon- sored by Olympus and Zeiss at the MBL and supported by a grant to A.M.K.(RR03820. NIH). FTX blocks a calcium channel expressed by Xenopus oo- cytes after injection of rat brain mRNA. J.-W. LIN, B. RUDY, AND R. LLINAS (Dept. Physiology and Bio- physics, NYU Medical Center, New York, NY 10016). Funnel-web spider toxin (FTX) has been shown to block Ca conduc- tances in cerebellar Purkinje cells and the presynaptic terminal of the squid giant synapse (Llinas et ai 1989, PNAS 86: 1689). To further characterize this toxin, we studied its effect on a calcium current (ICa) appearing in Xenopus oocytes after injection of rat brain RNA. This current is insensitive to organic calcium channel blockers; i.e.. dihydro- pyridine or u)-conotoxin (Leonard el al- 1 987, J. Neiirosd. 7: 875). Two aspects of the effect of FTX on the expressed lca were examined: ( 1 ) the calcium-activated chloride current (Iaicai): ancl <-) currents carried by barium ions through calcium channels (IBj). In the presence of 1 .8 mAl extracellular calcium, depolarizing pulses activated a mixture of inward and outward currents in the injected oocytes. Following the termina- tion of the pulses, a prolonged tail of the Cl current could be recorded with its amplitude reflecting the magnitude of Ca influx activated dur- ing the depolarizing pulses. Iaica> tail current exhibited a threshold of -40 mV and reached maximal amplitude between 0 and +10 mV. FTX partially (57%, n = 7) and irreversibly blocked the calcium acti- vated chloride current without changing its voltage sensitivity, and had a minimal effect on INa and IK. This block was concentration depen- dent; an EDW was obtained at 1 500 >, dilution of the crude venom, and the maximum blockade was typically achieved at 600x dilution (see also Llinas el al. 1989, these Abstracts). In the presence of 60 m/U Ba+ + and blockers of sodium and potassium currents, a voltage-activated 324 ABSTRACTS FROM MBL GENERAL MEETINGS inward current was recorded. This current (IBa), presumably mediated by calcium channels, showed little inactivation, had a higher threshold (-20 mV), and reached its peak amplitude between +20 to +30 mV. FTX blocked 1BJ partially and did not alter the time course of the cur- rent or its I-V characteristics. Furthermore, the extent of FTX partial block depended upon the level of RNA purification. The block is more extensive in the !Ba expressed from poly(A)-mRNA (66 ± 9. n = 4), than that from whole brain RNA (41 ± 1 1%, n = 12). Thus, more than one population of calcium channels may exist, and they would be ex- pressed in different proportions, depending on the degree of mRNA purification. Dose-response for FTX blockade of presynaptic IICl,, in the squid giant synapse. R. LLINAS, M. SUGIMORI, J-W. LIN, AND B. CHERKSEV (Dept. Physiology and Biophysics, NYU Medical Center, New York, NY 10016). The dose-response relationship for the blocking action of FTX (a toxin fraction from Agalenopsis a/wrta venom, Sugimori el al. 1988 Biol. Bull. 175: 308; Cherksey et al. 1988, Biol. Bull. 175: 304, Llinas el al. 1989, PNAS. 86: 1689) on the voltage-dependent presynaptic cal- cium current (ICj) in the squid stellate ganglion, was determined from voltage clamp measurements. In addition to the purified toxin, we tested the raw venom, and a synthetic poly-amine with an arginine at one end. constructed on the basis of chemical analysis of the FTX frac- tion (Cherksey et al. 1989. these Abstracts). These substances were added to the bathing solution in concentrations ranging from 0.2 to 190 nl/ml (volume of venom or liquid synthetic toxin in nl/ml seawa- ter). Each of the three fractions had an ED50 of 5 nl/ml and produced a total blockade at 80-100 nl/ml. The block caused by 100 nl/ml had a time course of about 20 min and was very slowly reversible. Comparing the degree of calcium current block with the reduction of the post-syn- aptic potential, we concluded that the effect of the toxin on synaptic release is totally ascribable to its calcium blocking effect. We reached a similar conclusion about the synthetic polyamine. But because the same volume of raw venom and the synthetic polyamine produced about the same degree of block, we conclude that the active toxin may be more potent, and that the structure of the naturally occurring poly- amine is therefore probably a variant of the synthetic product. Finally, we tested some polyamines with an arginine at each terminal of the chain, but I(a was not blocked. Thus, the polyamine may require the terminal amine to penetrate the channel and produce the block; the arginine group may hold the molecule in place. Modulation of the spontaneous and evoked responses of lagenar afferent s in the toadfish Opsanus tau, by elec- tric pulse stimulation of the efferent vestibular nuclei. RACHEL LOCKE AND STEPHEN M. HIGHSTEIN (Ma- rine Biological Laboratory, Woods Hole, MA). The lagena is one of eight acoustico-lateralis mechanoreceptors in- nervated by the efferent vestibular nuclei (EVN). The ultrastructural morphology of efferent-afferent interactions consists of an efferent ter- minal on innervated hair cells and a second terminal on innervated afferents. Firing patterns of lagenar afferents encode aspects of head movement, head position, or substrate-borne vibration. Animals were lightly anesthetized with Finquel. MS222 (Ayerst), partially paralyzed by an intramuscular injection of 0.05 mg/kg of pancuronium bromide, Pavulon (Organon). placed in a lucite experimental tank atop a servo- controlled rotary table and perfused through the mouth with recircu- lated seawater at 1 5°C. Stationary animals were subjected to periods of recording of afferent activity or were oscillated with sinusoidal stimuli in the yaw plane before during and after epochs of electric pulse stimu- lation ( 1 00 Hz. 0. 1 -ms wide pulses, < 1 00 ^ A amplitude constant cur- rent) via paired silver wire electrodes inserted visually into the EVN. Afferents range up to 16 nm in diameter and can be visually identified and penetrated with glass microelectrodes near the lagenar macula. Re- cordings reveal spontaneous excitatory post-synaptic potentials (EP- SPs) and action potentials (APs). Electrical stimulation of the EVN could either increase or decrease the spontaneous or mechanically evoked firing rates of afferents (number or frequency of APs) depending upon the particular afferent studied. Effects of stimulation were gener- ally stationary; i.e.. inhibition could never be converted into excitation by any experimental manipulation, and vice versa. Single, double, tri- ple, etc. pulse stimulation of the EVN evoked monosynaptic EPSPs and APs in afferents whose spontaneous and evoked rate of APs were increased by trains of stimuli. Spontaneous miniature EPSPs from transmitter released by hair cells (EPSPs persisted when EVN axons were severed) had their amplitude and time-to-peak reduced during EVN stimulation. This reduction of "synaptic noise" from hair cells is correlated with a reduction of spontaneous and evoked firing rates. Therefore, we suggest that the EVNs exert a dual control over lagenar pri mary afferents via the dual innervation of lagenar receptors. We pro- pose that actions of the axo-axonic synapses are responsible for in- creased firing rates while the axo-somatic synapses on hair cells may be responsible for reducing spontaneous and evoked firing rates. The effects of GABA on retinal horizontal cells: evidence for an electrogenic uptake mechanism. ROBERT PAUL MALCHOW (Department of Ophthalmology, Univer- sity of Illinois College of Medicine, Chicago, Illinois 60612). The inhibitory effects of GABA on neurons are thought to be termi- nated by uptake into neurons and glia surrounding the release site. Cer- tain classes of retinal horizontal cells avidly accumulate exogenously applied GABA. The present series of experiments were designed to de- termine if an electrophysiological correlate of the process of GABA up- take could be observed in such cells. Isolated cells from the retina of the skate (Raja erinacea and Raja ocellata) were obtained by enzymatic dissociation. Voltages and cur- rents from external horizontal cells were recorded using the whole-cell version of the patch-clamp technique. GABA (500 p.M) applied via pressure ejection from pipettes placed 10-20 nm from the cell somas elicited pronounced depolarizations of the cells. When cells were volt- age clamped at -70 mV, GABA produced a slow inward current of between 200-500 pA in magnitude; similar applications of 500 nM muscimol or 1 m.M (-)baclofen were without effect. The response to GABA was not blocked by superfusion with either 500 nM bicuculhne or 500 iihl pictrotoxin. However, the response was rapidly and revers- ibly eliminated by superfusion with sodium-free Ringer. The I-V rela- tionship of the GABA response was similar to that observed for gluta- mate uptake into glial cells (Brew and Attwell 1987. Nature 327): the current decreased as cells were progressively depolarized, but did not reverse even with cells held at +70 mV. These data indicate that the effects of GABA on horizontal cells are not due to the activation of either GABA., or GABAb receptors, but rather reflect electrogenic uptake of GABA into these cells. Skate retinal horizontal cells may prove to be an excellent model system with which to study the process of GABA uptake into neurons. The author acknowledges the generous support for this research pro- vided by a fellowship from the Grass Foundation. Electric organ discharge and electrosensory reafference in the little skate, Raja erinacea. JOHN G. NEW (Dept. NEUROBIOLOGY 325 of Neurosciences, A-001, School of Medicine, UCSD, La Jolla.CA 92093). The electric organs (EO) of the little skate are located laterally along the longitudinal axis of the tail. They are innervated by a series of spinal nerves that receive descending input from the electric organ command nucleus (EOCN) situated on the midline of the rostral medulla (Szabo 1955, J. Physiol. 47: 382-385). The weak and irregular discharge of the EOs consists of a head-negative, tail-positive waveform of 10-50 mV amplitude and approximately 70 ms duration (see Bass 1986, in Elec- troreception. Bullock and Heiligenberg, eds., J. Wiley, for review). The purpose of this study was to determine the effect of electric organ discharge (EOD) upon the activity of the animal's own electrosensory system. In alert decerebrated skates, stimulation of the EOCN with a brief, high-frequency train of pulses, results in synchronous discharge of the electric organs. This evoked EOD is indistinguishable from those evoked by tactile or electric field stimuli, or the animal's spontaneous activity. Recordings from anterior lateral line nerve (ALLN) fibers innervat- ing the electrosensory ampullae of Lorenzini demonstrate different re- sponses to the EOD depending upon the orientation of the receptor and the position of the ampullary pore on the body surface. ALLN fibers, with receptive fields in the caudal third of the pectoral fins, are strongly excited by the EOD (mean max. freq. = 90 spikes/s). whereas more rostral ampullae are driven more weakly or not at all. Those fibers in- nervating rostral ampullae and exhibiting modulation are differentially excited or inhibited depending upon ampullary orientation. Analysis of the internal and external electric fields indicates that the EOD acts similarly to an externally applied dipole. producing differential modu- lation of ALLN activity. This contrasts with the common-phase modu- lation produced by ventilatory activity. Supported by a fellowship from the Grass Foundation and NIH NRSANS-08114. Sti/bene derivatives or chloride replacement by imperme- ant unions dramatically alter a late component of the light scattering change in mammalian nerve terminals. A. L. OBAID, K. STALEV, J. B. SHAMMASH, AND B. M. SALZBERG. (University of Pennsylvania School of Medicine). Electrical stimulation produces large and rapid changes in the intrin- sic optical properties of the neurosecretory terminals of mammalian neurohypophyses. These signals, measured as changes in opacity, re- flect alterations in large-angle light scattering, and comprised at least three components. The first of these, the E-wave. coincides with the arrival of excitation in the terminals; the second, the S-wave, is inti- mately related to the secretion of arginine vasopressin and oxytocin, and the third and slowest component exhibits a complex N-shaped waveform that requires seconds to return to baseline. This slow change in the transparency of the tissue suggested that volume compensation associated with chloride movement might be implicated in determining the time course of the optical signal. Chloride replacement (90%) by impermeant anions such as gluconate. isethionate. methyl-sulfate, and methane-sulfonate (Ca activity matched to that of the control Ringer's solution) resulted in the reversible loss of the N-shaped waveform and a faster return to baseline with overshoot. The same effect was obtained, although irreversibly, upon the addition of either 0.2-0.5 mA/ 4-acet- amido-4-isothiocyano-stilbene-2,2'-disulfonic acid, disodium salt (SITS-Sigma) or 0.1-0.5 mA/ 4,4'-dinitrostilbene-2,2'-disulfomc acid, disodium salt (DNDS-Molecular Probes). This result suggests that the late component of the light scattering signal from the nerve terminals of the mouse neurohypophysis is prolonged by transmembrane chloride movements which may be coupled to volume changes in this tissue. We thank Drs. Wu Jian-Young, Harvey Fishman, and M. V. L. Ben- nett. Supported by USPHS grant NS 1 6824 and by fellowships from the Nuffteld Foundation ( K.S.) and the Short Term Experience in Research Program of the NIH (J.B.S.). Imaging learning-specific changes in the distribution of protein kinaseC. JAMES L. OLDS, DONNA L. McPHiE, AND DANIEL L. ALKON (NINDS-NIH, Bethesda, MD 20892). Protein kinase C (PKC) is an important second messenger in a wide variety of physiological systems. Ongoing work in this laboratory has demonstrated a statistically significant learning-specific change in the distribution of PKC in the rabbit hippocampus. Rabbits were classically conditioned in the nictating membrane para- digm, and sacrificed either 24 h or 3 days later. Brains were prepared for quantitative autoradiography by standard methods [Olds el al. 1989, Science (in press)]. Cryosections (20 ^m) were incubated with ['H]- phorbol-12,13-dibutyrate (PDBU) at 2.5 nMfor 60 min. Non-specific binding was consistently less than 9% in all assays. We analyzed film autoradiograms of rabbit brain sections and radioactive standards using with a computerized image analysis system. Conditioned (Group C) animals showed a 49% and 43% increase, respectively, in 'H-PDBU binding (P < 0.01. one way ANOVA) in the CA1 region of the dorsal hippocampus when compared with either un- paired control animals (Group UP), or naives (Group N). 24 h after conditioning. No significant difference was observed between Groups UP and N. Higher resolution transept-line analysis established that the stratum pynmidale to stratum oriens ratio (SP/SO) was significantly increased over controls, thus showing not only a change in the amount of the enzyme, but also a change in its distribution. An analogous, sta- tistically significant decrease in SP/SO was seen in group 3-C animals, which were sacrificed 3 days after the end of behavioral training. These results suggest that, after an initial increase in soma-localized PKC (24 h after conditioning), the enzyme migrates to areas corresponding to the dendritic compartments of CA 1 pyramidal cells (3 days after condi- tioning). This conditioning-specific change represents a modification in the pattern of PKC distribution that is dependent on behavioral re- tention time. A search for correlations in the spike activity of the Aplysia abdominal ganglion during the gill withdrawal reflex. D. SCHIMINOVICH, L. B. COHEN, A. I. COHEN, H.-P. HOPP, C. X. FALK, AND J.-Y. Wu (Department of Physiology, Yale School of Medicine). We previously made optical recordings of action potential activity in the Aplysia abdominal ganglion during the gill withdrawal reflex using voltage-sensitive dyes and a 124-element photodiode array. We have now searched for correlations in the spike activity of the optically de- tected action potentials. High correlation values were given to pairs of cells maintaining exactly the same time difference between some subset of spikes in each cell. The time differences were allowed to deviate by as much as 20 ms, although such deviations would reduce the value we assigned to the correlation. Correlations found at times of high spike activity (just after the touch to the siphon) were given lower values be- cause of an increased probability that the correlations were due to chance. 326 ABSTRACTS FROM MBL GENERAL MEETINGS Several trials of spike data from seven preparations were searched for correlations. No large correlations were found. To judge whether the small correlations we found were real, we looked for correlations in a file called "random." This file was obtained by taking the experimental spike file and randomly reassigning the time position of each spike in each cell in a region of 250 ms. In this way, the overall appearance of the experimental and random files were similar, while any correlations found in the experimental file should be weakened. However, the corre- lations found in these random files were just as large as those in the experimental file; we concluded that the correlations in the experimen- tal file were due to chance. We have considered two kinds of explanations for the absence of correlations. First, the optical recordings were only 20% to 40% com- plete, and we might have observed a special subset of cells (e.g., motor neurons) that were not correlated. On the other hand, the nervous sys- tem of the Aplysia may not use many large fast interactions but, in- stead, either small fast interactions that require the activity of many pre-synaptic cells or slow interactions that would not generate precise intervals between pre- and post-synaptic spikes. Supported by NIH grant N508437. On-line rapid determination of[Ca]i by means ofFura- II and high speed video imaging. M. SUGIMORI AND R. LLINAS (Dept. Physiology and Biophysics, NYU Medical Center, New York, NY 10016). Calcium concentration changes in the soma and dendrites of mam- malian Purkinje cells during spike activity have been determined with the calcium-sensitive dyes Arsenazo III (Ross and Werman 1986, J. Physio/. 389: 319) and more recently Fura-II (Tank el at. 1 988, Science 242: 633). In the latter study, a distinction was made between calcium entry, which occurs at the onset of plateau potentials in the peripheral dendritic branchlets, and that which occurs in the main dendritic arbor. The study suggested that the spontaneous activation of Purkinje cells is initiated by an inward calcium current at peripheral dendntes which, upon reaching sufficient amplitude, evoked calcium-dependent action potentials in the main dendritic tree. The measurements reported in that study were obtained at a maximum speed of 250 ms. With im- proved high speed imaging techniques (a photon counting camera and a high speed video recording system capable of recording 2.5 ms per frame), we have recorded the actual time course and distribution of calcium entry during single action potentials in Purkinje cells. Because the fluorescence measurements using Fura-II were made at only one light frequency (380 nm), the measurements indicate only relative cal- cium concentration changes in the cell cytosol. Simultaneous record- ings of light absorption by Fura-II, and of intracellular voltage, clearly indicate that the plateau potentials preceding the activation of Purkinje cells occur in the spiny branchlets. The full action potential is then observed in the main dendrites and is followed by a synchronous, anti- dromic invasion into the fine dendritic tree. Averaging the calcium sig- nal obtained immediately after the onset of a dendritic spike, and com- paring it to that prior to this spike, demonstrates a large calcium influx at the main dendntic field during the spike. Moreover such signals indi- cate that calcium concentration transients may last for periods of 10 to 1 5 ms after which the calcium is buffered or pumped from the cytosol. Thus, [Ca]i is very actively modulated. This technique also allows visu- alization of the Fura-II response to the calcium entry in the presynaptic terminal of the squid giant synapse following a single stimulation to the presynaptic fiber. Activation of the octavolateralis efferent system in the lat- eral line of free-swimming toadjish. T. C. TRICAS AND S. M. HIGHSTEIN (Washington University School of Medicine, Department of Otolaryngology, St. Louis, MO 63 110). The octavolateralis efferent system (OES) can be activated by multi- modal sensory stimuli (Highstein and Baker 1985, / Neitrophysiol. 54: 370) and has a predominantly inhibitory action on the firing of lateral line (LL) primary afferent fibers. We studied the OES in a semi-natural setting to determine if visual stimuli could activate it. Single LL affer- ents were recorded chronically with metal microelectrodes, for periods up to 9 days, in toadfish swimming freely in a small tank monitored by a video system. Primary afferents are grouped into four classes based upon their spontaneous interspike intervals: regular, irregular, burst- ing, and silent. Silent fibers were mechanically activated by mild 25- 90 Hz vibrations of the experimental tank. Trains of bright photic stim- uli (10-100 flashes/s for 1-5 s) increased spontaneous spike rates of irregular and bursting afferents, but decreased mechanically evoked afferent firing rates of silent fibers. To test a biologically relevant visual stimulus, a small prey fish was presented in a clear sealed chamber that eliminated all other sensory cues. After presentation of the prey from behind a movable blind, the spontaneous activity of some LL fibers decreased; mechanically evoked firing in silent fibers also decreased for the duration of the stimulus (up to 1 20 s) and recovered to pre-stimulus rates when the blind was closed. Thus, the OES can modulate the activ- ity of LL afferents via visual input pathways in biologically relevant contexts. Chaotic properties of quanta/ transmission at the skate neuro-electrocyte junction. J. VAUTRIN, J. HOLZAP- PLE, AND M. KRIEBEL (SUNY Health Science Center, Syracuse, NY 13210). Spontaneous synaptic activity at neuro-electrocyte junctions has been recorded using focal and intracellular recording techniques. After electrocyte dissociation with collagenase, nerve terminals remain in physiological contact with the electrocytes (see Dowdall el at., these Abstracts). Focally recorded miniature end-plate potentials (MEPPs) from dissociated electrocytes show broad amplitude and time-to-peak distributions that are skewed towards low values. Intervals and time-to- peak plots reveal interactions between spontaneous MEPPs. Successive MEPPs often show congruent rising phases, and arise in very bursty patterns. Breaks on MEPP rising phases and changes in slope indicate a sub-structure. Many treatments increase the skew-to-bell-MEPP ratio in frog or mammalian preparations (skew class includes slow- and gi- ant-MEPPs). Temperature changes have a reversible effect on the slope of the skate slow-MEPP. Electrocytes from intact electric organs show periods of MEPPs with the same, fast time course producing bell- shaped amplitude distributions with a subunit substructure. These peri- ods of bell-MEPPs alternate with periods of mainly skew-MEPPs. The rapid shifts (seconds) between the two regimes of the spontaneous re- lease process strongly suggest that each MEPP is not due to a performed packet of transmitter, but results from a deterministic process that dy- namically combines subumts in different numbers and rates. A simple model, combining subminiature end-plate currents in different num- bers and rates, is able to simulate all the observed MEPP amplitudes and time courses. Chaos theory shows that simple systems lead to com- plicated patterns. We demonstrated that a leaking faucet shows transi- tions, from a very organized regime with drops of regular size, to insta- ble regimes of various drop sizes that are organized in bursts. Records of drops showing classes of volumes and interval structure are remarkably similar to those of MEPPs. Supported by NSF 19694 and Association Francaise contre les My- opathies. PATHOBIOLOGY AND ENVIRONMENTAL STUDIES 327 Pathobiology and Environmental Studies Shell disease syndrome in Cancer crabs. ROSEMARIE BORKOWSKI AND ROBERT A. BULLIS (University of Pennsylvania, Laboratory for Marine Animal Health, Marine Biological Laboratory). Gross examinations and microbiological investigations were per- formed on 14 Cancer borealis (Jonah crab) individuals and 1 3 Cancer irroralus specimens (rock crab) afflicted with shell disease syndrome. The crabs were collected from near-shore waters between the Chesa- peake Bay and Martha's Vineyard. Manifestations of disease varied be- tween and within the two species, indicating a potential for more than one pathogenic process. Lesions in C. borealis: focal blackening without ulceration was the most common dorsal carapace abnormality. Mildly affected animals exhibited punctiform blackening of the dorsal shell, whereas coales- cence of numerous black foci throughout at least two-thirds of this re- gion was characteristic of extensive disease. Distribution of all such le- sions ranged from markedly symmetrical to decidedly asymmetrical. Lesions of the ventral carapace and appendages included punctiform erosions and small foci of blackening with, or without, central pinpoint ulceration. Lesions in C. irroralus: the most frequent dorsal carapace abnormal- ity was focal ulceration through the exoskeleton to underlying soft tis- sues. Shell adjacent to such lesions was blackened and the underlying endocuticle discolored. This ulcerative abnormality occurred at ran- dom sites and affected less than one-third of the dorsal shell with the remaining dorsal carapace being normal. Circular erosions arranged in a linear fashion and cracking of the exoskeleton were the primary le- sions of the ventral carapace and appendages. Eight of 1 3 crabs had lost entire appendages or dactylopodites. Blackening of the gills was noted in both species, as was accumula- tion of black sediment between the gill bases and exoskeleton. In C. borealis and C. irroralus. \'ibrio spp. were the most frequent bacterial isolates from normal and affected areas of the exoskeleton: Pseudomonas spp. ranked second in number of isolates. These findings confirm earlier work regarding the genera of bacteria associated with shell disease syndrome. This study is supported in part by grants from the Division of Re- search Resources, National Institutes of Health (P40-RR 1333-09); Northeast Fisheries Center, National Marine Fisheries Service; and the Madison Trust. Shell disease in impounded American lobsters. Homarus americanus. ROBERTA. BULLIS (University of Penn- sylvania, Laboratory for Marine Animal Health, Ma- rine Biological Laboratory). Lobsters held in impoundments during the winter months for long periods can show dramatic increases in shell disease. This unsightly condition affects marketability and can spread rapidly through a popu- lation leading to wholesale losses and decreased profitability. Twenty-five lobsters were sampled from an impoundment in Gran Manan Island. Nova Scotia, during a disease outbreak. Gross examina- tion revealed diffuse pitting erosions covering the entire exoskeleton. but most prominently on the carapace and dorsal abdominal segments. Lesions of the ventral surface were characterized by foci of hyperpig- mentation associated with abrasions and scratches. Crushing injuries were manifested by cracking of the exoskeleton and severe blackening of underlying tissues. Lesions of this type are thought to be associated with trauma induced by overcrowding and poor handling. Microbiological examinations revealed that most individuals had concurrent bacterial septicemia. Microbial isolates from the hemo- lymph. internal organs, and exoskeleton included chitinolytic, lipo- lytic. or both types of bacteria from the genera Acinetobacter. Flavo- bacter. and Pseudomonas. All three genera have previously been impli- cated in shell disease outbreaks. Physiological and biochemical disturbances in chitin synthesis, and a resulting breakdown in exoskele- tal repair, were hypothesized to be the result of stress-induced immuno- suppression. The progression of the disease was altered by placing lobsters in flow- through containment with three feedings per week of fresh squid and weekly removal of excreta. After six months, lesion severity decreased in moderately affected lobsters. This work indicates that, although control of ubiquitous chitinolytic bacteria in captive populations is difficult, disease problems may be alleviated by increased attention to hygiene (removal of excreta and selective culling), proper husbandry (adequate nutrition and a clean water supply), and wound avoidance (minimized overcrowding). This study was supported, in part, by a grant from the Division of Research Resources. National Institutes of Health (P40-RR1333-09) and the generosity of Paul's Lobster Co., Boston, MA. Marketing, ecological, and policy considerations related to the New England conch fishery and Hoploplana. ILENE M. KAPLAN, BARBARA C. BOYER, AND DA- NIELA HOFFMANN (Union College). Marketing procedures and ecological factors associated with the New England conch (Busycon) fishery, and related changes in Hoploplana. a turbellarian flatworm commensal in the mantle cavity of Busvcon, are examined. Conch and pot fishermen were observed at sea on a regu- lar basis, and interviews with fishermen, seafood buyers, processors, and fish market and restaurant owners were conducted. Data on size, width, weight, sex, and number of worms in the conch also were col- lected. Conch fishermen are paid $.25-$.55 a pound for conch in the shell; markets sell conch for $.59-$1.19 per pound in the shell, and $2.75- $3.25 per pound outside the shell; and conch is sold as conch salad for $4. 99-$6. 99 per pound. The record of annual fluctuations over the past 20 years reveals an overall increase in the pounds of conch meat har- vested in New England, as well as an increase in the market value. Conch supply has become more available due to increased fishing, and price fluctuations in the last few years have become less dramatic. In- creased fishing for conch has also resulted in heightened competition between pot fishermen using traditional techniques and larger stern trawlers. A total of 741 specimens of Busycon canaliculalum were examined (466 females and 264 males). Approximately 20% yielded 223 worms, 154 from females and 69 from males. Further monitoring of the conch fishery is suggested for future re- search and policy considerations. This research was supported by a grant from Earthwatch and Union College/Dana Fellowships. The authors also acknowledge the support of the Marine Policy Center, Woods Hole Oceanographic Institution, and the Marine Biological Laboratory. The sperm cell's silent spring: herbicides and pesticides. LEONARD NELSON (Medical College of Ohio, Toledo, OH 43699). In 1962, Rachel Carson alerted the world to the fact that the indis- criminate use of DDT and other chlorinated hydrocarbons led to repro- ductive failure and to the death of birds and other animals. Spermato- 328 ABSTRACTS FROM MBL GENERAL MEETINGS zoa are also adversely affected by these pesticides, organophosphates, and organometallic compounds. Water-insoluble agents dissolve in DMSO (dimethyl sulfoxide); and because of their lipophilicity, they readily penetrate plasma membrane lipoproteins. The high surface-to- volume ratio of flagella makes sperm cells particularly vulnerable. Ar- bacia punctulaia cells suspended in artificial seawater respond dose- and time-dependently. Paraoxon, an anticholinesterase insecticide, reversibly stimulates motile progression after 5 min (0.7-70 ftM). Dieldrin and lindane are chlorinated hydrocarbons that can stimulate the vertebrate central ner- vous system to convulsions. These compounds are effective in the mi- cromolar range: above 30 to 100 pM dieldnn inhibits; between 1 and 15 it stimulates initially; and lindane mainly depresses (0.4-13 pAI). Mirex. an inducer of the cytochrome P-450 system, is a persistent insec- ticide and ftre-retardant (carcinogen). This compound markedly stimu- lates sperm (0.2-1 nM), but initially depresses them at lower ranges. The organometallic ethyl mercuric chloride is a fungicide used to treat seeds. It inhibits sperm movement (1-5 ^A/). but slightly increases it (50-55 nA/), apparently by deregulating sulfhydryl control systems. Tributyltin acetate, used in boat bottom paint to reduce fouling, de- creases sperm movement between 0.065 and 65 nAI. These organic pesticides and their inorganic counterparts affect the function of sperm cells, which can serve as sensitive indices of toxicants in the environment. Supported by the Sage Foundation. Sensory Biology Subnose 1: tracking oceanic odor plumes with high spa- tiotemporal resolution. JELLE ATEMA, GREG GER- HARDT, PAUL MOORE, AND LAURENCE MADIN (Bos- ton University Marine Program, Marine Biological Laboratory, Woods Hole, MA 02543). In May 1989, we employed electrochemical microelectrode tech- niques to measure the fine structure of an oceanic odor plume at 1000 m depth off St. Croix, LISVI. To avoid interference from unknown chemical compounds, we used a dopamine tracer and a selective dopa- mine detection method used previously to measure dopamine diffusion in brain tissue. "Subnose" is an underwater detector that can sample specific chemi- cal compounds with the spatiotemporal resolution of biological noses or better. The detector was mounted on a 1.5-m long stick that could be operated by the mechanical arm of the Johnson-Sealink submers- ible. For "Subnose- 1" we averaged the signals from three graphite-ep- oxy type microelectrodes, each with a sampling surface of about 30 ^m diameter. An odor plume was made by releasing a 50 mA/ solution of dopamine, together with fluoresceine dye, from a moored platform about 15 m above the sea floor. The solution was released evenly through a small nozzle (3 mm diameter) from a 10-1 reservoir over a 45-min period. Ambient turbulence created a characteristically patchy plume carried down by a 1 0 cm/s current. The plume was located visu- ally and tracked for 25 min with Subnose- 1, from about 50 m down- current, up to about 5 m from the source. The patchy distribution of visible dye was represented as a series of sharp peaks when patches of dopamine were encountered by the electrodes. Several peak parameters were measured based on peak definitions given in Moore and Atema (1988, Biol. Bull. 174: 355). Peak heights and maximum onset slopes rose toward the source indicating steeper and higher peaks near the release point. Patch sizes (area) decreased toward the source. Other pa- rameters did not show spatial gradients. Although the present, visually guided track is a rather arbitrary series of encounters with odor, peak parameters such as peak height and slope could be used by aquatic ani- mals to orient and locate odor sources regardless of an external frame of reference (e.g.. seafloor). This information may be critical for midwater animals, and useful for benthic animals, in their orientation. Response of bluefish (Pomatomus saltatrix) to increased intracranial pressure (Cushing response). STEPHEN H. Fox, CHRISTOPHER S. OGILVY, AND ARTHUR B. Du- Bois (The John B. Pierce Foundation Laboratory, New Haven, CT 065 19). Bluefish have vasomotor responses to hemorrhage, or to head-up tilt- ing in air (Ogilvy el al. 1989, Biol. Bull. 176: 176-190). Increases of intracranial pressure (ICP) induce increases of blood pressure (BP) in mammals (Cushing response). We increased ICP via metal implants in the skull and observed heart rate (HR) and BP in the ventral aorta. BP and HR increased in all 21 fish challenged, whether or not the fish were anesthetized with tncaine ( 1 g/40 I seawater) or were rendered flaccid with pancuronium (0. 1 mg/kg). Mean BP increased from 83 (SE 4) to 1 26 (SE 4) mm Hg, and HR from 46 ± 4 to 75 ± 2 beats per min (BPM), as ICP was increased in 10 mm Hg steps of 20 s duration, from 3 ± 1 to 65 ± I mm Hg. in five bluefish. In five other blueftsh, atropine (20 fig/kg) caused an increase in control heart rate and attenuated the in- crease of HR and BP during increases in ICP. In five more, phentol- amine (400 Mg/kg) lowered the resting BP and blunted the increase of BP during increased ICP. Atropine and phentolamine in combination completely eliminated the Cushing response in two fish (and the re- sponse to intraarterial epinephrine (4 j/g/kg) in one of these); but in three other bluefish, the Cushing response was reduced yet not elimi- nated, and in two of these, epinephrine response was not blocked. The Cushing response starts in 3-5 s, stops 3-5 s after ICP drops, and can persist at least 30 mm if ICP remains elevated at 16 mm Hg. The re- sponse was repeatable after recovery from atropine and phentolamine. The Cushing response was elicited by pressure exerted on the eyeballs, which raises ICP, or by an increase of air pressure over the exposed brain. Based on these experiments, the BP response in bluefish is medi- ated through tachycardia and vasoconstriction. The role of this re- sponse in fish in their natural environment remains to be elucidated. A novel chemosensory system in fish: do rocklings (Cili- ata mustela, Gadidae) use their solitary chemoreceptor cells as fish detectors'* KURT KOTRSCHAL (University of Salzburg, Austria), ROB PETERS, AND JELLE ATEMA. Solitary chemoreceptor cells (SCO of unknown function and biolog- ical role are scattered throughout the epidermis of most fish. Rocklings are favorable models for SCC research, as their modified anterior dorsal fin (ADF) contains several million SCC, but no taste buds (Kotrschal cl al 1984, Zoonwrphology 104: 365-372). These secondary sensory cells make synapses exclusively with recurrent facial nerve fibers (Whi- tear and Kotrschal 1988. J. Zoo/. Lond. 216: 339-366), which termi- nate in a distinct subdivision of the brainstem facial lobe (Kotrschal and Whitear 1988. / Camp. Neural. 268: 109-120). Body mucus dilu- tions of heterospecifics elicited responses in summed potential record- ings (Peters el al. 1987, ./. Mar. Biol. Assoc. U.K. 67: 819-823; these abstracts). To test the hypothesis that the ADF is a "fish detector," the breathing frequency (BTF) of four groups of rocklings (sham, closed noses, ADF removed, closed noses plus ADF removed) was recorded prior to, and during, a 2-min addition ot stimulus to the tank water (Lophius plus Opsanux mucus solution 0. 1% stock, or squid extract, or mucus water and squid extract simultaneously, six replicates per fish, four fish per treatment ). Sham fish responded to mucus water with a significant drop SENSORY BIOLOGY 329 in BTF. In fish with closed noses, this response was weaker, hut still significant. From fish with removed ADF, no BTF response to mucus water could be detected, even when the nose was intact. Squid juice did not affect the BTF, whereas squid juice together with body mucus produced a significant drop of BTF in the sham group. We conclude that fish body mucus is a biologically relevant stimulus for the ADF SCC in rocklings. Olfaction seems essential to interpret the ADF input. We predict, and subsequently will test with neuroanatomical tech- niques, that the brain connections of the ADF system are similar to that of the external taste buds, and the inputs from the ADF and the olfactory organ are integrated at prosencephalic levels. The authors acknowledge the financial support of the Austrian Fonds zur Foerd. wiss. Forsch., proj. nr. J0367-BIO, the Royal Netherlands Acad. Sci., and the Dutch-Austrian cultural exchange program. Potential gradient information contained within the three-dimensional structure of a laboratory odor plume. PAUL MOORE, NAT SCHOLZ, LYNNE LA- COMIS, AND JELLE ATEMA (Boston University Marine Program, Marine Biological Laboratory, Woods Hole, MA 02543). Odor plumes serve as sources of information for many animals dur- ing chemically mediated orientation. The information contained within the odor plume is poorly understood because the spatial and temporal scales at which chemoreceptor cells and organs function have been difficult to match with conventional chemical detectors. With newly introduced electrochemical microelectrodes ( Moore el til. 1989, Chem. Senses 13: in press), we can sample certain chemical tracers, such as dopamine, at micrometer space scales and millisecond time scales. In this study, we used this high resolution measurement to sam- ple the three-dimensional structure of an odor plume under controlled laboratory conditions. An odor plume was created in a uni-directional seawater flume (90 x 250 x 20 cm). A 2-m:U dopamine tracer flowed (50 ml/min) through a Pasteur pipette ( 1 mm ID) located 9 cm from the bottom in the cross-sectional center of the flume; lobsters sample in a plane about 9 cm from the bottom. After allowing the plume to establish itself for 2 min, an odor profile was recorded for 3 min using a graphhe-epoxy capillary electrode and a computer-based recording system (IVEC-V). The sample sites were located at 25. 50, and 100 cm from the pipette mouth: at 3. 