THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board JOHN M. ANDERSON, Cornell University F. H. RUDDLE, Yale University JOHN O. CORLISS, University of Maryland BERTA SCHARRER, Albert Einstein College ^^ of Medicine DONALD P. COSTELLO, University of North Carolina HOWARD A. SCHNEIDERMAN, University'of California, Irvine PHILIP B. DUNHAM, Syracuse University MELym SpIEGEL> Dartmouth College CATHERINE HENLEY, University of GROVER C. STEPHENS, University of North Carolina California, Irvine GEORGE O. MACKIE, University of Victoria EDWARD O. WILSON, Harvard University W. D. RUSSELL-HUNTER, Syracuse University Managing Editor VOLUME 147 JULY TO DECEMBER, 1974 Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA. THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Penn- sylvania. Subscriptions and similar matter should be addressed to The Biological Bulletin, Marine Biological Laboratory, Woods Hole, Massachusetts. Agent for Great Britain: Wheldon and Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2. Single numbers, $5.00. Subscription per volume (three issues), $14.00. Communications relative to manuscripts should be sent to Dr. W. D. Russell-Hunter, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between May 23 and September 1, and to Dr. W. D. Russell-Hunter, P.O. Box 103, University Station, Syracuse, New York 13210, during the remainder of the year. Second-class postage paid at Lancaster, Pa. LANCASTER PRESS, INC., LANCASTER, PA. CONTENTS tl RY No. 1, AUGUST, 1974 Annual Report of the Marine Biological Laboratory ...................... 1 CALOW, P. AND J. B. JENNINGS Calorific values in the phylum Platyhelminthes : the relationship between potential energy, mode of life and the evolution of entoparasitism ....... 81 HEFFERNAN, J. MICHAEL AND STEPHEN A. WAINWRIGHT Locomotion of the holothurian Euapta lappa and redefinition of peristalsis 95 PARDY, ROSEVELT L. Some factors affecting the growth and distribution of the algal endo- symbionts of Hydra viridis ...................................... 105 POSTLETHWAIT, JOHN H. Juvenile hormone and the adult development of Drosophila ............ 119 PRICE, C. A., L. R. MENDIOLA-MORGENTHALER, M. GOLDSTEIN, E. N. BREDEN, AND R. R. L. GUILLARD Harvest of planktonic marine algae by centrifugation into gradients of silica in the CF-6 continuous-flow zonal rotor ....................... 136 READ, CLARK P., GEORGE L. STEWART, AND PETER W. PAPPAS Glucose and sodium fluxes across the brush border of Hymcnolepis ditninuta (Cestoda) ............................................. 146 SlVASUBRAMANIAN, P., S. FRIEDMAN AND G. FRAENKEL Nature and role of proteinaceous hormonal factors acting during pu- parium formation in flies ......................................... 163 SMITH, SCOTT, JAMES OSHIDA AND HANS BODE Inhibition of nematocyst discharge in hydra fed to repletion ............ 186 STOFFEL, Lois A. AND JERRY H. HUBSCHMAN Limb loss and the molt cycle in the freshwater shrimp, Palaemonetes kadiakensis .................................................... 203 SUGIMOTO, KEIJI AND MlTSUAKI NAKAUCITI Budding, sexual reproduction, and degeneration in the colonial ascidian, Symplegma rcptans ............................................. 213 WHITLATCH, ROBERT B. Food-resource partitioning in the deposit feeding polychaete Pccthiaria goiildii ........................................................ 227 YOUNG, L. G. AND L. NELSON The effects of heavy metal ions on the motility of sea urchin spermatozoa 236 REEVE, M. R. AND B. LESTER The process of egg-laying in the chaetognath Sagitta hispid a .......... 247 No. 2, OCTOBER, 1974 ALDRICH, JOHN CARLSON Allometric studies on energy relationships in the spider crab Libinia cmarg'mata (Leach) ............................................ 257 BURKETT, BARBARA N. AND HOWARD A. SCHNEIDERMAN Roles of oxygen and carbon dioxide in the control of spiracular function in Cecropia pupae ............................................... 274 BURKETT, BARBARA N. AND HOWARD A. SCHNEIDERMAN Discontinuous respiration in insects at low temperatures : intratracheal pressure changes and spiracular valve behavior ...................... 294 iii iv CONTENTS CHENG, THOMAS C. AND GARY E. RODRICK Identification and characterization of lysozyme from the hemolymph of the soft-shelled clam, Mya arenaria 311 CULLINEY, JOHN L. Larval development of the giant scallop Placopcctcn magellanicits (Gmelin) 321 FELL, PAUL E. Diapause in the gemmules of the marine sponge, Haliclona loosanofji, with a note on the gemmules of Haliclona ocnlata 333 GIBSON, RAY Histochemical observations on the localization of some enzymes asso- ciated with digestion in four species of Brazilian nemerteans 352 HILL, ROBERT B. AND JOSEPH W. SANGER Anatomy of the innervation and neuromuscular junctions of the radular protractor muscle of the whelk, Busy con canaliculatum (L.) 36*5 LOWRY, LLOYD F., ALFRED J. MCELROY AND JOHN S. PEARSE The distribution of six species of gastropod molluscs in a California kelp forest 386 MORIN, JAMES G. AND GEORGE T. REYNOLDS The cellular origin of bioluminescence in the colonial hydroid Obclia .... 397 MUKAI, HIDEO AND HIROSHI WATANABE On the occurrence of colony specificity in some compound ascidians .... 411 RICHARDSON, NANCY E. AND JAMES D. McCLEAVE Locomotor activity rhythms of juvenile Atlantic salmon (Salino salar) in various light conditions 422 SPIELMAN, ANDREW AND JOANN WONG Dietary factors stimulating oogenesis in Acdcs aegypti 433 VOGEL, STEVEN Current-induced flow through the sponge, Halichondria 443 YINGST, DOUGLAS The vertical distribution and reproductive biology of Pelogobia longi- cirrata (Annelida) in the central Arctic Ocean 457 Abstracts of papers presented at the Marine Biological Laboratory 466 No. 3, DECEMBER, 1974 ARNOLD, JOHN M. AND RICHARD E. YOUNG Ultrastructure of a cephalopod photophore. I. Structure of the photo- genic tissue 507 ARNOLD, JOHN M., RICHARD E. YOUNG AND MAURICE V. KING Ultrastructure of a cephalopod photophore. II. Iridophores as reflectors and transmitters 522 BOYER, JOHN F. Clinal and size-dependent variation at the LAP locus in JMytilus ednlis 535 CAINE, EDSEL A. Feeding of Ovalipes giiadnlpensis (Saussure) (Decapoda: Brachyura: Portunidae), and morphological adaptations to a burrowing existence . . . 550 DIETZ, THOMAS H. Body fluid composition and aerial oxygen consumption in the freshwater mussel, Ligumia subrostrata (Say) : effects of dehydration and anoxic stress 560 DONALDSON, SVEN Larval settlement of a symbiotic hydroid : specificity and nematocyst re- sponses in planulae of Proboscidactyla flavicirrata 573 CONTENTS v ENESCO, H. ESPER AND K. H. MAN Cytoplasmic DNA in sea urchin oogenesis studied by 3H-actinomycin D binding and radioautography 586 JOSEPH SON, ROBERT K. Factors affecting muscle activation in the hydroid Tubularia 594 KUSTIN, KENNETH, KAYE V. LADD, GUY C. McLEOD AND DAVID L. TOPPEN Water transport rates of the tunicate dona intestinalis 608 NEWTON, W. DONALD The accessory cell and yolk halo of the oocyte of the freshwater turbel- larian Hydrolvmax grisea (Platyhelminthes; Plagiostomidae) 618 PEARSE, VICKI BUCHSBAUM Modification of sea anemone behavior by symbiotic zooxanthellae : phototaxis 630 PEARSE, VICKI BUCHSBAUM Modification of sea anemone behavior by symbiotic zooxanthellae : expan- sion and contraction 641 VALIELA, IVAN, DANIEL F. BABIEC, "WILLIAM ATHERTON, SYBIL SEITZINGER AND CHARLES KREBS Some consequences of sexual dimorphism: feeding in male and female fiddler crabs, Uca pugnax (Smith) 652 WALKER, CHARLES WTAYNE Studies on the reproductive systems of sea-stars. I. The morphology and histology of the gonad of Asterias vulgaris 661 WALKER, GRAHAM The occurrence, distribution and attachment of the pedunculate barnacle Octolasmis mitlleri (Coker) on the gills of crabs, particularly the blue crab, Callinectes sapidus Rathbun 678 , Volume 147 Number 1 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board JOHN M. ANDERSON, Cornell University F. H. RUDDLE, Yale University JOHN O. CORLISS, University of Maryland BERTA SCHARRER, Albert Einstein College of Medicine DONALD P. COSTELLO, University of , North Carolina HoWARD A" SCHNEIDERMAN, Umvers^of ^^ PHILIP B. DUNHAM, Syracuse University MELVIN SPIEGEL, Dartmouth College CATHERINE HENLEY, University of GROVER C. STEPHENS, University of North Carolina California, Irvine GEORGE O. MACKIE, University of Victoria EDWARD O. WILSON, Harvard University W. D. RUSSELL-HUNTER, Syracuse University Managing Editor AUGUST, 1974 9 Printed and Issued by LANCASTER PRESS, Inc. PRINCE a LEMON STS. LANCASTER, PA. THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Pennsylvania. Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, Woods Hole, Massachusetts. Agent for Great Britain: Wheldon and Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2. Single numbers, $7.00. Subscription per volume (three issues), $18.00, (this is $36.00 per year for six issues). Communications relative to manuscripts should be sent to Dr. W. D. Russell-Hunter, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between May 23 and September 1, and to Dr. W. D. Russell-Hunter, P.O. Box 103, University Station, Syracuse, New York 13210, during the remainder of the year. Copyright © 1974, by the Marine Biological Laboratory Second-class-postage paid at Lancaster, Pa. INSTRUCTIONS TO AUTHORS THE BIOLOGICAL BULLETIN accepts original research reports of intermediate length on a variety of subjects of biological interest. In general, these papers are either of particular interest to workers at the Marine Biological Laboratory, or of outstanding general significance to a large number of biologists throughout the world. Normally, review papers (except those written at the specific invitation of the Editorial Board), very short papers (less than five printed pages), preliminary notes, and papers which describe only a new technique or method without presenting substantial quantities of data resulting from the use of the new method cannot be accepted for publication. A paper will usually appear within four months of the date of its acceptance. The Editorial Board requests that manuscripts conform to the requirements set below; those manuscripts which do not conform will be returned to authors for correction before review by the Board. 1. Manuscripts, Manuscripts must be typed in double spacing (including figure legends, foot-notes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 8$ by 11 inches. They should be carefully proof-read before being submitted and all typographical errors corrected legibly in black ink. Pages should be numbered. A left-hand margin of at least 1$ inches should be allowed. 2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and placed in correct sequence in the text. Because of the high cost of setting such material in type, authors are earnestly requested to limit tabular material as much as possible. Similarly, foot- notes to tables should be avoided wherever possible. If they are essential, they should be indi- cated by asterisks, daggers, etc., rather than by numbers. Foot-notes are not normally permitted in the body of the text. Such material should be incorporated into the text where appropriate. Explanations of figures should be typed double-spaced and placed on separate sheets at the end of the paper. 3. A condensed title or running head of no more than 35 letters and spaces should be included. Continued on Cover Three Vol. 147, No. 1 I BIOLOGICAL BULLETIN : PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY &./ THE MARINE BIOLOGICAL LABORATORY SEVENTY-SIXTH REPORT, FOR THE YEAR 1973 — EIGHTY-SIXTH YEAR I. TRUSTEES AND EXECUTIVE COMMITTEE (AS OF AUGUST, 1973) 1 II. ACT OF INCORPORATION 5 III. BYLAWS OF THE CORPORATION 5 IV. REPORT OF THE DIRECTOR 7 Addenda : 1. Memorials 10 2. The Staff 11 3. Investigators, Fellowships, and Students 22 4. Fellows and Scholarships 35 5. Training Programs 35 6. Tabular View of Attendance, 1969-1973 39 7. Institutions Represented 39 8. Friday Evening Lectures 41 9. Members of the Corporation 42 V. REPORT OF THE LIBRARIAN 72 VI. REPORT OF THE TREASURER. 73 I. TRUSTEES Including Action of 1973 Annual Meeting DENIS M. ROBINSON, Chairman of the Board of Trustees, High Voltage Engineering Cor- poration, Burlington, Massachusetts 01803 GERARD SWOPE, JR., Honorary Chairman of the Board of Trustees, Croton-on-Hudson, New York, New York 10520 ALEXANDER T. DAIGNAULT, Treasurer, 7 Hanover Square, New York, New York 10005 JAMES D. EBERT, Director and President of the Corporation, Director, Department of Embryology, Carnegie Institution DAVID SHEPRO, Clerk of the Corporation, Boston University, Boston EMERITI WILLIAM R. AMBERSON, Falmouth, Massachusetts PHILLIP B. ARMSTRONG, State University of New York, College of Medicine, Syracuse ERIC G. BALL, Marine Biological Laboratory DETLEV W. BRONK, The Rockefeller University 1 Copyright © 1974, by the Marine Biological Laboratory Library"of Congress Card No. A38-518 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY KENNETH S. COLE, National Institutes of Health PAUL S. GALTSOFF, Woods Hole, Massachusetts HARRY GRUNDFEST, College of Physicians and Surgeons RUDOLF T. KEMPTON, Vassar College DOUGLAS MARSLAND, Marine Biological Laboratory CHARLES W. METZ, Woods Hole, Massachusetts HAROLD H. PLOUGH, Amherst, Massachusetts A. C. REDFIELD, Woods Hole, Massachusetts CARL C. SPEIDEL, University of Virginia H. BURR STEINBACH, Woods Hole, Massachusetts ALBERT SZENT-GYORGYI, Marine Biological Laboratory W. RANDOLPH TAYLOR, University of Michigan CLASS OF 1977 JOEL E. BROWN, Vanderbilt University JOHN B. BUCK, National Institutes of Health JAMES CASE, University of California, Santa Barbara DONALD P. COSTELLO, University of North Carolina WILLIAM T. GOLDEN, New York, New York SHINYA INOUE, University of Pennsylvania STEPHEN W. KUFFLER, Harvard Medical School MALCOLM S. STEINBERG, Princeton University RAYMOND E. STEPHENS, Brandeis University and the Marine Biological Laboratory CLASS OF 1976 EVERETT ANDERSON, Harvard Medical School GEORGE H. A. CLOWES, Jr., Harvard Medical School PHILIP B. DUNHAM, Syracuse University TIMOTHY H. GOLDSMITH, Yale University ROBERT K. JOSEPHSON, University of California, Irvine C. LADD PROSSER, University of Illinois LIONEL I. REBHUN, University of Virginia ANDREW SZENT-GYORGYI, Brandeis University EDWARD O. WILSON, Harvard University CLASS OF 1975 WILLIAM J. ADELMAN, National Institutes of Health FRANCIS D. CARLSON, The Johns Hopkins University LAURA H. COLWIN, Queens College SEARS CROWELL, Indiana University CATHERINE HENLEY, University of North Carolina SAMUEL LENHER, Wilmington, Delaware JOHN W. MOORE, Duke University Medical Center W. D. RUSSELL-HUNTER, Syracuse University WALTER S. VINCENT, University of Delaware CLASS OF 1974 ROBERT D. ALLEN, State University of New York at Albany MICHAEL V. L. BENNETT, Albert Einstein College of Medicine TRUSTEES JOHX E. DOWLIXG, Harvard University HARLYN O. HALVORSON, Brandeis University J. WOODLAND HASTINGS, Harvard University RUTH HUBBARD, Harvard University JAMES W. LASH, University of Pennsylvania RICHARD S. MORSE, Wellesley, Massachusetts STANDING COMMITTEES EXECUTIVE COMMITTEE OF THE BOARD OF TRUSTEES DENNIS M. ROBINSON, ex officio WALTER S. VINCENT, 1975 ALEXANDER T. DAIGNAULT, ex officio PHILIP B. DUNHAM, 1975 JAMES D. EBERT, ex officio HARLYN O. HALVORSON, 1976 J. WOODLAND HASTINGS, 1974 JAMES F. CASE, 1976 JAMES W. LASH, 1974 LIHRAKY COMMITTEE CATHERINE HENLEY, Chairman BRUCE WARREN GARLAND E. ALLEN RUBERT ANDERSON FRED GRASSLE THOMAS J. M. SCHOPF IVAN VALIELA JAMES L. GERMAN III RESEARCH SERVICES COMMITTEE DONALD T. FRAZIER, Chairman ROBERT V. RICE W. J. ADELMAN ANDREW SZENT-GYORGYI ANTHONY LIUZZI DAVID YPHANTIS ROBERT B. BARLOW FRANCES BOWLES SUPPLY DEPARTMENT COMMITTEE MILTON FINGERMAN, Chairman CLIFFORD V. HARDING LAWRENCE B. COHEN RALPH HINEGARDNER SEARS CROWELL W. D. RUSSELL-HUNTER NIGEL DAW MELVIN SPIEGEL EARL WEIDNER INSTRUCTION COMMITTEE MALCOLM S. STEINBERG, Chairman BENJAMIN KAMINER F. D. CARLSON E. O. WILSON JOEL BROWN J. R. WHITTAKER RALPH QUATRANO FREDERICK LANG BUILDINGS AND GROUNDS COMMITTEE FRANK A. LOEWUS, Chairman GERTRUDE HINSCH EVERETT ANDERSON LEONARD NELSON D. EUGENE COPELAND LEON P. WEISS JAMES GREEN CHARLES WYTTENBACH CLELMER BARTELL ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY RADIATION COMMITTEE GEORGE T. REYNOLDS, Chairman JAMES MORIN J. R. COLLIER R. C. RUSTAD DANIEL S. GROSCH ANTHONY LIUZZI LASZLO LORAND RAYMOND E. STEPHENS RESEARCH SPACE COMMITTEE T. H. GOLDSMITH, Chairman H. JANNASCH J. W. HASTINGS RUTH HUBBARD J. F. CASE MICHAEL V. L. BENNETT COMMITTEE FOR THE NOMINATION OF OFFICERS WALTER S. VINCENT JAMES F. CASE HARLYN O. HALVORSON JOHN E. DOWLING JAMES W. LASH J. WOODLAND HASTINGS FOOD SERVICE COMMITTEE JOHN ARNOLD, Chairman NOEL DETERRA FR. J. D. CASSIDY A. FARMANFARMAIAN S. J. COOPERSTEIN RlTA GUTTMAN COMPUTER SERVICE COMMITTEE JOHN W. MOORE, Chairman MELVIN ROSENFELD, JR. ARNOLD LAZAROW NORMAN B. RUSHFORTH C. LEVINTHAL EDWARD F. MACNICHOL, JR. FRED DODGE FINANCE COMMITTEE ALEXANDER T. DAIGXAULT, Chairman JOHN W. MOORE SAMUEL LENHER ROBERT D. ALLEN WILLIAM T. GOLDEN SAFETY COMMITTEE ROBERT GUNNING, Chairman MANUEL P. DUTRA DONALD LEHY LEWIS LAWDAY CHRISTINE A. LEHY FREDERICK E. THRASHER EMPLOYEE RELATIONS COMMITTEE JOHN VALOIS, Chairman LUCENA BARTH, 1974 LEE BOURGOIN, 1975 ROBERT GUNNING, 1974 IVAN VALIELA, 1975 CATHERINE HENLEY, 1974 ANTHONY MAHLER, 1975 ACT OF INCORPORATION 5 II. ACT OF CORPORATION No. 3170 COMMONWEALTH OF MASSACHUSETTS Be it Known, That whereas Alpheus Hyatt, William Sanford Stevens, William T. Sedgwick, Edward G. Gardiner, Susan Minns, Charles Sedgwick Minot, Samuel Wells, William G. Farlow, Anna D. Phillips, and B. H. Van Vleck have associated themselves with the intention of forming a Corporation under the name of the Marine Biological Laboratory, for the purpose of establishing and maintaining a laboratory or station for scientific study and investigation, and a school for instruction in biology and natural his- tory, and have complied with the provisions of the statutes of this Commonwealth in such case made and provided, as appears from the certificate of the President, Treasurer, and Trustees of said Corporation, duly approved by the Commissioner of Corporations, and recorded in this office; Now, therefore, I, HENRY B. PIERCE, Secretary of the Commonwealth of Massachu- setts, do hereby certify that said A. Hyatt, W. S. Stevens, W. T. Sedgwick, E. G. Gardi- ner, S. Minns, C. S. Minot, S. Wells, W. G. Farlow, A. D. Phillips, and B. H. Van Vleck, their associates and successors, are legally organized and established as, and are hereby made, an existing Corporation, under the name of the MARINE BIOLOGICAL LAB- ORATORY, with the powers, rights, and privileges, and subject to the limitations, duties, and restrictions, which by law appertain thereto. Witness my official signature hereunto subscribed, and the seal of the Commonwealth of Massachusetts hereunto affixed, this twentieth day of March in the year of our Lord One Thousand Eight Hundred and Eighty-Eight. [SEAL] HENRY B. PIERCE, Secretary of the Commonwealth III. BYLAWS OF THE CORPORATION OF THE MARINE BIOLOGICAL LABORATORY (Revised February 11, 1"72) I. The members of the corporation shall consist of persons elected by the Board of Trustees. II. The officers of the Corporation shall consist of a Chairman of the Board of Trustees, President, Director, Treasurer and Clerk. III. The Annual Meeting of the members shall be held on the Friday following the second Tuesday in August in each year at the Laboratory in Woods Hole, Massachusetts, at 9:30 A.M., and at such meeting the members shall choose by ballot a Treasurer and a Clerk to serve one year, and nine Trustees to serve four years, and shall transact such other business as may properly come before the meeting. Special meetings of the members may be called by the Trustees to be held at such time and place as may be designated. IV. Twenty-five members shall constitute a quorum at any meeting. V. Any member in good standing may vote at any meeting, either in person or by proxy duly executed. VI. 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 mailing 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. 6 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY VII. The Annual Meeting of the Trustees shall be held promptly after the Annual Meeting of the Corporation at the Laboratory in Woods Hole, Massachusetts. Special meetings of the Trustees shall be called by the Chairman, the President, or by any seven Trustees, to be held at such time and place as may be designated, and the Secretary shall give notice thereof by written or printed notice, mailed to each Trustee at his address as shown on the records of the Corporation, at least one (1) week before the meeting. At such special meeting only matters stated in the notice shall be considered. Seven Trustees of those eligible to vote shall constitute a quorum for the transaction of business at any meeting. VIII. There shall be three groups of Trustees: (A) Thirty-six Trustees chosen by the Corporation, divided into four classes, each to serve four years. After having served two consecutive terms of four years each, Trustees are ineligible for re-election until a year has elapsed. (B) Trustees ex officio, who shall be the Chairman, the President, the Director, the Treasurer, and the Clerk. (C) Trustees Emeriti, who shall be elected from present or former Trustees by the Corporation. Any member of the Corporation in good standing who has attained the age of seventy years, or has attained the age of sixty-five and has retired from his home institution, and who has served a full elected term as a regular Trustee, shall be desig- nated Trustee Emeritus for life at the next annual meeting provided he signifies his wish to serve the Laboratory in that capacity. Any regular trustee who qualifies for emeritus status shall continue to serve as Trustee until the next Annual Meeting whereupon his office as regular Trustee shall become vacant and be filled by election by the Corporation. The Trustees ex officio and Emeriti shall have all the rights of the Trustees, except that Trustees Emeriti shall not have the right to vote. The Trustees and officers shall hold their respective offices until their successors are chosen and have qualified in their stead. IX. The Trustees shall have the control and management of the affairs of the Cor- poration. 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 appoint a Director of the Laboratory for a term not to exceed five years, provided the term shall not exceed one year if the candi- date has attained the age of 65 years prior to the date of the appointment. They may choose such other officers and agents as they may think best. They may fix the com- pensation and define the duties of all the officers and agents; and may remove them, or any of them except those chosen by the members, at any time. They may fill vacancies occurring in any manner in their own number or in any of the officers. The Board of Trustees shall have the power to choose an Executive Committee from their own num- ber, and to delegate to such Committee such of their own powers as they may deem expedient. They shall from time to time elect members to the Corporation upon such terms and conditions as they may think best. X. The Associates of the Marine Biological Laboratory shall be an unincorporated 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. XI. The consent of every Trustee shal be necessary to dissolution of the Marine Biological Laboratory. In case of dissolution, the property shall be disposed of in such REPORT OF THE DIRECTOR / manner and upon such terms as shall be determined by the affirmative vote of two-thirds of the Board of Trustees. XII. The account of the Treasurer shall be audited annually by a certified public accountant. XIII. These bylaws may be altered at any meeting of the Trustees, provided that the notice of such meeting shall state that an alteration of the bylaws will be acted upon. RESOLUTIONS ADOPTED BY THE TRUSTEES: I. RESOLVED: (A) The Executive Committee is hereby designated to consist of not more than ten members, including the ex officio members (Chairman of the Board of Trustees, Presi- dent, 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. (August 11, 1967.) (B) The Chairman of the Board of Trustees shall act as Chairman of the Executive Committee, and the President as Vice Chairman. A majority of the members of the Executive Committee shall constitute a quorum and a majority of those present at any properly held meeting shall determine its action. It shall meet at such times and places and upon such notice and appoint such sub-committees as the Committee shall deter- mine. (August 12, 1966). (C) The Executive Committee shall have and may exercise all the powers of the Board during the intervals between meetings of the Board of Trustees except those powers specifically withheld from time to time by the Board or by law. (August 16, 1963). (D) The Executive Committee shall keep appropriate minutes of its meetings, and its action shall be reported to the Board of Trustees. (August 16, 1963). II. RESOLVED: The elected members of the Executive Committee be constituted as a standing "Com- mittee for the Nominations of Officers," responsible for making nominations, at each Annual Meeting of the Corporation, and of the Board of Trustees, for candidates to fill each office as the respective terms of office expire (Chairman of the Board, President, Director, Treasurer, and Clerk). (August 16, 1963). III. RESOLVED: Any member of the Corporation in good standing who has attained the age of seventy years, or has attained the age of sixty-five and has retired from his home institution, shall automatically be designated a Life Member of the Corporation provided he signifies his wish to retain his membership in the Corporation. Life Members shall not have the right to vote and shall not be subject to the payment of any dues. (February 16, 1973). IV. REPORT OF THE DIRECTOR To: THE TRUSTEES OF THE MARINE BIOLOGICAL LABORATORY Speaking on the heritage of excellence at another famous institution, during the celebration of the one hundredth anniversary of the birth of William H. Welch, Alan Gregg said, "Something extraordinarily precious comes out of the close but entirely free association of men and women of superior character and capacity, because it lasts so ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY long and because it never comes from or appeal? to mediocrities." He could ha\e been describing the Marine Biological Laboratory. It is one of the Director's responsibilities to convey the spirit of the Laboratory not only to fellow scientists, but to foundations and the public at large. It is a formidable task. The greatness of the Laboratory cannot be assayed in a phrase. Now in its 87th year of service to world science, the Laboratory's research and justly renowned summer courses have traditionally been — and remain — at the cutting edges of their respective fields. We are justly proud of the twenty-nine winners of the Nobel Prize who have conducted various aspects of their work at the Laboratory, and of the nearly sixty members of the National Academy of Sciences who are also members of our Corporation (about ten per cent of the Corporation's membership). However a roster of distinguished in- vestigators tells only a part of the story. The life blood of the Laboratory is the student and beginning investigator, who finds in the unique scientific community of Woods Hole an exceptionally favorable environment for growth and maturation. The Lab- oratory's greatness lies, in large part, in its insistence that teaching and research are inextricably interwoven. Despite the inroads of inflation into our already meager resources, we have been firm in our resolve to increase the vitality of the Laboratory during the winter months. The appointment of E. F. MacNichol, Jr. as Assistant Director and his immediate and effective grasp of many problems of equipment and research services has enabled the Director to devote an increasing amount of his own time to fund-raising. During the year some two dozen foundations were visited; at least a dozen proposals have been submitted, with others in the offing. The Henry L. and Grace Doherty Charitable Foundation has pledged $500,000 toward the establishment of an Ecosystems Center at the Laboratory. However the grant is contingent upon our securing additional funds to support the Center. Of the $2.5 million dollars needed to underwrite the Center for the first five years, over a quarter of that amount has been pledged. In addition to the Doherty commitment the Clowes Family Foundation has given and pledged $40,000 per year for at least four years and the Grass Foundation an additional $40,000. I am also pleased to report that the Rowland Foundation will provide $110,000 in 1974 and 1975 in support of MacNichol's Laboratory of Sensory Physiology bringing the total amount for year- round programs to over $800,000. Securing the remaining funds is one of the priorities of our major fund raising campaign. In this year's efforts I have relied heavily on some of the "at large" officers and Trustees. Our Chairman, Denis Robinson and Treasurer, Mr. A. T. Daignault, and indeed all four at large Trustees have responded frequently to requests for assistance. Their location in New York, the seat of many of the foundations, has made Mr. Daignault and Mr. Golden especially vulnerable to the Director's calls. It is against this back- ground that I observe that the Trustees have moved to reapportion the composition of future classes. The Laboratory has thirty-six active Board members representing four classes of nine Trustees each. One new class is elected by Members of the Corporation at the annual meeting. In the past each class has included eight biologists elected by the Corporation and one "at large" member proposed by the Executive Committee and approved by the Corporation. In view of the shifting demands being placed on Trustees and the increasing need for our Board to represent a broader segment of the business and philanthropic population, it was felt by the Board that this ratio should be changed. Beginning in 1974 the Corporation Members will elect six biologist Trustees and approve the election of three at large Trustees by the Executive Committee. The first election under this new arrangement will take place at the 1974 Annual Meeting. REPORT OF THE DIRECTOR "Education for uncertainty" We are passionately dedicated to the principle that the main hope for progress against disease and the deterioration of the environment lies in research to which the student should have early exposure. The critical turn of mind fostered by research tempers the student's approach to the evolving base of science and its applications. It forms the basis of what John Whitehorn termed "education for uncertainty." In the past few years there has been a phasing out of federal grants for training programs. However, at least in 1974 the Laboratory is still receiving federal support for four of its summer courses and training programs. The National Institute of Child Health and Human Development is funding the Research Program in Reproductive Biology and the Embryology course. Dr. Eric Davidson heads the Embryology course and Dr. Robin A. Wallace is coordinator of the Research Program in Reproductive Biology. The National Institute of Neurological Diseases and Stroke has announced it will restore its support of the Frontiers in Research and Training Program. The Frontiers Program was initiated in the summer of 1971 on an experimental basis through support from the Institute. In 1973 the program suffered a serious cutback in funds. The Frontiers Program is designed to introduce increasing numbers of well-qualified scientists of minority ethnic groups into the neurosciences. It is an advanced training program primarily for individuals at the postdoctoral level who are seeking further experience in research or new ways of strengthening their capacity as instructors of neuroscience. The Program Coordinator is Dr. James G. Townsel. The National Science Foundation has announced its intention to support the Lab- oratory's new program in History of Evolutionary Biology on a one-summer trial basis in 1974. Dr. William Coleman of Johns Hopkins will head this summer's program which will focus on the development of evolutionary doctrine from the late 18th to the early 20th century. Support for three summer programs will come from private foundations and indi- viduals. The Grass Foundation continues to underwrite the Neurobiology course and the Waksman Foundation will support the Marine Ecology course. We expect Dr. Ernest B. Wright to continue his generous contribution to the Research Training Pro- gram in Excitable Membrane Biophysics and Physiology, to which IBM is also contributing. During the month of January, 28 undergraduate students spent the "short-term" or "mini-semester," as the period between the regular academic semesters is often called, studying developmental biology at the Laboratory. The students came from a wide variety of institutions that included the University of Wyoming, Texas A & I, Johns Hopkins University and Smith and Mount Holyoke Colleges. Dr. Louis E. DeLanney of Ithaca College was the coordinating instructor in coopera- tion with the Director. The teaching staff included Dr. David Walters of Boston University, Ms. Joanne Fortune, a laboratory instructor from Cornell, Doctors John and Annette Coleman from Brown University and Doctors Lester and Lucena Earth, resident scientific investigators at the MBL. During the students' four-week stay they resided in the MBL dormitory. Their schedule included two morning lectures five days a week, formal opportunities to use the laboratories in the afternoon and informal day and night laboratory sessions with Dr. DeLanney and Ms. Fortune. In addition to the scheduled programs there were 18 guest seminars which gave the students an opportunity to see how specialists in their field approach and resolve scientific problems. Investigators at the MBL and the Oceanographic Institution were among the speakers at the seminar. 10 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY The Oceanographic Institution gave the students an afternoon tour of its facilities and an opportunity to go aboard the KXORR and observe how a vessel can be used for seagoing research. The students also went out on the Laboratory's research vessel, the A. E. VERRILL, and participated in collecting and towing. Each student paid $580 to cover laboratory expenses and room and board. The students may receive four semester hours of credit at their own college for their intensive course work. Plans are under way to expand the program in January, 1975, to include more students and to offer a greater selection of courses. Already scheduled in addition to Develop- mental Biology is a Neurobiology course from the joint MBL and the Boston University Marine Program, which will be taught by Doctors MacNichol, Fred Lang and Alan Fein. A grant made jointly to Boston University and the Laboratory by the Alfred P. Sloan Foundation will provide funds for equipment and scholarships. Thus we continue to attempt, in John Shaw Billings' words, "To give to the world men who cannot only sail by the old charts, but who can make new and better ones for the use of others." Losses — and gains As 1973 drew to a close, Mr. Robert Kahler retired as Superintendent of Buildings and Grounds, after forty-seven years' service, twenty-five of them as Superintendent. His length of service is impressive, of course, but the numbers tell only a small part of the story of his effective and unflagging service to the Laboratory. Mr. Robert Gunning has been named to succeed Mr. Kahler. Dr. Arnold Lazarow, Professor and Chairman of the Department of Anatomy at the University of Minnesota, received the highest honor of the American Diabetes Associa- tion at its twenty-third annual meeting in Chicago. He was the recipient of the Associa- tion's Banting Medal and delivered the annual Banting Memorial Lecture. Seven members of the Corporation were elected to the National Academy of Sciences in 1974: Drs. Martin Gibbs, a member of the staff of the Research Program in Experi- mental Marine Botany; Jerome Gross; Jerard Hurwitz; Alex Novikoff; Ladd Prosser, Trustee (Class of 1976); Sarah Ratner and DeWitt Stetten. Election to membership in the NAS is considered to be one of the highest honors that can be accorded to an American scientist. The fact that about ten per cent of the Corporation's membership has been so honored (over five per cent of the total membership of the Academy in all fields are MBL biologists) continues to point to the Laboratory as one of the Nation's leading centers of research and communication in the life sciences. 1. MEMORIAL ELMER G. BUTLER BY MALCOLM S. STEINBERG All of his many friends at the Marine Biological Laboratory are saddened at the death of Elmer G. Butler, who passed away on February 23, 1972, after a brief illness. Born February 13, 1900, he graduated from Syracuse University and received the Ph.D. from Princeton University in 1926, where he shortly joined the faculty. Within eight years he rose to the chairmanship of the Department of Biology, a post he held until 1948. A native of Parish, New York, Elmer Butler was well known for his contributions to the study of regeneration in amphibian extremeties. His interest focused on the role of nerves in regeneration, and his last experiments dealt with the regeneration of ex- tirpated segments of the spinal cord itself. REPORT OF THE DIRECTOR 11 Active in many professional organizations, Dr. Butler was a Past President of AIBS, the American Society of Zoologists, and the Society for the Study of Development and Growth (now the Society for Developmental Biology). He had served as Chairman of N.I.H.'s Cell Biology Study Section and in editorial capacities for the Journal of Morphology, Developmental Biology, and the Journal of Experimental Zoology. A member of the Corporation of the Marine Biological Laboratory since 1945, Elmer Butler served at various times on the Evening Lecture Committee, the Library Com- mittee, the Buildings and Grounds Committee and the Nominations Committee. He was elected to four terms as Trustee (1951-55, 1955-59, 1960-64, 1964-68) and to three terms on the Executive Committee of the Board of Trustees (1952-54, 1956-58, and 1965 67). In all of these capacities he lent his council, always willingly and always in a quiet manner. A generous and congenial man, it was characteristic of Butler's gentle humor for him to turn to a friend during the 1968 Annual Meeting of the Corporation, after Director H. Burr Steinbach's announcement of his election to the office of Trustee Emeritus, and add with a broad smile, "an honor long overdue"! Eleanor and Elmer Butler had no children of their own, but one by one "adopted" Elmer's graduate students, all of whom retained and returned this devotion forever thereafter. We join in extending to Eleanor our deepest sympathy. 2. THE STAFF EMBRYOLOGY I. INSTRUCTORS ERIC DAVIDSON, Associate Professor of Biology, California Institute of Technology, Director of course GARY FREEMAN, Associate Professor of Biology, University of Texas, Associate Director of course FOTIS KAFATOS, Professor of Biology, Harvard University L. DENNIS SMITH, Associate Professor of Biology, Purdue University IGOR DAWID, Staff Member, Carnegie Institution of Washington, Department of Embryology JOSEPH GALL, Professor of Biology, Yale University II. SPECIAL LECTURERS JOSEPH ILAN, Case Western Reserve University EVERETT ANDERSON, Harvard University PAUL SCHROEDER, Washington State University RICHARD HALBERG, Cornell University PAUL GROSS, University of Rochester FRED WILT, University of California, Berkeley BARBARA HAMKALO, University of California. Irvine WILLIAM TELFER, University of Pennsylvania ROBERT BRIGGS, University of Indiana BURKE JUDD, University of Texas, Austin DAVID EPEL, Scripps Institution of Oceanography WALTER GEHRING, Yale University PETER WELLAUER, Carnegie Institution of Washington 12 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY III. TECHNICAL ASSISTANT CHRISTOPHER AMENSON, Graduate Fellow, California Institute of Technology IV. LABORATORY ASSISTANTS GREGORY BRITTEN, Marine Biological Laboratory BARBARA EHRLICH, Brown University GARY FREEMAN JOE GALL ERIC DAVIDSON FOTIS KAFATOS DENNIS SMITH IGOR DAWID EVERETT ANDERSON PAUL SCHROEDER RICHARD HALBERG PAUL GROSS FRED WILT BARBARA HAMKALO WILLIAM TELFER ROBERT BRIGGS BURKE JUDD JOSEPH ILAN DAVID EPEL WALTER GEHRING PETER WELLAUER V. LECTURES Is oogenesis necessary for subsequent development? Chromosome activity during oogenesis Synthesis and storage of RNA during oogenesis Specific protein synthesis during chorion formation in insects Protein synthesis during oogenesis and egg maturation Mitochondrial development during oogenesis Comparative ultrastructure of the female gamete Endocrine control of oocyte growth in annelids Specific protein synthesis during oogenesis in Xenopns The synthetic organization of the sea urchin egg Cytoplasmic polyadenylation of maternal RXA following activation of sea urchin eggs Visualization of transcription in oocytes Nurse cell function during oogenesis The developmental genetics of the "o" mutation in the Axolotl The organization of the Drosophila chromosome Regulation of mRNA translation during development Molecular mechanisms of fertilization and egg activation Genetic studies with reference to oocyte development Structure of ribosomal RNA PHYSIOLOGY I. CONSULTANTS ARTHUR H. BURR, Assistant Professor, Simon Fraser University LEON WEISS, Professor of Anatomy, Johns Hopkins University School of Medicine II. INSTRUCTORS JOHN J. CEBRA, Professor of Biology, Johns Hopkins University PIEN-CHIEN HUANG, Associate Professor of Biochemistry, Johns Hopkins University School of Hygiene and Public Health RU-CHIH C. HUANG, Associate Professor of Biology, Johns Hopkins University ISTVAN KRISKO, Associate Professor of Medicine, Baylor University School of Medicine BRIAN J. MCCARTHY, Professor of Biochemistry, University of California, San Francisco THOMAS D. POLLARD, Assistant Professor of Anatomy, Harvard University School of Medicine ROBERT A. PRENDERGAST, Associate Professor of Ophthalmology and Pathology, Johns Hopkins University School of Medicine REPORT OF THE DIRECTOR 13 GERALD WEISSMANN, Professor of Medicine, New York University School of Medicine DAVID A. YPHANTIS, Professor of Biochemistry and Biophysics, University of Con- necticut, Storrs III. SPECIAL LECTURERS FRANK AUSTIN, Professor of Medicine, Harvard University School of Medicine LUDWIG BRAND, Professor of Biology, Johns Hopkins University CHRISTOPHER CORDLE, Department of Biology, Johns Hopkins University ROBERT A. GOOD, Director, Sloan-Kettering Institute for Cancer Research DANIEL GOODENOUGH, Assistant Professor of Anatomy, Harvard University School of Medicine ELVIN A. KABAT, Professor of Microbiology, Human Genetics and Development, Columbia University, College of Physicians and Surgeons MANFRED KARNOVSKY, Professor of Biological Chemistry, Harvard University School of Medicine NEAL LANGERMAN, Assistant Professor of Biochemistry and Pharmacology, Tufts University School of Medicine FRITZ LIPMANN, Professor of Biochemistry, Rockefeller University JOHN MARCHALONIS, Head, Laboratory of Molecular Immunology, Walter and Eliza Hall Institute for Medical Research PETER SCHILLER, Laboratory of Chemical Biology, National Institutes of Health RAYMOND E. STEPHENS, Brandeis University DOUGLAS TAYLOR, Department of Biological Sciences, State University of New York at Albany GEORGE WEBER, Professor of Pharmacology, Indiana University School of Medicine IV. STAFF ASSOCIATES DENNIS BARRETT, Lecturer in Zoology, University of California, Davis GERALD A. COLE, Associate Professor of Epidemiology, Johns Hopkins University School of Hygiene and Public Health SYLVIA HOFFSTEIN, Research Associate, New York University School of Medicine SARAH HITCHCOCK, Research Associate in Biology, Brandeis University DfiLiLL MCCARTHY, Research Associate in Biochemistry, University of California Medical Center, San Francisco DENNIS POWERS, Assistant Professor of Biology, Johns Hopkins University ROBLEY WILLIAMS, JR., Associate Professor of Biology, Yale University V. COURSE ASSISTANTS MARTHA B. BARRETT, University of California, Davis CALVINA BAUMGARTNER, Johns Hopkins University VI. LECTURES THOMAS D. POLLARD 1. Introduction to motile systems II. The molecular basis of muscle contraction III. Cytoplasmic actin and myosin: the molecular basis of cell motility SARAH HITCHCOCK IV. The regulation of muscle contraction and cell motility THOMAS D. POLLARD V. The relation of cytoplasmic contractile proteins to movement 14 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY RAYMOND E. STEPHENS DAVID A. YPHANTIS R. WILLIAMS LUDWIG BRAND PETER SCHILLER CHRISTOPHER CORDLE DENNIS POWERS DENNIS BARRETT ISTVAN KRISKO R. C. HUANG P. C. HUANG BRIAN MCCARTHY ROBERT PRENDERGAST JOHN CEBRA ROBERT PRENDERGAST JOHN MARCHALONIS JOHN CEBRA JOHN MARCHALONIS ROBERT PRENDERGAST GERALD COLE GERALD WEISSMANN MANFRED KARNOYSKY GERALD WEISSMANN VI. Unique motile systems: bacterial flagella and peritrich spasmonemes VII. Microtubule related movements: cilia and flagella VIII. Microtubule related movement in the cytoplasm Physical biochemistry I. Physical biochemistry II. Physical biochemistry III. Oxygen-binding proteins I. Oxygen-binding proteins II. Application of fluorescence measurements to problems in biology Nanosecond fluorometry Resonant energy transfer Fluorescence polarization and time dependent anistrophy Physiological and genetic mechanisms of molecular adapta- tion to a changing environment I. Physiological and genetic mechanisms of molecular adapta- tion to a changing environment II. Control of gene expression in early development Mechanism of peptide elongation in protein biosynthesis Peptide chain initiation and termination Mechanism of peptide bond formation — function of ribo- somal proteins; species specificity Studies on transcription and translation of specific genes I. Studies on transcription and translation of specific genes II. Studies on transcription and translation of specific genes III. Fractionation and sequence analysis of radioactive nucleic acid I. Fractionation and sequence analysis of radioactive nucleic acid II. Fractionation and sequence analysis of radioactive nucleic acid III. Structure of interphase chromosomes Fractionation of chromatin — background and approaches Properties of fractionated chromatin Specificity of DNA/RNA hybridization reactions Unique sequence DNA/RNA hybridization in eukaryotes The immune response The activities of antibodies including their specific inter- actions with antigens Antibodies: the structural bases of antigen-binding speci- ficity and their genetic control Cell mediated immune responses Phylogeny of immune response Antibody forming cells and their precursors Receptors on lymphocytes Immunopathology Viral infections and consequences of an immune response Metchnikoff revisited — lysosomes and inflammation Biochemical correlates of phagocytosis Membranes — their integrity and disruption in tissue injury Lysosomes, microtubules and cyclic nucleotides REPORT OF THE DIRECTOR 15 FRANK AUSTEN Molecular basis of inflammation ROBERT GOOD Biological amplification systems in immunology FRITZ LIPMANN Nonribosomal polypeptide synthesis of some antibiotics NEAL LANGERMAN Calorimetric studies of thymidylate synthesis DOUGLAS L. TAYLOR The contractile basis of amoeboid movement: studies on isolated cytoplasmic models DANIEL GOODENOUGH Cell junctions and intracellular communication: bio- chemical and structural studies on isolated junctions ELVIN KABAT Attempts to make models of the polypeptide backbone of immunoglobulins and other proteins EXPERIMENTAL MARINE BOTANY I. CONSULTANTS MARTIN GIBBS, Brandeis University FRANK A. LOEWUS, State University of New York at Buffalo RALPH S. QUATRANO, Oregon State University JEROME A. SCHIFF, Brandeis University W. RANDOLPH TAYLOR, University of Michigan MICHAEL J. WYNNE, University of Texas II. INSTRUCTOR JOHN A. WEST, University of California, Berkeley III. SPECIAL LECTURERS JOE RAMUS, Yale University GORDON F. LEEDALE, University of Leeds JOHN LEE, City University of New York G. F. PAPENFUSS, University of California, Berkeley IV. COURSE ASSISTANTS RICHARD WETHEKBEE, University of Michigan KENNETH MALINOWSKI, Yale University EXPERIMENTAL INVERTEBRATE ZOOLOGY I. CONSULTANTS F. A. BROWN, JR., Professor of Zoology, Northwestern University C. LADD PROSSER, Professor of Physiology, University of Illinois CLARK P. READ, Professor of Biology, Rice University ALFRED C. REDFIELD, Woods Hole Oceanographic Institution W. D. RUSSELL-HUNTER, Professor of Zoology, Syracuse University JAMES CASE, Professor of Biology, University of California, Santa Barbara II. INSTRUCTORS ROBERT K. JOSEPHSON, University of California, Irvine, Director of course JAMES MORIN, University of California at Los Angeles, Associate Director of course M. J. GREENBERG, Florida State University 16 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY JOSEPH B. JENNINGS, University of Leeds RALPH G. JOHNSON, University of Chicago ANN E. KAMMER, Kansas State University, Manhattan CHARLOTTE P. MANGUM, College of William and Mary S. K. PIERCE, University of Maryland THOMAS J. M. SCHOPF, University of Chicago III. SPECIAL LECTURERS BARRY W. ACHE, Florida Atlantic University FRANK A. BROWN, Northwestern University FRED GRASSLE, Woods Hole Oceanographic Institution BERND HEINRICH, University of California, Berkeley P. HOCHACHKA, University of British Columbia L. KIRSCHNER, Washington State University HOWARD LENHOFF, University of California, Irvine C. LADD PROSSER, University of Illinois RUDI STRICKLER, Johns Hopkins University STEVEN VOGEL, Duke University L. WILKENS, Bryn Mawr IV7. ASSOCIATE STAFF T. GOLDSMITH, Yale University E. KRAVITZ, Harvard FRED LANG, Boston University V. COURSE ASSISTANTS JOHN C. CORNELL, University of California, Berkeley, teaching assistant MARY L. CORNELL, course secretary BRIAN D. KELLER, Johns Hopkins University, teaching assistant VI. LECTURES R. JOSEPHSON Introduction to Woods Hole and the course S. PIERCE Some thoughts on the immortal sponges R. JOSEPHSON Relationships between animal phyla Coelenterates T. SCHOPF The biology of the bryozoa M. GREENBERG Mollusca I. Mollusca II. J. MORIX Protochordates C. MANGUM Annelida A. KAMMER Arthropod I. Arthropod II. J. MORIN The enigmatic echinoderms T. SCHOPF Ecological genetics of marine invertebrates C. MANGUM Respiration I: gas exchange (homage to August Krogh) Respiration II: oxygen transport (from monomer to the mud flat) S. VOGEL Fluid and animals I : boundary layers and how they grow Fluid and animals II: flow adaptations of sessile animals M. GREENBERG Patterns of circulation in invertebrates R. JOSEPHSON The origin of nervous systems REPORT OF THE TREASURER 17 A. KAMMER The generation of rhythmic motor outputs J. MORIN Bioluminescence as diverse effectors F. LANG Arthropod neuromuscular systems M. GREENBERG Comparative physiology and pharmacology of molluscan muscle A. KAMMER Adaptations to environmental temperature B. HEINRICH Thermoregulation in sphinx moths J. JENNINGS Digestive physiology in invertebrates, especially the acoelomates I. J. MORIN Organization and coordination in colonial animals J. JENNINGS Digestive physiology in invertebrates, especially the acoelomates II. T. SCHOPF Extinction and sea floor spreading: crises in the history of life R. JOHNSON The amazing fossils of Mazon Creek or experimental invertebrate zoology 260 million years ago S. PIERCE Salinity tolerance, osmosis and cell volume regulation Water balance continued, and a look at excretory systems L. KIRSCHNER Ionic regulation and active transport MARINE ECOLOGY I. CONSULTANTS HARLYN O. HALVERSON, Professor of Microbiology, Brandeis University J. WOODLAND HASTINGS, Professor of Biology, Harvard University HOWARD L. SANDERS, Senior Scientist, Woods Hole Oceanographic Institution LAWRENCE B. SLOBODKIN, Professor of Biology, State University of New York at Stony Brook ROGER Y. STANIER, Professor of Microbiology, Institut Pasteur, Paris II. INSTRUCTORS HOLGER W. JANNASCH, Senior Scientist, Woods Hole Oceanographic Institution, in charge of program JANE GIBSON, Associate Professor of Microbiology, Cornell University EDWARD R. LEADBETTER, Professor of Microbiology, Amherst College KENNETH H. NEALSON, Assistant Professor of Marine Biology, Scripps Institution of Oceanography RALPH S. WOLFE, Professor of Microbiology, University of Illinois III. RESEARCH ASSOCIATES ANATOL EBERHARD, Assistant Professor of Chemistry, Ithaca College CAROLYN EBERHARD, Research Associate, Cornell University CHARLES C. REMSEN, Associate Scientist, Woods Hole Oceanographic Institution DONALD M. SPOON, Assistant Professor, Georgetown University CRAIG D. TAYLOR, Postdoctoral Investigator, Woods Hole Oceanographic Institution EDWARD O. WILSON, Professor of Zoology, Harvard University IV. LABORATORY ASSISTANTS CHRISTOPHER VAN RAALTE, Marine Biological Laboratory CHARLES B. MARKOWITZ, Amherst College 18 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY V. SPECIAL LECTURERS RICHARD \V. CASTENHOLZ, Professor of Botany, University of Oregon DAVID P. CHYNOWETH, Associate Professor, University of Michigan JOEL C. GOLDMAN, Assistant Scientist, Woods Hole Oceanographic Institution ALEX KEYNAN, Professor of Microbiology, The Hebrew University, Jerusalem RALPH MITCHELL, Professor of Applied Microbiology, Harvard University JOHN R. POSTGATE, Professor of Microbiology, The University of Sussex JOHN H. RYTHER, Senior Scientist, Woods Hole Oceanographic Institution HOWARD L. SANDERS, Senior Scientist, Woods Hole Oceanographic Institution WERNER STUMM, Professor of Chemistry, Federal University, Zurich JOHN M. TEAL, Senior Scientist, Woods Hole Oceanographic Institution HOLGER W. JANNASCH JANE GIBSON EDWARD R. LEADBETTER KENNETH H. NEALSON RALPH S. WOLFE RICHARD W. CASTENHOLZ DAVID P. CHYNOWETH JOEL C. GOLDMAN ALEX KEYNAN RALPH MITCHELL JOHN R. POSTGATE CHARLES C. REMSEN VI. LECTURES Introduction to microbial ecology I Introduction to microbial ecology II Continuous culture at low substrate concentrations Competition in continuous culture Ecological aspects of denitrification Experiments in deep-sea microbiology Photosynthetic microorganisms I Photosynthetic microorganisms II Movement in photosynthetic bacteria Biochemistry of CO2-fixation Biology of Bdellovibrio Making a living aerobically Enrichment cultures Aspects of microbial geochemistry I Aspects of microbial geochemistry 1 1 Microbial attack on hydrocarbons Amine utilization by bacteria Regulation of bacterial bioluminescence Basic bacterial genetics I Basic bacterial genetics II Survey of bioluminescent systems Marine bacterial viruses Anaerobic ways to make a living I Anaerobic ways to make a living II Anaerobic metabolic food chains I Anaerobic metabolic food chains II Iron bacteria Hot springs and their microorganisms I Hot springs and their microorganism? 1 1 Sulnde formation in sediments Continuous culture of photosynthetic organisms The bacterial endospore : an example of a dormant stage in microorganisms Microbial predation and chemotaxis Ecological consequences of recent knowledge of the bio- chemistry of nitrogen fixation Survival of microorganisms Fine structure survey of autotrophic and phototrophic bacteria REPORT OF THE DIRECTOR 19 JOHN H. RYTHER Aquaculture HOWARD L. SANDERS Oil spill ecology DONALD M. SPOON Protozoan ecology WERNER STUMM Biological mediation of redox processes in seawater Role of redox potential and ecological milieu of micro- organisms CRAIG D. TAYLOR Structure and methylation of co-enzyme M JOHN M. TEAL Experimental marsh ecology EDWARD O. WILSON Geographical ecology I Geographical ecology II NEUROBIOLOGY I. INSTRUCTORS MICHAEL V. L. BENNETT, Professor of Anatomy, Albert Einstein College of Medicine, co-director of course JOHN E. DOWLING, Professor of Biology, Harvard University, co-director of course ANTHONY L. F. GORMAN, Professor of Physics, Boston University School of Medicine RODOLFO R. LLINAS, Professor of Physiology and Biophysics, University of Iowa GEORGE PAPPAS, Professor of Anatomy, Albert Einstein College of Medicine VICTOR P. \\'HITTAKER, Sir William Dunn Reader in Biochemistry, Cambridge University II. SPECIAL LECTURERS GEORGE KATZ, Columbia University MALCOLM BRODWICK, Duke University J. S. McREYNOLDS, National Institute of Neurological Diseases and Stroke F. A. DODGE, JR., Rockefeller University J. E. HEUSER, University College, London, England S. D. Erulker, University of Pennsylvania F. F. WEIGHT, National Institute of Mental Health C. NICHOLSON, University of Iowa H. GAINER, National Institutes of Health M. J. DOWDALL, Cambridge University, England E. J. SIMON, New York University Medical Center F". HOSKIN, Illinois Institute of Technology M. JOHNSON, Medical Research Council, England E. A. KRAVITZ, Harvard Medical School N. W. DAW, Washington LIniversity III. LECTURES JOHN DOWLING Anatomical analysis of the vertebrate retina R. LLINAS Neurocytology of the cerebellar cortex : a model of general CNS morphology M. V. L. BENNETT The central dogma in neurobiology GEORGE KATZ Introduction to electrophysiological measurements A. L. F. GORMAN Introduction to anatomy and physiology of molluscan neurons GEORGE D. PAPPAS Membranes and fine structure of neurons and the fine structure of membranes The fine structure of the chemically-transmitting synapses 20 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY S. G. WAXMAN JOHN E. DOWLING A. M. BROWN A. L. F. GORMAN MALCOLM BRODWICK J. S. Me REYNOLDS R. LLINAS F. A. DODGE, JR. J. E. HEUSER S. D. ERULKAR M. V. L. BENNETT R. LLINAS M. V. L. BENNETT F. F. WEIGHT M. V. L. BENNETT R. LLINAS C. NICHOLSON M. V. L. BENNETT H. GAINER V. P. WHITTAKER M. J. DOWDALL V. P. WHITTAKER E. J. SIMON F. HOSKIN H. POLLARD D. SOIFER M. JOHNSON E. A. KRAVITZ J. E. DOWLING R. L. CHAPPELL J. E. DOWLING N. W. DAW J. E. DOWLING M. H. GOLDSTEIN The fine structure of electrically-transmitting synapses The structural specializations of the axon in relation to function Synaptic organization of the vertebrate retina Ionic movements in a molluscan neuron Passive membrane properties of molluscan neurons and extracellular diffusion Electrogenic Na+ pump Slow potassium conductance Comparative physiology of hyperpolarizing receptor potentials General properties of chemically mediated synaptic trans- mission— the giant synapse The quantum of synaptic transmission Origin and fate of synaptic vesicles Synaptic delay Postsynaptic actions Inhibition Receptor synapses Synaptic potentials in sympathetic ganglia Electrical synapses Interpretation of spikes in the CNS Neuronal circuits in the cerebellar cortex Cerebellum in the regulation of movement Field potential analysis in the cerebellum Physiology of electrical synapses Control of very simple effector systems Control of less simple effector systems Effects of synaptic activity on macromolecular metabolism of neurons Biochemistry of the neurone — introduction Subcellular fractionation techniques applied to the nervous system Synthesis and storage of neurotransmitters : parallels and contrasts Dynamics of vesicle formation and discharge Biochemical studies on the opiate and other receptors Cholinergic drugs The structure of chromatin granules (EM and X-ray) Cyclic nucleotides in neuronal function Neuronal microtubules Axonal degeneration Studies of synaptic chemistry in the lobster nervous system Review : anatomy, chemistry, and physiology of photo- receptors The Limulns lateral eye The median ocellus of the dragonfly: a "simple" retina The processing of visual information Neurophysiology of color vision Gross retinal potentials, glial potentials, and the problem of visual adaptation Processing of information in the auditory cortex REPORT OF THE DIRECTOR 21 THE LABORATORY STAFF HOMER P. SMITH, General Manager JANE FESSENDEN, Librarian JOHN J. VALOIS, Manager, Supply Department LEWIS M. LAWDAY, Assistant Manager, Supply Department FRANK A. WILDES, Controller ROBERT GUNNING, Superintendent, Build- ings and Grounds JIM A. HANCOCK, Manager, Department of Research Service SANDRA E. BELANGER, Assistant Editor, The Biological Bulletin EDWARD J. BENDER CHRISTINA BOWDEN FLORENCE S. BUTZ SHIRLEY J. DELISLE GENERAL OFFICE JOYCE C. GLASS ELAINE C. PERRY LORRAINE A. RUDDICK MARY R. TAVARES DAVID J. FITZGERALD ELIZABETH A. FUSELER LENORA JOSEPH LIBRARY HOLLY KARALEKAS ANTHONY J. MAHLER THERESA K. McKEE MAINTENANCE OF BUILDINGS AND GROUNDS JOHN F. ACKERSON ELDON P. ALLEN LEE E. BOURGOIN BERNARD F. CAVANAUGH JAMES S. CLARKE JOHN V. DAY MANUEL P. DUTRA MICHAEL W. EDDINS GLENN R. ENDS CHARLES K. FUGLISTER ELIZABETH J. GEGGATT RICHARD E. GEGGATT, JR. ELIZABETH KUIL DONALD B. LEHY RALPH H. LEWIS WILLIAM M. LOCHHEAD RICHARD C. LOVERING ALAN G. LUNN JOHN E. MAURER STEPHEN MILLS FREDERICK E. THRASHER FREDERICK E. WARD RALPH D. WHITMAN DEPARTMENT OF RESEARCH SERVICE GAIL CAVANAUGH LOWELL V. MARTIN CHRISTINE A. LEHY FRANK E. SYLVIA SUPPLY DEPARTMENT EDWARD G. ENOS, JR. JOYCE B. ENOS CLIFFORD A. GOUDEY DAVID H. GRAHAM ROBERT O. LEHY A. DICKSON SMITH EUGENE A. TASSINARI BRUNO F. TRAPASSO JOHN M. VARAO ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY 3. INVESTIGATORS, LILLIE, GRASS, AND RAND FELLOWS; STUDENTS Independent Investigators, 1973 ACHE BARRY W., Assistant Professor of Zoology, Florida Atlantic University ADELMAN, WILLIAM J., JR., Chief, Laboratory of Biophysics, Xational institutes of Health, National Institute of Neurological Diseases and Stroke AFZELIUS, BJORN A., Associate Professor, Wenner-Gren Institute, University of Stockholm, Sweden ALLEN, ROBERT DAY, Professor of Biology, State University of Xew York at Albany ANDERSON, EVERETT, Professor of Anatomy, Harvard Medical School ANDERSON, PETER J., Assistant Professor, University of Ottawa, Canada AONO, OSAMU, Associate Professor, Jichi Medical School, Tochigi-ken, Japan ARMSTRONG, CLAY M., Associate Professor of Physiology, University of Rochester ARNOLD, JOHN M., Associate Professor, University of Hawaii, Pacific Biochemical Research Center and Acting Director, Kewalo Marine Laboratory BAGCHI, MIHIR, Assistant Professor, Oakland University BALL, ERIC G., Professor Emeritus Biological Chemistry, Harvard Medical School BARKER, JEFFERY LANGE, Special Fellow, National Institutes of Health BARLOW, ROBERT B., JR., Associate Professor of Sensory Communication, Syracuse University BARNES, STEPHEN N., X.I.H. Postdoctoral Fellow, Vale University BARRETT, DENNIS, Assistant Professor, University of California, Davis BARTELL, CLELMER K., Assistant Professor, Louisiana State University in New Orleans BAUER, G. ERIC, Associate Professor of Anatomy, University of Minnesota BELAMARICH, FRANK A., Professor, Boston University BELL, WAYNE H., Instructor of Biology, Middlebury College BENNETT, M. V. L., Professor of Anatomy, Albert Einstein College of Medicine BERCKEN, JOZEF M. M. VAN DEN, Postdoctoral Research Associate, Duke University BERGSTROM, BEVERLY H., Postdoctoral Student, University of Hawaii BERNARD, GARY D., Associate Professor, Vale University School of Medicine BESSO, JOSEPH A., JR., X.I.H. Postdoctoral Fellow, Albert Einstein College of Medicine BEZANILLA, FRANCISCO M., Associate in Physiology, University of Rochester BLAUSTEIN, MORDECAI P., Associate Professor of Physiology and Biophysics, Washington University School of Medicine BORGESE, THOMAS A., Associate Professor of Biology, Lehman College — The City University of Xew York BORISY, GARY G., Associate Professor of Molecular Biology and Zoology, University of Wisconsin BOURNE, DONALD W., Research Biologist, Marine Research, Inc. BRANDT, PHILIP W., Associate Professor of Anatomy, Columbia University BRINLEY, F. J., JR., Associate Professor of Physiology, Johns Hopkins Medical School BRODWICK, MALCOLM S., N.I.H. Postdoctoral Fellow, Duke University BROWN, FRANK A., JR., Morrison Professor of Biology, Xorthwestern University BROWN, JOEL E., Professor of Anatomy, Vanderbilt University BUCK, JOHN, Chief, Laboratory of Physical Biology, Xyational Institutes of Health BURDICK, CAROLYN J., Associate Professor, Brooklyn College BURGER, MAX M., Professor and Chairman, Department of Biochemistry, Biocenter of the University of Basel, Switzerland BURR, A. H., Assistant Professor, Simon Fraser University CARBONE, EMILIO, Visiting Fellow, Xational Institute of Mental Health CASE, JAMES, Professor, University of California, Santa Barbara CAVAGNA, GIOVANNI A., Associate Professor of Physiology, Universita degli studi di Milano, Italy CEBRA, JOHN J., Professor of Biology, Johns Hopkins University CHAMBERS, EDWARD L., Professor of Physiology and Biophysics, University of Miami School of Medicine CHANDLER, W. K., Professor of Physiology, Yale Medical School CHAPPELL, RICHARD L., Assistant Professor, Hunter College of the City University of Xew York CLARK, ARNOLD M., Professor of Biological Sciences, University of Delaware CLARK, VERNON, Associate Professor of Biology, Xorth Carolina Central University REPORT OF THE DIRECTOR CLARK, WALLIS H., JR., Associate Professor of Biology, University of Houston CLUSIN, WILLIAM, Graduate Student, Albert Einstein Medical College COHEN, ADOLPH I., Professor of Anatomy and Research Professor of Ophthalmology, Washington University School of Medicine COHEN, LAWRENCE B., Associate Professor, Yale University School of Medicine COLE, GERALD A., Associate Professor, Johns Hopkins University, School of Medicine COLE, KENNETH S., Research Biophysicist, Laboratory of Biophysics, National Institutes of Health COOHILL, THOMAS P., Assistant Professor, Western Kentucky University COOPERSTEIN, SHERWIN J., Professor of Anatomy, University of Connecticut COSTELLO, DONALD PAUL, Kenan Professor of Zoology, University of North Carolina at Chapel Hill COUCH, ERNEST F., Assistant Professor, Texas Christian University Cox, BRAD, Guest Worker, Laboratory of Biophysics, National Institutes of Health CREMER, GERTRUD, Investigator, University of Minister, Germany CROWELL, SEARS, Professor, Indiana University DAVIDSON, ERIC H., Associate Professor, California Institute of Technology DAVILA, HECTOR, Research Associate, Yale University DAW, NIGEL W., Associate Professor, Washington University Medical School DAWID, IGOR B., Staff Member, Carnegie Institution of Washington DAWSON, M. JOAN, Postdoctoral Fellow, Columbia University DETERRA, NOEL, Assistant Member, The Institute for Cancer Research, Philadelphia DE\VEER, PAUL, Associate Professor of Physiology and Biophysics, Washington University School of Medicine DINGLE, ALLAN D., Associate Professor, McMaster University, Hamilton, Canada DODGE, FREDERICK A., Adjunct Associate Professor, Rockefeller University DONATI, FRANCOIS, Graduate Student, University of Toronto, Canada DOWDALL, MICHAEL JOHN, Senior Assistant in Research, Cambridge University, England DOWLING, JOHN E., Professor of Biology, Harvard University DOWNING, DAVID R., Assistant Professor of Systems Engineering, Boston University DuBois, ARTHUR B., Professor of Physiology, University of Pennsylvania School of Medicine DUNHAM, PHILIP B., Professor of Biology, Syracuse University EASTWOOD, ABRAHAM B., III., Postdoctoral Trainee, Columbia University EBERHARD, ANATOL, Associate Professor, Ithaca College EBERHARD, CAROLYN, Lecturer, Cornell University ERULKAR, SOLOMON D., Professor of Pharmacology, University of Pennsylvania FARMANFARMAIAN, A., Professor of Physiology, Rutgers University FEIN, ALAN, Postdoctoral Fellow, Johns Hopkins University FERGUSON, THOMAS, Professor and Chairman, Department of Biology, Delaware State College FERTZIGER, ALLEN P., Assistant Professor of Physiology, University of Maryland School of Medicine FINE, JACOB, Director of Shock Research Division, Harvard Surgical Unit, Boston City Hospital FINGERMAN, MILTON, Professor of Biology, Tulane University FISHMAN, HARVEY M., Assistant Professor of Biological Sciences, State University of New York at Albany FLOCK, AKE, Assistant Professor, Karolinska Institute, Stockholm, Sweden FRAZIER, DONALD T., Associate Professor, University of Kentucky FREEMAN, GARY, Associate Professor of Zoology, LTniversity of Texas at Austin FUKUMOTO, MAKOTO, Instructor, Nagoya City University, Japan FUTAMACHI, KIN J., Research Associate, Stanford University GAINER, HAROLD, Research Physiologist, National Institutes of Health GALL, JOSEPH, Professor, Yale University GEDULDIG, DONALD, Assistant Professor, University of Maryland, School of Medicine GIBBS, MARTIN, Professor, Brandeis University GIBSON, JANE, Associate Professor of Microbiology, Cornell University GILBERT, DANIEL L., Research Physiologist, Laboratory of Biophysics, National Institute-; of Health GOLDMAN, DAVID E., Professor of Physiology and Biophysics, Medical College of Pennsylvania 24 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY GOLDSMITH, TIMOTHY H., Professor and Chairman, Department of Biology, Yale University GOLDSTEIN, MOISE, H., JR., Professor of Electrical Engineering and Associate Professor of Biomedical Engineering, Johns Hopkins University GOODMAN, LESLEY J., Senior Lecturer, Department of Zoology, Queen Mary College, London University, England GORMAN, ANTHONY L. F., Professor of Physiology, Boston University School of Medicine GREEN, DANIEL G., Associate Professor, University of Michigan, Ann Arbor GREENBERG, MICHAEL J., Associate Professor of Biological Sciences, Florida State University GRIESS, GARY A., Instructor, University of Rochester Medical Center GROSCH, DANIEL S., Professor of Genetics, North Carolina State University GROSSFELD, ROBERT M., Assistant Professor, Cornell University GROSSMAN, ALBERT, Associate Professor, New York University Medical Center GRUNDFEST, HARRY, Emeritus Professor, Columbia University GUTTMAN, RITA, Professor of Biology, Brooklyn College of the City University of New York HALVORSON, HARLYN O., Director, Rosenstiel Basic Medical Sciences Research Center, Brandeis University HARA, TOMIYUKI, Professor, Department of Biology, Nara Medical University, Japan HARDING, C. V., Professor and Chairman, Department of Biological Sciences, Oakland University HASCHEMEYER, AUDREY E. V., Associate Professor of Biological Sciences, Hunter College of the City University of New York HENKART, PIERRE, National Institutes of Health HENLEY, CATHERINE, Research Associate in Zoology, University of North Carolina at Chapel Hill HERMAN, LAWRENCE, Professor, State University of New York, Downstate Medical Center HEUSER, JOHN E., Moseley Traveling Fellow of Harvard University, University College, London, England HILL, ROBERT B., Associate Professor of Zoology, University of Rhode Island HINSCH, GERTRUDE, Associate Professor, Institute for Molecular and Cellular Evolution, Uni- versity of Miami HITCHCOCK, SARAH E., Postdoctoral Research Associate, Brandeis University HODSON, ROBERT C., Assistant Professor, University of Delaware HOFFSTEIN, SYLVIA, Associate Research Scientist, New York University School of Medicine HOSKIN, FRANCIS C. G., Professor of Biology, Illinois Institute of Technology HUANG, P. C., Associate Professor of Biochemistry, Johns Hopkins University HUANG, R. C., Associate Professor of Biology, Johns Hopkins University HUBBARD, RUTH, Lecturer, Harvard University HUMPHREYS, SUSIE, Assistant Researcher, University of Hawaii HUMPHREYS, TOM, Associate Professor of Biochemistry, University of Hawaii INOUE, SADAYUKI, Assistant Professor, Montreal University JANNASCH, HOLGER W., Senior Scientist, Woods Hole Oceanographic Institution JENNINGS, JOSEPH BRIAN, Reader in Invertebrate Zoology, University of Leeds, England JOHNSON, MARTIN KEITH, Research Biochemist, Medical Research Council, Toxicology Unit, England JOHNSON, RALPH G., Professor, University of Chicago JOSEPHSON, ROBERT K., Professor, University of California, Irvine JOYNER, RONALD W., Graduate Student, Duke University KAFATOS, F. C., Professor, Harvard University KAMINER, BENJAMIN, Professor and Chairman, Boston University School of Medicine KAMMER, ANN E., Associate Professor, Kansas State University KANESHIRO, EDNA S., Assistant Professor, University of Cincinnati KATZ, GEORGE M., Assistant Professor, Columbia University KAWAI, MASATAKA, Research Associate, Columbia University KELLEY, B. J., JR., Assistant Professor of Biology, The Citadel KIRKHAM, JOHN BRIAN, Head of the Department of Electron Microscopy, Queen Mary College, London University, England KOCHAKIAN, CHARLES D., Professor and Director, Medical Center, University of Alabama in Birmingham KORN, HENRI, Visiting Professor, Harvard Medical School REPORT OF THE DIRECTOR 25 KRAVITZ, EDWARD A., Professor of \eurobiology, Harvard Medical School KRIEBEL, MAHLON E., Assistant Professor of Physiology, Upstate Medical Center, State Uni- versity of New York KRISKO, ISTVAN, Assistant Professor of Medicine and Biology, Johns