9, and 1 5 cm from the flume bottom, and at 0, 5, 10, 20, and 30 cm to the right and left of the pipette. These recordings were analyzed for odor pulse parameters by procedures de- veloped in Moore and Atema (1988, Biol. Bull. 174: 355-363). Initial three-dimensional analysis of parameters that describe odor pulse shapes and frequencies shows that spatial gradients of certain parame- ter values point to the source. Specifically, the frequency of large peak heights and large onset slopes increase toward the source, whereas other parameters decrease, and yet others remain constant (see also Atema el al. these Abstracts). Animals may use this gradient information during orientation behavior, and their chemoreceptor filter properties may be "tuned" to certain odor pulse parameters (see abstracts by Scholz el al.. and Voigt and Atema). Supported by NSF (BNS 88- 1 2952) to J.A. A novel chemosensory system in fish: electrophysiological evidence for mucus detection by solitary chemoreceptor cells in rocklings (Ciliata mustela, Gadidae). ROBERT C. PETERS (University of Utrecht, The Netherlands), KURT KOTRSCHAL, WOLF-DIETRICH KRAUTGART- NER, AND JELLE ATEMA. Rocklings possess anterior dorsal fins (ADF) that are modified com- pared to other teleostean dorsal fins. The ADF consists of about 60 vibratile rays with a web only at the base; the first ray is enlarged. The vibratile rays are beset with about 5,000,000 solitary chemoreceptor cells (SCC). These SCCs are designated as chemoreceptor cells on ultra- structural criteria (Whitear 1971, J. Zoo/. Land. 163: 237-264; Kotrschal et al. 1984. Zoomorphology 104: 365-372). Electrophysio- logical studies demonstrated "tuning" of the SCCs to mucoid stimuli (Peters et al. 1987, J. Mar Biol. Assoc. U.K. 67: 819-823). The present experiments were designed to further specify the ade- quate stimulus and to compare the responses of the SCCs to those of taste buds (TB) in the pectoral and pelvic fins. Rocklings were anaesthe- tized with MS-222, and silver wire electrodes were then implanted around the recurrent facial nerve innervating the SCCs and around the facial nerves innervating the TBs. Summed potential recordings were made in unrestrained fish. Both groups of sensory organs were stimu- lated with body mucus ofGadns. Ciliata. Solea. Pholis, Coitus. Mugil, and Zoarces. and with the amino acids L-Arg, L-Ala, L-Cys, L-Gly, and L-Asp(l mAf). The SCCs reacted to body mucus of all the fish mentioned above, except for that of conspecifics (Ciliata); the SCCs were insensitive to amino acids. The TBs, on the other hand, did not react to body mucus, but gave vigorous responses to amino acids. These results suggest a specific function for the SCCs. The sensitivity of these receptors to body mucus might represent a chemosensory com- pensation for the limited vision of the rocklings in their highly struc- tured and tidal habitat. This ability may help them discriminate be- tween conspecifics and predators. The authors gratefully acknowledge the financial support of the Royal Netherlands Academy of Sciences, the Fonds zur Forderung der Wissenschaftlichen Forschung in Osterreich proj. nr. J0367-BIO, and the Austrian-Dutch cultural exchange program. Chemo-orientation of the lobster, Homarus americanus, to a point source in a laboratory flume. NAT SCHOLZ, PAUL MOORE, LYNNE LACOMIS, AND JELLE ATEMA (Boston University Marine Program Marine Biologi- cal Laboratory, Woods Hole, MA 02543). Many animals use chemical cues contained within turbulent odor plumes for orientation, but the relevant parameters of the plume are not known. Orientation studies in aquatic environments have advan- tages over those in terrestrial environments because odor distribution patterns can be more accurately measured, and stimulus delivery more closely controlled. We chose the lobster, Homarus americanus. because of its size and ease of handling, and the extensive base of neurophysio- logical data on the properties of its chemoreceptors. Previous behav- ioral studies indicate that the lobster relies heavily on chemosensory input in its natural habitat and specifically on antennularchemorecep- tion for efficient orientation (Devine and Atema 1982, Biol. Bull 163: 144-153). In a flow-through flume (90 X 250 x 20 cm), lobsters oriented to- wards a stimulus (0.5 g/l homogenized and centnfuged mussel tissue in raw seawater) constantly flowing from a pipette two meters upcurrent. The flow parameters were identical to those used for detailed plume description in Moore ct al. (these Abstracts). Lobsters were placed on a mussel diet, and then starved for at least three days prior to testing. Tests were videotaped with a camera mounted directly overhead on a moveable track; tapes were digitized at 1 Hz with the rostrum as the reference point. From casual inspection of the resulting walking paths, we favor the hypothesis of direct chemosensory control of orientation (i.e.. "chemotaxis") rather than an innate behavior program (such as zig-zagging) triggered by chemical input. These preliminary results 330 ABSTRACTS FROM MBL GENERAL MEETINGS have established the lobster as a viable model for studying chemo-orien- tation and chemical information extraction in turbulent odor plumes. Supported by NSF(BNS 88-12952) to J.A. Responses of chemoreceptor cells to controlled temporal stimulus patterns. RAINER VOIGT AND JELLE ATEMA (Boston University Marine Program, Marine Biologi- cal Laboratory, Woods Hole, MA 02543). The lateral antennules of the lobster, Homarus americanus. play an important role in olfactory orientation. Taurine- (Tau) sensitive che- moreceptor cells on the antennules form a narrowly tuned and highly sensitive cell population. Tau occurs in high concentrations in lobster prey and in extremely low concentrations in coastal waters; this low natural background may enhance the value of Tau as a long distance orientation cue. Natural odor plumes can be described by stimulus pulse intensity, background concentration, and repetition rate of pulses. To determine the adaptation and dis-adaptation properties of chemoreceptor cells, we tested single chemoreceptors with a series of 10 Tau pulses in different concentrations and different background concentrations, and varied the interstimulus intervals. Chemoreceptor cells were recorded extracellularly with suction elec- trodes. Single cells were identified with a standard Tau pulse (7 X 1(T5 A/). A train of 10 pulses in one of four concentrations (7 -, 10 4A/to7 x 10~7 A/) was applied in a 10 7 M Tau background in one of three pulse intervals (2.5 s, 5 s, 10 s). Each of these 12 trains was separated by a 3-min recovery period. After 3 min of recovery in artificial seawa- ter, the series of trains was repeated in 10~6 M, 10~5 A/. 10~4 A/, and 10 3 M Tau backgrounds. Individual cells showed a wide range of cummulative adaptation. Stimulus-response functions revealed range fractionation. In general, shorter pulse intervals resulted in gradually stronger cumulative adap- tation. Weaker pulse concentrations caused less cumulative adapta- tion. All but the highest background had negligible effects on response magnitude, showing the efficiency of background adaptation. Cumula- tive adaptation occurred mostly during the first three stimuli, and mostly with strong pulse concentrations in low background. Thus, low-firing cells showed good response reproducibility (i.e., no cumulative adaptation) even with the shortest pulse interval, whether low firing rates were caused by internal cell properties, low stimulus concentration, or high background concentration. This state of adapta- tion may be natural for cells operating in odor plume conditions. Supported by a grant from NSF (BNS 88- 1 2952) to J.A. Spectral tuning to amino acids and mixture effects on antennular chemoreceptor cells in the lobster, Ho- marus americanus. ANNA WEINSTEIN, RAINER VOIGT, AND JELLE ATEMA (Boston University Marine Program, Marine Biological Laboratory, Woods Hole, MA 02543). Lobster antennules are chemoreceptor organs that play a major role in orientation behavior. Hydroxyproline-best and taurine-best cell pop- ulations dominate their amino acid tuning spectrum. We tested the response of chemoreceptor cells on the lateral antennule to 15 ammo acids and to 3 mixtures containing these ammo acids to determine the effect of search mixture composition on spectral tuning. The mixtures we used reflect the prominence of hydroxyproline (Hyp) and taurine (Tau) as best compounds for antennular chemore- ceptors. All three mixtures had a total concentration of 1.5 X 10~3 M. "Equimolar mixture" contained all the compounds in equal concentra- tion ( 10 4 A/). The other two mixtures were based on the amino acid composition of a lobster prey, Mytilus edulis, modified by adjusting concentrations of Hyp and Tau. "Tau mixture" contained Tau at 3 x 10"4 A/ and Hyp at 10~4 A/ "Tau-Hyp mixture" contained both Tau and Hyp at 10 4 A/ Thus, all mixtures contained Hyp at 10~4 A/. Tau mixture was used as the search mixture. The mixtures and compounds were injected at 1-min intervals into a carrier flow of artificial seawater which passed over the antennule. Action potentials were recorded with a suction electrode in excised antennules. Twelve of the 2 1 cells were narrowly tuned to Hyp: three exclusively to Hyp, and nine with secondary responses mainly to glycine and argi- nine. Three cells were narrowly tuned to Tau, showing consistently strong responses. Four cells were tuned to glutamate, one to serine, and one to leucine. For each Hyp-best cell, all three mixtures caused similar responses ( Wilcoxon, P > 0.05 ) that were significantly less than responses to Hyp. Therefore, searching with any one of these mixtures would reveal a similar receptor population. However, cells that are completely sup- pressed by a search mixture only appear as "by-products" in recordings where more than one cell can be discriminated with extracellular tech- niques. The results show that similar tuning spectra are revealed by searching with an equimolar "nonsense" mixture (Johnson and Atema 1983, Neurosd. Lett. 41: 145-150) and natural mixtures of amino acids, all containing the same Hyp concentrations. Supported by a grant from NSF (BNS 88-12952) to J.A. CONTENTS CONSISTENCY AND VARIABILITY IN PEPTIDE FAMILIES Greenberg, Michael J., and Michael C. Thorndyke Consistency and variability in peptide families: in- troduction 167 Steiner, D. F., S. J. Chan, S. P. Smeekens, G. I. Bell, S. Emdin, and S. Falkmer Evolution of peptide hormones of the isletsof Lang- erhansand of mechanisms of proteolytic processing 172 Ebberink, R. H. M., A. B. Smit, and J. van Minnen The insulin family: evolution of structure and func- tion in vertebrates and invertebrates 176 Thorndyke, Michael C., Jennifer H. Riddel I, David T. Thwaites, and Rodney Dimaline Vasoactive intestinal polypeptide and its relatives: biochemistry, distribution, and functions 183 Taylor, Ian L. Peptide YY: the ileo-colonic, gastric, and pancreatic inhibitor 187 Vigna, Steven R. Tachykininsand the bombesin-related peptides: re- ceptors and functions 192 Dockray, G. J. Gastrin, cholecystokinin (CCK), and the leukosul- fakinins 195 Price, David A., and Michael J. Greenberg The hunting of the FaRPs: the distribution of FMRFamide-related peptides 198 Kobayashi, Makoto, and Yojiro Muneoka Functions, receptors, and mechanisms of the FMRFamide-related peptides 206 Nagle, Gregg T., Sherry D. Painter, and James E. Blankenship The egg-laying hormone family: precursors, prod- ucts, and functions 210 Goldsworthy, Graham, and William Mordue Adipokinetic hormones: functions and structures 218 Rao, K. Ranga, and John P. Riehm The pigment-dispersing hormone family: chemis- try, structure-activity relations, and distribution . . 225 DEVELOPMENT AND REPRODUCTION Eckelbarger, Kevin J., Craig M. Young, and J. Lane Cameron Modified sperm ultrastructure in four species of soft-bodied echinoids (Echinodermata: Echinothu- riidae) from the bathyal zone of the deep sea 230 Jaeckle, William B., and Donal T. Manahan Growth and energy imbalance during the develop- ment of a lecithotrophic molluscan larva (Haliotis rufescens) 237 Jeyalectumie, C., and T. Subramoniam Cryopreservation of spermatophores and seminal plasma of the edible crab Scylla serrata 247 Jones, Meredith L., and Stephen L. Gardiner On the early development of the vestimentiferan tube worm Ridgeia sp. and observations on the ner- vous system and trophosome of Ridgeia sp. and Rif- tia pachyptila 254 Keough, Michael J. Variation in growth rate and reproduction of the bryozoan Bugula neritina 277 ECOLOGY AND EVOLUTION Garthwaite, Ronald L., Carl J. Berg Jr., and June Harrigan Population genetics of the common squid Loligo penlei LeSueur, 1821, from Cape Cod to Cape Hat- teras 287 Maki, J. S., D. Rittschof, A. R. Schmidt, A. G. Sny- der, and R. Mitchell Factors controlling attachment of bryozoan larvae: a comparison of bacterial films and unfilmed sur- faces . . 295 PHYSIOLOGY Stickle, William B., Martin A. Kapper, Li-Lian Liu, Erich Gnaiger, and Shiao Y. Wang Metabolic adaptations of several species of crusta- ceans and molluscs to hypoxia: tolerance and micro- calorimetric studies 303 ABSTRACTS Abstracts of papers presented at the General Scientific Meetings of the Marine Biological Laboratory 313 Volume 177 THE Number 3 BIOLOGICAL BULLETIN JAN 1 7 19 DECEMBER, 1989 Published by the Marine Biological Laboratory Reference: Binl. Bull 177: 33 1-337. (December, 1989) Female Sexual Receptivity Associated with Molting and Differences in Copulatory Behavior Among the Three Male Morphs in Paracerceis sculpta (Crustacea: Isopoda) STEPHEN M. SHUSTER1 Department of Zoology, University of California, Berkeley, California 94720 Abstract. Paracerceis sculpta, a sphaeromatid isopod crustacean inhabiting the northern Gulf of California, forms harem polygynous breeding aggregations in spon- gocoels of intertidal sponges. Males in this species occur as three distinct morphs; a-males are large and possess modified uropods and telsons, /3-males resemble females, and 7-males are small and inconspicuous. Females are semelparous, and sexual receptivity is associated with a terminal molt; the half-molted (sexually receptive) con- dition lasts 6-50 h. Field-collected premolt females do not contain sperm. Half-molted females possess sperm masses in both oviducts, and postmolt females contain sperm tails in their spent ovaries. The presence of an a- male does not affect the duration of female receptivity, but females can delay initiation of their reproductive molt if males are absent. Isolated premolt females are in- capable of resorbing uninseminated ova. Such females molt, but do not transport ova into their marsupium and die without reproducing. All three male morphs com- plete similar behavioral sequences during intromission. However, (3- and 7-males copulate quickly and abandon females immediately after copulation, while a-males copulate longer and retain females after mating. The du- ration of female receptivity may encourage multiple mating and thus influence relative fertilization success among the three male morphs. Introduction The details of copulatory behavior are poorly known for most sphaeromatid isopods (see Bowman and Received 29 May 1989; accepted 15 August 1989. 1 Present address: Department of Ecology and Evolution. University of Chicago. 915 E. 57th Street, Chicago, IL 60637. Kuhne, 1974; Buss and Iverson, 1981; Shuster, 1981). The dearth of such studies appears primarily due to the tendency for breeding pairs or harems of these animals to situate in cavities, burrows, or beneath benthic substrates (Menzies, 1954; Wieser, 1962; Glynn, 1968; Holdich, 1968; Jansen, 1971; Bowman and Kuhne, 1974; Elef- theriou et al., 1981; Buss and Iverson, 1981; Shuster, 1981, 1987b; Upton, unpubl.). Direct observation of re- productive activities are, therefore, difficult or impossi- ble for many species. The timing of female sexual receptivity appears funda- mental to the secretive nature of these isopopds. In all sphaeromatid species examined to date, female receptiv- ity immediately follows a molt (review in Ridley, 1983; Shuster, 1981, 1986). Most Crustacea are vulnerable to predators or mechanical damage in newly molted condi- tion. Thus, in ancestral populations, females that pre- ferred protected habitats prior to their reproductive molt may have enjoyed greater fecundity than females prefer- ring more exposed areas. The association of breeding females with sheltered lo- cations may have facilitated male attempts to monopo- lize access to sexually receptive females (Emlen and Or- ing, 1977). Most female sphaeromatids lack sperm stor- age organs (Menzies, 1954; Ridley, 1983;Shuster, 1986), permitting males that mate first to place their sperm clos- est to a female's ova (Parker, 1970). Although the effect of mating order on male fertilization success has rarely been examined in Crustacea (Diesel, 1988), males that mate first achieve the greatest fertilization success in spe- cies with reproductive tract morphology similar to that of female sphaeromatids (Parker, 1970; Ridley, 1983). By guarding females prior to their reproductive molt, an- 331 332 S. M. SHUSTER Figure 1. Parucerceis scttlptu «-male (a) and prcmolt adult female (b). Horizontal line = 1 mm (redrawn from Brusca, 1980). cestral male sphaeromatids may have enhanced their probability of mating first. Competition among males for mating priority is likely to have been intense in such spe- cies (Parker, 1970, 1978), and may have favored males capable of physically removing their mates from access by other males. An evolutionary history of male attempts to sequester their mates from reproductive competitors may have contributed to modern difficulties with observ- ing breeding sphaeromatids (Shuster, 1981 ). The timing as well as the duration of female receptivity can profoundly influence the intensity of sexual selection on males (Knowlton 1979; Shuster and Caldwell, 1989). Therefore, evolutionary analysis of male reproductive behavior must first consider how patterns of female sex- ual receptivity provide the context for male reproductive activities (Shuster. 1 986). With respect to the characteris- tics of copulation itself, it is first necessary to determine when and how long females are sexually receptive, as well as how often, and with how many males, females are willing to mate. In this paper, I report these details for Paracerceis sculpla, a sphaeromatid isopod crustacean inhabiting in- tertidal zones in the northern Gulf of California (Fig. 1). This species forms harem polygynous breeding aggrega- tions in the spongocoels of Leucetta losangelensis, a common intertidal sponge (Shuster 1986, 1987a). P sculpta males exhibit an unusual polymorphism involv- ing three morphologically distinct types (Fig. 2, Shuster, 1986; 1987a, b; 1989). Alpha-males are larger than fe- males, and possess robust telsons and elongated uropods. Alpha-males guard the entrance of spongocoels contain- ing gravid females and defend their harems against other «-males. Beta-males are smaller than «-males, lack uro- pod and telsonic modifications, and resemble sexually mature females in external morphology. Beta-males mimic female courtship behavior and enter spongocoels containing reproductive females by deceiving resident «- males. Gamma-males are smaller still, also lack telsonic modification, and use their rapid movements and small size to slip around the bodies of resident «-males and into spongocoels. Gamma-males, like /3-males, prefer spongocoels containing reproductive females (Shuster, 1986, 1987a; 1989). The copulatory behavior of the three male morphs is presently undescribed. Materials and Methods Co/lection of experimental animals Leucetta losangelensis sponges grow abundantly year- round in tidepools on the coquina limestone reefs at Playa de Oro and at Station Beach, approximately 3 km southeast of Puerto Penasco, Sonora, Mexico (Shuster, 1 986). Isopods used in experiments were obtained in col- lections of 50-200 individuals, made every 4 to 10-days between February 1984 and November 1985. Breeding aggregations of isopods were removed from spongocoels. placed in separate vials, and examined in the laboratory within 6 h of collection. Females were identified by the possession of mature ovaries visible through the ventral cuticle or embryos in the brood pouch. Males were iden- tified by the presence of external genitalia and were clas- sified as «-, 13- , or 7-males by their body size and external morphology. Further details of collection and animal maintenance procedures, as well as details of male and female life histories, are available in Shuster (1986). All animals were returned to the Gulf of California after ex- periments were completed. Figure 2. The relative body sizes of the three male morphs in Para- Tiivv xailpui. Left to right, a-male, 0-male, 7-male. Horizontal line 1 mm. FEMALE P. SCVLPTA SEXUAL RECEPTIVITY 333 3. Diagram representing changes in female morphology as- sociated with a sexual molt. Walking legs are not drawn, (a) Premolt female, (b) hall-molted female (note genital pores in black at the base of the 5th leg), (c) postmolt female. Sperm in the reproductive tracts of field-collected females Females were removed from samples, measured to the nearest 0. 1 5 mm using a stereomicroscope, and assigned to one of three categories describing their reproductive condition. Like many isopods, P. sculpta females un- dergo a biphasic molt that initiates their sexual receptiv- ity (Ridley, 1983; Shuster, 1986). Females shed the pos- terior half of their cuticle below the fourth pereonal seg- ment first, and several hours to several days later (see below) shed the anterior half of their cuticle. Unmolted females possessing mature ovaries were classified as "pre- molt" females (Fig. 3a). "Half-molted" females pos- sessed mature ovaries and had shed their posterior cuti- cles (Fig. 3b). "Postmolt" females had completed their reproductive molts and transported their fertilized ova into their ventral marsupium (Fig. 3c). A more detailed description of female reproductive condition is provided in Shuster (1986). Twenty females of each of the three reproductive con- ditions were chilled in a freezer for 10 min and then dis- sected in physiological saline under a stereomicroscope. The body cavities of half of the females were opened dor- sally and the other half opened ventrally to permit exam- • nation of upper and lower aspects of the reproductive organs. Oviducts are transparent, and sperm, if present, are visible under low magnification (7X, Shuster, 1986). The presence or absence of sperm in the oviducts of dis- sected females was recorded. Effect ofa-males on the initiation and duration of female sexual receptivity In nature, unmolted females are attracted to spongo- coels containing n-males. Reproductive molting, mat- ing, and brooding of young by females all occur within these spongocoels (Shuster, 1 987b). To determine (a) the duration between female arrival in the spongocoel and the onset of sexual receptivity, (b) the duration of sexual receptivity itself (i.e., the half-molted condition), and (c) the effect of the presence of an «-male on the length of these durations, 40 unmolted females were removed from samples within 2 h of collection and placed in sepa- rate 225-ml cups containing food (Amp/iiroci thalli, Shuster, 1986) and seawater. A single a-male was added to 20 of these cups and each female was examined every 4 h until she had completed her molt. Isolation of uninseminated premolt females To determine whether sexually mature, but uninsemi- nated females, were capable of resorbing ova for future reproductions, 14 unmolted females were maintained in 225-ml cups with food and weekly water changes until they died. All females were examined while undergoing their molts, and postmolt females were examined every other day for evidence of feeding, changes in cuticle con- dition, visible changes in the character of internal organs, and activity level. The copulatory behavior oj males Copulatory behavior was examined by placing indi- vidual half-molted females into a watch glass containing seawater and introducing a single «- (n = 8), fi- (n = 3) or y- (n = 5) male. Interactions were observed under a stereomicroscope and, after apparent copulation, the genital pores and ventral surfaces overlying the oviducts of females were inspected under a stereomicroscope at high power (70X) for evidence of sperm transfer. To de- termine if females mate more than once, a half-molted female was placed in a watch glass, and four males were individually introduced to the female for 10 min each in the following sequence: a-, /?-, 7-, a-. All behavioral interactions were recorded on tape. Results Sperm in the reproductive tracts of field-collected females Premolt females contained no sperm in their repro- ductive tracts (n = 21, Table I). This is not surprising Table I The location of spermatozoa \\-ithtn I lie reproductive tracts oj premolt. half-molt, and postmolt females collected from Leucetta spongocoels Sperm present Female condition In oviducts In ovaries No sperm Premolt 0 0 21 21 Half-molt 19 0 0 19 Postmolt 14 19 0 19 334 S. M. SHUSTER Figure 4. (a) Sperm in the oviduct of a half-molted female; (h) iso- lated spermatozoon; as = acrosome; md = portion of oviduct leading tomarsupium; od = oviduct; ov = ovum;sm = sperm mass; v = vagina. as the vaginae of premolt females are fused prior to the reproductive molt and are physically incapable of acco- modating male genitalia (details of female reproductive morphology in Shuster, 1986). All half-molted females collected in the field (n = 19) possessed a whitish sperm mass in each oviduct. As men- tioned above, females possess simple oviducts, with no sperm storage organs (Shuster, 1 986). The sperm masses of inseminated females were visible through the ventral cuticle and were located approximately 0.25 mm inside of this opening, directly within the lumen of the oviduct. Sperm in the oviduct formed a loosely organized bundle (Fig. 4a). Intact sperm are approximately 1-1.5 mm in length, consisting of a long tail and a flange-like acro- some that is clearly visible at 70X (Fig. 4b). Within the reproductive tracts of half-molt females, sperm are non- motile. Postmolt females transport ova through their oviducts and into ventral pouches that form the marsupium (Shuster, 1986). Spent ovaries form an H-shaped bag that lies over the brood pouches when viewed dorsally (details in Shuster, 1986). While no intact sperm were found in the reproductive tracts of postmolt females (n = 19), sperm tails, minus their acrosomes, were distrib- uted throughout the spent ovaries and occasionally found within the oviducts. The effect oj a-males on the initiation and duration of female sexual receptivity Premolt females retained with «-males molt signifi- cantly sooner than premolt females retained in cups alone (one-tailed U-test, P = 0.008). Premolt females iso- lated in cups molted about five days after capture (me- dian = 126.3 h, range = 6.0-208.0 h, x ± SD = 120.78 ± 70. 17, n = 20), whereas premolt females retained in cups with an «-male molted about three days after cap- ture (median = 87.0 h, range = 4.50-138.0 h, x ± SD = 70.35 ± 47.24 h, n = 20). However, the duration of sexual receptivity, i.e., the duration a female remains in a half-molted condition, was unaffected by the presence of an a-male. Sexual receptivity lasted about 24 h for iso- lated females (median h as half-molt = 24.0, range = 6.0-50.0, x ± SD = 26.34 ± 10. 1 7, n = 20), as well as for females retained with an «-male (median = 25.0 h, range = 7.5-4 1 .0 h, x ± SD = 25.09 ± 8.80 h, n = 20; li- test, P> 0.42). Isolation of uninseminated premolt females Uninseminated females do not resorb their ova. Iso- lated premolt females undergo sexual molts normally, and cease to feed like gravid females (Shuster, 1986). However, uninseminated females do not transport ova into their brood pouches. Within a few days after molt- ing, the unfilled brood pouches of isolated females be- come opaque and slightly distended. The oostegites be- come progressively shriveled in appearance and begin to project outward from the body, while the ventral pereon grows large and appears to fill with fluid. Ovaries become progressively more pale, and the ova within begin to ap- pear fuzzy and indistinct. Over several weeks, the ova diminish in size, but with no corresponding improve- ment in the physical condition of females. In several cases, the empty brood pouches of females became in- fested with fungi and protozoa while the females were still alive, and in all cases, isolated females became slug- gish, deteriorated in physical condition, and died within 81 days (x ± SD = 49.36 ± 19.90, n = 14). This period is comparable to the adult longevity of females that be- come gravid and release normal broods (Shuster, 1986). The copulatory behavior of males In watchglasses, a-males did not seem capable of de- tecting half-molted females from a distance of more than a centimeter. However, when females swam within this distance, «-males became active and lashed their anten- nae vigorously. When a-males contacted half-molted fe- males with one of their antennae or with one of their walking legs, they immediately grasped females, adjusted them into a ventral-to-ventral position with respect to their own body, and pressed first one and then the other of the females' genital pores to their own genitalia (Fig. 5). Before being grasped, and while males adjusted their positions, females moved actively and occasionally es- caped from males. However, once face-to-face with «- males, females became quiescent and permitted males to mate. Each intromission (two per female) involved insertion of both of the male's penes into each of the female's geni- tal pores, followed by rapid pumping of the male's first two sets of pleopods. Median intromission duration for «-males was 30 s (range = 15-120 s, x ± SD = 39.08 FEMALE P. SCULPTA SEXUAL RECEPTIVITY 335 Figure 5. Alpha-male (white) in copula with hall-molted female (black). Drawn from photograph. ± 28.43 s, n = 13) and the median duration of the entire copulatory sequence for a-males was 12.07 min (range = 4.40-19.92 min, x ± SD = 1 1.59 ± 4.60 min, n = 8). In one case, an «-male completed the above sequence, retained the female, and copulated again, intromitting both of the female's genital pores. However, most «- males copulated as described above, and relaxed their grip on the female within 10 min after mating. Females became increasingly active following copulation, and shortly after a-males appeared to relax, females freed themselves and swam off. Females examined after single copulatory bouts contained sperm masses in both ovi- ducts. These sperm masses appeared smaller than those routinely found in half-molted females collected in the field. On two occasions, females were left in the watch- glass with their a-male for an additional 4 h. Upon re- examination, both females found were embraced by their a-males, and both females contained sperm masses that appeared substantially larger than observed after their initial mating. Thus, females evidently mate more than once with an individual male. Copulatory behavior involving females and |tf-males was similar in character to that observed between fe- males and «-males. Beta-males grasped females and as- sumed a ventral-to-ventral position. Females became quiescent, permitted intromission, and were released by /3-males almost immediately after mating. Median intro- mission duration for /i-males was 62 (range = 30-79 s, x ± SD = 54.83 ± 20.26 s, n = 3) and the median copula- tory sequence duration was 2.97 min (range = 2.00-3.36 min, x ± SD = 2.78 ± 0.70 min, n = 3). All females copu- lating with /3-males contained sperm in both oviducts af- ter a single copulatory sequence. Gamma-males copulated, not by positioning females, but instead by climbing beneath them and positioning their own genitalia toward the females" genital pores. The mere contact of a male, not the act of being grasped and positioned, seems to stimulate quiescence, as females paired with y -males ceased moving soon after 7-males assumed positions beneath them. While «- and /i-males rapidly moved their anterior pleopods during copulation and engage in few or no thrusting movements, 7-males thrusted actively during copulation. Like /3-males, 7- males abandoned females soon after mating. Median in- tromission time for 7-males was 42 s (range = 25-57 1 s, x ± SD = 122.86 ± 198.85 s, n = 4) and median copula- tory sequence duration was 5.75 min (range = 3.00- 12.60 min, x ± SD = 6.78 ± 4. 10 min, n = 4), intermedi- ate in duration between that of «- and 0-males. All fe- males mating with 7-males contained sperm masses in both oviducts. While intromission times for a-, /3-, and 7-males were not significantly different (two-tailed Kruskal-Wallis test, P > 0.05). copulatory sequence times differed significantly among males (two-tailed Kruskal-Wallis H = 7.96, P < 0.01 ). Thus a-males evi- dently copulate longer than 0- and 7-males. The half-molted female introduced to an «-male, a /3- male, a 7-male, and another a-male, mated with all four males in rapid succession. Copulation in the first three cases proceeded normally. In the last case, the a-male ejaculated, but was apparently unable to place all of his ejaculate into the female's genital pores. Several sperm bundles were observed trailing out of the female across her ventral pereon after the last male released her. Copu- lations with all four males occurred within 30 min, and, within 5 min of the final mating, the female shed her an- terior cuticle and began transporting ova into her brood pouch. Thus female receptivity lasts from moments after females shed their posterior cuticle until moments before the anterior cuticle is shed. During this time, females evi- dently will mate with any nearby male. Discussion Although premolt females are attracted to spongocoels containing a-males and engage in courtship behavior (i.e.. premolt females are behaviorally receptive, Shuster, 1 986 ), the genital pores of premolt females are indistinct, and the oviducts of these females do not contain sperm. Actual sexual receptivity and copulation occur only in half-molted females, substantiating observations by Menzies (1954) and Ridley (1983). Females seem capa- ble of postponing receptivity if males are not available. This is reasonable for a species in which females must leave the habitat in which they mature to locate suitable reproductive habitat (Shuster, 1986). However, females cannot postpone their reproductive molt indefinitely even when isolated from males, and once females molt, the presence of a-males does not affect the duration of receptivity. While receptive, however, females will mate more than once and with more than one male. Females transport ova to their brood pouches almost 336 S. M. SHUSTER immediately after shedding the anterior portion of their cuticles, and females are sexually receptive until mo- ments before this molt. Fertilization may occur in the ovary, as sperm tails, minus their acrosomes, were abun- dant in spent ovaries. Within the oviduct, sperm are non- motile, and while females may induce an acrosome reac- tion that initiates syngamy (P. Talbot, pers. comm.), this reaction is unlikely to propel sperm the length of the fe- male reproductive tract. Females appear to possess con- tractile tissues in their oviducts that may facilitate sperm transport. Fertilization evidently does not occur in either the marsupium or in the oviduct itself, as has been sug- gested for some isopods( Ridley, 1983). Females exhibit a single 24-h period of sexual receptiv- ity. This interval seems somewhat brief until the poten- tial density of competing males in a spongocoel is consid- ered. Although most occupied spongocoels contain a sin- gle «-male, up to eight «-, 0-, and 7-males in various combinations may simultaneously occupy the same spongocoel (Shuster, 1987a; 1989). At such male densit- ies, 24 h of sexual receptivity introduces a high probabil- ity of multiple insemination for females, and thus pres- ents considerable opportunities for sperm competition among males. Although it may be physically impossible for P. sculpta females to complete the two phases of their reproductive molt in less than 24 h, other isopod females complete their reproductive molts in a few hours or less (Ellis, 1971; review in Ridley, 1983). Furthermore, uni- phasic molting (i.e., shedding of the entire cuticle at once) occurs in some marine isopoda (George, 1972). The possibility must therefore be considered that P. sculpta females somehow benefit from a period of sexual receptivity that is sufficient in duration to permit multi- ple mating. All three male types performed similar activities asso- ciated with the act of copulation and the transfer of sperm. Intromission durations among the three male morphs were also similar and were consistently rapid (the transfer, within 30 s, of several hundred 1-mm-long sperm by males no larger than 7 mm in length is an amazing feat by any standard). Furthermore, the fecun- dities of females mated to n-, ft-, and 7-males are not significantly different (Shuster, 1986; 1989). Despite these similarities, the amount of time males spent with individual females differed significantly among males. Alpha-males retained females for some time after mating, while 0- and 7-males released females immediately. This result makes sense given descriptions of the behavior and distribution of ft- and 7-males in spongocoels (Shuster, 1986, 1987b, 1989). Beta- and 7- males seem well-adapted as sperm competitors (Shuster, 1 987b, 1 989) and could maximize their contacts with fe- males by avoiding post-copulatory guarding (Parker, 1974). However, the fact that «-males release females at all before females complete their reproductive molt is surprising. If females may be inseminated by other males, «-males have little to gain by releasing their half- molted mates unless other females are present. In the field, «-males are often found gripping premolt or half-molt females with their walking legs (Shuster, 1986, 1987a). Furthermore, when single, receptive fe- males are present in spongocoels with both an «-male and a 7-male, 7-males rarely mate successfully, suggest- ing successful post-copulatory guarding by w-males (Shuster, 1989). Perhaps in a watchglass and in the ab- sence of other males, «-males are not stimulated to retain their mates. Observations of the copulatory behavior of these isopods within the confines of their natural repro- ductive habitat, as well as further analysis of patterns of female sexual receptivity in this species, may explain these apparently conflicting patterns of male guarding behavior and fertilization success. Acknowledgments I am grateful to Drs. Roy L. Caldwell, Frank A. Pi- telka, Vincent H. Resh, George K. Roderick, and Eld- ridge S. Adams III for their comments on earlier drafts of the manuscript. Logistical support for this research was provided by Mona Radice and Mable Lee of the Depart- ment of Zoology, University of California, Berkeley, the Center for the Study of Deserts and Oceans in Puerto Penasco, Sonora, Mexico, and the Environmental Re- search Laboratory in Tucson, Arizona. Financial assis- tance was provided by the Theodore Roosevelt Memo- rial Fund, and NSF dissertation improvement grant OCE-8401067, as well as by the University Regents' Fel- lowship, the Alice Galloway Memorial Fund, the Gradu- ate Student Research Allocations Fund and the Depart- ments of Zoology and Genetics of the University of Cali- fornia, Berkeley. Literature Cited Bowman, T. E., and H. Kuhne. 1974. Cymodella Kamfami. a new sphaeromatid isopod (Crustacea) from Australia, with notes on its mating behavior. Rec Anxl. Mils 29: 235-243. Brusca, R. C. 1980. Common Intcrlidal Invertebrates of the (jitlj <>l California. University of Arizona Press, Tucson. AZ. 5 1 3 pp. Buss, L. VV., and E. W. Iverson. 1981. A new genus and species of Sphaeromatidae (Crustacea: Isopoda) with experiments and obser- vations on its reproductive biology, interspecific interactions and color polymorphisms. Poslilla 184: 1-23. Diesel, R. 1988. Discrete storage of multiple mating sperm in the spi- der crab. Inachus phalangium. Natunvissenschaften 75: I48-I41). Elefthfriou, A., D. M. Holdich, and K. Harrison. 1980. The systemat- ics and ecology of a new genus of isopod (Sphaeromatidae) from the West Coast sandy beaches of India. Estuur. Coast. Mar. Sci. 11: 251-262. Ellis, R. J. 1971. Notes on the biology of the isopod Asellus loma- Icnsis Hartford in an intermittent pond. Trans. Am Mu'rosc. Soc. 90:51-61. FEMALE P. SCULPTA SEXUAL RECEPTIVITY 337 Kmlen, S. I ., and L. \V. Oring. 1977. Ecology, sexual selection and the evolution of mating systems. Science 197: 2 1 5-223. Glynn, P. \V. 1968. Ecological studies on the associations of chitons in Puerto Rico, with special reference to sphaeromid isopods. Bit/1. Mar. Sci. 18: 572-626. George, R. V. 1972. Biphasic moulting in isopod Crustacea and the finding of an unusual mode of moulting in the antarctic genus, (ilyi'loin>lii\ J \al. Hist. 6: 651-656. lloldich. D. M. 1968. Reproduction, growth and bionomics o(Dyna- nicne hulcnlata (Crustacea: Isopoda)../ Zoo/. Luml 156: 136-153. Jansen. K. P. 1971. Ecological studies on intertidal New Zealand Sphaeromatidae (Isopoda: Flabellifera). Mar. Biol. 11: 262-285. Kncmlton, N. 1979. Reproductive synchrony, parental investment and the evolutionary dynamics of sexual selection. Anim. Bchav. 27: 1022-1033. Menzies, R. J. 1954. The comparative biology of reproduction in the woodboring isopod crustacean Limnoria. Bull, Mux (.'oinp Zoo/. 112:361-3X8. Parker, G. A. 1970. Sperm competition and its evolutionary conse- quences in insects. Bin/. Rev 45: 525-567. Parker, G. A. 1974. Courtship persistence and female guarding as male time investment strategies. Behaviour 48: 157-184. Parker, G. A. 1978. Evolution of competitive mate searching. Ann. Rev Enlomol 23: 173-196. Ridley, M. 1983. The Explanation ol Organic Diversity: The Coin- parative Method and Adaptations for Mating. Oxford Science Publi- cations, Oxford. UK. 272 pp. Shuster, S. M. 1981. Sexual selection in the Socorro Isopod, Ther- mosphaeroma thcrmoptuliim (Cole and Bane) (Crustacea: Peracar- ida). Anim Bchav. 28:698-707. Shuster, S. M. 1986. The reproductive biology ofParaccrceix sciilpta (Crustacea: Isopoda). Ph.D. Thesis, University of California, Berke- ley. 279 pp. Shuster, S. M. 1987a. Alternative reproductive behaviors: three dis- crete male morphs in Pamccrceix xcitlplu. an intertidal isopod from the northern Gulf of California./ Crust. Biol. 7: 318-327. Shuster, S. M. 1987b. Male body size, not reproductive habitat char- acteristics predicts polygyny in a sexually dimorphic intertidal iso- pod crustacean, Paracerceis xailpta (Crustacea: Isopoda). Mem. Simp. Bio!. Mar. Univ. Anton. Baja Calif. 5: 71-80. Shuster, S. M. 1989. Male alternative reproductive strategies in a ma- rine isopod crustacean (Paract'rceix xciilptu): use of genetic markers to measure differences in fertilization success among «-, ft-, and 7- males. Evolution. 43. Shuster, S. M., and R. L. Caldwell. 1989. Male defense of the breed- ing cavity and factors affecting the persistence of breeding pairs in the stomatopod, Gonodactylus bredini (Crustacea: Hoplocarida). Ethology. 82: 192-207. VVieser, W. 1962. Adaptations of two intertidal isopods. I. Respira- tion and feeding in Nacxa hidenlaia (Adams) (Sphaeromatidae). J. Mar. Biol. Assoc. V. A' 42: 665-682. Reference: Hiol. Bull 177: 338-343. (December. The Role of the First Quartet Micromeres in the Development of the Polyclad Hoploplana inquilina BARBARA CONTA BOYER Department of Biological Sciences, Union College, Schenectady, New York 12308 and The Marine Biological Laboratory, II 'oods Hole, Massachusetts 02543 Abstract. The role of the first quartet micromeres at the eight-cell stage in the development of the polyclad flat- worm Hoploplana inquilina was analyzed with regard to specific contributions made by these cells and their func- tion in the determination of embryonic symmetry. The experimental series involved: (1) deletion of one micro- mere ( la or Ic versus Ib or Id); (2) deletion of two adja- cent micromeres; (3) deletion of three micromeres; (4) isolation of intact first quartets; and (5) isolation of mac- romere sets 1 A- ID. As the number of micromeres re- moved was increased, the larvae became progressively more abnormal, involving reduction in number of eyes, deficiencies in lobe development, and disturbance of em- bryonic symmetry. After deletion of three micromeres, none of the larvae exhibited normal morphology. These experiments indicate that the determination of embry- onic axes leading to a larva with bilateral symmetry may involve micromere-macromere interactions, as has been shown in molluscan embryos with equal cleavage. Iso- lated first quartets consistently formed spherical, bloated, transparent larvae with multiple eyes, suggest- ing that the macromeres play an inhibitory role in eye development. Isolated macromeres 1A-1D often failed to develop, and larval structures never differentiated. Thus, the relatively loose determination of the polyclad embryo involves both cytoplasmic localization and cell interactions. Introduction The establishment of cell fate has been investigated in a number of animal species having embryos that exhibit spiral cleavage. Much of our current understanding of the role that cytoplasmic determinants play in embry- Received 5 April 1989; accepted 21 September 1989. onic development derives from studies of higher Spiralia, which typically express early embryonic determination or mosaicism. Such investigations have revealed com- plexity and variety within these fundamentally mosaic systems. For example, cell interactions have been dem- onstrated in many spiralian embryos, and within the molluscs, two different mechanisms of achieving the em- bryonic axes of symmetry have evolved. Some turbellarian platyhelminthes have spiral cleav- age, and they are more primitive than the annelids and molluscs on which these investigations have been carried out. Therefore, studies of turbellarian development should provide insights, not only into the nature of cy- toplasmic localization, but also into the origin and evolu- tion of spiral cleavage and embryonic determination. The polyclads, in particular, have a typical quartet spiral cleavage that is strikingly similar to the pattern in anne- lids and molluscs, but there has been little experimental work done on the development of this group. The normal development of a polyclad flatworm has been most thoroughly examined in Hoploplana in- ilitilina (formerly Planocera inquilina} by Surface ( 1907). Cleavage follows the typical quartet spiral pattern in which the macomeres (A-D) divide to produce four quartets of micromeres. At the four-cell stage, the em- bryo consists of two equal-sized A and C blastomeres that are displaced toward the animal pole, and two equal- sized but larger B and D blastomeres that are in contact at the vegetal cross-furrow. [Surface (1907) reports that the D cell is larger than B, but I have not observed this to be the case.] The eight-cell stage (Fig. 1A) consists of the animal first quartet ( la-Id) and the vegetal macro- meres ( 1 A- 1 D). Later, when the cells of the fourth quar- tet (4a-4d) are produced, they are significantly larger than the macromeres (4A-4D) and contain most of the 338 DEVELOPMENT OF HOPLOPLANA 339 yolk. This is characteristic of the polyclads, distinguish- ing them from other spiralians. Gastrulation proceeds primarily by epiboly of the micromeres over the macro- meres. By the fourth day, the embryo begins to rotate, the lobes have started to form, and differentiated tissues are discernable. A fully developed Muller's larva (Fig. 1 B) is produced by the fifth day. The embryos of Hoploplana have the characteristic mosaic development of the Spiralia(Boyer, 1986, 1987). Blastomere deletion and isolation experiments on two- and four-cell embryos produced partial larvae with char- acteristic deficiencies associated with each type of experi- ment. "Half larvae," resulting from the separation or de- letion of two-cell embryos, were abnormal in body shape, and in the development of lobes and eyes. Dele- tion of one cell at the four-cell stage produced less anom- alous "three-quarter larvae" that were underdeveloped in one quadrant and often exhibited eye abnormalities. These results resemble those obtained from such experi- ments on higher spiralians — that is, development is fun- damentally mosaic — but the studies did not include analysis of embryos beyond the four-cell stage, and the mechanism of determination of the embryonic axes of symmetry was not examined. Experimental analysis of eight-cell stage (first quartet) embryos of some of the higher spiralians have focused on two different problems, the cytoplasmic contributions of the first quartet cells, and the role that these cells play in the determination of embryonic symmetry. Among the former studies are those of Wilson (1904) on Patella, Horstadius (1937) on Cerebral id us. Costello (1945) on Nereis, Clement (1967) on Ilyanassa, Morrill el al. (1973) on Lymnaea, and van Dam and Verdonk (1982) on Bithynia. In general, the results supported the concept of mosaic development, though there was evidence of some regulative capacity in Bithynia and Lymnaea. Determination of the embryonic axes of symmetry has been extensively studied in gastropod molluscs, a group in which two different mechanisms are involved. In gas- tropods with unequal cleavage, axis determination has occurred by the four-cell stage. At this time the D blasto- mere, which is larger than the others, inherits cytoplas- mic determinants specifying mesoderm and the dorsal quadrant. Thus, by the four-cell stage, the axes of bilat- eral symmetry have been established through an intracel- lular mechanism involving cytoplasmic localization (see Davidson, 1986, for review), and deletion of the first quartet at the eight-cell stage does not affect determina- tion of the dorso-ventral axis (van Dam and Verdonk, 1982). In gastropods with equal-sized blastomeres at the four- cell stage, the first quartet cells play a crucial role in estab- lishing the embryonic axes of symmetry (van den Biggel- aar, 1976, 1977; van den Biggelaar and Guerrier, 1979; Arnolds et al.. 1983; Martindale el al., 1985). These em- bryos remain radially symmetrical with equipotent quadrants until after formation of the third quartet. At this time, the first quartet micromeres contact the central macromere and induce it to become dorsal (van den Big- gelaar, 1976, 1977). If these interactions are delayed (Martindale et al.. 1985) or inhibited (van den Biggelaar and Guerrier, 1979), no D quadrant forms, and the em- bryo remains radially symmetrical. Thus, in these mol- luscs, intercellular interactions determine the embryonic axes of symmetry. That such closely related groups should have such different mechanisms for establishing the D quadrant is surprising, because the cleavage patterns and cell lineages of these animals are very similar, and both equal and un- equal cleavage occur widely in the molluscs. As Martin- dale et al. (1985) state, "We need to know more about the experimental embryology of other spiralians before the significance of the differences in the mechanisms which have been uncovered so far can be put into per- spective." Because the ancestral flatworms may have been the first animals to evolve bilateral symmetry, the mechanism of symmetry determination in this group is particularly significant. The Hoploplana embryo, with two equal sized A and C blastomeres and two larger but also equal sized B and D cells at the four-cell stage, does not appear to fit either model. The consistent pattern of defects occurring with deletion experiments on two- and four-cell embryos in- dicates that morphogenetic determinants specifying par- ticular larval structures are organized in the zygote, but Hoploplana does not appear to have a designated cell corresponding to the D blastomere of higher spiralians. Therefore, I have examined the eight-cell (first quartet) stage to determine not only the specific cytoplasmic con- tributions of the micromeres. but also whether their in- teractions with the macromeres are involved in establish- ing the axes of symmetry. Materials and Methods Specimens of Hoploplana inquilina were collected from the mantle cavity of Busycon and maintained in seawater in finger bowls. Gametes were removed from the spermaducal vesicles and uterus by piercing the or- gans with sharp needles. Fertilization, which produced naked zygotes that lack the egg-shell membrane, oc- curred when eggs and sperm were mixed in plastic petri dishes containing Millipore-filtered seawater. The blastomere deletion experiments were done by puncturing the selected cell with hand-pulled glass nee- dles, typically about one half hour after cleavage to the eight-cell stage. The experimental embryos were exam- ined carefully to be certain that each punctured cell had 340 B. C. BOVER B Figure I. Normal developmental stages of Hoploplana. (A) Eight- cell stage viewed from animal pole. (B) Muller's larva. AT: apical tuft; DL: dorsal lobe; LLL: left lateral lobe; LVLL: left ventrolateral lobe; OH: oral hood; RVLL: right ventrolateral lobe; PT: posterior tuft. category, not normally a product of deletion experi- ments on two- and four-cell embryos, appeared with de- letion of three and four micromeres or isolation of the intact first quartet. These were classified as "spherical" or "swollen" embryos. The former were less well developed than half larvae, with solid but poorly differentiated tis- sues, and were either one-eyed or eyeless. Larvae with the "swollen syndrome" had almost perfectly spherical morphology, were greater than 100 ^m, and showed very abnormal tissue development. The inner tissues typically were undifferentiated and a transparent space existed be- tween the outer ectoderm and inner undifferentiated mass. These forms often had multiple eyes (Fig. 2C). The data for all experimental categories are summarized in Table I. Deletion oj one nncroniere (la or Ic vs Ih or hi) The first two columns in Table I show the results of deletion of single first quartet micromeres at the eight- cell stage. Regardless of which micromere was deleted, 16-18% of the larvae were completely normal. Most lar- vae (85%) exhibited normal Muller's morphology with bilateral symmetry, so that observed abnormalities were cytolyzed completely. If deletion was incomplete, the embryo was discarded. The experimental series in- volved: (1) deletion of first quartet micromeres la or Ic versus Ib or Id (it is not possible to differentiate between the A and C or the B and D quadrants); (2) deletion of two adjacent micromeres; (3) deletion of three micro- meres; (4) isolation of the first quartet intact by killing the macromeres; and (5) isolation of the 1 A- ID macro- meres intact by deleting the first quartet. Experimental embryos were raised in Millipore-fil- tered seawater to which 100 units/ml penicillin and 200 ^g/ml streptocmycin were added. They were examined daily and analyzed on day six or seven for abnormalities in eye numbers, tufts, and morphology. Results The larvae became progressively more abnormal as the number of deleted micromeres increased. If larval morphology appeared normal, the larvae were catego- rized as Muller's larvae (Fig. IB). If they exhibited any abnormalities similar to those resulting from deletion of one cell at the four-cell stage, such as absence or trunca- tion of one or more lobes, they were termed "three-quar- ter larvae" (Fig. 2A). Larvae that were underdeveloped, asymmetric (lacking the normal axes of bilateral symme- try), and unrecognizable as Muller's larvae, were similar to those produced by deletion of one cell at the two-cell stage, and were called "half larvae" (Fig. 2B). A fourth B Figure 2. (A) "Three-quarter" larva. (B) "Half" larva. (C) "Swollen syndrome" larva. DEVELOPMENT OF HOPLOPL.4NA Table I Effccl of deleting micromeres and macromeres on eight-cell embryos of Hoploplana inquilina 341 -(laor Ic) n = 61 -dbor Id) n = 50 -2 mics. n = 27 -3 mics. n = 29 -4 macs. n = 22 -4 mics. n = 20 Normal 10(16%) 9(18%) 0 0 0 0 Morphology Muller's 52 (85%) 37(74%) 15(56%) 0 0 0 % Larvae 9(15%) 1 1 (22%.) 10(37%) 6(21%) 0 0 '/3 Larvae 0 2 (4%) 2(7%) 9(31%) 0 4(20%) Sphere/swollen 0 0 0 14(48%) 22(100%) 16(80%) Tufts Normal 38 (62%) 24 (48%) 4(15%) 0 0 0 -Apical 15(25%) 13(26%) 18(67%) 8(28%) 1 tuft: 5 (23%) 1 tuft: 2 (10%) —Posterior 5(8%) 5(10%) 0 0 -Both 3 (5%) 8(16%) 5(18%) 21 (72%) 17(77%) 18(90%) #Eyes 0 1 (2%) 0 2(7%) 8(28%) 1 (5%) 20(100%) 1 22(36%) 9(18%) 19(70%) 19(66%) 0 0 2 35(57%) 37(74%) 4(15%) 1 (3%) 4(18%) 0 3 3 (5%) 4(8%) 2(7%) 1 (3%) 8(35%) (1 4 0 0 0 0 7(32%) (1 5 0 0 0 0 2 (9%) 0 Minus sign indicates deletion. in the number of eyes or tufts. Three-quarter larvae, which also exhibited fundamental bilateral symmetry, constituted 15% of embryos with la or Ic deleted and 22% with absence of Ib or Id. In this category the right ventrolateral lobe was truncated or missing in two larvae, the left ventrolateral lobe in four, the dorsal lobe was de- ficient in one, and all lobes were abnormal in two. Only a very small number were categorized as half larvae. There was no significant difference in larval morphology related to deletion of la or Ic vs Ib or Id (x2 = 1-8, P>0.05). Deletion of one first quartet micromere resulted in loss of a tuft in approximately half of the embryos, with the apical tuft more commonly missing than the posterior tuft. A small number were missing both tufts. There was no significant difference in tuft abnormalities with re- spect to the particular cell deleted (x: = 1 .88, P > 0.05). Although most larvae had two eyes, there was a sig- nificantly greater proportion of one-eyed larvae when la or Ic was deleted (36%.) versus Ib or Id ( 18%) (x: = 5.2, P < 0.05). Of the single-eyed larvae, when la or Ic was deleted, 22 (50%.) had an eye on the right, 5 (23%) were left-eyed, and 6 (27%) had a centrally located eye. Dele- tion of Ib or Id resulted in 5 (56%) right-eyed and 2 (22%) left-eyed larvae. One had a central eye and in one the position of the single eye could not be determined. Deletion of two adjacent micromeres When two micromeres rather than one were deleted (Table 1, column 3), the number of larvae with normal (Muller's) morphology decreased significantly, while concomitantly the number of three-quarter larvae in- creased (x: = 7.09, P < 0.01). Of the latter, three had truncated or missing right ventrolateral lobes, five had comparable abnormalities of the left ventrolateral lobe, one had a deficient oral hood, one had an abnormal dor- sal lobe, one had all abnormal lobes, and one was missing both ventrolateral lobes. Similarly, the number of larvae with normal tufts decreased significantly (x2 = 22, P < 0.01 ). (For the chi square analysis, the data from both kinds of single micromere deletions were pooled because the larvae from the two groups were not significantly different in morphology or tufts.) There was a significantly greater number of one-eyed larvae with deletion of two adjacent micromeres com- pared with deletion of la or Ic (x2 = 12.02, P < 0.01) and deletion of Ib or ld(x2 = 26.3, P < 0.01). The ma- jority of these ( 1 2 or 63%) had a central eye, 3 ( 1 6%) were right-eyed, and 4(21%) were left-eyed. Deletion of three micromeres A drastic effect on development was seen when three micromeres were deleted (Table I, column 4). There were no morphologically normal larvae, and almost half fell in the highly abnormal category of spherical, swollen forms. Of the "three-quarter" larvae, one was missing the left ventrolateral lobe, two lacked both ventrolateral lobes, and in one all lobes were abnormal. Only one larva had two eyes; the remainder were almost all one-eyed or 342 B. C. BOYER eyeless, and 77% were missing both tufts. The number of one-eyed larvae was not significantly different between the two and three micromere experiments, but there was a significantly greater number of eyeless larvae in the lat- ter experiments. Isolation of the intact first quartet Thirty-eight experiments were done in which all four macromeres were deleted leaving the four first quartet cells. Only 22 (58%) survived (Table I, column 5), and all were spherical with 95%. exhibiting the "swollen syn- drome." Seventy-six percent of the larvae had more than two eyes, and 77% had no tufts. Isolation of the first quartet macromeres The results of deleting the entire first quartet so that only the macromeres 1A-1D remain are presented in Table I, column 6. Of 105 embryos in which the la- Id cells were deleted, only 20 survived. These were highly aberrant; 70% were categorized as "spheres," undergoing little development beyond cleavage and no tissue differ- entiation. Ten percent (2) became swollen, and a small number (4) could be characterized as half larvae. None had any eyes, and 90%- were tuftless. Discussion The results of micromere deletion experiments on Hop/op/ana corroborate the earlier four-cell deletion ex- periments suggesting that morphogenetic determinants are sequestered early in development, but that specific blastomeres do not receive consistently specific determi- nants. Eyes and tufts If the Hoploplana cell lineage conformed to the mol- luscan plan, only la and Ic should form eyes. Although deletion of one of these blastomeres results in one-eyed larvae in a statistically significant number of cases, dele- tion of the ventral and dorsal Ib and Id cells also pro- duces one-eyed larvae. Moreover, the majority have the normal two eyes. Thus, Hoploplana is more similar to Bithynia (van Dam and Verdonk, 1982) and Lymnaea (Morrill et a/., 1973), which may develop normally after deletion of first quartet micromeres, than to Ilyanassa (Clement. 1967), which appears to be more rigidly mo- saic. The la and Ic blastomeres cannot be distinguished from cells Ib and Id in the Hoploplana embryo; there- fore the basis for the preponderance of right-eyed lar- vae— whether simple non-random deletion, or more complex developmental processes — cannot be deter- mined. The occurrence of a single, centrally located eye is probably due to the disturbed cellular topography re- sulting from blastomere deletion. When two adjacent micromeres are deleted, the num- ber of one-eyed larvae is significantly larger than when one micromere is deleted; but the proportion does not reach 1 00% (15% still have the normal two eyes) as would be expected if eye development involved simply the lo- calization of eye determinants in opposite micromeres. The two eyeless larvae also suggest a more complex sys- tem than can be explained by strict cytoplasmic localiza- tion. Similarly, Morrill et al. (1973) found that paired eyes develop in Lymnaea when one, two, or three micro- meres are deleted at eight cells. Because cells other than la or Ic can produce eyes, they concluded that eye deter- mination involves some kind of induction and that nor- mal development is possible only when the cleavage pat- tern and cell arrangements are normal. When three micromeres are deleted in Hoploplana, al- most all of the larvae are one eyed (66%) or eyeless (28%), and isolated first quartet macromeres never produce any eyes. Thus, at least two micromeres must be present for two eyes to form. However, isolated first quartets com- monly develop supernumerary eyes, suggesting that the macromeres may play an inhibitory role in eye develop- ment. Similarly, Gather ( 1973) has demonstrated an in- hibitory role of the polar lobe in Ilyanassa (a macromere derivative), in the development of cilia by first quartet cells. As the number of micromeres deleted increases, tuft abnormalities also increase. Single micromere deletions suggest that tuft determinants are localized in the first quartet micromeres, but that tufts are equally likely to be absent upon the deletion of any blastomere. Deletion of two or more micromeres almost always results in larvae missing one or both tufts. Symmetry Embryos in which one or two micromeres have been deleted almost always develop bilateral symmetry, as is characteristic of Muller's and three-quarter larvae. How- ever, following deletion of three micromeres, most of the larvae fall into the half-larva or sphere/swollen catego- ries, exhibiting either asymmetry or radial symmetry. Thus, the determination of embryonic axes resulting in a larva with bilateral symmetry may involve an interaction between the micromeres and a central cross-furrow mac- romere, and a minimum number of micromeres (i.e.. at least 2) must contact the central macromere for axis determination to occur. In these characteristics, Hoploplana appears to be similar to Patella (van den Big- gelaar and Guerrier, 1979) and Lymnaea (Martindale etal.. 1985). DEVELOPMENT OF HOPLOPLANA 343 Survival Though embryos consisting of only four micromeres are much smaller than those of four macromeres, they have much greater developmental potential than the lat- ter. Their survival rate to day five is much higher, and they differentiate structures such as eyes, cilia, and some- times tufts, though they almost always remain spherical. Wilson (1904), Horstadius (1937), and Costello (1945) observed similar development of isolated first quartets of Patella. Cerebratulus, and Nereis, respectively. The swollen syndrome that characterizes this type of embryo in Hoploplana may result from micromeres attempting to spread over macromeres that are not there. Isolated, intact first quartet macromeres (1A-1D em- bryos), on the other hand, seldom survive, and those that do are radially symmetrical and exhibit no recognizable differentiations, such as gut, eyes, or tufts. Development of eight-cell vegetal halves in Patella. Cerebratulus, and Nereis is similar, though these embryos sometimes gas- trulated. These results are in contrast to those of first quartet deletions in Bithynia (van Dam and Verdonk, 1982), in which the resulting larvae were bilaterally sym- metrical and differentiated many larval organs, though they were missing the head. Conclusions The Hoploplana embryo apparently develops axes of bilateral symmetry only when at least two micromeres are present, and for completely normal development to occur, at least three micromeres are required. While the blastomeres apparently do express cytoplasmic localiza- tion during early cleavage, positional differences between blastomeres also appear to play a role in divergence of developmental pathways. Therefore the polyclads, with slightly unequal cleavage, and a rather loose early embry- onic determination involving cytoplasmic localization, also demonstrate some complex cellular interactions during development. These studies suggest that polyclad flatworms could be an appropriate model for an ances- tral form that might have given rise to the two different developmental pathways characterizing present day higher Spiralia. Acknowledgments This work was supported by a grant from Research Corporation, NSF grant DCB-88 17760, and the Union College Faculty Research Fund. Literature Cited Arnolds, W. J. A., J. A. M. van den Biggelaar, and N. H. Verdonk. 1983. Spatial aspects of cell interactions involved in the determi- nation of dorsoventral polarity in equally cleaving gastropods and regulative abilities of" their embryos as studied by micromere dele- tions in Lymnaea and Patella. Roux'sArch. Dev. Biol. 192: 75-85. van den Biggelaar, J. A. M. 1976. Development of dorsoventral po- larity preceding the formation of the mesentoblast in Lvmnaea stagnalis. Prof KUH. Neil. Akad. H'el. C79: 1 13-126. van den Biggelaar, J. A. M. 1977. Development of dorsoventral po- larity and mesentoblast determination in Patella vulgata. J. Mor- pliol. 154: 157-186. van den Biggelaar, J. A. M., and P. Guerrier. 1979. Dorsoventral polar- ity and mesentoblast determination as concomitant results of cellular interactions in the mollusk Patella vulgata. Dev. Biol 68: 462-47 1 . Boyer, B. C. 1986. Determinative development in the polyclad turbel- larian Hoploplana inqiulma. Int. J linen. Repro. De\'. 9: 243-251. Boyer, B. C. 1987. Development of in vitro fertilized embryos ol the polyclad tlatworm Ilop/o/i/aiui inquilina following blastomere sepa- ration and deletion. Roitx'x Arch Dev Biol. 196: 1 58-164. Gather, J. N. 1973. Regulation of apical cilia development by the po- lar lobe of llyanassa (Gastropoda: Nassariidae). Malacologia 12: 213-223. Clement, A. C. 1967. The embryonic value of the micromeres in llya- nassa obsoleta as determined by deletion expenments. I. The first quartet cells. / Exp. Zoo/. 166: 77-88. Costello, D. P. 1945. Experimental studies of germinal localization in Nereis. I. The development of isolated blastomeres. J. Exp. Zoo/. 100: 19-46. van Dam, W. 1., and N. H. Verdonk. 1982. The morphogenetic sig- nificance of the first quartet micromeres for the development of the snail Bithynia tentaculala. Roux'sArch. Dev. Biol. 191: 1 12-1 18. Davidson, E. H. 1986. Gene Activity in Early Development, third edi- tion. Academic Press, New York. Horstadius, S. 1937. Experiments on determination in the early de- velopment of Cerebratulus lacteux. Biol. Bull. 73: 3 1 7-342. Martindale, M. Q., C. Q. Doe, and J. B. Morrill. 1985. The role of animal-vegetal interaction with respect to the determination of dor- soventral polarity in the equal-cleaving spiralian. Lymnaea palus- trix. Rou.\'x Arch. Dev. Biol. 194: 281-295. Morrill, J. B., C. A. Blair, and VV. J. Larsen. 1973. Regulative develop- ment in the pulmonate gastropod, Lymnaea paliistris, as determined by blastomere deletion experiments. / Exp. Zool. 183: 47-56. Surface, F. M. 1907. The early development of a polyclad, Planocera im/nilina. Proc. Acad. Nat. Sci. Plula. 59: 514-559. Wilson, E. B. 1904. Experimental studies on germinal localization. II. Experiments on the cleavage mosaic in Patella and Dentalium J Exp. Zool. 1: 197-268. Reference: Bin/. Bull 177: 344-349. (December. 1989) Cloned cDNA and Antibody for an Ovarian Cortical Granule Polypeptide of the Shrimp Penaeus vannamei JAMES Y. BRADFIELD, ROBERT L. BERLIN, SUSAN M. RANKIN, AND LARRY L. KEELEY Laboratories for Invertebrate Neuwendocrine Research. Department of Entomology, Texas A&M University, College Station, Texas 77843-2475 Abstract. A cloned cDNA was generated to a transcript for a major ovarian polypeptide (200 kDa) of the South American white shrimp, Penaeus vannamei. The cloned cDNA hybridized to a single transcript in ovaries but not to RNA from the hepatopancreas or muscle. For immu- nodetection and quantitation, a monospecific polyclonal antibody was raised against the cDNA translation prod- uct expressed in bacteria. The antibody was used to show that the 200 kDa ovarian polypeptide accumulated in cortical granules during ovarian development to com- prise ~ 1 1% of the total ovarian protein and disappeared during early embryonic development. These studies be- gin to explain a gene-product relationship essential for reproduction in P. vannamei. Introduction Cortical granules are large and abundant in the mature oocytes of penaeid shrimp (Duronslet et a/., 1975; Clark and Lynn, 1977; Clark et ai. 1980; Anderson et ai, 1984; Tom et a!., 1987; Bell and Lightner, 1988; Tan- Fermin and Pudadera, 1989). However, despite the prominence of these organelles, shrimp cortical granule synthesis and function have received little attention. His- tological studies have shown that cortical granules con- tain glycoproteins and lack lipids (Tan-Fermin and Pu- dadera, 1989), but the precise nature of the protein moi- eties or other potential components has not been determined. In the present studies we have quantified cortical gran- ule size and abundance in mature oocytes of the South American white shrimp, Penaeus vannamei; isolated a cloned cDNA for a major cortical granule polypeptide; and, after making a polyclonal antibody against the ge- Received 23 March 1989; accepted 22 August 1989. netically engineered cDNA translation product, quanti- fied the amount of that cortical granule polypeptide dur- ing ovarian and very early embryonic development. The cloned cDNA and its corresponding antibody are sensi- tive probes for studies on regulation of a gene that is abundantly expressed during ovarian development in P. vannamei. Materials and Methods Animals Broodstock female P. vannamei were obtained from Laguna Madre Shrimp Farm (Los Fresnos, Texas), and Sea Critters Inc. (Tavernier, Florida). Shrimp were main- tained in 2000-1 circular tanks [biologically filtered, aer- ated, artificial seawater (35%»)] at 28°C under a 16L:8D (dim light) photoperiod, and fed squid and oysters twice daily. Under these conditions, unilateral eyestalk abla- tion induces ovarian development within two weeks (Rankin et ai, 1989). Shrimp spawn samples were ob- tained from Granada Genetics Inc. (College Station, Texas). Ovarian cDNA library construction Total RNA was isolated (Chirgwin et ai, 1979) from ovaries in mid-development (~150 nm oocyte diame- ter), and poly(A)+RNA was selected by two rounds of oligodeoxythymidylate-cellulose chromatography (Aviv and Leder, 1972). Double-strand cDNA was prepared according to Gubler and Hoffman (1983), made blunt- ended with T4 DNA polymerase, and ligated to EcoRl/ Noll linker-adaptors (Invitrogen). cDNAs > 1.5 kb (~2 iug) were isolated by electrophoresis in agarose and glass powder adsorption ( Vogelstein andGillespie, 1979). The cDN As were ligated to DNA of bacteriophage Xgt 1 1 344 SHRIMP CORTICAL GRANULE cDNA AND ANTIBODY 345 (Young and Davis, 1983), and packaged with a commer- cial extract (Gigapack Plus, Stratagene Cloning Sys- tems). Propagation on Escherichia coli Y1088 gave a li- brary of 1 .3 X 106 primary recombinants. Isolation of a cloned cDNA representing an abundant ovarian transcript Two mixed cDNA probes were used to isolate an ovar- ian cDNA representing a highly expressed transcript for a high molecular weight translation product. The first probe was 3:P-cDNA (Schleif and Wensink, 1981) pre- pared from total ovarian poly(A)+RNA. The second cDNA probe was ^P-cDNA prepared from 5-7 kb poly (A)+RNA isolated from low-melting-temperature aga- rose gel. Duplicate plaque lifts on nylon filters (NEN Re- search Products) were hybridized with the two probes in 50% formamide, 4 X SSPE (20 X is 3.6 M NaCl, 0.2 M phosphate buffer pH 7.4, 20 mAl EDTA), 2% SDS, 0.5% nonfat dry milk at 42°C overnight, and washes were in 15 mMNaCl, 10 mA/Tris-HCl pH 7.4, 1 mM EDTA, 1% SDS at 60°C. Several recombinant bacteriophage hy- bridized strongly with both cDNA probes. A 3 kb cDNA insert from one of these recombinants was used for the experiments described below. Production of a cDNA-derived antibody The 3 kb ovarian cDNA (see above) was subcloned into the /^-galactosidase C-terminal coding region of the plasmid expression vector pUR 291 (Rutherand Miiller- Hill, 1983) for E. coli JM101 transformation. A fusion polypeptide (part bacterial /3-galactosidase and part shrimp cDNA translation product) was induced by addi- tion of isopropyl-fJ-D-thiogalactopyranoside to cell cul- tures in log-phase. Cellular polypeptides were separated by SDS-PAGE (Laemmli, 1970). The fusion polypeptide (205 kDa) was visualized with KC1, electroeluted (An- drews, 1986), and quantified according to Lowry et al. (1951). A polyclonal antibody to the gel-purified fusion polypeptide was raised in 3-month-old New Zealand white rabbits. Freund's complete adjuvant was included in the primary injection of ~600 ^g protein, and the non-pyogenic adjuvant T1501 (gift from Dr. L. F. Woodard, Dept. Vet. Sciences, University of Wyoming) was used in booster injections ( ~ 1 50 ^/g protein). Rabbit serum was diluted with 1 5 mM phosphate buffer pH 6.4, and the IgG fraction isolated with a prepacked diethy- laminoethyl-cellulose ion exchange column (Pharmacia LK.B Biotechnology) according to the supplier's instruc- tions. The IgG was lyophilized and stored at -80°C until use. Ovarian transcript characterization Total RNA was isolated from selected shrimp tissues, denatured with MeHgOH (Bailey and Davidson, 1976), and separated by electrophoresis in 1.2% agarose gel. RNA was transferred to a nylon filter (Zeta-Probe, Bio- Rad Laboratories) and hybridized with the nick-trans- lated (Rigby el al.. 1977) cloned ovarian cDNA under conditions described above for cDNA isolation. Hybrid- ization was visualized with autoradiography. hnnnmohlot analysis Polypeptides were separated by SDS-PAGE (7.5% gel) and transferred to nitrocellulose using a semi-dry elec- troblotting apparatus (American Bionetics) according to the manufacturer's recommendations. Polypeptides re- acting with the fusion polypeptide antibody were visual- ized with an alkaline phosphatase detection system (Bio-Rad). Histochemistry Ovary fragments were fixed in 4% formaldehyde in Ott's artificial seawater (Spotte, 1 979), embedded in par- affin, and sectioned (5 ^m) (Sheehan and Hrapchak, 1980). Immunoreactivity in sections was located with the Bio-Rad alkaline phosphatase detection system. Control samples were incubated with IgG from unin- jected rabbits. Determination of oocyte and cortical granule parameters were made on sections of 62 cells with nuclear diameter > 44 urn, using a light microscope ( 160x) with ocular micrometer. Analysis of cell surface sections (n = 42) was used to determine the number of cortical granules per cell. Enzyme-linked immunosorbent assay (ELISA) Samples in 60 mM carbonate buffer pH 9.6 were ad- sorbed to Immulon 2 Plates (Dynatech) at 4°C overnight. The plates were washed with 0.33 M phosphate-buffered saline pH 7.2 (PBS), and PBS adjusted to contain 0.00 1 % Tween-20. Wells were blocked with ovalbumin (5% in carbonate buffer), washed, and incubated in PBS with rabbit IgG (see below). Immunoreactive protein was as- sessed at 405 nm with a peroxidase-linked antibody to rabbit IgG. Forquantitation, varying quantities of fusion protein were run as standards against which the crude tissue extracts were compared. This quantitation re- quired two preliminary experiments: (1) adsorption of the antibody with varying concentrations of a pUR 29 1/ E. coli lysate. This experiment determined the concen- tration of cell lysate necessary to effectively remove the antibodies directed toward contaminating E. coli pro- teins. The most abundant contaminant was the 1 1 5 kDa /3-galactosidase portion of the fusion protein. Adsorption of the antibody was essential to ensure that only the epi- topes common to both the fusion protein and the shrimp ovary polypeptide were detected in the ELISA. Only ap- propriately adsorbed antibodies were used in subsequent 346 J. Y. BRADFIELD KT AL kb 95- 4.4- II 24- 1.4- 03- Figure 1. Total RNA samples (5 ng/\ane) from mid-reproductive cycle Penaeus rannamei tissues were denatured, separated by electro- phoresis in agarose, transferred to nylon membrane, and hybridized with the cloned 3 kb ovarian cDNA. Lane 1, ovary; lane 2, hepatopan- creas; lane 3, muscle. RNA size markers are indicated at left. experiments. (2) Incubation of extracts of various con- centrations of pre-vitellogenic ovaries and nearly mature ovaries with various concentrations of fusion protein. This experiment demonstrated that the fusion polypep- tide and immunoreactive native ovarian protein were adsorbed to the incubation wells with equal efficiency. We assumed that the common epitopes on the two different molecules (i.e., fusion protein and ovarian poly- peptide) behaved identically in the ELISA. Results Transcript characterization Northern hybridization was performed to determine (1) the size of the highly expressed transcript(s) repre- sented by the cloned ovarian cDNA and (2) which tissues expressed transcripts for production of the major ovarian polypeptide. The cDNA hybridized with a single 6 kb transcript from a mid-cycle ovary (oocyte diameter ~ 150 nm) (Fig. 1). The cDNA did not hybridize with RNA from muscle or hepatopancreas of the same animal. Genetically engineered fusion polypeptide A genetically engineered polypeptide consisting of shrimp ovarian polypeptide and bacterial 0-galactosi- dase was generated for subsequent immunological iden- tification of the shrimp ovarian cDNA translation prod- uct. For generation of this fusion polypeptide, the 3 kb cloned cDNA representing the 6 kb ovarian transcript was inserted into plasmid pUR 291 (Rutherand Miiller- Hill, 1983). This recombinant construct was used for high-level expression of the fusion polypeptide in bacte- ria. The fusion polypeptide consisted of ~ 1 1 5 kDa j3- galactosidase combined with a ~90 kDa P. vannamei ovarian polypeptide sequence (Fig. 2). Iinnninoblot analysis A polyclonal antibody raised against the fusion poly- peptide was used to determine the size and tissue distri- bution of the ovarian 6 kb mRNA translation product. Immunoblot analysis of tissues from a shrimp with mid- cycle ovaries showed strong reaction with a ~200 kDa ovarian polypeptide (Fig. 3). There were smaller, faintly reacting polypeptides in the ovary, and one small react- ing polypeptide (~35 kDa) from the hepatopancreas. The antibody did not react with muscle or hemolymph samples. Histochemical analysis Localization of the ovarian polypeptide was accom- plished by immunocytochemical analysis. Immunoreac- tivity was found in the cortical granules of the mature oocytes (Fig. 4). The cortical granules were prominent club-shaped corticular organelles, ~38 ^m long, 12 nm in basal diameter, and ~ 1 .4 x 103 ^m3 in volume calcu- lated according to conical shape. Analysis of cell surface sections indicated that there were ~420 cortical gran- 45- Figure 2. E co/i JM 101 was transformed with plasmid pUR 291, and plasmid-encoded ff-galactosidase was induced by addition of IPTG. Cell extracts were separated by SDS-PAGE (7.5%) and stained with Coomassie Blue R. Lane 1 , native pUR 29 1 ; lane 2, recombinant pUR 291 containing the 3 kb ovarian cDNA. The heavy band at 1 16 kDa in lane 1 is unfused fi-galactosidase. The band at 205 kDa in lane 2 is a fusion consisting of /j-galactosidase linked to a polypeptide en- coded by the ovarian cDNA. SHRIMP CORTICAL GRANULE cDNA AND ANTIBODY 347 1234 1234 MrXlO-3 205- 116- 97- 66- • I ! B Figure 3. Pcnacits vannamei polypeptides were separated by SDS- PAGE, and stained with Coomassie Blue R (Panel A), or blotted onto nitrocellulose and incubated with a polyclonal antibody to the geneti- cally engineered fusion polypeptide (Panel B). Samples were from an animal with mid-cycle ovaries (oocyte diameter ~ 150 ^m). Lane 1, ovary: 2, hemolymph; 3, muscle; 4, hepatopancreas. Positions of mo- lecular weight markers are indicated at left. ules/cell, comprising ~ 10% of the oocyte volume. These estimates are based on a cell diameter of ~226 ^m (Ta- ble I ). This cell diameter measurement was an underesti- mate of that in fresh tissue as a result of (a) cell shrinkage (from 320 ^m, see below) during sample preparation and (b) imprecision in determining the cell center due to vari- ations in cell morphology. Developmental profile oft he 200 kDa ovarian polypeptide The contribution of the 200 kDa polypeptide to total protein was determined by ELISA during both ovarian development and embryonic development using the pu- rified fusion protein as the standard. The 200 kDa poly- peptide was not detected in a previtellogenic ovary (oo- cyte diameter = 0) (Fig. 5). It represented 4-5% of the total ovarian protein at oocyte diameter 200-240 fim and ~ 1 1% of the protein in a fully developed ovary (320 pm). In newly spawned eggs, the 200 kDa polypeptide comprised 3-4% of the protein. By 2 h after the spawn, the 200 kDa polypeptide had declined to <1% and was not detected in >4 h spawn. By 6 h after spawning, the first antennal, second antennal, and mandibular primor- dia were visible, and hatching occurred 16-18 h after spawning. Discussion Cortical granules are found widely in vertebrates and invertebrates; however, the size, abundance, and compo- sition of these granules varies among species (see Guraya, 1982, for review). As shown by light and electron micros- copy, cortical granules in mature ovaries of penaeid shrimp are large and plentiful (Duronslet et ai. 1975; Clark and Lynn, 1977; Clark et a/., 1980; Anderson et al. 1 984; Tom etai. 1987; Bell and Lightner, 1988;Tan- Fermin and Pudadera, 1989). We showed that, in P. van- namei, the cortical granules were club-shaped structures, about 38 jim long, and occupied about 10% of the vol- ume in fixed, mature oocytes (Fig. 4, Table I). In mature ovaries of Penaeus aitecits. the somewhat smaller but Figure 4. Localization of the 200 kDa ovarian polypeptide using immunocytochemistry. Ovarian sec- tions (5 Mm) were incubated with (Panel A) preimmune rabbit IgG and (Panel B) IgG from the rabbit inoculated with the gel-purified fusion polypeptide (see Fig. 2). Immunoreaction was visualized with an alkaline phosphatase-linked second antibody. Arrows indicate cortical granules. Scale bars = 100 nm. 348 J. Y. BRADFIELD ET AL Table I Estimates of oocyte and cortical granule parameters in mature ovaries* <>/ Penaeus vannamei Parameter Size ± SD Oocyte diameter Oocyte surface area Oocyte volume 226 ± 16 pm 1.6 * 10- MITT 6.0 • 10" pin' Cortical granule length Cortical granule diameter (a) Cortical granule diameter (b) Vol. of cortical granules 37.8 ± 4.3pm 12.0± 1.4pm 7.5 ± 1.2pm 1.4 • 10' pm' No. cortical granules/cell Vol. cortical granules/cell Vol. conical granules/cell vol (%) 420 ± 70 5.9 x |()5 pm' -10% * Measurements were taken from 5 pm sections alter fixation, em- bedding, deparafinization, and rehydration. Sections of 62 cells with nuclear diameter > 44 pm were measured for the various parameters. To determine cortical granule dimensions, three granules per cell sec- tion were measured for length, apical (a), and basal diameter (b). more numerous cortical granules occupy about 12% of the oocyte volume (Clark et a/., 1980). Histological anal- ysis shows that cortical granules of Penaeus monoclon stain positively with alcian blue-periodic acid Scruff, in- dicating glycoprotein components, and negatively with Sudan black, suggesting an absence of lipids (Tan-Fer- min and Pudadera, 1989). The studies described in this paper resulted in a cloned cDNA and antibody for one highly expressed cortical granule polypeptide of the shrimp, P. vannamei. Using these probes, regulation of shrimp reproduction at the level of a single gene can be examined for the first time. Bacterial expression of the cloned cDNA (Fig. 2) allowed production of a monospecific antibody directed against the translation product. Immunoblot analysis showed that the major immunoreactive polypeptide (200 kDa) and several faintly reacting, smaller polypeptides were ei- ther immunologically related products of more than one gene, or, alternatively, natural or artifactual cleavage products of the 200 kDa polypeptide. We regard the lat- ter possibilities as the more likely since the ovarian cDNA hybridized to only a single ovary-specific tran- script (Fig. 1 ). Most likely, the immunoreactive polypep- tide in the hepatopancreas originated in the ovary and, through leakage and absorption, was detected subse- quently in the hepatopancreas. We demonstrated by immunocytochemical analysis that the 200 kDa polypeptide was localized in cortical granules (Fig. 4) and that it accrued from virtually unde- tectable levels in the immature ovary up to ~ 1 1% of the total protein in the mature ovary of P. vannamei (Fig. 5). Because it is strikingly abundant (Fig. 5) and relatively insoluble (Rankin et a/.. 