Hopkins University School of Medicine KRUCZYNSKI, WILLIAM LEONARD, Assistant Professor of Biology, Hartwick College KUFFLER, STEPHEN W., Robert Winthrop Professor of Neurobiology and Department Chairman, Harvard Medical School KUHNS, WILLIAM J., Associate Professor, Department of Pathology, New York University School of Medicine KUNOV, HANS, Associate Professor, University of Toronto, Canada KUSANO, KIYOSHI, Associate Professor, Illinois Institute of Technology KUWASAWA, KIYOAKI, Assistant Professor, Tokyo Kyoiku University, Japan LALL, ABNER BISHAMBER, Assistant Professor, The City College of the City University of New York LANDOWNE, DAVID, Assistant Professor, University of Miami LASEK, RAYMOND J., Assistant Professor, Case Western Reserve University LASH, JAMES W., Professor of Anatomy, University of Pennsylvania School of Medicine LAZAROW, ARNOLD, Professor and Head, Department of Anatomy, University of Minnesota LEADBETTER, E. R., Professor of Biology, Amherst College LEAK, LEE V'., Professor and Chairman, Department of Anatomy, Howard University LEE, JOHN J., Professor of Biology, City College of the City University of New York LESTER, ROGER, Associate Professor, Boston University School of Medicine LEVIN, JACK, Associate Professor of Medicine, Johns Hopkins University School of Medicine LEVINTHAL, CYRUS, Professor of Biology, Columbia University LEVY, MILTON, Professor of Biochemistry, New York University Li TICK Y, RAYMOND JOHN, Professor of Pharmacology and Associate Professor of Medicine, University of Cincinnati LIUZZI, ANTHONY, Associate Professor, Lowell Technological Institute LLINAS, RUDOLFO, Head, Division of Neurobiology, Professor of Physiology and Biophysics, University of Iowa LOEWENSTEIN, W. R., Professor and Chairman, Department of Physiology and Biophysics, University of Miami School of Medicine LOEWUS, FRANK A., Professor of Biology, State University of New York at Buffalo LONGO, FRANK J., Assistant Professor, University of Tennessee Medical Units LOOMIS, WILLIAM F., JR., Assistant Professor, University of California, San Diego LORAND, L., Professor, Northwestern University MANALIS, RICHARD S., Assistant Professor of Physiology, University of Cincinnati MANGUM, CHARLOTTE P., Associate Professor of Biology, College of William and Mary MARCHALONIS, JOHN J., Head, Laboratory of Molecular Immunology, Walter and Eliza Hall Institute of Medical Research MCCARTHY, BRIAN J., Professor of Biochemistry, University of California, San Francisco McMAHON, ROBERT F., Assistant Professor of Biology, University of Texas at Arlington McREYNOLDS, JOHN S., Research Neurophysiologist, National Institute of Neurological Diseases and Stroke METUZALS, J., Professor in Charge of the Electron Microscopy Unit, University of Ottawa, Faculty of Medicine METZ, CHARLES B., Professor of Biology, University of Miami MICHAEL, ALLAN D., Independent Investigator, Marine Biological Laboratory MITCHELL, RALPH, Gordon McKay Professor of Applied Biology, Harvard University MOORE, JOHN W., Professor of Physiology, Duke University MOORE, LEE E., Associate Professor of Physiology, Case Western Reserve University MORIN, JAMES G., Assistant Professor, University of California, Los Angeles MOSESSON, MICHAEL W., Associate Professor of Medicine, State University of New York, Downstate Medical Center MOTE, MICHAEL L, Assistant Professor, Temple University MULLINS, L. J., Professor of Biophysics and Chairman, LTniversity of Maryland School of Medicine 26 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MtiRER, ERIK HOMANN, Research Assistant Professor, Temple University Health Sciences Center NAKAJIMA, SHIGEHIRO, Associate Professor, Purdue University NAKAJIMA, YASUKO, Associate Professor, Purdue University NARAHASHI, TOSHIO, Professor and Head of Division of Pharmacology, Duke University NEALSON, KENNETH, Assistant Professor, Scripps Institute of Oceanography NELSON, LEONARD, Chairman, Department of Physiology, Medical College of Ohio NICHOLLS, JOHN G., Associate Professor of Neurobiology, Harvard Medical School NICHOLS, RUTH ANN, Postdoctoral Research Associate, Purdue University NICHOLSON, CHARLES, Assistant Professor, University of Iowa NIELSEN, JENNIFER, Research Associate, Hunter College OHKI, SHINPEI, Research Associate Professor of Biophysical Sciences, State University of New- York at Buffalo PAPPAS, GEORGE D., Professor of Anatomy, Albert Einstein College of Medicine PEARLMAN, ALAN L., Assistant Professor of Physiology and Neurology, Washington University, School of Medicine PERSON, PHILIP, Chief, Special Research Laboratory, Veterans Administration Hospital, Brooklyn, New York PIERCE, SIDNEY K., JR., Assistant Professor, University of Maryland POLITOFF, ALBERTO L., Investigator, Albert Einstein College of Medicine POLLARD, THOMAS D., Assistant Professor of Anatomy, Harvard Medical School POUSSART, DENIS J. M., Associate Professor, Universite Laval, Quebec, Canada POWERS, DENNIS A., Assistant Professor, Johns Hopkins University PRENDERGAST, ROBERT A., Associate Professor, The Johns Hopkins University School of Medicine PRINCE, JEFFREY S., Postdoctoral Investigator, Woods Hole Oceanographic Institute PROSEN, EDWARD J., Physical Chemist, National Bureau of Standards PROSSER, C. LADD, Professor of Physiology, University of Illinois PRUSCH, ROBERT D., Assistant Professor of Biology, Brown University PRZYBYLSKI, RONALD J., Associate Professor, Case Western Reserve University QUATRANO, RALPH S., Associate Professor Plant Physiology, Oregon State University RAMON, FIDEL, Research Associate, Duke University REBHUN, LIONEL L, Professor of Biology, University of Virginia REINISCH, CAROL L., Research Fellow in Medicine, Harvard University RENCRICCA, NICHOLAS J., Assistant Professor of Biology, Lowell Technological Institute REUBEN, JOHN P., Associate Professor of Physiology, Columbia University REYNOLDS, GEORGE T., Professor of Physics and Director, Center for Environmental Studies, Princeton University RICE, ROBERT V., Professor and Head, Department of Biological Sciences, Mellon Institute of Science, Carnegie-Mellon University RIPPS, HARRIS, Professor of Ophthalmology and Physiology, New York University School of Medicine ROSE, BIRGIT, Research Scientist, Physiology and Biophysics, University of Miami School of Medicine ROSE, S. MERYL, Professor, Tulane University ROSENBAUM, JOEL L., Associate Professor, Yale University ROSENBLUTH, JACK, Professor of Physiology, New York University Medical Center Ross, WILLIAM N., Research Associate, Yale University Medical School RUSHFORTH, N. B., Chairman, Biology Department and Associate Professor of Biology, Case Western Reserve University RUSSELL, JOHN M., Postdoctoral Fellow, Research Associate, Washington University School of Medicine RUSSELL-HUNTER, W. D., Professor of Zoology, Syracuse University RUSTAD, RONALD C., Associate Professor of Radiology, Anatomy and Biology, Case W'estern Reserve University SALMON, EDWARD D., Postdoctoral, Brandeis University SALZBERG, BRIAN MATTHEW, Postdoctoral Fellow in Physiology, Yale University School of Medicine SCHEIN, STANLEY JAY, Graduate Student, Albert Einstein College of Medicine REPORT OF THE DIRECTOR 27 SCHIFF, JEROME A., Professor of Biology, Brandeis University SCHOPF, THOMAS J. M., Associate Professor, University of Chicago SCHUEL, HERBERT, Associate Professor of Biochemistry, State University of New York, Down- state Medical Center SCHUETZ, ALLEN W., Associate Professor, The Johns Hopkins University SEGAL, SHELDON J., Vice President, The Population Council SENFT, JOSEPH PHILIP, Associate Professor, Juniata College SHEPHARD, DAVID C., Assistant Professor of Anatomy, Case Western Reserve University SHEPRO, DAVID, Professor, Boston University SHRIVASTAV, BRIJ BHUSHAN, Research Associate, Harvard University School of Medicine SIEGEL, IRWIN M., Associate Professor Research Ophthalmology, New York University School of Medicine SIMON, ERIC J., Professor of Experimental Medicine, New York University Medical Center SINGER, IRWIN, Assistant Professor of Medicine, Hospital of the University of Pennsylvania Sisco, KENNETH L., Staff Associate, National Institute of Mental Health SMITH, L. DENNIS, Associate Professor, Purdue University SORENSON, A. L., Assistant Professor, Brooklyn College SORENSON, MARTHA M., Postdoctoral Fellow, Columbia University, College of Physicians and Surgeons SPIEGEL, EVELYN, Research Associate, Dartmouth College SPIEGEL, MELVIN, Professor of Biology, Dartmouth College SPOON, DONALD M., Assistant Professor of Biology, Georgetown University SPRAY, DAVID C., Postdoctoral Fellow, Duke University STARZAK, MICHAEL E., Assistant Professor, State University of New York at Binghamton STEINBERG, SIDNEY, Research Associate, Columbia University STEPHENS, LEE B., Professor of Biology, California State University, Long Beach STEPHENS, RAYMOND EDWARD, Associate Professor of Biology, Brandt-is University STOKES, DARRELL R., Postdoctoral, University of California, Irvine STKACHER, ALFRED, Professor and Chairman, Department of Biochemistry, State University of New York, Downstate Medical Center STUNKARD, HORACE W., Research Associate, American Museum of Natural History, New York SZAMIER, ROBERT BRUCE, Assistant Professor, University of Texas Medical School at Houston TAKASHIMA, SHIRO, Associate Professor, Moore School of Electrical Engineering, University of Pennsylvania TALAMO, BARBARA R., Research Fellow, Harvard Medical School TASAKI, ICHIJI, Chief, Laboratory of Xeurobiology, National Institute of Mental Health TAYLOR, DOUGLASS L., Postdoctoral, State University of New York at Albany TAYLOR, WILLIAM RANDOLPH, Emeritus Professor of Botany; Curator of Algae, University of Michigan TELFER, WILLIAM H., Professor of Biology, University of Pennsylvania TILNEY, LEWIS G., Associate Professor of Biology, University of Pennsylvania TOOLE, BRYAN, Assistant Professor of Medicine, Massachusetts General Hospital TRINKAUS, JOHN PHILIP, Professor of Biology and Master of Branford College at Yale Uni- versity, Yale L^niversity TROLL, WALTER, Professor, New York University Medical School TURNER, ROBERT SCOTT, JR., Research Associate, University of Basel, Switzerland UEMURA, ISAO, Research Associate, University of South Carolina VAUGHN, JAMES M., Postdoctoral Fellow, Woods Hole Oceanographic Institution VINCENT, W. S., Professor and Chairman, University of Delaware WALD, GEORGE, Higgins Professor of Biology, Harvard University WALL, BETTY J., Research Associate, Northwestern University WALLACE, ROBIN A., Staff Member, Biology Division, Oak Ridge National Laboratory WAMPLER, JOHN E., Assistant Professor of Biochemistry, University of Georgia WARASHINA, AKIRA, Visiting Fellow, National Institute of Mental Health WATKINS, DUDLEY T., Associate Professor, University of Connecticut Health Center WEIDNER, EARL, Assistant Professor of Zoology and Physiology, Louisiana State University WEIGHT, FORREST F., Chief, Section on Synaptic Pharmacology, National Institute of Mental Health 28 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY WEINSIEDER, ALLAN, N.I.H. Postdoctoral Fellow, Oakland University WEISENBERG, RICHARD, Assistant Professor of Biology, Temple University WEISSMANN, GERALD, Professor of Medicine, New York University School of Medicine WEST, JOHN A., Associate Professor of Botany, University of California, Berkeley WETHERBEE, RICHARD, Laboratory Assistant, University of Michigan WHITTAKER, J. RICHARD, Associate Member, Wistar Institute of Anatomy and Biology WHITTAKER, VICTOR P., Direktor, Abteilung fur Neurochemie, Max-Planck Institut fur Bio- physical Chemie, Gottingen, Germany WILLIAMS, ROBLEY C., JR., Assistant Professor, Yale University WILSON, EDWARD O., Professor of Zoology, Harvard University WILSON, WALTER L., Professor, Oakland University WOLFE, RALPH S., Professor of Microbiology, University of Illinois WOOD, DONALD S., N.I.H. Trainee, Columbia University, College of Physicians and Surgeons WYNNE, MICHAEL J., Associate Professor, The University of Texas at Austin YEH, J. Z., Postdoctoral Fellow, Duke University YOSHIKAMI, Duju, Research Fellow, Harvard Medical School YOUNG, LILY Y., Assistant Professor, Stanford University YPHANTIS, DAVID A., Professor Biochemistry and Biophysics, University of Connecticut ZIGMAN, SEYMOUR, Associate Professor of Ophthalmology and Biochemistry, University of Rochester Medical Center ZOLLMAN, JENNY R., Postdoctoral Trainee Fellow, Columbia University Lillie Fellow, 1973 MARCHALONIS, JOHN J., Head, Laboratory of Molecular Immunology, \Valter and Eliza Hall Institute of Medical Research Rand Fellow, 1973 HARA, TOMIYUKI, Professor, Nara Medical University, Nara, Japan Grass Fellows, 1973 FRAZIER, DONALD T., Associate Professor, Senior Fellow, University of Kentucky BESSO, JOSEPH A., JR., N.I.H. Postdoctoral Fellow, Albert Einstein College of Medicine FUTAMACHI, KIN J., Research Associate, Stanford University KAPLAN, EHUD, Graduate Student, Syracuse University KRAUSZ, HOWARD, Neurosciences Graduate Student, University of California at San Diego LISMAN, JOHN, Grass Fellow, Harvard University SIDES, PAUL J., M.D.-Ph.D. Candidate, Duke University Medical Center TALAMO, BARBARA R., Research Fellow, Harvard Medical School WILKENS, LON A., Summer Fellowship, Grass Foundation YOUNG, WISE, Student, University of Iowa Research Assistants, 1973 ALLPORT, SUSAN W., Marine Biological Laboratory AMENSON, CHRISTOPHER SCOTT, California Institute of Technology BAKER, DIANA, University of Hawaii BASS, ANDREW, Case Western Reserve University BAUMGARTNER, CALVINA A., Johns Hopkins University BAYER, D. SCOTT, Syracuse University BEGENISICH, PEGGY ELLEN, Washington University Medical School BEGENISICH, TED, University of Maryland BELL, PAUL B., JR., Yale University BERNE, GORDON DAVID, Marine Biological Laboratory BOSLER, ROBERT B., Harvard University BRITTEN, GREGORY H., Marine Biological Laboratory BRUNHOUSE. ROBERT F., Johns Hopkins University REPORT OF THE DIRECTOR 29 BURNETT, Louis E., JR., College of William and Mary BUTLER, PRISCILLA F., Lowell Technological Institute CARTWRIGHT, JOINER, JR., Pacific Biomedical Research Center, University of Hawaii CHAIRES, JONATHAN B., University of Connecticut CHAMBERLIN, CHARLES E., Harvard University CHAMBERLIN, MARGARET, California Institute of Technology CHONG, PHILIP CHOW, University of Ottawa CLARKE, BARBARA J., Tulane University CLIPPER, JONATHAN C., York LTniversity CLOUD, JOSEPH G., University of Wisconsin COOPERSTEIN, LAWRENCE, Princeton University CORDLE, CHRISTOPHER T., The Johns Hopkins University CORNELL, JOHN C., University of California, Berkeley CORNELL, MARY, Marine Biological Laboratory CROUSE, LINDA, New York University School of Medicine DEGROOF, ROBERT C., Duke University DUVA, JOSEPH MICHAEL, Herbert H. Lehman College of the City University of New York EHRLICH, BARBARA, Brown University EICKBUSH, THOMAS H., Johns Hopkins University FELDMAN, LANCE, University of Cincinnati FIERO, ALAN, State LTniversity of New York at Albany FINGERMAN, SUE WniTSELL, Tulane University Fox, RICHARD S., Temple University Medical School GARNER, JUDY, Case Western Reserve University GOLDBERG, DANIEL J., Yale University GOLDMAN, ROSALIND, Case Western Reserve University GRANET, MICHAEL A., Johns Hopkins University GRAY, ALFRED J., JR., Northwestern University GRISSETT, J. B., Florida State University HAGEN, JOHNNY, The City College of New York HAMER, RUSSELL D., Syracuse University HAMMOND, ROBERT D., Tulane University HARRIS, EDWARD M., Duke University HEDDEN, WILLARD L., Harvard University HENDERSON, FIONA, MaxPlanck Institute for Biophysical Chemistry, Gottingen, West Germany HITCHNER, SARABELLE L, State University of New York at Albany HUDSON, ALAN P., Hunter College HUSE, ALLISON, Worcester Polytechnic Institute INOUE, CHRISTOPHER, Marine Biological Laboratory JENSEN, THOMAS J., City College of New York JUMBLATT, JAMES EDWARD, Columbia LTniversity KEEM, KIRSTEN H., Purdue University KELLER, BRIAN D., The Johns Hopkins LTniversity KIEHART, DANIEL P., University of Pennsylvania KIMURA, JOHN, Duke University KIRSTEN, MARKHAM, New York University Medical School KLAG, MICHAEL J., Juniata College KOTCHABHAKDI, NAIPHINICH, University of Illinois KRIEGSTEIN, ARNOLD, New York University Medical School LANDRIGAN, JAMES ALEXANDER, Illinois Institute of Technology LARNER, ANDREW, Haverford College LASKER, HOWARD, University of Chicago LEE, DAVID, University of Ottawa LEE, RANG, John Hopkins University LIND, STUART, New York University Medical School LOWENHAUPT, MANUEL TSING, University of Cincinnati LUNDEN, MATS, Stockholm University, Stockholm, Sweden LYNCH, CARL, III, LTniversity of Rochester, 30 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MACBRIDE, ROBERT GORDON, Case Western Reserve University MACDANIEL, ROBERT PATRICK, University of Chicago MARKOWITZ, CHARLES BERNARD, Amherst College MARSHAK, ANN, University of Pennsylvania MELLOR, KIRK, University of Miami MEYEROWITZ, ELLIOT, Columbia University MOORE, MARILYN R., University of Connecticut Health Center MOOSEKER, MARK S., University of Pennsylvania MORRISSEY, PHILIP J., JR., Lowell Technological Institute MURPHY, SUSAN A., University of Connecticut Health Center NEAL, HAROLD, Glasgow University, Scotland NEUFELD, DANIEL A., Tulane University OERTEL, DONATA, University of California, Santa Barbara OLSON, SUELLEN A., Wayne State University OSMAN, RICHARD W., University of Chicago PATLAK, JOSEPH, University of California, Los Angeles PENCEK, TERRENCE L., Illinois Institute of Technology PERSELL, ROGER, Hunter College PHILIPS, WALTER JOHN, Marine Biological Laboratory PRICE, DAVID A., Florida State University PURPURA, KEITH, Albert Einstein College of Medicine RAUBACK, SYLVIA ANN THOMPSON, University of Miami REESE, RICHARD G., University of Texas at Austin RICH, BEVERLY, Brandeis University RIGG, JANE, California Institute of Technology ROBINS, DIANE MARA, Yale University ROSENFELD, ARLiNE, Temple University SCHER, HOWARD I., New York University Medical School SCHWAB, WALTER E., University of Maryland SINGLEY, CARL T., University of Hawaii SOROKA, MICHAEL D., Case Western Reserve University SPERLING, LINDA, Harvard University SPINDEL, ROBERT A., City College STAN WOOD, EMILY A., University of Delaware STARKUS, JOHN, Duke University STERNBACH, ROBERT, Boston University STICK, THOMAS J., Johns Hopkins University Medical School STILLER, RONALD A., Boston University SULANOWSKI, JACEK, University of Chicago SZONYI, ESZTAR, Boston University TASHIRO, JAY SHIRO, Kenyon College TAYLOR, BARBARA A., Brooklyn College THOMAS, JOSEPH M., Brandeis University THOMAS, SUSAN M., Harvard Medical School TUFTS, MEMPHIS DAMON, Case Western Reserve University WASSERSTEIN, PAUL, State University of New York, Downstate Medical Center WATTS, JOHN A., University of Maryland WESTLEY, STEPHEN K., Tulane University WHATLEY, NED, Gulf Coast Research Laboratory WHITLATCH, ROBERT B., University of Chicago WINN, WILLIAM MARSHALL, Johns Hopkins University YAU, KING- WAI, Harvard University YULO, TERESA S., University of Rochester Medical Center ZAKEVICIUS, JANE M., New York University School of Medicine Library Readers, 1973 ADELBERG, EDWARD A., Professor of Human Genetics, Yale University ANDERSEN, OLAF S., Guest Investigator, The Rockefeller University REPORT OF THE DIRECTOR 31 ANDERSON, RUBERT S., Independent Library Reader, Marine Biological Laboratory ARMSTRONG, PHILIP B., Professor Emeritus, Upstate Medical Center BURNSIDE, MARY BETH, Assistant Professor, University of Pennsylvania CASS, ALBERT H., JR., Assistant Professor, Albert Einstein College CASSIDY, FR. JOSEPH D., Assistant Professor of Biology, University of Notre Dame CHILD, FRANK M., Associate Professor of Biology, Trinity College CHURNEY, LEON, Professor of Physiology, Louisiana State University Medical Center CLEMENT, ANTHONY C., Professor of Biology, Emory University CLIFFORD, SISTER ADELE, Professor of Biology, College of Mount St. Joseph on the Ohio COHEN, M. W., Assistant Professor, McGill University COHEN, SEYMOUR S., American Cancer Society, Professor of Microbiology, University of Colorado Medical Center COLLIER, J. R., Professor of Biology, Brooklyn College COL WIN, ARTHUR L., Professor, Queens College, City University of Xew York COLWIN, LAURA HUNTER, Professor, Queens College, City University of New York COPELAND, DONALD EUGENE, Professor, Tulane University DELANNEY, Louis E., Professor of Biology and Chairman, Ithaca College DIXON, KEITH, Senior Lecturer in Biology, Flinders University of South Australia EDER, HOWARD A., Professor of Medicine, Albert Einstein College of Medicine EISEN, HERMAN N., Professor and Head, Department of Microbiology, Washington University School of Medicine FEINGOLD, DAVID S., Associate Professor of Medicine, Harvard Medical School- Beth Israel Hospital GABRIEL, MORDECAI L., Dean, School of Science, Brooklyn College GERMAN, JAMES L., Senior Investigator, and Director, Laboratory of Human Genetics, The New York Blood Center, and Cornell University Medical College GINSBERG, HAROLD S., Professor and Chairman, Depart ment of Microbiology, University of Pennsylvania GRANT, PHILIP, Professor of Biology, University of Oregon GREEN, JAMES \Y., Professor of Physiology, Rutgers University GUPTA, BRIJ L., Assistant Director of Research in Zoology, Cambridge University, England HOLZ, GEORGE G., JR., Professor and Chairman, Department of Microbiology, Upstate Medical Center HUNTER, ROBERT DOUGLAS, Assistant Professor, Oakland University ILAN, JOSEPH, Associate Professor, Case Western Reserve University School of Medicine ISSELBACHER, KURT J., Mallinckrodt Professor of Medicine, Harvard Medical School and Chief, Gastronintestinal LTnit, Massachusetts General Hospital KARUSH, FRED, Professor of Microbiology, University of Pennsylvania School of Medicine KEMPTON, RUDOLF T., Professor Emeritus of Biology, Vassar College KEYNAN, ALEX, Vice President of the University, Hebrew LTniversity of Jerusalem KIRSCHENBAUM, DONALD M., Associate Professor, College of Medicine, Downstate Medical Center KOSOWER, EDWARD M., Professor of Chemistry, Tel-Aviv University KOSOWER, NECHAMA S., Associate Professor, Tel- Aviv University KUFFLER, STEPHEN WT., Robert Winthrop Professor of Neurobiology and Department Chairman, Harvard Medical School LADERMAN, AIMLEE D., Instructor, Ramapo College of New Jersey LANDSBERGER, FRANK R., Assistant Professor of Chemistry, Indiana University LAUFER, HANS, Professor of Biology, LTniversity of Connecticut LEFEVRE, PAUL G., Professor of Physiology and Biophysics, State University of New York at Stony Brook LEIGHTON, JOSEPH, Professor and Chairman, Department of Pathology, The Medical College of Pennsylvania LENARD, JOHN, Associate Professor, Rutgers Medical School LOCHHEAD, JOHN H., Professor of Zoology, University of Vermont MARSLAND, DOUGLAS A., Research Professor Emeritus, New York University MEINS, FREDERICK, JR., Assistant Professor of Biology, Princeton University MONROY, ALBERTO, Professor and Director, C.N.R. Laboratory of Molecular Embryology, Italy 32 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MORRELL, FRANK, Professor of Neurology, Rush Medical College ODELL, GARRETT M., Assistant Professor of Mathematics, Rensselaer Polytechnic Institute OSCHMAN, JAMES L., Assistant Professor, Northwestern University PALMER, JOHN D., Chairman and Professor of Biology, New York University PLOCKE, DONALD J., Associate Professor and Chairman, Boston College ROSENKRANZ, HERBERT S., Professor of Microbiology, Columbia University ROTH, JAY S., Professor of Biochemistry, University of Connecticut ROWLAND, LEWIS P., Professor and Chairman of Neurology, University of Pennsylvania RUBINOW, SOL I., Professor of Biomathematics, Cornell University Medical College SAUNDERS, JOHN W., JR., Professor of Biological Sciences, State University of New York at Albany SCHLESINGER, R. WALTER, Professor and Chairman, Department of Microbiology, Rutgers Medical School, College of Medicine and Dentistry of New Jersey SCOTT, ALLAN C., Professor, Colby College SEECOF, ROBERT, Head, Department of Developmental Biology, City of Hope National Medical Center SHEDLOVSKY, THEODORE, Professor Emeritus, Rockefeller University SHEMIN, DAVID, Professor of Biochemistry, Northwestern University SHERMAN, IRWIN W., Professor of Zoology, University of California, Riverside SIEBURTH, JOHN McN, Professor of Microbiology and Oceanography, University of Rhode Island SONNENBLICK, B. P., Professor of Zoology, Rutgers University STEINBERG, MALCOLM S., Professor, Princeton University STRITTMATTER, PHILIPP, Professor and Head, Department of Biochemistry, University of Con- necticut Health Center TANZER, MARVIN L., Associate Professor of Biochemistry, University of Connecticut Medical School TEREBEY, NICHOLAS, Assistant Professor of Anatomy, New York University TUCKER, GAIL SUSAN, Postdoctoral, Columbia University College of Physicians and Surgeons TWAROG, BETTY M., Professor, Tufts University TWEEDELL, KENYON S., Professor of Biology, University of Notre Dame WAINIO, WALTER, Professor of Biochemistry, Rutgers — The State University of New Jersey WAKSMAN, BYRON H., Professor of Microbiology, Yale University WALL, PATRICK D., Professor of Anatomy, University College, London WEBB, H. MARGUERITE, Professor and Chairman, Department of Biological Sciences, Goucher College WTEISS, LEON, Professor of Anatomy, Johns Hopkins Medical School WHEELER, GEORGE E., Professor of Biology, Brooklyn College WICHTERMAN, RALPH, Professor of Biology, Temple University WIERCINSKI, FLOYD J., Professor of Biology, Northeastern Illinois University WILSON, THOMAS HASTINGS, Professor of Physiology, Harvard Medical School WITTENBERG, JONATHAN B., Professor of Physiology, Albert Einstein College of Medicine WOLKEN, JEROME J., University Professor of Biophysics, Carnegie-Mellon University YNTEMA, CHESTER L., Professor of Anatomy, State University of New York, Upstate Medical Center Students, 1973 All students listed completed the formal course program. Asterisk indicates completing post-course research program. ECOLOGY BALCH, WILLIAM E., University of Illinois *BECVAR, JAMES E., Harvard University *BREWER, LINDA A., Stanford University CURTIS, PAUL R., Eisenhower College Fox, GEORGE E., Syracuse University *GREANEY, GEORGE S., Johns Hopkins University *GREENBERG, EVERETT P., University of Massachusetts REPORT OF THE DIRECTOR 33 HYDE, CAMILLE L., University of Michigan *KEENAN, KATHERINE, City College of New York *KiEFER, LYNDA A., University of West Florida McKAY, MELVIN G., Dalhousie University SCHWALBE, CECIL R., University of Arizona *STEIN, CAROL L., Harvard University *STOLL, DENISE R., Boston University WALTON, DIANA S., University of Illinois EMBRYOLOGY *BAKER, J. B., University of Hawaii *BECKENDORF, S. K., California Institute of Technology *CHYBA, D. E., University of Wisconsin *COTTRELL, S. F., Rutgers University *FELDHERR, C., University of Pennsylvania *GABRIELLI, F., University of Wisconsin *GALAU, G. A., California Institute of Technology *HALL, L., University of British Columbia *HOBART, P. M., Wesleyan University *HUNSLEY, J. R., Michigan State University *KAHN, E. B., Northwestern University *LooMis, M. R., Indiana University *NEWMAN, S. A., University of Chicago *NORMAN, J. A., University of California *OSTRER, H., Columbia University *PINON, R., University of Washington *SNYDER, B. W., University of Michigan *STEEL, L. F., Cornell LJniversity *WYMAN, A. R., Harvard Medical School *ZUNIGA, M., Yale University EXPERIMENTAL BOTANY AGNES, SISTER CECILIA, Regis College AVERBACK, RAE RACHAEL, Bard College BRA\VLEY, SUSAN HOWARD, Wellesley College DEAN, JOHN M., University of Massachusetts DEHN, PAULA FAYE, DePauw University FULTON, ALLICE BORDWELL, Brown University LEVANDOSKI, DENNIS, Brooklyn College LUNN, PATRICIA ANN, Drew University POSTELLE, REBEKAH ANN, Tennessee Wesleyan College RILEY, WILLIAM DENIS, University of Massachusetts, Boston SCOTT, GORDON, University of Massachusetts STEAD, RICHARD B., University of Wisconsin STRAUS, JOHN WILLIAM, Earlham College WILLIAMS, DEBORAH W., Harvard University, Radcliffe College YEN, DUENYING, JENNETTE, Bryn Mawr College PHYSIOLOGY *BAILEY, STEPHEN, Washington University BARNITZ, JOY, New College BUBBERS, J. ERIC, Johns Hopkins University *BuRNS, ROY G., University of Edinburgh COHEN, ROBERT M., Amherst College DAVIS, FRANCIS M., Yale University 34 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY *DEAN, PAULA, Johns Hopkins University DsMoss, Harvard Medical School DODGE, ROBERT H., Texas A & M University *FALLON, ANN M., Yale University *FINDLEY, ROBERT, Yale University FITZGERALD, GREGORY J., University of Dayton FUJIWARA, KEIGI, University of Pennsylvania *GRAINGER, JAMES L., University of California, Santa Cruz HACKETT, CHARLES J., Wayne State University HASKELL, AMY, College of William and Mary HOFF, STEVEN, F., University of California, Los Angeles HOGG, JAMES C., McGill University *HOPPENSTEADT, FRANK, New York University *HUNT, RICHARD K., University of Pennsylvania JOHMANN, CAROL A., University of Rochester KAM, Zvi, University of California, San Diego KIRBY, CHRISTINE A., University of Pittsburgh KLEYN, JOHN G., University of Puget Sound KUPPERS, RUDOLF C., University of California, Northridge LITCHFIELD, WILLIAM J., University of Illinois *Lo\'GREN, TIMO N-E., University of New Mexico MADISON, CAROLYN R., Newark State College *McCuMBER, LARRY JOE, University of Florida MEDNIEKS, MAIJA I., DePaul University MILLER, STELLA C., Johns Hopkins University *PARKER, WENDY L., City College of New York REIMER, ERNEST M., Simon Fraser University RIBOLINI, ANN, State University of New York at Stony Brook *RiCE, CHARLES M., University of California, Davis *ROTHBLUM, LAWRENCE I., Hahnemann Medical College RUTZ, RICHARD E., JR., Michigan State University *SANGER, JOSEPH W., University of Pennsylvania *SCHACHER, SAMUEL M., Columbia University *TUAN, ROCKY S-c., Rockefeller University *WEISS, MARTIN B., University of Massachusetts, Amherst INVERTEBRATE ZOOLOGY BARTBERGER, CAROL A., University of Maryland BOLLEN, ANDREW W., Harvard College *BOUTON, REBECCA K., Florida Atlantic University BURSTEIN, NEAL, Stanford University *COLLINS, JAMES, Marquette University CRABTREE, ROBERT L., Ohio University DAY, DEBORAH, Universite' Aix-en- Provence, Marseille DONOHUE, NEIL, Drew University DWYER, ELLEN, Wilson College *£PSTEIN, KERRY, University of Oregon *FowLER, VELIA, Oberlin College FREADMAN, MARVIN, College of William and Mary FRIEDLANDER, MICHAEL, University of Illinois *GIRSCH, STEVE, Harvard University HALL, KATHLEEN, University of Minnesota HERBST, GARY N., University of Wisconsin JENKINS, CAROLE, University of Pittsburgh KEMP, WILLIAM M., University of Florida LEONARD, STEPHANIE, Oakland University REPORT OF THE DIRECTOR 35 LING, TSIN YUEN JEANNE, University of Maryland *MARTINI, MARY, University of California, Santa Barbara *MYKLES, DONALD L., University of California, Berkeley NATHN, SHARON, Simmons College OSBON, MARIA, California State University, San Francisco PRICE, CHRISTOPHER H., Syracuse University RADERMAN, RANDIE, State University of New York *REINGOLD, STEPHEN C, Cornell University *STOLPER, DAVID, Upstate Medical Center *S\VENSON, RANDOLPH P., Washington State University WHEELER, PATRICIA, University of California, Irvine ZEMKE, STEPHEN, Wesleyan University NEUROBIOLOGY BOHR, VILHEIM ALFRED, University of Copenhagen CALDWELL, JOHN HENDERSON, Washington University COUNTER, S. ALLEN, Harvard University EATON, ROBERT CHARLES, University of California FELDMAN, JACK L., University of Chicago FORSYTHE, DAVID WILLIAM, Princeton University HEYER, CAROLYN BAKER, University of Iowa KIMBLE, JUDITH E., University of Copenhagen KOVAC, MARC P., Stanford University POLLARD, HARVEY B., National Institutes of Health SAMPLES, JOHN RANDALL, Stanford University SRINIVASAN, MANDYAM VEERAMBUDI, Yale University FRONTIERS IN RESEARCH AND TEACHING PROGRAM CLARK, VERNON, North Carolina Central University STEPHENS, LEE B., California State University 4. FELLOWSHIPS AND SCHOLARSHIPS, 1973 Bio Club Scholarship: KATHERINE KEENAN, Ecology Course WENDY L. PARKER, Physiology Course Gary H. Calkins Scholarship: REBECCA K. BOUTON, Invertebrate Zoology Course CAROLE JENKINS, Invertebrate Zoology Course MICHAEL FRIEDLANDER, Invertebrate Zoology Course RANDIE RADERMAN, Invertebrate Zoology Course PATRICIA WHEELER, Invertebrate Zoology Course 5. TRAINING PROGRAM FERTILIZATION AND GAMETE PHYSIOLOGY RESEARCH TRAINING PROGRAM I. INSTRUCTORS CHARLES B. METZ, University of Miami, Program Director BJORN AFZELIUS, University of Stockholm, Sweden TOM HUMPHREYS, University of Hawaii JOE ROSENBAUM, Yale University WILLIAM TELFER, University of Pennsylvania ROBIN WALLACE, Oak Ridge National Laboratory 36 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY II. ASSISTANTS DIANE ROBINS, EM Assistant SYLVIA ANN THOMPSON RAUBACK, Program Secretary ROBERT SPINDELL, Photographic Assistant III. TRAINEES LESTER I. BINDER, Vale University MARGUERITE F. KNEBEL BOHM, University of Kansas DONALD C. CHANG, Rice University CHRISTOPHER T. CORDLE, Johns Hopkins University GLENN E. DANIELS, University of Pennsylvania MARK N. HILL, Miami University PAUL A. LEFEBVRE, Rockefeller University ANTHONY L. MESCHER, Ohio State FREDERICK W. MILLER, Miami University PHOEBE MOUNTS, University of Pittsburgh PAUL A. ROSENBERG, Albert Einstein Medical School BARRY W. SIMON, Tulane University ROGER D. SLOBODA, Rensselaer Polytechnic Institute ANDREA J. TENNER, University of California, San Diego DONNA L. VOGEL, Albert Einstein Medical School GEORGE B. WITMAN III., University of Chicago IV. LECTURES CHARLES METZ Mammalian sperm antigens and their roles in reproduction MARJORIE CRANDALL Molecular complementary of yeast glycoprotein mating factors BJORN AFZELIUS Sperm morphology and fertilization biology HERBERT SCHUEL Molecular constituents of egg cortical granules. Function of tryptic proteinase in fertilization JOEL ROSENBAUM Flagellar regeneration in Chlam ydomonas : A system for studying the in vivo synthesis and assembly of microtubule GEORGE WITMAN Directionality of microtubule assembly during flagellar regeneration in Chlamydomonas LESTER BINDER Directionality of microtubule assembly in vitro TOM HUMPHREYS Ribosomal and DNA-like RNA metabolism in sea urchin embryos SHELDON J. SEGAL Specific and heterospecific transfer of hormone action by RNA MAREO BURGOS Maturation of the mammalian spermatozoon EXCITABLE MEMBRANE PHYSIOLOGY AND BIOPHYSICS TRAINING PROGRAM I. CONSULTANTS \Y. J. ADELMAN, JR., National Institute of Neurological Diseases and Stroke K. S. COLE, National Institute of Neurological Diseases and Stroke J. W. MOORE, Duke University Medical Center L. J. MULLINS, University of Maryland, School of Medicine II. INSTRUCTORS D. E. GOLDMAN, Program Director, Medical College of Pennsylvania F. A. DODGE, Associate Director, IBM Corp. K. S. COLE, National Institute of Neurological Diseases and Stroke J. W. MOORE, Duke University Medical Center D. GILBERT, National Institute of Neurological Diseases and Stroke L. J. MULLINS, University of Maryland, School of Medicine M. P. BLAUSTEIN, Washington University REPORT OF THE DIRECTOR 37 W. K. CHANDLER, Yale University School of Medicine W. J. ADELMAN, JR., National Institute of Neurological Diseases and Stroke T. NARAHASHI, Duke University Medical Center E. S. MASORO, Medical College of Pennsylvania M. M. DEWEY, State University of New York at Stony Brook W. K. BLASIE, University of Pennsylvania H. LECAR, National Institute of Neurological Diseases and Stroke L. E. MOORE, Case Western Reserve University C. L. PROSSER, University of Illinois C. M. ARMSTRONG, University of Rochester R. DE HAAN, Carnegie Institution of Washington J. TASAKI, National Institute of Mental Health L. COHEN, Yale University School of Medicine H. FISHMAN, State University of New York at Albany J. Y. LKTTVIN, Massachusetts Institute of Technology R. S. MANALIS, University of Cincinnati III. ASSISTANTS P. GREIF, Haverford College G. STETTEN, Harvard College IV. TRAINEES R. F. ABERCROMBIE, University of Maryland I). M. EASTON, Florida State University W. F. GILLY, Washington University R. S. KASS, University of Michigan R. D. NATHAN, University of California, Los Angeles T. W. PEARSON, University of California, Davis J. F. PODUSLO, University of Pennsylvania J. M. SCHAEFFER, Purdue University E. E. SWENBERG, Brooklyn College R. E. YANTORNO, University of Pennsylvania V. LECTURES D. E. GOLDMAN Physical and electrochemistry of membranes I. Physical and electrochemistry of membranes 1 1. Physical and electrochemistry of membranes III. K. S. COLE Electrical properties of membranes The strategy of the voltage clamp J. \V. MOORE Operational amplifiers Preamplifiers Voltage clamp circuits D. GILBERT The biology of squid L. J. MULLINS Passive transport M. P. BLAUSTEIN Active transport F. A. DODGE Cable theory W. K. CHANDLER Voltage clamp currents in squid The Hodgkin-Huxley formulation Univalent ions and selectivity D. E. GOLDMAN Divalent cations W. J. ADELMAN, JR. The periaxonal space L. E. MOORE Myelinated nerve J. W. MOORE TTX and excitable membranes W. J. MASORO Chemical composition of membranes M. M. DEWEY Electron microscopy and membrane structure 38 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY H. LECAR W. K. BLASIE T. NARAHASHI D. E. GOLDMAN C. M. ARMSTRONG W. K. CHANDLER C. L. PROSSER R. DE HAAN D. E. GOLDMAN I. TASAKI L. COHEN H. FISHMAN J. Y. LETTVIN Artificial membranes X-ray diffraction of membranes Action of membrane "stabilizers" Action of membrane "labilizers" Excitability models I. Excitability models II. Excitability models III. Models for Na channels Models for K channels Striated muscle I. Striated muscle II. Striated muscle III. Smooth muscle Electrical activity and entrainment of heart cells in culture Surface charges in axon membranes Non-electrical stimulation of axons Axon perfusion and membrane behavior Optical properties of axons Noise in the squid axon The relation between semi-conductors and electrolytes with respect to membranes RESEARCH PROGRAM IN EXPERIMENTAL MARINE BOTANY I. SENIOR INVESTIGATORS MARTIN GIBBS, Brandeis University FRANK A. LOEWUS, State University of New York at Buffalo RALPH S. QUATRANO, Oregon State University JEROME A. SCHIFF, Brandeis University MICHAEL J. WYNNE, University of Texas, Austin II. ASSOCIATE INVESTIGATORS HANS GAFFRON, Florida State University ROBERT C. HODSON, University of Delaware MARY W. LOEWUS, State University of New York at Buffalo III. JUNIOR INVESTIGATORS AMITZUR BEN-AMOTZ, Weizmann Institute of Science, Israel SCOTT E. BINGHAM, University of Rhode Island JANICE ANN DERR, Washington University, St. Louis FRANKLIN FONG, University of California at Riverside ARTHUR C. LEY, Scripps Institute of Oceanography, San Diego DWIGHT G. PEAVEY, Brandeis University, Waltham, Massachusetts DENNIS SHEVLIN, Duke University, Durham, North Carolina STUART G. SIDDELL, University of Warwick, England NICOLE LEVEQUE, Pasteur Institute, Paris, France DAVID N. YOUNG, University of California at Berkeley IV. LECTURES D CHENEY N. BISHOP E. McCANDLESS D. A. WALKER A. TREBST Current research in Eucheuma, a carrageenan producer from the south Vitamin E and photosynthesis Biochemical consequences of the sex life of Chondrus Some aspects of photosynthetic induction Energy conservation in photosynthetic electron flow REPORT OF THE DIRECTOR 39 6. TABULAR VIEW OF ATTENDANCE, 1969-1973 1969 1970 1971 INVESTIGATORS — TOTAL 566 Independent 310 Library Reader 68 Research Assistants. . 188 STUDENTS — -TOTAL Invertebrate Zoology. Embryology Physiology Experimental Botany. Ecology TRAINEES — -TOTAL . TOTAL ATTENDANCE Less Persons represented in two categories. INSTITUTIONS REPRESENTED — -TOTAL. FOREIGN INSTITUTIONS REPRESENTED. 118 35 20 30 16 17 29 713 5 708 187 24 532 324 73 135 142 41 28 31 19 23 33 707 0 707 191 21 554 322 76 156 130 29 28 33 22 18 44 728 0 728 219 27 1972 561 328 76 157 119 38 19 31 14 17 46 726 1 725 210 25 1973 523 312 86 125 123 32 20 41 15 15 50 696 0 696 239 40 7. INSTITUTIONS REPRESENTED, 1973 Alabama, University of Albert Einstein College of Medicine American Museum of Natural History Amherst College Arizona, University of Bard College Beth Israel Hospital Boston City Hospital Boston College Boston University Boston University School of Medicine Brandeis University Brooklyn College, The City University of New York Brown University Bryn Mawr College California, University of, Berkeley California, University of, Davis California, University of, Irvine California, University of, Los Angeles California, University of, Riverside California, University of, San Diego California, University of, San Francisco California, University of, Santa Barbara California, University of, Santa Cruz California Institute of Technology California State University California State University, Long Beach Carnegie Institution of Washington Carnegie-Mellon University Case Western Reserve University Chicago, University of Cincinnati, University of Citadel, The City College, The City University of New York City of Hope National Medical Center Colby College College of Mount St. Joseph on the Ohio College of William and Mary Colorado, University of Colorado, University of, Medical Center Columbia University Columbia University, College of Physicians and Surgeons Connecticut, University of Connecticut, University of, Health Center Connecticut, University of, Medical School Cornell University Cornell University Medical College Dartmouth College Dayton, University of Delaware, University of Delaware State College De Paul University De Pauw University Drew University Duke University Duke University Medical Center Earlham College Eisenhower College 40 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY Emory University Florida, University of Florida Atlantic University Florida State University Georgetown University Georgia, University of Goucher College Gulf Coast Research Laboratory Hahnemann Medical College Hartwick College Harvard Medical School Harvard University Haverford College Hawaii, University of Herbert Lehman College, The City University of Xew York Houston, University of Howard University Hunter College, The City University of Xew York Illinois, University of Illinois Institute of Technology Indiana University Institute for Cancer Research, The Institute for Muscle Research Iowa, University of Ithaca College Johns Hopkins University, The Johns Hopkins University, The, School of Hygiene Johns Hopkins University, The, School of Medicine Juniata College Kansas, University of Kansas State University Kentucky, University of Kenyon College Louisiana State University Louisiana State University Medical Center Lowell Technological Institute Marquctte University Marine Research Foundation, Inc. Maryland, University of Maryland, University of, School of Medicine Massachusetts, University of Massachusetts General Hospital Massachusetts Institute of Technology Medical College of Ohio at Toledo Medical College of Pennsylvania Mellon Institute of the Carnegie-Mellon University Miami, University of Miami, University of, School of Medicine Michigan, University of Michigan State University Middlebury College Minnesota, University of National Bureau of Standards National Institute of Mental Health National Institute of Neurological Diseases and Stroke National Institutes of Health New College New Hampshire, University of New Mexico, University of New York Blood Center, The New York University New York University Medical Center New York University School of Medicine Newark State College North Carolina, University of, Chapel Hill North Carolina Central University North Carolina State University, Raleigh Northeastern Illinois University Northwestern University Notre Dame, University of Oak Ridge National Laboratory Oakland University Oberlin College Ohio State University Ohio University Oregon, University of Oregon State University Pennsylvania, University of Pennsylvania, Hospital of the University of Pennsylvania, University of, School of Medicine Pittsburgh, University of Population Council, The Portland State University Princeton University Puget Sound, University of Purdue University Queens College, The City University of New ~ York Radcliffe College Ramapo College of New Jersey Regis College Rensselaer Polytechnic Institute Rhode Island, University of Rice University Rochester, University of Rochester, University of, Medical School Rockefeller University, The Rush Medical College Rutgers — The State University Rutgers University Medical School Scripp's Institution ot Oceanography Simmons College Stanford University State University of New York, Downstate Medical Center State University of New Y'ork, Upstate Medical Center State University of New York at Albany State University of New York at Binghamton REPORT OF THE DIRECTOR 41 State University of New York at Buffalo State University of New York at Stony Brook Syracuse University Temple University Temple University Health Sciences Center Temple University Medical School Tennessee Medical Units, University of Tennessee Wesleyan College Texas A & M University Texas Christian University Texas, University of, Arlington Texas, University of, Austin Trinity College Tulane University Upsala College Yanderbilt University Yassar College Yeterans Administration Hospital, Brooklyn Virginia, University of Washington University Washington University School of Medicine Wayne State University Wellesley College Wesleyan University West Florida, University of Western Kentucky University Wilson College Wisconsin, University of Wistar Institute Woods Hole Oceanographic Institution Worcester Polytechnic Institute Yale University Yale University School of Medicine FOREIGN INSTITUTIONS REPRESENTED, 1973 Basel, The University of, Switzerland British Columbia, University of, Canada Calgary, University of, Canada Cambridge, University of, England CNR Laboratory of Molecular Embryology, Italy Copenhagen, University of, Denmark Dulhousie University, Canada Edinburgh University Scotland Flinders University of South Australia, Australia Glasgow University, Scotland Hebrew University Medical School, Israel Jichi Medical School, Japan Karolinska Institute, Sweden Leeds, University of, England London University, England Max Planck Institute for Biophysical Chem- istry, West Germany McGill University, Canada McMaster University, Canada Medical Research Council, Toxicology Re- search Unit, England Montreal, University of, Canada Mtinster, University of, Germany Nagoya City University, Japan Nara Medical University, Japan Ottawa, University of, Canada Pasteur Institute, France Queen Elizabeth College, England Queen Mary College, England Simon Fraser University, Canada Stockholm University, Sweden Tel-Aviv University, Israel Tokyo Kyoiku University, Japan Toronto, University of Canada Universita degli Studi di Milano, Italy Universite Laval, Canada University College, England Walter and Eliza Hall Institute of Medical Research, Australia Warwick, University of, England Weizmann, Institute of Science, Israel Wenner-Gren Institute, Sweden York University, England June 28 8. FRIDAY EVENING LECTURES, 1973 SAUL ROSEMAN Sugar transport in bacteria The Johns Hopkins University July 6 JOSEPH G. GALL. Yale University Repetitive DNA in Drosophila 42 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY July 13 KEITH R. PORTER Studies on cell topography by scanning micros- University of Colorado copy July 19 PATRICK D. WALL Somatic sensory mechanisms in relation to University College, London sensation Alexander Forbes Lecturer at MBL July 20 PATRICK D. WALL Pain mechanisms as a special example of somato- sensory mechanisms July 27 SHELDON J. SEGAL The population problem : a progress report The Population Council August 3 JOHN B. BUCK The seventh sense: rhythm and synchrony in National Institutes of Health fireflies and man August 10 ANNEMARIE WEBER Information transfer in the actin filament University of Pennsylvania, Medical School August 17 JEAN G. BAER Forging the first link University of Neuchatel Agassiz Memorial Lecturer August 24 M. S. BRETSCHER How are membranes put together ? Medical Research Council, Cambridge 9. MEMBERS OF THE CORPORATION, 1973 Including Action of 1973 Annual Meeting Life Members ADOLPH, DR. EDWARD F., University of Rochester School of Medicine and Den- tistry, Rochester, New York 14627 BEAM, DR. HAROLD W., Department of Zoology, State University of Iowa, Iowa City, Iowa 52240 BEHRE, DR. ELINOR M., Black Mountain, North Carolina 28711 REPORT OF THE DIRECTOR 43 BERTHOLF, DR. LLOYD M., 1228 Gettysburg Drive, Bloomington, Illinois 61701 BODANSKY, DR. OSCAR, Department of Biochemistry, Memorial Cancer Center, 444 East 68 Street, New York, New York 10021 BRADLEY, DR. HAROLD C., 2639 Durant Avenue, Berkeley, California 94704 BROWN, DR. DUGALD E. S., Cape Haze, Box 426, Placida, Florida 33946 BURDICK, DR. C. LALOR, The Lalor Foundation, 4400 Lancaster Pike, Wilming- ton, Delaware 19805 COLE, DR. ELBERT C., 2 Chipman Park, Middlebury, Vermont 05753 COWDRY, DR. E. V., 4580 Scott Avenue, St. Louis, Missouri 63110 DILLER, DR. IRENE C., 2417 Fairhill Avenue, Glenside, Pennsylvania 19038 DILLER, DR. WILLIAM F., 2417 Fairhill Avenue, Glenside, Pennsylvania 19038 FERGUSON, DR., JAMES K. W., 56 Clarkshaven St., Thornhill, Ontario, Canada FURTH, DR. JACOB, 99 Fort Washington Ave., New York, New York 10032 GALTSOFF, DR. PAUL, National Marine Fisheries Service, Woods Hole, Massa- chusetts 02543 GRAY, DR. IRVING E., Department of Zoology, Duke University, Durham, North Carolina 27701 HAMBURGER, DR. VIKTOR, Department of Zoology, Washington University, St. Louis, Missouri 63110 HESS, DR. WALTER, 787 Maple Street, Spartanburg, South Carolina 29302 HIBBARD, DR. HOPE, 366 Reamer Place, Oberlin, Ohio 44074 HISAW, DR. F. L., 5925 S. W. Plymouth Drive, Corvallis, Oregon 97330 HOADLEY, DR. LEIGH, 985 Memorial Drive, Cambridge, Massachusetts 02138 HOLLAENDER, DR. ALEXANDER, Biology Division, Oak Ridge National Labora- tory, Oak Ridge, Tennessee 37830 IRVING, DR. LAURENCE, University of Alaska, College, Alaska 99701 KAAN, DR. HELEN, P. O. Box 665, W'oods Hole, Massachusetts 02543 KAHLER, ROBERT, Box 423, Woods Hole, Massachusetts 02543 LYNN, DR. W. GARDNER, Department of Biology, Catholic University of America, Washington, D. C. 20017 MACDOUGALL, DR. MARY STUART, Mt. Vernon Apartments, 423 Clairmont Avenue, Decatur, Georgia 30030 MAGRUDER, DR. SAMUEL R., Rte 4, Box 177, Kevil, Kentucky 42053 MALONE, DR. E. F., 6610 North llth Street, Philadelphia, Pennsylvania 19126 MANWELL, DR. REGINALD D., Department of Biology, Syracuse University, Syracuse, New York 13210 MARSLAND, DR. DOUGLAS, 48 Church St., Woods Hole, Massachusetts 02543 MILLER, DR. JAMES A., Department of Anatomy, Tulane University, New Orleans, Louisiana 70112 PAGE, DR. I. H., Cleveland Clinic, Euclid at E. 93rd Street, Cleveland, Ohio 44106 PAYNE, DR. FERNANDUS, Wesley Manor, 1555 N. Main St., Frankfort, Indiana 46041 PLOUGH, DR. H. H., 15 Middle Street, Rt. 1, Amherst, Massachusetts 01002 POLLISTER, DR. A. W., Department of Zoology, Columbia University, New York, New York 10027 POND, SAMUEL E., 53 Alexander Street, Manchester, Connecticut 06044 SCHMITT, DR. FRANCIS O., 165 Allen Dale St., Jamaica Plain, Massachusetts 02130 SCHRADER, DR. SALLY, Duke University, Durham, North Carolina 27706 44 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY SCHRAMM, DR. J. R., Department of Plant Sciences, Indiana University, Bloom- ington, Indiana 47401 SEVERINGHAUS, DR. AURA E., 375 West 250th Street, New York, New York 10071 SMITH, DR. DIETRICH C., 216 Oak Forest Ave., Catonsville, Maryland 12128 SPEIDEL, DR. CARL C., 1873 Field Road, Charlottesville, Virginia 22903 STRAUS, DR. W. L., JR., Department of Anatomy, The Johns Hopkins University Medical School, Baltimore, Maryland 21205 STUNKARD, DR. HORACE W., American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024 TAYLOR, DR. WM. RANDOLPH, Department of Botany, University of Michigan, Ann Arbor, Michigan 48104 TEWINKEL, DR. Lois E., 4 Sanderson Ave., Northampton, Massachusetts 01060 TURNER, DR. C. L., Northwestern University, Evanston, Illinois 60201 WARREN, DR. FIERBERT S., % Leland C. Warren, 721 Conshohocken State Road, Penn Valley, Pennsylvania 19072 WEISS, DR. PAUL, The Rockefeller University, New York, New York, 10016 YOUNG, DR. D. B., Main Street, North Hanover, Massachusetts 02357 Regular Members ABBOTT, DR. BERNARD C., Department of Biological Sciences, University of Southern California, University Park, Los Angeles, California 90007 ABBOTT, DR. MARIE B., Resident Systematist, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 ACHE, DR. BARRY W., Department of Biological Sciences, Florida Atlantic University, Boca Raton, Florida 33432 ACHESON, DR. GEORGE H., Department of Pharmacology and Therapeutics, University of Cincinnati, College of Medicine, Eden and Bethesda Avenues, Cincinnati, Ohio 45219 ADELBERG, DR. EDWARD A., Department of Microbiology, Yale University Medical School, New Haven, Connecticut 06510 ADELMAN, DR. WM. J., JR., Building 36 — Room 2A 31, National Institutes of Health, Bethesda, Maryland 20014 AFZELIUS, DR. BJORN, Wenner-Gren Institute, University of Stockholm, Stock- holm, Sweden ALLEN, DR. GARLAND E., Biology Department, Washington University, St. Louis, Missouri 63110 ALLEN, DR. ROBERT D., Department of Biological Sciences, State University of New York at Albany, Albany, New York 12203 ALSCHER, DR. RUTH, Department of Biology, Manhattanville College, Purchase, New York 10577 AMATNIEK, ERNEST, 154 Bay Road, Huntington, New York 11743 AMBERSON, DR. WILLIAM R., Katy Hatch Road, Falmouth, Massachusetts 02540 ANDERSON, DR. EVERETT, Department of Anatomy and Laboratories of Human Reproductive Biology, Harvard Medical School, Boston, Massachusetts 02115 ANDERSON, DR. J. .\L, Division of Biological Sciences, Emerson Hall, Cornell University, Ithaca, New York 14850 REPORT OF THE DIRECTOR 45 ANDERSON, DR. RUBEKT S., Box 113, Woods Hole, Massachusetts 02543 ARMSTRONG, DR. CLAY Al., Department of Physiology, University of Rochester, Rochester, New York 14603 ARMSTRONG, DR. PHILIP B., Department of Anatomy, State University of New York, College of Aledicine, Syracuse, New York 13210 ARNOLD, DR. JOHN A^ILLER, Pacific Biomedical Research Center, 2538 The Mall, University of Hawaii, Honolulu, Hawaii 96822 ARNOLD, DR. WILLIAM A., Division of Biology, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830 ATWOOD, DR. KIMBALL C, 560 Riverside Drive, Apt 11L, New York, New York 10027 AUCLAIR, DR. WALTER, Department of Biology, Rensselaer Polytechnic Institute, Troy, New York 12181 AUSTIN, DR. COLIN RUSSELL, Physiological Laboratory, Cambridge University, Downing Street, Cambridge, England, U. K. AUSTIN, DR. MARY L., 506^ North Indiana Avenue, Bloomington, Indiana 47401 BACON, ROBERT, Church Street, Woods Hole, Massachusetts 02543 BAKALAR, DAVID, 35 Lapland Road, Chestnut Hill, Massachusetts 02167 BALL, DR. ERIC G., P. O. Box 406, Falmouth, Massachusetts 02541 BANG, DR. F. B., Department of Pathobiology, The Johns Hopkins University School of Hygiene, Baltimore, Maryland 21205 BARD, DR. PHILLIP, Department of Physiology, The Johns Hopkins University Aledical School, Baltimore, Maryland 21205 BARLOW, DR. ROBERT B., JR., Laboratory of Sensory Communication, Syracuse University, 821 University Avenue, Syracuse, New York 13210 BARTELL, DR. CLELMER K., Department of Biological Sciences, Louisiana State University of New Orleans, New Orleans, Louisiana 70113 BARTH, DR. LESTER G., Alarine Biological Laboratory, Woods Hole, Alassa- chusetts 02543 BARTH, DR. LUCENA, Marine Biological Laboratory, Woods Hole, Alassachusetts 02543 BARTLETT, DR. JAMES H., Department of Physics, University of Alabama, P.O. Box 1921, University, Alabama 35486 BAUER, DR. G. ERIC, Department of Anatomy, University of Alinnesota, Alinne- apolis, Alinnesota 55414 BECK, DR. L. V., Department of Pharmacology, Indiana University, School of Experimental Aledicine, Bloomington, Indiana 47401 BELAMARICH, DR. FRANK A., Department of Biology, Boston University, Boston, Massachusetts 02215 BELL, DR. ALLEN, RFD #1, Cambridge, Maine 04923 BELL, DR. EUGENE, Department of Biology, Alassachusetts Institute of Tech- nology, Cambridge, Massachusetts 02139 BENNETT, DR. MICHAEL V. L., Department of Anatomy, Albert Einstein College of Aledicine, Bronx, New York 10461 BENNETT, DR. MIRIAM F. Department of Biology, Colby College, Waterville, Maine 04901 BERMAN, DR. AIoNES, National Institutes of Health, Institute for Arthritis and Aletabolic Diseases, Bethesda, Alaryland 20014 46 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY BERNE, DR. ROBERT M., University of Virginia School of Medicine, Charlottes- ville, Virginia 22903 BERNHEIMER, DR. ALAN W., New York University College of Medicine, New York, New York 10016 BIGGERS, DR. JOHN DENNIS, Department of Physiology, Harvard Medical School, 25 Shattuck St., Boston, Massachusetts 02115 BISHOP, DR. DAVID W., Medical College of Ohio at Toledo, P.O. Box 6190, Toledo, Ohio 43614 BLANCHARD, DR. K. C., The Johns Hopkins University Medical School, Balti- more, Maryland 21205 BLOCK, DR. ROBERT, Adalbertstr. 70-8, Munich, Germany (13) BLUM, DR. HAROLD F., Department of Biological Sciences, State University of New York at Albany, Albany, New York 12203 BODIAN, DR. DAVID, Department of Anatomy, The Johns Hopkins University, 709 North Wolfe Street, Baltimore, Maryland 21205 BOETTIGER, DR. EDWARD G., Department of Zoology, University of Connecticut, Storrs, Connecticut 06268 BOLD, DR. HAROLD C., Department of Botany, University of Texas, Austin, Texas 78712 BOOLOOTIAN, DR. RICHARD A., Visual Science Productions, Box 24787, Westwood Village, Los Angeles, California 90024 BOREI, DR. HANS G., Leidy Laboratory, Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 BORGESE, DR. THOMAS A., Department of Biology, Lehman College, City Uni- versity of New York, Bronx, New York 10468 BORISY, DR. GARY G., Laboratory of Molecular Biology, University of Wis- consin, Madison, Wisconsin 53715 BORSELLINO, DR. ANTONIO, Institute di Fiscia, Viale Benedetto XV, 5 Geneva, Italy BOSCH, DR. HERMAN F., Marine Biological Laboratory, Woods Hole, Massa- chusetts 02543 BOWEN, DR. VAUGHN T., Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 BOWLES, DR. FRANCIS P., Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 BRANDT, DR. PHILIP WILLIAMS, Department of Anatomy, Columbia University, College of Physicians and Surgeons, New York, New York 10032 BRIDGMAN, DR. ANNA J., Department of Biology, Agnes Scott College, Decatur, Georgia 30030 BRINLEY, DR. F. J., JR., Department of Physiology, The Johns Hopkins Uni- versity Medical School, Baltimore, Maryland 21205 BRONK, DR. DETLEV W., The Rockefeller University, 66th Street and York Avenue, New York, New York 10021 BROOKS, DR. MATILDA M., Department of Physiology, University of California, Berkeley, California 94720 BROWN, DR. FRANK A., JR., Department of Biological Sciences, Northwestern University, Evanston, Illinois 60201 REPORT OF THE DIRECTOR 47 BROWN, DR. JOEL E., Department of Anatomy, School of Medicine, Vanderbilt University, Nashville, Tennessee 37203 BUCK, DR. JOHN B., Laboratory of Physical Biology, National Institutes of Health, Bethesda, Maryland 20014 BURBANCK, DR. MADELINE PALMER, Box 15134, Emory University, Atlanta, Georgia 30322 BURBANCK, DR. WILLIAM D., Box 15134, Emory University, Atlanta, Georgia 30322 BURDICK, DR. CAROLYN J., Department of Biology, Brooklyn College, Brooklyn, New York 11210 BURGER, DR. MAX M., Department of Biochemistry, University of Basel, CH. 4056-Klingelbergstrasse 70, Basel, Switzerland BURKY, DR. ALBERT J., Department of Biology, University of Dayton, Dayton, Ohio 45401 BURNETT, DR. ALLISON LEE, Department of Biology, Northwestern University, Evanston, Illinois 60201 CARLSON, DR. FRANCIS D., Department of Biophysics, The Johns Hopkins University, Baltimore, Maryland 21218 CARPENTER, DR. RUSSELL L., 60-H Lake Street, Winchester, Massachusetts 01890 CARRIKER, DR. MELBOURNE R., College of Marine Studies, University of Dela- ware, Field Station, Lewes, Delaware 19958 CASE, DR. JAMES F., Department of Biological Sciences, University of California, Santa Barbara, California 93106 CASS, DR. ALBERT H., JR., Department of Anatomy, Albert Einstein College of Medicine, Bronx, New York 10461 CASSIDY, REV. JOSEPH D., O.P., Department of Biology, University of Notre Dame, Notre Dame, Indiana 46556 CATTELL, DR. McKEEN, Cornell University Medical College, 1300 York Avenue, New York, New York 10021 CEBRA, DR. JOHN J., Department of Biology, Johns Hopkins University, Balti- more, Maryland 21218 CHAET, DR. ALFRED B., University of West Florida, Pensacola, Florida 32505 CHAMBERS, EDWARD L., Department of Physics and Biophysics, LIniversity of Miami School of Medicine, P.O. Box 520875, Biscayne Annex, Miami, Florida 33146 CHAPPELL, DR. RICHARD L., Department of Biological Sciences, Hunter College of the City University of New York, New York, New York 10021 CHASE, DR. AURIN M., Department of Biology, Princeton University, Princeton, New Jersey 08540 CHAUNCEY, DR. HOWARD H., 470 Palm Island S. E., Clearwater, Florida 33515 CHENEY, DR. RALPH H., 11 Park Street, Woods Hole, Massachusetts 02543 CHILD, DR. FRANK M., Department of Biology, Trinity College, Hartford, Con- necticut 06106 CITKOWITZ, DR. ELENA, Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 CLARK, DR. A. M., Department of Biological Sciences, LTniversity of Delaware, Newark, Delaware 19711 48 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY CLARK, DR. ELOISE E., National Science Foundation, 1800 G. Street, Washington, D. C. 20550 CLARK, DR. WALLIS H., National Oceanic and Atmospheric Administration, National Marine Fisheries Service, Biological Laboratory, 4700 Avenue U., Galveston, Texas 77550 CLARKE, DR. GEORGE L., Biological Laboratories, Harvard University, Cam- bridge, Massachusetts 02138 CLAYTON, DR. RODERICK K., Section of Genetics, Development and Physiology, Cornell UJniversity, Ithaca, New York 14850 CLEMENT, DR. A. C., Department of Biology, Emory University, Atlanta, Georgia 30322 CLOWES, DR. GEORGE H. A., JR., Harvard Medical School, Boston, Massa- chusetts 02115 COBB, DR. JEWEL P., Dean of the College, Connecticut College, New London, Connecticut 06320 COHEN, DR. ADOLPH L, Department of Ophthalmology, Washington University, School of Medicine, 4550 Scott, St. Louis, Missouri 67110 COHEN, DR. LAWRENCE B., Department of Physiology, Yale University, New Haven, Connecticut 06510 COHEN, DR. SEYMOUR S., Department of Microbiology, University of Colorado Medical School, Denver, Colorado 80220 COLE, DR. KENNETH S., Laboratory of Biophysics, NINDS, National Institutes of Health, Bethesda, Maryland 20014 COLLIER, DR. JACK R., Department of Biology, Brooklyn College, Brooklyn, New York 11210 COLWIN, DR. ARTHUR L., Division of Functional Biology, University of Miami, School of Marine and Atmospheric Sciences, 10 Rickenbacker Causeway, Miami, Florida 33149 COLWIN, DR. LAURA H., Division of Functional Biology, University of Miami, School of Marine and Atmospheric Sciences, 10 Rickenbacker Causeway, Miami, Florida 33149 COOPERSTEIN, DR. SHERWIN J., School of Dental Medicine, University of Con- necticut, Farmington, Connecticut 06032 COPELAND, DR. D. EUGENE, Department of Biology, Tulane University, New Orleans, Louisiana 70118 CORNELL, DR. NEAL W., Department of Chemistry, Pomona College, Claremont, California 91711 CORNMAN, DR. IVOR, 10A Orchard Street, W'oods Hole, Massachusetts 02543 COSTELLO, DR. DONALD P., Department of Zoology, University of North Caro- lina, Chapel Hill, North Carolina 27514 COSTELLO, DR. HELEN MILLER, Department of Zoology, University of North Carolina, Chapel Hill, North Carolina 27514 COUCH, DR. ERNEST F., Department of Biology, Texas Christian LIniversity, Forth Worth, Texas 76110 COUSINEAU, DR. GILLES H., Department of Biology, Montreal University, P. (). Box 6128, Montreal, P. Q., Canada CRANE, JOHN O., Box 145, Woods Hole, Massachusetts 02543 REPORT OF THE DIRECTOR 49 CRANE, DR. ROBERT K., Department of Physiology, Rutgers Medical School, Piscataway, New Jersey 08854 CRIPPA, DR. MARCO, CNR Laboratory of Molecular Embryology, Arco Felice, Naples, Italy GROUSE, DR. HELEN V., Institute for Molecular Biophysics, Florida State Uni- versity, Tallahassee, Florida 32306 CROWELL, DR. SEARS, Department of Zoology, Indiana University, Bloomington, Indiana 47401 DAK;NAULT, ALEXANDER T., 1114 Avenue of the Americas, New York, New York 10036 DAN, DR. JEAN CLARK, Department of Biology, Ochanomizu University, Otsuka, Bunkyo-Ku, Tokyo, Japan DAN, DR. KATSUMA, Misaki Biological Station, Misaki, Japan DANIELLI, DR. JAMES F., Center for Theoretical Biology, State University of New York, 4248 Ridge Lea Rd., Amherst, New York 14226 DAVIDSON, DR. ERIC H., Division of Biology, California Institute of Technology, Pasadena, California 91109 DAVIS, DR. BERNARD D., Harvard Medical School, 25 Shattuck Street, Boston, Massachusetts 02115 DAW, DR. NIGEL W., Department of Physiology, Washington University Medical School, 4566 Scott Avenue, St. Louis, Missouri 63110 DfiHAAN, DR. ROBERT L., Department of Anatomy, Emory University, Atlanta, Georgia 30322 DEPHILLIPS, DR. HENRY A., JR., Department of Chemistry, Trinity College, Hartford, Connecticut 06106 DETTBARN, DR. WOLF-DIETRICH, Department of Pharmacology, Yanderbilt University, School of Medicine, Nashville, Tennessee 37217 DEVILLAFRANCA, DR. GEORGE W., Department of Zoology, Smith College, North- ampton, Massachusetts 01060 DEWEER, DR. PAUL J., Department of Physiology, Washington University Medical School, St. Louis, Missouri 63110 DIEHL, DR. FRED ALISON, Department of Biology, University of Virginia, Char- lottesville, Virginia 22903 DIXON, DR. KEITH E., School of Biological Sciences, Flinders University, Bed- ford Park, South Australia DOOLITTLE, DR. R. F., Department of Chemistry, University of California, San Diego, La Jolla, California 92037 DOWLING, DR. JOHN E., Biological Laboratories, Harvard University, 16 Divinity Avenue, Cambridge, Massachusetts 02138 DRESDEN, DR. MARC H., Department of Biochemistry, Baylor College of Medi- cine, Houston, Texas 77025 DUDLEY, DR. PATRICIA L., Department of Biological Sciences, Barnard College, Columbia University, New York, New York 10027 DUNHAM, DR. PHILIP B., Department of Biology, Syracuse University, Syracuse, New York 13210 EBERT, DR. JAMES DAVID, Department of Embryology, Carnegie Institution of Washington, Baltimore, Maryland 21210 50 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY ECKERT, DR. ROGER O., Department of Zoology, University of California, Los Angeles, California 90024 EDDS, DR. MAC V., JR., South College, University of Massachusetts, Amherst, Massachusetts 01002 EDER, DR. HOWARD A., Albert Einstein College of Medicine, Bronx, New York 10461 EDWARDS, DR. CHARLES, Department of Biological Sciences, State University of New York at Albany, Albany, New York 12203 EGYUD, DR. LASZLO G., The Institute for Muscle Research, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 EHRENSTEIN, DR. GERALD, National Institutes of Health, Bethesda, Maryland 20014 EICHEL, DR. HERBERT J., Department of Biochemistry, Hahnemann Medical College, Philadelphia Pennsylvania 19102 EISEN, DR. ARTHUR Z., Division of Dermatology, Washington University, School of Medicine, St. Louis, Missouri 63130 EISEN, DR. HERMAN, Center for Cancer Research, Department of Biology, Massachusetts Institute of Technology, Rm 56-526, Cambridge, Massa- chusetts 02139 ELDER, DR. HUGH YOUNG, Institute of Physiology, University of Glasgow, Glasgow, Scotland, U. K. ELLIOTT, DR. GERALD F., The O. U. Research Unit, 11/12 Bevington Rd., Oxford, England, U. K. EPSTEIN, DR. HERMAN T., Department of Biology, Brandeis University, Waltham, Massachusetts 02154 ERULKAR, DR. SOLOMON D., Department of Pharmacology, University of Penn- sylvania Medical School, Philadelphia, Pennsylvania 19104 ESSNER, DR. EDWARD S., Sloan Kettering Institute for Cancer Research, 410 E. 68th Street, New York, New York 10021 ETTIENNE, DR. EARL M., Department of Biology, Oakland University, Rochester, Michigan 48063 EVANS, DR. TITUS C., State University of Iowa, Radiation Research Laboratory, College of Medicine, Iowa City, Iowa 52240 FAILLA, DR. P. M., Office of the Director, Argonne National Laboratory, Argonne, Illinois 60439 FARMANFARMAIAN, DR. ALLAHVERDI, Department of Physiology and Biochem- istry, Rutgers University, New Brunswick, New Jersey 08903 FAUST, DR. ROBERT GILBERT, Department of Physiology, University of North Carolina Medical School, Chapel Hill, North Carolina 27514 FAWCETT, DR. D. W., Department of Anatomy, Harvard Medical School, Boston, Massachusetts 02115 FERGUSON, DR. E. P., National Institute of General Medical Sciences, National Institutes of Health, Bethesda, Maryland 20014 FERTIZIGER, DR. ALLAN P., Department of Physiology, University of Maryland Medical School, Baltimore, Maryland 21201 FESSENDEN, JANE, Librarian, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 FINE, DR. JACOB, 8 Wolcott Road Ext., Chestnut Hill, Massachusetts 02167 REPORT OF THE DIRECTOR 51 FINGERMAN, DR. MILTON, Department of Biology, Tulane University, New Orleans, Louisiana 70118 FISCHER, DR. ERNST, 3110 Manor Drive, Richmond, Virginia 23230 FISHER, DR. JEANNE M., Department of Biochemistry, University of Toronto, Toronto 5, Ontario, Canada FISHMAN, DR. Louis, 143 North Grove Street, Valley Stream, New York 11580 FISHMAN, DR. HARVEY MORTON, Department of Physiology, University of Texas Medical Branch, Galveston, Texas 77550 Fox, DR. MAURKE S., Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 FRAENKEL, DR. GOTTFRIED S., Department of Entomology, LIniversity of Illinois, Urbana, Illinois 61801 FRAZIER, DR. DONALD T., Department of Physiology and Biophysics, Uni- versity of Kentucky, Lexington, Kentucky 40507 FREEMAN, DR. ALAN RICHARD, Indiana University Medical Center, Department of Psychiatry, Indianapolis, Indiana 46202 FREEMAN, DR. GARY L., Department of Zoology, University of Texas, Austin, Texas 78710 FREYGANG, DR. WALTER H., JR., 6247 29th Street, N. W., Washington, D. C. 20015 FRIES, DR. ERIK F. B., P.O. Box 605, WToods Hole, Massachusetts 02543 FULTON, DR. CHANDLER M., Department of Biology, Brandeis University Waltham, Massachusetts 02154 FUORTES, DR. MICHAEL G. F\, National Institute for Neurological Diseases and Stroke, National Institutes of Health, Bethesda, Maryland 20014 FURSHPAN, DR. EDWIN J., Department of Neurophysiology, Harvard Medical School, Boston, Massachusetts 02115 FYE, DR. PAUL M., Woods Hole Oceanographic Institution, Woods Hole, Massa- chusetts 02543 GABRIEL, DR. MORDECAI L., Department of Biology, Brooklyn College, Brooklyn, New York 11210 GAFFRON, DR. HANS, Department of Biology, Institute of Molecular Biophysics, Conradi Building, Florida State University, Tallahassee, Florida 32306 GALL, DR. JOSEPH G.. Department of Biology, Yale University, New Haven, Connecticut 06520 GELFANT, DR. SEYMOUR, Department of Dermatology, Medical College of Georgia, Augusta, Georgia 30904 GELPERIN, DR. ALAN, Department of Biology, Princeton University, Princeton, New Jersey 08540 GERMAN, DR. JAMES L., Ill, The New York Blood Center, 310 East 67th Street, New York, New York 10021 GIBBS, DR. MARTIN, Department of Biology, Brandeis University, Waltham, Massachusetts 02154 GIFFORD, DR. PROSSER, Dean, Amherst College, Amherst, Massachusetts 01002 GILBERT, DR. DANIEL L., Laboratory of Biophysics, NINDS, National Institutes of Health, Building 36, Room 2A-31, Bethesda, Maryland 20014 GILMAN, DR. LAUREN C., Department of Biology, University of Miami, Coral Gables, Florida 33146 52 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY GINSBERG, DR. HAROLD S., Department of Microbiology, University of Pennsyl- vania School of Medicine, Philadelphia, Pennsylvania 19104 GIUDICE, DR. GIOVANNI, University of Palermo, Via Archirafi 22, Palermo, Italy GOLDEN, WILLIAM T., 40 Wall Street, New York, New York 10005 GOLDMAN, DAVID S., Department of Physics and Biophysics, Medical College of Pennsylvania, 300 Henry Avenue, Philadelphia, Pennsylvania 19129 GOLDSMITH, DR. TIMOTHY H., Department of Biology, Yale University, New Haven, Connecticut 06520 GOOCH, DR. JAMES L., Department of Biology, Juniata College, Huntingdon, Pennsylvania 16652 GOODCHILD, DR. CHAUNCEY G., Department of Biology, Emory University, Atlanta, Georgia 30322 GORMAN, DR. ANTHONY L. F., Laboratory of Neuropharmacology, SMH, IRP, NIMH, St. Elizabeths Hospital, Washington, D. C. 20032 GOTTSCHALL, DR. GERTRUDE Y., 315 East 68th Street, Apartment 9M, New York, New York 10021 GOUDSMIT, DR. ESTHER M., Department of Biology, Oakland University, Rochester, Michigan 48063 GRAHAM, DR. HERBERT, National Marine Fisheries Service, Woods Hole, Massa- chusetts 02543 GRANT, DR. DAVID C., Davidson College, Box 2316, Davidson, North Carolina 28036 GRANT, DR. PHILIP, Department of Biology, University of Oregon, Eugene, Oregon 97403 GRASS, ALBERT, The Grass Foundation, 77 Reservoir Road, Quincy, Massa- chusetts 02170 GRASS, ELLEN R., The Grass Foundation, 77 Reservoir Road, Quincy, Massa- chusetts 02170 GREEN, DR. JAMES W., Department of Physiology, Rutgers University, New Brunswick, New Jersey 08903 GREEN, DR. JONATHAN P., School of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia GREEN, DR. MAURICE, Department of Microbiology, St. Louis University Medi- cal School, St. Louis, Missouri 63103 GREENBERG, DR. MICHAEL J., Department of Biological Sciences, Florida State University, Tallahassee, Florida 32306 GREGG, DR. JAMES H., Department of Zoology, University of Florida, Gainesville, Florida 32601 GREIF, DR. ROGER L., Department of Physiology, Cornell University Medical College, New York, New York 10021 GRIFFIN, DR. DONALD R., The Rockefeller University, 66 Street and York Avenue, New York, New York 10021 GROSCH, DR. DANIEL S., Department of Genetics, Garden Hall, North Carolina State University, Raleigh, North Carolina 27607 GROSS, DR. JEROME, Developmental Biology Laboratory, Massachusetts General Hospital, Boston, Massachusetts 02114 GROSSMAN, DR. ALBERT, New York University Medical School, New York, New York 10016 REPORT OF THE DIRECTOR 53 GRUNDFEST, DR. HARRY, Department of Neurology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 GUTTMAN, DR. RITA, Department of Biology, Brooklyn College, Brooklyn, New- York 11210 GWILLIAM, DR. G. F., Department of Biology, Reed College, Portland, Oregon 97202 HAJDU, DR. STEPHEN, National Institutes of Health, Bethesda, Man-land 20014 HALVORSON, DR. HARLYN O., Rosenstiel Basic Medical Science Research Center, Brandeis University, Waltham, Massachusetts 02154 HAMILTON, DR. HOWARD L., Department of Biology, University of Virginia, Charlottesville, Virginia 22903 HARDING, DR. CLIFFORD V., JR., Department of Biological Sciences, Oakland University, Rochester, Michigan 48063 HARRINGTON, DR. GLENN W., Department of Microbiology, University of Missouri, School of Dentistry, 650 E. 25th Street, Kansas City, Missouri 64108 HARTLINE, DR. H. KEFFER, The Rockefeller University, New York, New York 10021 HARTMAN, DR. H. BERNARD, Department of Zoology, University of Iowa, Iowa City, Iowa 52240 HASCHEMEYER, DR. AUDREY E. V., Department of Biological Sciences, Hunter College, 695 Park Avenue, New York, New York 10021 HASTINGS, DR. J. WOODLAND, Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138 HAUSCHKA, DR. T. S., Roswell Park Memorial Institute, 666 Elm Street, Buffalo, New York 14203 HAXO, DR. FRANCIS T., Department of Marine Botany, Scripps Institution of Oceanography, University of California, La Jolla, California 92038 HAYASHI, DR. TERU, Department of Biology, Illinois Institute of Technology, Chicago, Illinois 60616 HAYES, DR. RAYMOND L., JR., Department of Anatomy and Cell Biology, Univer- sity of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15219 HEGYELI, DR. ANDREW F., 10824 Middleboro Drive, Damascus, Maryland 20750 HENDLEY, DR. CHARLES 1)., 615 South Avenue, Highland Park, New Jersey 08904 HENLEY, DR. CATHERINE, Department of Zoology, University of North Carolina, Chapel Hill, North Carolina 27514 HERNDON, DR. WALTER R., Office of the Dean, College of Liberal Arts, 110 Administration Building, University of Tennessee, Knoxville, Tennessee 37916 HERVEY, JOHN P., Box B-5 Penzance Road, Woods Hole, Massachusetts 02543 HESSLER, DR. ANITA Y., 5795 Waverly Avenue, La Jolla, California 92037 HIATT, DR. HOWARD H., Office of the Dean, Harvard School of Public Health, 55 Shatttick St., Boston, Massachusetts 02115 HILL, DR. ROBERT BENJAMIN, Department of Zoology, University of Rhode Island, Kingston, Rhode Island 02881 HILLMAN, DR. PETER, Department of Biology, Hebrew University, Jerusalem HINEGARDNER, DR. RALPH T., Division of Natural Sciences, University of Cali- fornia, Santa Cruz, California 95060 54 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY HINSCH, DR. GERTRUDE W., Institute of Molecular Evolution, 521 Anastasia, University of Miami, Coral Gables, Florida 33134 HIRSHFIELD, DR. HENRY L., Department of Biology, Washington Square Center, New York University, New York, New York 10003 HODGE, DR. CHARLES, IV, Department of Biology, Temple University, Philadel- phia, Pennsylvania 19122 HOFFMAN, DR. JOSEPH, Department of Physiology, Yale University School of Medicine, New Haven, Connecticut 06515 HOLLYFIELD, DR. JOE C., Department of Ophthalmology, Columbia University, 630 W. 168th Street, New York, New York 10032 HOLTZMAN, DR. ERIC, Department of Biological Science, Columbia University, New York, New York 10032 HOLZ, DR. GEORGE G., JR., Department of Microbiology, State University of New York, Upstate Medical Center, Syracuse, New York 13210 HOSKIN, DR. FRANCIS C. G., Biology Department, Illinois Institute of Tech- nology, Chicago, Illinois 60616 HOUSTON, HOWARD, Preston Avenue, Meriden, Connecticut 06450 HUBBARD, DR., RUTH, Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138 HUMES, DR. ARTHUR G., Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 HUMMON, DR. WILLIAM D., Department of Zoology, Ohio University, Athens, Ohio 45701 HUMPHREYS, DR. TOM DANIEL, Pacific Biomedical Research Center, University of Hawaii, 41 Ahui St., Honolulu, Hawaii 96813 HUNTER, DR. BRUCE, Department of Zoology, Connecticut College, New London, Connecticut 06320 HUNZIKER, H. E., Main St., Falmouth, Massachusetts 02540 HURWITZ, DR. CHARLES, Basic Science Research Laboratory, VA Hospital, Albany, New York 12208 HURWITZ, DR. JERARD, Department of Molecular Biology, Albert Einstein College of Medicine, Bronx, New York 10461 HUXLEY, DR. HUGH E., Medical Research Council, Laboratory of Molecular Biology, Cambridge, England, U. K. HYDE, DR. BEAL B., Department of Botany, University of Vermont, Burlington, Vermont 05401 HYDE, L. ROBINSON, Princeton University, Princeton, New Jersey 08540 I LAN, DR. JOSEPH, Department of Anatomy, Case Western Reserve L^niversity, School of Medicine, Cleveland, Ohio 44106 INOUE, DR. SADAYUKI, Department of Biochemistry, University of Montreal, Montreal, P. Q., Canada INOUE, DR. SHINYA, 217 Leidy Building, Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 ISENBERG, DR. IRVIN, Department of Biochemistry and Biophysics, Oregon State University, Corvallis, Oregon 97330 ISSELBACHER, DR. KURT J., Massachusetts General Hospital, Boston, Massa- chusetts 02114 REPORT OF THE DIRECTOR 55 IZZARD, DR. COLIN S., Department of Biological Sciences, State University of New York at Albany, Albany, New York 12207 JACOBSON, DR. ANTONE G., Department of Biology, University of Texas, Austin, Texas 78710 JAFFE, LIONEL, Department of Biology, Purdue University, Lafayette, Indiana 47907 JANNASCH, DR. HOLGER W., Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 JENNER, DR. CHARLES E., Department of Zoology, University of North Carolina, Chapel Hill, North Carolina 27514 JENNINGS, DR. JOSEPH B., Department of Zoology, University of Leeds, Leeds LS2 9JT, England, U. K. JOHNSON, DR. FRANK H., Department of Biology, Princeton University, Prince- ton, New Jersey 08540 JONES, DR. E. RUFFIN, JR., Department of Biological Sciences, University of Florida, Gainesville, Florida 32601 JONES, DR. MEREDITH L., Division of Worms, Museum of Natural History, Smithsonian Institution, Washington, D. C. 20650 JONES, DR. RAYMOND F., Department of Biology, State University of New York at Stony Brook, Long Island, New York 11753 JOSEPHSON, DR. R. K., School of Biological Sciences, University of California, Irvine, California 92664 KABAT, DR. E. A., Neurological Institute, Columbia University, College of Physicians and Surgeons, New York, New York 10032 KAFATOS, DR. FOTIS C., Biological Laboratories, Harvard University, 16 Divinity Avenue, Cambridge, Massachusetts KAJI, DR. AKIRA, Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 KALEY, DR. GABOR, Department of Physiology, New York Medical College, Valhalla, New York 10595 KAMINER, DR. BENJAMIN, Department of Physiology, Boston University School of Medicine, Boston, Massachusetts 02118 KANE, DR. ROBERT F., Pacific Biomedical Research Center, 2538 The Mall, University of Hawaii, Honolulu, Hawaii 96822 KANESHIRO, DR. EDNA S., Department of Biological Sciences, University of Cincinnati, Cincinnati, Ohio 45221 KARAKASHIAN, DR. STEPHEN J., 124 Park Place, Brooklyn, New York 11217 KARUSH, DR. FRED, Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 KATZ, DR. GEORGE M., Department of Neurology, Columbia University, College of Physicians and Surgeons, 630 West 168th Street, New York, New York 10032 KAUFMAN, DR. B. P., Department of Zoology, University of Michigan, Ann Arbor, Michigan 48104 KEAN, DR. EDWARD L., Departments of Biochemistry and Ophthalmology, Case Western Reserve University, Cleveland, Ohio 44101 KELLEY, ROBERT E., Department of Anatomy, University of Illinois, College of Medicine, P. O. Box 6998, Chicago, Illinois 60680 56 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY KEMP, DR. NORMAN E., Department of Zoology, University of Michigan, Ann Arbor, Michigan, 48104 KEMPTON, DR. RUDOLF T., 924 Shore Drive, St. Augustine, Florida 32084 KEOSIAN, DR. JOHN, 18 Meadow Lane, Amityville, New York 11710 KETCHUM, DR. BOSTWICK H., Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 KEYNAN, DR. ALEXANDER, Authority for Research Development, Hebrew Uni- versity, Jerusalem, Israel KILLE, DR. FRANK R., 340 Albany Shaker Road, Londonville, New York 12211 KINDRED, DR. JAMES E. 2010 Hessian Road, Charlottesville, Virginia 22903 KING, DR. THOMAS J., Program Director, National Bladder and Prostatic Cancer Programs, Division of Cancer Grants, National Cancer Institute, Westwood Bldg., Km. 853, Bethesda, Maryland 20014 KINGSBURY, DR. JOHN M., Department of Botany, Cornell University, Ithaca, New York 14850 KIRSCHENBAUM, DR. DONALD, Department of Biochemistry, College of Medicine, State University of New York, 450 Clarkson Avenue, Brooklyn, New York 11203 KLEIN, DR. MORTON, Department of Microbiology, Temple University, Phila- delphia, Pennsylvania 19122 KLEINHOLZ, DR. LEWIS H., Department of Biology, Reed College, Portland, Oregon 97202 KLEYN, DR. JOHN (i., Department of Biology, University of Puget Sound, Tacoma, Washington 98416 KLOTZ, DR. I. A I., Department of Chemistry, Northwestern University, Evans- ton, Illinois 60201 KOHLER, DR. KURT, Biologisches Institut der ON Stuttgart, D-7 Stuttgart 60 Ulnier Str. 227, Germany KONIGSBERG, DR. IRWIN R., Department of Biology, Gilmer Hall, University of Virginia, Charlottesville, Virginia 22903 KORR, DR. I. M., Department of Physiology, Kirksville College of Osteopathy, Kirksville, Missouri 63501 KRAHL, DR. M. E., Department of Physiology, Stanford University, Stanford, California KRANE, DR. STEPHEN M., Massachusetts General Hospital, Boston, Massa- chusetts 02114 KRASSNER, DR. STUART MITCHELL, Department of Organismic Biology, Uni- versity of California, Irvine, California 92650 KRAUSS, DR. ROBERT, Dean, School of Science, Oregon State University, Cor- vallis, Oregon 97330 KRIEBEL, DR. MAHLON E., Department of Physiology, State University of New York, Upstate Medical Center, Syracuse, New York 13210 KRIEG, DR. WENDELL J. S., Northwestern Medical School, 303 East Chicago Avenue, Chicago, Illinois 60611 KRUPA, DR. PAUL L., Department of Biology, The City College of New York, 139th St. and Convent Avenue, New York, New York 10031 KUFFLER, DR. STEPHEN W., Department of Neurophysiology, Harvard Medical School, Boston, Massachusetts 02115 REPORT OF THE DIRECTOR 57 KUSANO, DR. KIYOSHI, Biology Department, Illinois Institute of Technology, 3300 Federal Street, Chicago, Illinois 61606 LAMARCHE, DR. PAUL H., 593 Eddy St., Providence, Rhode Island 02903 LAMY, DR. FRANCOIS, Department of Biochemistry, University of Sherbrooke, School of Medicine, Sherbrooke, Quebec, Canada LANCEFIELD, DR. D. E., 203 Arleigh Road, Douglaston, Lond Island, New York 11363 LANCEFIELD, DR. REBECCA C., The Rockefeller University, 66th Street and York Avenue, New York, New York 10021 LANDOWNE, DR. DAVID, Department of Physiology, University of Miami, Miami, Florida 33124 LANG, DR. FREDERICK, Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 LANSING, DR. ALBERT I., Department of Anatomy, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213 LASH, DR. JAMES W., Department of Anatomy, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 LASTER, DR. LEONARD, Executive Director, Assembly of the Life Sciences, National Academy of Sciences, 2101 Constitution Ave. N. W., Washington, D. C. 20418 LAUFER, DR. HANS, Department of Zoology and Entomology, University of Connecticut, Storrs, Connecticut 06268 LAUFER, DR. MAX A., Department of Biophysics, University of Pittsburgh, Pittsburgh, Pennsylvania 15260 LAVIN, DR. GEORGE I., 6200 Norvo Road, Baltimore, Maryland 21207 LAZAROW, DR. ARNOLD, Department of Anatomy, University of Minnesota Medical School, Minneapolis, Minnesota 55455 LEADBETTER, DR. EDWARD R., Department of Biology, Amherst College, Ain- herst, Massachusetts 01002 LEAK, DR. LEE VIRN, Department of Anatomy, Howard University, College of Medicine, Washington, D. C. 20001 LECAR, DR. HAROLD, Laboratory of Biophysics, National Institute of Neuro- logical Diseases and Stroke, National Institutes of Health, Bethesda, Mary- land 20014 LEDERBERG, DR. JOSHUA, Department of Genetics, Stanford Medical School, Palo Alto, California 94304 LEFEVRE, DR. PAUL G., Department of Physiology, State University of New York at Stony Brook, Stony Brook, New York 11790 LEIGHTON, DR. JOSEPH, Department of Pathology, Medical College of Penn- sylvania, 3300 Henry Ave., Philadelphia, Pennsylvania 19129 LENHER, DR. SAMUEL, 1900 Woodlawn Avenue, Wilmington, Delaware 19806 LERMAN, DR. SIDNEY, 49 Chaumont Square N. W., Cross Creek Parkway, Atlanta, Georgia 30329 LERNER, DR. AARON B., Yale Medical School, New Haven, Connecticut 06515 LEVIN, DR. JACK, Department of Medicine, The Johns Hopkins Hospital, Baltimore, Maryland 21205 LEVINE, DR. RACHMIEL, 2024 Canyon Road, Arcadia, California 91006 5S ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY LEVINTHAL, DR. CYRUS, Department of Biological Sciences, Columbia University, 908 Schermerhorn Hill, New York, New York 10027 LEVY, DR. MILTON, 39-95 48th Street, Long Island City, New York 11104 LEWIN, DR. RALPH A., Scripps Institution of Oceanography, La Jolla, California 92037 LEWIS, DR. HERMAN W., Genetic Biology Program, National Science Foundation, Washington, D. C. 20025 LING, DR. GILBERT, 307 Berkeley Road, Merion, Pennsylvania 19066 LINSKENS, DR. H. P., Department of Botany, University of Driehuizerweg 200, Nijmegen, The Netherlands LIPICKY, DR. RAYMOND J., Department of Pharmacology, College of Medicine, University of Cincinnati, Eden and Bethesda Avenues, Cincinnati, Ohio 45219 LITTLE, DR. E. P., 216 Highland Street, West Newton, Massachusetts 02158 LIUZZI, DR. ANTHONY, Division of Nuclear Science, Lowell Technological Insti- tute, Lowell, Massachusetts 01854 LLINAS, DR. RODOLFO R., Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa 52240 LOCHHEAD, DR. JOHN H., Department of Zoology, Life Sciences Building, Uni- versity of Vermont, Burlington, Vermont 05401 LOWENSTEIN, DR. WERNER R., Physiology and Biophysics, School of Medicine, University of Miami, P. (). Box 875, Miami, Florida 33152 LOEWUS, DR. FRANK A., Department of Biology, State University of New York at Buffalo, Buffalo, New York 14214 LOFTFIELD, DR. ROBERT B., Department of Biochemistry, University of New Mexico Medical School, Albuquerque, New Mexico 87106 LONDON, DR. IRVING M., Department of Medicine, Albert Einstein College of Medicine, New York, New York 10461 LORAND, DR. LASZLO, Department of Chemistry, Northwestern University, Evanston, Illinois 60201 LOVE, DR. WARNER E., Department of Biophysics, Johns Hopkins University, Baltimore, Maryland 21218 LURIA, DR. SALVADOR E., Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 LYNCH, DR. CLARA J., 4800 Fillmore Avenue, Alexandria, Virginia 22311 MAcNiCHOL, EDWARD F., JR., Marine Biological Laboratory, Woods Hole, Massachusetts 02543 MAHLER, DR. HENRY R., Department of Biochemistry, Indiana University, Bloomington, Indiana 47401 MALKIEL, DR. SAUL, Children's Cancer Research Foundation, Inc., 35 Binney Street, Boston, Massachusetts 02115 MANALIS, DR. RICHARD S., Department of Physiology, University of Cincinnati, College of Medicine, Eden and Bethesda Aves., Cincinnati, Ohio 45219 MANGUM, CHARLOTTE P., Department of Biology, College of William and Mary, Williamsburg, Virginia 23185 MARKS, DR. PAUL A., Columbia University, College of Physicians and Surgeons, New York, New York 10032 MARSH, DR. JULIAN B., Department of Biochemistry, University of Pennsylvania School of Dental Medicine, 4001 Spruce St., Philadelphia, Pennsylvania 19104 REPORT OF THE DIRECTOR 59 MAUTNER, DR. HENRY G., Tufts University School of Medicine, 136 Harrison Avenue, Department of Biochemistry and Pharmacy, Boston, Massachusetts 02111 MAXWELL, DR. ARTHUR, Provost, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 MAZIA, DR. DANIEL, Department of Zoology, University of California, Berkeley, California 94720 McCANN, DR. F RANGES, Department of Physiology, Dartmouth Medical School, Hanover, New Hampshire 03755 McCLOSKEY, DR. LAWRENCE R., Department of Biology, Walla Walla College, College Place, Washington 99324 McDANiEL, DR. JAMES SCOTT, Department of Biology, East Carolina College, Greenville, North Carolina 28734 McELROY, DR. WILLIAM D., Chancellor, University of California, San Diego, La Jolla, California 92037 MCLAUGHLIN, JANE A., Institute for Muscle Research, Marine Biological Labora- tory, Woods Hole, Massachusetts 02543 McREYNOLDS, DR. JOHN S., Laboratory of Neurophysiology, NINDB, National Institutes of Health, Bethesda, Maryland 20014 MEINKOTH, DR. NORMAN, Department of Biology, Swarthmore College, Swarth- more, Pennsylvania 19081 MELLON, DR. DEFOREST, JR., Department of Biology, University of Virginia, Charlottesville, Virginia 22903 MENDELSON, DR. MARTIN, Health Sciences Centers, State University of New York, Stony Brook, New York 1 1 790 METZ, DR. C. B., Institute of Molecular Evolution, University of Miami, Coral Gables, Florida 33146 METZ, DR. CHARLES W., Box 714, Woods Hole, Massachusetts 02543 MIDDLEBROOK, DR. ROBERT, Downsway, School Lane, Kirk Ella, Hull, England, U. K. HW10 7NR MILKMAN, DR. ROGER D., Department of Zoology, University of Iowa, Iowa City, Iowa 52240 MILLS, DR. ERIC LEONARD, Institute of Oceanography, Dalhousie University, Halifax, Nova Scotia, Canada MILLS, ROBERT, 56 Worcester Ct., Falmouth, Massachusetts 02540 AIiLNE, DR. LORUS J., Department of Zoology, University of New Hampshire, Durham, New Hampshire 03824 MIZELL, DR. MERLE, Department of Biology, Tulane LIniversity, New Orleans, Louisiana 70118 MONROY, DR. ALBERTO, CNR Laboratory of Molecular Embryology, 80072 Arco Felice, Napoli, Italy MOORE, DR. JOHN A., Department of Life Sciences, University of California, Riverside, California 92502 MOORE, DR. JOHN W7., Department of Physiology, Duke University Medical Center, Durham, North Carolina 27706 MORAN, DR. JOSEPH F., Jr., 23 Foxwood Drive, RR #1, Eastham, Massachusetts 02642 60 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MORIN, DR. JAMES G., Department of Zoology, University of California, Los Angeles, California 90052 MORLOCK, DR. NOEL, Department of Surgery, Detroit General Hospital, 1326 St. Antoine Street, Detroit, Michigan 48226 MORRILL, DR. JOHN B., JR., Division of Natural Sciences, New College, Sarasota, Florida 33478 MORSE, DR. RICHARD STETSON, 193 Winding River Road, Wellesley, Massa- chusetts 02184 MORSE, ROBERT W., Associate Director, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 MOSCONA, DR. A. A., Department of Zoology, University of Chicago, Chicago, Illinois 60637 MOTE, DR. MICHAEL I., Department of Biology, Temple University, Philadelphia, Pennsylvania 19122 MOUL, DR. E. T., Department of Biology, Rutgers University, New Brunswick, New Jersey 08903 MOUNTAIN, DR. ISABEL M., 2 Lilac Place, Thornwood, New York 10594 MULLINS, DR. LORIN J., Department of Biophysics, University of Maryland School of Medicine, Baltimore, Maryland 21201 MUSACCHIA, DR. XAVIER J., Department of Physiology, Medical Center, Uni- versity of Missouri, Columbia, Missouri 65201 NABRIT, DR. S. M., 686 Beckwith Street, S. W. Atlanta, Georgia 30314 NACE, DR. PAUL FOLEY, 5 Bowditch Rd., Woods Hole, Massachusetts NACHMANSOHN, DR. DAVID, Department of Neurology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 NARAHASHI, DR. TOSHIO, Department of Physiology, Duke University Medical Center, Durham, North Carolina 27706 NASATIR, DR. MAIMON, Department of Biology, University of Toledo, Toledo, Ohio 43606 NASON, DR. ALVIN, McCollum-Pratt Institute, The Johns Hopkins llniversity, Baltimore, Maryland 21218 NELSON, DR. LEONARD, Department of Physiology, Medical College of Ohio at Toledo, Toledo, Ohio 43614 NICHOLLS, DR. JOHN GRAHAM, Department of Neurobiology, Harvard Medical School, 25 Shattuck Street, Boston, Massachusetts 02115 NICOLL, DR. PAUL A., R.R. 12, Box 286, Bloomington, Indiana 47401 Nm, DR. MAN-CHIANG, Department of Biology, Temple University, Phila- delphia, Pennsylvania 19122 NOVIKOFF, DR. ALEX B., Department of Pathology, Albert Einstein College of Medicine, Bronx, New York 10461 NYSTROM, DR. RICHARD A., Hudson Valley Community College, 80 Vandenburgh Ave., Troy, New York 12180 OCHOA, DR. SEVERO, New York University, College of Medicine, New York, New York 10016 ODUM, DR. EUGENE, Department of Zoology, University of Georgia, Athens, Georgia 30601 OLSON, DR. JOHN M., Department of Biology, Brookhaven National Laboratory, Upton, New York 11973 REPORT OF THE DIRECTOR 61 OSCHMAN, UK. JAMES L., Department of Biological Sciences, Northwestern University, Evanston, Illinois 60201 PALMEK, UK. JOHN D., Department of Biology, New York University, New York, New York 10053 PALTI, DK. YORAM, Hebrew University School of Medicine, Department of Physiology, Box 1172, Jerusalem, Israel PAPPAS, DR. GEORGE D., Department of Anatomy, Albert Einstein College of Medicine, Bronx, New York 10461 PASSANO, DR. LEONARD M., Department of Zoology, University of Wisconsin, Madison, Wisconsin 53706 PEARLMAN, DR. ALAN L., Department of Physiology, Washington University School of Medicine, St. Louis, Missouri 63110 PERKINS, DR. C. D., 621 Lake Drive, Princeton, New Jersey 08540 PERSON, DR. PHILIP, Special Dental Research Program, Veterans Administration Hospital, Brooklyn, New York 11219 PETTIBONE, DR. MARIAN H., Division of Marine Invertebrates, U. S. National Museum, Washington, D. C. 20025 PFOHL, DR. RONALD }., Department of Zoology, Miami LIniversity, Oxford, Ohio 45056 PHILPOTT, DR. DELBERT E., MASA Ames Research Center, Moffett Field, California 94035 PIERCE, DR. SIDNEY K., JR., Department of Zoology, University of Maryland, College Park, Maryland 20740 POLLACK, DR. LELAND W., Department of Zoology, Drew University, Madison, New Jersey 07940 PORTER, DR. KEITH R., 748 llth Street, Boulder, Colorado 80302 POTTER, DR. DAVID, Department of Neurophysiology, Harvard Medical School, Boston, Massachusetts 02115 POTTS, DR. WILLIAM T. W., Department of Biology, University of Lancaster, Lancaster, England, U. K. PRENDERGAST, DR. ROBERT A., Department of Pathology and Ophthalmology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 PROSSER, DR. C. LADD, Department of Physiology and Biophysics, Burrill Hall, University of Illinois, LTrbana, Illinois 61803 PROVASOLI, DR. Luna, Haskins Laboratories, 165 Prospect Street, New Haven, Connecticut 06520 PRUSCH, DR. ROBERT D., Division of Biomedical Sciences, Brown University, Providence, Rhode Island 02904 PRYTZ, DR. MARGARET MCDONALD, 21 McCouns Lane, Oyster Bay, New York 11771 PRZYBYLSKI, DR. RONALD J., Department of Anatomy, Case Western Reserve University, Cleveland, Ohio 44101 RABIN, DR. HARVEY, Director, Department of Virology and Cell Biology, Bionetics Research Laboratories, 5510 Nicholson Lane, Kensington, Maryland 20795 RANKIN, DR. JOHN S., Department of Zoology, University of Connecticut, Storrs, Connecticut 06268 62 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY RANZI, DR. SILVIO, Department of Zoology, University of Milan, Via Celonia 10, Milan, Italy RATNER, DR. SARAH, Department of Biochemistry, The Public Health Research Institute of the City of New York, Inc., 455 First Avenue, New York, New York 10016 REBHUN, DR. LIONEL I., Department of Biology, Gilmer Hall, University of Virginia, Charlottesville, Virginia 22901 REDDAN, DR. JOHN R., Department of Biological Sciences, Oakland University, Rochester, Michigan 48063 REDFIELD, DR. ALFRED C., Maury Lane, Woods Hole, Massachusetts 02543 REINER, DR. JOHN M., Department of Biochemistry, Albany Medical College of Union University, Albany, New York 12208 RENN, DR. CHARLES E., Route 2, Hampstead, Maryland 21074 REUBEN, DR. JOHN P., Department of Neurology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 REYNOLDS, DR. GEORGE THOMAS, Palmer Laboratory, Princeton University, Princeton, New Jersey 08540 REZNIKOFF, DR. PAUL, 151 Sparks Ave., Pelham, New York 10803 RICE, DR. ROBERT VERNON, Mellon Institute, Carnegie-Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213 RICH, DR. ALEXANDER, Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 RICHARDS, DR. A. GLENN, Department of Entomology, University of Minnesota, St. Paul, Minnesota 55101 RICHARDS, DR. OSCAR W., Pacific University, College of Optometry, Forrest Grove, Oregon 97116 RIPPS, DR. HARRIS, Department of Opthalmology, New York University, School of Medicine, 550 1st Avenue, New York, New York 10016 ROBERTS, DR. JOHN L., Department of Zoology, LTniversity of Massachusetts, Amherst, Massachusetts 01002 ROBINSON, DR. DENIS M., 19 Orlando Avenue, Arlington, Massachusetts 02174 ROCKSTEIN, DR. MORRIS, Department of Physiology, University of Miami School of Medicine, P. O. Box 875 Biscayne Annex, Miami, Florida 33152 RONKIN, DR. RAPHAEL E., National Science Foundation, O.I.S.A., Washington, D. C. 20550 ROSE, DR. S. MERYL, Laboratory of Developmental Biology, Tulane University, F. Edward Hebert Center, Belle Chasse, Louisiana 70037 ROSENBAUM, DR. JOEL L., Kline Biology Tower, Yale University, New Haven, Connecticut 06510 ROSENBERG, DR. EVELYN K., Jersey City State College, Jersey City, New Jersey 07305 ROSENBERG, DR. PHILIP, Division of Pharmacology, University of Connecticut, School of Pharmacy, Storrs, Connecticut 06268 ROSENBLUTH, RAJA, # 10, 3250 West 4th Avenue, Vancouver 8, British Columbia, Canada V6K IR9 ROSENKRANZ, DR. HERBERT S., Department of Microbiology, Columbia Uni- versity, College of Physicians and Surgeons, New York, New York 10032 REPORT OF THE DIRECTOR 63 ROSENTHAL, DR. THEODORE B., Department of Anatomy, University of Pitts- burgh Medical School, Pittsburgh, Pennsylvania 15213 ROSLANSKY, DR. JOHN, 26 Albatross, Woods Hole, Massachusetts 02543 ROTH, DR. JAY S., Division of Biological Sciences, Section of Biochemistry and Biophysics, University of Connecticut, Storrs, Connecticut 06268 ROWLAND, DR. LEWIS P., Department of Neurology, Columbia University, College of Physicians and Surgeons, 630 W. 168th St., New York, New York 10032 RUBINOW, DR. SOL I., Cornell University, Medical College, Department of Biomathematics, New York, New York 10012 RUGH, DR. ROBERTS, Grosvenor Park, Apt. 1018, 10500 Rockville Pike, Rockville, Maryland 20852 RUSHFORTH, DR. NORMAN B., Department of Biology, Case Western Reserve University, Cleveland, Ohio 44106 RUSSELL-HUNTER, DR. W. D., Department of Biology, Lyman Hall, Syracuse University, Syracuse, New York 13210 RUSTAD, DR. RONALD C., Department of Radiology, Case Western Reserve University, Cleveland, Ohio 44106 RUTMAN, DR. ROBERT J., University of Pennsylvania, School of Veterinary Medicine, Department of Animal Biology, 3800 Spruce Street, Philadelphia, Pennsylvania 19104 RYTHER, DR. JOHN H., Woods Hole Oceanographic Institution, W^oods Hole, Massachusetts 02543 SAGER, DR. RUTH, Department of Biological Sciences, Hunter College, 695 Park Avenue, New York, New York 10021 SANBORN, DR. RICHARD C., Dean, Purdue University Regional Campus, 1125 East 38th Street, Indianapolis, Indiana 46205 SANDERS, DR. HOWARD L., Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 SATO, DR. HIDEMI, 217 Leidy Building, Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 SAUNDERS, DR. JOHN W., JR., Department of Biological Sciences, State University of New York, at Albany, Albany, New York 12203 SAZ, DR. ARTHUR KENNETH, Department of Microbiology, Georgetown Univer- sity Medical and Dental Schools, 3900 Reservoir Road, Washington, D. C. 20051 SCHACHMAN, DR. HOWARD K., Department of Biochemistry, University of California, Berkeley, California 94720 SCHARRER, DR. BERTA V., Department of Anatomy, Albert Einstein College of Medicine, Bronx, New York 10461 SCHIFF, DR. JEROME A., Department of Biology, Brandeis University, Waltham, Massachusetts 02154 SCHLESINGER, DR. R. WALTER, Department of Microbiology, Rutgers Medical School, P. O. Box 101, Piscataway, New Jersey 08903 SCHMEER, SISTER ARLINE CATHERINE, O.P., The Americal Medical Center of Denver, 6401 W. Colfax Ave., Denver, Colorado 80214 SCHMITT, DR. O. H., University of Minnesota, 200 T.N.C.E., Minneapolis, Minnesota 55455 64 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY SCHNEIDERMAN, DR. HOWARD A., Center for Pathobiology, School of Biological Sciences, University of California, Irvine, California 92664 SCHOLANDER, DR. P. F., Scripps Institution of Oceanography, La jolla, California 92037 SCHOPF, DR. THOMAS J. M., Department of the Geophysical Sciences, University of Chicago, 5734 S. Ellis Avenue, Chicago, Illinois 60637 SCHOTTE, DR. OSCAR E., Department of Biology, Amherst College, Amherst, Massachusetts 01002 SCHUEL, DR. HERBERT, Department of Biochemistry, State University of New York, Downstate Medical Center, 450 Clarkson Ave., Brooklyn, New York 11203 SCHUETZ, DR. ALLEN WALTER, The Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland 21205 SCHWARTZ, DR. TOBIAS L., Biological Sciences Group, University of Connecticut, Storrs, Connecticut 06268 SCOTT, DR. ALLAN C., Colby College, Waterville, Maine 02901 SCOTT, DR. GEORGE T., Department of Biology, Oberlin College, Oberlin, Ohio 44074 SEARS, DR. MARY, Box 152, Woods Hole, Massachusetts 02543 SEGAL, DR. SHELDON J., Population Council, The Rockefeller University, New York, New York 10021 SELIGER, DR. HOWARD H., McCollum-Pratt Institute, The Johns Hopkins University, Baltimore, Maryland 21218 SENFT, DR. JOSEPH P., Department of Biology, Juniata College, Huntingdon, Pennsylvania 16652 SHANKLIN, DR. DOUGLAS R., Pathologist-in-chief, University of Chicago, Chicago Lying-in Hospital, Chicago, Illinois 60637 SHAPIRO, DR. HERBERT, 6025 North 13th Street, Philadelphia, Pennsylvania 19141 SHAVER, DR. JOHN R., Department of Zoology, Michigan State University, East Lansing, Michigan 48823 SHEDLOVSKY, DR. THEODORE, The Rockefeller University, New York, New York 10021 SHEMIN, DR. DAVID, Department of Chemistry and Biological Sciences, North- western University, Evanston, Illinois 60201 SHEPHARD, DR. DAVID C., P. (). Box 44, Woods Hole, Massachusetts 02543 SHEPRO, DR. DAVID, Department of Biology, Boston University, 2 Cummington Street, Boston, Massachusetts 02215 SHERMAN, DR. I. W., Division of Life Sciences, University of California, Riverside, California 92502 SHILO, DR. MOSHE, Head, Department of Microbiological Chemistry, Hebrew University, Jerusalem, Israel SICHEL, DR. ELSA KEIL, Emeritus Professor of Biology, Trinity College, Burling- ton, Vermont 05401 SIEGEL, DR. IRWIN M., Department of Ophthalmology, New York University Medical Center, 550 First Avenue, New York, New York 10016 SIEGELMAN, DR. HAROLD W., Department of Biology, Brookhaven National Laboratory, Upton, New York 11973 REPORT OF THE DIRECTOR 65 SILVA, DR. PAUL C., Department of Botany, University of California, Berkeley, California 94720 SIMMONS, DR. JOHN E., JR., Department of Biology, University of California, Berkeley, California 94720 SIMON, DR. ERIC J., New York University Medical School, 550 First Avenue, New York, New York 10016 SJODIN, DR. RAYMOND A., Department of Biophysics, University of Maryland School of Medicine, Baltimore, Maryland 21201 SKINNER, DR. DOROTHY M., Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830 SMELSER, DR. GEORGE K., Department of Anatomy, Columbia University, New York, New York 10032 SMITH, HOSIER P., General Manager, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 SMITH, PAUL FERRIS, Church Street, Woods Hole, Massachusetts 02543 SMITH, DR. RALPH I., Department of Zoology, University of California, Berkeley, California 94720 SONNEBORN, DR. T. M., Department of Zoology, Indiana University, Blooming- ton, Indiana 47401 SONNENBLICK, DR. B. P., Department of Biology, Rutgers University, 195 University Avenue, Newark, New Jersey 07102 SORENSON, DR. ALBERT L., Department of Biology, Brooklyn College, Brooklyn, New York 11210 SORENSON, DR. MARTHA M., Department of Neurology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 SPECTOR, DR. A., Department of Ophthalmology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 SPIEGEL, DR. MELVIN, Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire 03755 SPINDEL, DR. WILLIAM, Belfer Graduate School of Science, Yeshiva University, Amsterdam Avenue and 186th Street, Bronx, New York 10461 SPIRA, DR. MICHA E., Department of Zoology, Hebrew University, Jerusalem, Israel SPIRTES, DR. MORRIS ALBERT, Veterans Administration Hospital, 1601 Perdido Street, New Orleans, Louisiana 70112 STARR, DR. RICHARD C., Department of Botany, Indiana University, Blooming- ton, Indiana 47401 STEINBACH, DR. H. BURR, Oceanic Foundation, Ma Kajuiu Point, Waimanalo Point, Hawaii 96795 STEINBERG, DR. MALCOLM S., Department of Biology, Princeton University, Princeton, New Jersey 08540 STEINHARDT, DR. JACINTO, Science Advisor to the President, Georgetown Uni- versity, Washington, D. C. 20007 STEPHENS, DR. GROVER C., Division of Biological Sciences, University of Cali- fornia, Irvine, California 92650 STEPHENS, DR. RAYMOND E., Department of Biology, Brandeis University, Waltham, Massachusetts 02154 66 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY STETTEN, DR. DE\VITT, National Institute of General Medical Sciences, National Institutes of Health, Bethesda, Maryland 20014 STETTEN, DR. MARJORIE R., Rutgers Medical School, New Brunswick, New Jersey 08903 STRACHER, ALFRED, Downstate Medical Center, State University of New York at Brooklyn, 450 Clarkson Avenue, Brooklyn, New York 11203 STREHLER, DR. BERNARD L., 1 Laguna Circle Drive, Agoura, California 91307 STRITTMATTER, DR. PHILIPP, Department of Biochemistry, University of Con- necticut, School of Medicine, Health Center, Hartford Plaza, Hartford, Connecticut 06105 SULKIN, DR. S. EDWARD, Department of Bacteriology, Southwestern Medical School, University of Texas, Dallas, Texas 75221 SUMMERS, DR. WILLIAM C., Huxley College, Western Washington State Uni- versity, Bellingham, Washington 98225 SUSSMAN, DR. MAURICE, Institute of Life Sciences, The Hebrew University, Jerusalem, Israel SWANSON, DR. CARL PONTIUS, Department of Botany, University of Massa- chusetts, Amherst, Massachusetts 01002 SWOPE, GERARD, JR., Blinn Road, Box 345, Croton-on-Hudson, New York 10520 SZABO, DR. GEORGE, Harvard School of Dental Medicine, 188 Longwood Avenue, Boston, Massachusetts 02115 SZAMIER, DR. ROBERT BRUCE, Department of Neurobiology, University of Texas Medical School, Houston, Texas 77025 SZENT-GYORGYI, DR. ALBERT, Institute for Muscle Research, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 SzENT-GvoRGvi, DR. ANDREW G., Department of Biology, Brandeis University, Waltham, Massachusetts 02154 TANZER, DR. MARVIN L., Department of Biochemistry, University of Connecticut, School of Medicine, Farmington, Connecticut 06032 TASAKI, DR. ICHIJI, Laboratory of Neurobiology, National Institute of Mental Health, National Institutes of Health, Bethesda, Maryland 20014 TAYLOR, DR. ROBERT E., Laboratory of Biophysics, National Institute of Neurological Diseases and Blindness, National Institutes of Health, Bethesda, Maryland 20014 TAYLOR, DR. W7. ROWLAND, Department of Oceanography, Chesapeake Bay Insti- tute, The Johns Hopkins University, Baltimore, Maryland 21218 TELFER, DR. WILLIAM H., Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 DETERRA, DR. NOEL, The Institute for Cancer Research, 7701 Burholme Avenue, Fox Chase, Philadelphia, Pennsylvania 19111 THALER, DR. M. MICHAEL, University of California, San Francisco, California 94102 TIFFNEY, DR. WESLEY N., Department of Biology, Boston University, Boston, Massachusetts 02215 TRACER, DR. WILLIAM, The Rockefeller University, New York, New York 10021 TRAVIS, DR. D. M., Department of Pharmacology, University of Florida, Gaines- ville, Florida 32601 REPORT OF THE DIRECTOR 67 TRAVIS, DR. DOROTHY FRANCES, 1918 Northern Parkway, Greenberry Woods, Baltimore, Maryland 21210 TRINKAUS, DR. J. PHILIP, Department of Biology, Yale University, New Haven, Connecticut 06520 TROLL, DR. WALTER, Department of Environmental Medicine, New York Univer- sity, College of Medicine, New York, New York 10016 TWEEDELL, DR. KENYON S., Department of Biology, University of Notre Dame, Notre Dame, Indiana 46556 URETZ, DR. ROBERT B., Department of Biophysics, University of Chicago, Chicago, Illinois 60637 VALIELA, DR. IVAN, Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 VALOIS, JOHN, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 VAN HOLDE, DR. KENSAL EDWARD, Oregon State University, Department of Biochemistry and Biophysics, Corvallis, Oregon 97331 VILLEE, DR. CLAUDE A., Department of Biochemistry, Harvard Medical School, Boston, Massachusetts 02115 VINCENT, DR. WALTER S., Department of Biology, LTniversity of Delaware, Newark, Delaware 19711 WAINIO, DR. W. W., Bureau of Biological Research, Rutgers University, New Brunswick, New Jersey 08903 WAKSMAN, DR. BYRON, Department of Microbiology, Yale University, New Haven, Connecticut 06510 WALD, DR. GEORGE, Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138 WALKER, DR. CHARLES A., Department of Physiology and Pharmacology, School of Veterinary Medicine, Tuskegee Institute, Tuskegee, Alabama, 36088 WALL, DR. BETTY }., Department of Biological Sciences, Northwestern Uni- versity, Evanston, Illinois 60201 WALLACE, DR. ROBIN A., P. (). Box Y, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37890 WANG, DR. A., Bedford Road, Lincoln Massachusetts 01773 WARNER, DR. ROBERT C., Department of Molecular and Cell Biology, Uni- versity of California, Irvine, California 92664 WARREN, DR. LEONARD, Department of Therapeutic Research, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 WATERMAN, Dr. T. H., 610 Kline Biology Tower, Yale University, New Haven, Connecticut 06520 WATKINS, DR. DUDLEY TAYLOR, Department of Anatomy, University of Con- necticut, Farmington, Connecticut 06268 WATSON, DR. STANLEY WAYNE, Woods Llole Oceanographic Institution, Woods Hole, Massachusetts 02543 WEBB, DR. H. MARGUERITE, Department of Biological Sciences, Goucher College, Towson, Maryland 21204 WEBER, DR. ANNEMARIE, Department of Biochemistry, University of Pennsyl- vania, School of Medicine, Philadelphia, Pennsylvania 19104 68 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY WEBSTER, DR. FERRIS, Associate Director for Research, Woods Hole Oceano- graphic Institution, Woods Hole, Massachusetts 02543 WEIDNER, DR. EARL, Department of Zoology, Louisiana State University, Baton Rouge, Louisiana 70803 WEISENBERG, DR. RICHARD, Department of Biology, Temple University, Phila- delphia, Pennsylvania 19104 WEISS, DR. LEON P., Department of Anatomy, The Johns Hopkins University, School of Medicine, Baltimore, Maryland 21205 WEISSMANN, DR. GERALD, Professor of Medicine, New York University, 550 First Avenue, New York, New York 10016 WERMAN, DR. ROBERT, Department of Zoology, Hebrew University, Jerusalem, Israel WHITING, DR. ANNA R., Woods Hole, Massachusetts 02543 WHITING, DR. PHINEAS, Woods Hole, Massachusetts 02543 WHITTAKER, DR. J. RICHARD, Wister Institute of Anatomy and Biology, 36th Street at Spruce, Philadelphia, Pennsylvania 19104 WICHTERMAN, DR. RALPH, Department of Biology, Temple University, Phila- delphia, Pennsylvania 19122 WIERCINSKI, DR. FLOYD J., Department of Biology, Northeastern Illinois Uni- versity, 5500 North St. Louis Avenue, Chicago, Illinois 60625 WIGLEY, DR. ROLAND L., National Marine Fisheries Service, Woods Hole, Massachusetts 02543 WILBUR, DR. C. G., Chairman, Department of Zoology, Colorado State LTniver- sity, Fort Collins, Colorado 80521 WILSON, DR. DARCY B., Department of Pathology, University of Pennsylvania, School of Medicine, Philadelphia, Pennsylvania 19104 WILSON, DR. EDWARD O., Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 WILSON, DR. T. HASTINGS, Department of Physiology, Harvard Medical School, Boston, Massachusetts 02115 WILSON, DR. WALTER L., Department of Biology, Oakland University, Rochester, Michigan 48063 WINTERS, DR. ROBERT WAYNE, Department of Pediatrics, Columbia University, College of Physicians and Surgeons, New York, New York 10032 WITTENBERG, DR. JONATHAN B., Department of Physiology and Biochemistry, Albert Einstein College of Medicine, New York, New York 10461 WRINCH, DR. DOROTHY, Department of Physics, Smith College, Northampton, Massachusetts 01060 WYNNE, DR. MICHAEL J., Department of Botany, University of Texas, Austin, Texas 78710 WYSE, DR. GORDON A., Department of Zoology, University of Massachusetts, Amherst, Massachusetts 01002 WYTTENBACH, DR. CHARLES R., Department of Zoology, University of Kansas, Lawrence, Kansas 66044 YNTEMA, DR. C. L., Department of Anatomy, State University of New York, Upstate^Medical Center, Syracuse, New York 13210 YOUNG, DR. DAVID KENNETH,^Smithsonian Institution, Fort Pierce Bureau, RFD #1, Box 194-C, Fort Pierce^Florida 33450 REPORT OF THE DIRECTOR 69 YPHANTIS, DR. DAVID A., Department of Biochemistry and Biophysics, Uni- versity of Connecticut, Stors, Connecticut 06268 ZACK, DR. SUMNER IRWIN, The Pennsylvania Hospital, University of Pennsyl- vania School of Medicine, Philadelphia, Pennsylvania 19104 ZIGMAN, DR. SEYMOUR, University of Rochester School of Medicine and Den- tistry, 260 Crittenden Boulevard, Rochester, New York 14620 ZIMMERMAN, DR. A. M., Department of Zoology, University of Toronto, Toronto 5, Ontario, Canada ZINN, DR. DONALD J., Department of Zoology, University of Rhode Island, Kingston, Rhode Island 02881 ZORZOLI, DR. ANITA, Department of Physiology, Vassar College, Poughkeepsie, New York 12601 ASSOCIATE MEMBERS ABELSON, DR. AND MRS. PHILIP H. ACKROYD, DR. AND MRS. FREDERICK w. ADELBERG, DR. AND MRS. EDWARD A. ADELMAN, DR. AND MRS. WILLIAM J. ALLEN, Miss CAMILLA K. ALLEN, MRS. NITA M. ALTON, MRS. BENJAMIN ANDERSON, DR. EVERETT ANDREWS, MR. WILLIAM R. ANTHONY, MR. AND MRS. RICHARD A. ARMSTRONG, MRS. PHILIP B. ARNOLD, DR. AND MRS. JOHN BACON, DR. CATHERINE L. BACON, MR. ROBERT BAKALAR, MR. AND MRS. DAVID BALL, DR. AND MRS. ERIC G. BALLANTINE, DR. AND MRS. H. T., JR. BANKS, MR. AND MRS. W. L. BARKER, MRS. ANSON ..BARROWS, MRS. ALBERT W. BARTOW, MRS. CLARENCE W. BARTOW, MRS. FRANCIS D. (S. R.) BARTOW, MRS. PHILIP K. BEALE, MR. AND MRS. E. F. BENNETT, DR. AND MRS. MICHAEL V. L. BERNHEIMER, DR. ALAN W. BIDDLE, DR. VIRGINIA BOETTIGER, DR. AND MRS. EDWARD G. BRADLEY, DR. CHARLES C. BRONSON, MR. AND MRS. SAMUEL C. BROWN, DR. AND MRS. DUGALD E. S. BROWN, DR. AND MRS. F. A., JR. BROWN, DR. AND MRS. THORNTON BUCK, MRS. JOHN B. BUFFINGTON, MRS. ALICE H. BUFFINGTON, MRS. GEORGE BUNTING, DR. MARY I. BURDICK, DR. C. LALOR BURROUGH, MRS. ARNOLD H. BURT, MR. AND MRS. CHARLES E. BUSSER, DR. AND MRS. JOHN H. BUTLER, MRS. E. G. CALKINS, MR. AND MRS. G. N., JR. CAMPBELL, MR. AND MRS. WORTHING- TON, JR. CARLTON, MR. AND MRS. WINSLOW G. CARPENTER, MR. DONALD F. CASHMAN, MR. AND MRS. EUGENE R. CHAMBERS, DR. AND MRS. EDWARD L. CHENEY, DR. AND MRS. RALPH H. CHRISTMAN, DR. AND MRS. GEORGE D. CLARK, DR. AND MRS. ARNOLD M. CLARK, MR. AND MRS. HAYS CLARK, MRS. JAMES McC. (Cynthia) CLARK, DR. AND MRS. LEONARD B. CLARK, MRS. LEROY (Edna A.) CLARK, MR. AND MRS. W. VAN ALAN CLEMENT, DR. AND MRS. A. C. CLEMENTS, MR. AND MRS. DAVID T. CLOWES, MR. ALLEN W. CLOWES, DR. AND MRS. G. H. A., JR. (Margaret) COCHRAN, MR. AND MRS. F. M ORRIS COFFIN, MR. AND MRS. JOHN B. 70 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY COHEN, MRS. SEYMOUR CONNELL, MR. AND MRS. W. J. COPELAND, MR. AND AtRS. PRESTON S. COSTELLO, MRS. DONALD P. CRAMER, MR. AND MRS. IAN D. W. CRANE, MR. JOHN CRANE, JOSEPHINE, Foundation CRANE, Miss LOUISE CRANE, MR. STEPHEN CRANE, MRS. W. CAREY CRANE, MRS. W. MURRAY CROCKER, MR. AND MRS. PETER J. CROSSLEY, MR. AND MRS. ARCHIBALD M. CROWELL, DR. AND MRS. SEARS CURTIS, DR. AND MRS. W. D. DAIGNAULT, MR. AND MRS. A. T. DANIELS, MR. AND MRS. BRUCE G. DANIELS, MRS. F. HAROLD DAY, MR. AND MRS. POMEROY DEMELLO, MR. FREDERICK DRAPER, MRS. MARY C. DuBois, DR. AND MRS. A. B. DuPoNT, MR. A. FELIX, JR. DYER, MR. AND MRS. ARNOLD EASTMAN, MR. AND MRS. CHARLES E. EBERT, DR. AND MRS. JAMES D. EGLOFF, DR. AND MRS. F. R. L. ELLIOTT, MRS. ALFRED M. ELSMITH, MRS. DOROTHY O. EWING, DR. AND MRS. GIFFORD C. FACHON, MRS. EVANGELINE M. FENNO, MRS. EDWARD N. FERGUSON, DR. AND MRS. J. J., JR. FINE, DR. AND MRS. JACOB FIRESTONE, MR. AND MRS. EDWIN FISHER, MR. FREDERICK S., Ill FISHER, DR. AND MRS. SAUL H. FRANCIS, MR. AND MRS. LEWIS W., JR. FRIES, MR. AND MRS. E. F. B. FYE, DR. AND MRS. PAUL M. GABRIEL, DR. AND MRS. MORDECAI L. GAISER, DR. AND MRS. DAVID W. GALTSOFF, DR. AND MRS. PAUL S. GAMBLE, MR. AND MRS. RICHARD B. GARFIELD, Miss ELEANOR GAYTON, MR. GARDNER F. GELLIS, DR. AND MRS. SYDNEY GERMAN, DR. AND MRS. JAMES L., Ill GIFFORD, MR. AND MRS. JOHN A. GIFFORD, DR. AND MRS. PROSSER GIFFORD, MRS. W. M. GILBERT, DR. AND MRS. DANIEL L. GlLCHRIST, MR. AND MRS. JOHN M. GILDEA, DR. MARGARET C. L. GILLETTE, MR. AND MRS. ROBERT S. GLASS, DR. AND MRS. H. BENTLEY GLAZEBROOK, MRS. JAMES R. GLUSMAN, DR. AND MRS. MURRAY GOLDMAN, DR. AND MRS. ALLEN S. COLORING, DR. IRENE P. GOLDSTEIN, MRS. MOISE H., JR. GOOD, Miss CHRISTINA GRAHAM, DR. AND MRS. HERBERT W. GRAHAM, MR. AND MRS. JAMES D., SR. GRANT, DR. AND MRS. THEODORE J. GRASSLE, MR. AND MRS. J. K. GREEN, Miss GLADYS M. GREENE, MR. AND MRS. WILLIAM C. GREER, MR. AND MRS. W. H., JR. GREIF, DR. ROGER L. GROSCH, MRS. DANIEL S. GRUSON, MR. AND MRS. EDWARD GULESIAN, MR. AND MRS. PAUL J. GUNNING, MR. AND MRS. ROBERT HAMLEN, MRS. J. MONROE HANCOX, CAPT. AND MRS. FREDERICK HANDLER, DR. AND MRS. PHILIP HANNA, MR. AND MRS. THOMAS C. HARE, DR. AND MRS. H. GERALD HARRINGTON, MR. AND MRS. R. D. HARVEY, DR. AND MRS. EDMUND N., JR. HARVEY, DR. AND MRS. RICHARD B. HASKINS, MRS. GARY L. HEFFERON, DR. RODERICK HELLMAN, MR. JOHN R. HENLEY, DR. CATHERINE HIBBARD, Miss HOPE HILL, DR. AND MRS. ALFRED T. HILL, MRS. SAMUEL E. HlRSCHFELD, MRS. NATHAN B. HOCKER, MR. AND MRS. LON HOPKINS, MRS. HOYT S. HOUGH, MR. AND MRS. GEORGE A., JR. HOUGH, MR. AND MRS. JOHN T. REPORT OF THE DIRECTOR 71 HOUSTON, MR. AND MRS. HOWARD E. HUETTNER, DR. AND MRS. ROBERT HUNZIKER, MR. AND MRS. HERBERT E. INOUE, MRS. SHINYA ISSOKSON, MR. AND MRS. ISRAEL JANNEY, MR. AND MRS. WISTAR JEWETT, MR. AND MRS. G. F., JR. JONES, MR. AND MRS. DEWITT, III JORDAN, DR. AND MRS. EDWIN P. KAHLER, MR. AND MRS. GEORGE A. KAHLER, MRS. ROBERT W. KAHN, DR. AND MRS. ERNEST KAIGHN, DR. AND MRS. MORRIS E. KEITH, MRS. HAROLD C. KEITH, MR. JEAN R. KENNEDY, DR. AND MRS. EUGENE P. KENEFICK, MRS. R. G. KEOSIAN, MRS. JESSIE KlNNARD, MR. AND MRS. L. R. KOHN, DR. AND MRS. HENRY I. KOLLER, DR. AND MRS. LEWIS R. LASSALLE, MRS. NORMAN LASTER, DR. AND MRS. LEONARD LAWRENCE, MR. FREDERICK V. LAWRENCE, MRS. WILLIAM LAZAROW, DR. AND MRS. ARNOLD LEMANN, MRS. LUCY B. LENHER, DR. AND MRS. SAMUEL LEVINE, DR. AND MRS. RACHMIEL LEVY, DR. AND MRS. MILTON LILLIE, MRS. KARL C. LOBB, PROF. AND MRS. JOHN LOEB, DR. AND MRS. ROBERT F. LONG, MRS. G. C. LORAND, MRS. L. LOVELL, MR. AND MRS. HOLLIS R. LOWENGARD, MRS. JOSEPH LURIA, DR. AND MRS. S. E. MACKEY, MR. AND MRS. WILLIAM K. MACNARY, MR. B. GLENN MACNICHOL, DR. AND MRS. EDWARD J..JR. MARKS, DR. AND MRS. PAUL A. MARSLAND, DR. AND MRS. DOUGLAS MARVIN, DR. DOROTHY H. MATHER, MR. AND MRS. FRANK J., Ill MAYOR, MRS. JAMES W., SR. MCCUSKER, MR. AND MRS. PAUL T. MCELROY, MRS. NELLA W. McGlLLICUDDY, DR. AND MRS. J. J. McKENZIE, MR. AND MRS. KENNETH C. MC-LANE, MRS. T. THORNE McLARDY, DR. AND MRS. TURNER MEIGS, MR. AND MRS. ARTHUR MEIGS, DR. AND MRS. J. WISTER METZ, MRS. CHARLES B. MEYERS, MR. AND MRS. RICHARD MILKMAN, DR. AND MRS. ROGER D. M CLONE Y, DR. ALBERT M. MONTGOMERY, DR. AND MRS. CHARLES H. MOORE, DR. AND MRS. JOHN W. MORRELL, DR. FRANK MORSE, MR. AND MRS. CHARLES L., JR. MORSE, MR. AND MRS. RICHARD S. MOSES, MR. AND MRS. GEORGE L. MOUL, MRS. EDWIN T. NEUBERGER, MRS. HARRY H. NEWTON, Miss HELEN K. NICHOLS, MRS. GEORGE NlCKERSON, MR. AND MRS. FRANK L. NORMAN, MR. AND MRS. ANDREW E. ORTINS, MR. ARMAND PACKARD, MRS. CHARLES PARK, MR. MALCOLM S. PARK, MR. AND MRS. FRANKLIN A. PARMENTIER, MR. GEORGE L. PATTEN, MRS. BRADLEY M. PENDERGAST, MRS. CLAUDIA PENDLETON, DR. MURRAY E. PENNINGTON, Miss ANNE H. PERKINS, MR. AND MRS. COURTLAND D. PERSON, DR. AND MRS. PHILIP PETERSON, MR. AND MRS. E. GUNNAR PHILIPPE, MR. AND MRS. PIERRE PORTER, DR. AND MRS. KEITH R. PROSSER, MRS. C. LADD PUTNAM, MR. AND MRS. W. A., Ill RATCLIFFE, MR. THOMAS G., JR. RAYMOND, DR. AND MRS. SAMUEL REDFIELD, DR. AND MRS. ALFRED C. RENEK, A!R. AND MRS. MORRIS REYNOLDS, DR. AND MRS. GEORGE 72 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY REZNIKOFF, DR. AND MRS. PAUL RIGGS, MR. AND MRS. LAWRASON, III RIINA, MR. AND MRS. JOHN R. ROBB, Ms. ALISON A. ROBERTSON, DR. AND MRS. C. W. ROBINSON, DR. AND MRS. DENIS M. ROGERS, MRS. JULIAN Ross, MRS. JOHN ROOT, MRS. WALTER S. ROWE, MRS. WILLIAM S. RUGH, DR. AND MRS. ROBERTS RUSSELL, MR. AND MRS. HENRY I). RYDER, MR. AND MRS. FRANCIS D. SAUNDERS, DR. AND MRS. JOHN W. SAUNDERS, LAWRENCE, Fund SAVERY, MR. ROGER SAWYER, MR. AND MRS. JOHN E. SCHLESINGER, MRS. R. WALTER SCHROEDER, MR. RlCHARD F. SCOTT, MRS. GEORGE T. SEARS, MR. AND MRS. HAROLD B. SHEMIN, DR. AND MRS. DAVID SHAPIRO, MRS. HOWARD E. SHEPRO, DR. AND MRS. DAVID SHERMAN, DR. AND MRS. IRVIN SMITH, DR. FREDERICK SMITH, MRS. HOMER P. SMITH, MR. AND MRS. VANDORN C. SPEIDEL, DR. AND MRS. CARL C. STEINBACH, DR. AND MRS. H. B. STETTEN, DR. AND MRS. DEWITT, JR. STUNKARD, DR. HORACE STURTEVANT, MRS. P. SWANSON, DR. AND MRS. CARL P. SWEENY, DR. AND MRS. THOMAS D. SWOPE, MR. AND MRS. GERARD L. SWOPE, MR. AND MRS. GERARD, JR. SWOPE, Miss HENRIETTE H. TAYLOR, DR. AND MRS. W. RANDOLPH THOMAS, DR. AND MRS. LEWIS TIETJE, MR. AND MRS. EMIL D. TODD, MR. AND MRS. GORDON F. TOLKAN, MR. AND MRS. NORMAN N. TOMPKINS, MRS. B. A. TRACER, MRS. WILLIAM TROLL, DR. AND MRS. WALTER TURNER, MRS. ROBERT VALOIS, MR. AND MRS. JOHN VEEDER, MRS. RONALD A. WAKSMAN, DR. AND MRS. BYRON H. WAKSMAN, MRS. SELMAN A. WALLACE, DR. AND MRS. STANLEY L. WANG, DR. AND MRS. AN WARE, MR. AND MRS. J. LINDSAY WARREN, DR. AND MRS. SHIELDS WATT, MR. AND MRS. JOHN B. WEISBERG, MR. AND MRS. ALFRED M. WEXLER, MR. AND MRS. ROBERT H. WHEATLEY, DR. MARJORIE A. WHEELER, MR. AND MRS. HENRY WHEELER, DR. AND MRS. PAUL S. WHEELER, DR. AND MRS. RALPH E. WHITELEY, MR. AND MRS. G. C., JR. WHITING, DR. AND MRS. PHINEAS W. WHITNEY, MR. AND MRS. GEOFFREY G., JR. WlCKERSHAM, MR. AND MRS. A. A. TlLNEY WlCKERSHAM, MRS. JAMES H., JR. WlCHTERMAN, DR. AND MRS. RALPH WlLBER, DR. AND MRS. CHARLES G. WILHELM, DR. HAZEL S. WILSON, MRS. EDMUND B. WILSON, MR. AND MRS. ROBERT E., JR. WlTMER, DR. AND MRS. ENOS E. WOLFE, DR. CHARLES WOLFINSOHN, MR. AND MRS. WOLFE WRINCH, DR. PAMELA N. YNTEMA, DR. AND MRS. CHESTER L. ZWILLING, MRS. EDGAR V. REPORT OF THE LIBRARIAN The library book section has expanded so rapidly that more room had to be found in the Lillie Building for additional shelving. Over 100 books were pur- chased with the Associates 1971 gift and a total of 710 books were added to this section this year. REPORT OF THE TREASURER 73 The Woods Hole Oceanographic Institution received a substantial amount of money for books and will order approximately 1,500 volumes in 1974. Therefore Room 306 will be converted to library stack space and all books relating to Ocean Engineering, Physics, Mathematics and Marine Policy will be shelved temporarily in that room on the third floor. Interlibrary loan requests continue to come in at a greater rate than the pre- vious year. We average about 20 to 30 requests per day. Two staff members left for other positions — Virginia Brandenberg, on the library staff for ten years, left to work in the Boston area and Cathy Glynn, Periodicals Librarian, also resigned. They have been replaced by Elizabeth Fuseler, formerly a librarian at the University of Pennsylvania, and Terry McKee, a 1973 graduate of Skidmore College. The library is currently receiving 2,624 serial titles and holdings now total 151,764 volumes. VI. REPORT OF THE TREASURER The market value of the General Endowment Fund and the Library Fund at December 31, 1973, amounted to $2,154,552 and the corresponding securities are entered in the books at a value of $1,633,192. This compares with values of $2,579,583 and $1,624,597 respectively, at the end of the preceding year. The average yield on the securities was 4.18% of the market value and 5.52% of the book value. Uninvested principal cash was $2, 433. Classification of thesecurities held in the Endowment Fund appears in the Auditor's Summary of Investments. The market value of the Pooled Securities at December 31, 1973, amounted to $1,088,355 as compared to book values of $930,340. These figures compare with values of $1,238,943 and $922,227 respectively, at the close of the preceding year. The average yield on the securities was 4.00% of the market value and 4.73% of the book value. Uninvested principal cash was in the amount of $823. The proportionate interest in the Pool Fund Account of the various funds, as of December 31, 1973, is as follows: Pension Funds 30.41% General Laboratory Investment 28.04% F. R. Lillie Memorial Fund 2.04% Other : Bio Club Scholarship Fund 0.53% Rev. Arsenius Boyer Scholarship Fund 0.64% Garry N. Calkins Fund 0.67% Allen R. Memhard Fund 0.12% Lucretia Crocker Fund 2.21% E. G. Conklin Fund 0.37% Jewett Memorial Fund 0.19% M. H. Jacobs Scholarship Fund 0.27% Herbert W. Rand Fellowship 18.84% Mellon Foundation 8.89% Mary Rogick Fund 1.94% Swope Foundation 4.90% 74 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY Donations from MBL Associates for 1973 amounted to $17,250 as compared with $10,755 for 1972. Unrestricted gifts from foundations, societies and com- panies amounted to $27,967. During the year we administered the following grants and contracts: MBL Insti- Investigators 'Framing tntional 4 NIH 3 NIH 1 NSF 5 NSF 1 Research Corp. 1 EPA 1 MWPC 1 NOAA 1 API 1 Damon Runyon Cancer 1 Merrill Trust 15 4 1 Most of the federally funded research grants and contracts provided for reim- bursement of indirect costs on a cost per square foot basis for the laboratory space assigned to each project. A provisional rate of $12.25 per square foot was in effect throughout the year. Training courses supported by NIH grants were funded for indirect costs at a rate of 8% of allowable direct costs. The following is a statement of the auditors: To the Trustees of Marine Biological Laboratory, Woods Hole, Massachusetts: We have examined the balance sheet of Marine Biological Laboratory as of December 31, 1973 and the related statements of operating expenditures and income and funds for the year then ended. Our examination was made in accordance with generally accepted auditing standards, and accordingly included such tests of the accounting records and such other auditing proce- dures as we considered necessary in the circumstances. We previously ex- amined and reported on the 1972 financial statements. In our opinion, the aforementioned financial statements (pages 75 to 79) present fairly the financial position of Marine Biological Laboratory at December 31, 1973 and 1972, and its operating expenditures and income for the years then ended, and the changes in its funds for the year ended Decem- ber 31, 1973, and in conformity with the accounting principles referred to in Note A to the financial statements applied on a consistent basis. The supplementary schedules (page 80) included in this report were obtained from the Laboratory's records in the course of our examination and, in our opinion, are fairly stated in all material respects in relation to the financial statements taken as a whole. Boston, Massachusetts March 22, 1974 COOPERS & LYBRAND REPORT OF THE TREASURER 75 It will be noted from the operating statement that the Laboratory activities for the year under review, amounted to a figure of a little over 1.8 million dollars, which amount is 150 thousand more than the preceding year. ALEXANDER T. DAIGNAULT, Treasurer MARINE BIOLOGICAL LABORATORY BALANCE SHEETS December 31, 1973 and 1972 Investments 1973 1972 Investments held by Trustee: Securities, at cost (approximate market quotation, 1973— $2,154,552; 1972— $2,579,583) $1,633,192 $1,624,597 Cash.. 2,433 1,850 1,635,625 1,626,447 Investments of other endowment and unrestricted funds: Pooled investments, at cost (approximate market quotation, 1973— $1,088,355; 1972— $1,238,943) 930,340 922,227 Other investments .... 1,366,230 1,094,752 Cash 1,571 31,516 Due from current fund- 189,168 131,927 $ 4,122,934 $ 3,806,869 Plant Assets Land, buildings, library and equipment, at cost 12,454,336 12,453,416 Less allowance for depreciation (Note A) . . 2,726,028 2,444,052 $ 9,728,308 $10,009,364 Current Fund Assets Cash 65,310 388,130 Accounts receivable (U. S. Government, 1973— $33,739 ; 1972— $88,186) 252,589 235,903 Inventories of supplies and bulletins 48,533 43,514 Other assets 5,912 5,581 Due to endowment funds. . (189,168) (131,927) 183,176 $ 541,201 76 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MARINE BIOLOGICAL LABORATORY BALANCE SHEETS December 31, 1973 and 1972 Invested Funds 1073 1972 Endowment funds given in trust for benefit of Marine Biological Laboratory.. $1,635,625 $1,626,447 Endowment funds for awards and scholarships: Principal.... 449,777 427,702 Unexpended income 65,715 60,795 515,492 488,497 Unrestricted funds functioning as endowment . . . 1,726,603 1,435,918 Retirement fund.... 357,439 316,512 Pooled investments — accumulated loss. . (112,225) (60,505) $ 4,122,934 $ 3,806,869 Plant Funds Funds expended for plant, less retirements 12,454,336 12,453,416 Less allowance for depreciation charged thereto 2,726,028 2,444,052 $ 9,728,308 $10,009,364 Current Fund Liabilities and Balances Accounts payable and accrued expenses 20,677 26,333 Advance subscriptions 35,899 28,293 Unexpended grants — research 39,987 77,774 Unexpended balances of gifts for designated purposes 124,124 87,949 Current fund.. (37,511) 320,852 $ 183,176 $ 541,201 The accompanying note is an integral part of the financial statements. Note A — Accounting Principles: The following accounting principles have been reflected in the accompanying financial statements: 1. Investments are stated at cost. 2. Investment income is recorded on a cash basis. 3. Operating income is recorded when earned. 4. Expenses are recorded on an accrual basis. 5. Depreciation has been provided for plant assets at annual rates ranging from 1% to 5% of the original cost of the assets. REPORT OF THE TREASURER 77 \O IO r^ -*t OO t^ \o 1 — I vO ON ON ON § r ^— i O Csl PD ON Csl o *— 1 »— 1 -, IO »-H t^- ON — 2 00 "0 o" Csl rsr fv\ """'•* ^^ OO ^"* ^O ™^ ^^ ^~^ O*1 '"O OO v( 1~» T ON C 2 O ON 2 OO r<0 o rvi •-I h "0 -f — ^ ^2^ "^ ^o ^] o -s: ^ ^ ^. uO'^l'^'Ot^iOrsi -^OO t-~ O 10 t^ O »— O O -H O 0 LO f Csl « OO \£ ON C 0 "0 rsi o «-i •* •3 o T— 1 oo 3 ON 0 3 IT) CM 00 "1 t^ 10 o ABORATORV HTURES AND INCOMI 1973 and 1972 1973 •*» -f" 0s . vo" ^ \C" — — -H ^H" ^ S ^ 00 00 00 2 !^s5^ u~>ooc>r^ t^ 111 ss'l" ^ cT a 01 C c Cs] C O •p- O 0 3 fN 3 "t 3 H i 3 ON H O N CO 06" 00 Csl Csl i w j S S PH L- OJ 4J * ' • O — a F— a 01 • • • • • -a O S '«•-•• en 4J u — -M tC 5 • nj O) gj •I 1 £ 1 1 11 1 | Administration Plant oneration Grant expendit Other 78 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY t^ oo o_ ONON -^^ -^ -^ 01 ro NO t^» LO <*O Ol t"*» LO NO O1 ON ON <°O rO '~( ^^~ ON LO OO ffj ^ 01 ON O r^- LO -t 0 O 00 CN -* CN ON" LO ^O "-'. 'r. ON" t^ CN *— 4 \^^ t^« ON Ol *-H f^N f^* ^N ^\ V^ ^^ l^J ^J* NO Ol LO ^2 ^O *™H t^. NO" of t-- OO r^ 01 ^o" ON <^ 1 00 CN O 00 O — i GOO -t-LOt^oOOON OILO OO OO^oivOOOLO S LO LO ON r— O ON 0 01 4— 1 § oo NO o -f t~-^ \^ LO t^^ o.] \^ ro *^* ^^ *~H oo" 00 oo 4—1 ON" —" NO 00 00 ON O' oo_ o 00 OO LO o" CN" 2 ON" I— LO O vo f~i »o oo \r. O—H OOO-tLOt^-00 ON CNLO OOOOrfOlO LO 00 ON LO LO ON r— O ON o «-H NO" o CO o" CN ^ OO — < O) ro O ON O LO 00" O 4— t ON" — NO OO ro 01 00 >/} C oo ON— OOO^LOt^-OO ON, •rfi ^^ 01 QN -rt1 ^ ^ — i — f* o-i LOLO OOOO^CNNO LO NO ON •*-* 0 o" ~ — O] O: ^^ Ol f*O O ON O LO -^vO LO t^~ Ol NO ro — f '—i ON" OO „ NO CN r ^H 01 1^ ON ON 1 NO 0 £ 01 00" OO -f OJ — u . . . . . . rt , ^ ' • i> Pl 4—- , o •^ 4-> Cu 0 g 15 u o a • -™ y1^ ^ -a r- . — * m •a 'j~. c 2 — aj LT Co T> "O 'Si « o.y u & o g •£ tt tJ *> tn 2 -u c r- •-1 P . — S u -a k X £ 5 "^ c _OJ oi d . § « £ •5 ^= 'i g * y §-| ^C-5,fe 1 » g:=i^ | £ .p< P Q J PQ co Apparatus. Supply. . . . Administration. Investment inco Gifts used for ci Allowance for in Grants for supp Other i! •~ ^ i - o U Reduction in plant f Rxcess current evuei The accompany REPORT OF THE TREASURER s s 00 j S ^ S —• oo cs o -f t— <-O O LO t^> ON OO OO O *i* NO _ ^ LO >-l o ON" « £ g O O 01 O O ^5 O1* . O °° ,9 *— v l/J t5? 1_ o . « LO t? •§ S o S I 8 o" R 5 g oo _- oo \O Ol OO ^H X 0) c O ON" 00 VO o to o t— ^^ \O r~ "0 l~- LO I-— O OO f& i- o ^ 2 S « S •sis f 4J "S Other gifts an Grants for rese il i w •8*8 4J X .2 s I- U a. x 2 w a ac- 0 o rt > O u o 3 rt o -i-1 ^&H_!O 80 ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY MARINE BIOLOGICAL LABORATORY SUMMARY OF INVESTMENTS December 31, 1973 Securities held by Trustee: General endowment fund: U. S. Government securities Corporate bonds Preferred stocks. Common stocks. Cost 25,065 595,553 ,51,094 670,397 Per- cent of Total 1.9 45.0 2.4 50.7 Market Quotations Per- cent of Total > 24,800 1.4 465,376 26.6 17,625 1.0 1,242,107 71.0 Investment Income 1973 1,813 29,025 1,650 41,401 1,322,109 100.0 1,749,908 100.0 73,889 General educational board endowment fund : U. S. Government securities. Other bonds Common stocks. . 51,113 145,965 114,005 16.4 46.9 36.7 50,592 109,290 244,762 12.5 27.0 60.5 Total securities held by Trustee $1,633, 192 $2,154,552 Investments of other endowment and unrestricted funds: 3,698 6,690 5,824 311,083 100.0 404,644 100.0 16,212 90,101 Pooled investments: U. S. Government securities 6,522 Corporate bonds 379,200 40.8 312,908 28.8 13,868 Preferred stocks 62,207 6.7 63,738 5.8 1,120 Common stocks 488,933 52.5 711,709 65.4 22,522 930,340 100.0 $1,088,355 100.0 44,032 Other investments: U. S. Government securities 27,764 1,323 Other bonds 15,029 750 Common stocks 588,016 21,697 Preferred stocks 2,871 104 Real estate 17,550 Short-term commercial notes 715,000 38,633 1,366,230 62,507 Total investments of other en- dowment and unrestricted funds $2,296,570 Total 196,640 Custodian's fees charged thereto (7,590) Total investment income $189,050 Reference : Biol. Bull., 147 : 81-94. (August, 1974) CALORIFIC VALUES IN THE PHYLUM PLATYHELMINTHES : THE RELATIONSHIP BETWEEN POTENTIAL ENERGY, MODE OF LIFE AND THE EVOLUTION OF ENTOPARASITISM P. CALOW AND J. B. JENNINGS Department of Zoology, University of Glasgow, Scotland, U. K., and Department of Pure and Applied Zoology, University of Leeds, England, U. K. An analysis of the calorific values (kcal/g) of seventeen species of animals from six phyla has shown that they have a skewed distribution with a modal fre- quency at or near the lower range limit (Sloboclkin and Richman, 1961). This was regarded as support for the hypothesis that natural selection generally favors production of the maximum number of progeny, rather than a high energy con- tent per unit weight. The latter is favored only sporadically and under special circumstances such as preparation for a period of fasting or stress (Slobodkin, 1962). To test this hypothesis further we have undertaken a survey of calorific values within the Platyhelminthes. This phylum shows a broad spectrum of life styles which range from the free-living predatory habits of most Turbellaria, through various degrees of facultative or obligate ecto- or entocommensalism (other Turbel- laria), to the obligate ecto- or entoparasitism of a few Turbellaria and all Mono- genea, Digenea and Cestoda (Jennings, 1971; 1973). The symbiotic species show high fecundity relative to the free-living forms (Hyman, 1951), apparently as an adaptation to their mode of life, and by the above hypothesis could be expected to have correspondingly lower calorific values. The entosymbiotic species also differ from the free-living ones in that they contain large amounts of carbohydrate, usually in the form of glycogen. This accounts for up to 17.73% of the dry weight in entocommensal turbellarians, 14.77% in digeneans and 26.99% in cestodes, but only 9.96% in free-living flat- worms (data from various sources, summarized by Jennings, 1973). The latter, though, generally store lipid and levels of up to 19.83% of the dry weight have been recorded (Jennings, 1957; Mettrick and Boddington, 1972). This dichotomy in the form in which potential energy is stored has long been regarded as stemming from differences in the oxygen tensions of the environments of free-living and entosymbiotic species. Habitats within other organisms, and especially alimentary tracts, often have low oxygen tensions (Campbell, 1931 ; von Brand, 1946; Rogers, 1949; Crompton, Shrimpton and Silver, 1965) and in such situations carbohy- drates may well be the most appropriate metabolic substrates. There is certainly much evidence, ably reviewed by Read (1968), that many parasitic flat worms have metabolic processes suited to life under semi-anaerobic conditions. However, as Crompton, et a!., 1965 emphasize, the partial pressure of oxygen in regions favored by entosymbiotes still usually exceeds the value of 5 mm Hg given by Hill 81 P. CALOW AND J. B. JENNINGS (1936) as the threshold necessary for satisfactory functioning of cytochrome oxidase. Further, some entosymbiotes live in fully aerobic situations and yet store glycogen (Halton, 1967; Mettrick and Jennings, 1969), so that this phenomenon cannot be linked directly with oxygen tension. Recognizing this, Jennings and Mettrick (1968) have suggested that glycogen storage, instead, is linked in some way with high fecundity. Carbohydrates, having lower calorific values than lipids (Cummins and Wuychek, 1971), may reduce whole body values for the ento- symbiotes and this notion therefore conforms to the hypothesis cited by Slobodkin and Richman (1961) that fecundity generally implies a low calorific value. Consequently, in addition to testing the general applicability of this hypothesis, by examining calorific values in the phylum Platyhelminthes, we have also ex- amined the validity of its particular implications for entosymbiotic animals. This has been done by selecting species for examination which are representative of the various life styles in this phylum so that any correlation between potential energy and mode of life would be revealed. MATERIAL AND METHODS The species investigated are listed systematically together with some details of their habitats and modes of life. The few species for which calorific values have already been published are also listed, with the authority, since the data for these will be incorporated with those from the present study. Turbellaria Rhabdocoela Syndesmis jranciscana (Lehmann). Entocommensal in the gut and coelom of Strongylocentrotits purpuratus Clark (Echinoidea), from Bodega Bay, California. Allocococla Plagiostoniiiiii sitlplinrcnin von Graff. Free-living, marine littoral, Filey Brigg, Yorkshire, England. Tricladida Bdelloura Candida (Girard). Ectocommensal on the gills and limbs of Limulus Polyphemus L. (Xiphosura), Cape Cod, Massachusetts. Dendrocoelum lactcinn (O. F. Miiller). Free-living, freshwater littoral, Loch Lomond, Scotland. Dugcsia lugubris (O. Schmidt). As D. lactcinn. Planana torva (O. F. Miiller). As D. lactcnm. Pulycclis nigra (O. F. Miiller). Free-living, freshwater littoral, Malham Tarn, Yorkshire, England. Procerodcs ulvae (Diesing). Free-living, marine littoral, Isle of Cumbrae, Scotland. Bipalhtm kezvense Moseley. Free-living, terrestrial. Data from Slobodkin and Richman, 1961. Dugesia tigrina (Girard). Free-living, freshwater littoral. From Slobodkin and Richman, 1961. CALORIFIC VALUES IN PLATYHELMINTHES Phagocata gracilis (Haldeman). As D. tigrina. From Minshall, G. W., un- published data, cited in Cummins and Wuychek, 1971. Phagocata inorgani (Stevens and Boring). As D. tigrina. From Teal, 1957. Phagocata woodworthi Hyman. As D. tigrina. From Teal, 1957. Monogenea Polyopisthocotylea Diclidophora mcrlangi (Kroyer). Ectoparasitic on the gills of the whiting Gadus merlangus (L.) (Teleostei), Whitby, Yorkshire, England. Entodbclla soleae Johnston. Ectoparasitic on the skin of the sole Solea solea (L.) (Teleostei), Plymouth, England. Polystoma integcrrimum Rudolphi. Entoparasitic in the urinary bladder of Rana temporaria L. (Amphibia), Adel, Yorkshire, England. Digenea Prosostomata Fasciola hcpatica L. Entoparasitic in the biliary system of domestic sheep Ovis arics L. (Mammalia), slaughtered in Glasgow, Scotland. Haplometra cylindracca Looss. Entoparasitic in the lungs of Rana temporaria L. (Amphibia), Adel, Yorkshire, England. Cestoda Cyclophyllidca Hymcnolcpis duninnta Rudolphi. Entoparasitic in the intestine of laboratory rats, Rattus norvcgicns Berkenhout, \\'istar strain (Mammalia), Department of Zoology, University of Toronto. Pseudophyllidea Triacnophorns nodulosus Pallas. Entoparasitic in the intestine of the pike Eso.r Indus L. (Teleostei), R. Ouse, Yorkshire, England. Tetraphyllidca EcheneibotJiriuin variable Van Beneden. Entoparasitic in the spiral valve of the skate Raia clai'ata L. (Elasmobranchii), Whitby, Yorkshire, England. The four freshwater species of triclads (D. lactcitin. I), ingnbris. P. torva and P. nigra ) were kept in the laboratory for two weeks prior to examination and fed daily on the oligochaete Tubijc.v, apart from one subgroup of P. nigra which were starved for this period. All other species were processed immediately after col- lection from the field, or removal from their hosts. Calorimetry Living healthy specimens were rinsed briefly in distilled water and then either freeze dried to constant weight (Syndcsiiiis franciscana and Hymenolepis duninnta ) 84 P. CALOW AND J. B. JENNINGS TABLE I Calorific values (kcal/g ash-free dry wt), Joule equivalents and ash contents of free living and symbiotic Platyhelminthes Species and mode of life kcal/g ash-free dry wt ±95% conf. interval kj/g ash- free dry wt Ash% dry wt ±95% conf. interval** Number of mea- surements Free living D. lacteum 6.316 ±0.317 26.439 5.82 ± 0.39 10 D. lugubris P. nigra P. sulphiireum P. torva 6.295 ± 0.222 6.420 ± 0.338 6.798| 6.382 ± 0.451 26.351 26.874 28.456 26.715 3.50 ± 0.27 6.10 ± 0.71 7.71 3.79 ± 0.42 10 10 1 10 P. ulvae 6.000 ± 0.258 25.116 6.95 ± 0.72 3 B. kewense 5.684 ± 0.124 23.793 * * D. tigrina P. gracilis P. morgani P. woodworthi Mean Ectocommensal 6.286 ± 0.338 6.377 ± 0.137 5.600f 5.600| 26.313 26.694 23.442 * * * * * * 6.216 ±0.258 26.019 5.65 ± 0.41 B. Candida Entocommensal 5.897f 24.685 10.09f 1 S. franciscana Ectoparasitic D. merlangi E. soleae Mean Entoparasitic (except Cestoda) F. hepatica H. cylindracea P. integerrimiim Mean Cestoda 5.080 ±0.216 5.372| 5.668f 21.265 22.487 23.726 11.72 ± 2.35 8.98f 6. 