1989), we speculate that the 200 kDa polypeptide contributes to the structural integrity of the cortical granules. This polypeptide is similar in size to a 180-193 kDa polypeptide found in cortical granules of the sea urchin Strongylocentrotus purpitratus (Kopf et a/., 1982; Villacorta-Moeller and Carroll, 1982). Whether sea urchin cortical granule polypeptides are im- munologically related to those of P. vannamei remains to be determined. Indeed, determination of similarities and differences between cortical granules within single oocytes is an area of active research. Among the crusta- ceans, ultrastructural observations indicated two types of cortical granules in the horseshoe crab, Limulus polyphe- /w«(Bannon and Brown, 1980); two in the fairy shrimp Tanymastix (Garreau de Loubresse, 1974); and four in the lobsters Homants americanus and H. gammarus (Talbot and Goudeau, 1988). In the sea urchin S. purpu- ratus, immunological analysis shows a marked heteroge- neity of distribution for a cortical granule polypeptide, suggesting differences between cortical granules in that species (Anstrom etal.. 1988). Cortical granule contents are thought to originate in the oocytes (see Guraya, 1982, for review). For example, in crustaceans, ultrastructural evidence indicates that at least one type of cortical granule, now termed "ring- shaped inclusions," of the crab Carcinus maenas (Goudeau, 1984) and the lobster Homarus (Talbot and Goudeau, 1988) are products of oocytic machinery. We are the first to apply molecular genetic analysis to dem- onstrate an ovarian origin for a major cortical granule polypeptide (Figs. 1-3). It is likely that this polypeptide is synthesized in the oocytes; however, this remains to be determined by in situ hybridization. The 200 kDa cortical granule polypeptide accumu- Oocyte diameter (urn) Spawn age (hr) Figure 5. Enzyme-linked immunosorbent assay of the 200 kDa ovarian polypeptide, in total protein from developing ovaries and spawn. The purified fusion polypeptide (Fig. 2) was the antigen and protein standard. The fusion polypeptide antibody was preadsorbed to an E. co/i lysate. SHRIMP CORTICAL GRANULE cDNA AND ANTIBODY 349 lated during ovarian development, from a virtually un- detectable level in a previtellogenic ovary to ~ 1 1 % of the total protein in a fully developed ovary. Within a few hours of spawning, the 200 kDa polypeptide had disap- peared (Fig. 5), presumably through exocytosis during the cortical reaction (see Clark el a/.. 1980). The cDNA and antibody that we have generated for this polypeptide can now be used to deduce the primary structure of a major component of the cortical granules and determine its site of synthesis precisely, to characterize the matura- tion of cortical granules; and to investigate the regulation of synthesis of this major ovarian component at the pre- translational and translational levels. Acknowledgments We thank Martha Lundberg, Pat Sistrunk, and Dr. John Ellison for technical assistance. This work was sup- ported by institutional grant NA 85AA-D-SD128 to Texas A&M University by the National Oceanic and At- mospheric Administration's Sea Grant Program, and by the Texas Advanced Technology Research Program. The research was conducted by the Texas Agricultural Experiment Station. Literature Cited Anderson, S. L., E. S. Chang, and VV. H. Clark Jr. 1984. Timing of postvitellogenic ovarian changes in the ridgeback prawn Sicyonia ;'ngtv»/s(Penaeidae) determined by ovarian biopsy. Aquaculnirt' 42: 257-271. Andrews, A. T. 1986. Electrophoresis. Theory. Techniques, and Bio- chemical and Clinical Applications. 2nd ed. Clarendon Press. Ox- ford. Anstrom, J. A., J. E. Chin, D. S. Leaf, A. L. Parks, and R. A. Raff. 1988. Immunocytochemical evidence suggesting heterogeneity in the population of sea urchin egg cortical granules. De\: Biol 125: 1-7. Aviv, H., and P. Leder. 1972. Purification of biologically active globin messenger RNA by chromatography on oligothymidylic acid-cellu- lose. Proc. Natl- Acad. Sci. USA 69: 1408-1412. Bailey, J. M., and N. Davidson. 1976. Methylmercury as a reversible denaturing agent foragarose gel electrophoresis. Anal. Biochem. 70: 75-85. Bannon, G. A., and G. G. Brown. 1980. Vesicle involvement in the egg conical reaction of the horseshoe crab. Limn/us polyphemus L. Dev. Biol. 76:410-417. Bell, T. A., and D. V. Lightner. 1 988. A Handbook of Normal Penaeid Shrimp Histology. Allen Press, Inc.. Lawrence, Kansas. Chirgwin, J. M., A. E. Przybyla, R. J. MacDonald, and \V. J. Rutter. 1979. Isolation of biologically active nbonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294-5299. Clark, VV. H., Jr., and J. VV. Lynn. 1977. A Mg* * dependent cortical reaction in the eggs of penaeid shrimp. J E.\p Zoo/. 200: 177-183. Clark, W. H., Jr., J. W. Lynn, A. I. Yudin, and H. O. Persyn. 1980. Morphology of the cortical reaction in the eggs of fcnacux in/can liiol Hull 158: 175-186. Duronslet, M. J., A. I. Yudin. R. S. Wheeler, and VV. H. Clark Jr. 1975. Light and fine structural studies of natural and artificially induced egg growth of penaeid shrimp. Proc. Meet. World Mart. SOL: 6: 105-122. Garreau de Loubresse, N. 1974. Etude chronologique de la mise en place des enveloppes de 1'oeuf d'un crustace phyllopode Tanyma\- li.\ laciimic J. Micros 20:21-38. Goudeau, M. 1984. Fertilization in a crab. III. Cytodifferentiation of vesicles enclosing ring-shaped elements involved in the cortical re- action. Gamete Res. 9: 409-424 Gubler, II., and B. J. Hoffman. 1983. A simple and very efficient method for generating cDNA libraries. Gene 25: 263-269. Guraya, S. S. 1982. Recent progress in the structure, origin, composi- tion, and function of cortical granules in animal eggs. Int. Rev. Cy- lol 78:257-360. Kopf, G. S.,G. W. Moy, and V. D. Vaquier. 1982. Isolation and char- acterization of sea urchin egg cortical granules. 7. Cell Biol. 95: 924- 932. Laemmli, U. K. 1970. Cleavage of structural proteins during the as- sembly of the head of bactenophage T4. Nature 277: 680-685. Lowry, O., N. Rosebrough, A. Farr, and R. Randall. 1951. Protein measurement with folin phenol reagent. / Biol. Chem 193: 265- 275. Rankin, S. M., J. V. Bradfield, and L. L. Keeley. 1989. Ovarian pro- tein synthesis in the South American white shrimp, Penaeus vanna- mei. during the reproductive cycle. Int. J Invert Reprod. Dev. 15: 27-33. Rigby, P. W. J., M. Dieckmann, C. Rhoads, and P. Berg. 1977. Labelling deoxyribonucleic acid to high specific activity in vitro by nick-translation with DNA polymerase I. J. Mol. Biol 113: 237-251. Ruther, I '., and B. Miiller-Hill. 1983. Easy identification of cDNA clones. EMBO ./ 2:1791-1 794. Schleif, R. F., and P. VV. VVensink. 1 98 1 . Practical Methods in Molec- ular Biology. Springer- Verlag. New York. Sheehan, D. C., and B. B. Hrapchak. 1980. Theory and Practice of Histology, 2nd ed., C. V. Mosby Co., St. Louis. Toronto and New York. Spotte, S. H. 1979. Fish and Invertebrate Culture: Water Manage- ment in Closed Systems, 2nd ed. Wiley Interscience, New York. Tom, M., G. Goren, and M. Ovadia. 1987. Localization of the vitellin and its possible precursors in various organs of Parapenaeus longi- rostris (Crustacea. Decapods. Penaeidae). Int. J Invert. Reprod. and Dev 12: 1-12. Talbot, P., and M. Goudeau. 1988. A complex cortical reaction leads to formation of the fertilization envelope in the lobster. Homarus. Gamete Res. 19: 1-18. Tan-Fermin, J. D., and R. A. Pudadera. 1989. Ovarian maturation stages of the wild giant tiger prawn. Penaeus monodon Fabricius. Aqiiacultwe 77: 229-242. Villacorta-Moeller, M. N., and E. J. Carroll Jr. 1982. Sea urchin em- bryo fertilization envelope: immunological evidence that soluble envelope proteins are derived from cortical granule secretions. Dev. Biol. 94:415-424. Vogelstein, B., and D. Gillespie. 1979. Preparative and analytical pu- rification of DNA from agarose. Proc: Natl. Acad. Sci. USA 76: 6 1 5- 619. Young, R. A., and R. VV. Davis. 1983. Efficient isolation of genes by using antibody probes. Proc. Natl. Acad. Sci. USA 80: 1 194-1 198. Reference: Biol. Bull 177: 350-355. (December, 1989) Larvae of a Nudibranch Mollusc (Phestilla sibogae) Metamorphose when Exposed to Common Organic Solvents J. TIMOTHY PENNINGTON1 AND MICHAEL G. HADFIELD Kewala Marine Laboratory, University of Hawaii, 41 Ainu St., Honolulu, Hawaii 96813 Abstract. Larvae of the nudibranch mollusc Phestilla sibogae metamorphosed when exposed to 10 of 14 or- ganic solvents. The active solvents included five alcohols and ethanolamine, acetonitrile, acetone, dichlorometh- ane, and toluene. Inactive solvents were ethylene glycol, DMSO, benzene, and hexane. These compounds span a wide range of polarities and contain a number of func- tional groups. Ethanol induced metamorphosis after 1- 5 days of exposure at 0.5-0.001 A/, and maximally in- duced about 65% of larvae to metamorphose in 3-5 days at 0. 1 A/. Ethanol was lethal to larvae above 0.75 M (ca. 4%). Methanol was lethal only above 1.75 A/ (ca. 1%), but produced less metamorphosis than ethanol at most concentrations. The natural inducer of metamorphosis in P. sibogae produced higher percentages of metamor- phosis more rapidly than did any of the solvents. The mechanism of metamorphic induction by the solvents is not known, but they probably interfere with a wide range of neuronal activities and trigger an existing metamor- phic pathway. Precompetent (young) larvae did not metamorphose in response to ethanol or methanol, but juveniles produced by exposure of competent (mature) larvae to ethanol or methanol survived to reproduce. Larvae of one other mollusc species also metamorphosed in response to ethanol, suggesting that larvae of other in- vertebrates may also be induced to metamorphose by or- ganic solvents. Larval biologists should be aware of this possibility. Introduction Many marine invertebrate species have complex life cycles, wherein a planktonic larval stage is both ecologi- Received 27 June 1989; accepted 30 August 1989. ' Present address: Hopkins Marine Station, Stanford University. Pa- cific Grove, CA 93950. cally and morphologically distinct from the following benthic juvenile and adult stages. The planktonic and benthic segments of these life cycles are joined by rela- tively rapid and often drastic metamorphoses (see papers in Chia and Rice, 1978). Such metamorphoses have been subjected to considerable study (reviewed by Meadows and Campbell, 1972; Crisp, 1974, 1976; Burke, 1983), as researchers have been interested both in the ecology of metamorphosis and in the physiological and morpho- genetic process of metamorphosis itself. In nature, metamorphosis is initiated by environmen- tal "cues" that are ecologically relevant; these cues in- duce larvae to metamorphose in sites where the probabil- ity of survival to adulthood is relatively high. For exam- ple, juveniles and adults of the nudibranch mollusc Phestilla sibogae Bergh eat only coral of the genus Po- rites, and larvae of P. sibogae metamorphose in response to a kairomone produced by these corals (Hadfield, 1977). However, pharmacological studies with P. sibo- gae and other invertebrates have revealed various neuro- active compounds that can also induce larvae to meta- morphose (Hadfield, 1977, 1984; reviewed by Burke, 1983). These compounds are of no apparent ecological relevance, but interfere with larval nervous systems and apparently activate pre-existing metamorphic pathways. Such "artificial inducers" of metamorphosis have been useful as pharmacological probes during the study of the control of metamorphosis. Workers in our laboratory have developed the Phes- tilla.Porites interaction as a system for the study of larval metamorphosis (reviewed by Hadfield and Pennington, in press). Here we present results of experiments with P. sibogae showing that organic solvents, and in particular ethanol and methanol, can serve as artificial inducers of metamorphosis. Additionally, we examine age-depen- 350 SOLVENT-INDUCED METAMORPHOSIS 351 dency of the response to ethanol and methanol, demon- strate that ethanol and methanol-metamorphosed nudi- branchs survive to reproduce, and show that larvae of at least one other gastropod species also metamorphose in response to ethanol. Materials and Methods Larval culture Routine larval culture methods for Phestilla sihogae have been fully described by Miller and Hadfield ( 1 986). Briefly, embryos and early veliger larvae were raised at 25°C in their egg masses until 5 days old, when they were manually hatched. Hatched veligers were maintained in seawater with antibiotics until "competent" to meta- morphose (see Hadfield, 1977, and below). Individual larvae of P. sibogae are either competent or not, but en- tire batches of larvae gradually become competent dur- ing days 7-9 of culture at 25°C. Larvae of Crucibulum spinositm were provided by J. L. Bell (University of Hawaii). The hatched veligers had been fed in unstirred beakers (see Bell, 1988) for 21 days when they were tested for metamorphic response to ethanol. Assays for metamorphic response to solvents Most assays were conducted in soap-washed or acid- cleaned (Galigher and Kozloff, 1964, p. 25) 2-ml plastic tissue culture wells (Fisher Cat. No. 08-772-1), but where noted, assays were performed in acid-cleaned glass slender dishes. Both types of containers were covered during experiments. In experiments in culture wells, about 20 larvae were pipetted along with 200 n\ of 0.22 ^m-filtered seawater (FSW) into each of 2 replicate wells, each containing 1 .8 ml of a given test solution. Test solu- tions consisted of organic solvents (analytical. HPLC, or photograde) dissolved in FSW, though where noted, a few comparisons employed MBL artificial seawater (Cavanaugh, 1956). The molar concentration of solvent in the final 2 ml of assay medium is reported. For those solvents relatively insoluble in water, a dilution series of solvent:seawater test solutions was vortexed and ob- served for disappearance of oily micelles. The most con- centrated test solutions lacking persistent micelles were used to prepare assays of these solvents; 5 M stocks were used to prepare assays of solvents sufficiently soluble in water. Larvae in the wells were counted and scored as meta- morphosed or not on each of several day's exposure to solvent. Larvae were determined metamorphosed when they lost their vela and larval shells, thus becoming juve- nile nudibranchs. All assays included both positive and negative controls. In negative controls, larvae were as- sayed as above in seawater alone, to control for any back- ground or "spontaneous" metamorphosis. In positive controls, a living chip of Forties compressa (1-9 mm1) was added to wells containing larvae and seawater to as- sess the competence of larvae to respond to the natural metamorphic inducer. Several series of larval assays were conducted with P. sibogae: ( 1 ) Detailed dose-response curves were constructed for both ethanol and methanol, spanning 5 orders-of- magnitude of alcohol concentration and 6 days of expo- sure, beginning with 1 1 -day-old larvae. Six experiments of two replicate wells per concentration were conducted with each alcohol. Several of these experiments addition- ally compared responses in FSW versus those in MBL artificial seawater. Another experiment in this series compared responses to ethanol in acid-cleaned glass dishes and the plastic culture wells. (2) In a second series of assays, 14 common organic solvents were surveyed for metamorphosis-inducing ac- tivity. These assays began with 11 -day-old larvae and were run for 2 days over wide solvent concentrations (from lethal or near-saturated to no discernible effect). (3) A third series of assays examined the effect of lar- val age (i.e., precompetent-to-competent larvae) on re- sponses to both ethanol and methanol. In these assays, larvae from the same culture but 6, 8, or 1 0 days old were exposed to coral, 0.5 M ethanol, 0.5 M methanol, or FSW alone and assayed for metamorphosis over the next 3 days. Four replicate wells of each of the alcohol treat- ments were used in this experiment. Sum'val and growth of solvent-metamorphosed juveniles An experiment was conducted to determine if juve- niles of P. sibogae resulting from solvent-induced meta- morphosis could survive and grow to adulthood. Eleven- day-old larvae were pipetted into culture wells contain- ingO.5 M ethanol, 0.75 M methanol, orO. 5 M methanol. Metamorphosed nudibranchs were removed from the solvent solutions over the next two days, counted, and transferred onto pieces of living coral. Juveniles from the different solvent solutions were placed on different pieces of coral, and each piece of coral was isolated in a small flow-through aquarium. The juveniles fed on the coral, grew, and became visible to the eye about 2 weeks after transfer. The nudibranchs were counted at this time and again at 24 days after transfer, when at least some in each basket had begun to lay eggs. Response o/"Crucibulum spinosum to ethanol Ten veligers of C. spinosum were pipetted into each of eight glass stender dishes. Two of the dishes contained 0.5 M ethanol in FSW, two contained 0.1 M ethanol in 352 J. T. PENNINGTON AND M. G. HADF1ELD 90-, 80- cn 'S 70- O 0. 60- o E soH o 1 ^ "c 30- 0.75 A/ (4%). (B) Response as a function of duration exposure to ETOH at 0. 1 , 0.0 1 , and 0.00 1 M; maximal larval response was assayed with a coral chip, and "spontaneous" metamor- phosis was monitored in wells containing larvae and seawater alone. FSW, and two contained 0.01 M ethanol and FSW. The final pair of dishes contained FSW alone. The larvae were scored for metamorphosis on each of the next 3 days; animals were determined metamorphosed when their vela had completely disappeared. Results Assays for metamorphic response to solvents Ethanol and methanol. Eleven-day-old veligersof/Vu'- stilla sibogae metamorphosed when exposed to both eth- anol (Fig. 1 A-B) and methanol (Fig. 2A-B). Very few of the larvae (<5%) metamorphosed over the 6 days of the experiments in filtered seawater alone, and 85-100% of the larvae metamorphosed within 1-2 days of exposure to a living chip of the coral Porites compressa (Figs. IB, 2B). Maximal response to ethanol was between these control values, with 60-80% oflarvae metamorphosing in 0.1 M ethanol (Fig. 1A). At least a few larvae meta- morphosed at all doses between 5 X 10~4 and 1 .0 A/, but larvae were always moribund and usually died in solu- tions at and above 0.75 M (ca. 4%) ethanol. Maximal response to methanol was lower, with 20-35% oflarvae metamorphosing in 0.1-1.0 M methanol (Fig. 2A). Effective doses ranged between 0.001 and 1.75 M, but a few metamorphoses occurred at lesser concentrations. 80- 70- 60- o .c Q. O E 50H o I 40H v 20- D_ 10 *— » Doy 6 •-« Doy 4 . « Day 2 10- 10- 10-J 10-' 10-' Concentration MEOH (molar) B o-G Carol «— • 0.5 M MEOH *— * 0.1 U MEOH •— • 01 M MEOH o — o Seowoter 100- 90- 80- 70- 60- 50- 40- 30- 20- 10- 123456 Exposure (days) Figure 2. Cumulative percent metamorphosis oflarvae ofPhestilla sihogae exposed to a series of methanol (MEOH) solutions for 6 days. Larvae were 1 1 days old at the beginning of experiments: plotted results are as described under Figure 1 . (A) Metamorphic response plotted as a function of MEOH concentration on Days 2, 4. and 6; larvae were moribund and most eventually died at concentrations > 1.75 A/ (7%). (B) Response as a function of duration exposure to MEOH at 0.5, 0.1, and 0.0 1 M: maximal larval response was assayed with a coral chip, and "spontaneous" metamorphosis was assayed in seawater alone. SOLVENT-INDUCED METAMORPHOSIS 353 Table I Organic no/rents ax.taycil for capacity lo induce metamorphosis oj Phestilla sibogae, listed hy decreasing polarity Compound Chemical family Maximum percent metamorphosis [molar cone.] Effective range (molar) Ethylene glycol diol none — Ethanolamine a mine 14 [0.01] 0.01 Methanol alcohol 44 [0.5] 0.001-1.75 Ethanol alcohol 34 [0. 1 ] 0.0005-0.75 n-Propanol alcohol 56 [0.05] 0.005-0.05 n-Butanol alcohol 33 [0.001] 0.001-0.01 Acelonitrile nitrile 35 [1.0] 0.1-1.0 DMSO sulfoxide none — — n-PentanoI alcohol 18 [0.002] 0.0005-0.005 Acetone ketone 83 [0.25] 0.05-1.0 Dichloromethane halidc 63 [0.3] 0.1-0.3* Benzene aromatic 3 [0.008] 0.008* Toluene aromatic 55 [0.003] 0.003* Hexane(s) alkane none — — * Indicates near-saturation. Larvae were moribund and usually died in methanol so- lutions above 1.75 M(ca. 7%). The time-course of the response to both ethanol and methanol was much slower than to coral. Very few larvae metamorphosed in response to ethanol during the first day of the experiments, with increasing percentages of metamorphosis on succeeding days (Fig. 1 B). Similarly, few larvae metamorphosed in methanol during the first day, with increasing percentages of metamorphosis over the second and third days (Fig. 2B). However, in meth- anol, few additional larvae metamorphosed after the third day. Percentages of larvae metamorphosing in response to the alcohols in FSW or MBL artificial seawater were sim- ilar (data not shown). Experiments conducted in acid- cleaned glassware also produced similar percentages of metamorphosis to those in plastic culture wells. How- ever, very few larvae metamorphosed in detergent- cleaned glass as compared to detergent-cleaned plastic or acid-cleaned ware of either material. Experiments were not conducted in detergent-cleaned glassware. Other organic solvents. Including ethanol and metha- nol, 14 organic solvents were surveyed for their capacity to induce metamorphosis, and at least 1 0 solvents did so. The solvents are listed in Table 1 in order of decreasing polarity or water solubility, along with the general chemi- cal family to which each belongs, the mean percent meta- morphosis produced by the most effective concentration (molarity) of a solvent, and the effective ranges of con- centration. Ethylene glycol, DMSO (dimethyl sulfoxide). and the hexane mixture did not produce any metamor- phosis; the 3% metamorphosis observed in benzene is probably also questionable. However, the remaining sol- vents all produced metamorphosis in several replicate wells, and most did so at a number of concentrations. The highest solvent concentrations that did not kill or clearly incapacitate the larvae generally produced the most metamorphosis. Response of precompetent veligers. Older, metamor- phically competent larvae metamorphosed when ex- posed to either ethanol or methanol, but younger, pre- competent larvae did not (Fig. 3A-C). When 6-day-old veligers were exposed to FSW or 0.5 M ethanol or metha- nol, essentially none metamorphosed over the next 3 days (days 6-9; Fig. 3A). At least some larvae became competent by days 8 and 9, as demonstrated by meta- morphosis in the presence of coral, but few larvae meta- morphosed in ethanol or methanol on these days, pre- sumably because of the lag in response to alcohols as de- scribed above. When 8- or 10-day-old larvae were treated similarly, larvae began metamorphosing in both alcohols by day 10 (Fig. 3B) or day 1 1 (Fig. 3C). In these latter assays, no larvae metamorphosed in FSW alone, but at least some larvae were competent to do so during the first day of the experiments as demonstrated by the responses to coral (Fig. 3B-C). Again, larvae responded to these alcohols more slowly and less strongly than they did to coral. Survival and growth of solvent-metamorphosed juveniles When veligers were induced to metamorphose with ethanol or methanol and then grown on living coral, over 50% survived until at least some nudibranchs began to 33 * Coral OETOH » MEOH • FSW _ -- * ' rs 0 • AjO — »fc ABO BO - AjO ABO 6 7 8 9 8 9 10 II 10 11 12 13 Days Old Figure 3. Cumulative percent metamorphosis of precompetent-to- competent larvae of Phcxtilla xibngac when exposed to either coral, 0.5 M ethanol (ETOH). 0.5 M methanol (MEOH). or filtered seawater (FSW) alone. Individual larvae are either metamorphically competent or not. but batches of larvae gradually become competent during days 7-9 at the culture temperatures used (25°C). Plotted results are means and S. D.'s of four replicate wells of each alcohol treatment and two replicates each of FSW and the coral treatment. (A) Response of 6-day- old, initially precompetent veligers. (B) Response of 8-day-old, mostly competent veligers. (C) Response of I0-day-old, fully competent veli- gers. 354 J. T. PENNINGTON AND M. G. HADF1ELD Table II Phestilla sibogae: survival of alcohol-induced nudibranchs until beginning oj egg-laying Initial no. Number Number Egg masses Metamorphic juveniles alive alive present? inducer (DayO) (Day 14) (Day 24) (Day 24) Coral 10 4 3 yes 0.25 A/ETOH 10 6 6 yes 0.5METOH 4 4 4 yes 0.75 A/MEOH 10 7 7 yes lay eggs (Table II). It was usually not possible to deter- mine which or how many of the nudibranchs in each basket had begun egg-laying. The observed survival rate was comparable to that observed when larvae were in- duced to metamorphose with coral (Table II), but there were no striking differences in survival among animals induced with the two ethanol doses or methanol. Nudi- branchs in all treatments appeared normal. These results indicate that ethanol and methanol induce a true meta- morphosis, producing juveniles that can grow to adult- hood. Response (>/ Crucibulum spinosum veligers to ethanol Veligers of C. spinosum metamorphosed by losing their vela when exposed to either 0.5 or 0.1 A/ ethanol for 1-3 days (Table III). No C. spinosum metamorphosed in either FSW or 0.01 M ethanol. Veligers in 0.5 M ethanol (ca. 1 .5%) did not swim actively, as did larvae in the less concentrated ethanol solutions or FSW alone, but re- mained partly retracted or swam weakly at the bottom of the dish. J. L. Bell has successfully repeated this experi- ment several times (pers. comm.). Discussion A wide variety of common organic solvents induce metamorphosis of competent larvae of Phestilla sibogae (Table I). For ethanol and methanol, the response is a true metamorphosis, because precompetent larvae do not "metamorphose" in response to these alcohols (Fig. 3), and because competent larvae induced to metamor- phose with ethanol or methanol can survive as juveniles and grow to adulthood (Table II). Other inducers of lar- val metamorphoses are suggested to be either ( 1 ) envi- ronmentally derived natural inducers, active at epider- mal receptors on larval surfaces, (2) neurotransmitters, their analogues or their precursors, presumed to be active either at surface receptors or internally on larval nerves, or (3) cations thought to cause sensory cell or neuron depolarizations that activate existing metamorphic path- ways (reviewed by Burke. 1983; Yool et ill.. 1986). Be- cause organic solvents do not normally occur in suffi- cient quantity to induce metamorphosis in natural sea- water, are not normally involved in regulation of nerve function, and are not electrically charged, they cannot fall into any of these categories. Instead, the solvents probably penetrate larval tissues and interfere with a wide range of nervous activities, somehow activating a metamorphic pathway. This explanation is in general agreement with what is known of the pervasive and often narcotic effects of organic solvents on mammalian ner- vous systems (reviewed by Browning, 1965). Because such a diversity of solvents successfully in- duced metamorphosis (Table I), specific functional groups of the solvent molecules are apparently not re- quired for the induction. However, three of the solvents tested produced absolutely no metamorphosis, contrary to what might be expected if the simple presence of dis- solved hydrocarbon was sufficient to induce metamor- phosis. When the solvents were arrayed in order of de- creasing polarity or water solubility (as in Table I), rela- tionships between solvent polarity and maximum percent metamorphosis or effective solvent concentra- tion were not apparent. Similarly, when arrayed by mo- lecular weight (not shown), no obvious relationships be- tween solvent molecular weight and percent metamor- phosis or effective concentration were apparent. These simple pharmacological considerations do not clarify the means by which the active solvents induce larvae to metamorphose. Nevertheless, it was generally true that the highest concentrations of solvents that were not obvi- ously toxic to larvae produced the highest percentages of metamorphosis. Other artificial inducers of metamorphosis of P. sibo- gae include choline chloride (Hirata and Hadfield, 1986) and excess K4 ions in seawater (Yool et al.. 1986). Addi- tionally, epinephrine produces "partial metamorpho- sis," wherein veligers of P. sibogae lose their vela but not their shells (Hadfield. 1 984). Choline maximally induces 60-70% of larvae to metamorphose at 3.75 X 10~3 Al. after a 2-3 day latent period during which little meta- morphosis occurs (Hirata and Hadfield, 1986). In con- Table III Ethanol-induced metamorphosis <>/ Crucibulum spinosum Percent metamorphosis Metamorphic inducer Day 1 Day 2 Day 3 Seawater 0 0 0 0.5 A/ETOH 5 45 75 0. 1 M ETOH 20 20 33 0.01 A/ETOH 0 0 0 Values are means of 2 dishes containing 10 larvae each. SOLVENT-INDUCED METAMORPHOSIS 355 trast, live coral usually induces nearly all larvae to meta- morphose within 48 h. The maximum percent metamor- phosis produced by choline, the range of effective choline concentrations, and the delayed response to choline are all similar to our present results with organic solvents. Maximally effective choline doses are also just beneath toxic levels, as observed here with many of the solvents. These similarities might suggest that the solvents and choline function in the same or a similar manner. Con- versely, the dose/response curves of the different solvents and choline are clearly different in some respects, and, unlike choline, the organic solvents cannot be involved in neurotransmitter biosynthesis (see Hirata and Had- field, 1986). Larval responses to both excess K+ (Yool el a/., 1986) and epinephrine (Hadfteld, 1984) are more rapid (i.e.. some response within 24 h) than to solvents or choline, and the maximum percentages of larvae to respond to K+ are higher (>90%) than for solvents or choline. These differences probably indicate that solvents, K+, and epi- nephrine operate via different mechanisms. Neverthe- less, solvents and K4 , at least, probably induce metamor- phosis through relatively nonspecific interference in ner- vous function. We have had the opportunity to test larvae of only one additional invertebrate species for a metamorphic re- sponse to an organic solvent. In these experiments, lar- vae of the gastropod Crucibulum spinosum readily meta- morphosed in response to ethanol. While these results are very limited, they suggest that competent larvae of other invertebrate species may also metamorphose in re- sponse to solvents. If so, ethanol in particular may prove of use to larval culturists as a widely available and rela- tively cheap and non-toxic inducer of metamorphosis. Conversely, in laboratories where organic solvents are commonly used, larval biologists should be aware that contamination of solutions by organic solvents may cause unwanted metamorphoses of competent larvae. Acknowledgments We thank Sister Phyllis Plantenberg for making the observations which led to this paper, L. Ho-Iseke and S. Miller for help in the laboratory, and J. Bell for larvae of Crucibulum spinosum. The work was supported by NSF Grant DCB8602 149 to M.G.H. Literature Cited Bell, J. L. 1988. Optimal feeding by gastropod larvae: patches and picoplankton. Am. Zoo/. 28: I67A. Browning, E. 1965. Toxicity ami Metabolism of Industrial Solvents Elsevier Pub., New York. 739 pp. Burke, R. D. 1983. The induction of metamorphosis of marine inver- tebrate larvae: stimulus and response. Can J. Zool. 61: 1701-1719. Cavanaugh, G. M. 1956. Formulae and methods 17 of the Marine Biological Laboratory Chemical Room. Woods Hole. Massachu- setts. Pp. 62-69. Chia, F. S., and M. E. Rice, eds. 1978. Settlement and Metamorpho- sis of Marine Invertebrate Lar\ae. Elsevier/North-Holland, New York. 290 pp. Crisp, D. J. 1974. Factors influencing the settlement of marine inver- tebrate larvae. Pp. 177-265 in Chemoreception in Marine Organ- isms. P. T. Grant and A. N. Mackie, eds. Academic Press. London. Crisp, D. J. 1976. Settlement responses in marine organisms. Pp. 83- 124 in Adaptations to Environment: Essavs on the Phvsiologv of Marine Animals. R. C. Newell, ed. Butterworths, London. Galigher, A. E., and E. N. Kozloff. 1964. Essentials of Practical Mi- crotechnique. Lea & Febiger, Philadelphia. 484 pp. Hadfield, M. G. 1977. Chemical interactions in larval settling of a marine gastropod. Pp. 403-4 1 3 in Marine Natural Products C 'hem- istry. D. J. Faulkner and W. H. Fenical, eds. Plenum Press, New York. Hadfield, M. G. 1984. Settlement requirements of molluscan larvae: new data on chemical and genetic roles. Aijiiaculture39: 283-298. Hadfield, M. G., and J. T. Pennington. In press. The nature of the metamorphic signal and its internal transduction in larvae of the nudibranch Phcstilla sihogae. Bull Mar Sci. 46(2). Hirata, K. Y., and M. G. Hadfield. 1986. The role of choline in meta- morphic induction of Phestilla (Gastropoda. Nudibranchia). Comp. Biochem. Physio/. 84C: 15-21. Meadows, P. S., and J. I. Campbell. 1972. Habitat selection by aquatic invertebrates. Pp. 27 1 -36 1 in Advances in Marine Biology. F. S. Russell and M. Yonge, eds. Academic Press, London. Miller, S. E., and M. G. Hadfield. 1986. Ontogeny of phototaxis and metamorphic competence in larvae of the nudibranch Phesttlla si- bogae Bergh (Gastropoda: Opisthobranchia). J E.\p. Mar. Biol. Eco/. 97: 95-1 12. Yool, A. J., S. M. Grau, M. G. Hadfield, R. A. Jensen, D. A. Markell, and D. E. Morse. 1986. Excess potassium induces larval meta- morphosis in four marine invertebrate species. Biol. Bull. 170: 255- 266. Reference: Biol. Bull 177: 356-362. (December, 1989) Population Structure, Larval Dispersal, and Gene Flow in the Queen Conch, Strombus gigas, of the Caribbean JEFFRY B. MITTON1, CARL J. BERG JR.:*. AND KATHERINE S. ORR:** 'Department of Environmental, Population and Organismic Biology, University of Colorado, Boulder, Colorado 80309, and -Marine Biological Laboratory, H 'oods Hole, Massachusetts 02543 Abstract. Genetic variation from 8 polymorphic en- zyme loci among 1 7 population samples of queen conch. Strombus gigas, exhibits similarity of allelic frequencies throughout the species distribution. Analyses of stan- dardized variances of allelic frequencies and of the fre- quencies of private alleles indicate that gene flow among populations in the Caribbean must be high. However, analyses of allelic frequencies clearly demonstrate that the populations are not panmictic. Bermuda is isolated from Caribbean populations, and there are numerous further examples of heterogeneity of allelic frequencies among populations within island groups. Limited data suggest that normal conch and samba, a slower growing, melanic form, are genetically differentiated. Introduction Gene flow, defined as the movement of gametes or in- dividuals from one place to another and incorporation of the genetic material into the recipient population, in- fluences both the population structure and geographic distribution of a species, as well as the adaptation of pop- ulations to their local environments (Slatkin, 1987). Gene flow is usually seen as a homogenizing force, pre- venting the differentiation of populations that exchange gametes or individuals (Mayr, 1963, 1970). Exchange of an average of just one individual per generation will pre- vent the fixation of neutral alleles arising from mutations within a population, regardless of its size (Wright, 1931). The importance of gene flow to population structure was Received 3 October 1988; accepted 1 8 September 1989. * Current address: Florida Marine Research Institute, Florida De- partment of Natural Resources, 13365 Overseas Highway, Marathon. Florida 33050. ** Current address: 297 Anglers Drive N., Marathon. Florida 33050. illustrated in a comparative study of the degrees of differentiation of populations of three species of Littor- ina differing in their modes of reproduction and poten- tial for gene flow (Berger, 1973, 1983). Larval L. littorea are planktonic, and hence have a high potential for gene flow. The egg cases of L. obtusata are attached to algae, and the algae may be detached from the substate and car- ried about by tides. Littorina saxatalis is restricted to the high intertidal and is ovoviviparous; because the eggs de- velop in the female until the juvenile adult stage, this spe- cies has little opportunity for gene flow. The potential for gene flow predicts the magnitude of genetic differentia- tion among populations. Populations of L. littorea are relatively homogeneous, populations of L. saxatalis are well differentiated, and populations of L. obtusata are intermediate in their degree of differentiation. It is difficult to measure gene flow directly, so many biologists infer levels of gene flow from distances of mi- gration or the potential for dispersal. But for several rea- sons these inferences can be seriously misleading. First, the movement of gametes and individuals may seriously overestimate gene flow. For example, pine pollen can be collected by ships 50 km from shore, yet studies of gene flow by pine pollen suggest that most genes move only a few dozen meters (Levin and Kerster, 1974). Pollen moving great distances may not reach receptive surfaces, or, having reached receptive surfaces, may have lost via- bility. Similarly, the flight of Euphydryas butterflies at- tests the potential for long distance dispersal, yet empiri- cal studies reveal little or no gene flow among popula- tions (Ehrlich et a/.. 1975). Clearly, the distances that individuals can move and the distances that genes typi- cally move can be profoundly disparate. Second, the homogenizing effect of gene flow may be overridden bv natural selection (Endler, 1973, 1977). A 356 GENE FLOW IN THE QUEEN CONCH 357 clear example has been described for the leucine amino- peptidase (Lap) polymorphism in the blue mussel, Alyti- /z30 min) for recovery from previous test stimuli. Neverthe- less, no escape withdrawal was observed in response to such stimuli, suggesting that the worm's responsiveness to moving shadow or abruptly decreased light intensity is dependent on its behavioral state. Escape responses in body fragments Isolated body fragments were screened for behavioral responsiveness to a moving shadow. The results (Fig. 5) showed that nearly half of the tail fragments, but no ante- rior or middle fragments, rapidly shortened in response to at least one of three test stimuli delivered at 30 min intervals. Failure of some test stimuli to evoke responses in otherwise responsive fragments did not appear to re- sult from insufficient recovery time. Rather, the lack of responsiveness appeared to be related to variations in the behavioral state of the fragment. For example, we noted that if tail fragments were quiescent just before testing, then rapid escape responses were often evoked. How- ever, if fragments were crawling or wriggling just before testing, then no rapid escape responses were evoked. These observations suggest that responsiveness to shadow may be reduced during such movements. Next, electrical correlates of rapid escape movements were examined in tail fragments exhibiting shadow re- sponsiveness. A very light touch stimulus anywhere on the fragment evoked one or more LGF spikes. As in intact worms, a single spike was never accompanied by a detectable muscle electrical response or movement (Fig. Frame no. 0 2 mm 10 11 Elapsed time (ms) 0 33 300 333 367 Figure -4. Frame-by-frame analysis of an escape response to an abrupt decrease in light intensity. Frame 0 shows the worm's tail lying at the air-water interface before onset of the stimulus. By the next frame ( 1 ) the lighting from above had been switched off, and between frames 9-1 1 the worm's tail rapidly withdrew from the water surface. OLIGOCHAETE ESCAPE REFLEX 367 ANTERIOR MIDDLE) POSTERIOR; * of fragments with shadow reflex 0/20 * of + tests for shadow reflex 0/60 0/30 0/90 31/68 65/204 Figure 5. Results from behavioral screening tests with isolated frag- ments from anterior, middle, and posterior locations. 6A), but two or more closely spaced spikes usually evoked such responses (Fig. 6B). The latency from a touch stimulus to the first LGF spike ranged from 3- 12ms. A moving shadow or abrupt decrease in light intensity also reliably evoked multiple LGF firing and muscle re- sponses (Fig. 6C). resembling those obtained in intact worms. The minimal rate of shadow movement required for reliably evoking escape shortening was determined by passing a shadow at varying speeds over the tail frag- ments. The mean minimal rate was 4.4 mm/s ± 2. 1 SD (n = 7 worms). This value appeared unaffected by changes in the orientation of the shadow relative to the longitudinal axis of the exposed tail. The latencies of electrical and behavioral responses to abrupt decreases in light intensity were also determined, as described for intact worms. The mean latency from stimulus onset to the first LGF spike was 375 ms ± 96 SD (n = 23 measurements; 13 worms). A videotaped se- quence of the behavioral response to the same stimulus is shown in Figure 7. The mean latency to the onset of such responses was 449 ms ± 67 SD (n = 24 measure- ments; 9 worms). Thus, although latencies to electrical and behavioral responses were somewhat longer in tail fragments than intact worms, the same general pattern of LGF responsiveness to photic stimulation was seen in both tail fragments and intact worms. Since results from isolated tail fragments suggested that the sensitivity of escape responses to shadow was lo- calized in tail segments, these fragments were examined for candidate photoreceptor cells. Candidate photo- and mechanoreceptors in posterior fragments Since physiological and behavioral results from iso- lated body fragments suggested that tail fragments were particularly responsive to shadow (as well as touch) these fragments were microscopically examined for candidate photoreceptors and mechanoreceptor cells. Light microscopy and scanning electron microscopy revealed no obvious ocellar-like structures on the cuticu- lar surface, although many complex arrangements of long and short ciliary processes, intermingled with large numbers of short microvillar projections, were evident on the cuticular surface (Fig. 8A). However, transmis- sion electron microscopy of the epithelial surface re- vealed candidate photoreceptor cells. These were similar to the simple phaosomal-type photoreceptors described in the anterior segments of leeches and earthworms (re- views by Sawyer, 1 986, and Jamieson, 1981). These cells were located on the dorsal surface of the tail and were sparsely distributed. Usually two, but occasionally as many as four, were found per segment. Cursory observa- tions in mid-body segments of three animals revealed no evidence of comparable cells. Figure 8B-C shows the candidate photoreceptors from two different worms. The cells have a large apical phao- some (varying in size from 1.5 to 4.0 nm) with a broad opening to the external cuticle. The phaosome contains numerous microvilli (0.10 to 0.1 5 ^m in diameter), most projecting centrally into the phaosome cavity and a few projecting through the phaosomal opening and into the cuticular layer. Other features of these cells include: (1) extensive tight junctions with surrounding epithelial • 1~ ^v^~4 Figure 6. LGF responses in isolated tail fragments. (A) One LGF spike was evoked by a mechanical stimulus (arrow). (B) Two LGF spikes, followed by a presumed longitudinal muscle potential (dot), were evoked by another, slightly stronger mechanical stimulus. (C) A train of LGF spikes and muscle response were evoked by a moving shadow. Time scale: 5 ms(A, B); 10ms(C). Voltage scale: 368 C. D. DREWES AND C. R. FOURTNER 2 mm Frame no. Elapsed time (ms) DARK 33 433 466 .500 .... 