10f 3 1 1 10 1 1 5.520 23.107 7.51 5.205 ±0.201 5.124f 5.372f 21.788 21.449 22.487 5.07 ± 0.37 6.53f 5.98f 5.234 ± 0.292 21.908 5.85 ±0.15 E. variable 5.164 ± 0.316 21.616 3.86 ± 0.39 5 H. dim in ut a 5.817 ±0.203 24.350 4.30 ± 0.41 10 T. nodulosus Mean Grand mean for all species 5.972f 24.999 2.l7f 1 5.651 ± 0.561 23.655 3.44 ± 0.67 5.841 ± 0.230 24.452 6.17 ± 0.08 f = insufficient material for replicates. * = data from other authors, % ash and number of measurements not specified. = data were transformed to arcsines before calculation of mean and confidence intervals, and have been retransformed for presentation (Rohlf and Sokal, 1969). or air dried at 60° C for twenty four hours. The dried specimens were ground into powder and calorific values determined for the entire animal. In H. diminuta individuals were also divided into three portions, consisting of immature, mature CALORIFIC VALUES IN PLATYHELMINTHES and gravid proglottids and each portion was then processed separately to detect any differences in energy reserves related to the different reproductive states. Calorific values were determined in a Phillipson micro-bomb calorimeter using standard calorimetric techniques (Phillipson, 1964). Each sample used was of at least 10 mg dry weight and usually was in excess of 15 mg. Routine corrections were applied for fuse wire "glow" and acid production (Paine, 1971). The ash contents were usually less than 10% of the dry weight (only four samples had > 10% ash) and consequently no correction was applied for endothermy (Paine, 1966). Due to the small size of most of the specimens and the difficulties of ob- taining suitable numbers there was not, in most cases, enough material for in- dependent determination of ash content in a muffle furnace. This information was routinely obtained, therefore, from the residues remaining in the bomb after firing. Data were discarded if fuse wire contamination was suspected. Replicate samples were used whenever the amount of material available per- mitted. When possible, up to ten subsamples were examined, means were com- puted together with fiducial limits and the latter were corrected for small sample g sizes (i.e., t. -T=, where t was obtained at the 95% probability level for n-- 1 \n degrees of freedom). Means for species with common modes of life were then averaged. Data were compiled in this fashion because it was considered that the resultant means and variances were of greater importance than those based on individual measurements. Calorific values for free-living triclads obtained by previous investigators were used either directly (e.g., those for B. keivense and D. tiyrina given by Slobodkin and Richman, 1961), or after adjustment to make them comparable with those obtained here. The values given by Teal (1957) for two species of Phagocata were obtained by dichromate digestion and expressed in terms of wet weight. To obtain an ash-free dry weight estimate it has been assumed, from experience in drying samples, that dry weight equals 25% of wet weight and that ash equals 5% of dry weight (the average ash value for the four freshwater triclads studied here being 5.038% as calculated from the data in Table I). RESULTS The potential energy contents of the 21 species of flat worms investigated are listed in Table I, together with their ash contents expressed as a percentage of the dry weight. Energy values are given in terms of kcal/g ash-free dry weight and S. I. (Joule) equivalents are also shown. Fiducial limits on the calorific values ranged between 2.14—7.06% of the means, which is generally considered to be ac- ceptable for biological samples. Before considering these results in detail it should be pointed out that, apart from random manipulative errors, two further factors may have influenced the data. First, no correction has been made for nitrogen production during combus- tion. This may have resulted in an underestimate of the calorific values, but since there appears to be little interspecific variation in nitrogen contents within the Platyhelminthes (Mettrick and Boddington, 1972) comparative treatment of the data is still legitimate (Kersting, 1972). Secondly, bomb combustion, rather than muffle furnace estimates of ash, can introduce systematic errors into calorific 86 P. CALOW AND J. B. JENNINGS MA Cestoda Species with common life styles FIGURE 1. Mean calorific values (kcal/g ash-free dry weight) of flatworms with common life styles. The values for lipid, carbohydrate and MA (the mean value for whole animals from other phyla) are from Cummins and Wuychek ( 1971). values. However, this depends on the- ash content of the materials involved and becomes more extreme as percentage ash increases. At ash levels below 10% (most of the species considered here) the effect becomes trivial and contributes less than ± 1% variation (Reiners and Reiners, 1972). The grand mean of 5.841 =t 0.230 kcal/g ash-free dry weight for all the species studied falls close to the mean value of 5.821 kcal/g ash-free dry weight previously reported for whole animals from a wide range of other phyla (data summarized by Cummins and Wuychek, 1971). Values ranged from a minimum of 5.080 kcal/g ash-free dry weight for the entocommensal S. franciscana to a maximum of 6.798 kcal/g ash-free dry weight for the free-living P. sulphureutn. CALORIFIC VALUES IN PLATYHELMINTHES 87 The mean calorific values for species sharing common life styles are summarized in Figure 1. This shows that the free-living flatworms have a value higher than the mean for whole animals from other phyla (MA) whereas the entocommensal and entoparasitic species, other than cestodes, have much lower values. The ectocom- mensal, ectoparasites and cestodes, however, have values very close to this mean. The differences in calorific values and ash contents hetween the three regions of the cestode H. diminnta are shown in Table II, together with values for the whole worm. The calorific values of immature and mature proglottids are similar to each other, but significantly smaller than that of the gravid proglottids (d -- 2.53, P =£= 0.02 ) . The value for the whole worm is only slightly less than that of the gravid component and is obviously elevated by this. The immature and mature proglottids are more comparable to the whole flatworms of other classes, in terms of growth and reproductive activity, than are the gravid proglottids and the mean calorific value for these two regions (5.554 kcal/g ash-free dry weight) is some- what closer to those for the entocommensal and entoparasitic species than is the value for whole cestodes. Figure 1 also shows average calorific values for pure lipids and carbohydrates. The range between these two extremes represents that which is theoretically possible in organic material but the actual range recorded here within the Platyhelminthes is considerably less. This is also the case in all other organisms so far studied (Cummins and Wuychek, 1971 ) . Starvation for two weeks caused a significant fall in the calorific value of the free-living triclad P. nigra. The value dropped from 6.420 kcal/g ash-free dry weight to 5.050 kcal/g ash-free dry weight, and the ash content fell from 6.10% dry weight to 5.98%. The average size decreased by approximately 30%. DISCUSSION Calorific values in the Platyhelminthes follow the general pattern observed in other phyla in that they are skewed towards the lower possible limit (i.e. the value for carbohydrate. Figure 1), with a mean of 5.S41 kcal/g ash-free dry weight and a range of 5.080-6.798 kcal/g ash-free dry weight. The distribution within this range, however, is not haphazard but shows a definite relationship to the mode of life. Entosymbiotic species, with the exception of the Cestoda, have values which cluster towards the lower limit of the range and are generally less than the mean level for animals from other phyla (MA of Figure 1 and referred to hereafter as "mean animal"), whereas values for the free-living flatworms cluster toward a higher level and have a mean in excess of that for the "mean animal." The ectosymbiotes and cestodes are intermediate between these two extremes. Fluctuations in ash-free calorific values around the "mean animal" level have been identified almost exclusively with fluctuations in tissue lipid concentration (Ostapenya and Sergeev, 1963). This is because proteins and carbohydrates have similar energy equivalents (circa 5.7 and 4.1 kcal/g ash-free dry weight, respec- tively), while lipids have far greater equivalents (ca. 9.5 kcal/g ash-free dry weight). Thus our results confirm that the evolutionary trend within the Platy- helminthes, assuming that the free-living forms are primitive, has been toward a reduction in the amount of lipid stored per unit weight with a concomitant reduc- tion in calorific values. P. CALOW AND J. B. JENNINGS The cestodes are somewhat atypical, when compared with the other ento- symbiotes, in that they have relatively high and variable calorific values (Table I). The latter probably result from variability in lipid contents; high lipid levels have been reported in Hymenolepis diminuta by Webb and Mettrick (1973) but much lower levels are recorded for several other cestodes (Smyth, 1966). In H. diimnuta gravid proglottids contain a higher proportion of lipid than do immature and mature ones (Fairbairn, Wertheim, Harpur and Schiller, 1961 ; Mettrick and Cannon, 1970) and this is reflected in their high calorific value (Table II). If the cestode strobila is viewed as a series of progressively maturing reproductive units, though, rather than as a single individual, it becomes legitimate to compare only the immature and mature proglottids with the whole individuals from other flatworm groups. The gravid proglottids contain the products of reproduction and are more comparable, therefore, to the cocoons of free-living flatworms. The im- mature and mature proglottids have calorific values closer to those of other ento- symbiotes than does the whole cestode (Table II) and so conform more closely to the basic evolutionary trend in the phylum. TABLE II Calorific values (kcal/g ash-free dry wt.), Joule equivalents and ash contents in Hymenolepis diminuta kcal/g ash-free dry wt ± 95% conf. interval kj/g ash- free dry wt Ash % dry wt ± 95% conf. interval Number of mea- surements Whole worm 5.817 ± 0.203 24.35 4.30 ±0.41 10 Immature proglottids Mature proglottids Gravid proglottids 5.559f 5.550 ± 0.241 5.997 ± 0.185 23.27 23.23 25.10 3.30f 3.54 ± 0.39 3.35 ±0.28 1 8 10 f = insufficient material for replicates. The link between calorific value, fecundity and mode of life would seem, there- fore, to be amply demonstrated in the Platyhelminthes. The precise nature of this link is at first difficult to define, but the situation becomes clearer when it is recognized that "survival of the fittest" must be interpreted in terms of the survival rate of offspring rather than entirely in terms of survival of the adult. Thus the "fittest" individual is one which produces the greatest number of viable offspring, and selection will operate in favor of using any energy available in excess of growth and maintenance requirements for progeny production rather than for ac- cumulation of endogenous food stores. In Slobodkin's words "there is a selective advantage in increasing fecundity but not adiposity" (Slobodkin, 1962, page 71). From this viewpoint it is the free-living flatworms which are atypical in the Platyhelminthes and the onus is put on explaining their high calorific values, re- sulting from lipid stores, rather than on explaining the low energy equivalents, resulting from glycogen stores, in the entosymbiotes. Lipid has the advantage over glycogen as a long-term energy reserve in that it allows more calories to be stored per unit weight of tissue or, conversely, less weight need be added per unit energy stored. Energy storage is essential in species likely to experience prolonged periods of hardship, and lipid accumulation CALORIFIC VALUES IN PLATYHELMINTHES with concomitant increase in calorific value occurs, for example, in birds (Odum, Marshal and Marples, 1965) and termites (Wiegert and Coleman, 1970) prior to migration. It has also been suggested that lipid storage is an adaptation for locomotion, because of the favorable weight/potential energy ratio (Stetten and Stetten, 1956) and this argument could be applied in the case of the free-living flatworms. The principal cause, though, behind the transformation of excess energy into lipids in the free-living species must be that it increases the chances of progeny survival by allowing individuals to survive poor feeding conditions and then subsequently breed. This strategy is obviously advantageous in iteroparous organisms like the triclad Turbellaria, which breed more than once in their life span. Here, survival of adults that might already have bred is equally as im- portant as survival of juveniles. This is in contrast to semelparous organisms, breeding only once in the life span, where juvenile survival may be more impor- tant than preservation of adults that might already have bred (Calow, 1973). Some acoel and rhabdocoel turbellarians are semelparous and it would be interest- ing to compare lipid and glycogen contents and calorific values in these at various stages in the life history. In freshwater molluscs, however, which are similarly semelparous glycogen storage predominates and in fact increases in preparation for over-wintering (Goddard and Martin, 1966). These animals, though, are slow- moving, which permits the storage of less potential energy per unit weight (vide Stetten and Stetten, 1956), and they are also herbivorous. This habit, as opposed to the carnivorous one, reduces the probability of either prolonged or intense starvation (Hairston, Smith and Slobodkin, 1960) so that long term energy stores, i.e. lipid. are less important. The triclads utilize lipid during starvation (Jennings, 1957; Boddington and Mettrick, 1971) and, as shown in the present study, this is accompanied by a decrease in calorific value. The importance of this ability, coupled with the tri- clad's capacity to resorb its tissues and then grow again when conditions permit, has been demonstrated in a classic study by Reynoldson (1966). Reynoldson showed that hatching of the eggs, in European triclad populations, is followed by gradual development of food shortage as the young grow. This slows and eventu- ally stops breeding. The triclads then decline in size, resorbing their tissues, and this is accompanied by some mortality which readjusts population size to food supply. The survivors are then able to feed, grow and reproduce in a further cycle, as conditions permit. The symbiotic flatworms, and especially the parasitic species, are buffered from such variations in food supply. Ectocommensals are perhaps the least buffered, for their nutritional physiology resembles that of free-living species (Jennings, 1968, 1973). In the species studied here the calorific value is higher than the mean for the ectoparasites (Fig. 1), which have a more stable food supply, but it is still lower than that of the free-living species. In entosymbiotic species, as in the ectoparasites, the food supply is relatively stable. The entocom- mensals feed on protozoa and other small organisms which are either co-com- mensals or components of the host's food ; in both cases they are abundant and readily available to the entocommensal flatworm (Jennings, 1973). In the ento- parasitic species the stability of the food supply is self-evident for tissue-feeders and sanguivores (Monogenea and Digenea), and this is also the case for the 90 P. CALOW AND J. B. JENNINGS Cestoda which live, almost without exception, in the intestine of vertebrates. Here the host actively maintains a constant level of amino acids, vitamins and other organic substances in the gut lumen, by means of the exocrine-enteric circulation (Read, 1950; Read, Simmons, Campbell and Rothman, 1960; Nasset and Ju, 1961; Nasset, 1965; Read, 1970), and this will buffer the cestode from any temporary changes in the quantity or composition of the host's ingesta. The adoption of an entosymbiotic life style, therefore, removes the need for long-term storage products because it makes available a stable source of food. The entosymbiote is freed from the need to form lipid reserves to ensure survival until it can breed and we suggest that this is the prime reason for the reduced calorific values of entosymbiotic flatworms. These reduced values may also be related to the fact that entosymbiotic species, once mature, tend to breed more or less con- tinuously throughout their life span and thus will be continually exporting lipid as a component of the eggs. Free-living species, whether semelparous or itero- parous, do not have this prolonged reproductive phase and their relatively high lipid content could be due to a build-up of lipid which is subsequently exported during reproduction. A comparable situation is seen, for example, in freshwater animals which accumulate inorganic salts before breeding for incorporation in the eggs. This hypothesis cannot be tested, as yet, due to lack of appropriate data. We suggest, though, that even if it were proved to be correct it could account for only a proportion of the discrepancy between the calorific values of symbiotic and non-symbiotic flatworms, and that reliability of the food supply is the major factor. The food supply of entoparasites would seem to be even more reliable than that of entocommensals, and so the evolution of entoparasitism is seen as a logical conclusion to the tendency to form symbiotic associations which is a dominant feature of the biology of the Platyhelminthes. The entosymbiotic life style, though, places a very considerable premium on production of offspring, because of the problems of dispersal between hosts. This can be met by using resources that in free-living flatworms would have to be diverted into lipid reserves. Such increased production of progeny presumably involves considerable expenditure of energy and since long-term storage is not in- volved glycogen rather than lipid is the most suitable substrate. It is readily metabolized and its high molecular weight and low intrinsic viscosity permit ac- cumulation in tissues without much effect on osmotic pressure (Stetten and Stetten, 1956). Thus, although there may be fairly rapid turn-over of glycogen in ento- symbiotes that are continually producing eggs, its properties are such that the tissues can always contain relatively large amounts. The occurrence of large quantities of glycogen in entosymbiotes appears, there- fore, to be basically an adaptation for high fecundity, necessitated by the mode of life, and is not primarily an adaptation related to environmental oxygen tension, as has been previously suggested. Nevertheless, as pointed out by Jennings H973), the presence of glycogen in entosymbiotes does constitute a pre-adaptation for life under conditions of low or variable oxygen tension, in that a suitable sub- strate is already available for partially or totally anaerobic respiration. A number of entosymbiotic flatworms, notably the Cestoda, have exploited this potential and colonized habitats such as the vertebrate intestine where the oxygen concentra- tion may be much reduced. In such species there is, without doubt, a direct CALORIFIC VALUES IN PLATYHELMINTHES (>1 relationship between their high glycogen content, the low environmental oxygen tension, and their modified respiratory physiology, hut on the ahove argument this relationship is seen to be proximate rather than ultimate in nature. An interesting consequence of our interpretation of the relationship between the nature of the food reserve, mode of life and level of fecundity, is that parasitism, and especially entoparasitism can be regarded simply as adaptive devices which favor high fecundity. This view, of course, is the reverse of the classical view, already cited, that high fecundity is a basic and essential adaptation for the parasitic habit. In explaining the distribution of calorific values within the Platyhelminthes we have emphasized the importance of reproductive strategy and food availability. This is because groups with different life styles, ranging from completely parasitic to completely free-living, fit exactly into the distribution spectrum from low to high caloric values and we believe that the only parameters which show the same distribution are fecundity and food availability. We dismiss the direct in- volvement of oxygen tension for reasons already stated but accept that mobility might be of some importance. For example, free-living forms may need to be more mobile than parasites and thus might require to store more energy per unit weight. This could account for their higher calorific values. However, such an argument presupposes that food stores are necessary anyway, and in this sense is less basic than our explanation in terms of fecundity and food supply. Further- more, the two possibilities are not mutually exclusive. There may, of course, be more subtle explanations. For example, there could be some biochemical disadvantages (other than those stemming from reduced oxygen tension) associated with lipid storage, and hence high calorific values, in the enclosed entosymbiotic environment. It is difficult, though, to envisage what these could he and so we present our argument as the most plausible within the limits of currently available information. We wish to thank Dr. D. F. Mettrick for supplying Hyuicnolcpis dindnuta and Syndesmis jranciscana, and for helpful discussion in the early stages of this work. We are grateful also to Dr. P. G. Moore for supplying Proccrodcs iilrac. This work was supported in part by a Natural Environment Research Council award, No. GT 4/69/OF/l 1 , to P. Calow and a Science Research Council research grant, No. 2016/4, to J. B. Jennings. SUMMARY 1. A comparison has been made of the calorific values (kcal/g ash-free dry weight) of 21 species of flat worms which exhibit, between them, the various life styles found in the phylum Platyhelminthes. 2. The grand mean calorific value for whole animals of all 21 species was 5.841 ± 0.230 kcal/g ash-free dry weight, which is close to the mean of 5.821 kcal/g ash-free dry weight reported for whole animals from a wide range of other phyla. 3. There is a direct relationship between mode of life and calorific value. Free- living flatworms have a mean value higher than the mean for whole animals in general, while entocommensal and entoparasitic species other than cestodes have 92 p. CALOW AND J. B. JENNINGS lower values. The ectocommensal, ectoparasites and cestodes have values close to the mean. 4. In the cestode H. diminuta the values for immature and mature proglottids (regions comparable to whole flat worms from other classes) are less than the mean for whole animals from other phyla and approach those for the entocom- mensal and entoparasitic flatworms. Gravid proglottids have a higher mean. 5. The ento symbiotic flatworms, with high fecundity relative to free-living species, thus conform to the hypothesis that fecundity is linked with low potential energy per unit weight. 6. The dichotomy in food reserves in the Platyhelminthes, with storage of lipid in free-living species and of carbohydrate in the symbiotic forms is considered, therefore, to be basically an adaptation to different reproductive patterns neces- sitated by different life styles. 7. It is suggested that the emphasis on carbohydrate as an energy source, rather than lipid, pre-adapted entosymbiotes for partial or total anaerobic respira- tion. The relationship between high glycogen content, low environmental oxygen tension and modified respiratory physiology occurring in some entoparasites is thus proximate rather than ultimate in nature. 8. The evolution of entoparasitism is seen as a logical conclusion to the trend toward the formation of symbiotic relationships, which is a characteristic feature of the Platyhelminthes, in that it removes the need for long-term reserves and thereby allows total emphasis on progeny production. It follows that the parasitic habit, and especially entoparasitism, provides conditions necessary for high fecund- ity so that the latter can be regarded as a direct and automatic consequence of the parasitic mode of life, rather than as a specific adaptation to it. LITERATURE CITED BODDINGTON, M. J., AND D. F. METTRiCK, 1971. Seasonal changes in the chemical composition and food reserves of the freshwater triclad Dugcsia tif/rin-a (Platyhelminthes: Turbellaria). /. Fish. Res. Board Canada. 28: 7-14. BRAND, T. VON, 1946. Anaerobiosis in Invertebrates. Biodynamica Normandy, Missouri, 328 pp. CALOW, P., 1973. The relationship between phenology, fecundity and longevity: a systems approach. Amcr. Natur., 107: 559-574. CAMPBELL, J. A., 1931. Gas tensions in the tissues. Physiol. Rev., 11 : 1-40. CROMPTON, D. W. T., D. H. SHRIMPTON AND I. A. SILVER, 1965. Measurements of the oxygen tension in the lumen of the small intestine of the domestic cluck. /. E.rp. Biol, 43 : 473-478. CUMMINS, K. W., AND J. C. WUYCHEK, 1971. Caloric equivalents for investigations in ecological energetics. Mitt. Int. Ver. Tlicor. Atuinv. Limnol., 18: 1-158. FAIRBAIRN, D., G. WERTHEIM, R. P. HARPUR AND E. L. SCHILLER, 1961. Biochemistry of normal and irradiated strains of Hyuienolcpis diminuta. E.vp. Parasitol., 11: 248-263. GODDARD, C. K., AND A. W. MARTIN, 1966. Carbohydrate metabolism. Pages 275-308 in K. W. Wilber and C. M. Yonge, Eds., Physiology of Mollusca, Volume 2. Academic Press, New York. HAIRSTON, N. G., F. E. SMITH AND L. B. SLOBODKIN, I960. Community structure, population control, and competition. . liner. \'attir., 94: 421^125. HALTON, D. W., 1967. Studies on glvcogen deposition in Trematoda. Comp. Biochem Physiol.,23: 113-120. HILL, R., 1936. Oxygen dissociation curves of muscle haemoglobin. Proc. Rov. Soe. Series B, 151: 256-276. CALORIFIC VALUES IN PLATYHELMINTHES 93 HYMAN, L. H., 1951. The Invertebrates: Platyhelminthes and Rhynchocoela. Tlic acoelotnate BUater/a. Volume II. McGraw-Hill Book Co., New York, 550 pp. JENNINGS, J. B., 1957. Studies on feeding, digestion and food storage in free-living flat- worms (Platyhelminthes: Turbellaria). Biol. Bull., 112: 63-80. JENNINGS, J. B., 1968. Nutrition and digestion. Pages 303-326 in M. Florkin and B. T. Scheer, Eds., Chemical Zoology, Volume II, Porifera, Coelenterata and Platy- helminthes. Academic Press, New York. JENNINGS, J. B., 1971. Parasitism and commensalism in the Turbellaria. Pages 1-32 in B. Dawes, Ed., Advances in Parasitology, Volume IX. Academic Press, New York. JENNINGS, J. B., 1973. Symbioses in the Turbellaria and their implications in studies on the evolution of parasitism. Pages 127-160 in Winona B. Vernberg, Ed., Symbiosis in the Sea. University of South Carolina Press, Columbia, South Carolina. JENNINGS, J. B., AND D. F. METTRICK, 1968. Observations on the ecology, morphology and nutrition of the rhabdocoel turbellarian Syndesmis franciscana (Lehman, 1946) in Jamaica. Carih. J. Sci., 8 : 57-69. KERSTING, K., 1972. A nitrogen correction for caloric values. Limnol. Oecanoi/r.. 17: 643- 644. METTRICK, D. F., AND M. J. BODDINGTON, 1972. The chemical composition of some marine and fresh-water turbellarians. Carib. J. Sci., 12: 1-7. METTRICK, D. F., AND C. E. CANNON, 1970. Changes in the chemical composition of Hymenolepis diminuta (Cestoda: Cyclophyllidea) during prepatent development within the rat intestine. Parasitology, 61 : 229-243. METTRICK, D. F., AND J. B. JENNINGS, 1969. Nutrition and chemical composition of the rhabdocoel turbellarian Syndesmis franciscana. with notes on the taxonomy of S. antillanim. Can J. Zool, 26: 2669-2679. NASSET, E. S., 1965. Role of the digestive system in protein metabolism. Fed. Proc., 24: 953-958. . NASSET, E. S., AND J. S. Ju, 1961. Mixture of endogenous and exogenous protein in the alimentary tract. /. \'ntr., 71 : 461-465. ODUM, E. P., S. G. MARSHALL AND T. G. MARPLES, 1965. The calorific content of migrating birds. Ecology. 46 : 901-904. OSTAPENYA, A. P., AND A. I. Sp.RGEEV, 1963. The caloric content of the dry substance of freshwater invertebrates used as food by fish, fin Russian] Voprosy Ikhtiologi, 3 : 177-183. ('Fish. Res. Bd. Canada, Transl. No. 874.) PAINE, R. T., 1966. Endothermy in bomb calorimetry. Limnol. Oceanoi/r., 11: 126-129. PAINE, R. T., 1971. The measurement and application of the calorie to ecological problems. Annual AYi'fVtc' of Ecology and Systematics, 2: 145-164. PHILLIPSON, J., 1964. A miniature bomb calorimeter for small biological samples. Oikos, 15: 130-139. READ, C. P., 1950. The vertebrate small intestine as an environment for parasitic helminths. Rice hist. P am phi.. 37 (2) : 1-94. READ, C. P., 1968. Intermediary metabolism of flatworms. Pages 327-357 in M. Florkin and B. T. Scheer, Eds., Chemical Zoology, Volume II, Porifera, Coelenterata and Platy- helminthes. Academic Press, New York. READ, C. P., 1970. Some physiological and biochemical aspects of host-parasite relations. /. Parasitol.. 56: 643-652. READ, C. P., J. E. SIMMONS, JR., J. W. CAMPBELL AND A. H. ROTHMAN, 1960. Permeation and membrane transport in parasitism : studies on a tapeworm-elasmobranch sym- biosis. Biol. Bull., 119 : 120-133. REINERS, W. A., AND N. M. REINERS, 1972. Comparison of oxygen bomb combustion with standard ignition techniques for determining total ash. Ecology. 53 : 132-136. REYNOLDSON, T. B., 1966. The distribution and abundance of lake-dwelling triclads — towards a hypothesis. Pages 1-71 in J. B. Cragg, Ed., Advances in Ecological Research. Volume III. Academic Press, New York. ROGERS, W. P., 1949. On the relative importance of aerobic metabolism in small nematode parasites of the alimentary tract. Aust. J. Sci. Res. Scries B., 2: 166-174. ROHLF, F. J., AND R. R. SOKAL, 1969. Statistical Tables. W. H. Freeman and Co., San Francisco, 253 pp. 94 P. CALOW AND J. B. JENNINGS SLOBODKIN, L. B.. 1962. Energy in animal ecology. Pages 69-101 in J. B. Cragg, Ed., . Itlninccs in Ecological Research, Volume I. Academic Press, New York. SLOBODKIN, L. B., AND S. RICHMAN, 1961. Calories/gm in species of animals. Nature, 191 : 299. SMYTH, J. D., 1966. The Physiology of Ccstodcs. Oliver and Boyd, Edinburgh, 279 pp. STETTEN, D. W., AND M. R. STETTEN, 1956. Glycogen turnover. Pages 291-307 in S. Graff, Ed., Essays in Biochemistry. Wiley, New York. TEAL, J. M., 1957. Community metabolism in a temperate cold spring. Ecol. Monogr., 27: ' 283-302. WEBB, R. A., AND D. F. METTRICK, 1973. The role of serine in the lipid metabolism of the rat tapeworm Hyinenolepis diiniinito. Int. J. Parasitol., 3 : 47-58. WEIGERT, R. G., AND D. C. COLEMAN, 1970. Ecological significance of low oxygen consump- tion and high fat accumulation by Nasiititennes costalis (Isoptera: Fermitidae). Bioscience, 20: 663-665. Reference : Blol Bull, 147 : 95-104. (August, 1974) LOCOMOTION OF THE HOLOTHURIAN EUAPTA LAPP A AND REDEFINITION OF PERISTALSIS .1. MICHAEL HEFFERNAN AND STEPHEN A. WAINWRIGHT Department of Zoology, Duke University, Durham, North Carolina Observation of the holothurian Euapta lap pa (J. Muller, 1853) reveals loco- motory waves that are unlike any previously described. According to the term- inology of Vies (1907), the wave is direct rather than retrograde, meaning that it moves in the same direction as the animal- — in this case, anteriorly. In Euapta, the wave is peristaltic and involves no elevation of the body cylinder's major axis. Direct peristaltic locomotory waves, previously described in the holothurians Stichopits panimensis and Astichopus niultifidus have an arch above the sub- stratum formed by the localized temporary elevation of the body cylinder's major axis (Parker, 1921; Glynn, 1965). Although retrograde waves in wormlike animals are well understood (Gray and Lissman, 1938; Clark, 1964), direct locomotory waves are less well known (Elder, 1973a, 1973b). This paper presents the first description of locomotion in Euapta and a resolution of the ambiguity of the term "peristalsis." "Peristalsis" is herein redefined as any muscular con- traction moving along a radially flexible tube in such a way that each component wave of circular, longitudinal or oblique muscular contraction is preceded or fol- lowed by a period of relative relaxation of all similarly oriented muscle within a given tubular segment. A classification of some known types of peristalsis is also presented. THE EXPERIMENTAL ANIMAL AND METHODS Euapta lap pa is an apodous holothurian with a wormlike appearance. The animal has an extremely flexible body which may attain an extended length of 1 m and a diameter of 2-4 cm. If disturbed, however, Euapta can contract regions of the body to about one third their extended lengths. The anterior end has a circumoral ring of 15 pinnate tentacles and the body cylinder, in a condition of tonus (resting muscular tension), has five radially symmetric rows of outpocketings called warts (Fig. 1). The animal has both circular and longitudinal muscles; the former are a continuous tube throughout the body; the latter are restricted to 5 strips arranged in radial symmetry and extending the length of the body. The animal's body surface is covered with many anchor-shaped calcareous ossicles (Hyman, 1955) approximately 0.4 mm long which aid in attachment to the sub- stratum. Specimens were obtained from Tropical Atlantic Marine Specimens, Big Pine Key, Florida, and kept in a 25 gallon aquarium filled with recirculated artificial sea water ("Instant Ocean") on a crushed shell substratum. Observation was the main technique and a continuous-feeding 35 mm movie camera (Grass) was used to record and "slow down" the motion. The camera's shutter was always 95 96 J. M. HEFFERNAN AND S. A. WAINWRIGHT open and images were formed by the flashing of a stroboscopic light in a dark room. For filming, the animal was transferred to a smaller and narrower tank provided with a centimeter grid as a background for reference. Living body wall was examined with a Wild M-5 dissecting microscope with polarizing attachments. RESULTS Warts (Fig. 1) are not permanent structures. Individual warts may be ob- served to shift or "slide" slightly along the major body axis, to increase or de- crease in size and to divide into two smaller warts or merge to form a larger wart. Warts do not divide or merge circumferentially. A single wart will collapse when its apex is pinched with forceps. Animals anesthetized with magnesium salts have a body cylinder of greater average diameter with no warts. Histological preparations failed to reveal any morphologically distinct structure that might account for the warts. In living animals, a number of warts were tagged by threading them with a suture. But subsequent vivisection did not re- veal any distinct correlated structures except that tags were always between two adjacent strips of longitudinal muscle (Fig. la). The five relatively narrow strips of longitudinal muscles are separated by five relatively wider strips of body wall lacking longitudinal muscles. This pattern of longitudinal muscle is super- imposed on the continuous circular muscle layer. Thus, we conclude that warts in Euapta and probably those in other similar apodous holothurians are not specific morphologically distinct points or structures as is suggested by other writers (Hyman, 1955). Instead, warts appear to be areas of limited contraction of circular muscle bounded by regions of greater degree of contraction of circular muscles. These holothurian warts then are analogous to the haustrations of the vertebrate large intestine (Guyton, 1971). Divisions of warts are effected by contraction of additional circular muscles, and merging occurs when a contracted ring of circular muscle relaxes. The locomotory wave (Pig. Ib) By observing the distance between points on the body wall, it is possible to tell when longitudinal or circular muscles are contracting. As the direct locomotory wave passes along a section of the body wall, contraction of longitudinal muscles (B) is the first evident occurrence. Warts and recognizable pigmented points on the body wall come closer together longitudinally in an accordion-like effect. Then, while longitudinal contraction is still under way, circular contraction occurs (C) as is evidenced by the decrease in diameter of the body cylinder and the "flatten- ing" of warts. This is followed by relaxation and extension of all body wall musculature (D) so that the entire body wall bulges outward. Expansion is fol- lower by a return to the original tonus (E) . Each locomotory wave starts at the posterior end of the animal by the same sequence of longitudinal and circular contraction. Each part of the body wall advances while under longitudinal contraction. The major axis of the body is not raised in the formation of an arch but rather an arch is formed by a decrease in diameter of the body cylinder. The body axis does not change its position. Con- LOCOMOTION OF EUAPTA 97 (a) FIGURE 1. Diagrams showing the external shape of the apodous holothurian Euapta lappa. (a) Seven warts are shown with the muscles that delineate them: solid lines indicate muscles in state of tonus ; broken lines indicate relaxed circular muscles ; (b) body of Euapta with a direct overlapping peristaltic wave passing from right to left; B, initial contraction of longitudinal muscle ; C, contraction of circular muscle and further longitudinal muscle con- tractions; D, relaxation of all muscles: body wall distended by coelomic pressure; E, state of tonus. traction of circular muscles releases the body wall from its attachment to the substratum, thus facilitating forward movement. The direct overlapping peristalsis described above is the most common method of locomotion in Euapta, but variations do exist as a result of the reduction or complete suppression of some of its components. One type of wave, direct longi- tudinal peristalsis, is characterized by longitudinal muscular contractions as ob- served in the direct overlapping peristalsis, but there is no accompanying wave of circular contraction. Direct longitudinal peristalsis is observed when the animal crawls across smooth glass surfaces and when it is climbing up a vertical aquarium wall clinging by its tentacles with part or all of its body hanging free below. Sometimes this direct longitudinal paristalsis passes along a body cylinder which is in a tonus state of circular contraction. At other times the body cylinder is constricted throughout part or all of its length by continuous contraction of cir- cular muscles. This obliterates all warts and holds the body diameter constant while waves of only longitudinal contraction pass along the body. In these two cases of waves of longitudinal muscle contraction passing along a body cylinder of unchanging diameter, it is clear that friction is minimal and offers little help or hindrance to forward motion. Direct longitudinal peristalsis is further considered in the discussion. Another type, direct overlapping partially circular peristalsis, has normal con- traction of longitudinal muscle but only the two ventral-most warts are retracted, as opposed to the full circumferential contraction by circular muscles seen in direct overlapping peristalsis. This wave of incomplete circular muscle contraction was observed on only two occasions and was never recorded on film. 98 J. M. HEFFERNAN AND S. A. WAINWRIGHT There is another wave which may be of interest. Viewed infrequently while the animal is resting, this circular contraction peristalsis is a wave of contraction of only circular muscles that seems to have no locomotory function. These different waves are not necessarily mutually exclusive behavioral events. A wave in Eitapta may start as a certain type; by the time it reaches the anterior end it may be modified to one of its variations. Intermediate forms exist. Upon first viewing En apt a move by direct overlapping peristalsis, it may be apparent why longitudinal contraction advances a more posterior point rather than pulling an anterior point posteriorly. Obviously circular contraction raises the body wall from the substratum while the still-attached body wall anterior to the locomotory wave acts as a fixed point of attachment. Thus, longitudinal con- traction will move detached points forward toward the fixed point. In addition, Euapta's calcareous ossicles are believed to be so situated that they catch the sub- stratum when pulled posteriorly and release their hold when pulled anteriorly (Hyman, 1955). It appears that opportunities for catching onto the substratum are maximum when the body is maximally inflated and the epidermis is stretched tightly over the ossicles. Portions of the body anterior to a locomotory wave are under slight longitudinal stretch which is increased by the approaching longitudinal contraction. Accordingly, these regions, just anterior to a locomotory wave, will have their ossicles well protruded and attached to the substratum, each forming a fixed point. During locomotion the tentacles also act as fixed points at the anterior end. Each tentacle in turn sweeps from its extended position in front of the animal, backward toward the mouth. It then reaches forward and attaches to the substratum creating a fixed point. A tentacle may subsequently detach and sweep back toward the mouth or it may first contract longitudinally and take up any "slack" left by a peristaltic wave before detaching. DISCUSSION When Euapta hangs from the vertical glass wall of an aquarium, the tentacles are the only fixed points of attachment. In this situation direct longitudinal peri- stalsis continues to serve a locomotory function. Each such wave gives the animal a boost as it climbs the glass. Each wave begins by pulling the posterior end upward and forward. The region of longitudinal compression thus formed is propagated anteriorly. This situation is analogous to that of a "slinky" toy, hang- ing vertically from one end. A "slinky" toy is a long easily extensible spring with a small spring constant, K, according to the equation F = - KX where F is the force generated by the spring and X is the linear displacement of an end of the spring stretched from its resting length. Such springs hung vertically will elongate to many times their resting length due to gravity alone. If the lower end of a vertically hung "slinky" is pushed up one foot and re- leased, a wave of compression is formed which travels upward while the lower end of the spring begins to fall back down. The upward travelling wave of longi- tudinal compression, upon reaching the upper end, provides an upward force op- posing the downward force of gravity. Meanwhile, as the lower end which was lifted up and then released begins to fall back down, the downward force of gravity acting upon the mass of the spring at the lower end is manifest in the LOCOMOTION OF EU . I/'/./ 99 motion of the falling segment of spring. Thus, as long as the lower part of the spring is still falling, the weight of this lower part of the "slinky" is not borne by a point of attachment at the upper end. However, the lower end of the spring is brought to an abrupt stop and would need to be stretched to continue any further downward motion. The point of attachment now must provide an upward force equal and opposite to the force of the downward acceleration of gravity acting upon the mass of the lower end of the spring ; the point of attachment reassumes support of the weight of the lower end of the spring. In addition, the fixed point of at- tachment must momentarily supply another upward force : one that is not even required when the fixed point is supporting the whole body weight. This upward force is needed to cause the rapid deceleration (an upward acceleration) of the falling mass of the lower end of the spring as it comes to a stop. In the instance of Eiiapta hanging and climbing by its tentacles, a wave of longitudinal compression reaches the anterior end, providing an upward force. This force causes an upward displacement of the anterior end. The tentacles for a moment do not have to support the full weight of the body. While some of the tentacles maintain their hold, others reach forward and secure a new hold before they must resume full support of the body mass and absorb the momentum of the falling lower body region. Definitions of "peristalsis" found in dictionaries are contradictory and con- fusing. We have given a description of the direct peristaltic locomotory waves found in Eitupta, which are very different from the direct peristaltic locomotory waves described for Asticliopus by Glynn (1965). Parker (1921 ) did not use the term "peristalsis" to describe the wave he observed in Stichopus but accounts of his work by text writers (Clark, 1964) indicate that the wave he observed is con- sidered by them to be peristaltic. Elder (19731)) states that locomotion in the apodous holothurian Lcptosynapta is achieved by means of direct peristaltic waves involving simultaneous longitudinal and circular muscle contraction. Duncan and Pickwell (1939, page 141) described the telescopic locomotion of dipteran larvae as being "similar to peristalsis." A more recent reviewer, (Hughes, 1965, page 233) however considers apodous larvae to be "entirely dependent on peristaltic movements of the body for propulsion." Direct pedal locomotory waves of gastropods and the direct peristaltic locomotory waves of Euapta both function by similar mechanisms. In each case the ventrum is lifted from the substratum and then advanced by contraction of longitudinal muscles. Yet the latter is termed peristaltic and the former is not. All too often an author, upon identifying a wave as peristaltic, believes he has adequately characterized the wave. Some authors (Code and Carlson, 1968; and Ritchie, Truelove, Ardran and Tuckey, 1971) work- ing with mammalian intestinal contractions; seem to have adopted the definition of peristalsis as a propulsive wave of contraction of part or all of the circular muscle. This working definition, however, excludes contractions observed in invertebrates which have long been termed peristaltic. Mammalian gastroenterologists have what is perhaps an even more confusing terminology problem. Balloons and open-tipped catheters inserted into the mam- malian gastrointestinal tract yield four distinct types of pressure wave recordings. These recordings include pressure changes caused by peristaltic as well as non- peristaltic contractions. Two of these four classifications differ only in the ampli- 100 J. M. HEFFERNAN AND S. A. WAIN WRIGHT tude of their fluctuation (Hightower, 1968). There is no clear accepted correla- tion between types of pressure recordings and the varied types of gastrointestinal movement recognized by direct observation and cineradiography ; segmenting move- ments, pendulum movements, peristalsis and peristaltic rushes. Some authors believe that the pendulum movements are identical with segmenting contraction. Pressure recording classifications tell us nothing about the mechanical process involved and they are valued mostly as diagnostic tools in medicine. Most important to our discussion is the fact that there is still disagreement as to the contraction sequence of circular and longitudinal muscle components in mammalian intestinal peristalsis (Bortoff and Ghalib, 1972; Gonella, 1972; Raiford and Mulinos, 1934; Bayliss and Starling, 1899; and Wood and Perkins, 1970). Obviously some standardization of terminology is necessary. The following, then, is an attempt to construct a useful definition of "peristalsis" consistent with its past and present usage. Peristalsis is the phenomenon of any muscular contraction moving along a radially flexible tube in such a way that each component wave of circular, longi- tudinal or oblique muscular contraction is preceded and/or followed by a period of relative relaxation of all similarly oriented muscle within a given tubular seg- ment. The meaning of this definition is not changed if the word "relaxation" is replaced by "contraction" and each "contraction" is replaced by "relaxation." In any form of peristalsis, any two tubular segments are subject to a sequence of qualitatively identical force vectors. Furthermore, in locomotory peristalsis, each region of the body must, at some time during or between the passages of waves, have no forward motion. This is in contrast to the continuous undulatory locomotory mechanisms of fishes or the rectilinear progression of the boa constrictor (Gray, 1968), where the body moves as a unit and at a constant rate. This definition is consistent with current and past use of the term "peristalsis" and includes the propagated waves of contraction found in vertebrate intestine, the retrograde locomotory waves found in earthworms and other coelomates, the locomotory waves characteristic of lepidopterous caterpillars (Barth, 1937), the direct locomotory waves of both dipteran larvae and Euapta and undoubtedly many more mechanisms of locomotion. Although the definition includes the locomotory waves of lepidopterous caterpillars, it excludes the sinusoidal loco- motory waves of Nereis which involve alternating component waves of longitudinal contraction on opposite sides of the body cylinder (Gray, 1968). The definition also excludes swimming movements of leeches and fish, pedal locomotory waves, and the serpentine, concertine, crotaline and rectilinear movements of snakes (Gray, 1968). The peculiar swimming movements of young adult Leptosynapta are similarly excluded (Costello, 1946). The term, however, remains broadly ap- plicable and an attempt to classify the types of peristalsis seems appropriate. In the following classification of peristaltic types, all locomotory peristalsis will be classified as either direct or retrograde. The word "locomotory" is omit- ted to avoid verbosity. Only the names of non-locomotory waves will be without this direct versus retrograde distinction. Although some types of non-locomotory peristalsis may be mechanistically equated with forms of locomotory peristalsis, this distinction between locomotory and non-locomotory peristalsis is useful. Transportive peristalsis as seen in the vertebrate intestine has not been fully characterized. Kosterlitz (1968), Raiford and Mulinas (1934), Wood and Per- LOCOMOTION OF EUAPTA 101 kins (1970) and Gonella (1972) believe longitudinal and circular muscle contract sequentially. Thomas and Baldwin (1971), Bortoff and Ghalib (1972) and Bay- liss and Starling (1899) propose a simultaneous contraction of both circular and longitudinal muscle followed by a subsequent simultaneous relaxation. A Circular contraction peristaltic wave is a wave of contraction of circular muscles only. Euapta has such a wave, and it cannot serve to pull the animal forward. It may serve some other purpose, as in the "ring of constriction" in Thyone (Pearse, 1908, page 266) which facilitates the detachment of tube feet from the substratum. The peristaltic wave described by Yamanouchi (1929) and Crozier (1915) may be circular contraction peristalsis. Yamanouchi's paper in- cludes a diagram depicting a wave very much like the direct overlapping peristalsis of Euapta and he cites Crozier as having already described this type of wave in detail. Crozier, however, describes peristalsis in Holothuria surinamensis originat- ing at any section of the body and slowly travelling either anteriorly or posteriorly. The body is not raised above the substratum and movement is accomplished primarily by the action of pedicles. Buccal tentacles and contraction of the body musculature play secondary roles in this locomotion. Partially circular relaxation peristalsis consists of a wave of relaxation passing as an incomplete ring along the circular musculature. Coelomic fluid pressure causes distension at the point of relaxation. The worm Arenicola seems to use this type of wave to irrigate its' tubule although the role of longitudinal muscula- ture is not delineated (Wells, 1961). In the case of Arenicola the wave of re- laxation is restricted to the dorsal region. A wave occludes the space between the worm's dorsal surface and the walls of the tubule thus pushing water in front of it. Longitudinal contraction peristalsis is a wave of contraction of only longitudinal muscles that is unrelated to locomotion. Sabella uses such a wave to irrigate its tubule (Mettam, 1969). Starting at one end of the body, a wave of longitudinal contraction shortens constant volume segments causing dilation. The dilated re- gion closely fits the smooth tubule and the propagated wave acts like a "piston" traveling the length of the tubule and pushing water in front of it. Sabella's "piston" is able to move with minimal frictional resistance since parapodia are retracted and abundant mucous glands provide lubrication. Narrow elongated segments grip the tubule wall with extended parapodia. Direct longitudinal peristalsis refers to wraves of contraction of only longi- tudinal muscles travelling in the same direction as the animal. Euapta is capable of this wave type. Direct arching peristalsis is observed in some lepidopterous caterpillars (Barth, 1937). Basically, a wave of contraction passes anteriorly along the dorsal longi- tudinal musculature and is followed by a wave of contraction of the dorso-ventral musculature. As the dorso-ventral musculature subsequently relaxes, a wave of contraction of the ventral musculature passes along the body tube. Although they haven't been analyzed into their component muscular contrac- tions, similarly described waves have been reported in holothurians. Parker (1921, page 205) reports a direct muscular wave in Stichopus which "resembles superficially the locomotion of a gigantic caterpillar." ("ilynn (1965, page 112) describes Astichopus's peristalsis in which "the posterior end is first elevated two to four centimeters from the substratum and then the wave moves forward form- ing an arch 2 cm high between ventrum and the underlying surface." 102 J. M. HEFFERNAN AND S. A. WAINWRIGHT Direct overlapping peristalsis is the locomotory mechanism most frequently observed in Enapta. This type of peristalsis moves in the same direction as the animal. A wave of contraction of the longitudinal musculature passes along the body and during the latter part of its duration, occurs simultaneously with a wave of contraction of the circular muscles. Relaxation of circular and longitudinal muscles occurs simultaneously. The body cylinder's major axis is not elevated. In the burrowing polychaete Polyphysia crassa (Elder, 1973a) and in the apodous holothurian Leptosynapta (Elder, 1973b), the contractions of circular and longi- tudinal muscules appear to be simultaneous. Direct overlapping partially circular peristalsis as demonstrated in Euapta is similar to direct overlapping peristalsis, differing only in the radial extent of circular contraction. In the former, contraction of circular musculature does not involve entire bands or rings of circular muscle as is the case for the latter. Telescopic locomotion (Duncan and Pickwell, 1939), a direct non-overlapping peristalsis has been reported in some dipteran larvae. A direct wave of circular muscle contraction increases the pressure of the body fluids thus causing the body to lengthen and, pushing against a posterior fixed point, moves the head forward. While the head anchors itself to the substratum, a direct wave of contraction of longitudinal muscles then pulls the posterior regions forward. This appears similar to the burrowing activity observed in Arcnicola (Truman, 1966). Direct dilation peristalsis is reported in Arenicola (Mettam, 1969). In this type of locomotion (Wells, 1949) a direct wave simultaneously relaxes both cir- cular and longitudinal musculature. Coelomic fluid pressure in the relaxed seg- ment causes the body wall to dilate and contact the sides of the tubule, thus acting as a fixed point. Once the wave of relaxation passes, both circular and longitudinal muscle return to tonus. As the longitudinal muscle returns to tonus, longitudinal shortening pulls the tubular segment toward the fixed point which has in the meantime moved anteriorly. Retrograde non-overlapping peristalsis is the method of locomotion of earth- worms and is described in detail by Gray and Lissman (1938), Clark (1964) and Child (1901). Waves of contraction of longitudinal muscles pass posteriorly caus- ing dilation of the body cylinder. A wave of contraction of circular muscles fol- lows immediately, but does not overlap the longitudinal wave, and causes elonga- tion of the body cylinder as well as causing it to push forward from more posterior fixed points. In some cases, as in earthworms, fixed points are provided by the increased body diameter. In some animals, fixed points are provided by other means of attachment. In the leech, for example, the suckers at each end of the body alternately serve as fixed points. Retrograde dilation peristalsis is reported in Sabella (Mettam, 1969). A retrograde wave of contraction passes along the longitudinal muscle as the seg- mental volume is kept constant. Longitudinal shortening pulls a segment an- teriorly toward a fixed point. Continued longitudinal shortening dilates the seg- ment until it become wedged against the tubule walls and itself acts as a fixed point. After a wave of longitudinal contraction, a segment extends anteriorly. These classes of peristalsis are not intended to be considered as the only possible classes. They do represent the different forms of muscular waves con- sidered to be peristaltic that we have encountered in the literature. Apodous holothurians comprise an unusual adaptation of hydrostatic skeletons. LOCOMOTION OF EUAPTA 103 Most animals that move by peristalsis have relatively thick body walls and high internal hydrostatic pressures. Eiiapta and other apodous holothtirians are low pressure creatures whose body wall is so thin that it is ruptured by the weight of coelomic fluid if the animal is lifted into the air by one end. The Caribbean Enapta and the Pacific Opheodcsoma live completely exposed on reef tops among corals and attached algae. They must normally be constant volume hydrostats— as are burrowing annelids. But these holothurians may, at any one time, contract the longitudinal muscles in the entire front (or rear) 2/3 of the animal and force the coelomic fluid into the remaining 1/3. This inflated 1/3 will have a larger diameter and no warts. Even those polychaetes that have incomplete septa can- not change their dimensions to this extent. As a result of their ability to change dimensions drastically, apodous holothurians can pass through cracks or holes in the substrate that are less than 1/3 their diameter. This recalls similar behavior of Peripatns (Manton, 1958) which is similarly correlated with lack of effective septa. The structural dependence of holothurian warts upon hydrostatic pressure is reminiscent of the tube feet found in many other echinoderms. Yet while a re- tracted tube foot is still an anatomically identifiable structure, a flattened wart on Euapta is totally indistinguisable from adjacent regions of body wall. We would like to thank Mrs. W. R. Heffernan for her selfless assistance in repeated typing and editing of the manuscript. SUMMARY 1. The apodous holothurian, Enapta lap pa, demonstrates four types of peristalsis. Direct overlapping peristalsis in which circular and longitudinal muscle components overlap is the most common. Other types are characterized by (a) being of longitudinal muscles only, (b) lack of circular muscle contraction along one side of the body and (c) being of circular muscles only (nonlocomotory). 2. "Warts" in apodous holothurians are not discrete structures but are areas of the body wall delimited by temporary contraction of circular muscles. 3. Peristalsis is defined as any muscular contraction moving along a radially flexible tube such that each component wave of circular, longitudinal or oblique muscle contraction is preceded or followed by a period of relative relaxation of all similarly oriented muscle. Definitions and nomenclature are given for 8 locomotory and 4 non-locomotory types of peristalsis. LITERATURE CITED EARTH, R., 1937. Muskulatur und bewgungsart der raupen zugleich ein beitrag zur spann- bewegung und schruckstellung der spannerraupen. Zoo}. JaJirb. Anat. Ontog. Ticre, 62 : 507-566. BAYLISS, W. M., AND E. H. STARLING, 1899. The movements and innervation of the small intestine. /. PhysioL, 24 : 99-143. BORTOFF, A., AND E. GnALiB, 1972. Temporal relationship between electrical and mechanical activity of longitudinal and circular muscle during intestinal peristalsis. Amer. J. Digestive Diseases, 17 : 317-325. CHILD, C. M., 1901. The habits and natural history of Stichostcmma. Amer. Natur., 35 : 975-1006. CLARK, R. B., 1964. Dynamics in Mctasoan Evolution. Clarendon Press, Oxford, 313 pp. 104 J. M. HEFFERNAN AND S. A. WAINWRIGHT CODE, C. F., AND H. C. CARLSON, 1968. Motor activity of the stomach. Pages 1903-1916 in C. F. Code and W. Heidel, Eds., Handbook of Physiology, Section VI, Vol. IV. Williams Wilkins Co., Baltimore. COSTELLO, D. P., 1946. The swimming of Leptosynapta. Biol. Bull., 90: 93-96. CROZIER, W. J., 1915. The sensory reactions of Holothuria surinamensis Ludwig. Zoo/. Jahrb. Abt. Allg. Zool. Physio!. Tiere, 35 : 233-292. DUNCAN, C. D., AND G. PICKWELL, 1939. The World of Insects. McGraw Hill Book Co., New York, 409 pp. ELDER, H. Y., 1973a. Direct peristaltic progression in the polychaete Polyphysia crassa. J. Exp. Biol., 58 : 637-655. ELDER, H. Y., 19735. Distribution and functions of elastic fibers in the invertebrates. Biol. Bull., 144: 43-63. GLYNN, P. W., 1965. Active movements and other aspects of the biology of Astichopits and Leptosynapta (Holothuroidea). Biol. Bull., 129: 106-127. GONELLA, J., 1972. Modifications of electrical activity of the rabbit duodenum longitudinal muscle after contractions of the circular muscle. Amcr. J. Digestive Diseases, 17: 327-332. GRAY, J., 1968. Animals Locomotion. Weidenfeld and Nicolson, London, 479 pp. GRAY, J., AND N. W. LISSMANN, 1938. Locomotory reflexes in the earthworm. /. E.vp. Biol., 15: 506-517. GUYTON, A. C., 1971. Te.rfbook of Medical Physiology. W. B. Saunders Co., New York, 1032 pp. HIGHTOWER, N. C, 1968. Motor action of the small bowel. Pages 2001-2024 in C. F. Code and W. Heidel, Eds., Handbook of Physiology, Section VI, Vol. IV. Williams Wil- kins Co., Baltimore. HUGHES, G. M., 1965. Terrestrial locomotion. Pages 227-254 in M. Rockstein, Ed., The Physiology of Insects, Vol. II. Academic Press, New York. HYMAN, L. H., 1955. The Invertebrates, Vol. IV. McGraw Hill Book Co., New York, 763 pp. KOSTERLITZ, H. W., 1968. Intrinsic and extrinsic nervous control of motility of the stomach and the intestine. Pages 2147-2171 in C. F. Code and W. Heidel, Eds., Handbook of Physiology, Section VI, Vol. IV. Williams Wilkins Co., Baltimore. MANTON, S. M., 1958. Habits of life and evolution of body design in arthropods. /. Linn. Soc. London Zool., 44 : 57-72. METTAM, C., 1969. Peristaltic waves of tubicolous worms and the problem of irrigation in Sabella pavonia. J. Zool. Proc. Zool. Soc. London, 158: 341-356. MULLER, J., 1853. Ueber den ban der echinoderman. Abh. Berlin Akad. Wiss. 1853: 123-129. PARKER, G. H., 1921. The locomotion of the holothurian Stichopus panamcnsis Clark. /. Exp. Zool., 33 : 205-208. PEARSE, A. S., 1908. Observations on the behavior of the holothurian Thyone briarcus (Leseur). Biol. Bull., 15 : 259-288. RAIFORD, T., AND M. G. MULINOS, 1934. The myenteric reflex as exhibited by the exteriorized colon of the dog. Amcr. J. Physiol., 110 : 129-136. RITCHIE, J. A., S. C. TRUELOVE, G. M. ARDRAN AND M. S. TUCKEY, 1971. Propulsion and retropulsion of normal colonic contents. Amer. J. Digestive Diseases, 16: 697-704. THOMAS, J. E., AND M. V. BALDWIN, 1971. The intestinal mucosal reflex in the unanesthetized dog. Amcr. J. Digestive Diseases, 16: 642-647. TRUEMAN, E. R., 1966. The mechanism of burrowing in the polychaete worm Arenicola marina (L.) Biol. Bull., 131 : 369-376. VLES, A., 1907. Sur les ondes pedieuses des molusques reptateurs. Conipt. Rend. Acad. Sci. Paris, 145 : 276-278. WELLS, G. P., 1949. Respiratory movements of Arenicola marina L. : intermittent irrigation of the tube, and intermittent aerial respiration. /. Mar. Biol. Ass. U. K., 28 : 447—464. WELLS, G. P., 1961. How lugworms move. Pages 209-233 in J. A. Ramsay and V. B. Wiggles worth, Eds., The Cell and the Organism. Cambridge University Press, New York. WOOD, J. D., AND W. E. PERKINS, 1970. Mechanical interaction between longitudinal and circular axes of the small intestine. Amer. J. Physiol., 218: 762-768. YAMANOUCHI, T., 1929. Notes on the behavior of Holothuria caitdina chilcnsis. Sci. Rep. Tohoku Univ., 4: 73-115. Reference : Biol. Bull, 147: 105-118. (August, 1974) SOME FACTORS AFFECTING THE GROWTH AND DISTRIBUTION OF THE ALGAL ENDOSYMBIONTS OF HYDRA VIRIDIS ROSEVELT L. PARDY Department of Dri'clopincutal and Cell Biology, University of California, Irvine, California 92664 Researches on the hydra-algae endosymbiosis over the last decade have pro- duced a clearer understanding of the role of the symbiotic zoochlorellae. The works of Muscatine and Lenhoff (1965a, 1965b) have shown that green hydra survive and bud longer than aposymbiotic (algal-free) hydra under starvation conditions. Moreover, Muscatine (1965) and Muscatine and Lenhoff (1963) have demonstrated that the algae may release soluble, photosynthetically fixed material to the host's cells. RoiTman and Lenhoff (1969) have shown that photosyn- thetically fixed 14CO2 appears, with time, in all of the host's major biochemical fractions but predominates in the animal's glycogen pool. Analysis of the products released by the algae in vitro have shown them to consist mainly of the disaccharide, maltose (Muscatine, 1965). The assumption is that maltose is translocated to the host by the algae in vivo and serves as nutritional supplementation for the host during periods of starvation. The ultrastructure of algae symbiotic with hydra has been studied by Park, Greenblatt, Mattern and Merril (1967) and Oshman (1967). These studies have shown that the algae are located in separate vacuoles within the digestive cells, and that the algae appear to be similar to free-living Chlorclla vulgaris, a view ad- vanced by Haffner (1925) and Beyernick (1890). Moreover the algae are seen to reside mainly in the base of the digestive cells (Whitney, 1907; Goetch, 1924; Haffner, 1925; Pardy and Muscatine, 1973). Beyernick (1890) and Haffner (1925) claim to have cultured algae from green hydra, but repeated attempts by myself and others (Muscatine, personal com- munciation) to grow isolated hydra algae in pure culture have failed. As a result of the apparent inability to effect an axenic culture of the algal symbionts it is suggested that the symbiosis is obligate for the algae. Only recently has it been shown that the algae may derive nutrition from the host. Cook (1972) demon- strated that algae in hydra become raclioactively labeled when the animals were fed 14C labeled Arteinia nauplii. Current work has shown that aposymbiotic (algal-free) Hydra viridis are able to recognize potential algal endosymbionts and reject non-symbiotic algae (Pardy and Muscatine, 1973). These workers have shown that symbiotic algae injected into the gastrovascular cavity of aposymbiotic hydra are phagocytosed and trans- ported to the base of the animals' digestive cells. With time the injected algae reestablish the symbiosis and restore the standing population of endosymbionts. Virtually nothing is known as to how the endocellular algal flora may be regulated or modulated by the host. Unstudied is the effect of various environ- mental conditions on the size and distribution of the resident algal flora. This 105 106 ROSEVELT L. PARDY paper describes the pattern of algal distribution in the hydra, how this pattern may be altered, and how continuous light, dark, and the animal's nutrition may affect the size of the algal population maintained by hydra. MATERIALS AND METHODS Maintenance of hydra cultures Stock cultures of Hydra viridis (Florida strain) were maintained in M solution (Muscatine and Lenhoff, 1965a) at room temperature and room illumination (approximately 50 foot-candles), hereafter referred to as ambient conditions. The animals were fed daily to repletion on freshly hatched Artemia nauplii. For dark experiments, animals were placed in 15 cm plastic petri dishes which were then wrapped with aluminum foil. These animals were exposed to no more than 2—4 minutes of light daily for counting, feeding and routine maintenance. In experi- ments requiring animals to be exposed to continuous illumination, animals in plastic petri dishes were placed on a white background 25 cm below two 20-watt Sylvania Gro-Lux lamps. Technique for counting the number of algae in Jiydra Endocellular algae were counted in digestive cells that had been isolated from hydra by maceration (David, 1973). In practice, either an individual or a piece of hydra was transferred to a drop of maceration fluid on a microscope slide. This solution consisted of glacial acetic acid, glycerine and water (1:1: lo-v:v). After 5 minutes in this fluid the individual cells were teased apart with fine dis- secting needles. Figure 1 is an example of a digestive cell prepared in this man- ner and demonstrates the presence of endocellular algae. The average number of algae/digestive cell was determined by counting directly the algae in digestive cells prepared by maceration. In an exponentially growing population of hydra there are individuals in all states of development, of varying size, and possessing a variable number of budding hydranths. Thus the selection of a "standard" experimental animal is somewhat arbitrary. For the present work a standard hydra was defined as an animal that had two budding hydranths in an advanced state of growth. To estimate the total number of algae in hydra, 10 standard animals were homogenized in 1 ml of M solution using a semi-micro tissue homogenizer. An aliquot of homogenate was transferred to a hemacytometer where the number of algae in the sample was determined. Counts from 5 groups of hydra were averaged and extrapolated to yield the average number of algae per hydra. In some experiments it was necessary to determine the standing crop of algae in an entire growing population of hydra. Since growing populations of hydra have individuals in various stages of development as mentioned earlier, a more useful index was found to be the number of algae per hydranth. Therefore the number of algae counted from homogenates of whole populations was expressed as the number of algae per hydranth. Inspection of intact hydra reveals that the animals are not uniformally green. The central region of the hydra appear greener than the stalk and base. Such GROWTH/DISTRIBUTION OF ALGAE IN HYDRA 107 an observation has been reported previously by Pardy and Muscatine (1973). As various environments and physiological factors might influence the distribution of algae within the animals, it was important to assess quantitatively the apparent heterogenous distribution of algal symbionts. To do this animals were divided into three zones (Pardy and Muscatine, 1973) : Zone 1, hypostome and tentacles; Zone 2, central growing region ; Zone 3, stalk and base. Zones were dissected from ten individual hydra (buds removed) and macerated separately. The number of algae in 20 digestive cells from each zone was counted. In experiments dealing with the effects of light and feeding on the number of algae per digestive cell (discussed below), only cells from Zone 2 were analyzed. Effect of light, dark and feeding on the nuniher of cndoccllnlar algae To determine the effect of light and nutrition on the number of endocellular algae, cultures of hydra were maintained under the following conditions: (1) ambient light conditions — fed every 24 hours, (2) ambient light conditions- starved, (3) constant illumination — fed every 24 hours, (4) constant illumination- starved, (5) constant darkness — fed every 24 hours, (6) constant darkness- starved. Animals for these experiments came initially from populations in exponential growth maintained under ambient light conditions and fed every 24 hours. Hydra from such cultures had 18 ±3.1 (Zone 2) algae per digestive cell (n — 100). At 24-hour intervals 100 digestive cells (20 from each of 5 animals were selected from each of the experimental groups — Zone 2) were examined and the number of algae in each cell was recorded. The data were expressed as the mean number of algae ± s.d. per digestive cell. GroivtJi rate of hydra Hydra's growth rate was measured according to the method of Loomis (1954). Standard animals with two buds were placed in 8 cm plastic petri dishes. Every morning the total number of hydranths was counted, the number recorded, and the animals fed to repletion. The average number of hydranths from duplicate experi- ments was plotted on semi-log paper against time. From a straight line fitted to these points, K, the growth rate constant was calculated using the standard equations for exponential growth. Growth rate of endosymbiotie algae Ten dishes each containing 20 hydranths were maintained in either continuous illumination or constant dark and fed every 24 hours. At 24-hour intervals fol- lowing the start of the experiment for a total of 5 days, the total hydra population in a dish was harvested and homogenized in 1 ml of M solution. The number of algae in an aliquot of the homogenate was determined by using a hemacytometer as described earlier and the average number of algae from two dishes was plotted on semi-log paper against time. Growth rates of the algae were calculated in the same manner as those for hydra. 108 ROSEVELT L. PARDY FIGURE 1. A digestive cell from green hydra prepared by maceration and photographed with phase optics. Symbiotic algae can be seen in the base of the cell. The animal cell nucleus with its prominent nucleohis is located centrally. Bar equals 10 microns. RESULTS Examination of cells from hydra prepared by maceration (Fig. 1) showed that the algal symbionts resided mainly in the base of the digestive cells. All digestive cells examined contained algae though individual digestive cells evidenced con- siderable range (2-30) in the number of algae contained in them. I found ap- proximately 1.5 X 105 ± 3.7 X 104 algae per hydra in a standard intact Hydra viridis, a value closely agreeing with that of Pardy and Muscatine (1973). Figure 2 is a histogram showing the distribution of algae in the three zones of hydra. It is evident that the algae are not uniformly distributed throughout the GROWTH/DISTRIBUTION OK ALGAE IN HYDRA 109 animal. Zone 2, the central growing region, averaged approximately 19 algae per digestive cell, whereas Zones 1 and 3 had fewer algae cells, averaging about 12 symbionts per digestive cell. Animals grown in continuous darkness became increasingly pale, however, I observed that Zone 3 appeared to be greener than Zones 1 and 2. As this condi- tion was different from animals maintained in continuous light (see above), I ZONE 1 5 10 15 20 algae per digestive cell 25 30 FIGURE 2. Histogram showing the distribution of algae in three zones of green hydra maintained in constant illumination and fed daily; Zone 1 — tentacles and hypostome, Zone 2 — central growth region, Zone 3 — stalk and basal disc. 110 ROSEVELT L. PARDY ZONE 1 5 10 15 20 algae per digestive cell 25 FIGURE 3. Histogram showing the distribution of algae in three zones of green hydra maintained in continuous dark for 6 days. Zones are the same as Figure 2. examined the distribution of algae in the three zones from hydra grown in the dark. Figure 3 is a histogram of the distribution of algae from hydra grown in the absence of light for 16 days. Compared to the zones from hydra reared in continuous light (Fig. 2), the results shown in Figure 3 may be summarized as follows : Zones 1 and 2 have fewer algae (averaging approximately 6 and 7 per digestive cell, respectively). Zone 3 had approximately the same number of algae (11 symbionts per digestive cell) as hydra grown in continuous light. These data show that the numbers and proportions of algae in Zones 1 and 2 (Fig. 2) GROWTH/DISTRIBUTION OF ALGAE IN HYDRA 111 are not fixed but may be altered by growing tbe animals in continuous darkness (Fig. 3). During growth in the dark, the number of algae are reduced in Zones 1 and 2, but apparently not in Zone 3. Table I shows the effects of the various illumination and feeding conditions on the number of algae in digestive cells from hydra. , Hydra which are fed and maintained under ambient conditions exhibited a nearly constant number of algae per digestive cell over the duration of the experiment. The remainder of the experimental results can be grouped in two classes : those in which there is a small, transient increase in algae followed by an approach to starting levels (Table I, ambient-starved, constant light-fed, constant light-starved), and those in which a pronounced, continuous decrease in the number of algae per digestive cell was observed (Table I, constant dark-fed, constant dark-starved). TABLE I Number of algae /digestive cell (Zone 2) from animals maintained under various conditions of light and feeding (see Materials and Methods). Data are expressed as mean ± s.d. of 100 digestive cells (20 from each of 5 hydra). Animals for the experiments were selected from a population of hydra that averaged 18 ±3.1 algae/digestive cell (Zone 2) on day 0 Experimental condition Day l 2 3 4 5 « Ambient-fed 18.2 ± 2.8 17.0 ± 2.0 17.0 ± 3.1 18.4 ± 2.6 17.9 ±3.2 Ambient-starved 20.4 ± 3.1 20.2 ± 4.1 20.1 ± 3.6 18.3 ± 2.8 18.0 ± 3.0 V Constant light-fed Constant light-starved Constant dark-fed 18.1 ± 4.6 20.0 ±4.1 13.5 ± 3.8 20.5 ±4.7 21.0 ± 5.7 12.3 ± 3.8 19.6 ± 3.8 22.2 ±3.8 10.6 ± 3.7 20.8 ±4.1 21.8 ±4.6 6.8 ± 2.3 19.1 ±4.5 18.3 ± 3.3 6.6 ± 3.1 Constant dark-starved 17.6 ± 3.8 17.0 ± 1.4 12.5 ± 3.4 11.0 ± 3.0 11.0 ± 3.2 Figure 4 is a semi-log plot of the increase in the number of algae and hydra in cultures maintained under continuous illumination and fed to repletion every 24 hours over a period of 4 days. The growth rate constant of hydra under these conditions was K -- 0.358 and of the algae, K = 0.380. The doubling time for the hydra and algae was approximately 1.9 days. The result of growing hydra under conditions of continuous dark while feeding every 24 hours is shown in Figure 5. It is evident that the hydra population keeps expanding exponentially though at a rate (K == 0.288) somewhat less than hydra maintained in continuous light (Fig. 2). The algae undergo only a very slight increase (K ~ 0.026) compared with those in the continuous light experi- ment (K== 0.380). When the average number of algae per hydranth from populations of hydra maintained under continuous light and dark is plotted against time on semi-log coordinates, the curves shown in Figure 6 are obtained. From these graphs it is apparent that the average number of algae per hydranth remains nearly constant in populations grown under continuous illumination. Conversely, the number of algae per hydranth in hydra grown in continuous dark declines (K = -- 0.232). Inspection of Figures 5 and 6 suggests that the decrease of algae observed in hydra which are grown in the dark may be correlated with the exponential growth 112 ROSEVELT L. PARDY of the animals. If the rate of algal decrease (K = -- 0.232) is corrected for the slight rate of algal division actually observed in the dark (K = 0.026) (Fig. 5), a new rate, K = - 0.258 is obtained. This value represents the theoretical rate of algal decline to be expected in dark-grown hydra if the algae were to cease dividing. Hence the rate of algal decrease should equal the growth rate of the animal hosts. The theoretical rate of algal decreases given above (K = 0.258) approximates the rate of animal multiplication (K == 0.288) and indicates that the loss of algae from hydra reared in the dark probably results from dilution of the standing crop of algae brought about by repeated animal cell division. The results from the experiment depicted in Table I (constant dark-fed), and Figures 4, 5 and 6 clearly emphasize the role of light in maintaining continuous maximum algal growth. Thus an experiment was devised to determine how long it would take the algae to repopulate hydra following depletion resulting from exponential-growth of the animals in the dark. Duplicate cultures of hydra were maintained in continuous darkness and fed every 24 hours. After 11 days it was determined that the hydra had approximately 7.1 X 103 algae per hydranth. The cultures were then placed in continuous light and the average number of algae per hydranth was determined every 24 hours for 4 days. Figure 7 shows that there o FIGURE 4. Growth of hydra and algae under continuous illumination and daily feeding- circles, hydranths ; triangles, algae ; growth rate of hydra, K — 0.388 ; growth rate of algae, K = 0.380. GROWTH/DISTRIBUTION OF ALGAE IN HYDRA 113 80^ 60- 40-1 o: a 20- 10- -8 -6 If) -42 L±J < CD _l < -2 2 DAYS i 3 FIGURE 5. Growth of hydra and algae in continuous darkness and with daily feeding- circles, hydranths; triangles, algae; growth rate of hydra, K = 0.288 ; growth rate of algae, K = 0.026. was a burst of algal multiplication over the first 2 days following the return of the animals to light. During this initial period the algae exhibited a growth rate of 1.37 — over three times the typical algal rate. After the period of rapid multiplica- tion the algal population reached and maintained a level typical of hydra grown in continuous light. DISCUSSION The effect of light and feeding on the growth of algae in hydra can be viewed as the consequence of two distinct but interrelated factors : the necessity of light for maximum algal multiplication, and tissue growth in hydra. Light has been shown to be a necessary condition for a variety of algal processes (photosynthesis, organic and inorganic nutrient assimilation, ion uptake) including division in certain strains of Chlorella vulgar is (Griffith, 1961). Algal multiplication in green hydra is strongly light dependent. Figure 5 shows that when hydra are transferred to the dark the algal population increased only slightly (K := 0.026). Under constant light and feeding, the number of hydra increases at an exponential rate of K = 0.358 and the algae increase at approximately the same rate (K - 0.380) (Fig. 4). These results are expected if animals in a growing population of hydra are to maintain a constant, optimum number of algae, and Table I, 114 ROSEVELT L. PARDY 6- cc Q 2- < _J < 0 I 2 DAYS \ 3 i 4 FIGURE 6. The number of algae per hydra in continuous light compared to the number of algae per hydra in animals maintained in continuous darkness. Both groups fed daily; the rate of decrease of algae, K = - 0.232 ; triangles, continuous light ; circles, continuous dark. ambient-fed constant light-fed, and Figure 4 indicate that the hydra do maintain a relatively constant number of algae during exponential growth in the light. 6- 4- (T Q LU O < r 2- 0 k= 1.372 2 DAYS ~r 4 FIGURE 7. Growth of algae in hydra after being transferred to continuous light following 11 days in constant darkness ; animals fed daily. GROWTH/DISTRIBUTION OF ALGAE IN HYDRA 115 When hydra is reared in the ahsence of light, the number of algae per digestive cell (Zone 2) and per hydranth rapidly declines (Table I, constant dark-fed; Fig. 6). TjTe._algae continue to growjn the dark thougjijvj?n/_^]x3wly (Fig. 5). The depopulation of algal symbionts observed in animals reared in the dark results from the growth and asexual reproduction of the host. The animals keep pro- liferating at a high rate relative to the algae and hence outgrow their symbionts. Further support for this argument comes from Table I, constant dark-starved, which shows that when hydra are starved in the dark fewer algae are lost from the digestive cells than when hydra are fed (Table I, constant dark-fed). In starved hydra the rate of animal growth decreases (Muscatine, 1961) and with time the hydra cease multiplying altogether. In the absence of hydra growth in the dark, the apparent loss of algae also stops. I have kept starved animals in the dark for up to 10 days without observing any loss of green color. It could be postulated that exponentially growing hydra maintained in the dark would eventually outgrow their algal symbionts. In fact, such an effect has not been observed. I have kept growing cultures of hydra in the dark for as long as four months with the animals maintaining a low level of infection that reproduces at the same rate as the hydra. *It is possible that in the dark the algae assume a heterotrophic mode of nutrition and derive all of their energy by the assimilation of metabolites from the hydra. The total number of heterotrophic symbionts that are maintained under dark conditions would thus be determined by density- dependent factors such as competition for nutrition available from hydra. Cook (1972) has shown that transfer of material from hydra to algae takes place. Hydra kept in both light and dark were fed 14C labeled Artemia nauplii. Assay of the algae after 48 hours showed that they had acquired 25-43% of the total radioactivity in the dark* versus 22-26% in the light. While the nature of the translocated substance(s) is not known, Cook's work (1972) clearly demonstrates that a flow of material from hydra to the algae takes place. This material is probably the source of heterotrophic nutrition for the algae. There is evidence suggesting that algae in the dark may be competing with hydra for metabolites. I have found that green hydra grows slower in the dark (Fig. 5, K == 0.288) than in the light (Fig. 4, K == 0.358) and that the animals growth rate may be depressed by as much as 20% in the dark. Thus in the dark there may be two levels of competition — between algae and between algae and host — the sum of which has the overall effect of depressing the growth of both symbiotic partners when they are in the dark. The absence of light, however, has an overall greater effect on the growth of the algae. My data show that hydra reared in the light have most of their endosymbionts located in the central growing region (Fig. 2). This situation may reflect that some essential nutrient (s) supplied by the hydra cells of the growth region and needed for maximal algal growth are limiting in the hypostomal (Zone 1) and basal (Zone 3) regions. Alternatively, hydra may actively regulate algal mitosis via some hormone-like mechanism, promoting division in the growth region and/or suppressing it in the basal and hypostomal region. Finally, algal mitosis might in some way be influenced by digestive cell mitosis. While there is no evidence to support these ideas, the work of Campbell (1967) is suggestive. Campbell (1967) demonstrated that while mitosis occurs throughout the body of the animal, 116 ROSEVELT L. PARDY the greatest amount of mitotic activity occurs in those cells (digestive, epithelial- muscular) of the central growing region with a substantially lower level in the base and hypostomal areas (Campbell, 1967). Recently, David and Campbell (1972) have found that the cell cycle of epithelial cells in the basal disk and pe- duncle of Hydra attenuate, is longer than those in the growing regions. The zonal distribution of algae that I have found in green hydra appears to correspond to the pattern of actively mitosing cells found in other hydra. Possibly the processes which initiate or are involved in mitosis of the hydra cells promote algal division. Hydra maintained in the dark exhibit a pattern of algal distribution (Fig. 2) differing from animals reared in the light (Fig. 3). In dark-reared animals there are more algae in the base (Zone 3) than in the other zones. Such a situation may be interpreted as follows : Loss of algae from hydra reared in the dark is a result of dilution of the standing crop of algae by the division of the digestive cells and is proportional to the growth rate of hydra. Because the animal cells constituting the various zones probably do not divide at the same rate (Campbell, 1967), the differential loss of algae observed in the various zones may reflect the differential mitotic activity of the digestive cells in these regions. The digestive cells of Zone 3 divide slower than the cells in the other zones and hence, in the dark, have proportionally more algae. Though direct experimental evidence is lacking, it is generally agreed that the endosymbiotic algal population in hydra is under some kind of regulation. The algae do not over-grow their hosts and seem to maintain a fairly constant number providing there is adequate light (Table I, ambient-fed, ambient-starved, constant light-fed). Possibly hydra actively regulates the division of its algal flora, main- taining a population compatible with the physiology and metabolism of the digestive cells. Alternatively, the ultimate number of algae in digestive cells (in the presence of light) may be density dependent. The quantity of algae could be limited either by competition among fellow symbionts for growth-promoting sub- stances supplied by the hydra or by growth-inhibiting factors released by the algae themselves. Hydra transferred to constant light, whether fed or starved, show a transient increase in the number of algae per digestive cell (Table I, constant light-fed, constant light-starved). Animals starved under ambient conditions also show this effect. If the algae are regulated by hydra, the slight algal increase ob- served may represent a temporary disruption in the control of algal division by the host or by a response to a change in physiological conditions, e.t instar 84 17 (20%) Posterior third First instar 87 1 (!',) Anterior third minus brain-ring gland complex First in>tar 99 5 (5%) Brain-ring gland complex First instar 74 14 (19', ) Imaginal disc Third instar 48 0 (.(»' , ) melanogaster were used as donors and hosts. The data for both were similar, but were more extensive for D. t'irilis, and so these appear in Table I. When abdomens of adult hosts were examined under the compound micro- scope, abnormalities were evident in 17 (20 per cent) of the 84 abdomens that had been implanted with the anterior third of first instar donors. Only 1 abnormality was encountered among 87 abdomens that had received the posterior third of first instar larvae. The abnormalities were localized in the cuticle immediately overlying the implant ; they were encountered only when the implant was superficial, never when it was deeply imbedded in the abdomen. As illustrated in Figure 1, the « r MM I Ml\,f ' II I 7 In 1 ttl \\- AC I / ' L A fC FIGURE 1. The effect of implanted juvenile endocrine organs on a metamorphosing host; A., Control adult abdominal cuticle from a host implanted with the posterior third of a first instar larva; B., Adult abdominal cuticle from a host implanted with the anterior third of a first instar larva ; C, Normal pupal cuticle. Symbols used are AB., abnormal bristles ; AC., aberrant cuticle; B., normal bristles; 1C., imaginal cuticle; PC., pupal cuticle. Scale bar equals 100 /u. JUVENILE HORMONE AND DROSOPHILA 123 abnormalities consisted of a localized zone closely resembling pupal cuticle and differing from adult cuticle in terms of the absence of pigment and hairs. So also, the bristles distinctive of adult cuticle were either absent or of aberrant size and shape. Vogt (1946) apparently observed similar inhibition of adult differentiation after the implantation of adult corpora allata into maturing third instar hosts of D. hydci. The ring gland of first instar larvae proved to be too small to be dissected and transplanted as such. Therefore, the brain-ring gland complex was removed and implanted into mature larval hosts (Table I). When the latter emerged as adults, 14 (19 per cent) of 74 individuals showed the typical integumentary defects. Here again, the local inhibition of metamorphosis was conditional upon the close proximity of implant and overlying cuticle. Control animals received the anterior third from which the brain-ring gland complex had been removed. Only 5 per cent of these animals showed the typical defect. This could be clue to the retention of the ring gland by some of the anteriors since the ring gland grasps the pharynx rather tenaciously. As a final control imaginal discs from mature larvae were im- planted into hosts of the same age, and none of these hosts showed any abnormali- ties. These studies, coupled with Vogt's (1946) results, show that the ring gland of adult Drosophila and the brain-ring gland complex of first instar Drosophila FIGURE 2. Effect of Cecropia juvenile hormone on the adult differentiation of the ab- domen of Drosophila; A., 0.05 /tig/animal ; B., 5 jig/animal. Symbols used are : AB., abnormal bristles; AC., aberrant cuticle; B., normal bristle; 1C., imaginal cuticle. Compare AC to Figure 1C. Scale bar equals 124 JOHN H. POSTLETHWAIT FIGURE 3. Scanning electron micrographs of : A., acetone treated control adult cuticle; B., juvenile hormone treated cuticle; C, pupal cuticle. Scale bar equals 20 p. cause local inhibition of abdominal metamorphosis. The aberrant cuticle is similar to pupal cuticle in that both lack pigment, bristles and hairs. In addition, sur- rounding the pupal-like regions are areas which have short, nicked, stubby or in- completely pigmented bristles. The molting of the implants in these studies, published elsewhere (Postlethwait, 1973), confirm Bodenstein's (1944) results and show that larval tissues must develop competence to respond appropriately to the environment at metamorphosis. Effect of c.vogenously supplied JH on flic metamorphosis of Drosophila Having ascertained the morphogenetic defects produced by implants containing juvenile endocrine organs, we centered further attention on the effects of topically applied, synthetic JH. Individuals receiving in excess of 0.05 p-g C18JH formed defective pharate adults which failed to emerge. As illustrated in Figures 2 and 3. the abdominal tergites of individuals receiving low doses of JH showed patches of cuticle identical to those produced by the implantation of brain-ring gland com- plexes. In the case of individuals treated with the highest doses, the entire "adult" abdomen was covered by an aberrant cuticle showing an almost complete suppres- sion of pigmentation and of bristles and hairs (Fig. 2B). The cuticle in the affected areas was impressively different from that of the controls (Fig. 1) and in many cases was indistinguishable from pupal cuticle over large areas. Although the aberrant cuticle could not be distinguished from pupal cuticle in whole mounts, it must be noted that the normal pupal cuticle lacks any projections or irregularities which in unstained whole mounts permit its positive and unambiguous identification. But we can conclude that exogenously supplied synthetic JH causes a general syndrome identical to that produced locally by an implanted active young larval brain-ring gland complex, and in both cases the aberrant cuticle is indistinguish- able from pupal cuticle. JUVENILE HORMONE AND DROSOPHILA 125 Developmental stages during metamorphosis which are most sensitive to exoge- nously supplied JH Effects on adult eclosion. Figure 4A records the percentage of flies which failed to eclose when specific doses of C18JH were administered at discrete ages ranging from the mid-third instar until one day after the initiation of pupariation. All except the lowest dose had major effects in blocking eclosion when adminis- 100 80 1/1 o _J u 60 < "• 40 20 -20 100 in 80 UJ _J t— I/I m , 60 O I 40 20 --.-" -1O O AGE •10 •2O -20 -10 0 AGE 10 20 60 120 180 240 300 360 -20 -10 10 20 AGE 100 m S Z o 80 60 40 20 D -10 0 AGE 10 20 FIGURE 4. Four parameters of JH activity and their relationship to age at time of treat- ment ; A., per cent failing to eclose vs. age ; B., failure of male genitalia to rotate ; C., per cent abnormal bristles ; D., per cent of bristles missing on the male 5th tergite and female 6th tergite; Stars, 5.0 /ig; open circles, 0.5 /*g; squares, 0.05 fig; filled circles, 0.005 /j.g. 126 JOHN H. POSTLETHWAIT tered during a certain critical period. The latter begins about 5 hours prior to the onset of pupariation and persists for approximately 25 hours thereafter. Maximal sensitivity includes the first 12 hours after pupariation. Effects on the rotation of the male genitalia. During a terminal phase of adult differentiation of Dipterous insects the male genitalia undergo a permanent rota- tion with respect to the long axis of the body (Scudder, 1971). In the lower Diptera the rotation is through 180° ; in the higher Diptera such as Drosophila, it is through 360° (Gleichauf, 1936). Genitalia rotation is partially or completely blocked in mosquitoes (Spielman and Williams, 1966) and Sarcophaga (Bhaskaran, 1972) derived from pupae exposed to JH analogues; the same is true for Droso- pliila according to Bryant and Sang (1968) and Madhavan (1973). These findings are confirmed in the present study. Thus, in Figure 4B, is recorded as a function of dosage and time of application the per cent by which the male genitalia failed to undergo the normal 360° rotation. Here again, the period of maximal sensitivity is the first 12 hours after the initiation of pupariation. Effects on bristle number and morphology. In order to quantify the inhibition of metamorphosis, a detailed study on the 5th abdominal tergite of males and the 6th of females was undertaken. These segments were chosen because they have both bristles and hairs (trichomes) and because the posterior tergites are affected to a greater degree than the anterior tergites. The male 5th tergite is char- acterized by 54 ± 4 bristles ; the female 6th tergite by 62 ± 3 bristles in normal flies or in controls treated only with acetone. Counts were made of the total number of bristles which had differentiated on these tergites, as well as the number which were abnormal. Figure 4D records the percentage of bristles (with reference to control values) as a function of dose and time of application. In Figure 4C is plotted the percent- age of abnormal bristles among those which had differentiated. Here again, we see that the effects are dose-dependent and that the sensitive period is from shortly before to about 15 hours after the initiation of pupariation. Effect of dose and site of application on integumentary dejects Topical application of the JH doses employed was not able to suppress the nor- mal adult differentiation of the integument of the head or thorax. Only an oc- casional individual which had received the highest dose (5 fig ) at the most sensi- tive stage showed a few abnormal bristles on the head, but no other trace of abnormal characters on either the head or thorax. By contrast, the adult dif- ferentiation of the abdomen was subject to inhibition by JH when the latter was applied during the sensitive period. The typical result was the formation of more or less extensive zones of unpigmented, hairless cuticle in which the bristles were either missing or of aberrant size and shape. Featureless zones within this cuticle were indistinguishable from pupal cuticle as shown above. These characteristics are illustrated in Figures 5B-5E in the case of four individuals that received graded doses of JH in the "white puparium" stage. For comparison, a normal adult integument is depicted in Figure 5A and a normal pupal integument in 5F. In the experiments summarized in Figure 6, graded doses of C18-JH were applied to individuals at a single highly sensitive stage — namely, at the outset of JUVENILE HORMONE AND DROSOPHILA V" = V ' ' • '<*'.•! •> :< 127 D FIGURE 5. Affect of increasing doses of C18JH on the sternites of Drosophila; A., acetone- treated control; B., 0.002 jtg; C, 0.005 Mg ; D., 0.05 jug; E., 0.5 /xg; F., Pupal cuticle; AB, abnormal bristles; AC, aberrant cuticle; 1C, imaginal cuticle. pupariation (white puparia). The several parameters of juvenilization are in each case plotted as a function of the logarithm of dose. Table II records the threshold doses, the saturating doses, and the doses that give 50% effects. When low doses of hormone were administered during the critical period, the posterior abdominal segments were more sensitive than the anterior segments and the abdominal sternites were routinely more effected than the tergites. Meanwhile, as mentioned above, the head and thorax remained insensitive at all stages. The insensitivity of the head and thorax was examined in two series of experi- ments. In the first of these the hormone was applied to the anterior ends of white prepupae rather than to the posterior ends. This shift in the site of application 128 JOHN H. POSTLETHWAIT 100 75 CO §50 0. co LU a: 25 .•*"" -V / / • / / • / / / :' ./a ; / -3 2 -1 LOG DOSE 0 FIGURE 6. Dose response curves ; stars, abnormal sternite bristles ; filled circles, failure to eclose ; squares, abnormal tergite bristles. had no effect on the differentiation of the adult head and thorax. Moreover, the effect on the abdomens was indistinguishable in the two cases except that the rota- tion of the male genitalia was somewhat more inhibited in the individuals that received the hormone at the hind end. TABLE II Quantitative aspects of the action of CIS JH on white puparia (Critical doses (/jg/individual)* Threshold 50% Effect 100% Effect Abnormal bristles on sternites 0.002 0.006 0.05 Failure to eclose 0.016 0.03 0.2 Inhibition of rotation of genitalia 0.005 0.03 0.2 Abnormal bristles on tergites 0.005 0.6 0.75 Each dose was topically applied in 0.3 jul acetone. JUVENILE HORMONE AND DROSOPHILA 129 TABLE III Relative abilities of JH analogues to block metamorphosis Approx- imate rank Analogue Number Designation Concen- tration /ig/g # of ani- mals tested % fail to eclose % with morpho- genetic effect on sternites Degrees male geni- talia rotated % control bristle # % aber- rant bris- tles per tergite 1 XIII epoxv 34 28 100 100 15 ± 12 36 ± 18 95 ± 10 geranyl sesamole 2 VIII iso-C!7-JH 34 15 87 100 54 ±40 70 ± 17 55 ±20 3 XXII ZR-515 34 22 100 100 100 ± 33 66 ± 21 65 ±34 4 IX C16-JH 34 15 60 100 67 ± 36 90 ± 12 39 ± 20 5 I C18-JH 34 47 76 100 87 ± 39 82 ± 15 36 ± 16 I C18-JH 17 15 27 100 240 ± 100 97 ± 13 19 ± 13 6 I +XXI C18-JH + 17 sesamex 17 15 33 100 200 ± 70 103 ± 10 13 ± 7 7 VI C17-JH 34 15 47 100 154 ± 52 91 ± 15 18 ± 15 8 XV R-20458 34 15 60 100 154 ± 79 102 ± 10 19 ±9 9 XVI R-19828 34 15 93 100 300 ± 42 104 ± 14 8 ±3 10 XIV geranvl sesamole 34 15 67 100 283 ± 67 104 ± 10 5 ±4 11 XII ClMe-Cl6-JH 34 13 0 100 265 ± 86 108 ± 11 4 ± 4 12 XI C16-JH-S 34 15 0 101) 308 ± 66 110 ± 11 4 ±5 13 III MBHF 34 15 0 100 342 ± 40 103 ± 14 1 ± 2 14 XX Williams-Law 34 15 0 78 360 ±0 107 ±11 0.4 ± 1 15 XIX EDCF 34 15 0 29 360 ± 0 102 ± 11 0.2 ± 0.8 16 IV Farnesenate 34 14 0 43 356 ± 1 1 102 ± 10 0.1 ±0.5 17 V Farnesol 34 14 0 50 360 ± 0 102 ± 9 0 ± 0 18 II C18-JH-N 34 14 7 14 360 ± 0 108 ± 10 0.1 ±0.1 19 X C16-JH-N 34 14 0 0 360 ± 0 110 ± 8 0 ± 0 20 VII C17-JH- aldehyde 34 15 0 0 360 ± 0 98 ± 7 0 ± 0 21 XVII Epoxy geranyl methyl benzoate 34 14 0 (1 360 db 0 102 ± 9 0 ± 0 22 XVIII geranyl methyl henzoate 34 10 0 0 360 ± 0 98 ± 12 0 ±0 23 XXI Sesamex 34 15 0 0 '360 ± 0 100 ± 12 0 ±0 24 Acetone 26 0 0 360 ± 0 100 ± 9 0 ± 0 In the second series of experiments the operculum was removed from puparia and the hormone was applied directly to the pupal head. (This operation cannot be carried out before the 12th hour since, prior to that time, the epidermis is firmly attached to the operculum.) In each of the treated individuals the head and thorax underwent essentially normal adult development. A few bristles were bent abnormally, but this occurred also on animals whose operculum was removed and who were then treated with acetone. The effects on the abdomen could not be distinguished from that seen in controls in which the hormone was applied to intact puparia of the same age. The relative abilities of different JH analogues to block metamorphosis in As a check on the general types of molecules which have JH activity Drosophila, I compared the morphogenetic effects of 22 different JH analogues (Table III). The conclusions drawn below are subject to the fact that the isomer composition of most of the compounds tested is unknown. The study shows that for Drosophila the synthetic JH analogues epoxy geranyl sesamole and ZR-515 130 JOHN H. POSTLETHWAIT are both more active than any of the hormones isolated from Lepidoptera (com- pounds I, VI and IX, Roller, Dahm, Sweely and Trost, 1967; Meyer, Schneider- man, Hanzmann and Ko, 1968; Dahm and Roller, 1970; Judy, Schooley, Dunhan, Hall, Bergot and Siddall, 1973). A comparison of JH analogue activities shows that for Drosopliila a methyl group at position 11 is better than ethyl since iso- C17-JH (VIII) is more active than C1S-JH (I), and C16-JH(IX) is more active than C17-JH (VI). An epoxy ring is better than a double bond between carbons 10 and 11 since C18-JH (I) is more active than methyl bishomofarnesoate (HI) and epoxy geranyl sesamole (XIII) is more active than geranyl sesamole (XIV). An epoxy ring is better than an imino ring or a thio ring between carbons 10 and 11 since C16-JH (IX) is more active than C16-JH-aziridine (X) or C-16-JH- episulfkle (XI) and C18-JH (I) is more active than C18-JH-aziridine (II). An ethyl group at position 7 is better than a methyl since iso-C17-}H (VIII) is more active than C16-JH (IX) and C18-JH (I) is more active than C17-JH (VI). A methyl group at position 7 is better than a chloromethyl since C16-JH (IX) is more active than chloromethyl-C16-JH (XII). Sesamole is more active than other modified benzenes in an ether link to geranol since epoxy geranyl sesamole (XIII) is more active than R-20458 (XV), R-19828 (XVI) or epoxy geranyl methyl benzoate (XVII). DISCUSSION These experiments show that a characteristic syndrome of abdominal cuticular abnormalities can be induced in DrosopJiila by either implanting a first instar brain-ring gland complex into a mature larva, or by topically treating a freshly pupariated Drosophila with Cecropia juvenile hormone. The characteristics of the aberrant cuticle involve a failure of the cuticle to pigment, or to form cell hairs. Surrounding these areas are bristles which are short, nicked, or incompletely meta- morphosed. The bristle irregularities confirm the results of earlier workers (Dearden, 1964; Bryant and Sang, 1968, 1969; Ashburner, 1970b; Madhavan, 1973). Since the entire brain-ring gland complex (including several endocrine organs) was transplanted, the precise inducer of the aberrant cuticle is in question. But the evidence is consistent with the interpretation that the corpus allatum part of the ring gland inhibits metamorphosis. First, either ring glands or brain-ring gland complexes of Calliplwra when transplanted into fifth instar hosts of Rhodnius in- duce the retention of larval cuticle in the adult (Wigglesworth, 1954). Secondly, adult corpora allata implanted into mature larval hosts inhibit normal adult develop- ment of the abdomen in Drosopliila hydci (Vogt, 1946) or Calliphora (Possompes, 1953), causing the same syndrome as appeared in our experiments. Thirdly, first instar ring glands of D. hydei implanted into adult females which had been de- prived of their ring glands induced vitellogenesis, whereas brains alone failed to induce egg maturation (Vogt, 1943). It therefore seems that the ring gland of flies produces a juvenile hormone, and it may be that transplanted young ring glands caused the aberrant cuticle in our transplants. The experimentally induced aberrant cuticle was indistinguishable from pupal cuticle, but that does not prove that it is pupal cuticle. Since pupal cuticle is JUVENILE HORMONE AND DRUSOPIIILA 131 devoid of unique and specific cuticular markers, its positive identification using purely morphological criteria is frustrated. Five higher flies other than Drosophila have been tested for their response to juvenile hormone analogues: Sarcophaga bullata (Srivastava and Gilbert, 1968, 1969; Bhaskaran, 1972), Sarcophaga argyrostonia (Fraenkel and Hisao, 1970), Calliphora erythrocephala ( Pihan and Bautz, 1970), Stomo.\-ys calcitrans (Wright, 1970; Wright and Rushing. 1973; Wright, Crookshank and' Rushing, 1973), and Mil sea domcstica (Herzog and Monroe, 1972; Cerf and Georghiou, 1972). These studies showed that JH analogues inhibited eclosion and metamorphosis and sup- pressed the differentiation of hairs, bristles and pigment characteristic of the adult fly. In three of these investigations histological sections were prepared. Srivas- tava and Gilbert (1968, 1969) interpreted their sections as showing two pupal cuticles, the second apparently being induced by JH. Wright (1970) and Bhas- karan (1972) found no second pupal cuticle nor a pupal-adult intermediate cuticle. Bhaskaran (1972) suggested that the "second pupal cuticle" described by Srivastava and Gilbert (1968, 1969) may be undifferentiated adult cuticle or the undigested inner layers of the first pupal cuticle. The latter explanation could not hold for the pupal-similar cuticle reported here for Drosophila since the normal and aberrant cuticle are contiguous in treated abdomens. The aberrant cuticle in Drosophila may be merely undifferentiated adult cuticle which, by chance, is indistinguishable from pupal cuticle. The available data do not allow us to distinguish between the possibilities that in Drosophila juvenile endocrine glands and synthetic Cecropia JH induce a second pupal cuticle in the abdomen, or rather an aberrant adult cuticle that cannot be distinguished from pupal cuticle in whole mounts. Biochemi- cal or immunological identification of pupal-specific cuticular protein could resolve the dilemma. The procedure developed here provides a simple quantitative method for the assay on Drosophila of substances suspected to possess JH activity. This test is a sensitive measure of JH activity — as little a 0.0015 ^g C18-JH can be detected. The dose-response curves show that for the posterior three sternites the dose resulting in 50% morphological inhibition of metamorphosis (ID50Morph., Slama, 1971) is 5 /Ag/g live weight. The ID-,0Morph. for topically applied JH is about 8 /Ag/g for Tenebrio, and 25 //.g/g for Pyrrhocoris (Slama, 1971 ; Williams, 1970). It therefore appears that the Drosophila abdomen is about as sensitive to JH as other insects. In many insect orders the male genitalia rotate to accommodate the female during copulation. In Diptera the rotation occurs during development and is permanent. This rotation involves the abdominal tissue surrounding the genitalia rather than the genitalia per sc (Gleichauf,, 1936; Scudder, 1971). One of the effects noted after JH application to Drosophila is failure of the male genitalia to complete rotation (Bryant and Sang, 1968). Failure to complete rotation occurs also after treatment of the mosquito Aedes acgypti (Spielman and Williams, 1966) and Sarcophaga (Bhaskavan, 1972) with JH analogues. The JH sensitive stage for the rotation in Drosophila is from shortly before pupariation until about 15 hours (Fig. 5B). This sensitive phase is well prior to the actual time of rotation for Drosophila, which does not occur until the late stages of metamorphosis (Gleichauf, 1936). 132 JOHN H. POSTLETHWAIT These results confirm that the abdomen is quite sensitive to JH while the head and thorax are not and the differentiation of the genitalia is less sensitive than the abdomen (Dearden, 1964; Bryant and Sang, 1969; Ashburner, 1970b; Madhavan, 1973). This difference is probably not due to the method of application since treatment on the anterior or posterior gives similar results and removal of the operculum to allow application of hormone directly to the pupal head does not cause the head and thorax to be affected. In addition, feeding the hormone after the method of Bryant and Sang (1969) elicits the same syndrome of abnormalilities which are described here for topical application. If the difference in response of different tissues is not an artefact it must have some basis in the developmental physiology of various body parts. The develop- ment of the epidermis of the imaginal abdomen differs from the development of the head, thorax, and genitalia in two significant respects. First, the adult abdomen of higher flies arises from clusters of histoblasts, which undergo little or no cell division from embryogenesis until the time of pupariation (Robertson, 1936; Garcia-Bellido and Merriam, 1971a; Guerra, Postlethwait and Schneiderman, 1973). The other integumentary structures arise during embryogenesis as sacs of cells, called imaginal discs. The imaginal disc cells proliferate rapidly during larval life, but slow down after pupariation (Becker, 1957; Bryant and Schneider- man, 1969; Bryant, 1970; Garcia-Bellido and Merriam, 1971b; Postlethwait and Schneiderman, 1971; Ulrich, 1971). Thus, when cell division rate is high in the abdomen, cell division rate is low in the other imaginal anlagen. Our results (Fig. 5) show that the time of JH sensitivity for the abdomen is precisely the time of rapid proliferation of abdominal histoblast cells. A correspondence of JH sensitivity and high proliferation rate appears to be a general phenomenon in the insect integument. For example, in Rhodnius, sensi- tivity to exogenously supplied JH analogues is greatest at 8 days after feeding, which is the time mitosis is beginning in the epidermis (Wigglesworth, 1963). In some cases an increased proliferation rate can be induced experimentally. Piepho and Heims (1952) demonstrated that regenerating cells of the waxmoth GaUcria inellonclla are much more sensitive to JH than undividing cells. This fact has been exploited in the very sensitive "wax test" for JH activity (Schneider- man and Gilbert, 1958). The second main difference between abdominal and anterior tissues is that the pupal cuticle of the abdomen is secreted at least largely by the larval abdominal cells, while the pupal cuticle of the anterior is secreted by the imaginal disc cells. So it is not clear whether the aberrant (pupal-like) adult cuticle in Drosophlla is secreted by larval or imaginal cells. It is clear for Sarcophaya that the aberrant adult cuticle is secreted by imaginal cells (Bhaskaran, 1971). Further histological studies are in progress to answer this question for Drosophila. The correspondence between cell division and JH activity has been thought to be fundamental to the action of the hormone. Williams and Kafatos (1971) have proposed that cell division in the presence of JH delays the activation of new gene sets, whereas cell division in the absence of JH permits new programs of genetic information to be utilized. Perhaps further hormonal research on Drosophila, in which genetics can be effectively used to investigate developmental problems, will lead to a better understanding of such problems. JUVENILE HORMONE AND DROSOPIULA 133 Thanks are clue to Professor Carroll M. Williams for helpful discussions during the course of this work and for comments on the manuscript and to Mr. Doug Sears for technical assistance. The research was supported by a grant from the Rockefeller Foundation awarded to Professor Carroll M. Williams, and NIH grant GM 19307 awarded to J. H. P. SUM MA in' Topical application of Cecropia juvenile hormone ( JH ) blocked metamorphosis of the abdominal integument of pupating Drosopliila larvae. JH treated adult flies retained more or less extensive areas of cuticle which could not be distinguished from pupal cuticle in whole mounts. Implantation into mature larvae of fragments of first instar larvae containing brains and ring glands provoked localized retention of patches of incompletely metamorphosed cuticle. An assay procedure for com- pounds with juvenile hormone activity was developed by defining the critical developmental stage and dose-response relationships. Twenty-two substances suspected of having juvenile hormone activity were assayed for their effect on Drosophila. These findings support the choice of Drosophila for resolving the genetic basis of juvenile hormone action. LITERATURE CITED AJAMI, A., 1973. Biochemical studies on the metabolism and biosynthesis of the cecropia juvenile hormone. Ph. D. thesis., Harvard I 'nirersity. ASHBURNER, M., 1970a. Function and structure of polytene chromosomes during insect development, .-Idvan. Insect. Pliysiol.. 7 : 1-95. ASHBURNER, M., 197()b. Effects of juvenile hormone on adult differentiation of Drosophila tnchmogaster. Nature, 227:187-189. BECKER, H. J., 1957. Uber Rontgenmosaikflecken und Defektmutationen am auge von Drosophila und die Entwicklungsphysiologie des Auges. Z. Indukt. Abstamm. Vcrerbungsl, 88 : 333-373. BHASKARAN, G., 1972. Inhibition of imaginal differentiation in Sarcophaga bullata by juve- nile hormone. /. E.rp. Zoo!., 182 : 127-142. BODENSTEIN, D., 1944. The induction of larval molts in Drosophila. liiol. Bull, 86 : 113-124. BRYANT, P. J., 1970. Cell lineage relationships in the imaginal wing disc of Drosophila melanogaster. Develop. Biol, 22 : 389-411. BRYANT, P., AND J. SANG, 1968. Lethal derangements of metamorphosis and modifications of gene expression caused by juvenile hormone mimics. Nature, 220 : 393-394. BRYANT, P., AND J. SANG, 1969. Physiological genetics of melanotic tumors in Drosophila melanoc/astcr. VI. The tumorigenic effects of juvenile hormone-like substances. Genetics, 62 : 321-336. BRYANT, P. J., AND H. SCHNEIDERMAN, 1969. Cell lineage, growth and determination in the imaginal leg discs of Drosophila melanogastcr. Develop. Biol., 20: 263-290. CERF, D. AND G. GEORGHIOU, 1972. Evidence of cross-resistance to a juvenile hormone analogue in some insecticide resistant houseflies. Nature, 239 : 401-402. CHAN, L., AND W. GEHRING, 1971. Determination of blastoderm cells in Drosophila mclano- gaster. Proc. Nat. Acad. Sci., 68: 2217-2221. DAHM, K. H., AND H. ROLLER, 1970. The juvenile hormone of the giant silk moth Hyalophora gloveri (strecker). Life Sci. II, 9 : 1397-1400. DEARDEN, M., 1964. Experiments on the effect of farnesol on the development of normal and Bar-eyed Drosophila. J. Insect. Pliysiol., 10 : 195-210. DOANE, W., 1973. Role of hormones in insect development. Pages 291-498 in S. J. Counce and C. H. Waddington, Eds., Developmental Systems: Insects, Volume 2. Academic Press, London and New York. 134 JOHN H. POSTLETHWAIT FRAENKEL, G., AND C. HSIAO, 1968. Alorphological and endocrinological aspects of pupal diapause in a fleshfly, Sarcophaga argyrostoma. J. Insect Physiol., 14 : 707-714. GARCIA-BELLIDO, A., AND J. MERRIAM, 1971a. Clonal parameters of tergite development in Drosophila. Develop. Biol, 26 : 264-276. GARCIA-BELLIDO, A., AND J. MERRIAM, 1971b. Parameters of wing imaginal disc development of Drosophila melanogaster. Develop. Biol., 24 : 61-87. GLEICHAUF, R., 1936. Anatomic uncl Variabilitat des Geschlechts apparates von Drosophila mclanogastcr (Meigen). Z. Wiss. Zool., 148: 1-66. GUERRA, M., J. POSTLETHWAIT AND H. SCHEXIDERMAN, 1973. The development of the imaginal abdomen of Drosophila melanogaster. Develop. Biol., 32 : 361-372. HERZOG, A., AND R. MONROE, 1972. EfTect of synthetic juvenile hormone and citric acid on housefly pupae, Musca domcstica L. Conip. Biochcni. Physiol., 41A : 481-486. JUDY, K., D. SCHOOLEY, L. DTXHAM, M. HALL, J. BERGOT AND J. SIDDALL, 1973. Isolation, structure, and absolute configuration of a new natural insect juvenile hormone from Manuca se.rta. Proc. Nat. Acad. Sci., 10: 1509-1513. KARLSON, P., AND G. HANSER, 1952. Uber die Wirkung des Puparisierungshormons bei der Wildform und der Mutante Iql von Drosophila melanogaster. Z. Natnrjorsch., 7b : 80-83. LAW, J. H., C. YUAN AND C. M. WILLIAMS, 1966. Synthesis of a material with high juvenile hormone activity. Proc. Nat. Acad. Sci., 55 : 576-578. MADHAVAN, K., 1973. Morphogenetic effects of juvenile hormone and juvenile hormone mimics on adult development of Drosophila. J. Insect Physiol., 19 : 441-453. MEYER, A. S., H. A. SCHNEIDERMAN, E. HANZMANN AND J. H. Ko, 1968. The two juvenile hormones from the Cecropia silk moth. Proc. Nat. Acad. Sci., 60 : 853-860. POSSOMPES, B., 1953. Recherches experimentales sur le determinisme de la metamorphose de Calliphora erythroccphala Meig. Arch. Zool. E.vp. Gen., 89 : 203-364. PIEPHO, H., AND A. HEIMS, 1952. Das Cuticula Muster der Schmetterlings larve und die hormonale Grundlage seiner Entstehung. Untersuchungen an der Wachsmotte Caller in niclanclla L. Z. Naturjorsch., 7b : 231-237. PIHAN, J. C., AND A. M. BAUTZ, 1970. Utilization des mimetiques de 1'hormone juvenile comme agents teratogenes de la morphogenese tegumentaire chez Calliphora erythro- ccphala Meig. f. R. Acad. Sci. Paris, 270D : 2973-2976. POSTLETHWAIT, J., 1973. Molting of Drosophila first instar larval cuticle induced by a meta- morphosing host. Drosoplwphila Inform. Serv.. 50: 186. POSTLETHWAIT, J., AND H. SCHNEIDERMAN, 1971. A clonal analysis of development in Drosophila melanogaster: Morphogenesis, determination and growth in the wild type antenna. Develop. Biol.. 24 : 477-519. POSTLETHWAIT, J., AND K. WEISER, 1973. Vitellogenesis induced by juvenile hormone in the female sterile mutant apterous-four in Drosophila melanogaster. Nature New Biol., 244 : 284-285. RAYLE, R. E., 1967. A new mutant in Drosophila mclanogastcr causing an ecdysone-curable interruption of the prepupal molt. Genetics, 56 : 583. ROBERTSON, C., 1936. Metamorphosis of Drosophila mclanogastcr including an accurately timed account of the principal morphological changes. /. Morphol., 59 : 357-399. ROLLER, H., K. H. DAHM, C. SWEELY AND B. M. TROST, 1967. The structure of the juvenile hormone. Angeiv. Chem., Intern. Ed., 6: 179-180. SCHNEIDERMAN, H., AND L. GILBERT, 1958. Substances with juvenile hormone activity in Crustacea and other invertebrates. Biol. Bull., 115: 530-535. SCUDDER, G., 1971. Comparative morphology of insect gentalia. Ann. Rev. Entomol., 16: 379-406. SHEARN, A., T. RICE, A. GAREN AND W. GEHRING, 1971. Imaginal disc abnormalities in lethal mutants of Drosophila. Proc. Nat. Acad. Sci., 68 : 2594-2598. SLAMA, K., 1971. Insect juvenile hormone analogues. Ann. Rev. Biochcni., 40: 1079-1102. SPIELMAN, A., AND C. WILLIAMS, 1966. Lethal aspects of synthetic juvenile hormone on larvae of the yellow fever mosquito, Acdcs acgypti. Science, 154: 1043-1044. SRIVASTAVA, T., AND L. GILBERT, 1968. Juvenile hormone: Effects on a higher Dipteran. Science, 161 : 61-62. JUVENILE HORMONE AND DKOSOPHILA 135 SRIVASTAVA, T., AND L. GILBERT, 1969. The influence of juvenile hormone on the meta- morphosis of Sarcophage bullata. J. Insect Physiol., 15: 177-189. STEWART, M., C. MURPHY AND J. W. FRISTROM, 1972. The recovery and preliminary char- acterization of x-chromosome mutants affecting imaginal discs of Drosphila melano- gaster. Develop. Biol., 27 : 71-83. ULRICH, E., 1971. Cell lineage, Determination, und Regulation in der weiblichen Genital- imaginalscheibe von Drosophila melanogaster. Develop. Biol., 20: 411-425. URSPRUNG, H., 1967. In vivo culture of Drosophila imaginal discs. Pages 485-492 in F. H. Wilt, and N. K. Wessells, Eds., Methods in Developmental Biology. Thomas Y. Crowell Co. VOGT, M., 1943. Zur produktion gonadotropen Homons durch Ringdriisen des ersten Larvin- stadiums bei Drosophila. Biol. Zb., 63 : 467-470. VOGT, M., 1946. Inhibitory effects of the corpora cardiaca and of the corpus allatum in Drosophila. Nature, 157: 512. WIGGLESWORTH, V. B., 1954. Secretion of juvenile hormone by the corpus allatum of Calli- phora. Nature, 174: 556. WIGGLESWORTH, V. B., 1963. The juvenile hormone effect of farnesol and some related com- pounds: Quantitative experiments. /. Insect Physiol., 9: 105-119. WIGGLESWORTH, V. B., 1970. Insect Hormones. Oliver and Boyd, 168 p. WILLIAMS, C. M., 1970. Hormonal interactions between plants and insects. Pages 103-132 in E. Sondheimer and J. B. Sineone, Eds., Chemical Ecology. Acad. Press, New York. WILLIAMS, C., AND F. C. KAFATOS, 1971. Theoretical aspects of the action of juvenile hormone. Mitt. Schzi'eiz. EntomoL GcselL, 44: 151-162. WRIGHT, J. E., 1970. Hormones for control of livestock arthropods. Development of an assay to select candidate compounds with juvenile hormone activity in the stable fly. /. Econ. EntomoL, 63 : 878-883. WRIGHT, J. E., AND D. D. RUSHING, 1973. Glycogen in pupal and adult stable flies as affected by an insect juvenile hormone analogue. Ann. EntomoL Soc. Amer., 66 : 274-276. WRIGHT, J. E., H. R. CROOKSHANK AND D. D. RUSHING, 1973. Glycogen phosphorylase activity in pharate adults of the stable fly and the effects of a juvenile hormone analogue. /. Insect Physiol., 19: 1575-1578. WYATT, C., 1972. Pages 385-490 in G. Litwack, Ed., Biochemical Actions of Hormones, volume 2. Academic Press, New York and London. Reference: Biol. Bull., 147: 136-145. (August, 1974) HARVEST OF PLANKTONIC MARINE ALGAE BY CENTRIFUGA- TION INTO GRADIENTS OF SILICA IN THE CF-6 CONTINUOUS-FLOW ZONAL ROTOR1 C. A. PRICE, L. R. MENDIOLA-MORGENTHALER, M. GOLDSTEIN, E. N. BREDEN AND R. R. L. GUILLARD 2 Particle Separation Facility, Department of Biochemistry and Microbiology, Rutgers University, New Brunswick, New Jersey 08903 The harvest of the more common algae from laboratory cultures rarely presents serious problems. However, the collection (or concentration) of living algal cells in good physiological condition from dilute cultures or from natural plankton populations is much more difficult. The two major devices commonly used are nitration and centrifugation. Filtration is carried out commonly on membranes of modified cellulose (Clarke and Sigler, 1963), with the aid of a suction pump. The greatest advantage of this method as a concentrating device is that it is able to collect microalgae or cells of very low density. However, concentration by nitration is limited to small volumes and leads to the eventual clogging of the filter by the packed cells when vacuum is applied. Several methods have been devised which avoid these problems. One involves the use of a reverse-flow vacuum (Dodson and Thomas, 1964) in which the pressure operates from above, making the process more gentle and avoiding the packing of cells. This method itself has been modified to allow a relatively large volume of water to be concentrated in a short period of time (20 liters to 300 ml in 3 hours) (Holm-Hansen, Packard and Pomeroy, 1970). A second process uses a direct vacuum but involves a stirring blade in the flask above the filter which prevents the particles from settling at all during the concentration process (Morris and Yentch, 1972). Continuous-flow centrifugation with the classical Foerst rotor or the Szent- Gyorgyi-Blum modification of the Sorvall rotor is another widely used method. This method is reasonably efficient, but sensitive algal cells may be damaged by pelleting against the rotor wall and the method is essentially unselective ; all particles with a sedimentation rate above some limiting value will be collected. The variant on zonal centrifugation known as continuous sample-flow with iso- pycnic banding (rf. Cline and Ryel, 1971 ) offers a number of theoretical advantages in the concentration (and simultaneous purification) of particles: large capacity, more efficient recovery at substantially lower speeds than are required for conven- tional continuous-flow centrifugation, and avoidance of pelleting. Plankton, including algae, have been collected in sucrose gradients in the B-XYI and K-II zonal rotors (Lammers, 1971), but we do not know either the efficiency of recovery or the integrity of the recovered algae. 1 Supported in part by a contract from the MARMAP Program of the National Marine Fisheries Service and an NSF Grant (GA 33288) to R. R. L. Guillard; paper of the Journal Series, New Jersey Agricultural Experiment Station, Cook College, Rutgers Uni- versity, New Brunswick, New Jersey 08903 ; contribution no. 3252 from the Woods Hole Oceanographic Institution. 2 Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543. 136 PLANKTONIC ALGAL HARVESTING 137 TABLE I Algae investigated Species Clone Class Approximate dimensions Origin Dunaliella tertiolecta Dun Chlorophyceae 6.5 X 10 M Unknown ; possibly de- rived from Plymouth #83 Pyramimonas sp. Pyr-I Prasinophyceae 8X15/1 San Francisco, Cali- fornia Tha lassiosira fluviatilis Synechococcus bacillaris Actin Syil Bacillariophyceae Cyanophyceae 20 M (isodiametric) 0.75 X 1.5 /x Long Island Sound (Gardiner's Island) Milford, Connecticut In any event it should be possible to harvest algae with centrifuge systems \vhich are much slower and less costly than the zonal centrifuges designed for the purification of viruses. We have, therefore, tested an alternative means of harvest- ing algal plankton in the CF-6 rotor (Casciato, 1973), a continuous-flow zonal rotor. MATERIALS AND METFIODS Organisms and conditions of growth The algal strains used in this work are listed in Table I, together with clonal designation, algal class, origin, and approximate size of each. Algae will be refer- red to by clonal designation in this paper. All clones are maintained by R. R. L. Guillard at the Woods Hole Oceanographic Institution. These particular species are taxonomically varied and cover a range of sizes, though all are in the smallest size class of the phytoplankton (the nannoplankton). Two clones (Dun and Pyr- 1 ) are naked flagellates ; the blue-green algal clone Syn has a multilayered wall with protein, carbohydrate, and mucopeptide components; the diatom Actin has a wall composed of a silica shell in addition to an organic matrix. We expected the structural variability to influence the density during centrifugation. The chlorophyte Dun was particularly iiseful because of its motility and general hardiness; a visual indication of its viability under gradient conditions and manipulation could be ob- tained by observing motility. All algae were grown in an enriched articificial seawater. A seawater base was made by using the major constituents (MgSOi, MgCl2, CaCl2, KG, NaHCO3, and NaCl) of ASP-M (McLachlan, 1964) plus 1 ml/1 of an 0.6% (w/v) solu- tion of HsBOs. This seawater base was then supplied with the nutrients of en- richment "f" (Guillard and Ryther, 1962) with all nutrients at half the concentra- tions specified. Cultures were grown at 20-24° C under continuous illumination of 3500-6000 lux from cool-white (Sylvania Co.) fluorescent lamps. Cultures were grown in Erlenmeyer flasks or carboys with aeration. Preparation of gradients The silica gradients were prepared as follows: Ludox AM (E. I. du Pont de Nemours and Co., Wilmington, Delaware), a preparation of colloidal silica with 138 PRICE ET AL. partial substitution of Si by Al, was allowed to dialyze against distilled water for at least two weeks; water was changed daily. The Ludox was then collected and used for several runs. The reason for this dialysis is that, contrary to the evidence of others with animal cells (Mateyko and Kopac, 1963; Pertoft, 1969), Ludox AM was found to be highly toxic to algal cells, marine algae as well as Euglena gracilis. We suspect that the toxicity is due to the presence of a biocide in the commercial product. Two weeks of dialysis does not completely remove the toxic quality The gradients used were: 40% to 70% v/v dialyzed Ludox for Dun, Pyr-I, Syn and 20% to 50% v/v dialyzed Ludox for Actin (see graphs in Results). Gradients of zero to 3% w/v NaCl ran counter to the Ludox to minimize gelling. The appropriate gradients were established by trial runs with step gradients in swinging buckets. The banding densities were monitored by the use of calibrated density beads (Clark Wilcox and Assoc., Los Altos, California). Gradients were also prepared from mixtures of Ficoll in seawater. Ficoll is a trade name for polysucrose, available from Pharmacia Fine Chemicals, New Market, New Jersey. Harvest and centrifugation Cells were harvested in the 620-ml CF-6 zonal rotor (lEC/Damon, Needham Heights, Massachusetts), in the PR-6 (IEC) centrifuge. The gradient con- sisted of 235 ml of starting solution, 220 ml of limiting solution and 115 ml of underlay. The starting and limiting solutions were mixed in the IEC gradient former and pump (#3651) to give a gradient which was linear with volume, and were pumped into the rotor at 40 ml/min. The rotor speed during loading was 1500 rpm. The sample was loaded at 100 or 200 ml/min by either the Harvard (#600-00) or the Cole-Parmer Masterflex (#WZ-1R031) peristaltic pump. The sample was then run for 15 min at 4000 rpm to band the particles, slowed to 1000 rpm, and the gradient displaced by pumping at 35 ml/min. The gradient was collected in 30-ml fractions in chilled graduated test tubes. In order to study the efficiency of collection, samples of the algal suspensions were fixed with 1% glutaraldehyde and the cell concentrations determined sub- sequently in a Coulter Counter by Dr. J.-J. Morgenthaler. Background counts were taken for comparison with counts from supernatants and original culture samples (see Table II). Analyses Growth. Relative cell concentration was measured by fluorometry of the chlorophyll (Knight, 1968). A 436 nm interference primary filter and a #66 Klett secondary filter were used in a Turner Model III fluorometer. Fluorescence was found to be proportional to cell number. Daily samples of about 4 ml were taken aseptically from the cultures in 250-ml culture flasks. Where dilution was necessary, it was done with artificial seawater. The blank was also artificial sea- water. Growth rates and progress curves were obtained by plotting relative fluorescence against time on a semilog scale (Fig. 3 a-d). Chlorophyll. The concentrations of cell material within the peak zones of the gradient were established by measuring the amounts of chlorophyll spectrophoto- PLANKTONIC ALGAL HARVESTING 139 metrically. This method was also used to compare the chlorophyll content of the original culture suspension with that of the harvested cells. One milliliter from a 30-ml gradient fraction or 100 ml of the culture suspension concentrated to 1 ml by centrifugation were placed in 4 ml of 100% acetone and allowed to remain in the cold and dark for 1/2 hour. They were then centrifuged at 2000 rpm for 10 min and the ahsorbance of the supernatant read at 652 nm (Arnon, 1949; Bruinsma, 1961). Polaroffraphy. Oxygen polarography was used to determine the rates of photo- synthesis and respiration of harvested and unharvested cells. Harvested cells from the tube corresponding to the peak concentration of cells were diluted 1 : 10 or 1 : 5 in artificial sea water or 100 ml of unharvested cells were centrifuged at 2000 rpm for 10 min and resuspended in 3 ml of medium. Oxygen exchange by the algal suspensions in the dark (respiration) and in the light (net photosynthesis) were measured with the Model 53 Biological Oxygen Monitor (Yellow Springs Instrument Company, Yellow Springs, Ohio). RESULTS Efficiency of collection The efficiency of collection of algae by the rotor was measured by comparing the concentration of algae in the original suspension with that in the effluent from the rotor. The four organisms studied cover a range of sizes that include most planktonic forms. The clean-out at any given speed is a function of the size of the alga (Table II). On this basis we have projected the flow rates required to yield > 90% clean-out. For the larger algae at 1500 rpm and 200 ml/min, clean-out was essentially quantitative, and flow rates can probably be increased severalfold without significant decrease in clean-out. For Synechococcus, which is of bacterial size, the clean-out was no better than 78% at maximum rotor speed and at the lower flow rate. Syncchococcns may be at the lower limit of particle TABLE 1 1 Clean-out of algal suspensions in the CF-6. Fractional "clean-out" is a measure of the efficiency of retention of particles in a continuous-flow rotor and defined as roriginal concentration — effluent concentration~\ L original concentration J The coefficient of variation of clean-out varied from 0.2 to 8% Cell number/ml Organism Flow-rate Fractional clean-out ml/min Initial Effluent Dun 6.5 X 10 /u 200 137,767 968 0.99 Actin 10-20 M 200 50,160 111 0.99 Pyr-I 8 X 15 M 200 71,585 269 0.99 Syn* 0.75 X 1.5 M 100 8,430 1818 0.78 150 8,430 1790 0.78 200 8,430 3115 0.63 Determined at rotor speed of 6000 rprn. All others at 1500 rpm. 140 PRICE ET AL. sample 2596 Ludox 5096 75% 100% Dun Actin Pyr-l Syn FIGURE 1. Diagram of isopycnic banding of different algae in step gradients of Ludox. an be separated tinder the present conditions. Very few algal species 1 as 5". bacillaris. size that can are as small as 5". bacillaris. Equilibrium densities The banding densities were initially estimated in step gradients of silica in swinging buckets. After centriftigation at 5000 rpm for 5 min, the banding pat- terns sketched in Figure 1 were obtained. These measurements showed that Dun, Actin and Pyr-I banded at approximately the same densities, while Syn banded at a much higher density. More precise distributions were determined from direct measurements in the continuous gradients recovered from the CF-6. The volumes in which the cells were recovered were in the order of 100 ml (Fig. 2 a-d), but substantial hetero- geneity in density was sometimes observed. Dun banded between p -- 1.07-1.12, but principally near 1.08. Actin banded between 1.09-1.12. Integrity Any method of harvest is of limited usefulness if the cells cannot be recovered physiologically intact and viable. We tested physiological state by polarographic measurements of respiration and photosynthesis, before and after harvest, for some of the algae. A summary of the data is presented in Table III. In the case of Dun, there was no loss of photosynthetic or respiratory activity after banding in Ludox gradients. For the other species, both the rates of photosynthesis and respiration were decreased to 1/3 to 1/5 of the original values after the cells were harvested from the Ludox gradients. To determine whether Ludox itself affected photosynthesis or respira- tion, separate experiments were run whereby polarography was measured on cells in the absence or presence of the silica sol (—66% Ludox) using the organisms Dun and Syn. In both cases there was an enhancement (doubling) of respiration in the presence of Ludox. However, photosynthesis was again lowered to about 1/4 in the case of Syn and essentially unchanged in the case of Dun, in the pres- ence of Ludox. There have been reports that Ludox catalyzes certain biological PLANKTONIC ALGAL HARVESTING 141 0.3 0.2 0. CD E Q. o l_ a n O 0.0 .0 0.5 0.0 Syn Actin Dun Pyr-I 0 1.0 O) E CL O O 0.5 0.0 10 15 20 10 15 20 Fraction number FIGURE 2. Recovery of algae after continuous sample flow with isopycnic banding in gradients of Ludox. Details of the method are described in the text. Cell concentration was estimated by monitoring the absorbance of the fractions at 652 nm ; Synechococcus bacillaris (Syn); Dunaliella tertiolecta (Dun.); Thalassiosira fluviatilis (Actin); Pyramimonas sp. (Pyr-I). 142 PRICE ET AL. TABLE 111 Respiration and photosynthesis of algae before and after centrifugation through Ludox gradients Rates of Oa exchange nmoles O-2 hi— i- fig chlorophyll-i Sample Dark Light Pyr-I (83/52) Original - 61.5 + 21 Ludox-harvested - 23.2 + 7.£ Syn (83/54) Original -229 +285 Ludox-harvested - 89 + 52 Dun (83/57-SB) Original 12.6 +136 Ludox-harvested 12.6 +210 oxidations (Slawson, Adamson and Mead, 1973). This may account for the observed increase in Oo uptake by the cells in the presence of concentrated amounts of Ludox. Viability was measured by inoculating fresh media with approximately equal numbers of cells before and after harvest. We measured growth from the rate of increase of chlorophyll ; although this method will not necessarily provide a measure of absolute growth rates, the existence of an abnormal lag phase will indicate if there are significant numbers of cells which are nonliving or whose growth rates have been rendered abnormal. Growth appeared to proceed normally except in the case of Syn, which showed a significant lag period (Fig. 3 a-d). DISCUSSION We have shown that several kinds of planktonic algae can be harvested from laboratory cultures by continuous-flow centrifugation into gradients of silica. The recovered cells are not seriously damaged in the process, as shown by their essenti- ally normal rates of growth in fresh medium. The rates of sample flow are such that 50 liters of culture of the smallest algal cells could be harvested in a four- hour interval. The absolute quantities are limited only by the capacity of the gradient, which is of the order of tens of grams of cells. Density gradient centrifugation is performed most often using sucrose as the gradient material. However, sucrose has the distinct disadvantage that the high osmotic potential of sucrose solutions will plasmolyze whole cells. Plasmolysis. in addition to the danger of working irreversible changes on the cells, has the effect of increasing the cell density. In fact, we found that none of our organisms could be floated in any concentration of sucrose-seawater up to 65 % w/w nor in sorbitol- sea water up to 60% w/w. Ficoll-seawater did provide a medium in which the cells could float and remain approximately normal in size and shape. We found that algae required up to 25% PLANKTONIC ALGAL HARVESTING 143 w/w Ficoll in seawater, a solution that is extremely viscous and difficult to pump in and out of the rotors ; in addition it is expensive. Ludox, a silica sol, is non-osmotic like Ficoll, but poses neither the problem of viscosity nor expense. However, it has some peculiarities of its own : It has a tendency to gel in the presence of salts. We found that low concentrations of Ludox were stable in 3% w/v NaCl, but higher concentrations were not. We were obliged therefore to employ Ludox gradients in which the dense end of the gradient is deficient in salt with respect to ordinary seawater. For stenohaline organisms 100 1000 - 300 100 8 120 4 Time (days) 8 12 FIGURE 3. Comparison of growth rates of algae before and after harvest by centrifuga- tion in Ludox gradients. Cell densities were estimated by the relative concentration of chlorophyll as determined by fluorescence. Similar numbers of cells were inoculated into fresh media at zero time. The absence of an abnormal lag period was taken as an indica- tion of viability; Synechococcus bacillaris (Syn); Dnnaliella tertiolecta (Dun); Thalassiosira fluviatilis (Actin) ; Pyramimonas sp. (Pyr-I). 144 PRICE ET AL. the deficit in osmotic potential can be compensated with sorhitol or other com- patible osmotica. We have also found some evidence of toxicity of Ludox. The movements of flagellates, such as Dnnaliella, ceased abruptly when ex- posed to even a low concentration of ordinary Ludox. We then found that this toxicity is removed by dialysis, but prolonged dialysis may substantially dilute the preparation. Silica sols free of added biocides have recently become available from Nalco Chemical Company, Chicago, Illinois. In some cases biological activity may be preserved by the addition of small amounts of polymers such as polyethylene glycol (Pertoft, 1969; Morgenthaler ct al, 1974). The lowered rates of respiration and photosynthesis obtained after the organisms were harvested from Ludox gradients may, as in the case of Pyr-I, reflect short- term physiological effects, since growth appeared to proceed normally or after a brief lag period. Of the organisms tried here, Dun seemed to be the least affected by centrifugation in gradients of Ludox, whereas Syn manifested physiological damage, as shown by the increased lag period as well as decreased rates in respira- tion and photosynthesis. Since Dun occurs naturally in brackish waters, its insen- sitivity to Ludox gradients may be related to its tolerance of low7 concentrations of salt. SUMMARY 1. Cells of four different phytoplankton species were harvested from cultures using gradients of colloidal silica (Ludox) in the CF-6, a continuous-flow zonal rotor. 2. The efficiency of harvesting (''fractional clean-out") was > 0.99 for three of the algae and > 0.78 for the smallest species (Synechococciis bacillaris). 3. Growth rates of the algae subcultured after harvesting in Ludox were similar to those unharvested controls, but SynecJiococcus showed a lag period of about four days after harvesting. 4. The rates of photosynthesis and respiration of cells harvested in Ludox gradients were, in most cases, about ^ to ^ of the pre-harvest rates. Dnnaliella tertiolecta showed no loss of photosynthetic or respiratory activity. 5. This method has application to the collection of algae from dilute cultures and from natural waters when cells are needed in good physiological condition. LITERATURE CITED ARNON, D. I., 1949. Copper enzymes in isolated chloroplasts. Polyphenol oxidase in Beta wilgaris. Plant Physio!., 24: 1-15. BRUINSMA, J., 1961. A comment on the spectrophotometric determination of chlorophyll. Bio- chim. Biophys. Acta, 52: 576-578. CASCIATO, R., 1973. Developments in continuous sample-flow zonal rotors. American Labora- tory, 5(4) : 35-44. CLARKE, W. J., AND W. F. SIGLER, 1963. A method of concentrating phytoplankton samples using membrane niters. Limnol. Occanogr., 8: 127-129. CLINE, G. B., AND R. B. RYEL, 1971. Zonal centrifugation. P~ges 168-204 in S. P. Colo\vick and N. O. Kaplan, Eds., Methods in Ensymology, Volume 22. Academic Press, N. Y. DODSON, A. N., AND W. H. THOMAS, 1964. Concentrating plankton in a gentle fashion. Limnol. Oceangr., 9 : 455-456. GUII.LARD, R. R. L., AND J. H. RvTHER, 1962. Studies on marine planktonic diatoms. I. Cyclotella nana, Hustedt, and Detonula conjervacea (cleve) Gran. Can. J. Microbiol., 8 ^229-239. PLANKTONIC ALGAL HARVESTING 145 HOLM-HANSEN, O., T. T. PACKARD AND L. R. POMEROY, 1970. Efficiency of the reverse flow filter technique for concentration of particulate matter. Linmol. Oceanogr., 13 : 350-355. KNIGHT, G. J., 1968. Biophysical parameters of organelle development : S/p coordinates of developing chloroplasts. Ph.D. thesis, Rutgers University, New Brunswick, New Jersey. LAMMERS, W. T., 1971. Insoluble material in natural water. Pages 593-638 in L. L. Ciaccio, Ed., Water and Water Pollution Handbook, Volume 2. Marcel Dekker, New York. MATEYKO, G. M., AND M. J. KOPAC, 1963. Cytophysical studies on living normal and neo- plastic cells. Part II. Isopycnotic cushioning during high speed centrifugation. Ann. New York Ac ad. ScL, 105(4) : 219-253. McLACHLAN, J., 1964. Some consideration of the growth of marine algae in artificial media. Can. J. Micobiol, 10 : 764-782. MORGEN THALER, J.