533 Figure 7. Frame-by-frame videotape sequence of rapid shortening in an isolated tail fragment in re- sponse to an abrupt decrease in light intensity. Frame 0 shows the tail fragment before onset of the stimulus. Rapid shortening occurred between frames 1 3 to 16. cells (Fig. 8C); (2) a basal nucleus; and (3) basal projec- tions into a radial nerve plexus directly beneath the epi- thelial layer. Thin sections through the body wall also provided de- tails regarding the ciliated epithelial cells (Fig. 8D, E). Both uniciliate and multiciliate cells were evident. The multiciliate cells generally have four to six cilia with a corresponding number of basal bodies. In the uniciliate cells, the cilium arises from a pit in the apical region of the cell and is surrounded by 10 microvillar projections, each containing an abundance of microfilaments and an electron-dense region adjacent to the ciliary process. The microvilli (approximately 0.3 ^m in cross-sectional area and 0. 1 5 /urn in width) are also interconnected via an or- ganized, electron-dense material (Fig. 8E). These two cil- iated cell types are nearly identical to the proposed epi- thelial mechanoreceptors described in the anterior seg- ments of earthworms (Knapp and Mill, 1971; Mill, 1982) and an aquatic oligochaete, Rhynchelmis linio- sella (Moritz and Storch. 1971). the latter representing the same family as Lnmbriculus (i.e.. Lumbriculidae). Discussion Adaptive significance and timing oj the shadow reflex Stimulus modalities that elicit rapid tail withdrawal in L. variegatus include touch, moving shadow, and abrupt decrease in light intensity. These modalities, as well as the specific escape movements they elicit, appear well matched to the worm's specific lifestyle and habitat. The worms are especially abundant in the shallow margins of ponds where their tails protrude several centimeters above the sediments and lie horizontally at the air-water interface (Fig. 1 ), a position that apparently facilitates gas exchange via the dorsal blood vessel. In this position, the exposed tail would be vulnerable to attack by subsurface, surface, or aerial predators. Because pond water is often highly turbid, and candi- date photoreceptor cells in exposed segments tend to be located in a dorsolateral position, mechanosensory, rather than photosensory, cues may be more important in the detection of subsurface predators (e.g.. aquatic in- sects or fish). On the other hand, photosensory cues may be important in signalling the approach of surface or ae- rial predators (e.g., amphibians or birds). This idea is consistent with observations that moving shadows above the water surface reliably elicit rapid tail withdrawal, ei- ther in small laboratory aquaria containing natural sedi- ments (Fig. 4), or in actual field settings (C. Drewes, unpub.). Although the electrophysiological and behavioral events during responses to photic stimuli appeared indis- tinguishable from those elicited by mechanosensory stimuli, the timing of escape responses to these two stim- ulus types differed markedly with respect to onset la- tency. The latency between a mechanical stimulus and the onset of LGF spiking was usually less than 12 ms, a value comparable to those in other terrestrial and aquatic oligochaetes (Drewes, 1984; Zoran and Drewes. 1987). In contrast, latencies following a photic stimulus were much longer, ranging from 250-375 ms. Such values are similar to those seen in other annelid escape responses to photosensory stimuli, such as giant fiber responses to shadow in polychaetes (Gwilliam, 1969) and S-cell re- sponses to light Hashes in leech (Laverack, 1969; Bagnoli elai. 1973). Several possible factors could contribute to these rela- tively long latency values. (1) The response time of the photoreceptors may be relatively slow. For example, in leech photoreceptor cells, the time from onset of a photic OL1GOCHAETE ESCAPE REFLEX 369 I Figure 8. Scanning and transmission microscopy of body wall of the tail of L. \aricKiilits. The scanning image (A) reveals a rough cuticular surface and an abundance of microvilli and cilia protruding from the surface. Note the long cilia (arrows) lying parallel to the surface. The cross-section (B) and frontal section (C) through the phaosomal region of the presumed photoreceptor reveals numerous microvilli (m) project- ing into the phaosome cavity (p) and cuticle (c). The presumed photoreceptors are linked to adjacent epithelial cells via tight junctions (arrows in C and D). A frontal section (D) of the body wall shows unicil- iary(uc)and multiciliary cells (me). Note the basal bodies (b) of the multiciliary cell. A cross-section (E) of the cilium from a uniciliary cell reveals the standard 9 + 2 cilium surrounded by ten microvillar processes. Adjacent microvilli are connected via a circular complex of extracellular fibers (e). Scale bars: A, C, 2.0 nm; B. D, 1 .0 /am; E, 0.25 /jm. stimulus to the first spike in the photoreceptor cells is approximately 80 ms(Lasansky and Fuortes, 1969; Fior- avanti and Fuortes, 1972; Peterson, 1984). (2) Conduc- tion and synaptic transmission from the peripheral pho- toreceptors to through-conducting fibers in the central nervous system may be time-consuming, especially if 370 C. D. DREWES AND C. R. FOURTNER pathways are small in caliber and include interposed sen- sory interneurons, as proposed in the leech (Kretz et ai. 1976). (3) Additional time may be consumed for spatial integration and parallel processing of segmental inputs onto giant fibers. These factors, although detrimental in terms of in- creasing response time, may be advantageous by provid- ing an opportunity for appropriate modulation of escape response sensitivity. For example, our results suggest that escape response sensitivity depends on the worm's be- havioral state, with tail segments being especially respon- sive to shadows when positioned horizontally at the air- water interface. This suggests that specific combinations of sensory cues may somehow interact to modulate es- cape responsiveness, a phenomenon that has been exper- imentally demonstrated for converging photosensory and mechanosensory inputs onto S-cell interneurons in the leech (Bagnoli etai, 1973). Candidate photoreceptor cells Photoreception in oligochaete worms is predomi- nantly mediated by extraocular. dermal photoreceptors (Steven, 1963; Yoshida, 1979; Welsch et ai. 1984). Evi- dence from two lumbricid earthworm species, Lumbri- cus terrestris and Eiseniafoetida, indicate that the photo- receptor cells have a characteristic microvillar organiza- tion. That is, the receptor cell contains a large, membrane-bound cavity (termed the "phaosome") which is lined by microvilli (review by Jamieson, 1981). Our results (Fig. 8) indicate that this same type of recep- tor cell exists in the epidermis of tail segments in the aquatic oligochaete, L. variegatus. Because isolated tail fragments are capable of reacting to decreased light in- tensity (Fig. 5), and because we have been unable to find other candidate photoreceptor cells in these fragments, we infer that the phaosomal photoreceptors somehow mediate the worms' reactions to these stimuli. The ultrastructural organization of these epidermal photoreceptor cells closely resembles that of the earth- worm, Eiseniafoetida, in having a relatively apical posi- tion in the epidermis, a free space in the center of the phaosome, and a microvillar-lined canal, which joins the phaosomal cavity with the extracellular space beneath the body wall cuticle (Hirata et ai, 1969). These features are also evident in photoreceptors of the medicinal leech (review by Sawyer, 1986) but not in the common earth- worm, Lumbricus terrestris. In the latter case, the micro- villar photoreceptor cells are located in the basal portion of the epidermis and the phaosome does not open to the outside (Rohlich el ai, 1970; Myhrberg, 1979). Based on microelectrode studies of leech photorecep- tor cells (Lasansky and Fuortes, 1969; Fiorvanti and Fu- ortes, 1972; Peterson, 1984), an abrupt increase in light intensity causes a large depolarizing receptor potential and marked increase in action potential firing in the cell. Since the adequate stimulus for eliciting an escape re- sponse in L. vuriegatus is an abrupt decrease in intensity, an interesting question arises as to the polarity of its pho- toreceptor electrical response. If the polarity is identical to that in the leech, then a shadow stimulus would be expected to hyperpolarize the receptor cell and excita- tion of the LGF system would therefore necessitate a step of disinhibition somewhere along the afferent pathway. Progress in resolving this question will require develop- ing a reliable means of identifying and recording from the photoreceptor cells in dissected preparations. This may be difficult in view of the relatively sparse distribu- tion of receptors cells in the epidermis and the proclivity for segmental autotomy when attempting to dissect this species. On the other hand, the associated high capacity for segmental regeneration by body fragments, in combi- nation with a segmental respecification (morphallaxis) during regeneration (Drewes and Fourtner, in prep.), may offer unusual opportunities for investigating behav- ioral, anatomical, and physiological correlates of devel- opmental plasticity in the shadow reflex of this species. Acknowledgments Some of this research was carried out at Iowa Lakeside Laboratory and we thank Dr. R. V. Bovbjerg, Director, for generously providing space and services there. We also thank A. J. Siegel, P. Bush, and R. Barone for assis- tance with electron microscopy. Literature Cited Bagnoli, P., M. Brunelli. and F. Magni. 1973. Afferent connections to the fast conduction pathway in the central nervous system of the leech llmulomcilianalm. Arch- Hal. Biol. Ill: 58-75. Clark, R. B. 1960. Habitation of the polychaete Nereis to sudden stimuli. I. General properties of the habituation process. Anim. Be- hiiv .8:83-91. Darwin, C'. 1881. The formation of vegetable mould through the ac- tion of worms, with observations of their habits. Murray, London. Dorset!, D. A. 1978. Organization of the nerve cord. Pp. 1 15-160 in Physiology of. Annelids. P. J. Mill, ed. Academic Press, New York. Dorsett, D. A. 1980. Design and function of giant fibre systems. Trends Neiiri>sci. 3: 205-208. Drewes, C. D. 1984. Escape reflexes in earthworms and other anne- lids. Pp. 43-91 in Neural Mechanisms of Startle Behavior. R C. Eaton, ed. Plenum, New York. Drewes, C. D., C. A. Callahan, and VV. M. Fender. 1983. Species spec- ificity of giant nerve fiber conduction velocity in oligochaetes. ( \in J. Zool .61:2688-2694. Evans, S. M. 1969. Habituation of the withdrawal response in nereid polychaetes. I. The habituation process in Nereis diversicolor. Biol. Bull 137:95-104. Fioravanli, R., and M. G. F. Fuortes. 1972. Analysis of responses in visual cells of the leech../ Physiol. 227: 173-194. Gnilliam, G. F. 1969. Electrical responses to photic stimulation in OLIGOCHAETE ESCAPE REFLEX 371 the eyes and nervous system of nereid polychaetes. Bint. Bull 136: 385-397. Hess, VV. N. 1925. Photoreceptors of Lwnbru its terrestris. with spe- cial reference to their distribution, structure and function. ,/ Mor- pAo/. 41: 63-93. Hirata. K., N. Ohsako, and K. Mabuchi. 1969. Fine structure of the photoreceptor cell of the earthworm, Eiscnia toelida. Rep. Fac. Sci. KtiKoshima L'nir. (Earth Sci. Biol.) 2: 127-142. Howell, C. D. 1939. The response to light in the earthworm. Pheret- ima agrestis Goto and Hatai, with special reference to the function of the nervous system. J. Exp. Zoo/ 81:231 -259. Jamieson, B.C. M. 1981. TheUltrastructureoftheOligochaeta. Aca- demic Press, New York. Knapp, M. F., and P. J. Mill. 1971. The fine structure of ciliated sen- sory cells in the epidermis of the earthworm Lumbricus terrestris. Tissue Cell 3: 623-636. Krasne, F. B. 1965. Escape from recurring tactile stimulation in Branchiomma vesiculosum. J Exp. Biol. 42: 307-322. Kretz, J. R., G.S. Stent, and \V. B. kristan. 1976. Photosensory input pathways in the medicinal leech. / Comp. Physiol 106: 1 -37. Lasansky, A., and M. G. F. Fuortes. 1969. The site of electrical re- sponses in visual cells of the leech, Hirudo medicinalis. J. Cell Bio/. -42:241-252. Laverack, M. S. 1969. Mechanoreceptors, photoreceptors and rapid conduction pathways in the leech, Hirudo medicinalis. J. Exp. Biol. 50: 129-140. Mill, P. J. 1975. The organization of the nervous system in annelids. Pp. 21 1-264 in "Simple" Nervous Systems. P. N. R. Usherwood and D. R. Newth, eds. Edward Arnold. London. Mill, P. J. 1978. Sense organs and sensory pathways. Pp. 63-1 14 in Physiology of. •{nnelids. P. J. Mill, ed. Academic Press, New York. Mill, P. J. 1982. Recent developments in earthworm neurobiology. Comp. Biochem Physiol. 73A: 641-661. Moritz, K., and V. Storch. 1971. Elektronenmikroskospische Unter- suchungen eines Mechanorezeptors von Evertebraten (Priapuliden, Oligochaeten). / Zellforsch. Mikro. Anal. 117: 226-234. Myhrberg, H. K. 1979. Fine structural analysis of the basal epidermal receptor cells in the earthworm (Lumbrieus lerrestris). Cell Tiss. Res. 203: 257-266. Nicol, J. A. C. 1948. The function of the giant axon of Mvxtcola 111- luntlihnlnm. Cancul J Res 26: 2 1 2-222. Nicol, J. A. C. 1950. Responses of Branchiomma vesiculosum ( Mon- tagu) to photic stimulation. J. Mar Biol. Assoe. U. A'. 29: 303-320. Nomura, F. 1926. Effect of light on the movements of the earthworm, Allolobophom toeiulu (Sa\.). Sci. Rep Tohoku Imp. Univ. I\ 1: 294-409. Peterson, E. L. 1984. Photoreceptors and visual interneurons in the medicinal leech. / Neiimbiol. 15: 413-428. Rohlich, P., B. Aros, and S. Viragh. 1970. Fine structure of photore- ceptor cells in the earthworm, Lumbricus terreslris. Z. Zellforsch. Mikro. Amu. 104: 345-357. Sawyer, R. J. 1986. Leeeh Biology and Behavior. I'D/. /. Anatom\; Physiology, and Behavior. Oxford LIniversity Press, New York. Steven, D. M. 1963. The dermal light sense. Biol. Rev. 38: 204-240. Unteutsch, \V. 1937. Uber den Licht- und Schattenreflex des Re- genwurms. Zoo/. Jahrbuch. Allg. Zoo/. Phyxiol. TiereSS: 69-1 12. Welsch, U., V. Storch, and K. S. Richards. 1984. Epidermal cells. Pp. 269-296 in Biology of the Integument, I'ol I Invertebrates. J. Bereiter-Hahn, A. G. Matolsty and K. S. Richards, eds. Springer- Verlag, New York. Yoshida, M. 1979. Extraocular photoreception. Pp. 581-640 in Handbook of Sensory Physiology. Vol. VII/6A, H. Autrum, ed. Springer-Verlag, New York. Zoran, M. J., and C. D. Drewes. 1987. Rapid escape reflexes in aquatic oligochaetes: variations in design and function of evolution- arily conserved giant fiber systems. J. Comp. Phvsiol.(k) 161:729- 738. Zoran, M. J., and C. D. Drewes. 1988. The lateral giant fibers of the tubificid worm, Brancluura sowerhyi: structural and functional asymmetry in a paired interneuronal system. / Comp. Neurol. 275: 76-86. Reference: Biol. Bull 177: 372-385. (December, 1989) Autotrophic Carbon Fixation by the Chemoautotrophic Symbionts of Riftia pachyptila CHARLES R. FISHER1, JAMES J. CHILDRESS', AND ELIZABETH MINNICH2 1 Oceanic Biology Group, Department of Biological Sciences and Marine Science Institute. University of California. Santa Barbara, California 93106: and 2Panlabs, 11804 N. Creek ParkwayS., Bothell, Washington 98011 Abstract. Preparations of trophosome tissue from Rif- tia pachyptila containing viable endosymbiotic bacteria were incubated with several substrates under a variety of conditions to characterize the symbionts physiologically. Of all the potential substrates tested, only sulfide stimu- lated carbon fixation by the trophosome preparations; neither hydrogen, ammonia, nor thiosulfate were effec- tive. Trophosome preparations did not oxidize UC- methane to either '4C-organic compounds or 14CO2, nor did they reduce acetylene under the conditions tested. Carbon fixation by the endosymbionts appears barotol- erant. The symbionts require both sulfide and oxygen to fix carbon through autotrophic pathways, but are inhib- ited by free oxygen and by sulfide concentrations in the 300 n\l range. Maximal rates of carbon fixation were documented in incubations in dilute Riftia blood, which protects the symbionts from the inhibitory effects of free sulfide and oxygen while providing them with an abun- dant pool of both substrates, bound by the vestimentif- eran hemoglobins. Introduction Symbioses between marine invertebrates and endo- symbiotic Chemoautotrophic sulfur bacteria were first discovered at deep sea hydrothermal vents (Cavanaugh cl al.. 1981; Felbeck, 1981). Similar associations are now well documented in several phyla of worms and in five molluscan orders found in a variety of habitats where reduced chemical species are present (Fisher, 1990). In all groups except the Vestimentifera, symbionts are housed in close proximity to the external environment. They are found in gill cells (most molluscs), in internal Received 2 August 1989; accepted 25 September 19S9. cells in very small animals (pogonophorans), or extracel- lularly, on the surface of the animal (oligochaetes, alvi- nellids, and thyasirids). These symbionts are therefore in close contact with the necessary metabolites carbon di- oxide, sulfide, and oxygen (Fisher, 1990). In contrast, the vestimentiferans examined to date harbor abundant en- dosymbiotic, sulfide-oxidizing, Chemoautotrophic bac- teria in an internal organ — the trophosome. This organ is highly vascularized and is located in the trunk of the worm; it has no close connections to ambient seawater (Jones, 1981). The trophosome comprises about 15% of the wet weight of the hydrothermal vent tubeworm, Riftia pa- chyptila, and hemoglobin-containing vascular and coe- lomic fluids account for at least another 30% (Childress et al., 1984). The bacterial volume is between 15 and 35% of the total volume of the trophosome, and esti- mates of bacterial density range from 3.7 to 10 x 104 cells/g trophosome (Cavanaugh et al., 1981; Powell and Somero, 1986). Because this organ is located in a coe- lomic cavity in the trunk of the animal, metabolites must be transported to the bacteria through the circulatory system. Vestimentiferan hemoglobins, found in both the hemolymph and coelomic fluid, bind hydrogen sulfide and oxygen independently and reversibly, preventing spontaneous oxidation of the sulfide while transporting it to the trophosome for use as an electron donor by the Chemoautotrophic endosymbionts (Arp and Childress. 1983; Childress et al., 1984; Fisher and Childress, 1984; Arp et al., 1987). The high affinity of the hemoglobins for sulfide also protects the animal cytochrome c oxidase system from poisoning by this potentially toxic molecule (Powell and Somero, 1983, 1986). Few studies have been done on the physiology of vesti- mentiferan symbionts, most likely because it is difficult 372 CARBON FIXATION BY RII-'TIA SYMBIONTS 373 to obtain living material. Belkin cl al. (1986) demon- strated that the symhionts from several individuals of R. pachyptila used sulfide, and not thiosulfate, as an elec- tron donor to fuel chemoautotrophic carbon fixation. Fisher el al ( 1988a) reported sulfide stimulated carbon fixation by the symbionts of another vestimentiferan. Wilmot and Vetter ( 1 990) recently reported that only ex- ogenously supplied sulfide (not thiosulfate or sulfite) is oxidized by the symbionts, and that oxygen consump- tion by both trophosome preparations and partially puri- fied symbionts is not inhibited by atmospheric levels of oxygen, or by sulfide concentrations below 2 mM. All of the vestimentiferans tested (including numerous individ- uals of R. pachyptila) have contained appreciable activi- ties of RuBP carboxylase/oxygenase, ATP sulfurylase, and adenosine-5'-phosphosulfate (APS) reductase in their trophosomes indicating that the symbionts are chemoautotrophic sulfur oxidizers (Felbeck, 1981; Fel- becketa/.. 1 98 1; Brooks et al., 1987; Fisher etal.. 1988b; Gary et al., 1989). While one study suggested that R. pa- chyptila trophosome homogenates could oxidize meth- ane (Fisher and Childress, 1984), another demonstrated that the intact symbiosis did not take up methane and that the hemolymph lacked a binding protein for meth- ane (Childress et al., 1984). In addition, stable carbon isotope studies have led some investigators to suggest that methane may be oxidized by some vestimentiferans (Ku\metal.. 1985). The role of vestimentiferan blood in protecting the en- dosymbionts of an unnamed hydrocarbon-seep escarpid (phylum Vestimentifera) from the toxic effects of hydro- gen sulfide, while supplying the endosymbionts with a large pool of bound hydrogen sulfide, has recently been reported (Fisher el al., 1988a). Although carbon fixation by the symbionts of R. pachyptila is dependent on the availability of oxygen (Belkin et al., 1986), investigators have only speculated about the role of vestimentiferan blood in protecting the symbionts from oxygen inhibi- tion of carbon fixation, while simultaneously providing a large pool of this required substrate, (Childress, 1987; Fisher etal.. 1988a). Here we report the results of several experiments car- ried out with trophosome preparations from Riftia pa- chyptila. A variety of potential electron donors were tested for suitability as an electron donor for the endo- symbionts. Methane oxidation by the endosymbi- onts was investigated using both UC methane and NaH'4CO:. The symbionts' ability to fix molecular N: was tested by the acetylene reduction method under a variety of conditions. Barotolerance of symbiont autot- rophy was studied in incubations of trophosome prepa- rations at high pressure. Finally, sensitivity of symbiont carbon fixation to free oxygen and sulfide, and the role of vestimentiferan blood in providing these substrates at appropriate activities for maximal rates of autotrophic carbon fixation, was investigated. Materials and Methods Experimental material The Riftia pachyptila used in these experiments were collected during three cruises: two to the Galapagos Rift (Galapagos 1985, and Galapagos 1988), and one to 13°N on the East Pacific Rise (Hydronaut in 1987). On all cruises, the animals were collected by submersible (either ALVIN or NAUTILE) and brought to the surface in a temperature insulated container. Upon recovery, the an- imals were placed in fresh, chilled seawater and trans- ferred to a refrigerated van for processing. Most of the animals used in this study were dissected immediately after recovery; some were transferred to flow-through pressurized aquaria (Quetin and Childress, 1980) for short term maintenance (less than three days) before be- ing used. The experiments reported here were conducted on tis- sue from living worms. Trophosome tissue is extremely fragile and deteriorates rapidly when the organ is even slightly damaged. When damaged, the tissue begins to take on a fuzzy pink appearance. In the early stages, this is just visible between the lobes of the trophosome tissue. This "fuzzy pink" effect may be due to the effects of lyso- zymes on the blood, which is normally a deep red color. Tissue from damaged individuals has always shown very low rates of carbon fixation in our studies. About 25% of externally undamaged animals (especially larger individ- uals) dissected immediately upon recovery contained a substantial portion of visibly degraded trophosome tis- sue. The incidence of degradation, and its intensity, in- creased dramatically in animals held at ambient pressure for even a few hours before dissection. No damaged tis- sue was used in this study. Preparation of Riftia saline Riftia saline was prepared based on the average con- centrations of inorganic salts measured in both vascular and coelomic fluids. The saline was titrated to pH 7.5 with NaOH before use. One liter of Riftia saline con- tains: 20.48 g NaCl (0.4 M); 0. 194 g KQ (2.6 mM); 6.22 g MgCl2*6H2O (30.6 mM); 1.65 g CaCl;*2H2O (1 1.2 mM); 4.53 g Na2SO4 (31.9 mM); and 1 1.91.5 g HEPES (50 mM) (Fisher etal., 1988a). Preparation and analysis of Riftia pachyptila blood The Riftia pachyptila blood used in these experiments was collected from living worms and, unless otherwise specified, was kept frozen at either -20 or -70°C until used. This blood was a mixture of coelomic fluid and 374 C. R. FISHER ET AL hemolymph from a number of individuals. Bound sul- fide was removed from the blood by acidifying to pH 5.5 with HC1, and purging the chilled blood (7°C) with a stream of nitrogen for 24 h. The blood was diluted with vestimentiferan saline solution before use in the tropho- some homogenate incubations. In two sets of experiments, fresh blood was used in the incubations. Blood from freshly collected R. pachyptila was neither acidified to remove bound sulfide, nor fro- zen, before use in the experiments in which free oxygen was varied in blood incubations (Fig. 2C and Table II). For these experiments, the chilled blood (pH 7.5) was saturated with oxygen by stirring under a stream of air for 15 min and then stripped of free (unbound) oxygen by stirring under a stream of nitrogen for an additional 45 min before use. Because it is difficult to strip the he- moglobins of oxygen, this treatment has little effect on the amount of oxygen bound by the hemoglobin. The result, therefore, is blood with the hemoglobin virtually saturated with oxygen, but containing very low concen- trations of free oxygen in solution. This blood stock was then loaded into a gas-tight syringe, and added to incuba- tion syringes containing saline solution with various con- centrations of oxygen to generate variable concentra- tions of free oxygen in the experimental syringes. In the experiments shown in Figure 5 (testing symbiont sulnde sensitivity), fresh coelomic fluid from worms maintained in pressure aquaria in the absence of sulfide was used be- cause it contained very low levels of bound sulnde (28 AtA/), and therefore did not require the acidification treatment to remove bound sulfide. The heme content of the blood mixture was deter- mined from the absorbance of a cyanomet hemoglobin derivative (Tentori and Salvati, 1981). We determined the capacity of separate aliquots of the bloods for sulfide by equilibrium dialysis of the blood in 30 mA/ citric acid phosphate buffer at 7°C and pH 7.5 ( Arp and Childress, 1983). Samples of the blood in dialysis tubing were al- lowed to come to equilibrium (24 h) with sulfide in the dialysate ( 1 mA/). The concentrations of sulfide in both blood and dialysate were analyzed with a gas chromato- graph (Childress el a/.. 1984). The difference between the sample sulfide and the dialysate sulfide was the amount of sulfide bound. Preparation and analysis o/Calyptogena magnifica scrum Dilute Calyptogena magnifica serum was used as an incubation medium in several experiments because the clam serum binds sulfide but not oxygen (Arp ct ai. 1984). The clam serum was collected during the "Gala- pagos 1985" expedition and was used in experiments conducted during the "Hydronaut" expedition. Clam blood was collected from freshly recovered clams and centrifuged for 3 min in a bench top centrifuge at about 2000 X g. Dissection of the clams, collection of the blood, and centrifugation was all conducted in a refriger- ated van (6-8°C). The serum was transferred to plastic scintillation vials, frozen at -20°C on board ship, then transferred to -70°C for storage in the laboratory. Before use, the serum was concentrated and cleared of precipi- tates by dialysis against distilled water for 16 h in a vac- uum concentrator. Binding capacity of the serum for sul- fide was determined as above for the Riftia blood. The binding capacity of the concentrated serum was 8.43 mM, and it was diluted with Riftia saline to a binding capacity of 2.0 mM before use as a serum stock in the trophosome incubations. Preparation of the trophosome homogenates Trophosome tissue containing symbiotic bacteria was dissected from living Riftia pachyptila and separated from the major blood vessels and gonads also present in this organ. A portion of the tissue (0.4- 1.0 g) was blotted for a few seconds on a paper towel to remove excess blood, weighed on a motion compensated shipboard bal- ance system (Childress and Mickel, 1 980), and then sub- merged in about 7 ml of chilled, deoxygenated (nitrogen purged) Riftia saline solution (Fisher et ai, 1988a). The tissue was gently homogenized for 5-10 s in a chilled, loose fitting Dounce type ground glass tissue homoge- nizer (2-4 passes), to rupture most of the bacteriocytes and disperse the symbionts. The homogenate was diluted to either 30 or 60 ml with additional deoxygenated saline and loaded into one or two glass 30-ml syringes, equipped with three-way valves, containing marbles to mix the homogenate. This entire procedure takes be- tween 5 and 10 min and, except for weighing, was con- ducted in a refrigerated van (6-8°C). A portion of this homogenate (0.1 ml) was fixed in 0.9 ml of 3% glutaral- dehyde in 0. 1 A/ phosphate-buffered 0.35 M sucrose (pH 7. 35) for later examination using epifluorescence micros- copy (Hobbie ct ai, 1977). For the experiments in which the effect of free oxygen concentration on carbon fixation was examined, the weighed tissue was transferred to a glove bag (in the re- frigerated van), and the homogenate was prepared and loaded into the 30-ml syringe under a nitrogen atmo- sphere. Nail "COj incubations The incubations were conducted in 10-ml glass sy- ringes (except the Galapagos 1985 experiments, which were conducted in disposable 10-ml plastic syringes) fit- ted with low dead volume teflon valves. Experiments were conducted in a refrigerated van that was main- CARBON RXATION BY RIFTIA SYMBIONTS 375 tained between 6 and 8°C (the incubation temperature during an experiment was constant, but the temperature inside the van varied slightly from day to day). Prior to preparation of the homogenate, between 6 and 1 8 sy- ringes were loaded with the incubation media and sub- strate concentrations appropriate for a given experiment. All media and substrate stock solutions were titrated to pH 7.5 before use. The incubation media contained vari- able amounts of Riftia blood or Calyptogena magnified serum diluted with Riftia saline, or else Riftia saline alone. The various sulfide concentrations used in the ex- periments were generated in the experimental syringes by adding appropriate amounts of a sulfide stock solu- tion (7 to 15 mA/) to the experimental syringes, using a three-way valve on the stock syringe. Similarly, variable oxygen concentrations were generated in the experimen- tal syringes by introducing a mixture of saline stock solu- tions of variable oxygen concentrations (also contained in syringes and introduced through three-way valves). Stock solutions of methane and hydrogen were prepared by bubbling a saline solution with the appropriate gas. The stock solutions of thiosulfate and ammonia were prepared from sodium thiosulfate and ammonium chlo- ride, respectively. Sulfide, methane, and inorganic car- bon concentrations in the blood and saline stock solu- tions, and oxygen concentration in the saline stocks, were determined using a gas chromatograph (Childress ct al. . 1 984). (O: concentrations were not directly quanti- fiable in blood by this method.) Ammonium concentra- tion in the stock solution was verified by flow injection analysis (Willason and Johnson, 1986). To confirm that stimulated carbon fixation was through autotrophic pathways. 1 0 mA/ DL-glyceraldehyde (a feedback inhibi- tor of RuBP carboxylase-oxygenase; Stokes and Walker, 1972) was added to one syringe in most of the experi- ments (Fisher el al.. 1988a). To start the incubations, NaH14COj was added to the trophosome preparation, and 2.5 ml of the labeled prep- aration was drawn into each of the syringes, which al- ready contained incubation media with the desired levels of sulfide, oxygen, or other substrates. The final concen- trations of NaH14CO3 used in these experiments ranged from 0.1 to 1.0 j'Ci/ml depending on the experiment. Activity of NaH'4CO, in the syringes was determined by scintillation counting of replicate samples of the tropho- some preparation stabilized with hyamine hydroxide. After all the experimental syringes had been filled and the contents mixed (a process that took between 4 and 7 min). replicate samples (0.1 ml) were removed from each syringe and acidified for scintillation counting of the fixed carbon. Replicate samples (0.1 ml) were similarly removed and assayed from each syringe at 10- to 20-min intervals for the next 60 to 80 min. Carbon fixation rates were calculated, following the methods of Strickland and Parsons (1972), from the measured concentration of in- organic carbon in the incubation media, the measured specific activity of that inorganic carbon pool, and the amount of acid-stable I4C found in the samples at each time point. Fixation rates were calculated after subtracting first sample values from the values measured at the later time points. This method of analysis compensated for the variable rates of carbon fixation before, and during, the first few minutes that the preparations were exposed to the substrates in the experimental syringes (Fisher ct al.. 1988a). Trophosome preparations that failed to show a carbon-fixation rate greater than 0.2 ^mol/g/h under any experimental conditions were considered to contain, at best, only marginally viable symbionts, and no data from these preparations are reported here. Incubations under pressure In the experiments designed to test the effect of pres- sure on the carbon fixation rate by the trophosome prep- arations, identical paired syringes were prepared, and one of each pair was incubated in a pressure vessel at 100 atm. One hundred atm was considered a sufficient pressure because this pressure supports extended sur- vival of the tubeworms that cannot survive at 1 atm (Childress et al.. 1984). For pressure incubations, the glass syringes were suspended in an acrylic pressure ves- sel (Quetin and Childress, 1980), with a fine teflon tube extending from a luer lock fitting on the syringe to the outside through high pressure valves. Thus, samples could be taken from the syringes without releasing the pressure around the incubations. The pressure vessel was inverted repeatedly so that the marbles in the syringes would mix the samples. Rate calculations Determination of the appropriate carbon fixation rate for analysis was often problematic because, under some conditions, the fixation rates were distinctly non-linear over the course of the incubation. These situations were handled as follows: if the carbon fixation showed no sys- tematic increases or decreases during the incubation (did not appear to increase or decrease significantly as a func- tion of time), then rates were calculated by linear regres- sion of all data points. This was the situation in most of the incubations in Riftia blood, and some of the saline experiments. If the rates in all of the syringes during an experiment decreased over time, then rates were calcu- lated from the same portion of the experiment for all sy- ringes (the first two, three, or four time points). In some of the saline and serum incubations, where either sulfide or oxygen was low, the rates decreased as the limiting substrate was exhausted, and the rates were therefore cal- 376 C. R. FISHER ET AL culated from the first two or three points. In a few incu- bations where the substrate (oxygen or sulfide) was ini- tially present in slightly inhibitory concentrations, the rates increased over the course of the incubations as the substrate was depleted. The rates in these experiments were also calculated from the first few time points. Ex- amples of these situations can be seen in Fisher el ul. (1988a). The specific experiments in which rates were calculated from less than all five points are indicated in the results section. Data were always treated consistently within an experiment. 14 C-methane incubations Trophosome samples from two individuals of R. pa- ehyplila were also tested for the ability to use methane as a carbon or energy source. I4CH4, synthesized microbio- logically as described by Daniels and Zeikus ( 1 983), hav- ing a specific activity of 4 X lO"1 dpm/^1, was used as a tracer for methane oxidation by trophosome prepara- tions. Contamination of the labeled methane was less than 0.01% as determined by gas proportion counting. Trophosome homogenates were prepared as described above, and 10 ml of the homogenate was placed in each of six 35-ml serum vials. The headspace was flushed with a stream of nitrogen for 1 min, then the vials were capped with butyl rubber stoppers and crimped with aluminum seals. The headspaces over the vials were adjusted using a gas-tight syringe to remove nitrogen and inject air and methane to produce the desired concentrations of dis- solved oxygen (~100 ^M) and methane, as estimated from Bunsen coefficients. Two methane concentrations were tested in each experiment: 10 and 100 pM in the first experiment and 5 and 10 ^M in the second. UCH4 tracer stock (500 /jl) was added to each vial 30 min after injecting the cold methane to initiate the incubations. Duplicate samples and a formalin-killed control were in- cubated for each methane concentration. The incuba- tions were terminated 1 h after introducing the labeled methane by adding 200 n\ of 5 N NaOH to each. The seals were then removed from each vial and the stoppers replaced by another, with a piece of Whatman #1 filter paper soaked with 100 n\ of phenethylamine suspended from a wire into the headspace. The vials were re-sealed, and 500 n\ of concentrated H;SO4 was added to each vial by injection. The filters were allowed to absorb the CO: released from the liquid for 24 h, and then removed and placed in 10 ml of 3a70 fluor( National Diagnostics) and counted in a liquid scintillation counter. Replicate 100- A/l samples of the homogenate were degassed and assayed for acid stable I4C by liquid scintillation counting. Results Potential symhiont substrates The results of experiments conducted during the Gala- pagos 1 985 expedition should be regarded as preliminary because these techniques were developed during this ex- pedition. Nine sets of experiments were conducted dur- ing this expedition, with preparations of trophosome material from nine individuals of Riftia pachyptila. In each of these experiments, six to ten syringes were run simultaneously. Only hydrogen sulfide (among electron donors tested in these experiments) significantly stimu- lated carbon fixation: by as much as 140% in saline, and by as much as 880%i when incubations were conducted in 50%. Riftia blood. (Control syringes containing blood but no trophosome preparation did not fix carbon.) Con- siderably higher rates of carbon fixation (ten- to twenty- fold) were found when the incubations were conducted in dilute Riftia blood as compared to incubations in sa- line alone. Neither methane (140-500 nM, 10 incuba- tions, 5 worms), hydrogen ( ~ '/? saturated, 5 incubations, 3 worms), ammonia (50 and 100 nAl, 6 incubations, 3 worms), nor thiosulfate (0.05 and 0.5 mM, 6 incuba- tions, 3 worms) had a significant effect on the rate of car- bon fixation in these experiments. All of these compari- sons were conducted under both aerobic (0.1-0.2 mA/ oxygen )and micro-aerobic (0.02-0. 04 m M oxygen) con- ditions. Trophosome tissue from two individuals of R. pachyp- tila tested during the Galapagos 1988 expedition did not oxidize 14C-methane to either '4C-organic compounds or'4CO:. Trophosome tissue from two individual worms was tested for the ability to fix N2 by the acetylene reduction method during the Galapagos 1985 expedition. Tissue from both individuals was tested under both aerobic and micro-aerobic conditions in both saline and dilute Riftia blood, with and without sulfide ( 100 ^M) as an energy source. Results were negative (no appearance of ethyl- ene) in all 12 incubations, which lasted either 9 or 16 hours. Effects of pressure on carbon fixation Seven experiments, conducted during the Hydronaut expedition, were designed to test the effects of pressure on carbon fixation by Riftia pachyptila trophosome preparations (Table I). The carbon fixation rates for the incubations conducted in saline (3 experiments) were calculated from the first three points only, because the rates decreased significantly during the incubations (Fig. la). Similar decreasing rates were observed during many of the other incubations in saline, but the phenomenon was most pronounced in these incubations; a possible cause was the increased time involved in beginning the experiments in pressure vessels, an additional 1 5-20 min after preparing the homogenate. No significant rate de- crease was observed in the blood incubations during the experiments (Fig. IB). Typical results of this study are CARBON FIXATION BY Rlh'TIA SYMBIONTS Table I /•.7/iv/s i>t I'rc'i'iitrciin carhon fixation rale of Rif'tia pachyptila trophosome tissue 377 Exp. Pres. (ATM) [Sulfide] (f,M) Carbon fixation rates [jimol 'h ' (r:)] Ratio of P/ A* Low Med. High Low Med. High Inhib.** Low Med. High S 1 100 35 65 260 2.40(1.0) 2.49(.72) 2.80(.83) 1.26 1 .00 1.35 1 35 65 260 1.91 (.63) 2.49(.62) 2.07(.18) S2 100 25 50 200 1.63 (.25) 2.191.27) 1.33 (.16) 1.55 1.60 1.06 1 25 50 200 1.051.28) 1.37(.53) 1.261.31) 0.49 (.22) S3 100 30 60 230 3.82 (.25) 3.38M5) 2. 32 (.08) 1.12 1.02 1.63 1 30 60 230 3.41 (1.05) 3.32 (.60) 1.42 (.68) 1.271.20) B 1 100 55 230 570 5.8 (.32) 8.7 (1.33) 7.7 (.77) 0.69 0.94 0.91 1 55 230 570 8.4 (.26) 9.3 (1.12) 8.5 (.71) 2.7 (.17) B2 100 85 335 840 9.3 (.56) 10.3 (.67) 5.0 (.56) 0.89 1.00 0.98 1 85 335 840 10.4 (.66) 10.3 (1.65) 5.1 (.91) 5.0 (.35) B3 100 1 10 395 1060 7.221.57) 6.11 (.87) 2.46 (.32) 0.99 1.04 1 .08 1 1 10 395 1060 7.301.72) 5.90(.74) 2.27(.19) 3.831.18) B4 100 1 10 395 1060 8.5 (.11) 9.0 (.40) 2.5 (.12) 0.90 0.98 1.19 1 1 10 395 1060 9.4 (.26) 9.2 (.39) 2.1 (.12) 3.0 (.08) Standard errors of the slopes of the linear regressions are presented in parentheses next to the rales. Linear regressions (and rates) were calculated from the first three time points only for the saline incubations. S — experiments conducted in Rif'tia saline. B — experiments conducted in dilute Rifiia blood (binding capacity about 330 pM). Low, Med.. and High are related to each other within an experiment and refer to the sulfide concentrations given in the table for each experiment. * P/A is the ratio of the rate at 100 ATM to the rate at 1 ATM in that experiment. '* The inhibitor ( 1 0 m A/ D.L-glyceraldehyde) was added to a syringe containing the "Med." level of sulfide in the saline experiments and "Low" level of sulfide in the blood experiments. depicted in Figure 1. In the saline incubations, the car- bon fixation rates under pressure were the same or slightly higher (averaging 30% higher) at all three sulfide concentrations (Table I), but there were no significant differences in the slopes of the linear regressions (rates) in pairwise comparisons within experiments. When the incubations were conducted in blood, carbon fixation rates were slightly lower under pressure. The only sig- nificant differences in the slopes of the paired pressure and ambient incubations occurred in run Bl at 55 juA/ sulfide (Table I). Effects of oxygen concentration on carbon fixation rates Ten experiments conducted during the Hydronaut ex- pedition were designed to determine the effect of free ox- ygen on the carbon fixation rates of trophosome prepara- tions. The results of three experiments conducted in sa- line and four experiments conducted in 8% Calyptogena magnified serum were substantially the same (Table II). Maximal fixation rates in saline and dilute clam serum were recorded at initial oxygen concentrations between 72 and 1 54 nM, with higher oxygen concentrations sig- nificantly inhibiting carbon fixation. These rates were calculated from the first two or three points when the carbon fixation pattern was distinctly non-linear, as it was at all limiting oxygen concentrations (see Fig. 2A). The difference between limitation and inhibition was ev- ident from the shape of the curves at the different oxygen concentrations (compare the fixation patterns at 53 and 160 nM O2 in Fig. 2A). The rates presented in Table II for the saline and clam serum experiments are higher than the actual measured inorganic carbon incorpora- tion rates for the 60 to 70 min incubations because they were calculated from the initial, linear portion of the curves (see Fig. 2 A). When incubations were conducted in dilute Rifiia blood (with O: saturated hemoglobin), none of the incu- bations appeared oxygen limited (the carbon fixation rates were linear at all oxygen concentrations throughout the experiments. Fig. 2B), and oxygen inhibition of car- bon fixation was evident at much lower free (unbound) oxygen concentrations than in the saline and clam serum experiments (data shown in Fig. 2B are typical of the blood incubations). This is most easily visualized in Fig- ure 3, where all of the experiments are summarized. Each point in Figure 3 represents a rate calculated from one incubation, and the rates are presented as a percent of the maximal rate observed with that preparation. The maximum rates calculated from each experiment are given in Table II. 378 C. R. FISHER ET AL 2.5 20 C o ra O 10 20 30 40 50 60 Time (min) Time (min) Figure 1. Effects of 100 atm pressure on carbon fixation by Rift in pachyptila trophosome preparations at three sulfide concentrations. Open symbols and dotted lines represent data from incubations at am- bient pressure and closed symbols represent data from incubations at 100 atm: (A) incubations in Riftia saline without blood (Saline 1, Table I): squares, 35 nM initial sulfide; circles. 