-J., C. A. PRICE, J. M. ROBINSON AND M. GIBBS, 1974. Photosynthetic activ- ity of spinach chloroplasts after isopycnic centrifugation in gradients of silicia. Plant Physiol., in press. MORRIS, I., AND C. S. YENTCH, 1972. A new method for concentrating phytoplankton by filtration with continuous stirring. Linmol. Oceanogr., 17: 490-495. PERTOFT, H., 1969. The separation of rat liver cells in colloidal silica-polyethylene glycol gradients. Exp. Cell Res,, 57 : 338-350. SLAWSON, V., A. W. ADAMSON AND J. F. MEAD, 1973. Autoxidation of polyunsautrated fatty esters on silica. Lipids, 8(3) : 129-134. Reference: Biol. Bull., 147: 146-162. (August, ll>74) GLUCOSE AND SODIUM FLUXES ACROSS THE BRUSH BORDER OF HYMENOLEPIS DIMINUTA (CESTODA)1 CLARK P. READ - GEORGE L. STEWART, AND PETER W. PAPPAS 3 Department of Biology, Rice University, Houston, Texas 77001 Tapeworms are a favorable material for studying the transport of organic solutes across membranes. The outer surface of adult tapeworms is comprised of a coenocytic epithelium with a highly developed brush border (Lumsden, 1966). Since tapeworms have no digestive tract, nutrients from the external medium must cross this external tegument. In addition, the intact tapeworms can be easily and rapidly maniupulated, making them ideal subjects for the study of transport kinetics. Previous studies have shown that glucose transport in several species of tape- worms is Na+-dependent (von Brand and Gibbs, 1966 ; Fisher and Read, 1971 ; Dike and Read, 1971a; Pappas and Read, 1972a; Pappas, Uglem and Read, 1973a). The present study was undertaken to characterize in some detail the relationship of glucose and sodium fluxes in the rat tapeworm, Hyinenolcpis dhnimita. MATERIALS AND METHODS Hymenolepis diminitta was maintained in the beetle, Tenebrio niolitor, and male Sprague Dawley rats (Holtzman Co.) Rats weighing 60-80 g were infected with 30 cysticerocoids (H. dimimtta) and worms were recovered 10 days post-infection. Worms from several rats were pooled and randomized into groups of 5, and pre- incubated for 15 min at 37 C in 15 ml of Krebs-Ringer saline containing 25 HIM tris(hydroxymethyl)aminomethane-maleate buffer at pH 7.4 (KRT of Read, Roth- man and Simmons, 1963), or in KRT with the NaCl replaced isomotically with KC1, LiCl. choline-Cl, or tris(hydroxymethyl)aminomethane-Cl. Other modifications of salines are described in the text. After preincubation worms were blotted on filter paper, placed in 5 ml of KRT (or the appropriate ion-substituted saline) con- taining radioactive substrate, and incubated for 2 min at 37 C. To terminate incuba- tion each worm group was removed and rapidly rinsed in KRT, blotted on filter paper, and extracted overnight in 2 ml of 70% ethanol. Radioactivity in aliquots of the ethanol extracts and incubation media was determined using a gas-flow counter or liquid scintillation spectrometer ; influx rates were calculated by com- paring the radioactivity extracted from worms with the specific activity of the incubation media. The ethanol extracted worms were dried overnight at 95 C and weighed. Uniformly labeled 14C-D-gluc<>se and ~2Na+ (as Nad) were ob- tained from Amersham /Searle Corp. !This study was supported in part hy grants fAI-01384 and 2-TO1-AI-00106) from The N.T.H., U. S. Public Health Service. 2 Deceased, December 24, 1973. 3N.I.H. Postdoctoral Fellow, 5-FO2-AI-45 108-02. Present Address: Department of Zoology, The Ohio State University, Columbus, Ohio 43210. 146 NA+ AND GLUCOSE FLUXES IN I IYMENOLEPIS RKSULTS 147 Dike and Read (1971a) showed that the influx of glucose across the brush border of H. diniinuta is sensitive to Na+ in the incubation medium. Our pre- liminary experiments indicated that preincubation of worms in media without Na4 also affects glucose influx. To examine this effect further, worms were pre- incubated for varying time periods in Na+-free KRT with tris as the replacement 8 16 24 Time ( m i n ) 32 FIGURE 1. The effect of perincubation time in Na+-free KRT on glucose influx (J'o = Aimoles glucose absorbed/g ethanol extracted dry \vt/2 min) in Hymenolepis diniinuta. Worms were preincubated in Na^-free KRT for predetermined time periods (abscissa), and then trans- ferred to Na+-free media containing 1 HIM "C-glucose for 2 min. In this experiment, Na+ was replaced with tris. Each point is the mean of 3 replicates. cation, and then incubated for 2 min in Na+-free KRT containing 1 mM 14C-glucose. The results shown in Figure 1 demonstrated that a 12 min preincubation of worms in Na+-free KRT was sufficient to inhibit glucose influx (JV.) 96%. To determine the reversibility of the effect of Na+ deletion on glucose influx, worms were preincubated in Na+-free media for 30 min. incubated in KRT for varying time periods, and subsequently incubated for 2 min in KRT containing 1 HIM 14C-glucose. The results shown in Figure 2 demonstrated complete reversal of the effect of Na+ deletion. In all subsequent experiments, worms were pre- 148 READ, STEWART AND PAPPAS 14 10 12 17 22 Time (min) 27 32 FIGURE 2. Reversal of the effects of Na+ deletion on the influx of glucose (J'o, as in Fig. 1) in Hymenolepis diminittu. Worms were preincubated in Na+-free KRT (tris as the replacement cation) for 30 min, followed by incubation in KRT ([Na+] = 154 meq/1) for pre- determined time periods (abscissa), and subsequently incubated in 1 niM 14C-glucose in KRT for 2 min. The single square point represents glucose influx in a control group which was not incubated in Na+-free KRT. Each point is the mean of 3 replicates. incubated for 15 min in media containing the same cation concentrations in the final incubation medium. It seemed desirable to examine the effects of various other cations as replace- ments for Na+ in the medium. The effect on glucose influx of totally replacing Na+ with Li+, K+, choline, or tris is shown in Table I. The much lowered influxes in media where K+, tris, or choline were used as replacements for Na+ were similar. However, glucose influx with Li+ as the replacement cation was significantly higher (P < 0.05 by Student's t test) than influxes obtained with the other substituting TABLE I A comparison of 0.5 mu glucose influx (Jo1 = nmoles glucose absorbed/ g ethanol extracted dry wt/2 min) in Hymenolepis diminuta in salines with the Na+ totally replaced with various cations and KRT. Each value is the mean ± S.E. of 3 replicates Replacement Cation Jo' Jo1 in KRT K+ Choline Tris 0.086 ± 0.005 0.076 ± 0.006 0.031 ± 0.004 0.56 ± 0.03 5.66 ± 0.32 4.96 ± 0.28 5.28 ± 0.31 4.58 ± 0.22 NA+ AND GLUCOSE FLUXES IN HYMENOLEPIS 149 cations, suggesting that Li+ may partially replace Na+. This "replacement" effect disappeared when increasing amounts of Na+ were added to media containing Li+. Glucose influx was a hyperbolic function of Na+ concentration regardless of the replacement cation (Fig. 3). Although the data did not indicate that K+ competes with Na+ in glucose influx, this possibility was examined further by determining the effects of K+ concentra- tions on glucose influx at a constant glucose concentration (0.5 HIM) and a sub- optimal, fixed Na+ concentration (25 meq/1) (isosmolarity was maintained by add- 6r 40 80 120 [Na^mE/L 160 FIGURE 3. The influx of 0.5 HIM 14C-glucose (J'c as in Fig. 1) in Hyincnolcpis diininittu as a function of the Na+ concentration ([Na+]) in media with the Na+ replaced with various cations. Replacement cations were K+ (solid circles), Li+ (open circles), tris (open squares), or choline (solid triangles). Each point is the mean of 3 replicates. ing an appropriate amount of tris-Cl). There was no effect of K+ on glucose influx at K+ concentrations ranging from 0 to 100 meq/1. Glucose influx as a function of glucose concentration followed apparent Michaelis-Menten kinetics in media with fixed Na+ concentrations of 154, 50, 25, or 10 meq/1 (Fig. 4). The apparent transport constant (Kt) for glucose influx was affected only slightly with decreasing Na+ concentrations, but there was a marked decrease in the maximal influx of glucose (Ticmax) with decreasing Na+ concentration (Table II). When glucose influx was tested as a function of Na+ concentration at several fixed glucose concentrations, it was apparent that glucose influx was a hyperbolic 150 READ, STEWART AND PAPPAS function of Na+ concentration, regardless of the glucose concentration. The data also showed that the Na+ concentration necessary to achieve JiGmax/2 changed significantly at different glucose concentrations (Fig. 5, Table II). The kinetic parameters describing these glucose-Na+ interactions are summarized in Table II. While the above experiments demonstrated the dependence of glucose influx on Na+, and suggested coupling of glucose and Na+ fluxes, more direct evidence 30 20 10 48 [GLUCOSE] 12 15 20 FIGURE 4. The influx of uC-glucose (J'c as in Fig. 1) in Hymcnolcpis diminuta as a function of glucose concentration in media with different fixed Na+ concentrations. Na+ con- centrations of 154 meq/1 (solid triangles), 50 meq/1 (solid squares), 25 meq/1 (open circles), and 10 meq/1 (solid circles) were used. In all experiments, deleted Na+ was replaced with tris. Each point is the mean of 3 replicates. of coupling required a demonstration that Xa+ influx (JjNa) was related to the presence of glucose in the external medium. As shown in Table III, a very large Na+ influx was associated with glucose influx. Before carrying out experiments designed to evaluate coupling coefficients (pNa/ J1,;), it was necessary to determine that the influxes of 14C-glucose and 22Na+ fol- lowed first order kinetics for the time period employed. Groups of worms were incubated in solutions containing either 25 meq/1 2-'Na+ and 5 mM glucose, or 25 meq/1 Na+ and 5 mat 14C-glucose. Samples were removed at 15 sec intervals, washed for approximately 2 sec in KRT and extracted in 70% ethanol. The influxes of Na+ and glucose were a linear function of time for at least 2 min (Fig. 6). To determine coupling cofificients, using labeled Na+ and labeled glucose, incuba- tions were conducted for 2 min with less than a 2 sec postincubation rinse in KRT. NA* AND GLUCOSE FLUXES IN HYMENOLEPIS 151 The results of such an experiment are shown in Figure 7. In this experiment groups of worms were incubated in KRT containing 25 meq/1 22Na+ and varying glucose concentrations, or 25 meq/1 Na+ with varying concentrations of 14C-glucose. The plot of Na+ influx versus glucose influx has a slope of 1.6 by regression anal- ysis, indicating the coupling coefficient was at least 2. The intercept of the line in Figure 7 represents Na+ influx not associated with glucose influx. As expected, the glucose-coupled Na+ influx was a hyperbolic function of glucose concentration (Fig. 8). It should be noted that the concentration of Na+ yielding a half-maxi- mum rate of Na+ influx in the presence of 5 HIM glucose was about 154 meq/1. In a different type of experiment, the glucose concentration was held constant and TABLE II A summary of the kinetic parameters describing uC-glucose influx in Hymenolepis diminuta us. functions of Na+ and glucose concentrations (meq/1 and W/M, respectively). The graphic analysis of V vs. F/rjSJ (Dixon and Webb, 1964), with calculated regression lines, was used to deter- mine these parameters Summary of Figure 4 [Na+] Ja! m«* Kt** 154 26.3 1.4 50 17.1 1.0 25 14.7 0.9 10 1.1 0.8 Summary of Figure 5 [Glucose] Jo' mal Ktt 5 28.0 56 1 10.1 43 0.5 6.8 40 0.1 1.7 34 * maximal glucose influx = jumoles absorbed/g ethanol extracted dry wt/2 min. ** that [[glucose ~] necessary to achieve 5 JG' max. t that [Na+] necessary to achieve £ Jo1 max. the Na+ concentration varied. Results of such an experiment are presented in Table IV. When 22Na+ influx in the presence of glucose was corrected for the amount of Na+ entering the worms in the absence of glucose, the apparent coupling coefficient varied as an inverse function of the Na+ concentration. That is, with decreasing Na+ concentrations the coupling coefficient increased. Thus, the coupling coefficient appeared independent of the glucose concentration over a 50-fold range, but inversely dependent on the Na+ concentration over a 20-fold range. When 22Na+ influx was measured as a function of Na+ concentration in the absence of glucose, it was found to be a nonlinear function of Na+ concentration, This suggested that 22Na+ influx in the absence of glucose did not occur by simple diffiusion. This was examined further by adding varying quantities of unlabeled Na+ in the presence of a fixed quantity of 22Na+. If Na+ influx was by diffusion alone, the addition of unlabeled Na+ should not have affected the influx of 22Na+. 152 READ, STEWART AND PAPPAS 40 60 80 100 160 FIGURE 5. Influx of "C-glucose (JV, as in Fig. 1) in Hymenolepis dimwit ta as a function of Na+ concentration in media with different glucose concentrations. Glucose concentrations of 5 HIM (solid circles), 1 HIM (open circles), 0.5 mM (solid squares), and 0.1 HIM (solid triangles) were used. In all experiments, deleted Na+ was replaced with tris. Each point is the mean of three replicates. Such was not the case (Fig. 9). The data indicated that in the absence of glucose about 60% of the 2L'Na+ influx at this concentration (10 meq/1) occurred by a mediated process, and about 40% by diffusion. TABLE 1 1 1 Influx of nNa+ (Jiva1 = ^equivalents absorbed/g ethanol extracted dry wt/2 min) ay a function of the Na+ concentration ([Na+'], meq/1) in Hymenolepis diminuta in the presence or absence of 5 OTM glucose. Each value is the mean ± S.E. of 3 replicates [Na+] (with glucose) (without glucose) 5 15 25 50 100 3.07 ± 0.28 8.78 ± 0.51 14.33 ± 1.21 26.33 ±2.11 36.52 ± 2.93 0.7 ±0.1 0.9 ± 0.1 1.38 ± 0.01 1.64 ± 0.02 3.22 ± 0.02 NA+ AND GLUCOSE FLUXES IN HYMENOLEPIS 153 In the absence of external Na+, glucose influx in H. diminuta as a function of glucose concentration appeared saturable, with an approximate pGmax and Kt of 0.15 /Ainoles/g ethanol extracted dry wt/2 min and 0.5 mM, respectively. The inhibition of 0.1 mM 14C-glucose influx by unlabeled glucose indicated that at least of the glucose influx in Na+-free media was mediated. 8r 0 30 60 90 Time (sec) 120 FIGURE 6. Fluxes of 5 HIM 14C-glucose (J'G = Atmoles absorbed/g ethanol extracted dry wt, left ordinate) and 25 meq/1 ~Na (J'sa — /^equivalents absorbed/g ethanol extracted dry wt, right ordinate) into Hymenolepis dimimtta as a function of time; J'G = solid circles, J'xa = solid squares. Delected Na+ was replaced with tris. Each point is the mean of 3 replicates. The above observations suggested that further studies of the mediated influx of 14C-glucose in the absence of Na+, and of the mediated portion of 22Na+ movement in the absence of glucose, were desirable. The glycoside phlorizin was a competitive inhibitor of 14C-glucose influx in H. diminuta (Fig. 10). In the absence of external Na+, phlorizin abolished over 70% of the residual 14-C- glucose influx suggesting that the same glucose transport system was involved in the pres- ence and absence of external Na+. On the other hand, phlorizin had no effect on 154 READ, STEWART AND PAPPAS the influx of "Na+ in the absence of glucose, suggesting the possibility of mediated Na+ influx which was independent of the glucose transport system. DISCUSSION The data of the present paper are consistent with the basic elements of Crane's (1962, 1965) Na+-gradient hypothesis for the transport and accumulation of sugars (see also Schultz and Curran, 1970). Previous studies (Pappas, Uglem and Read, 1974) have shown that glucose is indeed accumulated by H. diminuta, and cotrans- 15 12 J' 9 Na ' FIGURE 7. The relationship of 14C-glucose influx (J'o as in Fig. 1) and "Na+ influx (J*Na — /^equivalents absorbed/g ethanol extracted dry wt/2 min) in Hymcnolcpis diiniinttti when the Na+ concentration was 25 meq/1, and the glucose concentration was 0, 0.1, 0.25, 0.5, 1 and 5 HIM. Deleted Na+ was replaced with tris. Each point is the mean of 3 replicates. port of glucose and Na+ across the brush border of H. ciiininiita is demonstrated unequivocally in the experiments described herein. When Na+ is deleted from the medium, K+, choline, or tris do not substitute for Na+ in enhancing glucose influx, while Li+ appears to substitute to a small extent. The effects of Li+ disappear with increasing Na+ concentrations. Similar effects of Li+ on glucose influx in the tapeworm Calliobothrium verticillatum were reported by Pappas and Read (1972a). Pappas et al. (1973a) found that K+ com- petes with Na+ in the glucose transport system of the larval tapeworm Taenia crassi- ceps, and a similar interaction between K+ and Na+ has been observed in the tetra- NA' AND GLUCOSE FLUXES IN HYMENOLEPIS 155 14 10 Na 2 3 [GLUCOSE] mM FIGURE 8. Influx of 25 meq/1 "Na+ (J'xa as in Fig. 7) in Hyinciiolepis diminuta as a function of glucose concentration. All data were corrected for that Na+ movement which occurs in the absence of glucose. Each point is the mean of 3 replicates, and deleted Na+ was re- placed with tris. phyllidean C. vcrticillatum by Fisher and Read (1971) and Pappas and Read (1972a). Competition between Na+ and K+ in Na+-dependent transport systems in mammalian tissues is suggested by the data of Nathans, Tapley and Ross ( 1960) , Kipnis and Parrish (1965), and Crane (1965). In contrast to the above systems, K+ does not appear to interact with the Na+-coupled glucose transport system of H. diminuta. Further, there is no evidence that glucose transport in H . diminuta is inhibited in K+-free media, as has been observed in some mammalian tissues (Riklis and Quastel, 1958; Bihler and Crane, 1962). When the external Na+ TABLE IV The coupling coefficient (Jxa^/Jo') for Na+-couf>led glucose influx in Hymenolepis diminuta a.v a function of Na+ concentration of the external medium ([_Na+~], meq/l). In each experiment the Na+ concentration was held constant and the glucose concentration varied from 0.1 to 5 mM. Cou- pling coefficients are calculated slopes of plots of JN100 >100 1.33 ± 0.06 * Time required for retraction of anterior segments after injection as % of time required for Ringer's-injected control. ** Time required for tanning of whole larvae after injection as % of time required for Ringer's-injected control. *** Quotient of the pretanning period of the posterior part of ligated larvae over that of the anterior part. The same notations are used in the subsequent tables. The cytosol fraction of the CNS was injected into intact and ligated hind parts of red-spiracle larvae (Table II). Injection of 10 fig of cytosol protein per host accelerated tanning to such an extent that intact hosts were still crawling when tanning had already set in. The effect of this quantity of cytosol on retraction was relatively small. Thus, it appears that CNS extract has a greater effect on tanning than on retraction of the anterior segments whereas the opposite is true of hemolymph. Precipitation of active factors with ammonium sulfate The hemolymph proteins precipitated with various concentrations of ammonium sulfate were assayed for ARF and PTF activities. The 45-70% ammonium TABLE II Effect of injection of various quantities of cytosol fraction of CNS homogenate into red-spiracle larvae of S. bullata Quantity of cytosol protein of CNS injected (jug) Retraction % Control Tanning Control P/A 0 (Ringer sol.) 10 100 86 ± 5.0 ,100 74 ± 3.6 1.28 ± 0.01 0.73'± 0.05 5 95 ± 2.1 79 ±4.1 0.80 ± 0.0 3 >100 85 ± 1.61 0.91 ± 0.06 2 >100 >100 1.02 ± 0.03 1.5 >100 >100 1.10 ±0.005 0.75 >100 >100 1.18 ± 0.02 168 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL sulfate fraction retained about 75% of the activity of both factors and, most im- portantly, did not darken and subsequently become toxic to the test animals, so this material was used for further fractionation. Ninety ,ug of protein from a 45-70% ammonium sulfate fraction had to be injected to get ARF activity equiv- alent to 1 unit, and 4 times more protein had to be injected to get a similar amount of PTF activity. Ammonium sulfate precipitation of the cytosol fraction of the CNS was unsuc- cessful and very little PTF and ARF activities were retained in a 0-70% ammonium sulfate precipitate. Therefore, attempts to precipiate the activity were discontinued, and the original cytosol fraction was used for further fractionations. Ultrafiltration The 45-70% ammonium sulfate fraction of hemolymph was filtered through Amicon permselective membranes and tested for activity. The ARF activity passed through an XM-300 filter but was retained by an XM-100 filter indicating that the molecular weight of ARF was between 100,000 and 300,000. The frac- tion retained by XM-300 showed all of the tanning (PTF) activity. The cytosol fraction of the CNS was filtered through Amicon filters and the fractions retained by the filters were assayed for activity. The fraction retained by an XM-300 filter" (molecular weight above 300,000) retained all the PTF activity, and also showed slight ARF activity when higher concentrations of the fraction were injected. Taken together the ultrafiltration experiments indicated that the gross molecular weight of ARF of hemolymph is between 100.000 and 300,000; and that of PTF from hemolymph and CNS is above 300,000. Heat treatment The 45-75% ammonium sulfate fraction of the hemolymph from orange puparia was dialyzed overnight and diluted 10 times with Ringer solution. Aliquots were then heated for 3 minutes at 60°, 80° and 100° C, then cooled and centrifuged at 10,000 g for 10 minutes. At 60° C there was no precipitation but the sample turned milky. At 80° C there was slight precipitation and at 100° C most of the proteins had precipitated. After centrifugation the supernatants were concentrated by filtering through an XM-50 filter and asasyed for ARF and PTF activities. The results indicated that ARF was destroyed by heating at 100° C whereas PTF was not. Heating at 100° C for 60 minutes still gave considerable PTF activity upon injection of 5 /xg per larva. Since the ultrafiltration studies showed that the fraction retained by the XM-300 filter had tanning activity and heating experiments revealed that PTF was not de- stroyed by heat treatment, the question arose as to whether heating had any effect on the size of the molecules which were originally quite large. As revealed in Table III, a high PTF activity is present in the unheated material as well as in the supernatant of the preparation heated to 100° C. In the unheated material PTF activity was present only in the fraction retained by XM-300 whereas after heating to 100° C almost all the activity was present in the fraction which passed through XM-100 but was retained by the XM-50 ultrafilter. HORMONAL CONTROL OF PUPARIATION 169 TABLE III Comparison of the effect of injection of the active ammonium sulfate (45-70%) fraction of hemolymph before and after heat-treatment at 100° Cfor 3 minutes. The fractions were filtered through Amicon ultrafilters before assaying for tanning activity. The figures in parentheses are quantities of protein (/*g) injected per host. The figures are means from 20 observations P/A* Molecular weight'of the fraction Unheated. 45-70% fraction Supernatant of heated 45-70% fraction Unfractionated 0.86 (600) 0.71 (30) >300,000 daltons 0.88 (60) 1.0 (30) 100,000 to 300,000 daltons 1.00 (60) 1.16 (30) 50,000 to 100,000 daltons 1.21 (60) 0.76 (30) Less than 50,000 daltons 1.19 (60) 1.22 (30) * Ringer injected controls had a P/A ratio of 1.26. Ill order to see whether the tanning activity is stable to heating, aliquots of the cytosol fraction of the CNS homogenate were heated for 3 minutes at 60° C, 80° C and 100° C. The fractions turned milky but there was no precipitation. The entire milky solution was assayed for PTF activity and the results clearly demonstrated that tanning activity was not lost after heating to 100° C for 3 minutes. There also remained an apparent, albeit very slight acceleration of retraction of the anterior segments after heating to 100° C. This will be taken up in some later experiments (Table IV). To find out whether heating broke down the cytosol PTF molecules, the heat treated fractions were filtered through XM-300 ultrafilters and their protein content determined. There was no change in the protein content of the fraction retained by XM-300. Enzyme hydrolysis Treatment of the 45—70% ammonium sulfate fraction of hemolymph with trypsin and pepsin for 1-2 hours at 37° C revealed that both ARF and PTF are inactivated by these proteolytic enzymes. Fraenkel ct al. (1972) reported that PTF activity of CNS was somewhat resistant to pepsin. This was investigated in greater detail with the XM-300 TABLE IV Effect of injection of heat-treated cytosol fractions of CNS into intact and ligated hind parts of red-spiracle larvae of S. bullata Injected fractions Quantity of protein in- jected Gig) Retraction % control Tanning P/A Ringer solution 0 100 1.3 ±0.05 Cytosol, unheated 15 76 ± 2.6 0.66 ± 0.02 Cytosol, heated to 60° C for 3 min. 15 74 ± 3.2 0.73 ± 0.02 Cytosol, heated to 80° C for 3 min. 15 91 ± 2.9 0.70 ± 0.09 Cytosol, heated to 100° C for 3 min. 15 93 ± 1.6 0.79 ± 0.00 170 S1VASUBRAMANIAN, FRIEDMAN AND FRAENKEL retained fraction of CNS cytosol, and the previous observation confirmed. How- ever, ultrafiltration of the pepsin treated fraction through XM-300 membranes revealed that digestion was incomplete over the time of the experiment, and that about half of the material with a molecular weight of above 300,000 had not been cleaved. Hence the apparent PTF activity after enzyme activity could be due to the presence of undigested molecules. These data on the effect of proteolytic enzymes indicate that both ARF and PTF activities are destroyed after enzyme hydrolysis, which is one more evidence of their protein nature. Gel filtration of hemolymph fraction Since the results of the ultrafiltration experiments indicated that the approxi- mate molecular weight range of the ARF active material was between 100,000 and 300,000. Bio-Gel P-300 was chosen for gel filtration. The 45-70% ammonium sulfate fraction was redissolved in Ringer solution and about 250 mg protein applied to a column with a bed volume of about 75 ml. One milliliter fractions were col- lected at a rate of 3 ml per hour. The protein content of each fraction after the void volume (27 ml) was deter- mined and every third sample was tested for ARF activity by injecting a volume equivalent to 15 ^.g of protein into each red-spiracle larva. The fractions eluted between 41 ml and 47 ml showed high ARF and tanning activity. These were pooled, concentrated by filtering through an XM-100 filter and then used for electrophoresis. From a plot of molecular weight as a function of elution volume using standard proteins with known molecular weights such as catalase (240,000), lactic dehydrogenase (140,000) and alkaline phosphatase (80,000) (Fig. 1) the molecular weight range of the active fractions was found to be be- tween 180,000 and 220,000. Electrophoresis About 500 /Ag of hemolymph protein from the ARF active fractions after gel filtration was layered on each gel tube and electrophoresed till the marker dye reached about 0.5 cm from the bottom of the gel. The gels were then fixed, stained, and examined. No difference could be seen between whole hemolymph and gel filtration fractions with regard to the number and position of bands. In each case there were about 13 bands (Fig. 2A & B). To examine PTF activity, the 45-70% ammonium sulfate precipitate was heated to 100° C for 3 minutes, the supernatant concentrated by filtering through an XM-50 ultrafilter and the material remaining above the filter electrophoresed. There were only four major bands in the gel (Fig. 2C). The cytosol fraction of the CNS and the fraction that was retained by the XM-300 filter were electrophoresed by the same procedure. In both cases the bands were very faint. There were about 11 bands in the whole cytosol fraction and only 5 in the fraction retained by the XM-300 filter (Fig. 2D & E). The pattern was obviously different from that of the hemolymph fraction. In order to extract the hemolymph proteins from the gel, 7 gels were run simultaneously using the ARF active fractions of gel filtration and at the end of HORMONAL CONTROL OF PUPARIATION 171 electrophoresis the gels were removed from the tubes and one of them was fixed and stained in Coomassie brilliant blue. The other six gels were sliced into two unequal lengths, the upper (cathodal side, slice #11 of 4 cm in length), and lower (anodal side, slice #1 of 3 cm in length), and the slices pooled. Elution and test- ing of the proteins from these pooled fractions showed that the lower part of the gel, i.e., slice #1, was more active than the upper part (Table V). 60 50 40 30 Alkaline phosphatase Lactic dehydrogenase Catalase 10 20 25 Molecular Weight x 10 FIGURE 1. Molecular weight determination based on the flow rate through Bio-Gel P-300 column. Total bed volume was 75 ml. Flow rate was 3 ml per hour. The area in dotted line represents the region where the hemolymph fractions were active for ARF. Knowing that ARF activity was present in the lower part of the gel, other gels were run and sliced into 14 equal parts of 0.5 cm length each (the total length of the gel was usually about 7 cm). Each slice was crushed into smaller bits in 1 ml Ringer solution, extracted overnight at 0-4° C, and the eluants tested as usual. The slices were numbered 1-14 beginning from the anodal end. As shown in Table V a relatively high ARF activity was seen in the extracts of slice #5. The next step was to electroelute the proteins directly from the gel. Since the approximate location of the activity was known (slice #5, Table V), gels were electrophoresed for various periods of time after the dye front emigrated out of the gel and examined by staining procedures. It was observed that by about 172 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL R .25 .50 .75 .00 ARF- - ARF- - RTF PTF- PTF- B FIGURE 2. Electrophoresis of active fractions from hemolymph and CNS on 7% acrylamide gels. Arrows indicate the active regions ; A., Whole hemolymph ; B., Bio-gel P-300 active fraction of hemolymph; C, PTF-active supernatant of heat treated ammonium sulfate (45- 70%) fractions of hemolymph; D., Cytosol fraction of CNS; E., Cytosol fraction of CNS retained by XM-300 filter. TABLE V Effect of injection of extracts obtained from electrophoresis of an active ARF fraction of hemolymph The results have been averaged from 20 observations. The hemolvmph was processed in a P-300 column and the active fractions used for gel electrophoresis Gel slice number* Amount of protein injected (/ig) % Control time Retraction Tanning I (Lower) II (Upper) 10 10 40.5 75.6 66 82 1 3.5 100 100 2 5 100 100 3 5 100 100 4 13.5 88 91 5 5.5 81 79 6 5.5 95 94 7 12.5 95 97 8 6.5 100 100 9 6.0 96 92 10 3.5 96 94 11 5.5 90 89 12 7.0 100 100 13 9.0 100 100 14 7.0 100 100 * See text. HORMONAL CONTROL OF PUPARIATION 173 TABLE VI Summary of purified /inn steps of anterior segments retraction factor (ARF) from one ml of hemolymph (retraction in 50% of control time equals one Sarcophaga unit for ARF). Fractions Total protein Total number of Specific activity Purification f«1H Yield (mg) ARF units (units/mg) (%) Hemolymph 205 1990 9.7 1 100 45-70% ammonium sulfate precipitate 125 1388 11.1 1.13 69.7 Bio-Gel P-300 fractions 20 1333 66.6 6.86 67 Electro-eluted fractions 0.56 933 1666 171.7 46.9 3^ hours after the start the active region had migrated to the lowest part of the gel. Therefore, collection of fractions was begun about 3] hours after the start of electrophoresis. Ten fractions, of 0.7 ml volume, were collected at intervals of 5 minutes and each was assayed for ARF activity. Fraction #3 collected exactly 3^ hours after the start of electrophoresis seemed to be most active. However, injection of even 6 /*g (10 units) of this protein did not accelerate tanning in the ligated hosts, indicating a complete absence of PTF. A summary of the purification steps of the hemolymph retraction factor is given in Table VI. It has been purified about 170 times. The PTF containing fraction of hemolymph (both the supernatant of 100° C-heated 45-70% ammonium sulfate fraction, and the unheated 45-70% ammonium sulfate fraction retained by the XM-300 filter) were electrophoresed and the gels sliced into 6 equal parts. The proteins were eluted and the concentrated eluants were assayed for PTF activity. In both cases, the eluant from the gel slice closest to the dye front contained the highest PTF activity. As shown in Table VII, the PTF has been purified 200 times. Since there is no gain in purification by ammonium sulfate precipitation it might be advisable in future to omit this procedure and to prevent darkening of the hemolymph by the same heating step used for purification. About 300 /ug of XM-300 retained protein from the CNS cytosol fraction was electrophoresed in 7% acrylamide gels. The gels were sliced and the proteins eluted, re-concentrated by filtering through XM-300 and, finally, tested for activity. The results showed that the eluant from the slice closest to the dye front contained TABLE VII Summary of purification steps of puparium tanning factor (PTF) of 1 ml of hemolymph. A P/A ratio of 1 equals 1 Sarcophaga unit for PTF Fraction Total protein (mg) Total number of PTF units Specific activity Purifica- tion fold Yield (%) Hemolymph 205 685 3.3 1 100 45-70% ammonium sulfate precipitate 125 348 2.8 — 51 Supernatant of 100° C heat treated am- monium sulfate precipitate 1.65 330 200 61 48 Eluted from electrophoresed gel slice 0.3 200 667 202 29.2 174 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL the PTF activity and injection of 0.6 /*g of this protein gave a P/A ratio of 1. The entire series of steps, beginning from the cytosol fraction of the CNS, increased the purity of PTF 5 fold (Table VIII). The electrophoretic runs of the active fractions from hemolymph and the CNS have shown that ARF and PTF are distinct entities which have different mobilities in gels run under conditions specified above. ARF is located in the middle third of the gel, and PTF in the lowest part of the gel. Estimation of molecular weight of the active fractions The molecular weight of the most active electroeluted ARF fraction of hemo- lymph was determined by SDS-gel electrophoresis. Five separate runs of the sample gave a single band with an average mobility of 0.37. From the semi-log plot of marker proteins and their mobilities (Fig. 3) the molecular weight of ARF was estimated to be about 90,000. Bio-gel P-300 column chromatography showed TABLE VIII Summary of purification steps of puparium tanning factor from 100 CNS of mature larvae of Sarcophaga bullata P Fraction Total protein (mg) Total number of PTF units Specific activity Purifica- tion fold Yield (%) Cytosol 1.71 570 333 1 100 Cytosol retained by XM-300 0.690 460 667 2 80.6 Eluted from the gel slice 0.120 200 1665 5 35 the molecular weight of the active sample to be between 180,000 and 220,000 (Fig. 1). Since the subunit molecular weight appears to be about 90,000, the molecular weight of the native protein with high ARF activity is probably about 180,000 with 2 subunits of 90,000 each. By the same procedure, the mobility of the gel-slice eluant of hemolymph having high PTF activity was found to be 0.74, which means a molecular weight of the subunit of PTF of about 26,000. Since the native protein had a molecular weight of more than 300,000, the molecule must have at least 12 subunits with a molecular weight of 26,000 each. (Fig. 3). The active PTF fraction of the CNS eluted from the gel slice gave a single band with an average mobility of 0.41. From a semi-log plot of the marker proteins and their mobilities (Fig. 3) the molecular weight of PTF was estimated to be about 80.000. Since the native protein had a molecular weight of more than 300,000, this PTF fraction contains at least 4 subunits with a molecular weight of 80,000 each. Molecular weight determinations by SDS-gel electrophoresis gave different mobilities (Fig. 3) and subunit molecular weights for ARF and PTF. The subunit molecular weight of ARF of hemolymph is about 90,000. The PTF from the CNS has a subunit molecular weight of about 80,000 and that from the hemolymph about 26,000. HORMONAL CONTROL OF PUPARIATION 175 Is there ARF activity in the CNS? The ease of demonstrating PTF activity in the CNS is accentuated by the dif- ficulty of showing ARF activity. In fact, only very high concentrations of the CNS fractions (cytosol and the fraction retained by XM-300 filter) injected into intact 20 10 8 I 6 a 4 Qj V, ARF PTF(CNS) BSA' V •\Catalase Alcoholdehydrogenase V 3 PTF(blood) 0.2 0.4 0.6 Mobility 0.8 1.0 FIGURE 3. Semilog plot of molecular weight as a function of mobility of marker proteins using SDS-gel electrophoresis ; Vi-V4 : Proteins of potato yellow dwarf virus with known molecular weights; ARF: Anterior retraction factor; PTF: Puparium tanning factor; BSA Bovine serum albumin. 176 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL red-spiracle larvae result in slightly accerelated anterior segment retraction (Table II). There are two possible explanations for this. Either the fractions do contain ARF in low concentration or, alternatively, the highly accelerated tanning and hardening of the cuticle might physically somehow make the hosts with- draw their mouth hooks before controls do. In order to clarify this point the above two fractions were injected with or without a-MDH into intact red-spiracle larvae of Sarcophaga biillata, a-MDH ( (DL)-a-Methyl-a-hydrazmo-£-(3,4-dihydroxy- phenyl) propionic acid monohydrate) is a potent DOPA decarboxylase inhibitor and injection of 1 m^-mole of this material into adult fleshflies has been shown to inhibit tanning almost completely (Seligman, Friedman and Fraenkel, 1969). As shown in Table IX, the cytosol fraction had ARF activity even when the tanning process was inhibited and this activity was lost after heating at 100° C for 3 minutes. However, the fraction retained by the XM-300 filter caused no acceleration of mouth hook retraction, when injected with a-MDH. Therefore the accelerated TABLE IX Effect of injection of the CNS fractions with or without a-MDH into red-spiracle larvae of S. bullata. The figures are means from 20 observations. The quantities injected per host larva were, CNS fraction: 10 fj.g; MDH: 1 nip-mole % Control 1' 1 d*~ L1U11 Retraction Tanning Ringer 100 100 a-MDH 100 * Cytosol 70 68 Cytosol plus a-MDH 81 * Cytosol heated to 100°C plusa-MDH 100 * Cytosol retained by XM-300 filter 83 79 Cvtosol retained by XM-300 filter "plusa-MDH 100 * * No tanning takes place in presence of a-MDH. mouth hook retraction that occurs after injection of this fraction is probably not due to the presence of ARF but is due to some physical process that forces the larvae to withdraw their mouth hooks when the tanning process has been greatly accelerated. The results of this experiment also indicate that the ARF which is present in the whole cytosol fraction probably has a MW below 300 K, since no activity re- mains after filtration through an XM-300 filter. Can the hormonal factors (ARF and PTF} replace ecdysone? In order to see whether or not the hormonal factors can replace ecdysone in whole or in part with regard to pupariation, several active fractions were injected into larvae in 3 different stages either with or without simultaneous injection of /3-ecdyosne. Since 0.018 p.g of /3-ecdysone per larva was reported to be the critical dose for Sarcopliaga pcregrina (Ohtaki, Milkman and Williams, 1967), a sub- threshold dose of 0.01 ,ug and a suprathreshold dose of 0.3 /xg of /3-ecdysone were injected. The subthreshold dose was used on the assumption that the hormonal HORMONAL CONTROL OF PUPARIATION 177 factors might need some ecdysone in order to enable them to penetrate the target tissues. The fractions were injected at two different periods : either simultaneously with ecdysone or an hour after ecdysone injection. The purpose of injecting an hour after ecdysone was to give the fractions an opportunity to act if they, like ecdysone, have a short half life, but can only function after ecdysone has exerted its action. The samples injected included (a) 8 jul of active hemolymph from orange puparia (which is equivalent to 15 ARF units and 5 PTF units), (b) five /Ag each (equivalent to about 10 units each) of ARF and PTF from the fractions eluted from electrophoresed gels, or (c) ten /xg (equivalent to 5 PTF units) of cytosol fraction of the CNS. Three different stages of host larvae used for injections were (i) pre-critical stage larvae, about 36 hours before pupariation which contain no or very little ecdysone, (ii) early post-critical stage larvae, about 12 hours before pupariation which contain some ecdysone and (iii) X-irradiated (10,000 R) pre-critical stage larvae. In the X-irradiated larvae, the synthesis of /?-ecdysone is temporarily blocked (Sivasubramanian, Ducoff and Fraenkel, 1974) and hence such larvae are suitable to test the activity of the hormonal factors in the absence of ecdysone. The results of this experiment were completely negative. Neither the hormonal factors from the hemolymph nor the CNS induced puparium formation. They accelerated puparium formation, as seen in all the previous experiments, only in a red-spiracle stage larva, i.e., long after ecdysone has triggered the initial events of pupariation. DISCUSSION In a previous report two alternative models for the hormonal events that occur during puparium formation were proposed (Fraenkel et al., 1972). One of these assumed that the synthesis of "X-factors" which were responsible for the accelera- tion of puparium formation and tanning was induced by ecdysone, and that the active substances in the CNS were entirely different from those of the hemolymph and merely accelerated the release of the X-factors into the hemolymph. The other model suggested that the X-factors in the hemolymph and CNS were essen- tially identical substances that were produced in the brain, stored in the peripheral nerve terminals and released into the hemolymph under the influence of ecdysone. The experimental evidence obtained in this investigation support the latter scheme. Based on the results of the present work and the available information in the literature a comprehensive model for the control of puparium formation in fleshflies is presented in Figure 4. According to this model, the retraction of the anterior segments, the longitudinal contraction to form a barrel shape, and the shrinkage of the cuticle that results in a smooth cuticle are under direct neural control of the CNS. The retraction of the anterior segments is also controlled indirectly through a neurosecretion from the CNS, as is also tanning. The neurosecretory material synthesized in the brain contains 2 distinct entities, the anterior retraction factor (ARF) which controls the retraction of the anterior segments and the puparium tanning factor (PTF) which controls tanning, and these neurosecretions are assumed to be stored in the peri- pheral nerve endings in all segments of the larva. The molting hormone, ecdysone, whose synthesis and liberation is controlled by another neurosecretion, ecdysio- 178 S1VASUBRAMANIAN, FRIEDMAN AND FRAENKEL tropin, triggers the release of the neurosecretory materials from the peripheral nerves into the hemolymph. Ecdysone also induces the synthesis of various ma- terials such as nucleic acids, enzyme proteins, tanning substrates and other co- factors which enable the released neurosecretory material to bring about retraction of the anterior segments and tanning. Thus, the CNS, both directly and indirectly controls the entire process of puparium formation in flies. Arguments in favor of this scheme will be presented in this discussion. Of the various processes shown in the scheme, the prothoracotropic effects of neurosecretion and ecdysone-induced protein synthesis during pupariation are well documented in the literature. The involvement of both the CNS and what we now know as ecdysone during molting, and the relationships bet/ween the CNS and ecdysone were demonstrated in the pioneering work of Williams (1947). Re- ports on ecdysone-induced protein synthesis during pupariation, which is a special molting process, abound in the literature (Sekeris, 1965; Karlson and Sekeris, 1966; Sekeri, Sekeris and Karlson, 1968; Arking and Shaaya, 1969; Sahota and Mansing, 1970; Wyatt and Wyatt, 1971; Thomasson and Mitchell, 1972). How- ever, except for the enzyme DOPA decraboxylase which is involved in tanning, the relationship of these proteins to the various morphological events that occur during pupariation is not clear. As will be explained later, at least some of these proteins might coordinate with ARK and PTF in bringing about the retraction of the anterior segments and tanning. Though no experiments have so far been performed by us, it is clear from the studies of Osborne and his colleagues (Osborne, 1967; Finlayson and Osborne, 1968; Osborne, Finlayson and Rice, 1971) that neurosecretion is stored in nerve endings. Recently they have demonstrated (Osborne et al., 1971) the existence of neurosecretory nerve endings in 3 insects, including the blowfly larva. They suggest that the site of release of neurosecretory material is covered only by a flimsy layer of connective tissue and that the material is probably released into the immediate vicinity of the muscle and diffuses into the hemolymph. They were not sure about the function of the neurosecretory material and assumed that it might have some trophic function. What causes the release of neurosecretory material from the neurosecretory nerve endings and when it is released is little understood. Ecdysone has been showyn to be capable of inducing the release of neurosecretory material from cultured brains of a cockroach (Marks, Itycheriah and Leloup, 1972). Besides, there have been reports on the action of ecdysone on the plasma membrane (Kambysellis and Williams, 1971b). In a previous report (Fraenkel et al., 1972) it was demon- strated that the hormonal factors responsible for the acceleration of the retraction of the anterior segments and tanning can appear in the blood even in the absence of CNS. In that experiment larvae were ligated pre-critically (i.e., no ecdysone was present in the system), and the blood in the ligated hind parts became active after ecdysone injection when the cuticle started to tan. Several experiments in the present investigation have indicated that the hormonal factors in the neuro- secretory material of the CNS are essentially identical to those that appear in the hemolymph during pupariation (see below). This, coupled with the fact that the CNS from even much younger larvae (3 day old) contain accelerating factors (Zdarek and Fraenkel, 1969; Fraenkel et al., 1972) tempts us to suggest that the HORMONAL CONTROL OF PUPARIATION 179 CENTRAL NERVOUS SYSTEM Neuro- muscular effects Neuro- secretory effects Ecdysiotropin triggers the release of Neurosecretion migrates ECDYSONE Causes the release of Neurosecretion stored in the periphera 1 nerve endings \ \ \ Neurosecretion directly injected I I Triggers the synthesis of into ' hemolymph as Proteins and other cofactors (available in red-spiracle larva) Anterior segments retraction factor (ARF) Puparium tanning factor (PTF) Longitudinal contraction to the barrel shape Cuticula r shrinkage to smoother! the puparium Retraction of the anterior segments Ta n n i n g of the puparium PUPARIATION FIGURE 4. A comprehensive scheme of the interrelationships between neuromuscular and neuroendocrine events that occur during puparium formation in the fleshfly Sarcophaga bullata P. ; explanation in text. 180 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL Hemol y mph Homogenize and centrifuge at 105,000 g 60 min. Cytosol (ARF, RTF) Pellet Ultrafiltration through XM-3OO Amicon membrane 45-70% Ammonium sulfate precipitate (ARF, PTF) Ult raf i 1 1 rat ion through Amicon membranes Fraction passed (ARF) Fraction retained (RTF) Retained XM-300 Heated to 100° C, for 3 mi n., centrif uged the supernatant passed through XM-300, but retained by XM-50 (RTF) \ Passed by XM-300 but retained by XM-IOO filter (ARF) Gel f i I tra t ion through Bio-Gel P-300 column Fractions 41-47 Electrophoresis in 7% acrylamide gels t ARF t RTF Electroelution of ARF and RTF 1 SDS-Gel electrophoresis FIGURE 5. Scheme of the steps taken in the isolation of ARF and PTF from the hemolymph and CNS of Sarcophaga bullata. neurosecretory material containing these factors (ARF and PTF) is already stored in peripheral nerve endings in a pre-critical stage larva and is eventually released into the blood by the action of ecdysone prior to puparium formation. This assump- tion would be strengthened by a demonstration of the disappearance of neurosecre- tory material from the nerve terminals just prior to pupariation. HORMONAL CONTROL OF PUPARIATION 181 TABLE X Comparison of the effects of various treatments on the retraction factor and tanning factor from the hemolymph of orange puparia of S. bullata Treatment ARF PTF (1) Heating at 100° C for 3 minutes (2) Ultrafiltration (3) Acrylamide gel electro- phoresis (4) Subunit molecular weight as determined by SDS- gel electrophoresis (5) Injection of the purest fraction obtained by electroelution Activity lost Activity passed through XM-300, but retained by XM-100 filter Activity in the middle ^ of the gel 90,000 Accelerates the retraction of anterior segments only Activity remains Activity retained by XM-300 filter Activity in lower § of the gel 26,000 Accelerates tanning only Direct and conclusive evidence has been obtained to show that there are two distinct factors for retraction and tanning. Figure 5 summarizes the various steps by which the separation of ARF and PTF from hemolymph, and PTF from the CNS was achieved. The properties of the two factors isolated from the hemolymph are different with regard to heating, ultrafiltration, electrophoresis, and subunit molecular weight (Table X). The CNS extracts, when injected directly into the larva effect the retraction of anterior segments and tanning (Table II), the tanning effect being more pre- dominant than the retraction effect. Nevertheless, the question arises as to whether the factors are identical to those in the hemolymph, and if so, why the tanning effect is predominant. Some of the characteristics which lend credence to the idea that the factors in the two tissues are identical have been listed in Table XI. HOW- TABLE XI Comparison of the properties of the ARF and PTF from hemolymph and CNS extracts Property Hemolymph CNS ARF PTF ARF PTF Ultrafiltration through XM-300 filter Heating at 100° C for 3 Passed through filter Destroyed Retained by filter Not destroyed Passed through filter Destroyed Retained by the filter Not destroyed minutes Acrylamide gel electro- phoresis Activity in mid- dle J of the gel Activity in the lower | of the gel * Activity in the lower 5 of the gel * Since the concentration of ARF was very low in the CNS extracts it was not isolated and hence the information is not available. 182 SIVASUBRAMANIAN, FRIEDMAN AND FRAENKEL ever, there is a significant difference in heat resistance between CNS and hemo- lymph PTF's. That from the CNS is not broken down upon heating at 100° C whereas PTF from hemolymph is partially split (Table III). We do not know the reason for this difference. Although the factors in the CNS and the hemolymph appear very similar, the relative quantities of ARF and PTF are quite different. In unfractionated hemo- lymph, injection of about 100 /xg of protein accelerates retraction of the anterior segments in 50% of control time; but to induce any appreciable tanning activity at least 300 /Ag must be injected (Table I). However, it appears that the amount of CNS protein to affect PTF activity is one fifth of that which affects ARF activ- ity. Hence, in the CNS the concentration of PTF must be higher than that of ARF and in the hemolymph the opposite must be true. We have no explanation as to why such differences in amount exist when they are secreted from one source. The studies of Zdarek and Fraenkel (1969) demonstrated that the PTF of the CNS originates in the pars intercerebralis of the brain. Since their experiments were performed with ligated larvae, the existence of ARF in the CNS was not known at that time. However, as described above, ARF also is present in the cytosol fraction of the CNS and it is possible that it, too, originates from the pars inter- cerebralis. Ecdysiotropin and bursicon also originate from the same region. Are these two hormones related to either of our hormonal factors ? It appears that with regard to ARF, the answer is an unequivocal no. ARF has a molecular weight of about 180,000 with two subunits of 90,000 each, whereas the molecular weights of ecdysiotropin and bursicon are well below 50,000. Furthermore, ARF is completely destroyed upon heating at 100° C for 3 minutes while the other two hormones of the brain are partially heat resistant. What then, of the relationship between PTF and the other brain hormones? With a subunit molecular weight of about 26,000, and resistance to high tempera- ture, it is possible that PTF may be related to ecdysiotropin. However, all pre- vious attempts to isolate ecdysiotropin have ended up with multiple factors and the situation is presently too confusing to make any meaningful comparisons. PTF as isolated from the CNS is most probably not related to bursicon. It is retained above the XM-300 filter even after heating and has a subunit molecular weight of about 80,000 which is about twice the size of bursicon. However, the situation with regard to hemolymph PTF is slightly confusing. This fraction, which is initially retained by the XM-300 filter, is partially broken down when heated to 100° C and some of the activity then passes through the XM-300 filter and is retained by the XM-50 filter. After further purification by preparative acrylamide gel electrophoresis both heated and unheated PTF contain- ing fractions appear to have a subunit molecular weight of 26,000. It appears that passage through the gel causes certain changes in the structure of the unheated high molecular weight material. One might argue that the hemolymph PTF has a molecular weight of about 52,000 with two subunits of 26,000 daltons, and that the subunits of this molecule are attached to a bigger molecule leading to its reten- tion on the XM-300 filter when unheated. If this is true, then the question arises as to whether it might be related to bursicon which has a molecular weight of about 40,000 (Fraenkel, Hsiao and Seligman, 1966) and is also a tanning hormone secreted by the pars intercerebralis of the brain. The answer is most probably no because, (a) by heating at 100° C for 3-5 minutes 40-90% of bursicon activity HORMONAL CONTROL OF PUPARIATION 183 is lost (Fraenkel et al, 1966) but PTF activity is not (Table IV), and (b) in acrylamide gel electrophoresis bursicon appears at an Rf of 0.3-0.4 ; but PTF comes very close to the dye front. Aside from the above, the injection of adult hemolymph active for bursicon into red-spiracle larvae did not accelerate tanning, and conversely, injection of PTF active hemolymph from orange puparia did not induce tanning in neck-ligated newly emerged adult flies (Zdarek and Fraenkel, 1969). As shown in previous experiments, moderately pure samples taken from acryla- mide gels are independent in their action. The ARF fraction effects only the re- traction of anterior segments, and the PTF fraction effects only tanning. However, these factors are not effective in pre-critical stage larvae which lack ecdysone, in the same stage larvae when injected with subthreshold doses of ecdysone, nor when injected into early post-critical stage larvae (about 10-12 hours before pupariation). They are effective only if injected into red spiracle stage larvae. Obviously, the active factors cannot replace ecdysone in effecting pupariation, and it is probable that they need ecdysone-dependent substances (e.g., enzyme proteins, other co-factors), which become available only in the red-spiracle stage, but one can only speculate about the role of such substances in their action. According to Kambysellis and Williams (1971a, 1971b, 1972), an undialyzable "macromolecular factor" from the hemolymph indispensable for the maturation of spermatozoa in the silkworm, Samia cynthla, is present in high titer in the hemo- lymph during and immediately after pupation. It is inactive in cultures of intact testes in the absence of ecdysone, but promotes spermatogenesis of germinal cysts if the cysts are removed from testes and then cultured. This suggested that the function of ecdysone here is to alter the permeability of the testis walls to facilitate the entry of the macromolecular factor. It is possible that during pupariation ecdysone might function in a similar way, aside from releasing the neurosecretory material from the neurosecretory nerve terminals. To conclude our discussion of the scheme presented in Figure 4, there is now evidence of direct control by the brain of certain muscular events during puparia- tion, such as longitudinal contraction to the barrel -shape and cuticular shrinkage to smoothen the puparial surface (Sivasubramanian. Ducoff and Fraenkel, 1974). Thus, the experimental evidence obtained in the course of this study confirms a previous report with regard to the existence of two different proteinaceous factors which separately effect retraction of the anterior segments and tanning during pupariation. It has also now been demonstrated that the factors in the hemo- lymph and CNS have a number of common properties, which makes it most probable that the source of the material in the blood is the CNS. Even though the scheme in Figure 4 so far only applies to a single fly species, Sarcophaga bnllata. we believe it to be valid for all cyclorrhaphous flies. It is also known that PTF occurs in the corpora cardiaca of Periplancta, the brain of a bug, Pyrrhocoris apt cms, and the brain of adult flies (Zdarek and Fraenkel, 1969), which suggests that it may play some role in insects which tan after molting. How far the model can be applied to other insects is difficult to predict at this point. This study was supported by NSF grant GB-2322 and NIH grant 5-K6-GM- 18,498 made to G. F., and USPH grant AI 06345 made to S. F. 184 S1VASUBRAMANIAN, FRIEDMAN AND FRAENKEL SUMMARY Injection of hemolymph from orange puparia or of CNS-extracts from larvae into red-spiracle larvae of Sarcophaga bullata accelerates the onset of pupariation in relation to retraction of the anterior segments and tanning. By the use of ammonium sulfate precipitation, heat treatment, ultracentrifugation, ultrafiltration, gel nitration, and electrophoresis, two proteinaceous factors were isolated and par- tially purified. One of these, called ARF (anterior retraction factor), accelerates retraction and has no tanning activity. The other, called PTF (puparium tanning factor), accelerates tanning and has no ARF activity. The ARF has a molecular weight of about 180,000, is heat labile, and was purified about 170 times. The PTF has a molecular weight of about 312,000, is heat stable, and was purified about 200 times. Similar factors were found in hemolymph and CNS-extracts, however, hemolymph was more active in ARF, and CNS in PTF. These factors are most probably not related to two other brain hormones, ecdysiotropin and bursicon, because of differences in molecular weights, electrophoretic mobilities, and distribution. Based on the similarities in the nature of the ARF and PTF in hemolymph and CNS it is suggested that they originate in the CNS, are stored in peripheral nerve endings, and released through the action of ecdysone into the hemolymph during pupariation. The factors have no effect in the precritical (36 hours before pupariation) and early post-critical (10-12 hours before pupariation) stage larvae. They are functional only in a red-spiracle larva (2-4 hours before pupariation). This sug- gests that they cannot replace ecdysone in effecting pupariation, and that they re- quire ecdysone-dependent enzyme proteins or other co-factors which become avail- able only in the red-spiracle larva. The role of such substances may be to alter the permeability of target cell walls in order to facilitate the entry of ARF and PTF. LITERATURE CITED ARKING, R., AND E. SHAAYA, 1969. Effect of ecdysone on protein synthesis in the larval fat body of Calliphora. J. Insect Physiol., 15 : 287-296. DAVIS, B. J., 1964. Disc electrophoresis. II. Method and application to human serum pro- teins. Ann. Nnv York Acad. Sci., 121 : 404-427. FINLAYSON, L. H., AND M. P. OsBORNE, 1968. Peripheral neurosecretory cells in the stick insect (Carausius morosus) and the blowfly larva (Phormia terrac-norae) . J. Insect Physiol.. 14: 1793-1801. FRAENKEL, G.. C. HSIAO, AND M. SELIGMAN, 1966. Properties of bursicon : An insect protein hormone that controls cuticular tanning. Science, 151 : 91-93. FRAENKEL, G., J. ZDAREK AND P. SIVASUBRAMANIAN, 1972. Hormone factors in the CNS and hemolymph of pupariating fly larvae which accelerate puparium formation and tanning. Biol. Bull.. 143 : 127-139. KAMBYSELLIS, M. P., AND C. M. WILLIAMS, 1971a. In vitro development of insect tissues. I. A macromolecular factor prerequisite for silkworm spermatogenesis. Biol. Bull., 141: 527-540. KAMBYSELLIS, M. P., AND C. M. WILLIAMS, 1971b. In vitro development of insect tissues. II. The role of ecdysone in the spermatogenesis of silkworms. Biol. Bull., 141 : 541-552. KAMBYSELLIS, M. P., AND C. M. WILLIAMS, 1972. Spermatogenesis in cultured testes of the cynthia silkworm: Effects of ecdysone and prothoracic glands. Science. 175: 768-769. KARLSON, P., AND C. E. SEKERIS, 1966. Ecdysone, an insect steroid hormone, and its mode of action, Rcc. Progr. Hormone Res., 22 : 473-502. HORMONAL CONTROL OF PUPARIATION 185 LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR AND R. J. RANDALL, 1951. Protein measure- ments with the folin phenol reagent. /. Biol. Chcm., 193: 265-278. MAIZEL, J. V., JR., 1964. Preparative electrophoresis of proteins in acrylamide gels. Ann. New York Acad. Sci., 121 : 382-390. MARKS, E. P., P. T. ITTYCHERIAH AND A. M. LELOUP, 1972. The effect of p-ecdysone on insect neurosecretion. /. Insect Physio!., 18 : 847-850. OHTAKI, T., R. D. MILKMAN AND C. M. WILLIAMS, 1967. Ecdysone and ecdysone analogues : Their assay on the fleshfly Sarcophaga pcregrina. Proc. Nat. Acad. Sci.. 58 : 981-984. OSBORNE, M. P., 1967. The fine structure of neuromuscular junctions in the segmental muscles of the blowfly larva. /. Insect Physio!.. 13: 827-833. OSBORNE, M. P., I. H. FINLAYSON AND M. J. RICE, 1971. Neurosecretory endings associated with striated muscles in three insects (Schistocerca, Carausius, and Phormia) and a frog. Z. Zctlforsch.. 166: 391-404. SAHOTA, T. S., AND A. MANSING, 1970. Cellular response to ecdysone: RNA and protein syn- thesis in larval tissues of oak silkworm Antheraea perni. J. Insect Phvsiol., 16: 1649-1654. SEKERI, K. E., C. E. SEKERIS AND P. KARLSON, 1968. Protein synthesis in subcellular frac- tions of the blowfly during different developmental stages. /. Insect Pliysiol. 14 : 425-431. SEKERIS, C. E., 1965. Action of ecdysone on RNA and protein metabolism in the blowfly, Calliphora erythroccphala. Pages 149-167 in P. Karlson, Ed., Mechanisms of Hormone Action. Academic Press, New York. SELIGMAN, M., S. FRIEDMAN AND G. FRAENKEL, 1969. Bursicon mediation of tyrosine hydro- xylation during tanning of the adult cuticle of the fly, Sarcophaga Inillata. J. Insect Physio!., 15 : 553-561. SIVASUBRAMANIAN, P., 1973. Puparium formation in the fleshfly, Sarcophaga bullata P. I. Effects of proteinaceous factors present in the hemolymph and the CNS of the larva. II. Effect of X-irradiation. Ph.D. thesis. University of Illinois, Vrbana, 114 pp. SIVASUBRAMANIAN, P., H. S. DUCOFF AND G. FRAEXKEL, 1974. Effect of x-irradiation on pupariation in the fleshfly, Sarcophaga bullata. J . Insect Physiol.. 20 : in press. THOMASSON, W. A., AND H. H. MITCHELL, 1972. Hormonal control of protein granule accum- ulation in fat bodies of Drosophila melanogaster larvae. /. Insect Phvsiol., 18: 1885- 1900. WEBER, K., AND M. OSBORNE, 1969. The reliability of molecular weight determination by dodecyl sulfate-polyacrylamide gel electrophoresis. /. Biol. Cliciu.. 244: 4406-4412. WILLIAMS, C. M., 1947. Physiology of insect diapause. II. Interaction between the pupal brain and prothoracic glands in the metamorphosis of the giant silkworm Platvsantia cccropia. Biol. Bui.. 93 : 89-98. WYATT, S. S., AND G. R. WYATT, 1971. Stimulation of RNA and protein synthesis in silk- moth pupal wing tissue by ecdysone in vitro. Gen. Comp. Endocrinol. 16: 369-374. ZDAREK, J., AND G. FRAENKEL, 1969. Correlated effects of ecdysone and neurosecretion in puparium formation (pupariation) of flies. Proc. Nat. Acad. Sci., 64: 565-572. Reference: Biol. Bull., 147: 186-202. (August, 1974) INHIBITION OF NEMATOCYST DISCHARGE IN HYDRA FED TO REPLETION l SCOTT SMITH, JAMES OSHIDA AND HANS BODE Department of Developmental and Cell Biology, School of Biological Sciences, University of California, Irvine, California 92664 The nematocysts of coelenterates are commonly assumed to be independent effectors ; that is, each nematocyst discharges in direct response to environmental stimuli, usually a combination of mechanical and chemical factors provided by the prey (e.g. Pantin, 1942; Picken and Skaer, 1966). Though stimulation of dis- charge is apparently not influenced by the animal itself, there is some evidence that inhibition of nematocyst discharge is. The sea anemone, Calliactis parasitica, will use the nematocysts of the tentacles to make a preliminary adhesion to a whelk shell. Once a solid attachment is formed by the pedal disk, the tentacles will not react to a second whelk shell indicating an inhibition of discharge (Davenport, Ross and Sutton, 1961). Ross and Sutton (1964) demonstrated that the tentacles of the sea anemone, Stomphia, will not capture prey when swimming which, though not strictly proven, suggests a control of nematocyst discharge. More recently, Sandburg, Kanciruk and Mariscal (1971) have shown that the anemone, Cal- liactis tricolor, greatly reduces nematocyst discharge upon feeding to repletion. Similarly, hydra will no longer capture shrimp larvae after feeding to repletion (Hyman, 1940; Burnett, Lentz and Warren, 1960) although Burnett et al. (1960) did not attribute this phenomenon to the inhibition of nematocyst discharge. In this report, we have extended the description of nematocyst discharge behavior of hydra fed to repletion. The results indicate that inhibition of discharge does exist and is probably due to a rise in concentration of a metabolite (s). MATERIALS AND METHODS Hydra attenuate, were used in all experiments. They were maintained in M solution (Lenoff and Brown, 1970), fed daily with Artemia nauplii, and washed 6-8 hours after feeding. All feeding experiments were conducted with adult hydra having 1-2 buds. Each hydra was placed in 1 ml (in some experiments) or 10 ml (in others) of M solution. The hydra was offered a given number of shrimp, 1-10, depending on the experiment, which were placed repeatedly within the ring of tentacles until the animal caught them. Several minutes later (6-20), the number of dead and alive shrimp remaining outside the hydra was counted under a dissecting microscope (25 X). From these numbers, the number killed and ingested was determined. After removal of the uningested shrimp, another group of shrimp (the same num- ber as before) was offered. This procedure was repeated until the hydra would no longer ingest or kill shrimp, and the number captured was greatly reduced. 1 This work was supported by grants from the National Science Foundation (GB-29284) and the National Institute for Child Health and Development (HD-08086-01). 186 INHIBITION OF NEMATOCYST DISCHARGE 187 Inhibition of the killing response, or stenotele discharge, was defined as five consecutive rejections of offered shrimp. A rejection was defined as a repeated contact ( > 4 ) of one tentacle or three contacts of at least two tentacles by a shrimp which elicited neither a capture nor a killing response. Also, shrimp which were captured, escaped, and were still swimming normally 30 seconds thereafter were considered as rejected. At least two different shrimp had to be involved in the five consecutive rejections. Shrimp which contacted the tentacles and were not cap- tured, but became immobile within 5-10 seconds were counted as killed. For grafting experiments, hydra were bisected above the budding region, and the two apical halves were grated together by threading the pieces, back-to-back, on 12 Ib nylon fishline (Stren, 0.013 in diameter). Sleeves of polyethylene tubing- were placed at either end of the graft to prevent migration of individual pieces. After 60 minutes, the graft was removed to M solution and used in feeding experi- ments 4-7 hours later. Nematocysts were discharged from the tentacles by electric shock as follows. Hydra were placed in M solution containing 2-10"4 M fructose- 1, 6-diphosphate (which enhances stenotele discharge ; Lentz and Barnett, 1962) in a leucite chamber fitted with silver electrodes at two ends. Single pulses of current (40V, 5 mAmp, 13 msec duration) were delivered every 5 seconds by a Grass SD6 Student Stim- ulator. The percentage of stenoteles discharged was measured by counting the number of stenoteles in the tentacles before and after shocking as follows. Animals were placed in a drop of M solution on a glass slide, flattened slightly but not killed with a coverslip, and the visible stenoteles on the upper half of each tentacle counted with phase microscopy. The shocking procedure resulted in the discharge of 60% of the stenoteles in 10 minutes, and 90% in 40 minutes. By 40 minutes, about 50% of the other three nematocyst types were estimated to be discharged. The animals were not damaged by this procedure and were used for experiments 60 minutes after treatment. Dilute and concentrated homogenates of Artcmia were prepared as follows. Specimens of Artemia were centrifuged at 2000 g for 2 minutes. The loosely pelleted animals were transferred to a glass tissue homogenizer and homogenized. The homogenate was centrifuged at 300 g for 5 minutes to remove the skeletal parts. The resulting supernatant was termed dilute homogenate. The concentrated homogenate was prepared by pouring a thick suspension of Artcmia onto a filter paper funnel to remove the fluid. The wet mass of shrimp was homogenized and the skeletal parts removed as before. Either homogenate was introduced into the gastral cavity of a hydra with a 10 /xl Hamilton syringe tipped with a polyethylene needle. Normally, two injections spaced 15 minutes apart were made. After another 30 minutes, the ability of the animal to capture and kill shrimp was tested. RESULTS Pattern of inhibition oj ncinatocvst discharge dnrin;/ jecding To put the nematocyst discharge behavior during feeding in context, a descrip- tion of the capture, killing and ingestion of prey by hydra is necessary. Prey (usually Artcmia nauplii in the laboratory) swimming close to the tentacles contact the long cnidocils of the desmonemes and cause discharge of these nematocysts. Each nematocyst everts a short, thick thread which wraps itself around any rod- 188 SMITH, OSHIDA AND BODK like object, such as, a spine on the swimming appendage of a shrimp. The trapped, struggling shrimp contacts the tentacle repeatedly and very soon hits the shorter cnidocil of a stenotele nematocyte. The nematocyst is discharged, everting a hollow thread which penetrates the shrimp, and a toxin is injected (Hessinger, Lenhoff and Kalian, 1973; Shapiro, 1968) which almost instantly paralyzes the shrimp, and probably kills it. Thereafter, the Artemia is transported to the hypo- stome by the tentacle and ingested. After taking in a number of shrimp by repeat- ing this process several times, ingestion ceases. The number of shrimp ingested varies from day to day but for a group of animals of similar size is reasonably constant on a given day. Since it is the stenotele and desmoneme nematocysts which are involved in feeding, the pattern of their discharge during the entire feeding process was first examined. Individual one-day starved hydra were placed in 1 ml M solution and a single Artemia placed within the crown of tentacles and the response noted. This was repeated until the animal no longer ingested or killed the shrimp larvae. A typical pattern with three or four stages emerged. At first, the starved animals quickly captured, killed, and ingested the shrimp. In contrast to earlier observa- tions (Burnett et a/., 1960), shrimp were caught and killed on an out-stretched tentacle. Curling of the tentacle after capture to aid in bringing the shrimp in contact with a stenotele cnidocil did not seem to be necessary. After the ingestion of a number of shrimp, specimens of Artemia, were caught and killed but not ingested. Later shrimp were captured but not killed and eventually escaped. Such shrimp frequently struggled violently for up to two minutes making innumerable contacts with one or more tentacles before escaping. They were found to be swimming normally at least thirty minutes later indicating that they had not been struck by a stenotele nematocyst. This behavior indicates that discharge of the stenotele nematocyst had become inhibited. Finally, a stage was reached where shrimp could contact several tentacles in succession without capture. Though this was not carefully documented, and inhibition was not as complete as for stenotele discharge, a distinct inhibition of desmoneme discharge was apparent. The behavior, or feeding activity, of six animals is shown in Figure 1 (solid line ) in which each of the responses is given a numerical value : 3 = capture, kill and ingestion; 2 == capture, kill, but no ingestion; 1 = capture and escape; and 0 = no capture. The pronounced decreasing trend, which is typical, clearly depicts the change in response of the animal (= change in nematocyst discharge behavior) during the course of feeding. The increasing cessation of killing indicates that stenotele discharge becomes increasingly inhibited. The decreased activity and in- hibition of nematocyst discharge is not due to a factor accumulating in the medium because the same behavior is obtained if the medium is changed after every five shrimp offered, or if the experiment is carried out in fifty ml of M solution. As a second measure of the increasing inhibition of stenotele discharge, the time re- quired to kill a shrimp after capture was determined for each consecutive shrimp caught and killed by the animal. With increasing number of shrimp killed and ingested, the time of killing increases implying an increasing inhibition of stenotele discharge or a raising of the threshold for stenotele discharge. Burnett et al. (1960) also found that after feeding to repletion killing of shrimp ceased and their capture occurred far less frequently than at the beginning of feeding INHIBITION OF NEMATOCYST DISCHARGE 189 3.0 X CD "O c o 2.0 CD c CD CD UL 1.0 1-5 610 11-15 16-20 21-25 26-30 Consecutive Artemia offered FIGURE 1. Quantitation of feeding behavior of hydra expressed as the feeding activity index. Individual hydra were offered one Artemia at a time. The response of the hydra to each Artemia was given a numerical value, the feeding activity index : 3.0 = capture, kill and ingestion of the shrimp; 2.0 = capture, kill, but no ingestion ; 1.0 = capture and escape. Each point represents the average value of responses to five consecutive shrimp/hydra and the average of six hydra. Normal animals expressed by (filled circles) ; animals with a truncated gastral cavity by (open circles). 