65 ^Al initial sulfide; triangles, 260 pM initial sulfide; +, 1 atm, 65 phf initial sulfide and 10 mA/ D,L- glyceraldehyde. (B) Incubation in dilute Riftia blood with binding ca- pacity of about 330 nA/(Blood 4, Table I): squares, 1 10 nM initial sul- fide; circles, 395 nM initial sulfide; triangles, 1060 nM initial sulfide; +, 1 atm, 1 10 tiM initial sulfide and 10 mA/D,L-glyceraldehyde. Effects of sulfide concentration on carbon fixation rates Experiments were conducted during the 1 988 Galapa- gos expedition to determine the optimum concentra- tions of sulfide for chemoautotrophic carbon fixation by Riftia pachyptila trophosome preparations. The results of the four experiments conducted in saline, in which the maximal carbon fixation rates were greater than 1 ^mole g~ ' h~ ' , are shown in Figure 4. Remember that the actual concentrations during the incubation are lower than the initial concentrations presented in the figure legends due to spontaneous oxidation of sulfide in the saline incuba- tions. The inhibitory levels of sulfide in the saline incuba- tions are, therefore, maximum values. The initial sulfide concentrations that were maximally stimulatory to car- bon fixation by the trophosome preparations incubated in saline ranged from 250 to 350 nM in the four experi- ments (Fig. 4). The trophosome preparations were sul- fide limited at initial concentrations below 175 to 250 nM, and sulfide inhibition became apparent at concen- trations of 350 to 500 nM in these experiments (Fig. 4). Ten experiments were conducted in various concen- trations of Riftia blood that had been collected during the Hydronaut expedition, stored at -20°C for several months, and treated as described in the methods section to remove bound sulfide. We consider the results of these experiments preliminary because subsequent analysis of the blood used indicated that it did not have the affinity. O .0 X c o A ^ re O O E Figure 5. Rate of carbon fixation by Riftia pacliypliln trophosome preparations as a function of initial total sulfide concentration. A and B represent two separate experiments with trophosome preparations from different worms. Incubations under all three conditions in each experiment were conducted simultaneously. Squares are rates from in- cubations in saline without added blood: triangles are rates from incu- bations in dilute ( 1 3%) fresh coelomic fluid with a binding capacity of 1 20 fi.M sulfide; circles are rates from incubations in more concentrated (53% in A and 40% in B) fresh coelomic fluid with a binding capacity of 360 and 480 pM sulfide, respectively. The effects of pressure on the blood incubations are also slight. It significantly inhibited carbon fixation at lowest sulfide concentration tested (55 fiM), with only slight effects at sulfide concentrations above that, and no effect at concentrations that were clearly inhibitory to carbon fixation. These small effects could be due to either a direct pressure effect on blood sulfide binding or an in- direct effect stemming from the altered equilibrium be- tween HS and H:S. Effects of oxygen concentration on carbon fixation The possibility that vestimentiferan endosymbionts might be sensitive to free (unbound) oxygen is suggested by the observation that many free-living sulfur bacteria are microaerophilic (Krieg and Hoffman, 1986). Al- though Wilmotand Vetter(1990) found no oxygen inhi- bition of oxygen consumption by trophosome prepara- tions, their study does not address the possibility that ox- ygen may inhibit autotrophic carbon fixation by the symbionts, as oxygen consumption (or sulfide oxidation ) and carbon fixation are not tightly coupled in sulfur bac- teria (Kelly, 1989). Experiments designed to explore the sensitivity of vestimentiferan symbiont carbon fixation to oxygen were conducted in Riftia saline. In these exper- iments, the maximum rate of carbon fixation was re- corded at oxygen concentrations of approximately 100 nM (Table II, Fig. 3A), but the shape of the curves (car- bon fixation vs. time) suggested increasing substrate limi- tation over time at the lower oxygen concentrations (Fig. 2A). Due to the reactivity of sulfide and oxygen, either of these substrates could have become limiting during the experiments under those conditions. In a first at- tempt to stabilize the sulfide without affecting the free oxygen in the experiments, incubations were conducted in dilute Calyptogena magnified serum. This serum was used because it binds sulfide but not oxygen (Arp et ai, 1984). The results of the experiments in dilute C mag- nified serum were essentially the same as those in saline alone (Table II, Fig. 3B). This suggests that oxygen — not sulfide — is the limiting substrate. With variable substrate limitation removed from the experimental design by the presence of Riftia hemoglo- bin, the effects of free oxygen on the carbon fixation rate of the trophosome preparations were easily visible (Fig. 3C). At free oxygen levels around air saturation (220 nM), carbon fixation by the trophosome preparations was inhibited by 60 to 70%, while at the maximum ambi- ent oxygen levels the animals are exposed to /// situ (100 fiM), inhibition ranged from about 10 to 40% (Fig. 3C). Oxygen inhibition of carbon fixation may be caused by the well-documented oxygenase function of the pri- mary carboxylating enzyme in the symbionts, ribulose 1,5-bisphosphate carboxylase/oxygenase (RuBPC/O, EC 4.1.139) (Tabita, 1988). However, the experimental conditions of high total inorganic carbon concentration used in these experiments (Table II) would minimize that effect, especially at the lower oxygen concentrations. The predicted inhibition of carbon fixation due to the oxygenase function of RuBPC/O can be calculated using a substrate specificity factor, which compares the relative rates of the carboxylase and oxygenase reactions at any given concentrations of CO: and O2 (Tabita, 1988). The specificity factor has not been experimentally deter- mined for RuBPC/O of Riftia pachyptila symbionts, but a factor of 50 is in the range of the lowest values reported for prokaryotic RuBPC/O with both large and small sub- units (Tabita, 1988); R pachyptila symbiont RuBPC/O contains both large and small subunits, (Stein el ai, 1989). Using this factor to predict the percent of oxygen 382 C. R. FISHER ET AL inhibition of carbon fixation due to the oxygenase func- tion of the enzyme (RuBPC/O) would, therefore, result in a maximum estimate. Based on the free CO2 concen- trations [calculated from the 2CO2 concentrations mea- sured in the experimental syringes (Table II), our experi- mentally determined relationship between Pco, and 2CO2 in the saline, and the solubility of CO2 in the sa- line], we predict no more than 8.4% inhibition of carbon fixation at 220 ^Af oxygen at the lowest CO2 concentra- tion employed (in the serum incubations. Table II). A range of 2. 5 to 5.7% inhibition by 220^A/O2 in the other incubations is due to RuBPO. The same calculations predict between 1.2 and 4.2% inhibition of carbon fixa- tion by 1 10 nM O2 due to the oxygenase function of RuBPC/O in these experiments. The most probable explanation for the degree of oxy- gen inhibition of carbon fixation found in this study is that the symbionts, like many free-living microaero- philes, are sensitive to toxic forms of oxygen, such as H2O2, O2 or OH- (Krieg and Hoffman, 1986), even though activity of some detoxifying enzymes has been demonstrated in extracts of R. pachyptila trophosome (Blum and Fridovich, 1984). Catalase is not present in R. pachyptila trophosome, but moderate levels of peroxi- dase have been demonstrated in trophosome extracts and might defend against H2O: (Blum and Fridovich, 1984). Activity of superoxide dismutase against superox- ide ions has also been demonstrated in R. pachyptila tro- phosome, but the activity was substantially lower than that in muscle tissue. Furthermore, the activity was non- linear after less than 1 min in cell-free preparations (Blum and Fridovich, 1984), and, therefore, may be in- active in our longer term trophosome experiments. Sev- eral microaerophiles contain activities of these protective enzymes and yet are sensitive to toxic forms of oxygen (reviewed by Kreig and Hoffman, 1986). One possible explanation for this is that intracellular protective en- zymes of the microaerophile Campylobacter fetus are ap- parently ineffective against exogenous O2 and H2O2, which could be adversely affecting the microbial cell sur- face (Hoffman ct ai. 1979). The subcellular location of the protective enzymes in Riftia symbionts is unknown. The possibility that symbionts are sensitive to toxic forms of oxygen is further substantiated by three other observations. (1) Exposure to moderate levels of oxygen during preparation of a trophosome homogenate sub- stantially lowers the activity of the preparation. (2) High- est rates of carbon fixation in saline incubations occurs when homogenates were prepared under virtually anaer- obic conditions (Table II). (3) When the same prepara- tion is incubated in dilute Riftia blood and in saline, the incubations in blood fix carbon at consistently higher rates. Unequivocal demonstration of symbiont sensitiv- ity to toxic forms of oxygen awaits studies that demon- strate relief of oxygen inhibition by substances that quench toxic forms of oxygen (see review by Kreig and Hoffman, 1986). Whatever the cause of the observed reduction in car- bon fixation rate as a function of free oxygen concentra- tion, the role of blood in mitigating this effect, while pro- viding an abundant pool of oxygen, is evident (Table II. Figs. 2, 3). That the highest carbon fixation rates were observed when free oxygen in solution approached zero (the P50 of Riftia hemoglobin is around 2 nM at 7.5°C; Arp and Childress, 1981) indicates that the symbionts have an even higher affinity for oxygen and can remove it from the blood. The maximum levels of free oxygen possible in the blood of living worms in situ would be equal to the maximum oxygen levels in the surrounding water — 1 10 /uA/in the ambient (non-vent) bottom water (Johnson et al., 1988a). According to detailed surveys of oxygen and sulfide conditions at the central clump of R. pachyptila at the Rose Garden vent on the Galapagos Rift, the worms are exposed to levels of oxygen that vary from 1 10 /iA/to undetectable over very short or moder- ate time scales (Johnson et al., 1988b). Thus, levels of oxygen in blood fresh from the plume would be expected to vary considerably. The anatomy of Riftia and the properties of its blood buffer its symbionts from the ex- tremes in ambient oxygen concentration (which range from limiting to inhibitory), and allow maximal rates of carbon fixation by providing the symbionts with an abundant pool of oxygen, while maintaining low levels of free (unbound) oxygen in the trophosome. Effects of sul fide concentration on carbon fixation The experiments designed to address this question can be divided into three groups: experiments in saline; ex- periments in mixed blood with reduced sulfide binding capacity, and comparative experiments in saline and fresh coelomic fluid. The maximum rates of carbon fixation in experiments in saline occurred at initial sulfide concentrations be- tween 250 and 350 nAf (Fig. 4). The differences in the sulfide concentrations yielding maximum carbon fixa- tion rates could be due to real differences in sulfide sensi- tivity of the symbionts from different individuals. How- ever, the inherent instability (and unpredictability) of unbound sulfide and oxygen in solution cast some doubt as to the actual levels of sulfide present during these incu- bations; the apparent differences could be due to differ- ing rates of auto-oxidation of sulfide in the different incu- bations and resultant variations in actual sulfide concen- trations during the incubations. Two general conclusions stand out from the data col- lected in the incubations with blood of reduced binding capacity. First, even this blood protects the symbionts CARBON FIXATION BY R1FT1A SYMBIONTS 383 from the inhibitory effects of sulfide demonstrated in the saline incubations (Fig. 4). Second, carbon fixation by R. pachyptila symbionts is not inhibited by levels of sulnde below the binding capacity of the blood. In the experiments with fresh coelomic fluid, as found in the other experiments reported here and in previous work (Fisher et a/.. 1988a), the blood protected the sym- bionts from the inhibitory effects of sulfide at sulnde con- centrations below the binding capacity of the blood. The maximal rates of carbon fixation in the blood incuba- tions were also higher than the maximum rates in the corresponding saline incubations. However, unlike the symbionts of a hydrocarbon seep escarpid vestimentif- eran which were inhibited by total sulfide concentrations greater than 70% of capacity (Fisher et ai, 1988a), Riftia pachyptila symbionts were not inhibited by sulfide until the concentration surpassed the binding capacity of the dilute blood (Fig. 5). The exact level of free sulfide that was inhibitory to R. pachyptila symbionts cannot be de- termined due to the paucity of data points around the carbon fixation maxima (Fig. 5). However, it is clearly above the 100 nM that is inhibitory to the seep escarpid (Fisher el ai, 1988a), and more likely closer to 350 juA/ as suggested by the saline incubations (Fig. 4). Another recent study (conducted during the same cruise as these experiments) demonstrated that oxygen consumption by trophosome preparations was not inhibited by sulfide concentrations up to 2 mM (Wilmot and Vetter, 1990). The results of that study and this one are not contradic- tory; most likely they reflect the relatively "loose" cou- pling between sulfide oxidation and carbon fixation, and indicate that trophosome preparations can oxidize sul- fide without concomitant carbon fixation. An earlier study of sulfide stimulation of carbon fixation by Riftia symbionts showed an optimum of 600 ^M, however, these workers' trophosome preparations had very low levels of activity and apparently no effort was made to exclude or quantify blood contamination in the prepara- tions (Belkin et at., 1986). Thus, while their results are in general agreement with ours, a precise comparison can- not be made. The lesser sensitivity to free sulfide of autotrophic car- bon fixation by the symbionts of R. pachyptila, as com- pared to the seep escarpid, is probably a reflection of the ambient sulfide levels the animals are exposed to in their respective environments. The highest sulfide levels mea- sured around the plumes of the seep escarpid are below 3 nAl, and even if these vestimentiferans take up sulfide across their body wall, as has been suggested (MacDon- ald et a/., 1989), the symbionts are likely never exposed to sulfide levels in the blood approaching saturation. In fact, the highest level reported in the blood of freshly col- lected seep vestimentiferans is 1 14 juM (Childress et al., 1986) which, assuming the blood has similar capacities to R. pachyptila blood, corresponds to a free sulfide con- centration below 1 p.M (Fisher et al.. 1988a). This as- sumption is validated by preliminary experiments with blood from the seep escarpid, which indicate that the blood binds sulfide and is of similar concentration to that of R. pachyptila (A. J. Arp, pers. comm.). R. pachyptila is exposed to sulfide levels in the 300 nM range in situ (Johnson et al., 1988a), and total sulfide levels in the blood of freshly collected R. pachyptila as high as 9 mM have been reported (Childress et al., 1984). Therefore, free sulfide levels in the blood of R. pachyptila are apt to be much higher than those found in the seep escarpid, and R. pachyptila symbionts are apparently adapted to these higher concentrations. The symbionts of both species are located in vacuoles within host cells (bacteriocytes) in the trophosome, and not directly exposed to the blood in situ. The conditions inside the vacuole and host bacteriocyte will certainly affect both the total amount of sulfide and the ionic spe- cies to which the symbionts are exposed. However, the high degree of vascularization of the trophosome (Jones, 1988) implies that the blood exerts considerable influ- ence on the concentrations of sulfide and oxygen in the bacteriocytes, and that the maximum concentrations to which the symbionts are exposed are almost certainly a reflection of the highest free concentrations of these sub- stances in the blood. Despite the differences between the species, the role of the blood with respect to sulfide in the intact R. pachyp- tila symbiosis is basically the same as for the seep escar- pid (Fisher et al., 1988a). That role is to provide the sym- bionts with an abundant supply of sulfide, while main- taining free sulfide at levels that allow maximal rates of carbon fixation. Conclusion R. pachyptila individuals appear to grow rapidly (Fus- tec et al., 1987; Hessler et al., 1988; Roux et al.. 1989). Because these animals are apparently dependent upon their symbionts for at least their bulk nutritional carbon requirements (see review by Fisher, 1989), the tropho- some must be a very productive chemoautotrophic or- gan. Shipboard studies with live animals under pressure suggest that the oxygen consumption rate of R. pachyp- tila is in the range of 0.44 to 1. 52 ^mole oxygen g"1 h'1 in the absence of sulfide (Childress et al., 1 984). Assuming a RQ of 1 , these data suggest that the intact symbiosis re- quires an input of organic carbon at about the same rate. Since the trophosome accounts for 15.3 ± 4.9% of the worms wet weight (Childress el al.. 1984), the tropho- some must incorporate inorganic carbon into organic compounds at at least 10 /umole carbon g trophosome"1 h ' to meet the metabolic needs of the intact symbiosis. 384 C. R. FISHER ET AL The higher rates reported in this study (20 to 28 C g~' h~') suggest that the symbionts can meet the bulk nutritional carbon requirements of the intact symbiosis through chemoautotrophy, even assuming an efficiency of 50% or less in the transfer of nutritional carbon from symbiont to host. This calculation also supports the con- tention that the activities of the preparations used in this study are reasonable, and that preparations with signifi- cantly lower activity are suboptimal. The stable carbon isotope composition of these ani- mals has been interpreted as reflecting carbon limited symbionts (Rau, 1981, 1985; Fisher el ai, 1988c). This interpretation implies a high rate of consumption of in- organic carbon by the symbionts and also suggests a tro- phosome with high metabolic activity (both of which are supported by this study). The high sulfide and oxygen binding capacities of the hemoglobins of R. pachyptila are therefore necessitated by both the relatively large quantity of trophosome and its high autotrophic capac- ity. Additionally, the affinity of the blood for sulnde and oxygen allows the symbionts access to these large pools of bound substrates without exposing the symbionts to high free concentrations of either substance, thereby sup- porting maximal rates of carbon fixation by the symbi- onts. Vestimentiferans rely on a finely tuned symbiosis for their survival. Both their anatomy and the properties of vestimentiferan hemoglobins are adapted for symbiosis with a specific type of sulfide-oxidizing symbiont. The host tube-worms reap the benefits of an autotrophic life style, while providing their symbionts with an environ- ment which free-living sulfide-oxidizing bacteria can only regard with envy. Acknowledgments This work was supported by NSF grants OCE83- 1 1257 and OCE86-09202 to JJC and OCE86-10514 to JJC and CRF. We would like to thank M. Wells and J. Favuzzi for technical assistance with apparatus and R. Van Buskirk, V. Vanderveer, R. Kochevar, and D. Gage for technical assistance during the cruises. Thanks are also due to the captains and crews of the RV Melville, RV Thomas Thompson, and N/O Nadir, Drs. A. Alayse- Danet and H. Felbeck the expedition leader and chief scientist of the RV Thomas Thompson respectively dur- ing project Hydronaut, as well as the sub crews and pilots of the submersibles Alvin and Nautile, without whom this work would not have been possible. This manuscript has benefited from discussions with and comments by R. Kochevar, J. Favuzzi, G. Somero, and R. Trench. Literature Cited Anthony, C. 1982. The Biochemistry <>/ Methylolrophs Academic- Press, London. 431 pp. Arp, A. J., and Childress, J. J. 1981. Blood function in the hydro- thermal vent vestimentiteran tube worm. Science 213: 342-344. Arp, A. J., and J. J. Childress. 1983. Sullide binding by the blood of the deep-sea hydrothermal vent tube worm Rijiia pachyptila. Sci- ence 219: 559-561. Arp, A. J., J. J. Childress, and C. R. Fisher Jr. 1984. Metabolic and blood gas transport characteristics of the hydrothermal vent bivalve Calyplogena magnijtca. Phvsiol Zoo/ 57: 648-662. Arp, A. J., J. J. Childress, and R. D. Vetter. 1987. The sulphide- binding protein in the blood of the vestimentiferan tube-worm Ril- lia pachyptila J Lxp Biol 128: 139-159. Belkin, S., D. C. Nelson, and H. VV. Jannasch. 1986. Symbiotic as- similation of CO2 in two hydrothermal vent animals, the mussel Bathymodiolus tliermoplrilus and the tube worm Rijiia pachyptila. BioLBull, 170: 110-121. Blum, J., and I. Kridovich. 1984. Enzymatic defenses against oxygen toxicity in the hydrothermal vent animals Riflia pachyptila and Cu- lyptogena muKnifica. Arch. Biochcm. Biuphys. 228: 6 1 7-62U. Brooks, J. M., M. C. Kennicutt, C. R. Fisher, S. A. Macko, K. Cole, J. J. Childress, R. R. Bidigare, and R. D. Vetter. 1987. Deep-sea hydrocarbon seep communities: evidence for energy and nutri- tional carbon sources. Science 238: 1 1 38- 1 1 42. Cary, S. C., H. Felbeck, and N. D. Holland. 1989. Observations on the reproductive biology of the hydrothermal vent tube worm, Ril- tia pachyptila Mar Eco/. Prog. Ser. 52: 89-94. Cavanaugh, C. M., S. L. Gardiner, M. L. S. Jones, H. \V. Jannasch, and J. B. \\aterbury. 1981. Prokaryotic cells in the hydrothermal vent tube worm, Rittia pachyptila: possible chemoautotrophic sym- bionts. Science 213: 340-342. Childress, J. J. 1987. Uptake and transport of sulfide in marine in- vertebrates. Pp. 23 1-239 in Comparative Physiology: Life in II 'ater and on Land, P. Dejours, L. Bolis, C. R. Taylor, and E. R. Weibel, eds. IX-Liviana Press. Padova. Childress, J. J.. and T. J. Mickel. 1980. A motion compensated ship- board precision balance system. Deep-Sen Res. 27a: 965-970. Childress, J. J., A. J. Arp, and C. R. Fisher Jr. 1984. Metabolic and blood characteristics of the hydrothermal vent tube-worm Rijtiapa- chypnla. Mar. Biol. 83: 109-124. Childress, J. J., C. R. Fisher, J. M. Brooks, M. C. II. Kennicutt, R. Bidigare, and A. F. Anderson. 1986. A methanotrophic marine molluscan (Bivalvia, Mytilidae) symbiosis: mussels fueled by gas. Science 233: 1306-1308. Daniels, L., and J. G. Zeikus. 1983. Convenient biological prepara- tion of pure high specific activity '4C-labelled methane. ./ Labelled Comp Radiophar 20: 17-24. Distel, D. L., D. J. Lane, G. J. Olsen. S. J. Giovannoni, B. Pace, N. R. Pace, D. A. Stahl, and H. Felbeck. 1988. Sulfur-oxidizing bacte- rial endosymbionts: analysis of phylogeny and specificity by 16S rRNA sequences. J Bacterial. 170: 2506-25 10. Felbeck, II. 1981. Chemoautotrophic potential of the hydrothermal vent tube worm, Ri/iiu pachyptila Jones (Vestimentifera). Science 213:336-338. Felbeck, H., J. J. Childress, and G. N. Somero. 1981. Calvin-Benson cycle and sulphide oxidation enzymes in animals from sulphide- nch habitats. Nature 193: 291-293. Fisher, C. R. 1990. Chemoautotrophic and methanotrophic symbio- ses in marine invertebrates. Rev Aquat. Sci. Fisher, C'. R., Jr., and Childress, J. J. 1984. Substrate oxidation by trophosome tissue from Riftia pachyptila Jones (Phylum Pogono- phora). Mar. Biol. Lett. 5: 171-183. Fisher, C. R., J. J. Childress, and N. K. Sanders. 1988a. The role of vestimentiferan hemoglobin in providing an environment suitable for chemoautotrophic sulfide-oxidizing endosymbionts. Symhiosis 5:229-246. CARBON FIXATION BY RIFT1A SYMBIONTS 385 Fisher, C. R., J. J. Childress, A. J. Arp, J. M. Brooks, D. Dislel, J. A. Favuzzi, S. A. Macko, A. Newton, M. Powell, G. N. Somero, and T. Soto. 1988b. Physiology, morphology, and biochemical com- position ofRifliapachyptilaat Rose Garden in 1985. Deep-Sen Rc.s 35: 1745-1758. Fisher, C. R., J. J. Childress, and J. M. Brooks. 1988c. Are hydro- thermal-vent Vestimentifera carbon limited?. tin. /vol. 28: 128a. Fustec, A., D. Desbruyeres, and S. K. Juniper. 1987. Deep-sea hydro- thermal vent communities at 1 3°N on the East Pacific Rise: micro- distribution and temporal variations. Biol. Ocean. 4: 121-164. Messier, R. R., VV. M. Smithey, M. A. Boudrais, C. H. Keller, R. A. l.ulz, and J. J. Childress. 1988. Temporal change in megafauna at the Rose Garden hydrothermal vent. Deep-Sea Rex 35: 1681- 1710. Hobbie, J. E., Daley, R. J., Jasper, S. 1977. Use of nucleopore niters for counting bacteria by fluorescence microscopy. .I/'/'/ Environ. Microhiol 33: 1225-1228. Hoffman, P. S., H. A. George, N. R. Krieg, and R. M. Smibert. 1979. Studies of the microaerophilic nature of Campylobacter Ic- tus subsp. leiiini- II. Role of exogenous superoxide anions and hy- drogen peroxide. Can. J. Microbiol. 25: 8-16. Issacs, N. S. 1981 . Liquid Phase High Pressure (. 'henusiry John Wi- ley and Sons. 414 pp. .1.1 1 HUM 1 1. II. VV. 1983. Microbial processes at deep sea hydrothermal vents. Pp. 677-709 in Hydrothermal Processes at Sea/loor Spread- infi Centers. P. A. Rona ct a/., eds. Plenum Press New York. Jannasch, H. \V. 1989. Chemosynthetically sustained ecosystems in the deep sea. Pp. 147-166 in Anlolrophic Bacteria, H. G. Schlegel and B. Bowien, eds. Science Tech. Publ., Madison and Springer- Verlag, Berlin. Jannasch, II. VV., and D. C. Nelson. 1984. Recent progress in the mi- crobiology of hydrothermal vents. Pp. 1 70- 1 76 in C 'urrent Perspec- tives in Micnihial Ecology. M. J. Klug and C. A. Reddy, eds. Am. Soc. Microbiol., Washington. Johnson, K.S., J. J. Childress, R. R. Hessler, C. M. Sakamoto- Arnold, and C'. I,. Beehler. 1988a. Chemical and biological interactions in the Rose Garden hydrothermal vent field. Deep-Sen Res 35: 1723- 1744. Johnson, K. S., J. J. Childress, and C. L. Beehler. 1988b. Short term temperature variability in the Rose Garden hydrothermal vent field. Deep-Sea Res. 35: 1 7 1 1 - 1 722. Jones, M. L. 1981. Riltia pachypnla Jones: observations on the vesti- mentiferan worms from the Galapagos Rift. Science 213: 333-336. Jones, M. L. 1988. The vestimentifera, their biology and systematic and evolutionary patterns. Oceanol Ada SP 8: 69-82. Jones. M. I.., and S. L. Gardiner. 1988. Evidence for a transient di- gestive tract in vestimentifera. Proc. Biol. Soc ll'asli. 101: 423-433. Kelly, D. P. 1989. Physiology and biochemistry of unicellular sulfur bacteria. Pp. 193-218 in Autotrvphic Bacteria. H. G. Schlegel and B. Bowen, eds. Science Tech. Pub. Madison, Wl. Krieg, N. R., and P. S. Hoffman. 1986. Microaerophily and oxygen toxicity. Ann Rev Microhiol. 40: 107-130. Kulm, L. D., E. Suess, J. C. Moore, B. Carson, B. T. Lewis, S. D. Ritger, D. C. Kadko, T. M. Thornburgh, R. W. Embley, W. D. Rugh, G. J. Massoth, M. G. Langseth, G. R. Cochranc, and R. L. Scammen. 1985. Oregon subduction zone: venting, fauna, and carbonates. Science 231: 561-566. Macdonald, I. R., G. S. Boland, J. S. Baker, J. M. Brooks, M. C. Ken- nicutt II, and R. R. Bidigare. 1989. Gulf of Mexico hydrocarbon seep communities. II. Spatial distribution of seep organisms and hydrocarbons at Bush Hill. Mar. Biol. 101: 235-247. Mickel, T. J., and J. J. Childress. I982a. Effects of pressure and pres- sure acclimation on activity and oxygen consumption in the bathy- pelagic mysid (inalhophausia ingens Deep-Sea Res. 29: 1293- 1301. Mickel, T. J., and J. J. C'hildress. 1982b. Effects of temperature, pres- sure and oxygen concentration on the oxygen consumption rate of the hydrothermal vent crab Bythoxrueu thermyilrun (Brachyura). Physio! '/Mol. 55: 199-207. Powell, M. A., and G. N. Somero. 1983. Blood components prevent blood poisoning of respiration of the hydrothermal vent tube worm Riftiapaeliyplila. Science 219: 297-299. Powell, M. A., and G. N. Somero. 1986. Adaptations to sulfide by hydrothermal vent animals: sites and mechanisms of detoxification and metabolism. Biol Hull 171: 274-290. Quetin, L. B., and Childress, J. J. 1980. Observations on the swim- ming activity of two bathypelagic mysid species maintained at high hydrostatic pressures. Deep-Sea Res 27A: 383-391. Rau, G. H. 1981. Hydrothermal vent clam and tube worm 13C/12C: further evidence of nonphotosynthetic food sources. Science 213: 338-340. Rau, G. H. 1985. I3C/12C and 15N/14N in hydrothermal vent or- ganisms: ecological and biogeochemical implications. Bull. Biol. Soc. ll'as/i. 6: 243-248. Roux, M., M. Rio, E. Schein, R. A. Lutz, L. W. Fritz, and L. M. Ra- gone. 1989. Mesures /// suit de la croissance des bivalves et des vestimentiferes et de la corrosion des coquilles au site hydrotherm- nal de 13°N (dorsale du Pacifique oriental). C R Acad. Sci. Paris. Ser.II13Q8: 121-127. Smith, R. P., and R. E. Gosselin. 1979. Hydrogen sulfide poisoning. J. Occup.Med. 21:93-97. Southward, E. C. 1988. Development of the gut and segmentation of newly settled stages of Rulgeia (Vestimentifera): implications for relationship between Vestimentifera and Pogonophora. / Mar. Biol. Assoc. U. K 68: 465-487. Stein, J., M. Haygood, and II. Felbeck. 1989. Diversity of RuBP car- boxylase genes in sulfur oxidizing symbioses. Proc. 4lh Endocy- tobio. Coll.. Lyon, France. July. 1989 (in press). Stokes, C. M., and D. A. Walker. 1972. Photosynthesis by isolated chloroplasts: inhibition by DL-glyceraldehyde of carbon dioxide as- similation. Biochem J 128: 1 147-1 157. Strickland, J. D. H., and T. R. Parsons. 1972. A practical handbook of seawater analysis, 2nd ed. Bull. Fish. Res. Board Can 167: 1- 310. Tabita, F. R. 1988. Molecular and cellular regulation of autotrophic carbon dioxide fixation in microorganisms. Microhiol Rev. 52: 155-189. Tentori, L., and A. M. Salvati. 1981. Hemoglobmometry in human blood. In Hemoglobins. E. Antonni, L. Rossi-Bernardi. and E. Chi- ancone, eds. Melh. En-ymol. 76: 7(17-7 14. VVillason, S. VV., and K. S. Johnson. 1986. A rapid, highly sensitive technique for the determination of ammonia in seawater. Mar. Biol. 91:285-290. Wilmot, D. B. Jr., and R. D. Vetter. 1990. The bacterial symbiont from the hydrothermal vents tubeworm Rifiia pachyptila is a sulfide specialist. Mar. Bio/ Reference: Bi,>l Bull. 177: 386-39?. (December. 1989) Gills as Possible Accessory Circulatory Pumps in Limulus polyphemus M. A. FREADMAN1 AND W. H. WATSON III2* 1 Marine Biological Laboratory, H 'oods Hole, Massachusetts 02543 ami -Department of Zoology. University of New Hampshire, Durham. New Hampshire 03824 Abstract. Heart electrical activity (ECGs), gill closer muscle potentials (EMGs), and blood pressures in the heart and the branchiocardiac canals, were measured in adult horseshoe crabs (Limulus polyphemus} during var- ious activities. During ventilation, hyperventilation, and swimming, large transient increases in pressures (10-35 cm H:O) occur in the branchiocardiac canals, which carry blood from the gills to the heart. These pulses of positive pressure are related to, and apparently caused by, gill plate closing. During quiescent periods, with no ventilatory activity, there are no pressure pulses in the canals, but the pressure is still greater than zero. We found covariation of heart and ventilation rates during intermittent ventilation, hyperventilation, gill cleaning, and swimming, as well as evidence of transient periods of phasic coordination. The heart appears to be weakly entrained to the gill rhythm by phasic cardioregulatory nerve input. The preferred phase of heartbeats, with re- spect to gill rhythm, was 0.5, or 1 80 degrees out of phase. In some animals, intra-cardiac pressures were enhanced when the heart and gill rhythms were entrained. We sug- gest that rhythmic movements of the gill plates enhance the flow of low pressure blood returning from the body to the heart. Thus, ventilatory appendage movements may constitute an accessory blood pumping mechanism in Limulus. Introduction Many invertebrates, including annelids, molluscs, ar- thropods, and echinoderms, have highly developed cir- culatory and respiratory systems. In many cases the mor- phology is well known, but the functional properties and Received 24 March 1989; accepted 22 September 1989. * Address for correspondence. interactions of these convective systems during normal behavior have not been examined in detail. Evidence from research on decapod crustaceans suggests that in- vestigations of the respiratory and circulatory systems of aquatic arthropods might yield valuable information for understanding the operating conditions and physiologi- cal role of coupling between these systems. Close coordi- nation and coupling between circulatory and respiratory systems under a variety of environmental conditions has been reported frequently (Homams arnericanus. Cancer productiis. McMahon and Wilkens, 1975, 1977; Young and Coyer, 1979; Cancer magister, Wilkens et at., 1974; Carcimisnuienus. Young, 1973; Cancer borealis, Cancer irroratus. Coyer, 1977). Nevertheless, the physiological significance of respiratory/circulatory coupling is not fully understood. The coordination of cardiac and ventilatory activity in Limulus polyphemus presents another interesting exam- ple, worthy of further investigation. Watson and Wyse (1978) reported frequency covariation (heart and venti- lation rates changed together) and phasic coordination (heart beats occurred in phase with the gill plate move- ments) of heart and ventilatory systems under normoxic and hypoxic conditions, as well as during intermittent ventilation, hyperventilation, gill cleaning, and swim- ming. The frequency covariation observed in Limulus, as well as many other organisms, may serve primarily to adjust the volume of blood circulated to the rate of oxy- gen uptake. On the other hand, the adaptive significance of phasic coordination between the gills and heart is not clear. Watson and Wyse (1978) suggested that phasic co- ordination makes the two blood pumping systems more efficient. While the role of the gills as an accessory "heart" has been suggested previously (Patten and 386 LIMVLUS GILLS AS CIRCULATORY PUMPS 387 Cannula 1st Bronchiocardiac Canal / Figure 1. Dorsal cutaway illustration of Linntliis /vlypliciniix, illus- trating the position of catheters and electrodes. Redenbaugh, 1899; Lockhead, 1950), little physiological evidence is available to support these interpretations. The present contribution provides evidence that the gill plates help circulate the blood in Limn/us, and docu- ments the hemodynamic relationships between the car- diac and ventilatory systems of intact horseshoe crabs during all of their known respiratory behaviors. In addi- tion, we present evidence suggesting that, in some cases, phase coupling between the heart and gill rhythms en- hances cardiac output. In crustaceans, there is good evidence that frequency covariation between the gills and heart is mediated by the cardioregulatory nerves (Field and Larimer, 1975a, b; Young, 1978). In a few instances investigators re- corded phasic activity in the cardioregulatory nerves that appeared to be coupled to the ventilatory rhythm; they suggested that this type of activity would help phase-lock the heart rhythm to scaphognathite movements. In Lim- ulus the cardioregulatory nerves also appear to be impor- tant in the tonic regulation of heart rate and in the coor- dination between the cardiac and ventilatory rhythms (Watson, 1979). However, phase coupling between the heart and gills could also be mediated by direct pressure cues. In this study we found that phase coupling between heart and gill rhythms was associated with rhythmic bursts in the cardioregulatory nerves. These data suggest that in Limulus both tonic and phasic coordination between the heart and gills is regulated by the nervous system. BCC Pressure Gill EMG ECG 20 10 •0 B D cmH20 -20 -10 -0 155 Figure 2. The relationship between pressure (cm H2O) in a branchiocardiac canal (BCC, 2nd canal), the electromyogram (EMG) of the gill closer muscle (20) of the 2nd gill plate and heart activity (ECG) during rest (A) and increasing amplitudes of ventilation ( B-D). Similar records obtained during hyperventi- lation and swimming are presented in a subsequent figure. Pressures are relative to ambient in the water column. 388 M. A. FREADMAN AND W. H. WATSON III Materials and Methods 28 Specimens of Limit/us polyphemus (18-25 cm cara- pace width) were either obtained from the Marine Bio- logical Laboratory or collected from Buzzards Bay, Fal- mouth, Massachusetts. Animals were maintained in ei- ther a recirculating seawater system (University of New Hampshire, Durham, New Hampshire) or a 1200-liter tank continuously supplied with fresh seawater (MBL, Woods Hole, Massachusetts) at 13-15°C (30-33%») and fed mussel bits every few days. Survival and health of the animals in the laboratory over several months were excellent. Twelve animals were used in our blood pres- sure experiments; and six were used for neurophysiology. Electrical activity of the heart was recorded with pairs of 40- or 45-gauge stainless steel wire (annealed, epoxy coated for insulation) inserted through small holes in the dorsal midline of the opisthosoma, adjacent to the car- diac ganglia. Electrical activity of ventilatory muscles was recorded with similar wires inserted into muscle 20 (Patten and Redenbaugh, 1899) through small holes in the overlying cuticle (Fig. 1 ). Implants were held in place on the cuticle surface with a small amount of methylcya- noacrylic adhesive or dental wax. Pressures in the heart and in the branchiocardiac ca- nals (which convey blood from the gills to the pericardia! cavity) were measured with polyethylene tubing (PE 100) and pressure transducers (Statham P23ID). Cathe- ters with beveled tips were inserted into the heart or one of the branchiocardiac canals through small holes drilled in the overlying cuticle, and cemented in place with quick-drying epoxy (Fig. 1 ). Prebranchial blood pressure was also measured at the junction of the mesosoma and telson. Catheters were kept as short as feasible (about 1 8 cm), and their insertion positions were confirmed at the termination of experiments by dissection. At times, we noted reduced frequency and amplitude responses in the pressure records irrespective of the activity of the animal; this was due to clotting at the proximal end of the cathe- ters and was alleviated by gentle flushing. Most experi- ments were conducted in cold seawater ( 1 3°C) to reduce the tendency of blood clotting in the catheters. Following surgical procedures, animals were allowed to rest for 4-5 h or overnight before measurements were made. Oxygen in the seawater of the test chamber was maintained above 90% saturation during the acclima- tion period and the experiments. Experiments were car- ried out in 80-liter Plexiglas chambers filled with seawa- ter and arranged with a mirror underneath the test cham- ber for observation of the gill appendages. Wire electrodes were attached to a sliding rack above the chamber to prevent fouling of the implants. Some of the crabs were restrained in the chamber on a Plexiglas shelf with a hole situated under the animals abdomen, so the 24 £ i 16 z < u \M/ HYPERVENTILATION • o o QUIESCENT I I I _L _L J 246 10 AMPLITUDE INTEGRATED EMG (ARBIT. UNITS) Figure 3. The relationship between strength of gill contractions and peak BCC pressure. Pressure in the 2nd branchiocardiac canal and inte- grated electromyogram (EMG) of a gill closer muscle were obtained during quiescent behavior, increasing ventilatory activity and hyper- ventilation in two different animals (open and closed circles). The pres- sures recorded from the BCC's were proportional to the strength of gill plate contractions. gill appendages could ventilate freely. Others were al- lowed to move and swim freely about the chamber. Sig- nals of heart and ventilatory muscle electrical activity, as well as the pressure waveforms from the heart and bran- chiocardiac canal catheters, were recorded individually, or simultaneously, on adjacent channels of a Grass Poly- graph (Model 7D). Cardioregulatory nerve recordings were obtained us- ing suction electrodes from animals restrained in a Plexi- glas chamber as described above. A small window was cut in the carapace to expose the cardioregulatory nerves. All signals (EMG, ECG, and extracellular potentials) were detected with Grass P5 1 1 amplifiers, displayed on an oscilloscope, photographed with a Grass C4 camera, and recorded on magnetic tape. Recordings from iso- lated ventral nerve cords were carried out as described in rt/. (1980). Results General respiratory behavior During experiments lasting 7- 1 6 h, most specimens of Linutlus polyphemus exhibited several types of ventila- tory activity: intermittent ventilation, hyperventilation, gill cleaning, and swimming. Intermittent ventilation in LIMULUS GILLS AS CIRCULATORY PUMPS A B cm H2O HO 389 BCC Pressure Gill EMG -•MH 44444 15 s Figure 4. Pressure in one branchiocardiac canal, gill EMG and ECG during hyperventilation (A) and swimming, as indicated by the large amplitude gill EMG's(B). Limulus occurs as alternating bouts of ventilation and apnea. Both hyperventilation and swimming are charac- terized by a large increase in rates of gill plate movements over that observed in shallow ventilation (Knudsen, 1973). The major difference between the two behaviors is that during swimming the legs also move in phase with the gills, while during hyperventilation leg movements are independent of gill activity. During gill cleaning, the gill plates move across the midline and flick the inner lobe of a gill plate between the book gill lamellae of the opposite side. In all the aforementioned ventilatory ac- tivities, heart rate was always positively correlated with the rate of ventilation. These respiratory and locomotor behaviors of Limulus have also been reported in earlier papers (Watson and Wyse, 1978; Watson, 1980a, b). Pressures in the branchiocardiac canals during rest, ventilation, hyperventilation, and swimming The branchiocardiac canals (BCCs), as the term im- plies, carry blood from the gills to the pericardia! cavity and heart. Five pairs of canals correspond to the five gill appendages (Fig. 1 ). Blood from the venous circuit passes through an extensive gill network (Lockhead, 1950; Jo- hansen and Petersen, 1975) before returning to the heart. We measured pressures in one canal, while simulta- neously recording the electrical activity of that gill's closer muscle, during rest (i.e.. no ventilatory activity oc- curring), ventilation, hyperventilation, and swimming. During rest, no pressure pulses occur in the canal, al- though the pressure may be greater than zero (Fig. 2A). As the amplitude of ventilation increases, pressure in the canal rises substantially (3-20 cm H2O), as does the am- plitude and frequency of the recorded EMG and the heart rate (Fig. 2B-D). To examine more closely the relationship between the strength of gill contractions (as monitored by EMG ac- tivity) and the resulting BCC pressures, we plotted inte- grated EMG amplitudes versus the magnitude of canal pressures, during several different intensities of ventila- tory activity, ranging from shallow ventilation to hyper- ventilation. There is a direct relationship throughout the range of pressures measured (Fig. 3). This relationship was consistent during all of our experiments, whether the animals were restrained or freely moving. We also recorded pressures in a BCC during hyperven- tilation and swimming. Examples of these results are shown in Figure 4. Hyperventilation was sometimes in- termittent, and with the onset of this behavior, large pres- sure pulses occur in the canal (Fig. 4A). During swim- ming, which is also periodic, the record shows essentially the same result, but the recorded pressures are more er- ratic (Fig. 4B). This is probably due to movement arti- facts introduced to the liquid-filled pressure catheter sys- tem during swimming as the animal moved up and down in the chamber. Prebranchial pressures (junction of soma and telson) are low during rest ( 1-2 cm H2O, n = 9) and increased only slightly during ventilatory periods (2.5-3 cm H;O). Therefore, the blood reaching the gills is at a low pres- 390 M. A. FREADMAN AND W. H. WATSON III sure, and the rhythmic movements of the gills are likely important for circulating venous blood dorsally to the heart. Frequency covariation between heart and gill rhythms To assess the relationship between heart pressures and BCC pressures, we obtained simultaneous recordings of closer muscle electrical activity (EMG), BCC pressures, heart electrical activity (ECG), and intracardiac blood pressure (Fig. 5A). Throughout all of our experiments, a strong frequency coordination existed between the two rhythmic systems. When gill ventilation increased, so did the pressures in the BCC's (Figs. 2-5), and, presumably, the rate of venous return. Thus, from a hemodynamic perspective, frequency covariation between heart and gill pumps insures that the venous return is coordinated with the tonic output of the heart. Phasic coordination between heart and gills Most of the animals examined did not exhibit strong phase coupling between the heart and gill rhythms (see Fig. 6 for illustration of how phase coupling was deter- mined). However, a small proportion (10-20%) of the animals examined either strongly phase coupled, or drifted in and out of the coupling mode. When phase coupling did occur, the preferred phase was usually close to 0.5 (Fig. 6), indicating that the heart beat 180° out of phase with the gills. Why is 0.5 the preferred phase? If there is some hemodynamic advantage, then the output of the heart should be maximal when the two pumps are 180° out of phase. To test this hypothesis, we recorded heart and BCC pressures for approximately 1 h, and then plotted peak cardiac pressures versus phase of the heart- beat with respect to the gill rhythm, on a beat-to-beat basis. We found that, in some of the animals, intra-car- diac pressure remained relatively constant regardless of ventilatory activity (Fig. 7B). In three animals, however, the cardiac pressures were greatest when the heart and ventilatory rhythms were 180° out of phase (Fig. 7A). Thus, at least in some animals, phase coupling may en- hance cardiac output. Cardioregulatory nerve activity during phase coupling If the heart is entrained by the nervous system, it must receive phasic timing cues through the cardioregulatory nerves. Furthermore, this timing information must have a fixed phase relationship to the ventilatory rhythm so it can entrain the two oscillatory systems. We recorded cardioregulatory nerve activity from intact, immobilized animals while monitoring heart ECGs and gill plate EMGs (Fig. 8). During periods when the gill plates were making large, rapid movements, there was a phasic com- BCC Pressure Gill EMG Heart Pressure r L cm H,O LB ECO H-,O • H.H.M.^.. ».»*-.».»• Figure 5. Pressures (cm H:O) and electrical activity in the respira- tory and circulatory systems of Limulus, The top tour records show pressure in Pc branchiocardiac canal (BCC), EMG of the 1s1 gill muscle, pressure in the heart and the ECG. The bottom four traces are the same physiological processes and include a section where ventilatory activity ceased. Pressures are relative to ambient in the water column. ponent to the cardioregulatory nerve activity (Fig. 8 A, C). Bursts comprising several different units tended to occur during and just after contractions of the gill plate closer muscle. Periods of silence followed for the remain- der of the gill interburst interval. When this type of car- dioregulatory nerve activity was present the heart was phase locked to the ventilatory cycle (see top plot. Fig. 8). But when ventilation slowed and became weaker, cardioregulatory nerves lacked phasic activity, and phase coupling did not occur (Fig. 8B, D, and top of Fig. 8). To eliminate the possibility that phasic activity was due to sensory modulation, we recorded cardioregula- tory nerve activity from isolated ventral nerve cords. When there were strong bursts in the ventilatory motor nerves, we also recorded bursts in the cardioregulatory nerves that were phase locked to the ventilatory central pattern generator (CPG) (Fig. 9). During quiet ventila- tion, however, only tonic firing was present. Thus, phasic cardioregulatory nerve activity is a component of the ventilatory central pattern generator. Although these data do not prove that the phasic cardioregulatory nerve activity is responsible for entraining the heart to the ven- tilatory rhythm, we suggest that phase coupling is most LIMULUS GILLS AS CIRCULATORY PUMPS A. Calculation of Phase Relationship between Heart and Gill Rhythms Heart Pressure ECG BCC Pressue Gill EMG 391 lOcmH.0 Phase B. Phase Histogram 60 - C IS 50 - LH Ou 40 - Heatbeats o o o MK^H • : WssA ixxxxxxi YSSSM 01 02 03 04 05 06 07 08 09 1 Phase Figure 6. Phase relationships between heart and gill rhythms. A. Actual records of BCC and heart pressures, gill EMG, and ECG. showing how the phase relationship between the two rhythms was calcu- lated. Phase is calculated by dividing the latency (time between onset of a gill muscle burst and onset of cardiac ganglion burst) by the duration of the concurrent gill interburst interval (time between the begin- ning of one burst and the beginning of the next burst). B. Histogram demonstrating phase preference of a cardiac ganglion burst (recorded as ECG) with respect to concurrent ventilatory event. The number of heartbeats from a particular phase were then totaled and a histogram constructed. Heartbeats per bin indicates number of events falling in each decile of phase (e.g.. between 0.2 and 0.3). In this experiment, there was a very strong phase preference of 0.5 indicating the heartbeat was 180° out of phase with ventila- tory activity. We did not find such strong phase preference in all animals or in any single animal at all times. But. when phase preferences existed, they were usually close to 0.5. likely mediated by neural rather than hemodynamic mechanisms. Discussion Circulatory systems have traditionally been divided into two types: closed systems, in which the blood is al- ways in distinct vessels separated physically from inter- cellular fluids; and open systems, with large, ill-defined cavities, in which the blood is not separated from the in- tercellular fluids. The open systems have been regarded as poorly developed and capable of generating only low pressures and low blood flow velocities. The open circu- latory system of Limulus polyphemus is well developed and appears to have few of the large sinuses normally found in open circulatory systems (Shuster, 1978; Red- mond el a/.. 1982). Our study indicates that high pres- sures occur in various portions of the circulatory system, particularly in the heart and branchiocardiac canals. Sumwalt (1933) and Abbott et al. (1969) also reported high pressures in the heart of Limulus. Redmond et al. (1982) measured blood pressures in and around the heart and body of resting horseshoe crabs in water and air. They attributed the moderate variability 392 M. A. FREADMAN AND W. H. WATSON III A. Heart Entrained to Gill Rhythm z u B Ed Phase B. Heart and Gill Rhythms Out of Phase V u Q. DJD C cd JS CJ Phase Figure 7. Influence of heart/gill phase relalionship on cardiac pres- sures. On a beat-to-beat basis the phase relationship between the heart and the gill rhythms was calculated and plotted versus the amplitude of the pressure change during the subsequent heartbeat. In some animals, when the heart and gills were approximately ISO" out of phase (0.5). the pressures recorded from the heart were higher (as illustrated by data from the animal plotted in A). In other animals there was no apparent influence of heart/gill phase on cardiac output (as shown by data from a different animal plotted in B). of their data to differing activity levels. We monitored ventilatory muscles and, from previous experience (Wat- son and Wyse, 1978), could identity quiescent behavior, intermittent ventilation, hyperventilation, gill cleaning, and swimming. These are the most characteristic activi- ties that involve the gill appendages (Watson and Wyse, 1978; Watson, 1980a, b). Rhythmic pressure changes in the BCCs were always associated with movements of the gills, and their amplitudes varied with the type and strength of gill movements. Thus, variability in pressures recorded in the heart and BCCs are probably due to the animal switching from one type of respiratory activity to another. It is likely that the rhythmic movements of the gill ap- pendages of Limit/us serve to enhance gas exchange and to pump oxygenated venous blood to the pericardia! cav- ity. It has been reported that blood moves from the large ventral sinuses through the gill lamillae and BCCs to the pericardial cavity (Patten and Redenbaugh, 1899; Lock- head, 1950; Redmond et ai, 1982). We have also ob- served blood moving in this manner in juvenile horse- shoe crabs. In this study we recorded large pressure pulses in the branchiocardiac canals during normal ven- tilation, hyperventilation, and swimming — all impor- tant components of the daily activities of horseshoe crabs. These changes in pressure always occur during contraction of the gill closer muscles, and their ampli- tudes are proportional to the magnitude of the gill plate movements. We never detected pressure surges in the branchiocardiac canals during quiescent behavior, when the gill plates were not moving, but the heart was still beating. Thus, it is likely that venous return to the heart is assisted by the gill appendages, which may serve as an accessory heart in Linntlus. Although pressure pulses recorded from the BCC's are likely caused by movements of the gill appendages and result in the flow of blood from the gills to the pericardial cavity, additional studies will be required to prove this hypothesis. In particular, it is important to determine if valves are present in the BCCs to rectify flow, and to con- tinuously monitor the dynamics of blood flow in the ca- nals. Apparently, enhanced blood flow from ventilatory movements in horseshoe crabs is established in early life stages. We had the opportunity to observe blood flow in young specimens ofLimulus (5 mm carapace width) un- der a dissecting microscope, because animals this small are quite transparent. With each ventilatory movement and during swimming activity, blood rapidly pulses into the pericardial cavity and heart. We also observed covari- ation of heart and ventilatory rates. Occasionally, horseshoe crabs exhibit phase coupling between the heart and gill rhythms. Our analyses suggest that, in some animals, this coupling may increase cardiac output. It may also allow the animal to do the same amount of circulatory work more efficiently, but we have no data to support this hypothesis at present. The pre- ferred phase of almost all animals that demonstrated phase coupling was approximately 0.5, or 180° out of phase. Why would this relationship between two systems lead to enhanced blood flow? When the gill plates close, due to the contraction of gill muscle 20, oxygenated blood is forced dorsally through the BCCs to the pericar- dial cavity. If the heart and gills are 180° out of phase, then the heart will be relaxing as blood enters the pericar- dial cavity (see Fig. 6). When the heart relaxes, blood en- ters the heart through the ostia. Thus, maximal blood is available to the heart during the filling phase of its con- traction-relaxation cycle. This appears to increase the L1MULUS GILLS AS CIRCULATORY PUMPS 393 1.0 o.sh LJ 0.6 CO < £ 0.4 0.2 0.0 2R20J C.N.R9W Heart" CYCLE NUMBER =t i==±; 50Mv| 4 sec Figure 8. The relationship between bursting in the cardiac nerves and the degree of coupling between the heart and gill rhythms. The top graph shows a sequential phase plot of heartbeats with respect to the concurrent gill interburst intervals (Phase = heart latency/duration of the gill burst interval). In this particu- lar experiment, there was good coupling between the heart and gill rhythms during segments A and C, and drifting, or no phase coordination during B and D. Representative segments of records from these periods are shown below ( A-D). During periods of coupling (A and C) there was discrete bursting in cardiac nerve 9. while during periods of dotting (B and D). the cardiac nerve activity was more weakly phasic. This suggests that coupling may be related to the degree of phasic activity in the cardiac nerves. force of the next heartbeat (Fig. 7) and, as a consequence, may lead to an increase in cardiac output. We cannot estimate the quantitative contribution of blood flow to the heart from ventilatory movements of the gill appendages, or cardiac output from our data be- cause we did not measure blood flow. Measurements of cardiac output in Limiilus at rest, calculated from the Pick equation, are reasonably high (Mangum el al.. 1975) and we expect the values to increase during activ- ity. The use of microcatheter tip flow probes should al- low measurements of cardiac output; then the possibility that phase coupling in Limiilus leads to enhanced car- diac output can be more rigorously tested. Phase coupling between Limiilus heart and gills ap- pears to be mediated by the cardioregulatory nerves. Bursts of action potentials, which are phase locked to the ventilatory CPG, can be recorded from the cardioregula- tory nerves in intact animals and in isolated preparations (Figs. 8, 9). When bursts are present in some animals, the heart is coupled to the gill rhythm; when the bursts are E M I CN 10 sec Figure 9. Rhythmic bursts recorded from cardioregulatory nerves in isolated ventral nerve cords. It is possible to monitor motor patterns which represent ventilation by recording from the External (E), Medial (M)and Internal (I) branchial nerves in the isolated nerve cord. Strong ventilatory motor patterns produced by the isolated ventral nerve cord are accompanied by rhythmic bursts in cardioregulatory nerve #10 (CN). which emanates from the 2nd abdominal ganglion. This indicates that the ventilatory central pattern generator may also provide timing information to the cardioregulatory nerves, and thus to the heart. 394 M. A. FREADMAN AND W. H. WATSON III not present, the two pattern generators run indepen- dently. But the heart may also be receiving timing input from the blood pressure pulses in the BCCs. To separate these two influences and to determine which is the con- trolling factor, we used a motor to move all the gill plates at a frequency that was slightly different from the output of the ventilatory CPG. We monitored the output of the ventilatory CPG by recording EMGs in the gill muscles. We found that the heart was phase coupled to the ventila- tory CPG, but not to the imposed gill movements. In other words, the heart was entrained by some type of neural input that was phase locked to the ventilatory CPG, but was out of phase with pressure pulses from the BCCs (unpub. data). Thus, we conclude that phase cou- pling between the heart and gills in Limulus is mediated by the cardioregulatory nerves. The coordination between cardiac and ventilatory rhythms in Limulus is very similar to that in decapod crustaceans. Decapods also exhibit frequency and phase coordination between the two systems (Wilkens cl a/.. 1974; McMahon and Wilkens, 1975; Wilkens, 1976; Field and Larimer, 1975a, b; Young and Coyer, 1979). The frequency coupling in decapods is not as tight as in Limulus, but the phase coupling is very similar. It only occurs in a small percentage of the animals studied and ranges from strong coupling to a drift and lock mode, to nonexistent. In decapods, cardioregulatory nerves also appear to mediate phase coupling. Young (1978) showed that cardioinhibitory units, and perhaps cardioaccelera- tory units as well, occasionally fired in bursts that were phase locked to the scaphognathite rhythm. In Limulus, the inhibitory and excitatory units in the cardioregula- tory nerves have not yet been identified. Nevertheless, in both decapods and Limulus, phase coupling appears to be mediated by cardioregulatory nerves that transmit in- formation from ventilatory CPGs in the CNS to the car- diac ganglion. Further experiments are necessary to demonstrate which units are involved and to examine why coupling strength is so variable. The movement of fluids in circulatory systems of in- vertebrates is accomplished by a variety of mechanisms, including tubular "hearts," sac-like hearts, chambered hearts, contracting blood vessels, several types of acces- sory pumps and body movements (Schmidt-Nielsen, 1983). In general, locomotion and associated body movements are important in blood flow in many small and large animals, coupling metabolic needs with deliv- ery of nutrients and oxygen to the tissues. This strategy appears to be an important facet of hemodynamics. The horseshoe crab provides yet another example of the efficiency of this organization. The rapid movements of the gill plates that are associated with swimming also serve to ventilate the book gill lamellae and pump the oxygenated blood to the pericardia! canal. Acknowledgments We thank Dr. Arthur B. DuBois (John B. Pierce Foun- dation, New Haven, Connecticut), the Grass Instrument Company (Quincy, Massachusetts), and Dr. John Sasner (University of New Hampshire) for loaning pressure transducers for our initial experiments. Dr. Daniel G. Gibson (Worcester Polytechnic Institute) taught us how to collect very young horseshoe crabs in the field and also pointed out their transparent character. Some of the ex- periments were conducted in Dr. Gordon Wyse's labora- tory (University of Massachusetts, Amherst) as part of W.H.W. Ill's thesis work. We are extremely grateful to him for his advice and the use of his facilities. Dr. John Sasner and Dr. Gordon Wyse also made numerous help- ful criticisms on earlier versions of this manuscript. This research was supported in part by the Ocean Industries Program at Woods Hole Oceanographic Institution (M.A.F.), a Grass Fellowship to W.H.W. Ill, a Faculty Research Grant from the University of New Hampshire (W.H.W. Ill), and a grant from NINCDS (19053, W.H.W. III). Literature Cited Abbott, B. C, F. Lange, I. Parnas, VV. Parmley, and E. Sonnenblick. 1969. Physiological and pharmacological properties of Limulus heart. Pp. 232-243 in E.\perientia Supplement 15, Comparative Physiology of ike Heart. Current Trends Birkhauser Verlag. Stutt- gart. Coyer, P. E. 1977. Responses of heart and scaphognathite rates in Cancer horcalis and C irroratus to hypoxia. Comp. Biochem. Phvs- ii>l. 56A: 165-167. Field, L. H., and J. L. Larimer. 1975a. The cardioregulatory system of crayfish: neuroanatomy and physiology. / Exp. Biol. 62: 519- 530. Field, L. H.. and J. L. Larimer. 1975b. The cardioregulatory system of crayfish: the role ofcircumesophageal interneurons. / Exp. Bin/. 62:531-544. Johansen, K., and J. A. Petersen. 1975. Respiratory adaptations in Limulus polyphemus (L.). Pp. 129-145 in Eco-p/iysiologyofEstua- rine Animals. F. J. Vernberg, ed. University of South Carolina Press. South Carolina. Knudsen, F. I. 1973. Muscular activity underlying ventilation and swimming in the horseshoe crab, Limulus polvphemus. Biol. Bull. 144: 355-367. Lockhead. J. H. 1950. Xiphosura polyphemus. Pp. 360-381 in Se- lected Invertebrate Types. F. A. Brown, ed. John Wiley, New York. Mangum, C. P., M. A. Freadman. and K. Johansen. 1975. The quan- titative role of hemocyamn in aerobic respiration of Lirmdus poly- phemus. J. Exp. Biol. 191: 279-285. McMahon, B. R., and J. L. Wilkens. 1975. Respiratory and circula- tory responses to hypoxia in the lobster Homarus americanus. J Exp. Biol. 62:637-655. McMahon, B. R., and J. L. Wilkens. 1977. Periodic respiratory and circulatory performance in the red rock crab Cancer productiis. J Exp. Biol 202: 363-374. Patten, W., and VV. A. Redenbaugh. 1899. Studies on Limulus II. The nervous system of Limulus polvphemus with observations upon the general anatomy. / Morphol. 16: 91-200. LIMULUS GILLS AS CIRCULATORY PUMPS 395 Redmond. J. R., D. D. Jorgensen, and G. B. Bourne. 1982. Circulatory physiology ofLimulus. Pp. 133-146 in Physiology and Biology of Horseshoe Crabs, 3. Bonaventura. C. Bonaventura, and S. Tesh eds. Alan R. Liss. New York. Schmidt-Nielsen, K. 1983. Animal Physiology- Cambridge Llniver- sity Press. New York. 6 1 9 pp. Shuster, C. N. Jr. 1978. The circulatory system and blood of the horseshoe crab. U. S. Department of Energy. Federal Energy Regu- latory Commission. 14pp. Sumwall, H. M. 1933. The blood pressure ofLimulus. Biol. Bull 65: 372. Watson, W. H. Ill, and G. A. Wyse. 1978. Coordination of the heart and gill rhythms in Limulus. J Comp. Physiol. 124: 267-275. Watson, W. H. III. 1979. The respiratory behavior of Limulus poly- phennts. PhD Thesis, LIniversity of Massachusetts. Amherst. Watson, W. H. III. 1980a. Limulus ^^ cleaning behavior. J. Comp. Physiol. 141:67-75. Watson, W. H. HI. 1980b. Long-term patterns of gill cleaning, venti- lation and swimming in Limulus. ./. Comp. Physiol. 141: 77-85. Wilkens, J. L., L. A. Wilkens, and B. R. McMahon. 1974. Central control of cardiac and scaphognathite pacemakers in the crab. Can- cer magisier. J Comp Physiol. 90: 89-104. Wilkens, J. L. 1976. Neuronal control of respiration in decapod Crus- tacea. Fed. Proc 35: 2000-2006. Wyse, G. A., D. H. Sanes, and W. H. Watson III. 1980. Central neu- ral motor programs underlying short- and long-term patterns of Limulus respiratory behavior. J. Comp. Phvsiol. 141: 87-92. Young, R. E. 1973. Nervous control of ventilation in the shorecrab Carcimi.i maenux. Ph.D. Thesis, University of St. Andrews. Young, R. E. 1978. Correlated activities in the cardioregulatory nerves and ventilatory system in the Norwegian lobster, Nephrops norvegicus (L.). Comp. Biochem. Physio/. 61 A: 387-394. Young, R. E. and P. E. Coyer. 1979. Phase coordination in the cardiac and ventilatory rhythms of the lobster Homarus americanus. J Exp. Biol. 82: 53-74. Reference: Biol. Bull 111: 346-400. (December. 1989) Inactivation of the Corpora Allata in the Final Instar of the Tobacco Hornworm, Manduca sexta, Requires Integrity of Certain Neural Pathways from the Brain LOUIS SAFRANEK AND CARROLL M. WILLIAMS Department of Cellular and Developmental Biology, Harvard University, Cambridge, Massachusetts 02138 Abstract. Neither the implantation of active CA nor treatment with O-ethyl,S-phenylphosphoramidothiolate (EPPAT), a potent inhibitor of the juvenile hormone es- terase ( JHE). prevented metamorphosis of final instar to- bacco hornworms. However, a combination of the two treatments often blocked metamorphosis and caused the formation of supernumerary larvae or larval-pupal inter- mediates. So also, in conjunction with EPPAT treat- ment, unilateral severance of the medial nerve from the brain to the corpus cardiacum-corpus allatum complex often resulted in abnormal supernumerary or intermedi- ate larval forms. Thus, clearance of JH from mature hornworm larvae prior to metamorphosis appears nor- mally to depend on two mechanisms: ( 1 ) cessation of JH production by inhibition of the CA via the innervation of these glands, and (2) destruction of previously secreted existing JH via production of JHE. In the present experi- ments, each of these mechanisms appeared fully able to clear JH sufficiently to permit normal metamorphosis, because only simultaneous interruption of both mecha- nisms led to formation of supernumerary larvae. Acting in concert as they do late in larval life, these two mecha- nisms ensure the timely and thorough clearance of JH in preparation for metamorphosis. Introduction Juvenile hormone (JH) is indispensable for mainte- nance of the larval condition in Lepidoptera as it is in Received 6 September 1988; accepted 25 September 1989. Abbreviations: JH = juvenile hormone; JHA = juvenile hormone analog; JHE = juvenile hormone esterase; br-cc-ca = brain-corpora cardiaca-corpora allata complexes; CA = corpora allata; EPPAT = O- ethyl.S-phenylphosphoramidothiolate: LD = long day; SD = short day: bl = black larva strain of the tobacco hornworm. many other orders of insects. Conversely, metamorpho- sis normally presupposes the virtual absence of JH and can regularly be derailed by the timely application of JH or JH analogs to mature larvae. Thus the transition from the larva's feeding life-style to the metamorphic se- quence entails elimination of JH. This prerequisite could potentially be satisfied either by curtailing production of JH or by augmenting its destruction. In fact, persuasive evidence supports regulation by both pathways. Thus, production of JH by the corpora allata (CA) declines late in larval life (for review, see Feyereisen, 1985): simulta- neously, breakdown of JH to inactive metabolites is en- hanced by the increased production of enzymes inacti- vating JH such as the so-called "juvenile hormone ester- ase" (JHE) (for review, see Hammock, 1985). In the present communication, we focus on regulation of JH clearance in the final instar of the tobacco horn- worm Manduca sexta. Prior studies of this species have documented a rapid increase in the activity of JHE dur- ing the final instar (Vince and Gilbert. 1977; Sparks et ai. 1983). The importance of this increase has been sup- ported by the demonstration that administration of spe- cific JHE inhibitors can delay metamorphosis by pro- longing the final instar just as can be accomplished by administering JH or its active analogs (Sparks et ai. 1983). Moreover, studies on hornworm CA have shown that their activity declines during the final larval instar (Bhaskaran et at., 1980). JH bioassays have directed at- tention to multiple avenues of CA regulation, including deployment of humoral and neural inhibitors as well as withdrawal of humoral and neural stimulators (Bhask- aran and Jones, 1980; Bhaskaran et ai. 1980; Bhaskaran, 1981). However, neither interruption of JH breakdown nor interference with CA regulation has successfully pro- 396 NEURAL INHIBITION OF CORPORA ALLATA 397 voked supernumerary larval molts or done more than delay the normal pupation of final instar hornworms. Thus, the normal regulatory mechanisms that are critical for JH clearance in mature hornworm larvae remain un- certain. In the experiments described here, we document the key role played by JHE, as well as that of a brain- centered neural inhibition of JH secretion by the CA. Materials and Methods Hornworms were reared as previously described (Sa- franek and Williams, 1980) under a long-day (LD, 17L: 7D) or a short-day (SD, 12L:12D) photoperiod at 25°C. The first day of each instar was termed Day 1 . Bioassays of brain-corpora cardiaca-corpora allata complexes (br- cc-ca) were performed in black (bl) hornworm larvae during the first half of the photophase as previously de- scribed (Safranek and Riddiford, 1975). Implantation of these complexes occurred 1-6 h prior to the onset of head capsule slipping in host larvae, and hosts were scored 48 h later after completion of the molt. The scor- ing system was derived from that previously described (Truman el ctl., 1973) and is summarized here under Re- sults. Denervation of the CA was accomplished in CO2- an- esthetized larvae under Ringer's solution (Safranek and Williams, 1980). A small, three-sided flap of integument was raised near the center of the head capsule, thereby exposing the brain in the center of the field. With forceps placed behind the brain, the latter was delicately tipped forward — a maneuver that permitted adequate exposure of the two small nerves from one hemisphere of the brain to the ipsilateral corpus cardiacum-corpus allatum com- plex. After both nerves had been inspected, the one to be transected was securely grasped with forceps and care- fully severed with scissors. The integumentary flap was replaced and the wound sealed by melted wax. In sham- operated preparations, both nerves were identified on one side but not manipulated. In all experiments, indi- viduals were carefully examined after completion of the molt at the end of the fifth instar for JH-dependent mor- phological abnormalities of the types previously de- scribed (Truman el al., 1974). The juvenile hormone esterase inhibitor O-ethyl,S- phenylphosphoramidothiolate (EPPAT) was a gift of Drs. T. C. Sparks and B. D. Hammock. It was dissolved in ethanol for topical administration to the dorsal thorax or anterior abdomen using a Hamilton syringe and a re- peating dispenser. Results Larcal hornworm CA are inactivated prior to metamorphosis To document inactivation of hornworm CA prior to metamorphosis, br-cc-ca complexes were dissected from o o CO .*.h. Jte4i! Jl WEIGHT (grams) Figure 1. Activity of fifth-instar larval corpora allata in the black larval assay as a function of fifth-instar larval weight. Brain-cc-ca com- plexes from feeding nfth-instar larvae of various weights were singly implanted for assay into pharate fifth black hornworm larvae. These assay larvae were scored after completion of the molt to the fifth instar using the following scoring system: 0 = fully black pigmentation; 0.5 = grey head or thorax; 1.0 = green head or thorax; 1.5 = green head or thorax with faint green pigmentation apparent in the abdomen; 2.0 = grey-green abdomen; 2.5 = excessive melanization restricted to cu- ticular creases across the abdomen; and 3.0 = normal pigmentation. Each point represents assay of one brain-cc-ca complex. SD fifth-instar hornworms and implanted for bioassay directly into fourth-instar black hornworm larvae as de- scribed under Materials and Methods. All operations were performed on bl larvae during the first half of the photophase so that the implanted complexes were pres- ent shortly before the onset of head capsule slipping by the bl larvae, at which stage the bl fifth-instar pigmenta- tion can be modified by exposure to JH. The bl assay is useful when comparing the activity of br-cc-ca com- plexes producing amounts of hormone less than that re- quired to generate the assay's maximum score of +3. As summarized in Figure 1, the complexes of larvae weigh- ing less than 8 g on the morning of assay were nearly always active, whereas those from larvae weighing more than 8 g at that stage of the photocycle were nearly always inactive. Larvae weighing over 8 g during the first half of the photophase usually initiated metamorphosis (sig- naled by dorsal vessel exposure and onset of the wander- ing behavior) during the next scotophase; smaller larvae delayed metamorphosis for one or more days. Thus, in these experiments, complexes actively secreting JH were 398 L. SAFRANEK AND C. M. WILLIAMS regularly identified from larvae that were at least 36 h before the onset of the wandering period, whereas inac- tive complexes were found when donor larvae would have initiated wandering in less than 12 h. Thus, inacti- vation of the CA occurs from 1 2 to 36 h before the visible onset of the wandering period. D enervated active CA block metamorphosis only in conjunction with EPP AT administration A series of experiments were designed to evaluate the developmental responses to implantation of denervated CA in the presence or absence of EPPAT. For these stud- ies we selected Day 2 LD fifths weighing 3.3-5.0 g, be- cause the above data suggested that all larvae of this stage would have CA actively secreting JH. One group re- ceived implants of single CA from larvae of a similar weight and stage. A second group received application of 20 nmol of EPPAT (2 ^D every 8 h, beginning late on Day 2 and continuing until dorsal vessel exposure sig- naled the onset of metamorphosis. This dose and sched- ule of administration had been shown in preliminary ex- periments neither to delay the onset of metamorphosis by intact fifth-instar hornworms nor to give rise to mor- phological aberrations in the resulting pupae. A third group received implantation of a single CA on Day 2 fol- lowed by 20 nmol of EPPAT (2 /ul) every 8 h beginning on Day 3 and continuing until the appearance of dorsal vessel exposure or of head capsule slipping. Three con- trol groups received either a sham-operation or topical ethanol application (2 jul) every 8 h or a combination of sham-operation and topical EPPAT application (20 nmol EPPAT in 2 ^1 ethanol every 8 h). All larvae were restored to diet after initiation of treatment. Results of these experiments are summarized in Table I. Larvae that received both implantation of isolated CA as well as topical application of EPPAT were the only group substantially different from the controls. Of the 23 larvae in this group, 12 formed larval-pupal intermedi- ates similar to those seen after JH treatment of late fifth- instar larvae. Among these individuals, head capsule slipping occurred on average at Day 10 of the instar. Even the 1 1 larvae in this group that did not form inter- mediates appeared, nevertheless, to have been affected by the treatment: wandering was typically delayed until Day 10 of the instar and only 7 of the 1 1 pupated nor- mally, 3 others died prior to pupation; and 1 displayed multiple abnormalities but no larval features. These findings were in contrast to the development of the con- trol groups and of the groups in which either implanta- tion of denervated CA or topical EPPAT application were singly employed: all groups initiated metamorpho- sis with, at most, slight average delays, and all individuals survived to form normal pupae. Table I Effects of treatment wiih EPPAT. denervated C A. or both Days to wander % Larval-pupal Treatment Number or molt* intermediate# Sham operation 9 7/0 0 Sham (ethanol) application 19 6/0 0 Implantation of active CA 9 7/0 0 EPPAT application 19 6/0 0 Sham operation plus EPPAT application 20 8/0 0 CA implantation plus EPPAT application 23 10/10 52 * Mean day of fifth instar on which initiation of the wandering period or of head capsule apolysis was first noted. Treatment was initiated on LD fifth-instar larvae weighing 3.3-5.0 g late on Day 2. # Intermediates ranged from sixth-instar larvae of nearly normal ap- pearance to largely pupal forms with retained prolegs and mandibles. The remainder exhibited patches of larval and pupal cuticle distributed in patterns characteristic of fifth-instar larvae treated with JH analogs (Truman cl ni. 1974). Few supernumerary larvae or intermediates were able successfully to complete ecdysis at the end of the fifth instar. nor did any feed substantially or undertake further development before death. We also attempted to abort the normal metamorpho- sis of fifth-instar larvae similar to those above (LD, Day 2, 3.5-5.0 g) by implanting individual loose CA from fourth-instar larvae (n = 22), or Day 1 fifth-instar larvae (n = 13), or adult females (n = 7). CA from all these stages are highly active in the black larval bioassay or in a pupal bioassay (Safranek and Williams, 1987; unpub. results). In no instance did the implanted CA prevent pu- pation or provoke retention of larval features. Another experiment used older LD fifth instars ap- proximately 30 h prior to the onset of the wandering pe- riod. Implantation of a single CA from a Day 1 fifth-in- star (n = 1 3) or treatment with EPPAT (100 nmol every 8 h) (n = 9) did not prevent normal pupation. By con- trast, combination of the two treatments once again led to the occurrence of sixth-instar larvae or larval-pupal intermediates in 5 of 8 larvae. Thus, as in the previous experiments, neither implantation of CA nor adminis- tration of EPPAT deployed individually substantially al- tered either the time course or character of metamorpho- sis: only combination of the two treatments regularly prevented normal pupation and led to retention of larval characteristics. Not all EPPAT-treated preparations containing dener- vated CA formed larval-pupal intermediates or supernu- merary larvae. Though we are uncertain of the reason for this, we suspect that the dosage of EPPAT used here may have been inadequate to ensure a response by all larvae. Our limited supply of EPPAT prevented extensive exam- NEURAL INHIBITION OF CORPORA ALLATA 399 Table II Effects of in situ denervalion oj the CA coupled with EPPAT application Number % Larval-pupal Operation oflarvae# intermediates* Sham 27 0 Sever a lateral nerve to an active CA 19 1 1 Sever a medial nerve to an active CA 2? 64 # The numbers represented include only those surviving to molt. Four sham-operated preparations died postoperatively, as did three preparations with a severed lateral nerve, and five with a severed medial nerve. * AH individuals not forming larval-pupal intermediates formed pu- pae that were normal except for minor aberrations about the head be- lieved secondary to the surgery. Larval-pupal intermediates demon- strated the same range of forms described in Table I. The heads of all preparations forming largely larval intermediates were dissected after the molt and in every case the persistent section of the originally oper- ated nerve was confirmed. ination of the response to higher EPPAT dosages. More- over, we wished to avoid the use of EPPAT at higher dos- ages that cause developmental delays even in intact lar- vae without denervated CA (Sparks el a!.. 1983; pers. obs.)- Our limited experience with higher EPPAT dos- ages up to 200 nmol every 8 h suggests that more aggres- sive EPPAT-treatment to larvae with loose or denervated CA might well result in an even higher percentage of lar- val-pupal intermediates than described in the present ex- periments. Denervation ofCA in situ can block metamorphosis in conjunction with EPPAT administration Table II summarizes a final series of experiments on fifth-instar hornworms; here we succeeded in unilaterally denervating active CA without otherwise disturbing them. In these experiments we used mature SD fifth in- stars 24-36 h prior to the outset of the wandering period. Larvae at this stage could be removed from diet postop- eratively without the significant developmental delay routinely encountered after operation and starvation of younger larvae. Either the medial or the lateral nerve from the brain to the corpus cardiacum of one side was transected as described under Materials and Methods. After the operation, all individuals were removed from diet and received an initial dose of 120 nmol of EPPAT followed by 50 nmol every 8 h. As shown in Table II, transection of a medial nerve to a CA resulted in the for- mation of a larval-pupal intermediate by the majority of larvae. Discussion Denervation of the CA figured critically in the aborted metamorphosis of the larval-pupal intermediates gener- ated in the present experiments. Either implantation of active CA or denervation of in situ CA often generated supernumerary larvae and larval-pupal intermediates when these maneuvers were deployed along with admin- istration of the potent JHE inhibitor EPPAT. These ab- errant forms were indistinguishable from the supernu- merary larvae and larval-pupal intermediates seen after administration of large doses of JHA to mature feeding fifth-instar larvae (Truman el al.. 1974). We attribute the present findings to the inability of the brain to inactivate the denervated CA: the latter's continued secretion of JH could not be countered by the EPPAT-inactivated JHE. Hence, the secreted JH remained active and caused re- tention of larval features. The present experiments document inactivation of hornworm CA late in the feeding portion of the final in- star. Suppression of JH synthesis by the CA through the neural circuitry to those glands has been suggested in ear- lier work on the hornworm (Bhaskaran el al. 1980) as well as in other species (for review, see Feyereisen, 1 985 ). The anatomical basis for neural inhibition of the CA by the brain has also been documented: the hornworm CA demonstrate a complex innervation arising from the brain, including axons from multiple neurosecretory cells whose cell bodies lie in several regions of the brain (Nijhout, 1975; Carrow et al., 1984). Most of the docu- mented neurosecretory innervation of the CA from the brain is via the medial nerve (termed NCC I-II in the work of Nijhout, 1975) whose section in the present ex- periments coupled with EPPAT treatment often resulted in the production of larval-pupal intermediates. Implantation of active CA can alone effectively pre- vent the pupation of allatectomized fourth-instar larvae and starved immature fifth-instar larvae (Bhaskaran and Jones, 1980; Bhaskaran et al., 1980). The efficacy of im- planted CA in these cases contrasts with our inability to block the pupation of mature, feeding fifth-instar larvae merely by implantation of active CA. In the present ex- periments denervation of the CA was of morphogenetic consequence only when deployed concurrently with the JHE inhibitor EPPAT. This is presumably explained by the high levels of JHE activity found in feeding fifth-in- star larvae, levels manifestly sufficient to inactivate JH produced by denervated CA of every stage examined. These experiments highlight the important role of JHE in the elimination of circulating JH prior to meta- morphosis. Nevertheless, production of JHE does not seem to be required for clearance of JH sufficient to per- mit metamorphosis. For example, while treatment of fifth-instar larvae with EPPAT in dosages larger than 4(1(1 L. SAFRANEK AND C. M. WILLIAMS those deployed here can slightly delay the onset of meta- morphosis (Sparks el ai. 1983; pers. obs.), EPPAT ad- ministration alone does not prevent metamorphosis or lead to the production of larval-pupal intermediates. This failure presumably reflects the concurrent suppres- sion of JH production by the CA and the clearance of residual JH by alternative routes. Thus, the present experiments document that the dis- appearance of JH necessary for the onset of metamor- phosis normally occurs through two mechanisms: the in- activation of JH secretion by the CA and the production of JHE. Each mechanism can apparently alone clear sufficient JH to permit metamorphosis. Together they es- tablish a potent system for the timely and thorough elim- ination of JH during the final days of larval life. Acknowledgments This work was initially supported by the N.I.H., con- tinued with the generous support of Rohm and Haas, Inc., and completed with support from the N.S.F. Literature Cited Bhaskaran, G.. and G. Jones. 