190 SMITH, OSHIDA AND BODE when the animal was starved. They have offered the interesting interpretation that the lack of capture was not due to the inhibition of desmoneme discharge, but in- stead, was due to the rapid release of the discharged desmoneme from the tentacle. Further, the cessation of killing was not due to the inhibition of stenotele discharge but rather due to the lack of opportunity of fully active stenoteles to fire since the shrimp were rapidly lost from the tentacles. Normally nematocysts are discarded after discharge, but only after enough time has elapsed (2-5 minutes) so that the captured and killed prey can be ingested. If a rapid release existed, then the time a captured shrimp remained attached to the tentacle after repletion should be much shorter than before repletion. This possibility was examined in H. attcnuata by measuring the time elapsed between capture and kill before and after repletion, and TABLE I Length of time captured Artemia remain attached to tentacles at three stage of feeding activity. Individual hvdra were offered four shrimp at a time and the time elapsed between capture and kill or escape measured with a stop watch. Each time value is the average elapsed time for the indicated number of shrimp captured. The error is expressed as the mean deviation Expt. Hydra Capture, kill & ingestion Capture, kill & release Capture & escape Number of shrimp Time (sec) Number of shrimp Time (min) Number of shrimp Time (min) 1 1 15 20.7 — — 3 38.3 2 14 47.9 1 130 5 47.4 3 13 54.5 2 36.5 5 44.8 4 10 42.6 7 74.7 7 185.3 5 12 128.6 1 42 7 38.3 Average 58.9 ± 27.9 70.8 ± 31.5 70.8 ± 45.8 2 1 20 16.3 16 31.6 4 60.8 2 11 9.6 23 19.4 5 11.0 3 20 11.8 20 40.5 4 29.0 4 12 8.0 30 24.6 — — Average 11.4 ± 2.6 29.0 ± 7.0 33.6 ±18.1 the time between capture and escape after cessation of killing had set in. As shown in Table I, the time that a captured shrimp remained actively struggling on the tentacle after repletion and cessation of killing was longer than before repletion. Instead of a rapid release, captured shrimp remain attached to the tentacle longer after feeding to repletion in H. attcnuata. Thus, cessation of killing is not due to lack of contact of the shrimp with active stenoteles, but due to the inhibition of stenotele discharge. Similarly, lack of capture sometime after feeding to repletion is probably due to cessation of desmoneme discharge instead of a rapid release of discharged desmonemes. Three other aspects of the nematocyst discharge pattern were examined. Repletion in the animal could be a response simply to the number of shrimp ingested, or it could involve a time factor. Thus, cessation of ingestion may set INHIBITION OF NEMATOCYST DISCHARGE 191 TABLE II Effect of the amount of time elapsed on the total number of Artemia ingested. Error is the mean deviation Feeding regime E.xpt. Number of hydra Average number of shrimp ingested One shrimp at a time Two shrimp, a 1 hour pause, then one shrimp at a time Four shrimp, a 1 hour pause, then one shrimp at a time 1 8 7 7 6.4 ± 2.6 9.3 ± 2.9 9.1 ± 1.0 Fed to repletion, a 2 hour pause, fed again until repletion Fed 6 shrimp, a 2 hour pause, fed until repletion 2 12 14 18.4 ± 5.4 16.6 ± 2.7 in after feeding for a certain time period, regardless of the number of shrimp ingested. To decide between these alternatives, animals were fed under different regimes in two experiments as shown in Table II. In each experiment, the hydra reached repletion after having ingested a similar number of shrimp even though the total period of time until repletion was reached was different. Thus, the time interval from the onset of feeding to repletion does not play a role. The difference between the two experiments in the total number of shrimp ingested per hydra reflects the day-to-day variability in the number of shrimp a hydra will consume. 0) en ro 4-1 C c/> 0) O) c to E 0) 111 _a E D 100 80 - 60 c O u TJ 03 I 40 en 20 0,'0 1 2345678 Time after inhibition of stenotele discharge (hr) FIGURE 2. Kinetics of recovery from inhibition of stenotele discharge. The results of two experiments are presented together. Each point is the average of 4-6 hydra. The two curves are the percentage of Artcniin killed (filled circles) and the percentage ingested (open circles) compared to starved controls. 192 SMITH, OSHIDA AND BODE How long stenotele discharge remains inhibited, once inhibition was reached, was also investigated. Animals were fed to inhibition. Then, at intervals over a period of several hours, a group (different group for each interval) of animals was tested for their ability to ingest and kill shrimp. They were offered shrimp until cessation of killing set in. As shown in Figure 2, the animals do not ingest any and kill very few shrimp until they regurgitate the remainder of their previous meal, which occurs about four hours after feeding to repletion. There- after, the recovery to the state of a starved animal requires an additional four to five hours. A pronounced change from severe inhibition to a state of some killing and ingestion takes place shortly after regurgitation. Finally, the question arose as to whether or not the stenoteles mounted along the body column, which are not involved in the feeding process, were also inhibited from discharging after feeding to repletion. These stenoteles will also kill shrimp upon contact but do so without previous capture since there are no desmonemes mounted on the body column. The behavior of these stenoteles was examined by offering shrimp to the bodies of starved and repleted animals and then measuring the time required for the body to kill 15 shrimp. In the process of offering the TABLE III Effect of feeding to repletion on the discharge of stenoteles mounted on the body column. Each hydra was offered about 120 specimens of Artemia and the time required for the body to kill 15 shrimp measured. The number of contacts of shrimp with the body and tentacles were measured on average as 20 contacts /min for the tentacles and 12 contacts /min for the body. The error is expressed as the mean deviation State of feeding Number of hydra Average time required for body to kill 15 Artemia (min) Number of shrimp killed by tentacles in the time required for body to kill 15 shrimp Starved Fed to repletion 5 14 5.4 ± 2.5 19.5 ± 4.1 —40 4.9 ± 2.0 shrimp to the body, many were caught and killed by the tentacles, which proved to be a useful internal control. As seen in Table III, there is a striking dif- ference in the behavior of body and tentacle stenoteles in repleted animals. The bodies of one-day starved hydra killed the 15 shrimp 3^4 times faster than did the bodies of repleted animals. Thus, the body stenoteles of the repleted animals were inhibited to an extent. However, they were far less inhibited than the stenoteles of the tentacles. During the same time period, the stenoteles of the starved animals killed at least 10 times as many shrimp as the tentacles of repleted animals, and probably more since there was no room for more than forty shrimp on the tentacles. Thus, the inhibition of discharge affects those stenoteles involved in feeding more strongly. The inhibition of nematocyst discharge is due to a gastral cavity filled with shrimp The foregoing experiments indicate that the ability to discharge stenoteles and desmonemes is sharply inhibited in animals fed to repletion. The inability to kill more shrimp could be due merely to an exhaustion of fully mature stenoteles or an INHIBITION OF NEMATOCYST DISCHARGE 193 inhibitory feedback effect caused by discharging a given number of the stenoteles in the tentacles. A lack of mature stenoteles seems unlikely in the light of an experiment by Burnett ct al. (1960) which we corroborated. If the captured and killed shrimp are removed from the tentacles, thus preventing ingestion, a hydra will kill at least twice as many than if allowed to ingest them. Thus, the tentacles contain at least twice as many mature stenoteles as are used at the time of inhibi- tion of killing. Also, there is no apparent feedback on discharge due to a loss of the stenoteles. The availability and feedback arguments were examined more directly in the following experiments. First, the percentage of stenoteles used to feed to inhibition was determined by counting the number of stenoteles in the tentacles before and after feeding as described in Materials and Methods. As shown in Table IV, 21% TABLE IV Percentage of stenoteles discharged by animals fed to inhibition. A II errors are expressed as the mean deviation Hydra Number of Artemia Average number of stenoteles/ 5 tentacle* Percentage of stenoteles discharged (%) Killed Ingested Before feeding After inhibition 1 30 12 93.6 ± 30.9 64.2 ± 7.4 31.4 2 27 16 53.6 ± 9.1 52.0 ± 9.4 3.0 3 27 15 66.2 ± 10.9 50.8 ± 12.5 23.2 4 23 10 63.2 ± 12.7 34.6 ± 6.4 45.2 5 20 11 56.8 ± 6.6 39.8 ± 7.3 30.0 6 28 23 66.6 ± 12.7 47.4 ± 4.7 28.8 7 21 17 53.0 ±11.2 45.2 ± 15.0 14.7 8 23 16 51.2 ± 7.6 35.7 ± 6.8 30.3 9 26 18 55.0 ± 10.2 51.0 ± 13.2 7.3 10 15 10 34.0 ± 7.3 33.6 db 11.7 Ave 0.1 rage 21.1 ± 12.1 * Each value is the average of 4-6 ^-tentacles. of the stenoteles had been used. Sixteen per cent were used in a second experi- ment. If these percentages represented either the number of available mature stenoteles or the percentage of stenoteles discharged required to achieve inhibition of discharge by a feedback mechanism, then removal of 50% of the stenoteles before feeding should completely prevent killing of prey. This possibility was tested as follows. Two groups of hydra were selected. The stenoteles of one group were counted with the coverslip method. Then, both groups were subjected to electric shock, as described in Materials and Methods, for various periods of time (5-15 minutes). The group of animals previously counted were counted again. More than 60% of the stenoteles had been discharged. The second group of animals was placed in M solution, allowed to recover from the treatment for two hours, and then offered shrimp. As shown in Table V, they killed and ingested a normal number of shrimp. These results argue against availability of mature stenoteles and feedback due to a number of discharged stenoteles as the cause of the observed 194 SMITH, OSHIDA AND BODE TABLE V Effect of large numbers of discharged stenoteles on feeding behavior. Errors are expressed as the mean deviation Animals assayed for Animals assayed for feeding behavior number of stenoteles Length of after electric shock electric Expt. shock treatment Average number of shrimp Percentage of (min) No. Hydra Number of hydra stenoteles remaining Killed Ingested (%) 1 0 4 29.5 ± 1.5 19.0 ± 0.5 — (100)* 10 4 27.5 ± 2.2 17.5 ± 2.2 3 36.4 15 4 25.0 ± 3.0 20.0 ± 3.0 3 21.9 2 0 4 14.8 ± 1.7 12.0 ± 1.5 — (100)* 5 4 17.2 ± 4.8 14.5 ± 3.2 1 87.8 10 4 10.8 ± 3.2 9.2 ± 4.8 1 42.0 15 4 11.0 ± 2.5 8.0 ± 1.5 1 43.2 * 100% was assumed, not measured. Percentage of stenoteles remaining ^, Stenoteles/5 Tentacle After ,-=! Stenoteles/5 Tentacle Before /N) X 100 cessation of killing and support the view that stenotele discharge has become inhibited. To obtain evidence that the inhibition of discharge was directly related to the gastral cavity filled with shrimp, the following pair of experiments were carried out. If a filled gastral cavity is related to inhibition, then an animal with a smaller gastral cavity should ingest fewer shrimp and display the observed pattern of nematocyst discharge inhibition (Fig. 1) after a smaller number of offered shrimp than a normal animal. To test this possibility, animals \vere bisected beneath the ring of tentacles, the apical part allowed to heal for 24 hours, and then offered Artemia one at a time. As expected these smaller animals ingested fewer shrimp compared to normal animals (Table VI), and they displayed the cessation of killing and capturing behavior much earlier than the normal animals [compare smaller animals (dotted line) with normal animals (solid line) in Fig. 1]. These results indicate that the filled volume, but not the absolute number of shrimp, is significant. In a second experiment, two one-day starved animals were bisected in the mid- gastric region and grafted together as shown in Figure 3. The graft was allowed TABLE VI Number of shrimp ingested as a function of the size of the gastral cavity; Group A = animals bisected beneath ring of tentacles and allowed to heal; Group C = control. Errors expressed as the mean deviation Expt. # Hydra Group A Group C 1 12 2.9 ± 0.6 26.9 ± 2.4 2 12 4.5 ± 0.7 13.0 ± 1.5 3 8 5.3 ± 0.8 24.0 ±1.1 INHIBITION OF NEMATOCYST DISCHARGE 195 FIGURE 3. Grafting procedure for double-headed animals. to heal (which requires about 1--J hours; Bibb and Campbell, 1973) and then fed 4-7 hours later. In these experiments, the two heads had a common gastral cavity ; that is, a wall had not yet formed between the two halves. If both heads of a starved animal were offered shrimp simultaneously, both killed and ingested shrimp in equal amounts indicating that the graft combination had no effect on nematocyst discharge in either head. If one head was fed to repletion and inhibition of killing, and then, the second head fed, the second head ingested essentially no shrimp and captured or killed ^— 5 as many as the first head (Table VII). A similar effect was found, but not as pronounced if a wall partitioning the cavity existed. This indicates the inhibition need not be transmitted through the gastral cavity alone. Clearly, the presence of the shrimp in the gut caused the inhibition of the discharge of stenotele nematocysts in the second head. These results taken together with those of the earlier experiments provide more evidence that inhibition of stenotele discharge is the reason the killing of shrimp ceases after feeding to repletion and that a filled gastral cavity is in some way responsible for the inhibition of stenotele discharge. This result is not unexpected since animals normally cease eating when their stomachs are full. TABLE VII Effect of filled gastral cavity on the feeding behavior of the second head of a two-headed animal. Errors expressed as the mean deviation Animal Time after graft (hrs) First head : number of shrimp Second head: number of shrimp Captured Killed Ingested Captured Killed Ingested 1 4 15 13 11 10 7 0 2 4 9 7 3 9 3 0 3 4 16 14 9 4 2 0 4 6 21 17 11 2 2 1 5 6 24 20 13 5 4 0 6 6 18 18 8 3 1 0 7 6 19 19 14 2 1 0 8 6 26 23 12 6 2 1 9 6 20 20 7 6 4 0 10 6 26 24 10 3 1 0 11 7 21 17 10 3 2 0 12 7 11 11 5 3 3 1 13 7 15 9 3 3 3 0 Average 18.5 ±4.2 16.3 ± 4.2 8.9 ± 3.0 4.5 ± 2.0 2.7 ± 1.2 0.23 ± 0.35 196 SMITH, OSHIDA AND BODE Properties of a filled g astral cavity responsible for inhibition of ncuiatocyst discharge The next question that arose was what kind of information is transmitted from filled gastral cavity to the nematocysts in the tentacles. In animals higher on the evolutionary scale, several different kinds of mechanisms involved in satiation or the cessation of ingestion are known to exist. One is based on the physical disten- tion of the walls of the stomach (Janowitz and Grossman, 1949), which in several species of insects, is monitored by stretch receptors (Finlayson and Lowenstein, 1958). A second is based on the levels of metabolites, such as glucose, in the blood (Mayer and Thomas, 1967). As another, Mook (1963) has suggested the involvement of osmoreceptors in the stomach. Also, controls of ingestion related to the physical process of eating have been shown to exist (Grossman, 1955; LeMagnen, 1971). In analogy with other animals cessation of ingestion and inhibi- tion of nematocyst discharge in hydra may be based on one or more of these mechanisms. TABLE VIII Effect of extreme distention of the body wall on feeding behavior. Feeding was initiated 0-25 minutes after injection. The variable starting time had no effect. The errors are expressed as the mean deviation State of body wall Number of hydra Number of shrimp Captured Killed Ingested Control Distended 3 7 12.0 ± 2.7 14.3 ± 4.1 11.3 ± 3.1 11.1 ± 3.0 9.0 ± 0.7 8.6 ± 1.8 To test the idea that the hydra recognizes physical distension of the body wall, an air bubble (1.0-1.5 /A ) was injected into the gastral cavity of a hydra with a microliter syringe tipped with a polyethylene needle. The bubble, with a volume 10 times greater than the empty gastric cavity (0.1 /xl, Bode, unpublished results), extended the walls more than those of an animal fed to repletion. The hydra was then offered 3^4- Artemia at a time and the response noted. The results Table VIII) indicate that the hydra behaved as a starved animal killing and ingesting as many shrimp as a one-day starved animal. With the increasing number of shrimp ingested, the bubble decreased in volume, but the total extended volume of the gastral cavity remained unchanged. These results indicate that severe disten- tion of the body walls surrounding the gastral cavity has no effect on killing, hence on stenotele discharge in the tentacles. Thus, it is unlikely that the nature of the formation is due to mechanical stretching. Burnett ct al. (1960) carried out a similar experiment by injecting glass beads into the gastral cavity and reported that the animals killed half as many shrimp as did a set of control animals. The discrepancy between the two results may be more apparent than real for the experi- mental animals of Burnett et al. (1960) were compared with control animals tested in a previous experiment, and as noted above, the number of shrimp killed and ingested can vary greatly from day to day. An observation made in an experiment described earlier supports the con- clusion that the distended body wall of the gastral cavity is not responsible for the INHIBITION OF NEMATOCYST DISCHARGE 197 inhibition of stenotele discharge. Animals fed to repletion have a fully extended gastral cavity. Four hours later, shortly before regurgitation of the remaining food, the volume of the gastric cavity had returned close to that of a one-day starved animal, yet the animal killed very few of the offered shrimp ; that is, stenotele discharge was still inhibited to a large degree. If simple physical dis- tention of the body wall were related to inhibition of stenotele discharge, we would have expected the extent of stenotele discharge, we would have expected the extent of stenotele discharge to be more like that of a starved animal than like a repleted one. 15 1 12 Q. E i_ co 01 .Q E --o-- _L _L 10 20 30 40 Time after decapitation (min) 50 60 FIGURE 4. Decay of inhibition of stenotele discharge in severed heads. At each time interval after severance, a group (3-5) of hydra were offered 15 shrimp and the number killed within two minutes noted. The two curves are for heads of repleted animals (filled circles) and for heads of starved animals (open circles). A related possibility concerns "neck formation" in the animal. Some time after feeding to repletion, the portion of the body column directly beneath the ring of tentacles elongates, that is, stretches, becoming very narrow like a neck (Blanquet and Lenhoff (1968). A stretch receptor located in this region could monitor ingestion and hence be involved in the control of nematocyst discharge. How- ever, this is unlikely because neck formation takes place after inhibition of steno- tele discharge sets in. Since physical distention of the body walls is probably not involved, the inhibi- tion of stenotele discharge may be due to the accumulation of a specific metabolite somewhere in the animal or to a general osmotic effect due to the accumulation of ions and metabolites in the gastral cavity. If so, gradual changes in inhibition 198 SMITH, OSHIDA AND BODE might be expected as the level of (say) the metabolite rose or fell in the animal. The slow recovery from inhibition by animals after regurgitation of the remainder of the previous meal is consistent with this view, as are the three following experi- ments. Removing the head from the filled distended body of a repleted hydra releases the inhibition of stenotele discharge to some extent after 15 minutes (Burnett et al., 1960). This was examined more fully by decapitating hydra, and thereafter, perioically offering 15 shrimp to each of a group of severed heads (a different group was used for each time point) and noting how many were killed within two minutes. The head of a starved animal killed Artemia immediately after severance and behaved as an intact animal by killing essentially all shrimp offered (Fig. 4). Thus, simply the removal of the head, as was expected from earlier work (e.g., Pantin, 1942), had no effect on stenotele discharge. In contrast, a head removed from a repleted animal killed only one or two shrimp in two minutes. Hence, stenotele discharge was still inhibited. With increasing time after severance, TABLE IX Effect of different feeding regimes on the lag time between repletion and cessation of killing. Errors are expressed as the mean deviation Number of Number of Number of hydra shrimp offered at a time before Time required to feed to repletion shrimp offered at a time after Lag time between repletion and inhibition Number of shrimp killed after repletion repletion repletion 5 10 25.0 ± 1.6 10 20.0 ± 3.2 9.2 ± 1.4 5 10 27.2 ± 5.0 2 19.0 ± 6.8 5.8 ± 2.2 5 2 67.4 ± 5.1 10 13.6 ± 3.7 13.2 ± 6.4 4 2 93.5 ± 8.2 2 6.5 ± 3.8 2.8 ± 1.6 the head killed more of the offered shrimp, and by 30 minutes, behaved as a severed head of a control animal. The result indicates that the inhibition of discharge decays slowly which is consistent with the gradual loss of a metabolite by either degradation or leakage. The lag time between cessation of ingestion and inhibition is also of this general length suggesting that the inhibition of stenotele discharge sets in gradually. If the lag time is due to a gradual rise in the concentration of a metabolite, then the feeding regime (the number of shrimp offered and ingested per time interval) may affect the length of the lag time. To examine this, hydra were fed either two or ten shrimp at a time until repletion. As was expected (Table IX), the time required to reach repletion was three times longer for animals fed two shrimp at a time than for those fed ten at a time since repletion is a function of a gastral cavity filled with shrimp. Thereafter, animals fed to repletion of both groups were offered shrimp at short intervals until killing ceased. The lag time was, indeed, affected by the initial feeding regime. Animals fed to repletion by offering shrimp ten at a time had a lag of — • 20 minutes compared to ^ 10 minutes for those offered two at a time (Table IX). The difference in lag times is consistent with the view that a certain amount of metabolism must take place before the con- centration of some metabolite reaches levels which inhibit stenotele discharge. INHIBITION OF NEMATOCYST DISCHARGE 199 Hydra fed ten shrimp at a time reach repletion in 25 minutes, but the digestion must continue another twenty minutes before the concentration is high enough. Animals fed shrimp two at a time require longer to reach repletion ( — ' 75 minutes) and, therefore, have more time to accumulate a metabolite (s). This would explain why the lag time between repletion and inhibition of killing is shorter in this case. Further, if the lag time is a function of metabolism, the length of the lag time should be independent of the number of shrimp killed after repletion has set in. This point was examined in the same experiment. As shown in Table IX, animals fed to repletion by either regime were then offered either two or ten shrimp at short intervals until killing ceased. Those hydra offered 10 shrimp at a time killed twice as many as those offered two at a time indicating that the numbers of shrimp killed after repletion have no effect on the lag time. TABLE X Effect of two concentrations of homogenates of Artemia injected -into the gastral cavity on the feeding behavior. Errors are expressed as the mean deviation Feeding activity Number of shrimp of experimental animals as Expt. Injected solution Number of hydra compared to controls (%) Killed Ingested Killed Ingested 1 M solution 5 20.8 ± 3.8 8.2 ± 2.6 Dilute homogenate 5 14.8 ± 4.7 2.0 ± 1.6 71 24 2 M solution 4 17.5 ± 3.0 7.5 ± 1.8 Dilute homogenate 4 9.5 ± 4.0 0 54 0 3 M solution 5 20.2 ± 1.4 8.8 ± 0.7 Concentrated homogenate 7 5.1 ± 1.9 0 25 0 4 No injection 4 22.5 ± 2.0 10.2 ± 1.8 M solution 6 20.8 ± 2.6 9.5 ± 2.5 Concentrated homogenate 7 7.4 ± 2.5 0 35 0 Finally, if inhibition of stenotele discharge is due to a metabolite either obtained directly from the Artcmla or by metabolizing a component of the Artemia, then the degree of inhibition may be a function of the concentration of the original metabolite in the gastral cavity. This was tested as follows. Two different con- centrations of Artemia homogenate were prepared as described in Materials and Methods. Two injections, 15 minutes apart, of either homogenate into the gastral cavity yielded hydra that had the extended appearance of animals fed to repletion. Thirty minutes later, these animals were offered shrimp four at a time and their ability to kill and ingest the shrimp analyzed in the usual manner. As shown in Table X, the dilute homogenate inhibited killing by only 30-50% as compared to controls injected with M solution, whereas the concentrated homogenate inhibited ingestion completely and inhibited killing by 65-75%. In experiment 4, some animals that had not been injected were also tested, and it is evident that injection as such has no effect on the ingestion and killing behavior. These results are con- sistent with the hypothesis that a metabolite (s) is involved and supports an experi- ment carried out by Burnett et al. (1960). They injected homogenates of starved hydra and hydra repleted with shrimp and found that the latter, but not the former, 200 SMITH, OSHIDA AND BODE reduced the killing of shrimp. The fact that the dilute homogenate, estimated to be 5-10 times less concentrated than the concentrated homogenate, is significantly less effective suggests that the metabolite (s) must be present in rather high concentra- tion. Also, these results underscore the earlier result that the distended body walls are not responsible for the inhibition of nematocyst discharge since the dilute and concentrated homogenates extended the body walls to the same degree but have dif- ferent effects on the killing activity. DISCUSSION The results presented above indicate that stenotele, and to a lesser extent desmoneme, discharge can be inhibited by a hydra fed to repletion. This adds to the increasing body of evidence (see Picken and Skaer, 1966) that nematocytes are not completely independent effectors. It has been suggested (Burnett ct al., 1960) that the inhibition in hydra is more apparent than real because after an animal has fed to repletion, desmonemes discharged while contacting shrimp are released so rapidly that contact with the still active stenoteles is avoided. For H. attenuate,, the length of time between discharge and release of desmonemes does not change after reaching repletion, yet the killing of Artcuiia ceases, and, in time, far fewer are captured indicating that the discharge of stenoteles and desmonemes has become inhibited. This effect is not due to a lack of available mature stenoteles and desmonemes nor due to feedback inhibition based on the number of nematocysts discharged for the following reasons: (a) the animal will kill more shrimp than normal if prevented from ingesting captured and killed shrimp; (b) removal of more than half the available stenoteles has no effect on feeding whereas feeding to repletion requires less than 20% of the stenoteles; (c) animals with a truncated gastral cavity but normal set of tentacles will capture and kill far fewer shrimp than an animal with a normal-sized gastral cavity; (cl) if the first head of a two-headed animal with a common gastral cavity is fed to repletion and inhibition of killing, the second head will kill very few shrimp even though it has a normal complement of stenoteles in the tentacles. All of these experiments strongly support the original indications that stenotele discharge is inhibited after feeding to repletion. Other observations suggest that there are a variety of circumstances in which a hydra will not capture and kill shrimp, and indicating an inhibition of nematocyst discharge. Hydra subjected to electrical shock will not capture shrimp for 30 minutes after treatment. Animals squashed mildly under a coverslip will not feed for 60 minutes thereafter. Similarly, in a stock culture, some animals will feed normally and others will not feed at all. These observations are understandable in the teleological sense that a "sick" hydra or one fed to repletion has no need for food and, hence, will inhibit the discharge of nematocysts involved in obtaining food. In this connection, it is interesting that stenoteles mounted on the body column which are not involved in feeding are inhibited far less (~ 5 times less) than those mounted in the tentacles. This suggests that the mechanism of inhibition is somewhat specific for tentacle stenoteles. It is not known if the discharge of the two types of isorhizas are subject to inhibition. As to possible mechanisms underlying the inhibition of stenotele and desmoneme discharge, some information has been obtained. The experiments with two-headed INHIBITION OF NEMATOCYST DISCHARGE 201 animals and those with truncated gastral cavity f(c) and (d) in an earlier paragraph of the discussion] demonstrate that inhibition is correlated with a gastral cavity filled with shrimp, hut is independent of the absolute number of shrimp in- gested. Though an animal with a rilled gastral cavity has greatly distended body ways, suggesting physical distention as the mechanism, this is not the case. Animals fed to repletion 3 to 4 hours previously no longer have a distended gastral cavity, yet; stenotele discharge was still severely inhibited. Also, distending the body walls with an air bubble does not cause inhibition and, in fact, has no effect on feeding at all. Finally, two different concentrations of Artcmia homogenate injected into the gastral cavity, though distending the body walls to the same extent, had markedly different effects on the number of shrimp killed and ingested during subsequent feeding. The results of other experiments are consistent with a metabolite (s) as the basis of the mechanism of inhibition. The lag time between cessation of ingestion and inhibition of stenotele discharge is of the order of 10-20 minutes while the decay time of inhibition in severed heads is about 30 minutes. These gradual processes are much too long to be based on mechanisms involving the nervous system but are consistent with slow rise or fall in concentration of a metabolite. Further, the fact that the concentrated Artcmia homogenate was far more effective than the dilute homogenate in inhibiting stenotele discharge indicates not only that the Artcmia provided some component that caused inhibition but that the component must be in relatively high concentration. These data suggest that the Artcmia provides a (or several) specific metabolies(s) that is directly involved or is altered by the hydra and then involved in the inhibition of nematocyst discharge. The results are also consistent with an inhibition based on an osmotic effect due to a general rise in all metabolite and ion concentrations. The logical extention of this work is to determine whether inhibition is due to a general osmotic effect or due to a threshold concentration of a specific metabo- lite (s). The answer to this question will help clarify whether or not the basic mechanisms involved in satiation in hydra and more highly evolved animals are similar. SUMMARY 1 ) The pattern of nematocyst discharge was observed in hydra fed shrimp until repletion. Some time after cessation of ingestion stenotele discharge ceases, and at a later time desmoneme discharge is greatly diminished. Lack of capture of shrimp is not due to a rapid release of desmonemes in animals fed to repletion. 2) Stenotele discharge remains severely inhibited until regurgitation of the re- mainder of the previous meal. Thereafter, stenotele discharge becomes pro- gressively less inhibited and is normal 4-5 hours later. 3) Discharge of stenotele mounted on the body column is partially inhibited in animals fed to repletion, but far less inhibited than for the stenoteles in the tentacles. 4) Cessation of stenotele discharge after feeding to repletion is not due to the absence of mature stenoteles, as there are at least twice as many mature stenoteles present in the tentacles as are needed to feed to repletion. 5) Inhibition of stenotele discharge is correlated with a gastral cavity filled with shrimp, but is not correlated with the absolute number of shrimp ingested. 202 SMITH, OSHIDA AND BODE 6) Physical distention of the gastral cavity due to the ingested shrimp is not the cause of inhhition of stenotele discharge. 7) The changes in inhibition of stenotele discharge are gradual. Stenoteles in heads severed from animals fed to repletion recover from inhibition of discharge in thirty minutes. The lag time between cessation of ingestion and inhibition of discharge is ten-twenty minutes. 8) A concentrated homogenate of Artcmia injected into the gastral cavity inhibits stenotele discharge to a greater extent than does a dilute homogenate. 9) The results indicate that hydra can inhibit stenotele and desmoneme dis- charge, and that a metabolite (s), obtained from the Arteinia is involved in the inhibition. LITERATURE CITED BIBB, C., AND R. D. CAMPBELL, 1973. Tissue healing and septate desmosome formation in hydra. Tissue and Cell, 5(1) : 23-35. BLANQUET, R. S., AND H. M. LENHOFF, 1968. Tyrosine enteroreceptor of hydra. Its function in eliciting a behavior modification. Science, 159 : 633-634. BURNETT, A. L., T. LENTZ AND M. WARREN, 1960. The nematocyst of hydra. I. The ques- tion of control of the nematocyst discharge reaction by fully-fed hydra. Ann. Soc. Roy. Zool. Belg., 90 : 247-267. DAVENPORT, D., D. M. Ross AND L. SUTTON, 1961. The remote control of nematocyst dis- charge in the attachment of Calliactis parasitica to shells of hermit crabs. Vie Milieu, 12: 197-209. FINLAYSON, L. H., AND O. LOWENSTEIN, 1958. The structure and function of abdominal stretch receptors in insects. Proc. Roy. Soc. London Series B, 148: 433-449. GROSSMAN, M. L, 1955. Integration of current views on the regulation of hunger and appetite. Ann. New York Acad. Set., 63 : 76-89. HESSINGER, D., H. M. LENHOFF AND L. B. KAHAN, 1973. Haemolytic, phospholipase A and nerve-affecting activities of sea anemone nematocyst venom. Nature Neiu Biology, 241: 125-127. HYMAN, L. H., 1940. The Invertebrates: Protozoa Through Ctcnophora. New York, McGraw Hill Book Co. JANOWITZ, H., AND M. I. GROSSMAN, 1949. Some factors affecting the food intake of normal dogs and dogs with esophagostomy and gastric fistula. Amcr. J. Physiol., 159: 143-148. LEMAGNEN, J., 1971. Advances in studies on the physiological control and regulation of food intake. Progr. Physiol. Psycho!., 4 : 204-261. LENHOFF, H. M., AND R. D. BROWN, 1970. Mass culture of hydra : an improved method and its application to other aquatic invertebrates. Laboratory Animals. 4: 139-154. LENTZ, T. L., AND R. J. BARNETT, 1962. The effect of enzyme substrates and pharmacological agents on nematocyst discharge. /. Exp. Zool.. 149 : 33-38. MAYER, J., AND D. W. THOMAS, 1967. Regulation of food intake and obesity. Science, 156: 328-337. MOOK, D. G., 1963. Oral and postingestional determinants of the intake of various solutions in rats with esophageal fistulas. /. Cotnp. Physiol. Psychol., 56 : 645-659. PANTIN, C. F. A., 1942. The excitation of nematocysts. /. Exp. Biol., 19: 294-310. PICKEN, L. E. R., AND R. J. SKAER, 1966. A review of researches on nematocysts. Pages 19-50 in W. J. Rees, Ed., The Cnidaria and Their Evolution. Academic Press, New York. Ross, D. M., AND L. SUTTON, 1964. Inhibition of the swimming response by food and of nematocyst discharge during swimming in the sea anemone Stomphia coccinca. J. Exp. Biol., 41: 751-757. SANDBERG, D. M., P. KANCIRUK AND R. N. MARISCAL, 1971. Inhibition of nematocyst dis- charge correlated with feeding in a sea anemone, Calliactis tricolor (Leseur). Nature, 232: 263-264. SHAPIRO, B., 1968. Purification of a toxin from tentacles of the anemone Condylactis gigantca Toxicon, 5 : 253. Reference: Biol. Bull., 147: 203-212. (August, 1974) LIMB LOSS AND THE MOLT CYCLE IN THE FRESHWATER SHRIMP, PALAEMONETES KADIAKENSIS LOIS A. STOFFEL AND JERRY H. HUBSCHMAN Department of Biological Sciences, Wright State University, Dayton, Ohio 45431 In crustaceans, the presence of a chitinous exoskeleton imposes growth limita- tions on an organism. An animal must molt or shed its exoskeleton periodically in order to grow. Molt cycle control is thought to be regulated by two hormones, a molt-inhibiting hormone and a molting hormone. The X-organs are a group of neurosecretory cell bodies, located near the surface of the medulla terminalis in the crustacean eyestalk, which are presumably responsible for producing the molt- inhibiting hormone (MIH). MIH is then thought to inhibit the Y-organs, the postulated molting glands of crustaceans, from secreting the molting hormone (MH) which acts on the epidermis to initiate premolt or the preparations for ecdysis (Passano, 1960). Another characteristic of many crustaceans is the ability to regenerate append- ages which are lost. Regeneration is dependent upon molting. If a limb is removed at the beginning of the intermolt cycle, a limb bud is produced which grows during premolt (stage D of the intermolt cycle), but not until the animal molts is the limb fully formed and functional. Several molts are required to restore the limb to its original size (Goss, 1969). Since the molting process is under hormonal control, regeneration and normal growth are closely interrelated in crustaceans. Several studies have been conducted to examine the relationship between regeneration and the molt cycle in crustaceans. Zeleny (1905) noted in three species of crustaceans: the fiddler crab, Gelasimus (~- Ucaf) pitgilator, the pistol shrimp, Alphcus and the crayfish, Cambanis propinqiiiis, that an excessive loss of limbs led to an increased rate of molting. Loss of limbs was noted as a molt-accelerating factor for the land crab, Gecar- cinns lateralis (Bliss, 1956, 1959). Under conditions unfavorable for molting (dry sand, air not saturated with moisture) over a 6-month period, five out of five land crabs missing 6-8 limbs completed premolt limb regeneration and molted, while only one out of nine controls missing 1-2 limbs molted. Skinner and Graham (1970) systematically examined in Gecarcinus the effect of loss of limbs on the initiation of molting under favorable conditions. Their results showed that the loss of numerous (6-8) limbs, both walking legs and chelae, caused the organisms to enter proecdysis almost immediately. Similarly, several species of crabs, including the marsh crab, Sesarma reticu- latum (Passano and Jyssum, 1963), the green crab, Car emus maenas, the blue crab, Callinectcs sapidus, and the two fiddler crabs, Uca pitgilator and U. pugnax (Skinner and Graham, 1972), entered proecdysis and molted after induced multiple autotomy. Two recent reports have mentioned unsuccessful attempts to accelerate the rate of molting by excessive limb loss. Passano and Jyssum (1963) found that 203 204 L. A. STOFFEL AND J. H. HUBSCHMAN the removal of six walking legs failed to stimulate several large green crabs to enter procedysis and molt. Also, the spider crab, Libinia emarginata, did not respond to limb loss during a ten week observation period (Skinner and Graham, 1972). The present study was undertaken to determine how the loss of four walking legs affects the duration of the molt cycle of the freshwater shrimp, Palacmonctcs kadiakcnsis Rathbun. Previous studies on limb loss have dealt primarily with Reptantia, crustaceans with heavy well-mineralized exoskeletons and adapted for crawling. Except for Zeleny's ( 1905 ) preliminary experiment with the pistol shrimp, Alphcns, the effect of the removal of several appendages in Natantia, crustaceans lacking well-calcified exoskeletons and adapted for swimming, has not been examined. Also, in the literature surveyed, only the overall effect of the removal of append- ages on molting was noted. Since the intermolt cycle in crustaceans can be sub- divided into stages depending upon physiological and morphological changes in the organism (Drach, 1939, 1944; Passano, 1960; Stevenson, 1968, 1972; Stevenson et al., 1968), this paper represents a study of the effect of limb loss at the different stages of the molt cycle. MATERIALS AND METHODS The freshwater shrimp used in this investigation were collected from the western end of Sandusky Bay, Lake Erie. At the start of an experiment, each organism was placed in a 4V' Carolina culture dish. The shrimp were fed daily with newly-hatched Art cm-la saliva nauplii. Five groups of organisms were established depending upon the stage (A, B, C or D) of the intermolt cycle in which legs would be amputated, with the controls set up as a separate group. All the experimental organisms were held for five consecutive molts. Between the third and fourth laboratory molts, four walking legs (3rd and 5th pairs) were amputated under 10 X magnification of a dissecting microscope. Using dissecting scissors, each leg was cut between the coxa and the basis. The control organisms were left intact for the five successive molts. Staging the molt cycle in this investigation was determined according to the methods described by Passano (1960) for Natantia which were later modified for Palactnonetes kadiakcnsis (Thompson, 1964). Although the stages can be sub- divided in some detail, only the four major stages of the intermolt cycle, A, B, C and D, were considered. Stage A is the stage immediately following ecdysis and is easily distinguished by the hollow appearance of the antennal scale and uropod setae. It was at this stage that the group A shrimp had four walking legs ampu- tated following the third laboratory molt. Stage B occurs when the conical bases of the antennal scale spines are formed. Organisms in group B had four walking legs removed during stage B of their third laboratory intermolt cycle. Stage C occurs when 30% of the intermolt cycle has been completed and no sign of proecdysis is evident. The individuals in group C had four appendages amputated during stage C. Finally, stage D, proecdysis, which occupies 60% of the entire intermolt cycle, is subdivided quite extensively for natantians. The group D shrimp had the amputation of legs performed during late stage D, immediately prior to molt. This stage (D4) is recognized by the new setae appearing as rods LIMB LOSS AND THE MOLT CYCLE 205 projecting into the anteimal scales and uropods. The development of new setae, upon which recognition of these stages depends, could he seen under a dissecting microscope with 30 X magnification. The laboratory room was maintained on a strict photoperiod of 14 hours of light and 10 hours of darkness which coincided approximately with the normal outdoor photoperiod. Room temperature ranged from 20° C through 26° C, with an average of 24.5° C. RESULTS Response to limb loss In the freshwater shrimp, Palaemonetes kadiakcnsis, the removal of four walk- ing legs caused an acceleration of the molting process (Fig. 1). The average time required for 50% of each group to complete the five laboratory molts (T50) was : group A = : 45 days, group B = ; 44 days, group C : 45 days, group D = 47 days, controls =- 52 days. When all of the organisms subjected to limb loss had completed the four intermolt cycles (Tioo = 65 days), only 92% of the controls had finished and it was not until six days later that all the controls had completed the five laboratory molts (Tioo™ 71 days). It was observed that on completion of the four intermolt cycles, stage A, stage B and stage C organisms had regener- ated the four walking legs, but stage D organisms had not completely regained their limbs. Response to limb loss during stage A The amputation of four appendages immediately following a molt leads to the subsequent intermolt period being significantly shortened (Fig. 2). The 120 indi- viduals in group A had a standard mean intermolt cycle of 13.6 days prior to limb loss. The standard mean intermolt cycle is found by taking the mean of inter- molt cycles #1 and #2 ((13.4+ 13.8)/2). The mean intermolt cycle following limb loss (intermolt cycle #3) was 8.8 days which represents a decrease in intermolt cycle duration of 43.4%. The second postoperative intermolt cycle was 10.4 days. While this was longer in duration than the first, it was still reduced from the standard mean intermolt cycle by 24%. Confidence intervals for all the means were figured with a 0.99 degree of con- fidence (Table I). The confidence intervals of the means for the two postoperative intermolt cycles of stage A organisms do not overlap the two preoperative inter- molt cycle mean intervals, hence for a 99% degree of confidence the decrease in intermolt cycle duration following limb loss is significant. Response to limb loss during stage B The removal of four walking legs during stage B of the third intermolt cycle also caused a shortening of the first postoperative molt (Fig. 2). The molt cycle duration between the third and fourth laboratory molts was reduced by 27%. Also, as in group A shrimp, the second postoperative intermolt cycle was slightly longer in duration than the first postoperative intermolt cycle although it had not returned to normal. Considering that all group B shrimp had four walking legs removed on the second day following the third molt, actually 8% of the intermolt 206 L. A. STOFFEL AND T. H. HUBSCHMAN 30 35 40 45 50 TIME (DAYS) 60 65 70 FIGURE 1. Effect of the removal of four walking legs on the duration of four successive intermolt cycles of Palacmonetcs. All of the organisms were held for two consecutive inter- molt cycles before limbs were amputated. During the third intermolt cycle four legs were removed from 120 animals which were in stage A of that intermolt cycle (open triangle), from 51 animals which were in stage B of that intermolt cycle (open arc), from 51 animals which were in stage C of that intermolt cycle (open ellipse), and from 50 animals which were in stage D of that intermolt cycle (open square). The 50 control organisms (open hexagon) were left intact for the five successive molts. TABLE I Effect of limb loss on molt cycle duration Stage A B C D X* Number of organisms 120 (125) 51 (54) 51 (54) 50 (54) 50 (50) Mean length (mm) 26.3 23.7 23.7 25.0 24.7 Mean intermolt cycle (days ± S.D.) f#l pre-amputation J. „ 13.4 ± 1.9 13.8 ± 2.3 12.2 ± 1.9 12.9 ± 2.2 12.0 ± 2.0 12.5 ± 2.2 12.7 ± 1.5 12.9 ± 1.9 12.7 ± 2.2 13.1 ± 2.2 post-amputation-] (_# 4 8.8 ± 0.9 10.4 ± 1.4 9.2 ± 0.9 9.8 ± 1.4 11.1 ± 1.4 9.9 ± 1.4 13.4 ± 2.2 8.1 ± 1.1 13.3 ± 2.4 13.6 ± 2.6 Confidence # 1 0.5 0.7 0.7 0.5 0.8 intervals for #2 0.5 0.8 0.7 0.7 0.8 the means (days) #3 0.2 0.3 0.5 0.8 0.9 #4 0.3 0.5 0.5 0.4 1.0 Range (days) # 1 9-21 9-19 8-15 10-16 8-18 #2 9-19 8-17 9-17 9-16 9-19 #3 7-12 7-11 8-14 9-18 8-19 #4 8-14 7-13 8-13 6-12 9-20 * Controls; brackets indicate initial number of organisms. LIMB LOSS AND THE MOLT CYCLE 207 cycle had already been completed when limb loss occurred. Thus, the remaining intermolt cycle following amputation was 61% of a standard mean intermolt cycle. The two postoperative intermolt cycles show significant decreases from the two preoperative intermolt cycles when the confidence intervals of the means are com- pared. 20 18 16 12 10 8 1 INTERMOLT CYCLE NUMBER FIGURE 2. Plot of the means of the two preoperative and the two postoperative intermolt cycles of Palaemonetes. The group A organisms (open circle) had four walking legs removed during stage A following the third laboratory molt. The group B organisms (closed circle) had four walking legs removed during stage B of the third intermolt cycle. The group C organisms (open square) had four walking legs removed during stage C of the third intermolt cycle. The group D organisms (closed square) had four walking legs removed during stage D of the third intermolt cycle. The control organisms (open triangle) were left intact for four successive intermolt cycles. 208 L. A. STOFFEL AND J. H. HUBSCHMAN Response to limb loss during stage C Limb loss during stage C of the third intermolt cycle caused a shortening of two postoperative intermolt cycles (Fig. 2). The group C shrimp, which had a mean intermolt cycle of 12.2 days prior to limb loss, showed a slight reduction in the first postoperative molt (9%) and then a further reduction in the second post- operative molt (19%). Again, if the amount of intermolt cycle completed prior to limb removal is con- sidered, 30% of the entire intermolt cycle had passed before limbs were removed in stage C individuals and thus, the length of the intermolt cycle following amputa- tion was actually 61% of the standard mean intermolt cycle. As for the confidence intervals of the means, only the second postoperative intermolt cycle was a significant decrease. Response to limb loss during stage D When the amputation of four appendages occurred during stage D, the dura- tion of the postoperative intermolt cycle was not affected, but the second intermolt cycle following limb loss was definitely shortened (Fig. 2). The third intermolt cycle of group D organisms showed a 5% increase over the standard mean inter- molt cycle. The second postoperative intermolt cycle, however, was reduced by 37% following the removal of appendages. The decrease of the second postoperative intermolt cycle was a significant de- crease over the two preoperative intermolt cycles. Molt cycle duration in control organisms The control organisms, those organisms left intact and held for five consecutive molts, had four fairly consistent mean intermolt cycles (Fig. 2). Each molt cycle increased slightly over the previous one: 12.7 days to 13.1 days ( + 3%) to 13.3 days (+2%) to'l3.6days (+276). The confidence intervals of the means show that for a 0.99 degree of confidence there is not a significant difference among the four intermolt cycle durations. Sex as a factor in limb loss No difference was noted between males and females with respect to the response to limb loss as both sexes do show that the removal of several appendages does accelerate molting (Table II ). The only factor which seems to affect the duration of the intermolt cycle between males and females is the size of the organism as females are larger (26.0 mm) than males (23.7 mm) and therefore, as shown in Table II, have mean intermolt cycles of slightly greater duration. Mortality The mortality rate in this investigation was 3.8% (13/337). Death following the removal of four walking legs made up only a small per cent of the total mortality (0.9%), while natural deaths (prior to the third molt) made up the largest per cent of deaths (1.8%). Other reasons for not completing the five laboratory molts included accidental deaths and severe injury resulting from ecdysis. LIMB LOSS AND THE MOLT CYCLE 209 TABLE II Comparison of the effect of limb loss between male and female shrimp Males Females Stage A Number of organisms 44 76 Mean length (mm) 24.4 27.4 Mean intermolt cycle (days) #1 12.9 13.9 #2 12.4 14.6 #3 8.4 9.1 #4 9.7 10.8 Stage B Number of organisms 26 25 Mean length (mm) 23.6 23.8 Mean intermolt cycle (days) #1 12.2 12.2 #2 12.8 13.0 #3 8.9 9.6 #4 9.4 10.2 Stage C Number of organisms 22 29 Mean length (mm) 22.5 24.7 Mean intermolt cycle (days) #1 11.4 12.4 #2 11.6 13.1 #3 10.7 11.4 #4 9.4 10.3 Stage D Number of organisms 24 26 Mean length (mm) 24.0 26.0 Mean intermolt cycle (days) #1 12.5 12.9 #2 12.2 13.5 #3 12.7 14.1 #4 7.6 8.5 Controls Number of organisms 23 27 Mean length (mm) 23.4 25.8 Mean intermolt cycle (days) #1 12.3 13.2 #2 12.7 13.3 #3 13.1 13.5 #4 13.0 14.2 DISCUSSION The experimental results reported here demonstrate that the loss of four walk- ing legs induces molting in the freshwater shrimp, Palaemonetes kadiakensis. They show that regardless of when limb loss occurred during the intermolt cycle, acceleration of molting ultimately resulted. These results support similar findings for several brachyuran species, including the land crab (Skinner and Graham, 210 L. A. STOFFEL AND J. H. HUBSCHMAN 1970, 1972) ; the marsh crab (Passano and Jyssum, 1963) ; and the green crab, the blue crab and two fiddler crabs (Skinner and Graham, 1972). Consequently, for at least one species of Natantia, the response to extensive limb loss is similar to that of Reptantia. The fact that the removal of walking legs did not affect the mortality rate suggests that molts induced by extensive limb loss are normal ecdyses. Skinner and Graham's (1972) work also showed this for Gccarcimts with very low mortality rates following ecdysis after limb loss. In an attempt to explain this molt-accelerating effect of limb loss, Skinner and Graham ( 1970) discounted the most obvious possible conclusion, that is, that a molt-inhibiting hormone is present in the walking legs. They suggested that the loss of body mass could prevent the secretion of the molt-inhibiting hormone of the X-organs. However, it was shown that the loss of tissue could not be the stimulus responsible for molt initiation since the loss of two chelipeds, which make up approximately 35% of the entire body weight, in the land crab did not accelerate molting while the loss of eight walking legs, which contribute 11-17% of the total body weight, did hasten molting (Skinner and Graham, 1972). Skinner and Graham (1972, page 230) then proposed that the "severing of a critical number of nerves" is the stimulus which induces molting. A possible explanation of the results of the present study is that the loss of several walking legs stimulates the neurosecretory cells of the X-organs via nervous impulses to stop releasing (and also, perhaps, producing) the molt-inhibiting hor- mone. This would then allow the Y-organs to secrete their postulated molting hormone, and premolt, stage D of the intermolt cycle, would be initiated. In sup- port of this hypothesis it should be noted that the length of the intermolt cycle following limb loss in three stages. A, B and C, closely approximates 60% of the previous intermolt cycle, which is considered the per cent occupied by proecdysis. The effect of limb loss on the molt cycle duration of the first postoperative intermolt cycle in group D individuals was not noted. This is to be expected since the orga- nisms were already in stage D, proecdysis, when amputation of four walking legs occurred. However, if the duration of the second postoperative molt is com- pared to the previous intermolt cycle, it only amounts to 60.4% of that intermolt period suggesting that excessive limb loss stimulates the organism to enter premolt as soon as possible following the molt. This hypothesis that stage D is initiated immediately following limb loss seems reasonable, but since molting is a cyclic event, it would seem that stages A, B and C of the intermolt cycle could not be omitted entirely. Hence, the shortening of the intermolt cycle following the removal of appendages could be the result of stages A, B and C being shortened proportionately with the animal molting in the absolute minimal time. Skinner and Graham (1972) did find that eyestalk removal caused land crabs to molt sooner than crabs stimulated to molt by limb removal. However, death usually occurred prior to or at ecdysis in eyestalkless crabs and not in crabs missing appendages. This suggests possibly that the removal of limbs, while stimulating the X-organs to stop releasing MIH, allows the change in concentration from molt-inhibiting hormone to molting hormone to be gradual enough for the organism to hasten through all the stages of the inter- molt cycle prior to premolt and thus, a normal molt could occur. LIMB LOSS AND THE MOLT CYCLE 211 Some explanation can be given for the reports which have noted that the loss of limbs has no effect on molt cycle duration. The most recent mention of inhibition of molting in crustaceans by extensive appendage loss was that the loss of all ten legs in Gecarcinns inhibited molting rather than accelerated it (Skinner and Graham, 1972). These crabs, which had lost their limbs in combat with other crabs, showed no sign of regeneration after 90 days in the laboratory. Starvation, which sometimes inhibits molting (Passano, 1960), was not a factor in this situation since the crabs were hand-fed. As for multiple autotomy failing to induce molting in the green crab, Carcintis maenas, Passano and Jyssum (1963) indicated that these crabs were perhaps in terminal anecdysis (stage C4T), when molting and growth in the animals have ceased. They described the crabs that they subjected to limb removal as large males and found no signs of approaching molt when they were kept in the labora- tory for several months. Skinner and Graham (1972) did find that limb loss in- duced molting in four green crabs, but they did not mention the size of the individuals. These results suggest possibly that prior to terminal anecdysis, the removal of several appendages can stimulate molting in Carcimis maenas, but once the crab has ceased to molt, extensive limb loss cannot break terminal anecdysis. Finally, no molting occurred in the spicier crab, Libinia emarginata, following the removal of 6-8 walking legs. None of the individuals, including eyestalkless and control organisms, molted in the laboratory during the ten week observation period (Skinner and Graham, 1972). Since the organisms were held in com- munity tanks during the experiment, it was suggested that perhaps in this organism, as in Gecarcinus (Bliss and Boyer, 1964), privacy at the time of molt is a "critical factor" which prevents crabs from molting in the presence of another crab. In this sense, molting would be prevented by the lack of privacy overriding any stimulus to molt from limb loss. SUMMARY 1. The loss of four walking legs induces molting in the freshwater shrimp. Palaemonetes kadiakcnsis. This study shows that at least in one species of Natantia, excessive limb loss accelerates the molting process as has been shown in several reptantian species. 2. Removal of four appendages (two pairs of walking legs) during the four major stages of the molt cycle, A, B, C and D, produced varying results. When the appendages were removed during stage A or stage B of the intermolt cycle, the first molt following limb loss was accelerated. When limbs were removed during stage C of the molt cycle, two successive molts following limb loss were accelerated. Finally, when limb removal occurred during stage D of the intermolt cycle, the first postoperative molt was not affected, but the second postoperative intermolt cycle was shortened significantly. 3. The sex of the organism did not affect the response to the loss of numerous appendages. 4. Mortality rates due to the loss of four walking legs were negligible. 5. A hypothesis was proposed to explain this molt-accelerating effect of limb loss. 212 L. A. STOFFEL AND J. H. HUBSCHMAN LITERATURE CITED BLISS, D. E., 1956. Neurosecretion and the control of growth in a decapod crustacean. Pages 56-75 in K. G. Wingstrand, Ed., Bcrtil Hanstroin: Zoological Papers in Honour of his Sixty-fifth Birthday, November 20, 1956. Zoological Institute, Lund, Sweden. BLISS, D. E., 1959. Factors controlling regeneration of legs and molting in land crabs. Pages 131-164 in F. L. Campbell, Ed., Physiology of Insect Development. University of Chicago Press, Chicago. BLISS, D. E., AND J. R. BOYER, 1964. Environmental regulation of growth in the decapod crustacean, Gecarcinus latcralis. Gen. and Comp. Endocrinol., 4: 15-41. DRACH, P., 1939. Mue et cycle d'intermue chez les Crustaces Decapodes. Ann. Inst. Occanogr. Monaco, 19: 103-391. DRACH, P., 1944. Etude preliminaire sur le cycle d'intermue et son conditionnement hormonal chez Leander serratus (Pennant). Bull. Bio]. France Bclg., 78: 40-62. Goss, R. J., 1969. Principles of Regeneration. Academic Press, New York and London, 287 pages. PASSANO, L. M.t 1960. Molting and its control. Pages 473-536 in T. H. Waterman, Ed., The Physiology of Crustacea. I. Metabolism and Growth. Academic Press, New York and London. PASSANO, L. M., AND S. JYSSUM, 1963. The role of the Y-organ in crab proecdysis and limb regeneration. Comp. Biochem. Physiol.. 9: 195-213. SKINNER, D. M., AND D. E. GRAHAM, 1970. Molting in land crabs: stimulation by leg re- moval. Science. 169 : 383-385. SKINNER, D. M., AND D. E. GRAHAM, 1972. Loss of limbs as a stimulus to ecdysis in Brachyura (true crabs). Biol. Bull., 143: 222-233. STEVENSON, J. R., 1968. Metecdysial molt staging and changes in the cuticle in the crayfish Orconcctcs sanborni (Faxon). Crustaceana, 14: 169-177. STEVENSON, J. R., 1972. Changing activities of the crustacean epidermis during the molting cycle. Amcr. Zool., 12 : 373-380. STEVENSON, J. R., R. H. GUCKERT AND J. D. COHEN, 1968. Lack of correlation of some proecdysial growth and developmental processes in the crayfish. Biol. Bull., 134: 160-175. THOMPSON, E. M., 1964. Gross external changes of the adult freshwater prawn, Palacinonetcs kadiakensis, during the molting cycle. M.S. thesis, Ohio State University, 18 pages. ZELENY, C, 1905. Compensatory regulation. /. E.vf. Zool., 2: 1-102. Reference: Biol Bull, 147: 213-226. (August, 1974) BUDDING, SEXUAL REPRODUCTION, AND DEGENERATION IN THE COLONIAL ASCIDIAN, SYMPLEGMA REPTANS1 KEIJI SUGIMOTO AND MITSUAKI NAKAUCHI Zoological Institute, Faculty of Science, Tokyo Kyoiku University, Bunkyo-ku, Tokyo 112 and Department of Biology, Kochi University, Asakura, Kochi 780, Japan Ascidians of the subfamily Botryllinae propagate asexually by pallial budding and, in some genera, by vascular budding (Berrill, 1950; Abbott, 1953; Oka and Watanabe, 1957, 1959; Milkman, 1967). Since Bancroft's pioneer work (1903), many developmental problems have been studied in botryllid ascidians, but most of the studies have been made on Botryllus or Botrylloides, so our knowledge of the budding in other synstyelid ascidians is very limited. Berrill (1940) described the mode of the budding of Symplegma viride, and his work is the only earlier study of budding in this genus. Symplegma reptans is one of the common synstyelid ascidians in Japanese waters, forming thin, flat colonies. This species is easily cultured on glass plates, providing colonies that are easy to handle and to operate upon, and are clearly observable from both sides. In these points Symplegma is much like Botryllus, but it has additional features which are desirable for some types of study : ( 1 ) the zooids of Symplegma are "independent," each zooid possessing its own individual atrial opening; (2) the life span of each zooid is much longer than that of Botryllus; and (3) Symplegma reproduces sexually throughout the year. Employing this species a series of descriptive and experimental studies have been carried out by the present authors. In this paper we present our observations on the life history of the species, and give special attention to the quantitative and dynamic aspects of budding. This and coming papers will show that 5". reptans is excellent material for various kinds of study. MATERIAL AND METHODS For a taxonomic description of Symplegma reptans (Oka) see Tokioka (1949). Studies were carried out at the Usa Marine Biological Station of Kochi Uni- versity, from March to May, 1969, at a sea water temperature of 15°-21° C, and at the Shimoda Marine Biological Station of the Tokyo Kyoiku University, from May to July, 1972, at a sea water temperature of 18°-24° C. Colonies of Symplegma collected in the field were brought back to the laboratory with their original substrata. Here the growing margin of the colony was detached, and fragments of this cut by a razor blade were fastened by thread to glass plates and allowed to attach and grow. Oozooids were obtained by placing cultured colonies containing mature zooids in a bowl, and collecting the larvae released from them with a pipette. Then the tadpoles were settled at desired positions on a glass plate. Glass plates bearing fragments of colonies or settled larvae were 1 Contribution no. 234 from the Shimoda Marine Biological Station. 213 214 K. SUGIMOTO AND M. NAKAUCHI FIGURE. 1. Development of stigmata in the oozooid. (A) 5-day oozooid with 7 pairs of protostigmata, ventral view. (B) 20-day oozooid with 2 rows of stigmata and 8 protostigmata on one side of the branchial basket, ventral view. cultured in a slidebox set in the bay. Materials were brought back daily to the laboratory, and repeated observations on particular zooids were made there. For histological study the materials were fixed in cold 5% phosphate-buffered glutaraldehyde at pH 7.4 for 3 hours and embedded in Epon. The blocks were sectioned serially at 1 //, and stained with 1% toluidin blue. OBSERVATIONS Development, bud formation, and degeneration of the oosooid In the laboratory most of the larvae are released during the morning hours (10-12 A.M.), and begin to metamorphose within 4-5 hours after release. At a temperature of about 15° C, heart-beat begins 2 days after attachment of larva, and 1-2 days later the apertures perforate and feeding begins. By this stage the body of the zooid is nearly transparent owing to consumption of the yolk with which the newly metamorphosed zooid is filled. The first bud is formed on the oozooid about 10 days after metamorphosis. At a temperature of 20°-23° C, the time schedule is speeded up : it takes 1-2 days after metamorphosis for perfora- tion of the apertures, and only 3 days after metamorphosis the first bud is produced. The oozoid continues to grow and gives off a series of 4—7 buds during its life span. In about 30 days after metamorphosis the oozooid begins to degenerate. The time of degeneration is little affected by temperature. Colonies grow faster LIFE HISTORY OF SYMPLEGMA 215 under warmer conditions, so the size of the colony at the time degeneration of the oozoid occurs varies with the temperature. Degeneration occurs more or less suddenly. It is not preceded by a decrease in the growth rate or by a slowdown of the heart-beat, both of which are observed in the degeneration of blastozooids. A young functional oozooid has 7-9 protostigmata on each side of the pharynx (Fig. 1A). The anterior 3-4 protostigmata are subdivided successively from anterior to posterior during the growth of the oozooid (Fig. IB. By the time degeneration occurs the subdivision of the anterior 3-4 pairs is completed, but the posterior 3-5 ones remain undivided. The oozooid thus has at least 3 pairs of protostigmata throughout its life. The blastozooids produced by successive buddings of an oozooid spread out radially and surround the oozooid. These zooids in turn produce buds, so at the time of degeneration the oozooid is situated at the center of the colony composed of 70-80 zooids at various stages of development. No gonad is formed in the oozooid before it degenerates. Even in oozooids which were surrounded by blastozooids with visible gonads no sign of gonad formation was observed. Development and bud formation of the blastozooid As stated above, the first bud, or first blastozooid, of an oozooid is formed about 10 days after metamorphosis at about 15° C. The bud arises from a definite site on the antero-ventral wall of the right atrial chamber. This site is very close to the base of a test vessel (Fig. 2A). The second and the subsequent buds are also formed in the same region. The epidermis and atrial epithelium of this region thicken and form a double-layered hemispheric vesicle about 50 ^ in diameter at this stage. With further development the bud becomes nearly spherical. The inner layer of the bud then detaches from the parental atrial epithelium whereas the outer layer remains continuous with the parent wall by a broad stalk through which the blood flows (Fig. 2B). Within 24 hours after the formation of a hemi- sphere vesicle the stalk is broken. However, a new test vessel arises from the anterior wall of the bud (Fig. 2C) ; this fuses with a test vessel of the parent, so the developing bud is again connected with the parental circulatory system. The developing bud, or blastozooid, forms its own first bud at a very early stage of development, when it is only 3-4 days old (17-20° C) and is not yet functional (Fig. 2D). At this stage the developing zooid is about 0.3 mm in length and 0.26 mm in width. The heart becomes functional when the developing zooid is 6-7 days old (17°-20° C) and has reached an average size of 0.39 X 0.29 mm (Fig. 2E). One day after the heart begins to beat the apertures become perforated (Fig. 2F). After this the zooids feed actively and grow quickly (Fig .3). Each developing zooid gives rise to a series of 3-6 buds during the period of development and growth (Fig. 4). The budding phase of a blastozooid comes to an end at about the time a pair of gonads appear clearly. At this stage the zooid measures about 1.7 mm in length, and 1.3 mm in width. The oocyte of this stage is about 35 ^ in diameter. As a rule, no bud formation occurs after this stage, and the gonad develops quickly. In other words, the budding, or growing phase gives place to the sexual reproduction phase. It takes, on an average, 216 K. SUGIMOTO AND M. NAKAUCHI *&:*•' / 0.1 mm 0.2mm FIGURE 2. Development of a bud arising from an oozooid. (A) Appearance of bud vesicle. (B) Bud vesicle with a broad test vessel. (C) 1-day bud with a secondary test vessel. (D) LIFE HISTORY OF SYMPLEGMA 217 2.5 2.0 1.0 0.5 blostozooid 10 20 30 Days FIGURE 3. Growth of an average oozooid and an average blastozooid. Abbreviations are the same as Figure 2. 14.6 days (18°-24° C) to reach this stage after the "birth" of a blastozooid as a bud rudiment (Figs. 5, 6). The buds formed on the wall of a developing blastozooid develop in the same manner as the buds formed on an oozooid. No significant differences have been found between the time schedule of development of buds derived from an oozooid and that of buds derived from a blastozooid. Blastozooids derived from blast- ozooids also produce a succession of 3-6 buds during their budding phase. They stop bud formation on an average of 12.8 days after their birth, and enter the sexual phase. Buds and young zooids spread out to the periphery of the colony, and older zooids remain in the central region. A large colony thus consists of two kinds of zooids with respect to reproductive activity ; zooids in sexual phase which occupy the central region of the colony and zooids in the asexual phase occupying the periphery. The intervals of the successive buddings in each oozooid and in each blastozooid are summarized in Table I. At about 17° C, the first budding of the oozooid takes place later than at about 20° C. However, the intervals between successive 3-day bud in which the first bud of the next generation is being formed. (E) 6-day bud in which weak heart-beat is seen. (F) 8-day blastozooid which has begun feeding. Arrows indicate gonadial rudiment; bl (b2), first (second) bud; br, thickening of peribranchial wall as a bud rudiment ; sec. tv, secondary test vessel ; st, stomach ; tv, test vessel. 