1980. Neuroendocrine regulation of corpus allatum activity in Manduca sexla- the endocrine basis for starvation-induced supernumerary larval moult. J. Insect Physiol. 26:431-440. Bhaskaran, G., G. Jones, and D. Jones. 1980. Neuroendocrine regu- lation of corpus allatum activity in Manduca sexta: sequential neu- rohormonal and nervous inhibition in the last-instar larva. Proc. Nail. Acad. Sci. USA 77: 4407-44 1 I . Bhaskaran, G. 1981. Regulation of corpus allatum activity in last in- star Manduca sexta larvae. Pp. 53-82 in Current Topics in Insect Endocrinology and Nutrition, G. Bhaskaran, S. Friedman, and J. G. Rodriguez, eds. Plenum Publishing, New York. Carrow, G. M., R. L. Calabrese, and C. M. Williams. 1984. Architecture and physiology of insect cerebral neurosecretory cells. J. \eurosci.4: 1034-1044. Feyereisen, R. 1985. Regulation of juvenile hormone liter: synthesis. Pp. 391-429 in Comprehensive Insect Physiology. Biochemistry, and Pharmacology. Vol. 7, G. A. Kerkut and L. I. Gilbert, eds. Per- gamon Press. New York. Hammock, B. D. 1985. Regulation of juvenile hormone liter: degra- dation. Pp. 431-472 in Comprehensive Insect Physiology. Biochem- istry, and Pharmacology. Vol. 7, G. A. Kerkul and L. I. Gilbert, eds. Pergamon Press, New York. Nijhout, H. F. 1975. Axonal palhways in Ihe brain-relrocerebral complex of Manduca sexta (L.) (Lepidoplera: Sphingidae). Int. J. Insect Morplml. Embryol. 4: 529-538. Safranek, L., and L. M. Riddiford. 1975. The biology of Ihe black larval mulanl of the tobacco hornworm, Manduca sex/a. J. Insect Physiol. 21: 1931-1938. Safranek, I.., and C. M. Williams. 1980. Sludies of Ihe prolhoracico- Iropic hormone in Ihe lobacco hornworm. Manduca sexla. Biol. Bull 158: 141-153. Safranek, I.., and C. M. Williams. 1987. Sludies of Ihe ecdysiotropic activily of juvenile hormone in pupae of Ihe lobacco hornworm. Manduca sexla. Biol. Bull. 172:299-306. Sparks, I. C, B. D. Hammock, and L. M. Riddiford. 1983. The haemolymph juvenile hormone eslerase of Manduca sexla (L.) — inhibilion and regulalion. Insect Btochem. 13: 529-541. Truman, J. T., L. M. Riddiford, and L. Safranek. 1973. Hormonal conlrol of cuticle coloration in the lobacco hornworm. Manduca sexta. basis of an ultrasensitive bioassay for juvenile hormone. J. Insect Physiol 19: 195-203. Truman, J. T., L. M. Riddiford, and L. Safranek. 1974. Temporal pallerns of response lo ecdysone and juvenile hormone in Ihe epi- dermis of the lobacco hornworm. Manduca sexta. Dev Biol 39: 247-262. Vince, R. k.. and L. I.Gilbert. 1977. Juvenile hormone eslerase acliv- ily in precisely limed lasl inslar larvae and pharale pupae of Man- duca sexta Insect Biochem. 1: 1 15-120. INDEX ABRAMSON, STEVEN, see Giulia Celli, 3 1 7 ABRM. 206 Accessory heart, 386 Acid phosphatase, 110 Actions ofFMRFamide-like peptides, 206 Activation of the octavolateralis efferent system in the lateral line of free-swimming toadfish, 326 Adipokinetic hormones: functions and structures, 2 1 8 Adrenocortical cells, 3 1 3 Aiptasia pallida, 130 AKH family. The. 2 18 Alcohol. 350 ALKON, DANIEL L., see James L. Olds, 325 ALK.ON, D. L.. C. COLLIN, 1. LEDERHENDLER, R. ETCHEBERIGARRAY, P. HUDDIE, M. SAK.AKIBARA, S. REDLICH, E. YAMOAH, A. PAPA- GEORGE, AND T. NELSON. Functional and structural conse- quences of activation of protein kinase C (PKC) and injection of G-protein substrates of PKC in Hermissenda neurons, 320 Amoeba cysts, 1 10 Amoebomastigote transformation, 1 10 Analysis of edge birefringence observed near refractive index steps in myofibrils and KC1 crystals using high resolution polarized light microscopy and spatial Fourier filtering. 318 Annelids, 363 Annual Report of the Marine Biological Laboratory, 1 Anomia simp/ex, 83 Anomiid, 83 Anoxia, 154 Antarctic, 77 Anti-fertility agent, 316, 317 Arachidonate and docosahexanoate as messengers in stimulus-re- sponse-coupling: evidence for effects of G-proteins in marine sponge aggregation, 3 1 7 Arhacia differentiation, 317 ARMSTRONG, PETER B.. see James P. Quigley. 316: and Frederick R. Rickles,319 Arterial perfusion of FMRFamide-related peptides potentiate trans- mission at the giant synapse of the squid, 322 Arthropod neuropeptides, 225 Aslerias rubens, 1 4 1 Asteroid larvae. 77 ATEMA, JELLE, GREG GERHARDT, PAUL MOORE, AND LAURENCE MADIN, Subnose 1 : tracking oceanic odor plumes with high spatio- temporal resolution, 328 ATEMA, JELLE. see Kurt Kotrschal. 328: Paul Moore, 329; Robert C. Peters, 329; Nat Scholz, 329; Rainer Voigt. 330; and Anna Weinstein. 330 ATP-models, 313 ALIGUSTINE, GEORGE J.. see Luis R. Osses, 146 Autotrophic carbon fixation by the chemoautotrophic symbionts of Rifiia pachyptila, 372 B Bacterial aggregates within the epidermis of the sea anemone Aiptasia pallida. 130 Bacterial films. 295 Bacterial symbionts. 130 Bag cells, 210 BANDIVDEKAR, A. H., S. J. SEGAL, AND S. S. KOIDE, Binding of 5- hydroxytryptamine to isolated plasma membranes of Spisula ga- metes, 314 BARLOW, ROBERT B., AND EHUD KAPLAN, What is the origin of pho- toreceptor noise? 32 1 BARRY, SUSAN, R., see Luis R. Osses, 146 Bathyal eehinoid sperm ullrastructure, 230 BELL. G. I., see D. F. Steiner, 1 72 BERG JR., CARL J., see Jeftry B. Milton, 356; and Ronald L. Garth- waite, 287 BERGLES, DWIGHT E., AND SIDNEY L. TAMM. Cell motility and cy- toskeleton, 3 1 3 BERLIN, ROBERT L.. see James Y. Bradfield, 344 BERNAL. JUAN, AND BARBARA EHRLICH, G-proteins modulate cal- cium currents in Paramvcium and Helix neurons, 32 1 Bilateral asymmetry in the shell morphology and micro-structure of early ontogenetic stages of Anomia simplex. 83 Binding of 5-hydroxytryptamine to isolated plasma membranes of Spisitla gametes. 314 Binding of gossypol and its analog to sperm proteins from Arhacia. Chaetopterus, and Spisula, 3 1 7 Biochemistry. 247 Biphasic modulation of calcium-dependent potassium current in pitu- itary tumor cells examined with the perforated patch clamp tech- nique. 323 Birefringence, 318 BISGROVE. BRENT W.. see Annette L. Parks, 96 BLANKENSHIP, JAMES E.. see Gregg T. Nagle, 210 Blood pressure, 328. 386 Bluefish, 328 Bombesin, 192 BORKOWSKI. ROSEMARIE, AND ROBERT A. BuLLis, Shell disease syn- drome in Cancer crabs, 327 BOSCH, ISIDRO, Contrasting modes of reproduction in two Antarctic asteroids of the genus Porania, with a description of unusual feed- ing and non-feeding larval types, 77 BOYER, BARBARA C., AND GWENDOLYN A. WALLACE, Early cleavage and the role of the macromeres in the development of the polyclad flatworm Hoploplana. 3 1 5 BOYER, BARBARA C., see Ilene M. Kaplan. 327 BOYER, BARBARA CONTA, The role of the first quartet micromeres in the development of the polyclad Hoploplana inquilina, 338 BRADFIELD, JAMES Y., ROBERT L. BERLIN. SUSAN M. RANKJN, AND LARRY L. KEELEY, Cloned cDNA and antibody for an ovanan cortical granule polypeptide of the shrimp Penaeus vannamei. 344 Bryozoan growth and reproduction, 277 Bryozoan larvae and bacterial films, 295 Buccal muscle, 206 BiiKtila. 277, 295 BULLIS, ROBERT A., see Rosemane Borkowski, 327 BULLIS, ROBERT A., Shell disease in impounded American lobsters, Homarus americanus. 327 Busvcon. 327 Ca-sensitivity, 313 Ca:* channel, 315 401 402 INDEX TO VOLUME 177 Calcification, 318 Calcification and proton transport in algae. 3 1 8 Calcium channel. 321.3:3 currents. 3:2 Calcium currents recorded in cells of anterior pituitary slices using the patch clamp technique. 322 Calcium-ATPase. 3 1 8 Calmidazolium. 314 Calsequestnn, 318 CAMERON, J. LANE, see Kevin J. Eckelbarger, 230 Cancer prevention. 3 1 7 (W/ic'iTspp.. 327 Cardioregulation, 386 CARTA, CARL A.. See Frederick R. Rickles. 3 1 9 Cell cycle, 316 Cell fusion induced hy a radio-frequency electric field, 314 Cell motility and cytoskeleton, 3 1 3 CELLI, GIULIA, JONATHAN MCMENAMIN-BALANO, STEVEN ABRAM- SON, KATHLEEN HAINES, JOANNA LESZCZINSKA, AND GERALD WEISSMANN, Arachidonate and docosahexanoate as messengers in stimulus-response-coupling: evidence for effects of G-proteins in marine sponge aggregation, 3 I 7 CELLI, GIULIA, see William Riesen, 319 Central control. 3:6 CHAN, S. J., see D. F. Steiner. 1 12 CHANG, DONALD C, AND QIANG ZHENG. Studies of membrane and cytoskeletal structures by electroporation using a radio-frequency electric field, 3 1 3 CHANG, DONALD C., see Qiang Zheng, 314 Chaos. 3:6 Chaotic properties of quanlal transmission at the skate neuro-electro- cytejunction, 326 Chemical orientation. 329 Chemo-orientation of the lobster, llonuirus amcncanus. to a point source in a laboratory flume, 329 Chemoautotrophy, 372 Chemoreception, 329, 330 CHERKSEY, B., R. L.LINAS, M. SUGIMORI, AND J.-W. LIN, Preliminary molecular structure of FTX and synthesis of analogs that block ICa in the squid giant synapse. 32 1 CHERKSEY, B., see R. Llinas, 324 CHILDRESS, JAMES J., see Charles R. Fisher, 372 Chili noclasia. 327 Chlorinated hydrocarbons, 327 Cholecystokinin, 195 Choline acetyltransferase, 322 Chromatin bodies, 1 10 Chromatophores. 225 Chromidia, 1 10 Cilia, 320 dona cilia, 313 Ciima inleslinalis, 317 Circulation, 328, 386 Cis-unsaturated fatty acid, 3 1 7 Cloned cDNA and antibody for an ovarian cortical granule polypeptide of the shrimp Pcnacux vannamci. 344 Cnidaria, 1 30 Coagulation. 319 COHEN, A. I., see D. Schiminovich, 325 COHEN, L. B., see D. Schiminovich, 325 COLLIN, C., see D. L. Alkon. 320 Color change. 225 Comparative aspects of gossypol action. 3 1 6 Conch. 327 Consequences of dispersal. 277 Consistency and variability in peptide families: introduction. 167 Contrasting modes of reproduction in two Antarctic asteroids of the genus Poriinui. with a description of unusual feeding and non- feeding larval types, 77 Copulatory behavior, 33 1 CORCORAN, GERAL YN, see Walter Troll, 3 1 7 Corpora allata, 396 Cortical granule, 344 COTTRELL, G. A., E. STANLEY, M. SUGIMORI. J.-W. LIN, AND R. LLI- NAS, Arterial perfusion of FMRFamide-related peptides potenti- ate transmission at the giant synapse of the squid, 322 Crab. 247. 327 Crasxo\lri'ti Ytrxtnii'ii, 154 Crustacea. 331. 344 Crustacean hormones, 225 Cryopreservation of spermatophores and seminal plasma of the edible crab Scy/la .wrruui. 247 Cryoprotectant, 247 Cushing response. 32S Cytoplasmic localization, 338 Cytoskeleton, 3 13. 314 D D'ALESSIO, GIUSEPPE, RENATA PICCOLI, AND NELLO Russo, Opioids in invertebrates, 3 1 7 DAPI, 1 10 Deep-sea reproduction, 230 Deletion of oncogenes, 3 1 7 DERIEMER, SUSAN A., MEYER B. JACKSON, AND ARTHUR KON- NERTH. Calcium currents recorded in cells of anterior pituitary slices using the patch clamp technique. 322 Desiccation resistance, 1 10 Development. 77. 96 Development of trophosome in Ridgeia sp.. 254 Digestive system, development in Ridgeia sp.. 254 DIMALINE, RODNEY, see Michael C. Thorndyke, 183 Direct development in the sea urchin Phyllacanthus /WV;.V/M/;H.V (Ci- daroidea): phylogenetic history and functional modification, 96 Disease, 327 Dispersal, 356 Dissoconch, 83 DNA repair, 320 DOCKRAY, G. J.. Gastnn, cholecystokinin (CCK), and the leukosulfa- kimns, 195 Does calsequestrin facilitate calcium diffusion along the endoplasmic reticulum of eggs? 3 1 8 Dogfish shark and sea robin, 319 Dose-response for FTX blockade of presynaptic I (Ca) in the squid gi- ant synapse, 324 DOWDALL, M.. G. PAPPAS, AND M. KRIEBEL, Properties of detached nerve terminals from skate electric organ: a combined biochemi- cal, morphological, and physiological study, 322 DREWES, C. D., AND C. R. FOURTNER, Hindsight and rapid escape in a freshwater oligochaete, 363 DuBois, AR i HUR B., see Stephen H. Fox, 328 Early cleavage and the role of the macromeres in the development of the polyclad flatworm Ilup/oplana. 315 Early development in Ridgeia, 254 EBBFRINK, R. H. M., A. B. SMIT, AND J. VAN MINNEN, The insulin family: evolution of structure and function in vertebrates and in- vertebrates, 176 Echinoderm reproduction, 230 Echinoidea, 230 ECKELBARGER, KEVIN J.. CRAIG M. YOUNG, AND J. LANE CAMERON. Modified sperm ultrastructure in four species of soft-bodied echi- noids (Echinodermata: Echinothuriidae) from the bathyal zone of the deep sea. 230 Ecology and life history of an amoebomastigote, Paratetrainini*, ingii- SHX. from a microbial mat: new evidence for multiple fission. 1 10 Effects ofhypoxia and anoxia on survival, energy metabolism and feed- ing of oyster larvae (Cravso.v/rtv/ virgimi'ti. Cimelin), 154 Effects of GABA on retinal horizontal cells: evidence for an electro- genie uptake mechanism. The, 324 INDEX TO VOLUME 177 403 Efferent control of sensory processes, 324 Egg-laying hormone family: precursors, products, and functions. The, 210 EHRLICH. BARBARA, see Juan Bernal, 32 1 Elasmobranch, 324 Electric organ. 322, 324 Electric organ discharge and electrosensory reafference in the little skate. 324 Electrofusion. 313, 314 Electrophysiology, 329 Electroporation, 313.314 Electroreception. 324 ELPHICK, MAURICE R.. ROLAND H. EMSON, AND MICHAEL C. THORNDYKE. FMRFamide-like immunoreactivity in the nervous system of the starfish Axlena.s ritbens. 1 4 1 Embryonic determination, 3 1 5 Embryonic symmetry, 314, 338 EMDIN. S., see D. F. Steiner, 1 72 EMSON, ROLAND H., see Maurice R. Elphick, 1 4 1 Encysting Clilorellti. 110 Endocrinology, 323 Endoplasmic reticulum, 318 Endosymbiotic bacteria of Ridgeia sp. and Riliia paehypiila, 254 Endotoxin, 319 Energetics. 237 Energy imbalance in nonfeeding larvae. 237 Energy metabolism. 154 ENZIEN. MICHAEL, HEATHER I. MCKHANN. AND LYNN MARGULIS. Ecology and life history of an amoebomastigote, Paraleiramnn.^ jugnsus. from a microbial mat: new evidence for multiple fission, 110 Epidermal chemoreception. 328 Escape reflex, 363 ETCHEBERIGARRAY, P., see D. L. Alkon, 320 Evolution. 176, 183 Evolution of peptide hormones of the islets of Langerhans and of mech- anisms of protcolytic processing, 1 72 Facilitated diffusion. 3 1 8 Factors controlling attachment of bryozoan larvae: a comparison of bacterial films and unnlmed surfaces, 295 FALK, C. X., see D. Schiminovich, 325 FALKMER, S., see D. F. Steiner, 172 Feeding. 154 Female sexual receptivity associated with molting and differences in copulatory behavior among the three male morphs in Paraeereeis xailpin (Crustacea: Isopoda), 331 FINK, RACHEL D., see Barbara E. Maclay, 316 FISHER, CHARLES R.. JAMES J. CHILDRESS. AND ELIZABETH MIN- NICH, Autotrophic carbon fixation by the chemoautotrophic sym- bionts ofRiftia pachyptila. 372 Fluorescent study of sensory neurons in normal and regenerating squid embryos, A 316 FMRFamide, 141,322 FMRFamide-like immunoreactivity in the nervous system of the star- fish Asterias ruhenx. 1 4 1 FMRFamide-related peptides, 206 FOURTNER, C. R., see C. D. Drewes. 363 Fox, STEPHEN H., CHRISTOPHER S. OGILVY, AND ARTHUR B. Du- Bois, Response of bluefish (Pomatomus saltatrix) to increased in- tracranial pressure (Cushing response), 328 FREADMAN, M. A., AND W. H. WATSON III. Gills as possible accessory circulatory pumps in Limulus polyphemus, 386 FTX. 323, 324 FTX blocks a calcium channel expressed by Xenopus oocyles after in- jection of rat brain mRNA, 323 FULLER, S. CYNTHIA, RICHARD A. LUTZ, AND YA-PING Hu, Bilateral asymmetry in the shell morphology and micro-structure of early ontogenetic stages on Anomia simplex, 83 Functional and structural consequences ot activation of protein kinase C (PKC) and injection of G-protein substrates of PK.C in Hermis- scnda neurons, 320 Functions, receptors, and mechanisms of the FMRFamide-related pep- tides. 206 Fura II, 325 G G-proteins modulate calcium currents in Paramccium and Helix neu- rons, 321 GABA, 324 GARDINER, STEPHEN L.. see Meredith L. Jones, 254 GARTHWAITE, RONALD L., CARL J. BI;RC; JR., AND JUNE HARRIGAN, Population genetics of the common squid Loligo pealei LeSueur, 1821, from Cape Cod to Cape Hatteras, 287 Gastrin. cholecystokinin (CCK), and the leukosulfakinins, 195 Gastrin-releasing peptide, 192 Gene flow. 356 Geographic variation, 356 GERHARDT, GREG, see Jelle Atema, 328 Germinal vesicle breakdown, 3 1 5 Giant axons of Ridgeia sp. and Riltui paehypiila, 254 Giant nerve fibers, 363 Giant-synapse, 322 Gill withdrawal reflex. 325 Gills as possible accessory circulatory pumps in Limulus polyphemus, 386 GLOGOWSKI, MARY ANN, see Edward E. Palincsar, 1 30 GNAIGER, ERICH, see William B. Stickle, 303 GOLDSWORTHY. GRAHAM. AND WILLIAM MORDUE, Adipokinetic hormones: functions and structures, 2 1 8 Gossypol. 316, 317 Gossypolone, 3 1 7 GREENBERG, MICHAEL J., AND MICHAEL C. THORNDYKE, Consis- tency and variability in peptide families: introduction, 167 GREENBERG, MICHAEL J., see David A. Price, 198 Growth and energy imbalance during the development of lecitho- trophic molluscan larva (/laliolis riifeseenx), 237 Growth variation, 277 H HADFIELD, MICHAEL G., see J. Timmothy Pennington. 350 HAINES, KATHLEEN, see Giulia Celli. 3 1 7 HARRIGAN, JUNE, see Ronald L. Garthwaite, 287 Heart muscle. 206 Heart-gill coordination. 386 HEFFLIN, BROCKTON, see George M. Langford, 313 Hemodynamics, 386 Hermissenda, 323 High resolution sampling, 328 Highspeed video, 325 HIGHSTEIN, STEPHEN M., see Rachel Locke, 324; and T. C. Tricas, 326 Hindsight and rapid escape in a freshwater oligochaete, 363 Hippocampus, 325 HOFFMANN, DANIELA, see Ilene M. Kaplan, 327 HOLZAPPLE, J., see J. Vautrin, 326 HOPP, H.-P.. see D. Schiminovich. 325 Horizontal cells, 324 Horseshoe crab, 386 Hu, YA-PING, see S. Cynthia Fuller, 83 HUDDIE, P., see D. L. Alkon, 320 Hunting of the FaRPs: the distribution of FMRFamide-related pep- tides. The. 198 Hydrothermal vents. 372 5-Hydroxytryptamine receptor types on SpisnUi gametes. 3 1 5 Hypoxia, 154 I Identification of myosin in dogfish shark and sea robin lens epithelium, 319 404 INDEX TO VOLUME 177 Image analysis, 318 Imaging learning-specific changes in the distribution of protein kinase C. 325 Immunocytochemistry. 141,323 Impermeant anions, 325 //; v/vo vectorial labeling of scallop gill ciliary membrane by NHS-LC- biotin, 320 INOUE, SHINVA. TED INOUE, ROBERT A. KNUDSON, AND RUDOLF OL- DENBOURG, Very high resolution and dynamic stereo images of neurons. 322 INOUE, TED, see Shinya Inoue, 322 INOUE, SHINYA, see Rudolf Oldenbourg, 318 Insect neuropeptides, 218 peptides, 176 Insulin, 172, 176 Insulin family: evolution of structure and function in vertebrates and invertebrates. The, 176 Interphase paniculate tubulin revisited, 316 Intracranial pressure. 328 Invertebrates, 317 Involvement of Ca:* channels in 5-hydroxytryptamine-induced oocyte maturation in Spi.fitla. 3 1 5 Ion channels, 323 Islet peptides, 172 Isopoda, 331 Isozymes, 111) JACKSON, MEYER B.. see Susan A. DeRiemer, 322 JAECKLE, WILLIAM B., AND DONAL T. MANAHAN, Growth and energy imbalance during the development of a lecithotrophic molluscan larva (Halinlis ru/i-sccns), 237 JAFFE, LIONEL F., Does calsequestnn facilitate calcium diffusion along theendoplasmic reticulum of eggs? 3 18 JEYALECTUMIE, C., AND T. SUBRAMONIAM, Cryopreservation of sper- matophores and seminal plasma of the edible crab Scvlla serrala, 247 JONES, MEREDITH L., AND STEPHEN L. GARDINER, On the early devel- opment of the vestimentiferan tube worm Ridgeia sp. and obser- vations on the nervous system and trophosome of Ridgeia sp. and Riftia pachyptila, 254 JONES. WARREN R., see Edward E. Pahncsar, 1 30 Juvenile hormone esterase, 396 K KADAM, A. L., P. A. KADAM, S. J. SEGAL. AND S. S. KOIDE. Involve- ment of Ca2+ channels in 5-hydroxytryptamine-induced oocyte maturation in Spisitla. 3 1 5 KADAM, A. L., see P. A. Kadam, 3 1 5 KADAM. P. A., A. L. KADAM, S. J. SEGAL, AND S. S. KOIDE, 5-hydroxy- tryptamine receptor types on Spixulii gametes. 3 1 5 KADAM, P. A., see A. L. Kadam, 3 1 5 KAPLAN, EHUD, see Robert B. Barlow, 32 1 KAPLAN, ILENE M., BARBARA C. BOYER, AND DANIELA HOFFMANN, Marketing, ecological, and policy considerations related to the New England conch fishery and Iloploplana. 327 KAPPER, MARTIN A., see William B. Stickle, 303 KEELEY, LARRY, L.,see James Y. Bradfield. 344 KEOUGH, MICHAEL J., Variation in growth rate and reproduction of the bryozoan Bugula ncritina, 277 Kinetid, 1 10 Kinetosome, I 10 KNUDSON, ROBERT A., see Shinya Inoue, 322 KOBAYASHI, MAKOTO, AND YOJIRO MLINEOKA. Functions, receptors, and mechanisms of the FMRFamide-related peptides, 206 KOIDE, S. S., see A. H. Bandivdekar, 314; A. L. Kadam, 315; and P. A. Kadam, 3 1 5 KONNERTH, ARTHUR, see Susan A. DeRiemer, 322 KOTRSCHAL, KURT, ROB PETERS. AND JELLE ATEMA, A novel chemo- sensory system in fish: do rocklings (CiV/oU inustcla. Gadidae) use their solitary chemoreceptor cells as fish detectors? 328 KOTRSCHAL, KURT, see Robert C. Peters, 329 KRAMER, RICHARD H.. AND EDWIN S. LEVITAN, Biphasic modulation of calcium-dependent potassium current in pituitary tumor cells examined with the perforated patch clamp technique. 323 KRAUTGARTNER, WOLF-DIETRICH, see Robert C. Peters, 329 KRIEBEL, M., see J. Vautrin. 326; and M. Dowdall. 322 KUZIRIAN. ALAN M., see Ebenezer Yamoah. 323 Labyrinth, 326 LACOMIS. LYNNE, see Nat Scholz, 329; and Paul Moore, 329 LANGFORD, GEORGE M., SANDRA A. MURRAY, BROCKTON HEFFLIN, AND KATHLEEN J. PENNY, Quantitative motion analysis of vesicle movement in Y-l adrenocortical cells and the use of fluorescent probes to identity the organelles, 3 1 3 Larvae. 154.350 Larvae of a nudibranch mollusc (Phi'stillu sitwgae) metamorphose when exposed to common organic solvents. 350 Larval development, 237 settlement, 295 Learning, 320. 325 LEDERHENDLER, I., see D. L. Alkon. 320 Lens epithelium, 3 19, 320 LESZCZINSKA, JOANNA, see Giulia Celli, 3 1 7 Leukosulfakinins. 195 Leusine aminopeptidase, 110 LEVITAN, EDWIN, S., see Richard H. Kramer, 323 Ligand-receptor interactions, 319 Light scattering, 325 Limit/its. 319,386 LIN, J.-W., B. RLIDY, AND R. LLINAS, FTX blocks a calcium channel expressed by Xcnupux oocytes after injection of rat brain mRNA, 323 LIN, J.-W., see B. Cherksey, 321; G. A. Cottrell, 322; and R. Llinas. 324 Liu. Ll-LiAN. see William B. Stickle, 303 LLINAS, R.. M. SUGIMORI, J.-W. LIN, AND B. CHERKSEY, Dose-re- sponse for FTX blockade of presynaptic I (Ca) in the squid giant synapse. 324 LLINAS R., see B. Cherksey, 321, G. A. Cottrell, 322; J.-W. Lin. 323; and M. Sugimon, 326 Lobster. 327, 329, 330 LOCKE, RACHEL. AND STEPHEN M. HIGHSTEIN, Modulation of the spontaneous and evoked responses oflagenar atferents in the toad- fish Opsanus tan. by electric pulse stimulation of the efferent ves- tihular nuclei. 324 Locusts. 2 1 8 Loligopealei, 287 LOWE, KRIS, see Nancy S. Rafferty. 319; and Seymour Zigman, 320 LLITZ, RICHARD A., see S. Cynthia Fuller, 83 .'itugniili.'i. 1 76 M MACLAY. BARBARA E., AND RACHEL D. FINK, A fluorescent study of sensory neurons in normal and regenerating squid embryos, 316 MADIN, LAURENCE, see Jelle Atema, 328 MAKI. J. S., D. RITTSCHOF, A. R. SCHMIDT. A. G. SNYDER, AND R. MITCHELL, Factors controlling attachment of bryozoan larvae: a comparison of bacterial films and unfilmed surfaces, 295 MALCHOW. ROBERT PAUL, The effects of GABA on retinal horizontal cells: evidence (bran electrogenic uptake mechanism. 324 Male polymorphism, 331 MANAHAN, DONAL T., see William B. Jaeckle, 237 Manduca se.Ma. 396 MANN, R., see J. Widdows, 1 54 INDEX TO VOLUME 177 405 MARGULIS, LYNN, see Michael Enzien. 1 10 Marine policy, 327 Marketing, ecological, and policy considerations related to the New En- gland conch fishery and Hophplana. 327 Mastigote. 1 10 MASTRO, JOSEPH L., see Edward E. Palincsar. 1 30 McCoNNAUGHEV, TED, Calcification and proton transport in algae. 318 MCDONALD, JOHN K... see Sharon L. Milgram, 318 MCKHANN. HEATHER I., See Michael Enzien, 1 10 MCMENAMIN-BALANO, see Giulia Celli, 317; and William Riesen, 3 1 9 McPHiE. DONNA L., see James L. Olds, 325 Meiosis. 316 Membrane 5-HT receptor, 314 Metabolic adaptations of several species of crustaceans and molluscs to hypoxia: tolerance and microcalorimetric studies, 303 Metalioproteinases of sea urchin embryo and sponge: detection by gela- tin-substrate polyacrylamide gel electrophoresis, 316 Metamorphosis. 350 Microbial habitats. 1 10 mats, 1 10 Microcoleus, 1 10 Microscopy. 322 MILGRAM, SHARON L.. JOHN K. MCDONALD, AND BRYAN D. NOE, Regulation of insulin release from pancreatic islet cells by norepi- nephnne and neuropeptide Y, 3 1 8 VAN MINNEN, J., see R. H. M. Ebberink, 1 76 MINNICH. ELIZABETH, see Charles R. Fisher, 372 MITCHELL, R.. see J. S. Maki, 295 Mithramycin, 1 10 MITTON. JEFFRY B., CARL J. BERG JR.. AND KATHERINE S. ORR. Pop- ulation structure, larval dispersal, and gene flow in the Queen conch. Slrombiis gigas. of the Caribbean, 356 Mixture suppression, 330 Modified sperm ultrastructure in four species of soft-bodied echinoids (Echinodermata: Echinothumdae) from the bathyal zone of the deep sea, 230 Modulation of the spontaneous and evoked responses of lagenar affer- ents in the toadfish Opsanus lau. by electric pulse stimulation of the efferent vestibular nuclei. 324 Molecular cloning, 344 Mollusc, 206, 350 Molluscan egg-laying hormones. 2 1 0 insulin-related peptide. 176 neuropeptides, 210 peptides, 176 Molting, 331 MOORE, PAUL. NAT SCHOLZ, LYNNE LACOMIS, AND JELLE ATEMA. Potential gradient information contained within the three-dimen- sional structure of a laboratory odor plume. 324 MOORE, PAUL, see Jelle Atema, 328: and Nat Scholz, 329 MORDLIE, WILLIAM, see Graham Goldsworthy, 218 Morphology, 320 Mosaic development, 338 Mucoid stimuli. 329 Multiple fission. 1 10 MUNEOKA, YOJIRO. see Makoto Kobayashi. 206 MURRAY, SANDRA A., see George M. Langford, 3 1 3 Muscle. 318 Mychonastea desiccatus, 1 10 Myosin. 319 NELSON, LEONARD, The sperm cell's silent spring: herbicides and pesti- cides. 327 NELSON, T., see D. L. Alkon. 320 Nerve terminals. 325 Neurohormones. 1 76 Neurokinin 1 receptor, 192 Neurokinm 2 receptor. 192 Neurokinin A, 192 Neurokinin B, 192 Neurons, 322 Neuropeptide, 141. 183 Neuropeptide V. 3 1 8 NEW, JOHN G., Electric organ discharge and electrosensory reafference in the little skate, 324 NEWELL. R. I. E.. see J. Widdows. 1 54 Nicotinamide suppresses Arhacia piincliiliila development. 3 1 7 NOE, BRYAN D., see Sharon L. Milgram, 3 1 8 Noise, 321 Norepinephrine. 3 1 8 Novel chemosensory system in fish: do rocklings (Ciliaia muslela. Gad- idae) use their solitary chemoreceptor cells as fish detectors? A. 328 Novel chemosensory system in fish: electrophysiological evidence for mucus detection by solitary chemoreceptor cells in rocklings, A. 329 Nucleolus, 316 o OBAID, A. L.. K. STALEY, J. B. SHAMMASH, AND B. M. SALZBERG, Stilbene derivatives or chloride replacement by impermeant anion dramatically alter a late component of the light scattering change in mammalian nerve terminals, 325 Odor plume. 328. 329 OGILVY. CHRISTOPHER S., see Stephen H. Fox, 328 OLDENBOURG. RUDOLF, AND SHINYA INOUE, Analysis of edge bire- fringence observed near refractive index steps in myofibrils and KC1 crystals using high resolution polarized light microscopy and spatial Fourier filtering. 3 1 8 OLDENBOURG, RUFOLF, see Shinya Inoue. 322 OLDS, JAMES L., DONNA L. McPmE, AND DANIEL L. ALKON, Imaging learning-specific changes in the distribution of protein kinase C. 325 On the early development of the vestimentiferan tube worm Ridgeia sp. and observations on the nervous system and trophosome of Ridgeia sp. and Riftia pachyptila. 254 On-line rapid determination of [Ca]i by means of Fura-II and high speed video imaging, 326 Ontogeny of serotonergic neurons in Hermissenda: a preliminary study. 323 Opioids in invertebrates, 3 1 7 Opsanus lau, 324 Optical recordings, 325 Organic solvents, 350 ORR. KATHERINE S.. see Jeffry B. Mitton, 356 Osmotic artifacts. 3 1 4 OSSES, Luis R., SUSAN R. BARRY, AND GEORGE J. AUGUSTINE. Pro- tein kinase C activators enhance transmission at the squid giant synapse, 146 Ovarian protein, 344 Oxygen. 372 Oyster, 154 N NAGLE. GREGG T.. SHERRY D. PAINTER, AND JAMES E. BLAN- KENSHIP, The egg-laying hormone family: precursors, products, and functions, 210 Near-U V effects on the thymidine incorporation into dogfish lens, 320 Neisseria, 3 1 7 Paddle cilia occur as artifacts in veliger larvae of Spisit/a solidissima and Lyrodus pedicellatus, 314 PAINTER, SHERRY D.. sec Gregg T. Nagle. 210 PALINCSAR, EDWARD E., WARREN R. JONES, JOAN S. PALINCSAR. MARY ANN GLOGOWSKI, AND JOSEPH L. MASTRO. Bacterial ag- gregates within the epidermis of the sea anemone Aiplasia pallida. 130 406 INDEX TO VOLUME 177 PALINCSAR, JOAN S., see Edward E. Palincsar. 1 30 Pancreatic islets. 318 PAPAGEORGE, A., see D. L. Alkon, 320 PAPPAS, G., see M. Dowdall, 322 Paracerceis, 331 Paramccium, 321 Paratctramilus, 1 10 PARKS, ANNETTE L., BRENT W. BISGROVE, GREGORY A. WRAY, AND RUDOLF A. RAFF. Direct development in the sea urchin Phvlla- canthus parvispinus (Cidaroidea): phylogenetic history and func- tional modification, 96 PENNINGTON, J. TIMOTHY, AND MICHAEL G. HADFIELD, Larvae of a nudibranch mollusc (Phestilla sibogae) metamorphose when ex- posed to common organic solvents, 350 PENNY, KATHLEEN J., see George M. Langford, 313 Peptide, 183 families, 167, 210. 225 structure, 218 VY, 187 Peptide YY: the ileo-colonic, gastric, and pancreatic inhibitor, 187 PETERS, ROB, see Kurt Kotrschal, 328 PETERS, ROBERT C.. KURT KOTRSCHAL. WOLF-DIETRICH KRAUT- GARTNER, AND JELLE ATEMA. A novel chemosensory system in fish: electrophysiological evidence for mucus detection by solitary chemoreceptor cells in rocklings, 329 Phosphatidylinositol, 323 Photoreceptor, 321, 363 Photosynthesis, 318 Phyllacanthus purvixpinux, 96 PICCOLI, RENATA, see Giuseppe D'Alessio, 3 1 7 Pigment dispersing hormone family: chemistry, structure-activity re- lations, and distribution. The, 225 Pituitary, 322 Polyclad, 315 Polyclad turbellaria, 338 Pomalomus xallalnx. 328 Population genetics of the common squid Loli.i;<> pculci LeSueur, 1821, from Cape Cod to Cape Hatteras. 287 Population structure, larval dispersal, and gene flow in the Queen conch, Strombus gigas, of the Caribbean, 356 Potential gradient information contained within the three-dimensional structure of a laboratory odor plume. 329 Potentiation, 322 Preliminary molecular structure of FTX and synthesis of analogs that block ICa in the squid giant synapse, 32 I Presynaptic blockers, 32 1 PRICE, DAVID A.. AND MICHAEL J. GREENBERG, The hunting of the FaRPs: the distribution of FMRFamide-related peptides, 198 Prodissoconch, 83 Prohormone processing. 1 72 Properties of detached nerve terminals from skate electric organ: a combined biochemical, morphological, and physiological study 322 Propionyl esterase, I 10 Protein I ofNcixxcriti. 317 Protein kinaseC. 146. 320, 325 Protein kinase C activators enhance transmission at the squid giant syn- apse, 146 Protein phosphorylation. 146 Purkmje cell, 325 Quanta, 322, 326 Quantitative motion analysis of vesicle movement in Y-l adrenocorti- cal cells and the use of fluorescent probes to identify the organelles 313 Queen conch, 356 QUIGLEY, JAMES P., AND PETER B. ARMSTRONG, Metalloproteinases of sea urchin embryo and sponge: detection by gelatin-substrate polyacrylamide gel electrophorcsis. 316 QUIGLEY, JAMES P., see Frederick R Rickles, 319 R Radial nerve cord, 1 4 1 RAFF, RUDOLF A., see Annette L. Parks, 96 RAFFERTY, KEEN A., see Nancy S. Rafferty, 3 1 9 RAFFERTY. NANCY S.. KRIS LOWE, KEEN A. RAFFERTY, AND SEY- MOUR ZIGMAN. Identification of myosin in dogfish shark and sea robin lens epithelium, 319 RAFFERTY, NANCY S.. see Seymour Zigrnan, 320 RANKIN. SUSAN M., see James Y. Bradfield, 344 RAO, K. RANGA, AND JOHN P. RIEHM, The pigment dispersing hor- mone family: chemistry, structure-activity relations, and distribu- tion. 225 Rat brain mRNA, 323 Receptor, 206 Receptor types, 3 1 5 REDLICH, S.. see D. L. Alkon, 320 Regeneration, 316 Regulation of insulin release from pancreatic islet cells by norepineph- rine and neuropeptide Y, 3 1 8 Reproduction, 96 Response of bluefish (Pomalonnm saltatrix) to increased intracranial pressure (Cushing response), 328 Responses of chemoreceptor cells to controlled temporal stimulus pat- terns, 330 Retina. 324 Rhodopsin. 321 RICKLES. FREDRICK R., PETER B. ARMSTRONG, CARL A. CART A, AND JAMES P. QUIGLEY, Spontaneous coagulation of Lnnulus amebocyte releasate in the absence of detectable endotoxin. 3 1 9 RIDDELL, JENNIFER H., see Michael C. Thorndyke, 183 Ridgeiasp., 254 RIEHM, JOHN P., see K. Ranga Rao, 225 RIESEN, WILLIAM. GIULIA CELLI. JONATHAN MCMENAMIN-BELANO, AND GERALD WEISSMANN, Secretion of Microciona prolifera ag- gregation factor (MAF ) is required for marine sponge aggregation: quantitative analysis by means of a new assay for MAF, 319 Ri/iiti pticliypiila. 254 RITTSCHOF, S. D., see J. S. Maki, 295 Role of the first quartet micromeres in the development of the polyclad Hoploplumi inquilina, The. 338 RUDY, B., see J.-W. Lin, 323 Russo, NELLO, see Giuseppe D'Alessio. 3 1 7 SAFRANEK, Louis, AND CARROLL M. WILLIAMS, Inactivation of the corpora allata in the final instar of the tobacco hornworm, Man- itiicu xc\tu, requires integrity of certain neural pathways from the brain, 396 SAKAKIBARA, M.. see D. L. Alkon, 320 SALZBERG. B. M., see A. L. Obaid, 325 Scallop. 320 SCHIMINOVICH, D., L. B. COHEN, A. I. COHEN, H.-P. HOPP, C. X. FALK, ANDj.-Y. Wu, A search for correlations in the spike activity of the ,-lplvxui abdominal ganglion during the gill withdrawal re- flex, 325 SCHMIDT, A. R., see J. S. Maki, 295 SCHOLZ, NAT. PAUL MOORE. LYNNE LACOMIS, AND JELLE ATEMA, Chemo-onentation of the lobster, fhimarux americanus. to a point source in a laboratory Hume, 329 SCHOLZ, NAT, see Paul Moore, 329 Scyllu .wrrata. 247 Sea anemone, 130 Sea urchin, 96 Sea urchin morphogenesis, 316 Search for correlations in the spike activity of the Ap/rxui abdominal ganglion during the gill withdrawal reflex. A, 325 Secretion, 146 Secretion of Microciona prolifera aggregation factor (MAF ) is required for marine spone aggregation: quantitative analysis by means of a new assay for MAF, 319 INDEX TO VOLUME 177 407 SEGAL, S. J., see A. H. Bandivdekar, 314; A. L. Kadam, 315; and H. Ueno, 3 1 7 SEGAL, SHELDON, AND HIROSHI UENO, Comparative aspects of gossy- pol action. 316 Seminal plasma. 247 Sensory adaptation, 330 Sensory neurons, 3 1 6 Seratonin, 314, 315, 323 Serotonin-sensitive K+-channels, 206 Shadow response. 363 SHAMMASH, J. B.. see A. L. Obaid, 325 Shell disease in impounded American lobsters, llomarus amcricanus, 327 Shell disease syndrome in Cancer crabs, 327 Shell microstructure, 83 morphology, 83 SHORT. GRAHAM. AND SIDNEY TAMM, Paddle cilia occur as artifacts in veliger larvae of Spisula solidissima and Lvruditx pcdici'llatus, 314 Shrimp, 344 SHLISTER. STEPHEN M.. Female sexual receptivity associated with molting and differences in copulatory behavior among the three male morphs in Paracerceis sculpta (Crustacea: Isopoda). 33 1 SMEEKENS, S. P., see D. F. Steiner, 172 SMIT, A. B., see R. H. M. Ebberink. 1 76 SNYDER. A. G., see J. S. Maki, 295 Spatial gradients, 329 Spectral tuning to amino acids and mixture effects on antennular che- moreceptor cells in the lobster, Homarm amcricunit\. 330 Sperm cells silent spring: herbicides and pesticides. The, 327 Sperm molility. 315, 327 ultrastructure, 230 Spermatophore, 247 Spider toxin, 321 Spike correlations, 325 Spisula gametes, 314 oocytes. 3 1 5 Sponge cell reaggregation. 316 Spontaneous coagulation of Limulus amebocyte releasate in the ab- sence of detectable endotoxin, 319 Squid, 287 embryo, 3 1 6 giant synapse, 324 STALEY, K., see A. L. Obaid, 325 STANLEY, E., see G. A. Cottrell, 322 Starfish. 141 STEINER, D. F., S. J. CHAN, S. P. SMEEKENS, G. I. CELL, S. EMDIN, AND S. FALKMER, Evolution of peptide hormones of the islets of Langerhans and of mechanisms of proteolytic processing, 1 72 STEPHENS, R. E.. see L. Warren, 320 Stereoscopy, 322 STICKLE, WILLIAM B., MARTIN A. KAPPER. LI-LIAN Liu, ERICH GNAIGER, AND SHIAO Y. WANG. Metabolic adaptations of several species of crustaceans and molluscs to hypoxia: tolerance and mi- crocalorimetric studies. 303 Stilbene derivatives or chloride replacement by impermeant anion dra- matically alter a late component of the light scattering change in mammalian nerve terminals, 325 Stimulus-response coupling, 319 Stock structure, 287 Stratified microhial community, 1 10 Structure and behavior, 328 Studies of membrane and cytoskeletal structures by electroporation us- ing a radio-frequency electric field. 3 1 3 Submersible, 328 Subnose 1 : tracking oceanic odor plumes with high spatiotemporal res- olution, 328 SUBRAMONIAM, T, see C. Jeyalectumie. 247 Substance P. 192 SUGIMORI, M.. AND R. LL.INAS, On-line rapid determination of [Caji by means of Fura-II and high speed video imaging, 326 SUGIMORI, M., see B. Cherkscy, 32 1 ; G. A. Cottrell. 322; R. Llinas, 324 Sulfide. 372 SUPRENANT, KATHY A., Interphase paniculate tubulin revisited, 316 Survival, 154 Symbiosis, 130,372 Synapse, 326 Synaptic transmission, 146 Tachykinins and the bombesin-related peptides: receptors and func- tions, 192 TAMM, SIDNEY L., see Dwight E. Bergles. 313; and Graham Short, 3 1 4 TAMSE, CATHERINE, see Ebenezer Yamoah, 323 Taste systems, 328 TAYLOR, IAN L., Peptide YY: the ileo-colonic, gastric, and pancreatic inhibitor, 187 Teleosts, 328 Thiocapsa. 1 10 THORNDYKE. MICHAEL C., JENNIFER, H. RIDDELL, DAVID T. THWAITES, AND RODNEY DIMALINE, Vasoactive intestinal poly- peptide and its relatives: biochemistry, distribution, and functions, 183 THORNDYKE, MICH AELC, see Maurice R. Elphick, 14 hand Michael J. Greenberg, 167 THWAITES, DAVID T., see M. C. Thorndyke, 1 83 Toadfish, 324 Tobacco hornworm, 396 TRICAS, T. C., AND S. M. HIGHSTEIN, Activation of the octavolateralis efferent system in the lateral line of free-swimming toadnsh, 326 Trochophore larva ofRidgcia sp., 254 TROLL, WALTER, AND GERALYN CORCORAN, Nicotinamide sup- presses Arhacin puncliilala development, 3 1 7 Trophosome of Ridgi-ia sp. and Rift in paclivptila, 254 Tubulin, 320 u UENO, H.. AND S. J. SEGAL, Binding of gossypol and its analog to sperm proteins from Arhacia. Chaetopterus, and Spisula. 317 UENO, HIROSHI. see Sheldon Segal. 3 1 6 Undulipodia. 1 10 Uptake, 45Ca:*. 315 UV radiation, 320 Variation in growth rate and reproduction of the bryozoan Bugula nen- tina. Ill Vasoactive intestinal polypeptide and its relatives: biochemistry, distri- bution, and functions. 183 VAUTRIN. J., J. HOLZAPPI.E, AND M. KRIEBEL, Chaotic properties of quantal transmission at the skate neuro-electrocyte junction, 326 Veliger larvae, 314 Very high resolution and dynamic stereo images of neurons, 322 Vesicle movement, 3 1 3 Vestibular, 324, 326 Vestimentifera. 254, 372 Video microscopy, 3 1 3 VIGNA, STEVEN R., Tachykinins and the bombesin-related peptides: receptors and functions, 192 VIP and related peptides, 1 83 VOIGT, RAINER, ANDJELLE ATEMA, Responses of chemoreceptor cells to controlled temporal stimulus patterns, 330 VOIGT, RAINER, see Anna Wemstein, 330 W WALLACE, GWENDOLYN A., see Barbara C. Boyer, 3 1 5 WANG, SHIAO Y,, see William B. Stickle, 303 408 INDEX TO VOLUME 177 WARREN, L.. AND R. E. STEPHENS, //; vivo vectorial labeling of scallop gill ciliary membranes by NHS-LC-biotin, 320 WATSON III. W. H., see M. A. Freadman, 386 WEINSTEIN, ANNA, RAINER VOIGT, AND JELLE ATEMA, Spectral tun- ing to ammo acids and mixture effects on antennular chemorecep- tor cells in the lobster, llnnuinn i/iin'iicuiin.\. 330 WEISSMANN. GERALD, see Giulia Celli, 317; and William Riesen, 3 1 9 What is the origin of photoreceptor noise? 32 1 WIDDOWS, J., R. I. E. NEWELL, AND R. MANN, Effects of hypoxia and anoxia on survival, energy metabolism and feeding of oyster larvea (Crassoslrea virginifa. Gmelin), 1 54 WILLIAMS. CARROLL M., see Louis Safranek. 396 WRAY, GREGORY, A., see Annette L. Parks, 96 Wu, J.-Y., see D. Schiminovich, 325 YAMOAH, E., see D. L. Alkon, 320 YAMOAH, EBENEZER, ALAN M. KUZIRIAN, AND CATHERINE TAMSE, Ontogeny of serotonergic neurons in Hermissenda: a preliminary study. 323 YOUNG. CRAIG M., see Kevin J. Eckelbarger. 230 ZHENG. QIANG, AND DONALD C. CHANG, Cell fusion induced by a radio-frequency electric field, 314 ZHENG, QIANG, see Donald C. Chang. 3 1 3 ZIGMAN, SEYMOUR. KRIS LOWE, AND NANCY S. RAFFERTY, Near- Li V effects on the thymidme incorporation into dogfish lens, 320 ZIGMAN, SEYMOUR, see Nancy S. Rafferty, 319 5 I 5 ;• CONTENTS BEHAVIOR Shuster, Stephen M. Female sexual receptivity associated with molting and differences in copulatory behavior among the three male morphs in Paracemis sculpta (Crustacea: Isopoda) 331 DEVELOPMENT AND REPRODUCTION Boyer, Barbara Conta The role of the first quartet micromeres in the de- velopment of the polyclad Hoploplana mquilma .... Bradfield, James Y., Robert L. Berlin, Susan M. Rankin, and Larry L. Keeley Cloned cDN A and antibody for an ovarian cortical granule polypeptide of the shrimp Penaeus vanna- 338 344 Pennington, J. Timothy, and Michael G. Hadfield Larvae of a nudibranch mollusc (Phestilla sibogae) metamorphose when exposed to common organic solvents . 350 ECOLOGY AND EVOLUTION Mitton, Jeffry B., Carl J. Berg Jr., and Katherine S. Orr Population structure, larval dispersal, and gene flow in the queen conch, Strombus gigas, of the Car- ibbean 356 PHYSIOLOGY Drewes, C. D., and C. R. Fourtner Hindsight and rapid escape in a freshwater oligo- chaete 363 Fisher, Charles R., James J. Childress, and Elizabeth Minnich Autotrophic carbon fixation by the chemoauto- trophic symbionts of Riftia pachyptila 372 Freadman, M. A., and W. H. Watson III Gills as possible accessory circulatory pumps in Lim- ulus polyphemus 386 Safranek, Louis, and Carroll M. Williams Inactivation of the corpora allata in the final instar of the tobacco hornworm, Manduca sexta, requires integrity of certain neural pathways from the brain 396 Index to Volume 177 . 401 MBI. WHOI UBRARV