218 K. SUGIMOTO AND M. NAKAUCHI 0 I 234567 No. of buds FIGURE 4. Variation in the number of buds produced per blastozooid. buddings in oozooids at 20° C is almost same as that in blastozooids at 20° C; in both, budding occurs every 2-3 days (Fig. 7). The 6th and 7th buddings tend to occur after a longer interval than earlier ones. The blastozooids surrounding an oozooid sometimes close their budding phase before the oozooid begins to degenerate. The buds produced by old oozooids and aged blastozooids tend to become abortive. Sexual reproduction and degeneration of the blastosooid The oozooid reproduces only asexually during its life span. The blastozooid reproduces asexually until the gonads have developed to the stage where a pair of tn 7 T3 o 6 o M 5 "•V- 0 4 0 z 3 2 1 0 15 19 Days FIGURE 5. Age in days of blastozooids at the time of formation of the last bud. LIFE HISTORY OF SYMPLEGMA 219 tv3 O.5 mm • tv1 tv2 FIGURE 6. The 13-day blastozooid which has just stopped bud formation; end, endostyle ; g, gonad; h, heart; int, intestine; st, stomach; tv(l-3), test vessel. TABLE I Average intervals between tivo successive buddings in each zooid (15°-21°C) Oozooid Blastozooid Metamorphosis to 1st budding Blastozooid "birth" to 1st budding lst-2nd budding 2nd-3rd budding 3rd-4th budding 4th-5th budding 5th-6th budding 6th-7th budding 10.8 (days) 2.3 2.5 4.0 2.8 4.0 8.0 4.1 (days) 2.2 2.2 2.5 3.0 4.0 220 K. SUGIMOTO AND M. NAKAUCHI • .' 3 Days Colony A K . . M "nlnnv R L L_ D I- 1_ U t. L L- 1_ U L? C L L L • L LULL? L M Colony C L L D •— L L L • L_ l-I_ L 1— * \- L_ [_ l_ * L FIGURE 7. Genealogies of three colonies originating from oozooids. Nomenclature of the zooids is taken from Berrill (1940). The end of lines indicates the end of observation of the zooids ; D, degeneration of the zooid ; M, metamorphosis ; solid circle, appearance of gonads. testicular follicles and an ovarian follicle are clearly observable under a microscope in living materials. The sexual phase starts with this stage, but the gonad rudiment appears much earlier. It appears as a mass of lymphocytic cells in the genital tracts around the time when the first bud is formed in the developing blastozooid (Fig. 8A). In living materials so many blood cells are found in the genital tracts that it is impossible to identify the gonadial rudiment, but in fixed material identification is easier. The gonadial rudiment grows as the zooid grows (Figs. 8B-8D). As shown in Figures 9A-9D, the eldest oocyte enters to quick-growing phase about 20 days (18-24 C) after the birth of the blastozooid. In about 5 days the grow- ing oocyte reaches its maximum size, 220 ^ in diameter. LIFE HISTORY OF SYMPLEGMA 221 0.1 mm A 20/J D FIGURE 8. Formation of the gonadial rudiment by an aggregation of lymphocytic cells ; (A) frontal section of a 4-day blastozooid, showing the position of the gonadial rudiment; (B) gonadial rudiment of a 4-day blastozooid; (C) longitudinal section of a 5-day blastozooid; (D) gonad of an 11 -day blastozooid; be, branchial cavity; gr, gonadial rudiment; o, ovary; pc, peribranchial cavity ; st, stomach ; t, testis. 222 K. SUGIMOTO AND M. NAKAUCHI * • 4 0.1 mm *. * B D FIGURE 9. Development of the gonad ; (A) 11-day zooid ; (B) 15-day zooid ; (C) 24-day zooid ; (D) 28-day zooid; FE, fertilized egg; ME, mature egg, O, ovary; S, brood sac; T, testis. LIFE HISTORY OF SYMPLEGMA 223 With enlargement of the oocytes, the shape of the zooid changes from ellipsoidal to triangular. In the anterior region on both sides of the body the epidermal and peribranchial wall protrude laterally, and the anterior part of each peri- branchial cavity expands laterally. A brood sac derived from the ovary protrudes into each expanded cavity (Fig. 10A). After the eggs have reached maximum size they are ovulated in the brood sac where fertilization occurs and development of the tadpole larvae takes place. In the early stage of gonad development, 6-7 oocytes are seen in the ovary. Before these oocytes reach their maximum size, new young oocytes appear. Fur- ther oocytes also appear as the enlarged ova are ovulated. So, in the aged zooid, 5-7 oocytes of various stages and about 3 ova are seen in the ovary. Sometimes, the larva developing in a brood sac moves the tail actively. At a temperature of 18°-24° C, it takes 30 days from the birth of the blastozooid, and 5 days from fertilization, to the liberation of the first larva. Some of the blastozooids have been observed to degenerate in 60-70 davs after their birth. In these zooids the internal organs except for the heart begin to disintegrate, and the zooids become transparent (Fig. 10B). These zooids con- tain a few larvae or ripening eggs, but before degeneration is complete all the larvae are released. From the number of cases oberved we cannot say that the life span of the blastozooid is limited to 60-70 days, but it is clear that blastozooids live for 60 days or more, and that at least some of them degenerate before the death or degeneration of the whole "colony." Time schedules of development for an average oozooid and an average blastozooid are sho\vn in Figure 11. DISCUSSION While the mode of budding of 5. rcptans is shown to be fundamentally identical with that of S. viride as reported by Berrill (1940), several new facts are revealed by the present study. The time schedule of budding and degeneration of zooids were previously unknown, and the development of the buds arising from the oozooid was "described only partially" by Berrill (1940). Repeated observations on particular oozooids in the present study reveals that the oozooid makes its first FIGURE 10. (A) Fully mature 35-day blastozooid with larvae; (B) Degenerating 65-day blastozooid in which all organs but the heart have disintegrated ; g, gonad ; lar, larva. 224 K. SUGIMOTO AND M. NAKAUCHI OOZOOID (Days) 0 attachment of larva BLASTOZOOID appearance as bud rudiment (Days) _ 0 10- 20- 30- heartbeat feeding first bud formation degeneration re •O O o. detachment from mother zooid first bud formation heartbeat feeding end of bud formation growth and maturation of oocytei maturation of the eldest egg fertilization liberation of 1st larva degeneration -10 -20 -30 -70 FIGURE 11. Time schedules of development in average oozooids and blastozooids at about 20° C. bud about 10 days (17°-20° C) after metamorphosis and thenceforth it con- tinues to produce buds at intervals of 2-8 days until it degenerates. The oozooid produces 4—7 buds during its life span of about 30 days. It is also shown that in oozooids neither complete subdivision of the proto- stigmata nor gonad formation takes place before degeneration begins, so the oozooid is distinguishable from the blastozooids in the same colony by these morphological characters. It is of interest that, likewise, no gonad is formed in the oozooid of the synstyelid Metandrocarpa taylori (Watanabe, 1970) or in that of the pelagic tunicate Doliolum (Berrill, 1950). As to the number of the buds produced by each blastozooid, Berrill (1940) states that in S. viridc, "As far as can be determined, this (fourth bud) represents the last of the bud series." The present study shows that in 5. reptans the fourth bud is not always the last bud, and that the number of buds produced by a developing blastozooid varies from 3 to 6. LIFE HISTORY OF SYMPLEGMA In 5". reptans the budding phase of a blastozooid comes to an end about the time when a pair of gonads becomes clearly visible, and as a rule no budding takes place in mature zooids. However, the rudiment of gonad is shown to be formed early in the asexual phase, so here the sexual phase is not so clearly segregated from the asexual one as in Hydra (Loomis and Lenhoff, 1956; Loomis, 1959) or in the hydroid Podocorync (Braverman, 1962). At an early stage of gonad formation the lymphocytes were found to aggregate in the genital tracts. The same phenomenon was seen in Distomas variolosus (Newberry, 1968) and in Botryllus primigenus (Mukai and Watanabe, 1972). These observations suggest that lymphocytes are usually involved in gonad forma- tion in synstyelid ascidians. One of the most important problems is whether or not these lymphocytes are on the so-called germ cell line. Growing oocytes at various stages have been observed in the common blood system of Botryllus (Berrill, 1941 ; Izzard, 1968; Mukai and Watanabe, 1972), but such oocytes are not found in the test vessels of S. reptans. As far as the microscopical sections show, germ cells seem to arise directly from the mass of the lymphocytes. The lymphocytes of ascidians are also known to play an important role in the budding of Botryllus (Oka and Watanabe, 1957; Milkman, 1967), Botrylloides (Oka and Watanabe, 1959), and Perophora (Freeman, 1964). These facts call for further studies on the developmental capacities of ascidian lymphocytes. It would also be desirable to trace the origin of the lymphocytes found in the genital tracts, to see whether the germ cells of ascidians are determined in an early stage of development as in mammals and insects. As to the life spans of the zooids of Symplegma, Berrill (1940) states, "unlike Botryllus, the constituent zooids of a colony of Symplegma persist while the colony lives and are not brief transients." Our observations show that oozooids of 6". reptans degenerate abut 30 days after metamorphosis, while the blastozooids of the same colony remain healthy. The oozooid of Aplidium calijornicum also de- generates in 23-32 days after metamorphosis, whereas all the blastozooids of the same colony are very active (Nakauchi, unpublished). The factors governing the degeneration of particular individuals in a Symplegma colony are being investigated. The authors wish to express their thanks to Professor H. Watanabe of Tokyo Kyoiku University for his valuable advice and encouragement during this study, and to Professor D. P. Abbott of the Hopkins Marine Station of Stanford Uni- versity for his critical reading of the manuscript. The authors are also grateful to the staff of the Usa Marine Biological Station and the Shimoda Marine Biological Station for their kindness in providing many conveniences throughout the course of the study. SUMMARY The life history of Symplegma reptans was studied by repeated observations on particular living animals and by observations on sectioned materials. The oozooid produces its first bud about 10 days (17°-20° C) after meta- morphosis, and produces a series of 4-7 buds by the time the oozooid begins to degenerate. The degeneration occurs about 30 days after metamorphosis. No 226 K. SUGIMOTO AND M. NAKAUCHI gonacl is formed in the oozooid even at the time when gonads are seen in the blastozooids of the same colony. Of the 7-9 protostigmata present in the newly metamorphosed oozooid, the anterior 3-4 are subdivided by the time of degenera- tion, but the posterior 3-5 remain undivided throughout the life of the oozooid. A blastozooid makes its own first bud 3-4 days (17°-20° C) after its first appearance, when it is still scarcely more than "bud" itself. The growing blast- ozooid makes a succession of 3-6 buds at intervals of 2-4 days. Bud formation by a blastozooid come to an end about the time its gonad become macroscopically visible. Blastozooids live for 60 days or more, but at least some of them degenerate before the death or degeneration of the colony. Microscopical sections suggest that the germ cells are formed directly from lymphocytes. LITERATURE CITED ABBOTT, D. P., 1953. Asexual reproduction in the colonial ascidian Metandrocarpa faylori Huntsman. Unir. Calif. Publ. Zool., 61 : 1-78. BANCROFT, F. W., 1903. Variation and fusion of colonies in compound ascidians. Proc. Calif. Acad. Sci. (Sci. 3), 3: 137-186. BERRILL, N. J., 1940. The development of a colonial organism : Symplegma riridc. Biol. Bull. ,79: 272-281. BERRILL, N. J., 1941. The development of the bud in Botryllus. Biol. Bull, 80: 169-184. BERRILL, N. J., 1950. The Tunicata with an Account of the British Species. Ray Society, London. 354 pp. BRAVERMAN, M. H., 1962. Studies in hydroid differentiation. I. Podocoryne carnca, culture methods and dioxide induced sexuality. Exp. Cell Res., 26 : 301-306. FREEMAN", G., 1964. The role of the blood cells in the process of asexual reproduction in the tunicate Perophora viridis. J. Exp. Zool., 156 : 157-183. IZZARD, C. S., 1968. Migration of germ cells through successive generations of pallia! buds in Botryllus schlosscri. Biol. Bull., 135 : 424. LOOMIS, W. F., 1959. Feedback control of growth and differentiation by carbon dioxide ten- sion and related metabolic variables. Pages 253-293 in D. Rudnick, Ed., Cell Organisms and Milieu. The Ronald Press Co., New York. 326 pp. LOOMIS, W. F., AND H. M. LENHOFF, 1956. Growth and sexual differentiation of Hydra in mass culture. /. Exp. Zool., 132 : 555-573. MILKMAN, R., 1967. Genetic and developmental studies on Botryllus schlosscri. Biol. Bull., 132: 229-243. MUKAI, H., AND H. WATANABE, 1972. Gametogenesis in Botryllus primigcnus. Zool. Mag. (Tokyo} (in Japanese), 81 : 268. NEWBERRY, A. T., 1968. The gonads and sexual cycle of the polystyelid ascidian Distomus variolosus Gaertner. J. Morphol., 126: 123-162. OKA, H., AND H. WATANABE, 1957. Vascular budding, a new type of budding in Botryllus. Biol. Bull., 112:225-240. OKA, H., AND H. WATANABE, 1959. Vascular budding in BotrvUoidcs. Biol. Bull, 177: 340- 346. TOKIOKA, T., 1949. Contribution to Japanese ascidian fauna II. Notes on some ascidians col- lected chiefly along the coast of Kii peninsula. Publ. Seto Mar. Biol. Lab., 1 : 39-64. WATANABE, H., 1970. Asexual reproduction in compound ascidians (I). Zool. Mag. (Tokyo) (in Japanese), 79: 131-143. Reference : Blol. Bull, 147: 227-235. (August, 1974) FOOD-RESOURCE PARTITIONING IN THE DEPOSIT FEEDING POLYCHAETE PECTIN ARIA GOULDII ROBERT B. WHITLATCH The University of Chicago, Committee on Evolutionary Biology, 5734 South Ellis Avenue, Chicago, Illinois 60637 and the Marine Biological Laboratory, Woods Hole, Massachusetts 02543 Although many benthic marine environments often support large populations of deposit feeding invertebrates, few studies have examined food-resource partition- ing in this group of animals. Most studies dealing with different modes of feeding concentrate primarily upon the morphological characteristics of feeding structure (see Yonge, 1928). Sanders, Goudsmit, Mills and Hampson's (1962) study of the deposit feeders associated with the intertidal sand flats of Barnstable Harbor, Massachusetts, was one of the first attempts to categorize different types of ma- terial actually being ingested by detritovores. One of their major conclusions from the examination of the gut contents of 36 species of invertebrates was that most species exhibited generalistic feeding habits. Other studies (e.g., Whitlatch, 1972) have shown some deposit feeders to be highly selective in their choice of particles. The importance of deposit feeders in marine benthic communities is demon- strated by studies dealing with their effects on sediment stability and trophic struc- ture (Rhoads and Young, 1970), distributions of other benthic organisms (Rhoads and Young, 1971) and water turbidity and nutrient recycling (Rhoads, 1973). Gordon (1966) has shown that Pectinaria gouldii (Verrill, 1873), a de- posit feeding polychaete, has an annual sediment working rate of about 600 gm of sediment/worm. Using averages of 10 and 40 worms/m2, he concluded that the sediment of one square meter and 6 cm deep could be completely reworked in 15 and 6 years, respectively. Because of the high rates of sediment working, this polychaete could have many effects upon the benthic environment. In the present study, P. gouldii was examined to ascertain feeding selectivity and the resultant effects upon the benthic community. METHODS Specimens of P. gouldii were collected at low tide from Little Sippewisset salt marsh, a shallow tidal enbayment located on the eastern shore of Buzzards Bay, north of Woods Hole, Cape Cod, Massachusetts (41°34'30", 70°38'20"). The exact position of individual worms in the flat was determined by gently sliding a hand across the sediment ; the tubes of the worms could easily be located in this manner. Twenty centimeter diameter cores were then taken and the sediment was removed with the worms in place. The sediment cores were brought to the labora- tory and cooled at 4° C until time of dissection. The cores were vertically dissected (always within two hours of collection), and the worms were removed 227 ROBERT B. WHITLATCH and immediately frozen. A sediment sample was taken at the level of the mouth and preserved in 70% alcohol. The sediment sampled near the mouths of the worms was analyzed to deter- mine particle size, percentage particle type abundance, and percentage organic material available to the organisms. Two subsamples of the sediment were stained with histological stains. Mercuric bromphenol blue was used for staining protein- containing material (Mazia, Brewer and Alfert, 1953). Humason (1967) states that this method appears to stain most proteins and peptides. Periodic acid Schiff reagent was used for staining carbohydrates. Humason (1967) mentions that this technique will stain a variety of substances including most protein- carbohydrate complexes. Sudan black B used for staining lipid material (Huma- son, 1967) was discontinued after preliminary staining of the sediment showed no reaction. The stained sediment was mounted on a slide and twenty random micro- scopic fields were recorded to determine the percentage of stained and unstained material. The particle size distribution of the samples was obtained by measuring the first 300 particles encountered on the slide. Individuals of P. gonldii were removed from their tubes and sediment was col- lected from the foregut (as near the mouth as possible) and the intestine (after the digestive gland) with a modified Pasteur pipette. After the gut fractions were washed several times in distilled water to remove mucus, they were preserved in 70% alcohol. The procedures outlined above for the sediment fractions were em- ployed to determine abundance of particle types and size distributions for these samples. The sediment and gut contents were examined at 200 X using a bright- field microscope with a calibrated eye-piece micrometer. The results of the analysis of abundance of different particle types are expressed as percentage particle abundance. This measure is more important than volumetric and weight estimates in feeding studies since it reflects the relative amount of dif- ferent types of particles an organism encounters while feeding. The measure, therefore, is valuable in determining food selection of different fractions of the environmental food-complex by the organisms. The statistic used to determine feeding selectivity is described by Ivlev (1961) and is calculated as E' : : (r} — pi)/(rj + pi). For the ith food type, r; equals the percentage ingested by the predator and pi is the percentage of that food type avail- able in the environment. The coefficient is finitely bounded and evenly distributed about zero. Zero indicates non-selective feeding; values from -1 to 0 indicate avoidance ; and values of 0 to +1 indicate feeding preference. Additional worms were brought to the laboratory for studies dealing with feed- ing habits. Fresh worms and sediment were collected and placed into 50 ml beakers. Observations of the feeding behavior were recorded using a vertically mounted dissecting microscope. RESULTS Feeding habits P. gouldii, which lives in a tapered tube composed of a single layer of large mineral grains, is normally positioned obliquely below the surface of the sediment. The head of the worm, located at the lower and larger end of the tube, is 1-5 cm FOOD SELECTION IN A DEPOSIT FEEDER 229 Q. 0> -o CD 60 50 40 30 20 10 o o o o o o o o 50 20 30 40 Body Length (mm) FIGURE 1. Relationship between body length of Pectinaria gonldii and depth of feeding. below the surface. A highly significant relationship (r := 0.90, F<0.05) exists between the length of P. gonldii and the depth at which the worm feeds (Fig. 1 ). Smaller individuals are found feeding closer to the surface than larger individuals. P. gonldii has long golden paleae used strictly for digging and numerous long, ciliated, grooved tentacles which bring sediment to the mouth. When feeding, the tentacles independently select particles from all areas near the mouth. The selected particles are brought to the mouth and rejected particles fall to the bottom of a feeding cavern which the worm creates while feeding. The selected sediment is either ingested or transported between the worm's body and tube to the surface of the flat where it is deposited around the posterior end of the tube. As the worm slowly moves through the sediment, the small excavated cavity continually collapses and fills in with sediment from the sides. Previous ecological studies on the genus Pectinaria include Gordon's (1966) work on the sediment reworking activities of P. gonldii and Waston's (1927) description of the natural history of P. koreni, a European species. Particle composition of the environmental food-complex The particle composition of the sediment obtained near the mouths of fifteen specimens of P. goitldii is presented in Table I. The classification of different possible food sources available to the polychaete are as follows: (a) Encrusted minerals — mineral grains encrusted with living matter, organic debris, or other encrustations. Encrusted material was determined by visual observations. Steele and Baird (1968) have found that nearly all of the organic component in a sandy 230 ROBERT B. WHITLATCH TABLE I Per cent particle type abundance of sediment sampled near the mouth of Pectinaria gouldii Particle type A B C D E F G H I J K L M N 0 Mineral >75 n encrusted 3.8 6.3 13.4 13.2 9.2 7.5 7.7 1.3 3.7 0.6 0.9 5.9 4.4 6.7 7.0 Mineral >75 ^ not encrusted 5.6 8.7 28.9 16.0 12.3 5.6 13.5 2.1 1.8 1.4 2.3 9.8 6.7 9.4 13.9 Mineral 25-75 n encrusted 0.0 0.0 0.5 0.0 0.8 0.0 0.0 0.3 0.5 0.2 0.1 0.4 0.2 0.1 0.4 Mineral 25-75 M not encrusted 9.3 10.8 6.7 8.3 7.9 7.6 8.6 6.2 7.4 5.7 6.1 14.0 12.4 9.7 10.1 Mineral <25 M 44.6 41.9 32.5 44.1 41.9 53.4 50.4 52.5 50.7 53.9 49.9 46.2 45.8 49.7 41.9 Floe aggregates 31.5 27.0 16.6 14.0 19.3 21.3 15.9 30.2 32.8 33.3 31.5 18.8 25.7 19.5 22.4 Fecal fragments 3.0 3.2 1.1 1.8 0.6 2.9 2.3 2.2 1.7 1.8 3.9 3.7 3.3 0.8 2.5 Plant fragments 0.9 0.7 0.0 0.3 0.4 0.5 0.0 0.5 0.2 0.6 0.7 0.3 0.5 0.5 0.8 Dead diatoms 0.8 0.8 0.3 1.8 0.8 0.9 1.5 2.5 0.3 1.7 2.7 0.4 0.4 1.0 0.9 Others 0.5 0.6 0.0 0.5 1.6 0.3 0.1 2.2 0.9 0.8 1.9 0.5 0.6 2.6 0.1 beach existed as encrustations upon sand grains. This includes bacteria and other micro-organisms described by Anderson and Meadows (1969) and Batoosingh and Anthony (1971). It seems likely, therefore, that encrusted mineral grains could be an important food source for deposit feeding organisms. (b) Fecal material — fecal pellets and fragments found in the sediment. The importance of fecal material as a food source for invertebrates has been demonstrated by Newell (1965) and Johannes and Satomi (1966). (c) Floe aggregates — amorphous ma- terial consisting primarily of organic and inorganic debris. The origins of this material may be from both plant and animal sources and is usually referred to as detritus. Considerable research has been directed in defining detritus as a food stuff (Darnell, 1967; Fox, 1950; Odum and de la Cruz, 1967; Adams and Angelovic, 1970). (d) Plant fragments — dead and decomposing plant material. Odum and de la Cruz (1967) have shown that decomposing plant material is actually a better food source than living tissue. The composition of the different particle types associated with the sediment is quite consistent between samples (Table I). Mineral grains less than 25 /x are the most abundant particle type. Mineral grains and floe aggregates comprise more than 90% of the total sediment. Feeding selectivity From his study of sediment working of P. gouldii collected from Barnstable Harbor, Gordon (1966) concluded that the polychaete w?as a non-selective deposit feeder and did not sort particles for ingestion according to size. His conclusions were based upon comparisons of particle size distributions in the gut to that of sediment collected on the surface of the sand flat \vhere the worms were feeding. Gordon failed, however, to note the length of individual worms or sample the sediment that the worms were directly feeding upon. In the present study, P. gouldii is found to be selectively feeding upon the sediment at Little Sippewisset salt marsh. Figure 2 shows that larger worms are selecting, on the average, larger particles than smaller worms (r = 0.85, P < 0.05 ) . This may be the result of two factors. Smaller individuals of P. gouldii feed closer to the surface than larger individuals. Since the average FOOD SELECTION IN A DEPOSIT FEEDER 231 150 en e o> I o> o o Q_ a; CT> O !00 50 o O o o o o 20 30 40 Body Length (mm) 50 FIGURE 2. Relationship between body length of Pcctiimria i/ouldii and length of average-sized particle ingested by the polychaete. particle size of the sediment increases slightly with depth, the animals may be feeding in completely different sedimentary environments. Secondly, the size selection may be the result of a morphological constraint placed upon the organisms (e.g., mouth size, ability of the feeding tentacles to obtain particles, etc.) that changes with size (age) of the polychaete. Feeding selectivity coefficients are presented in Figure 3 for the six most abundant particle types found in the foregut (more than 95% of the particles Mineral < 75 /i 1 — r Mineral >75/i Encrusted Mineral >75/i Not Encrusted * Mineral 25-75/1 Floe Aggregate Fecel Material I ' I -1.0 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1.0 FIGURE 3. Electivity coefficients of different particle types selected by Pectinaria gouldii (data averaged for fifteen worms). 232 ROBERT B. WHITLATCH CO 80 70 60 50 20 §30 ^20 10 °o O o o o o o o o 0° o o o o o — O — o o 100 200 300 Mineral Length 400 500 600 FIGURE 4. Relationship between mineral length found in the sediment where Pectinaria gouldii is feeding and percentages of minerals having encrustations. Mineral lengths are clumped into 25 M size groups. ingested). The data indicate that P. gouldii consistently prefers large encrusted mineral grains greater than 75 /*. Nonencrusted mineral grains greater than 75 ^ and encrusted mineral 25-75 ju. show positive electivity values in most of the worms. Fecal material and floe aggregates were selected in the majority of animals, though fecal material was selected more often. In all the polychaetes studied, there was avoidance (probably due to a mechanical or morphological constraint) for mineral grains less than 25 ^,. The percentage of encrusting material on the mineral grains is highly cor- related with grain size (Fig. 4). Mineral grains less than 25 JJL are never encrusted while grains greater than 75 /x are increasingly encrusted (lO-SO*/^). P. youldii may be preferentially selecting large grains to increase the amount of organic matter ingested. As already noted, fecal material and floe aggregates are highly important food sources for deposit feeders and may be selected for the same reason. Sediment staining Table II gives the results of the periodic acid Schiff (PAS) staining of the sediment near the mouth, sediment ingested and sediment sampled from the intestine of P. gouldii. Analysis of the sediment samples near the mouth of the worms indicates an average of 32.7% of the particles are possible sources of organic mat- ter. Possible sources of organic matter refers to the total material that could be regarded as a food source (as outlined above) for the polychaetes. The total FOOD SELECTION IN A DEPOSIT FEEDER 233 PAS stained material averages 13.9%. Of the 32.7% possible organic material, less than half (average of 42.6%) was stained by the carbohydrate stain suggesting that not all of the material is organic in nature. Studies by Waksman and Hotchkiss (1937), Anderson (1940), Mare (1942) and George (1964) have strongly suggested that only about 10-20% (by weight) of the organic matter in marine sediments is in a biologically utilizable form. Analysis of the sediment from the foregut of the worms shows an average of 42.7% possible organic matter of which 20.5% was PAS stained. There was an average of 30.9% possible organic material obtained from the intestine of the worms of which 14.3% stained by the PAS method. The mercuric bromphenol blue (MBB) method stained only a very small amount of material in the sediment. Of four samples analyzed, less than 0.4% of the material showed MBB reactions. Less than 0.1% of the ingested fraction obtained from the foregut of the worms stained with the MBB method, while sedi- ment sampled from the intestine showed no reaction to this stain. DISCUSSION Calculations of the assimilation efficiency of P. gouldii show that the worms re- move on the average 30% of the possible organic material from ingested sediment. Assimilation efficiencies of PAS-stained sediment average 29.1% for the fifteen worms. Gordon (1966) reports that P. gouldii removed almost 50% of the organic material (by weight) from each gram of sediment worked. George (1964) states that the deposit feeding polychaete Cirriformia tentaculata removes only 7.9% (by weight) of the organic matter present in the sediment. Adams and Angelovic (1970) found that the polychaete Glyccra dibranchiota has an assimilation rate of 41% (based upon the amount of 14C respired as CO2) when fed detritus and 34^ when fed undecomposed eel grass. Since the values presented in this study are not volumetric or by weight it is difficult to make comparisons with assimilation rates presented in the other studies. However, the values presented are well within the range of the other studies on assimilation efficiences. TABLE II Per cent particle abundance of carbohydrate staining (periodic acid Schiff) A B C D E F G H I J K L M N 0 Sediment near mouth of P. gouldii Per cent "possible" organic 38.9 37.3 31.1 33.1 32.4 31.2 25.9 30.8 32.1 36.5 33.3 29.1 36.5 29.4 33.0 Per cent stained 10.2 5.9 13.1 14.0 13.7 12.4 9.4 11.5 15.7 16.5 18.2 18.6 17.6 13.8 18.6 Per cent organic stained 26.2 15.8 42.1 42.3 42.3 39.7 36.3 37.3 48.9 45.2 54.6 63.9 48.2 46.9 56.4 Sediment in foregut of P. gouldii Percent "possible" organic 41.9 37.9 48.1 45.7 36.5 43.3 53.9 41.6 46.1 40.1 47.3 42.2 43.5 36.2 36.9 Per cent stained 18.6 17.9 21.9 16.5 13.9 18.6 18.2 23.9 31.4 24.5 28.1 13.3 23.4 21.3 15.4 Per cent organic stained 44.4 47.2 45.5 36.1 38.1 42.9 33.8 57.5 61.2 61.1 59.4 31.5 53.8 58.8 41.7 Sediment in hindgut of P. gouldii Per cent "possible" organic 28.1 28.3 33.2 33.3 28.4 31.4 37.2 32.5 40.1 33.5 29.5 26.1 22.7 28.2 31.2 Per cent stained 12.9 13.1 16.1 11.7 11.4 8.9 17.0 17.2 27.0 17.9 13.5 11.7 15.3 10.0 11.4 Per cent organic stained 45.9 46.3 48.5 35.1 40.1 28.3 45.7 52.7 68.1 53.4 45.8 44.8 67.6 35.5 36.5 % "Possible" assimilated 3.5.2 25.3 30.9 27.3 22.2 27.5 30.9 21.9 13.0 16.4 37.6 38.1 47.8 22.1 15.4 % Stained assimilated 30.6 26.8 26.5 29.1 17.9 52.2 6.6 28.0 14.0 26.9 51.9 12.0 34.6 53.1 25.9 234 ROBERT B. WHITLATCH Data presented in this study provide some interesting results concerning the possible effects of the polychaete on the benthic environment. The worms con- centrate an average of 32.7% possible organic material found in the sediment to 42.7% by selectively feeding. Through the concentration of organic material, the animals are channeling large amounts of organic material to the surface where it can become available to other organisms. The feeding activities of P. gouldii could have important effects upon the recycling of nutrients for the salt marsh tropic chain. The organic fraction deposited upon the surface may even be sus- pended in the water column. Rhoads (1973), for example, has shown that re- suspended fecal material of benthic deposit feeders in Long Island Sound may represent a significant food source for commercially important suspension feeding molluscs. The use of histological staining techniques for examination of sediments is a biologically meaningful way to ascertain the qualitative and quantitative composition of marine sediments. The staining methods, although very generalized in their staining affinities, are valuable in determining the nature of the food sources for benthic animals and those portions that may be biologically utilizable. I wish to thank Drs. R. G. Johnson and T. J. M. Schopf for reviewing and improving the manuscript. This research was supported by NSF grant GA 35819 to Ralph G. Johnson, University of Chicago and the Marine Biological Laboratory. SUMMARY 1. A study of the food-resource partitioning in the deposit-feeding polychaete Pectinana gouldii collected from Little Sippewisset salt marsh, Massachusetts, shows that, on the average, larger worms select larger particles than smaller worms. Comparisons of ingested sediment with sediment collected where the animals were feeding indicate that the polychaetes prefer organic-encrusted mineral grains, floe aggregates, and fecal material. 2. Histological stains w-ere used to determine the percentage particle abundance of different possible food sources and fractions ingested by the polychaetes. Mer- curic bromphenol blue (MBB) was used to stain protein-containing material and periodic acid Schiff reagent (PAS) was used to stain carbohydrate-protein com- plexes. Total possible organic material in the sediment averaged 32.7%. Very little of the sediment (less than 0.4%) stained with MBB, while an average of 13.9% of the sediment stained with PAS. Of the total possible organic matter, only about one-half stained with the PAS reagent suggesting not all of the material is organic in nature. 3. Analysis of the sediment ingested by the worms averaged 42.7% possible organic matter, of which 20.5% was PAS-stained. Calculations of the assimilation efficiencies of P. gouldii show that the worms remove, on the average 30% of the possible organic matter and 29.1% of the stained material from the sediment. LITERATURE CITED ADAMS, S. M., AND J. W. ANGELOVIC, 1970. Assimilation of detritus and its associated bac- teria by three species of estuarine animals. Chesapeake Sci., 11: 249-254. FOOD SELECTION IN A DEPOSIT FEEDER 235 ANDERSON, D. O., 1940. Distribution of organic matter in marine sediments and its availability to further decomposition. /. Mar. Res., 2 : 225-235. ANDERSON, J. G., AND P. S. MEADOWS, 1969. Bacteria on intertidal sand grains. Hydro- biologia, 33 : 33-46. BATOOSINGH, E., AND E. H. ANTHONY, 1971. Direct and indirect observations of bacteria on marine pebbles. Can. J. Microbiol, 17 : 655-664. DARNELL, R. N., 1967. The organic detritus problem. Pages 374-375 in G. H. Lauff, Ed., Estuaries. American Association for the Advancement of Science, Publication 83. Fox, D. L., 1950. Comparative metabolism of organic detritus by inshore animals. Ecology, 31: 100-108. GEORGE, J. D., 1964. Organic matter available to the polychaete Cirrijormia tentaculata (Montagu) living in an intertidal mud flat. Limnol. Occanogr., 9: 453-455. GORDON", D. C, JR., 1966. The effects of the deposit feeding polychaete Pectinaria gouldii on the intertidal sediments of Barnstable Harbor. Limnol. Occanogr., 11: 327-332. HUMASON, G. L., 1967. Animal Tissue Techniques. W. H. Freeman, San Francisco, 468 pp. IVLEV, V. S., 1961. Experimental Ecology of the Feeding of Fishes. Yale University Press, New Haven, Conn., 302 pp. JOHANNES, R. E., AND M. SATOMI, 1966. Composition and nutritive value of fecal pellets of a marine crustacean. Limnol. Occanogr., 11 : 191-197. MARE, M. F., 1942. A study of a marine benthic community with special reference to the micro-organisms. /. Mar. Hiol. Ass. U. K., 25 : 517-554. MAZIA, D., P. A. BREWER AND M. ALFERT, 1953. The cytochemical staining and measurement of protein with mercuric bromphenol blue. Biol. Bull., 104: 57-67. NEWELL, R. C., 1965. The role of detritus in the nutrition of two marine deposit feeders, the prosobranch Hvdrobia ulvae and the bivalve Macoma baltliica. Proc. Zool. Soc. London, 144 : 25-45'. ODUM, E. P., AND A. A. DE LA CRUZ, 1967. Particulate organic detritus in a Georgia salt marsh-estuarine ecosystem. Pages 376-382 in G. H. Lauff, Ed., Estuaries. American Association for the Advancement of Science, Publication 83. RHOADS, D. C., 1973. The influence of deposit-feeding benthos on water turbidity and nu- trient recycling. Amer. J. Sci., 273 : 1-22. RHOADS, D. C., AND D. K. YOUNG, 1970. The influence of deposit-feeding organisms on sedi- ment stability and community structure. /. Mar. Res., 28 : 150-178. RHOADS, D. C., AND D. K. YOUNG, 1971. Animal-sediment relations in Cape Cod Bay, Massachusetts. II. Reworking by Molpadia oolitica (Holothuroidea). Mar. Biol. 11: 255-261. SANDERS, H. L., E. M. GOUDSMIT, E. L. MILLS AND G. E. HAMPSON, 1962. A study of the intertidal fauna of Barnstable Harbor, Massachusetts. Limnol. Occanogr., 7 : 63-79. STEELE, J. H., AND I. E. BAIRD, 1968. Production ecology of a sandy beach. Limnol. Occanogr., 13 : 14-25. WAKSMAN, S. A., AND M. HOTCHKISS, 1937. On the oxidation of organic matter in marine sediments by bacteria. /. Mar. Res., 1 : 101-118. WATSON, A. T., 1927. Observations on the habits and life history of Pectinaria (Lagis) korcni. Proc. Trans. Liverpool Biol. Soc., 42 : 25-60. WHITLATCH, R. B., 1972. The ecological life history and feeding biology of Batillaria zonalis (Bruguiere). Masters thesis, University of the Pacific, Stockton, California, 157 pp. YONGE, C. M., 1928. Feeding mechanisms in the invertebrates. Biol. Rev., 3 : 21-76. Reference: Biol. Bull., 147: 236-246. (August, 1974) THE EFFECTS OF HEAVY METAL IONS ON THE MOTILITY OF SEA URCHIN SPERMATOZOA 1 L. G. YOUNG AND L. NELSON Department of Physiology, Emory University, Atlanta, Georgia 30322; Department of Physiology, Medical College of Ohio, Toledo, Ohio 43614; and Marine Biological Laboratory, IVoods Hole, Massachusetts 02543 Intense activity of sea urchin spermatozoa follows dilution in sea water. The high performance level is of limited duration ; metabolism and motility decline to a low steady state and fertilizing capacity diminishes progressively with time (Tyler and Tyler, 1966). This so-called "dilution effect" has been attributed to the action of trace amounts of heavy metal ions present in seawater on the respiratory enzymes of the spermatozoa (Rothschild and Tuft, 1950; Rothschild and Tyler, 1954). Although there are numerous reports on the effect of various metal ions on the Oo consumption of sea urchin spermatozoa, quantitative data on the action of heavy metals on spermatozoan motility are not generally available. Rothschild and Tuft (1950) reported that small amounts (4 X 10~5 M) of CuQo or ZnClo added to dense suspensions (>4X 10s sperm/ml) of Echinus esculent its spermatozoa increased their level of Oo consumption, but had no effect on the respiration of dilute suspensions (<4x 108 sperm/ml). While these authors did not provide actual measurements, they stated that neither copper nor zinc had any effect on the swimming speed of the spermatozoa. Mohri (1956) found that CuCU or ZnClo at low concentration (10~5 M) accelerated the Oo uptake of Hctniccntrotits pulcherrimus spermatozoa, but at a higher concentration (10~4 M) inhibited it. According to Barron, Nelson, and Ardao (1948), 5 X 10~6 M HgClo stimulates the respiration of Arbacia punctulata spermatozoa, but a concentration of 10"4 M inhibits their Oo consumption. These investigators did not relate the effects on Oo uptake to the motile activity of the cells. Additional evidence for the sensitivity of sea urchin speramtozoa to the toxic effects of heavy metal ions comes from experiments on the ability of chelating agents to prolong both the motility and fertilizing capacity of sea urchin sperma- tozoa. Rothschild and Tyler (1954) found that 10 micromolar ethylenecliamine- tetraacetic acid (EDTA) depressed dilution-induced increase in Oo uptake and de- layed senescence of Echinus cscnlcntns spermatozoa. Tyler (1953) used a variety of agents which bind heavy metal ions — amino acids, diethyldithiocarbamic acid (DEDTC), 8-hydroxyquinoline (oxine), and a-benzoinoxime (cupron) — to pro- long the life span and fertilizing capacity of sea urchin and sand dollar spermatozoa. Mohri (1956) also found that chelating agents would suppress the respiratory in- creases observed on dilution of Hemicentrotus pulchcrr'nuits sperm. The investi- gators concluded that these agents act to depress respiration, and thereby prolong sperm motility and viability, by binding heavy metals ordinarily present in the seawater. i Supported in part by NIH Research Grants HD 06491-01 and HD 03266-05. 236 HEAVY METALS AND SPERM MOTILITY 237 Although it is often assumed that increased O2 consumption by the spermatozoon is accompanied by increased motile activity and decreased O2 consumption by de- creased motile activity both Robbie (1946) and Rothschild (1948) have demon- strated that spermatozoan motility is not a simple function of oxygen uptake. In this report, using a convenient and objective determination of swimming speed, we explore quantitatively the effects of alterations in the environmental concentrations of heavy metal ions on the motile activity of Arbacia punctulata spermatozoa. MATERIALS AND METHODS Spermatozoa were collected from sea urchins, Arbacia punctulata, by injecting 0.53 M KC1 into the perivisceral cavity. The shed sperm were concentrated at 1000 X g in a clinical, table top centrifuge for 3 minutes, the supernatant fluid removed, and the packed sperm stored at 4° C. Immediately prior to the motility rating tests, samples of the spermatozoa were diluted to a final concentration of 10T ± 10% sperm/ml of filtered seawater, determined by optical density measure- ments of the sperm cell suspensions (Nelson, 1972). This concentration was selected empirically because it fell within the optical density range appropriate to the motility determinations. Five milliliters of seawater suspensions of spermatozoa were added to round colorimeter tubes containing either: (1) 0.5 ml distilled water as control, (2) 0.5 ml 1% formaldehyde to kill the cells, or (3) 0.5 ml of the reagent being tested. The contents were mixed thoroughly by inverting twice. The swimming speed of the spermatozoa was determined by the centrifuge- orientation, optical density method using a Bausch and Lomb Spectronic 20 Colorim- eter at a wavelength of 540 nm and a clinical centrifuge. Measurements of sperm cell density were made immediately after mixing the sperm suspensions into the reagents and after each of three four-minute centrifugations at 120 X g. The method depends on the fact that, under the force generated in a mild gravitational field, the normally, slightly positively geotropic Arbacia sperm cells become oriented centrifugally and swim with decreased randomness to the bottom of the tube. While the dead cells also become similarly oriented, they no longer swim and are only minimally sedimented at 120 X g. Therefore, the rate of decrease of optical density (AOD). which was previously determined to be linearly proportional to the swimming speed, may be calculated after correction for any slight displacement of non-motile cells, and in comparison to the AOD of the untreated control cells, viz., M := AODx/AOD,. X 100. Changes in the speed of the swimming sperm cells are expressed in terms of per cent of control swim rate. All experimental procedures were carried out in an air-conditioned room (22- 23° C) and initial swimming speeds were obtained within 5 minutes of dilution of the spermatozoa in filtered seawater. Branham (1966) and Timourian and Watchmaker (1970) have reported that the motile activity of sea urchin sperma- tozoa remains stable for approximated 30 minutes after dilution in seawater. The reagents tested included: CuCl? 500 HM to 100 ^M, ZnClo 5 /XM to 10 HIM, MnCl2 5 /JLM to 10 HIM. HgCli. 10 nivi to 1 HIM, EDTA 10 /AM to 5 HIM (adjusted to pH 7.8 with NaOH before use). The natural concentrations, in /xEq liter, of 238 L. G. YOUNG AND L. NELSON TABLE I Dose- and time-dependence of zinc effects. Motility, expressed as per cent of control motility (100%'}, of Arbacia punctulata spermatozoa after each of a series of three 4-minute centrifngations at 120 X g; W1 ± 10% sperm in 5 ml filtered Woods Hole seawater; temperature 22-23° C. Each value represents the mean of individual determinations made on spermatozoa collected from three different Arbacia punctulata, ± standard deviation % Control motility Centrifugation # 1 Centrifugation #2 Centrifugation #3 2500 22 ± 5.6 20 ± 5.6 20 ± 4.8 500 36 ± 2.2 34 ± 2.8 30 ± 2.8 250 44 ± 2.6 29 ± 0.6 17 ± 4.1 50 80 ± 7.2 52 ± 5.8 30 ± 4.6 25 155 ± 10.8 111 ±4.6 97 ± 8.7 5 113 ± 5.0 105 ± 7.1 107 ± 10.6 2.5 98 ± 3.6 101 ± 1.0 100 ± 1.0 certain heavy metal ions in seawater (using data adapted from Richards, 1972) are zinc — 0.3, copper — 0.1, manganese — 0.1, and mercury — 0.002. RESULTS Zinc Figure 1 shows the response of Arbacia punctulata spermatozoa to increases in environmental zinc. Sea water contains approximately 0.3 /xEq zinc/liter (Richards, 1972) ; increasing the zinc concentration to 5 /xEq/ liter causes little more than a 10% increase in swim speed while a tenfold increase in zinc concentration (to 25 juEq/liter) speeds up the sperm by 55%. Increasing the amount of environmental zinc to 50 /xEq/liter slows the swimming speed of the spermatozoa to 20% below control values. As the concentration of zinc is increased from 250 to 500 to 2500 ^Eq/liter of seawater, motility declines from 44% to 36% to 22% of control levels. However, we had to increase the concentration of zinc in the medium to 15 mEq/liter in order to immoblize the sperm cells completely. Prolonged contact with zinc ion, even at concentrations which initially stimulate motility, has an apparently debilitating effect on the motile activity of the sperma- tozoa. Table I shows that, following the addition of 25 /J£q of zinc, Arbacia sperm cells swim at 155% of the control speed after the first four minutes centri- fugation, 111% of the control swim rate after the second four-minute Centrifugation. At 250 ,uEq zinc/liter, motility declines from 44% to 29% to 17% of the control speed. In both higher (0.5 to 2.5 mEq) and lower (2.5 to 5 /xEq) concentrations of added zinc, there is no significant change in the swimming speed of the Arbacia spermatozoa after the first 4 minutes Centrifugation. Copper The concentration of copper in seawater is equivalent to about 0.1 //.Eq/liter. The effect of increases in environmental copper on the swimming speed of sperma- HEAVY METALS AND SPERM MOTILITY 239 tozoa collected from each of two sea urchins is illustrated in Figure 2. Addition of 0.5 juEq copper speeds the sperm by approximately 10%. In the presence of 2.5 /*Eq added copper, sperm motility falls to 86 or 66% of control values, decreases to 37 or 58% of control values in 5.0 /xEq excess copper, to 31 or 36% of control values in 25 /AEq excess copper, and ceases completely in the presence of 50 160 1- ~-2 -3 -4 -5 -6 LOG [ZnCJ2] ADDED TO FSW -7 FIGURE 1. Effect of ZnQ2 on the swimming speed of Arbacia punctulata spermatozoa. Varying concentrations of ZnCU added to filtered Woods Hole seawater containing 107 ± 10% Arbacia punctulata spermatozoa elicit a biphasic dose-dependent response. Abscissa represents the zinc chloride concentrations added, in log moles per liter seawater/sperm suspensions; ordinate is the swim rate of treated sperm expressed as per cent of the control rate in sea- water. The points and ranges indicate the mean and standard deviations of individual deter- minations made on spermatozoa collected from three individual Arbacia. The incubation varied from 22-23° C on different days. 240 L. G. YOUNG AND L. NELSON 120- 80- o u •3 •4 -5 -6 Log [MnCI2 or CuCI2] added to FSW FIGURE 2. Effects of CuCU and MnCU on the swimming speed of Arbacia pnnctulata spermatozoa. MnCla and CuCl2 added to filtered Woods Hole seawater containing 10' ± 10% Arbacia pnnctulata spermatozoa decreased their swimming speed in proportion to the concen- tration. Abscissa is concentrations of metal ion added, in log moles per liter of seawater/sperm suspensions: copper — open and closed squares, manganese — open and closed circles (open and closed data points from replicate runs). Ordinatc is the swim rate of treated sperm expressed as per cent of control rate in a seawater. Each point represents an individual determination of swim rate expressed as a per cent of control rate in seawater at a temperature of 22-23° C. Note that while the seawater content of Mn2+ and Cu2+ both equal about 10~7 equivalents per liter, the sperm cells are more sensitive to excess copper by about two orders of magnitude. excess copper. Copper appears to exert its effect on sperm motility almost imme- diately after addition to the sperm cell suspensions; further significant declines in swimming speed do not occur after the first 4 minutes centrifugation. Manganese Woods Hole seawater contains approximately the same concentration of manganese as it does copper. But as shown in Figure 2. manganese exerts much less of an inhibitory effect on Arbacia sperm motility than does copper. Addition of 0.5, 2.5 or 5 /nEq manganese/liter to the sperm cell suspensions does not measurably affect the swimming speed. Excess manganese up to 25 juEq/liter causes only a 4 or 15% reduction in motility and further increases result in only a slow decline in Arbacia sperm swimming activity ; in the presence of 2.5 mEq manganese/liter — approximately three orders of magnitude greater than the nor- mal seawater concentrations of this heavy metal ion — motile activity is still about one-quarter of the control value. HEAVY METALS AND SPERM MOTILITY 241 Mercury The effect of HgCU on the motility of Arbacia punctulata spermatozoa was measured after incubation of the cells with mercuric ion for fixed intervals of time. Figure 3 (solid line) is an example of the changes in motility of sperm cells collected from a single sea urchin after 8 minutes incubation in 5 X 1OS to 10~3 M/ A marked increase in swim speed is observed at mercury concentrations between 10~4 and 10~7 M. In 5 X 10~6 M HgCU Arbacia sperm swim at about 270% of the control rate. In higher concentrations of mercury, motility declines but is still 10% above control values in 10~a M HgCl^. As the concentration of mercuric ion is decreased below 10^7 M, acceleration gradually declines and the 240 ^160 o o c 80 o u 0 -2 -3 -4 -5 -6 Log [HgCIJ -7 -8 FIGURE 3. Effect of HgCU on the swimming speed of Arbacia punctulata spermatozoa. The HgCU added to filtered Woods Hole seawater containing 10* + 10% Arbacia punctulata spermatozoa shows pronounced dose-dependent and time-dependent effect. The abscissa is the log molar concentration of mercury in seawater/sperm suspensions : open square — 8 minutes incubation ; solid triangles — 16 minutes incubation ; open circles — 24 minutes incubation. The onlinate shows the swim rate of treated sperm expressed as per cent of control rate in seawater, temperature 22-23° C. 242 L. G. YOUNG AND L. NELSON 100 80 560 o :E 040 +-> c O 020 -20 -2.5 -30 -3.5 -4.0 -4.5 -50 Log [EDTA] FIGURE 4. Effect of EDTA on the swimming speed of Arbacia punctulata spermatozoa. EDTA added to filtered Woods Hole seawater containing 107 + 10% Arbacia punctulata spermatozoa per milliliter depresses the swimming speed at effective concentrations. The abscissa is in log molar concentration of EDTA in seawater/sperm suspensions. The ordinate is the swim rate of treated sperm expressed as per cent of the control rate in seawater. The points and ranges represent the mean and standard deviations of separate determinations made on spermatozoa collected from three individual Arbacia punctulata. Incubation temperature was 22-23° C. effect is negligible at 5 X 10~s M. Moreover, the stimulatory effect of HgClo on Arbacia sperm motility is transitory. The dashed lines in Figure 3 show the de- cline in swim speed observed after 16 and 24 minutes incubation in various concentrations of HgClo. For example, in S X 10~6 M HgClo motility is 170% above control level after the first centrifugation (8-minute measurement), 30% above control level after the second centrifugation (16-minute measurement) and 33% below control level after the third centrifugation (24 minutes measurement). Since an initial stimulation and subsequent decline in sperm motility was ob- served at all concentrations of HgClo tested, the effect of preincubation with mercuric ion on sperm swimming speed was determined. The results of a ten- minute preincubation of the sperm cells in 10~; to 10~7 M HgClo essentially con- firm the observations recorded on sperm cell delayed response illustrated in Figure 3. HEAVY METALS AND SPERM MOT1L1TY 243 EDTA EDTA, in excess of 100 /*M/1 of seawater, exerts a profoundly depressant effect on the swimming speed of sea urchin spermatozoa (Fig. 4). Motility was measured in concentrations of EDTA ranging from 5 X 10 3 to 5 X 10~5 molar. In Figure 4 we note the following: in seawater suspensions containing less than 10"* Moles of EDTA per liter, motility appears normal. A twofold increase in EDTA concentration, to 2 X 10~4 M, causes a 17% decrease in the swimming speed of the spermatozoa. Another doubling of the EDTA concentration, results in a further decline in motility to 64% of control values. In 6 X 10"4 M EDTA swim speed declines to only 21% of control rates. As the concentration of EDTA is further increased, the motility of the Arbacia spermatozoa continues to decline at a slow rate and in 2 X 10 3 M EDTA the sperm are virtually immotile. DISCUSSION From the evidence we may conclude that the motile activity of Arbacia punc- Itilata spermatozoa is affected in a concentration-dependent and, in some instances, time-dependent, manner by the heavy metal ion composition of the environment. Although we are not prepared at present to specify the particular sites within the spermatozooon on which these metal ions exert their action, previous studies indicate that the effects on motility may in part depend on heavy meal interaction with protein sulfhydryl groups to form mercaptides. Barren et al. (1958) suggested that soluble — SH groups are necessary for the activity of enzymes essential for spermatozoan viability. Sea urchin spermatozoa exhibit a biphasic response to increases in the environ- mental concentration of zinc ions. A three- to thirty fold increase of the zinc in the seawater causes the Arbacia pnnctidata spermatozoa to swim between 13 and 55%> faster than control sperm in normal seawater. This acceleration of the swim rate is only transitory since, within 30 minutes of the addition of 5 to 25 /xEq zinc/liter to the seawater in sperm suspensions, motility has declined to control levels. Larger excesses of zinc, above 50 juEq/liter, do not stimulate sperm motility, but instead cause an immediate fall in swimming speed which continues to decline as the time of incubation in zinc ion is increased. Zinc, at a concentration of 25 /iEq/liter, which according to Rothschild and Tuft (1950) has no effect on the oxygen uptake of dilute suspensions of Echinus spermatozoa, exerts a short- term stimulatory effect on the motility of Arbacia spermatozoa. Therefore zinc is probably acting in other ways than only by combining with the soluble sulfhydryl groups involved in regulation of sperm cell metabolism. Morisawa and Mohri (1972) found that both the sperm tails and isolated microtubules of the sea urchin Pseudocentrotns depresses appear to concentrate zinc, and they concluded that this ion may play an important role in the contractile process. The fairly large excess of zinc needed to overcome the initial acceleration of Arbacia sperm motility indi- cates that, at concentrations above 50 /xEq/liter, zinc's inhibitory effect may be due to binding at active sites on contractile proteins. In fact, Utida and co- workers (1956) report that other divalent cations, Cd2+, Co2+ and Ni2+, increase ATP hydrolysis by washed sperm tails of the sea urchin Hdiocidaris crassispina at 1 niM/1, while Cu2+ and Zn2+ activate the enzyme at 10"4 molar but inhibit at 244 L. G. YOUNG AND L. NELSON the higher concentration. Our data on swimming speed (see Fig. 1) bear a most striking resemblance to the graphic representation of Utida, Maruyama and Nanao (1956) relating sperm tail apyrase activity to the zinc concentration. In the range that we examined, both copper and manganese ions inhibit Arbacia sperm motility, but while we found copper to be about 100 times more potent an inhibitor of sea urchin sperm swimming than was managanese, still, at the lowest concentration — 0.5 /XM — CuClo actually stimulates motility by about 10%. An average hundredfold increase above environmental manganese causes an 8% decrease in swim speed while an equivalent increase in the copper content of sea- water depresses Arbacia sperm motility by 77%. The comparatively large amounts of manganese which must be added to the sperm cell suspensions before significant changes in motile activity can be detected suggest that manganese may, among other things, have only limited access to active groups which are important for propulsive activity. Alternatively, we cannot overlook the report of Garbers, Lust, First and Lardy (1971) that the sperm cells of the sea urchin, Stronglyocentrotus are re- markably endowed with guanyl cyclase, and to a lesser extent, adenyl cyclase, both of which are highly Mn'J+ dependent. Agents which influence cyclic nucleotide metabolism affect both motility and respiration of spermatozoa according to Garbers ctal. (1971). Since small amounts of copper (2.5 /uEq liter) in the sea water/sperm sus- pensions cause a measurable decrease (34%) in swim speed, it seems likely that copper may act at essential regulatory sites within the spermatozoan flagellum. The role of copper in metabolism is well documented ; excess copper ions exert their effects at a variety of enzymatic sites. Morisawa and Mohri (1972) believe that cytochrome C oxidase in the midpiece of sea urchin spermatozoa accounts for most of the copper found in these sperm while Barnes and Rothschild (1950) claim that sea urchin sperm are able to bind copper to the extent of 300 times the amount in seawater. Rothschild and Tuft (1950) have shown that copper arrests the decline in oxygen consumption of dilute suspensions of sea urchin spermatozoa. Mercury, in contrast to zinc and to managanese and copper, had not until recently been regarded as a biologically significant constituent of seawater. Never- theless this heavy metal ion is becoming increasingly important as a pollutant. \Ye found Arbacia sperm cells to be extremely sensitive in both dose-dependent and time-dependent manner to supplements of mercury. Addition of mercuric chloride to 5 X 10~6 M (or 3 X 10s molecules of mercury salt/spermatozoon at a working dilution of 107 sperm/ml seawater), accelerates the sperm by about 170%. This increase, however, is only transitory and the mercuric ion evinces its ultimate toxicity by causing the abrupt fall in motile activity that rapidly supersedes the initial rise. For example, prolonging the incubation period from 8 to 24 minutes in 5 X 10"'' M mercury causes about a 200% decrease in the swimming speed of the sea urchin sperm from 170%> above control values after 8 minutes incubation, to 29% above control values after 16 minutes incubation, to 33% below control values after 24 minutes incubation in mercury. Barron ct al. (1948) observed that low concentrations of mercury (5 X 10~6 M) caused an 88% increase in the respiratory activity of Arbacia pnnctitlata spermatozoa while higher concentrations of mercury (10~4 M) caused complete inhibition of the respiration. This they attributed to the fact that low levels of mercury, in combining with soluble — SH HEAVY METALS AND SPERM MOTILITY 245 groups that are important in the regulation of cell respiration, abolish that regulatory function, while with further increases, the mercury next combines with fixed — SH groups on essential proteins to inhibit enzymatic activity. Although we observed that low concentrations of mercury (5 X 1O6 M) caused an initial large increase in motile activity, we were still able to record motile activity in spermatozoa suspended in sea water containing mercury concentrations as high as 10~4 M. When exposed to this concentration of mercuric ion, the sperm swam at 133% of control rate after 8 minutes, at 54% of control after 16 minutes, and at 41% of control after 24 minutes exposure. Nevertheless some critical amount of certain of the heavy metal ions must be necessary for optimum motility since EDTA drastically reduces the swimming speed of dilute suspensions of Arbacia spermatozoa. The effect of EDTA in in- creasing the life span and fertilizing capacity of sea urchin spermatozoa by removing trace metals from seawater has been discussed by several investigators (Roths- child and Tyler, 1954; Mohri, 1956; Tyler, 1953).' Rothschild and Tyler (1954) calculate that one micromole/liter of EDTA is sufficient to bind all the trace metal ions in ordinary seawater. In our experiments, 10 ° M EDTA had no effect on the swimming speed of Arbacia spermatozoa. However, the spermatozoa used in this study were in dilute suspension and it is probable that low steady state conditions already prevailed even though we completed our first measurements within 5 minutes of dilution. Decreases in swimming speed did not become apparent until the EDTA was raised to 2 X 10~4 M/l. At this concentration motility was 83% of control values while in 8 X 10~4 M EDTA, motility further decreased to 13% of control values. Since EDTA did not affect Arbacia sperm swim speed except at high concentrations and since the concentration range of its effectiveness was sharply limited, it is possible that EDTA brought about a decrease in sperm motility by interfering with transport of divalent cations from cell surface sites as well as by removing trace metals from the seawater. In this paper we have shown that, although a critical level of certain heavy metal ions must be maintained for optimum sea urchin sperm motility, an excess of these ions in the seawater adversely affects the propulsive activity of the spermatozoa. While metal ions can induce changes in motility, the mechanism of their effects on specific sperm enzymes and on the fertilizing capacity of the sea urchin spermatozoa remains to be investigated. In any case, as a probable consequence of the accumulation of heavy metals as pollutants in our oceans, large deficits in the reproductive capabilities of marine invertebrates will begin to become more apparent. SUMMARY Optimum motility of sea urchin spermatozoa for a period adequate to initiate the process of fertilization requires an apparently critical level of certain heavy metal ions. Increase of some of the divalent cations above the "normal" seawater con- tent accelerates or depresses the swimming speed in dose- or time-dependent fashion (or both). The different patterns of Arbacia sperm swimming speed re- sponse to the individual cation supplements (Cu, Zn, Mn, Hg) may reflect dif- ferences in rate of penetration into the cell, binding of atcive groups or selective inhibition of as yet unspecified enzymes at or below the cell surface which directly 246 L. G. YOUNG AND L. NELSON or indirectly contribute to regulation of flagellar contractility. The concentra- tions tested ranged from 500 HM up to 10 HIM. The "optimum" concentrations fell between 1 to 10 micromolar Cu, Zn and Hg on initial exposure, while Mn was moderately inhibitory at these levels. EDTA, up to 10~4 molar, exerts no adverse effect on sperm propulsion, while 8 X 10"* M almost completely blocks the motility. Within this short concentration span, the EDTA appears not only to bind essential seawater cations, but may also deplete those intracellular regulatory cations which otherwise may exist in a state of dynamic equilibrium with the seawater. LITERATURE CITED BARNES, H., AND LORD ROTHSCHILD, 1950. A note on the copper content of sea-urchin semen and sea water. /. Exp. Biol, 27 : 123-125. BARRON, E. S. G., L. NELSON AND M. I. ARDAO, 1948. Regulatory mechanisms of cellular respiration. II. The role of soluble sulfhydryl groups as shown by the effect of sulfhydryl reagents on the respiration of sea urchin sperm. /. Gen. Ph\siol., 32(2) : 179-190. BRANHAM, J. M., 1966. Motility and aging of Arbacia sperm. Biol Bull., 131 : 251-260. CAREERS, D. L., W. D. LUST, N. L. FIRST AND H. A. LARDY, 1971. Effects of phos- phodiesterase inhibitors and cyclic nucleotides on sperm respiration and motility. Biochemistry, 10: 1825-1831. MOHRI, H., 1956. Studies on the respiration of sea-urchin spermatozoa. II. The cytochrome oxidase activity in relation to the dilution effect. /. E.rp. Biol., 33(2) : 330-337. MORISAWA, M., AND H. MOHRI, 1972. Heavy metals and spermatozoan motility. E.rp. Cell Res., 70: 311-316. NELSON, L., 1972. Quantitative evaluation of sperm motility control mechanisms. Biol. Reprod.,6: 319-324. RICHARDS, F. A., 1972. Table 61. Pages 467-469 in P. L. Altnian and D. S. Dittmer, Eds., Biological Data Book, volume 1. [2nd edition] FASEB, Bethesda, Maryland. ROBBIE, W. A., 1946. The quantitative control of cyanide in manometric experimentation. /. Cell. Comp. Physiol., 27 : 181-209. ROTHSCHILD, LORD, 1948. The physiology of sea urchin spermatozoa. Lack of movement in spermatozoa. /. E.vp. Biol., 25 : 344-352. ROTHSCHILD, LORD AND P. H. TUFT, 1950. The physiology of sea-urchin spermatozoa. The dilution effect in relation to copper and zinc. /. Exp. Biol., 27 : 59-72. ROTHSCHILD, LORD AND A. TYLER, 1954. The physiology of sea-urchin spermatozoa. Action of versene. /. Exp. Biol., 31 : 252-259. TIMOURIAN, H., AND G. WATCHMAKER, 1970. Determination of spermatozoan motility. Develop. Biol., 21 : 62-72. TYLER, A., 1953. Prolongation of life-span of sea urchin spermatozoa and improvement of fertilization-reaction, by treatment of spermatozoa and eggs with metal-chelating agents (amino acids, versene, DEDTC, oxine, cupron). Biol. Bull., 104: 224-239. TYLER, A., AND B. S. TYLER, 1966. The gametes ; some procedures and properties. Pages 639-682 in R. A. Boolootian, Ed., Physiology of Echinodermata. J. Wiley, Inc., New York. UTIDA, S., K. MARUYAMA AND S. NANAO, 1956. The effect of zinc on the apyrase activity in suspensions of the tail of sea-urchin spermatozoa. Jap. J. Zool., 12 : 19-23. Reference: Biol Bull, 147: 247-256. (August, 1974) THE PROCESS OF EGG-LAYING IN THE CHAETOGNATH SAG ITT A HISPID A * M. R. REEVE AND B. LESTER Riisenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida 33149 The process of egg-laying was described in Sagitta bipunctata by Boveri (1890), Stevens (1905, 1910), Bordas (1920) and Ghirardelli (1968), in S. hisplda by Conant (1896), in S. hc.raptcra by Grassi (1883), in S. elegans by Stevens (1905, 1910) in S. sctosa by Dallot (1967, 1968), in S. Inflata by Ghirar- delli (1968), and in Spadella cephaloptera by Vasiljev (1925) and Ghirardelli (1968). These authors do not agree as to how the ovum is fertilized by the sperm, from which it is separated by a layer of germinal epithelium known as the crescent. Disagreement also exists concerning the manner in which the fertilized egg reaches the exterior, the nature of the opening through which it leaves the body, and even on the number of ducts in the female reproductive system. Reeve and Walter (1972a) described mating in Sagitta hisplda, during which one or both spermatophores may be exchanged, following which the sperm migrate down the body of the animal to the nearest gonopore. Upon reaching it, they separate into two groups, one of which enters the adjacent gonopore while the other crosses over the body to the gonopore on the opposite side. The sperm enter the gonopores and move rapidly up a duct which lies on the outer side of each ovary. The observations reported below describe events beyond this point which result in the laying of eggs, and indicate that in Sagitta hisplda, at least, there is one tube which serves both as a sperm duct and an oviduct, and that the eggs are extruded by contractions of the ovary wall and body wall. METHODS Sagitta hisplda Conant, a neritic species of chaetognath of approximately 10 mm mature length, is readily obtainable in plankton tows made from the laboratory dock on Biscayne Bay (Miami), and can be cultured in the laboratory for periods of several weeks. The detailed methodology of culture was provided by Reeve and Walter (1972b). In summary, animals were maintained in transparent acrylic 80-liter aquaria and fed natural zooplankton. The observations reported below were made on many such populations over a period of two years by withdrawing orga- nisms from aquaria, placing them in small, shallow dishes or upon slides, and observing them live under the microscope. Of hundreds of animals examined in this way, numerous photomicrographs were made of organisms at various stages in the sexual cycle. In some cases individual animals were retained for several hours up to 2 days in 250-ml containers for further observation on the progress 1 Contribution of the Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida 33149. 247 248 M. R. REEVE AND B. LESTER of egg-laying in the live animal. At least 30 animals, having been examined at various stages of egg-laying, were fixed in Bourn's solution, sectioned (8 p.) and stained with eosin and haematoxylin. This material was examined and photographed up to magnifications of X 1000. The external appearance of the gonopore region was also examined by scanning electron microscopy after preparation as described by Cosper and Reeve (1970). Observations on egg deposition were made on ma- ture populations of chaetognaths in 80-liter aquaria into which had been transplanted three of the commonest local benthic plants (Thalassia testudinniu, Syringodium filijonnc and Diplanthera wrightii). The surfaces of the aquaria and of the plants (the latter having been removed to dishes) were examined for the presence of eggs after 48 hours. OBSERVATIONS Fertilisation Following mating, sperm can be seen in live animals examined microscopically (see figures in Reeve and Walter, 1972a) making vigorous movements in the duct running along the outer side of each ovary, and in prepared sections (Figs. 1A and B). Although the act of fertilization was not seen, at the cellular level, it appeared to be effected by passage of the sperm through a stalk consisting of two cells (Fig. 1A) which appear, in section, to be highly vacuolar, and somewhat in- distinct at their boundaries, as though on the point of degenerating. The stalk, (accessory fertilization cells, see Discussion) is part of the germinal epithelium from which the immature ovum grows out, and through which it maintains an attachment to the epithelium as it develops. Migration of mature eggs out of the ovary The egg in Figure IB is fully mature and about to migrate from ovary to duct; it is separated from the duct by only a thin strand of germinal epithelium. In order to reach the duct, the egg must pass through the germinal epithelium. It appears to do this following degeneration of the stalk cells, or their being engulfed by the egg, by insinuating itself very gradually (Figs. 1C-F) through the resulting pore. First the egg becomes oval, rather than spherical, and then it becomes pear shaped. The small lobe of the pear is pushed through the pore and into the duct. Once through, it is gradually enlarged until the major part of the egg has migrated, the egg being somewhat dumb-bell shaped during this process. Then the remainder of the egg slips through, till the migration is complete. This process seems to us to be accomplished by several means. The egg itself appears to make squeezing movements, though these are perhaps of too precise a nature to be termed ameboid. Also we have observed peristaltic waves of contrac- tions passing along the ovary from anterior to posterior together with the appear- ance of bands circumscribing the ovary wall (Fig. 1G). Possibly these contrac- tions help to propel the eggs through the crescent. In addition, pressure of the gut and even the normal muscular contractions of swimming movements may aid the eggs in their migration. The time required for all the mature eggs to complete their migrations, which are performed randomly rather than in unison, varies from 15 to 20 minutes. EGG-LAYING IN SAGITTA 249 FIGURE 1. A— maturing egg; B— mature egg; C to F— stages in migration of egg to duct; G — ovary with mature unmigrated eggs showing contractions of ovary walls; H — eggs in duct immediately prior to laying. Figures A-C and H are from sections, D-G from live material ; width of body approximately 0.5 mm ; a — accessory fertilization cells ; b — body wall ; d — duct ; e— maturing egg; ei— mature unmigrated egg; 62— migrating egg; e3— migrated egg; cw— egg cell wall; g— germinal epithelium; go— gonopore ; gu— gut ; s— sperm. 250 M. R. REEVE AND B. LESTER As more and more eggs pass through the germinal epithelium into the duct, they become crowded (Figs. ID, E and F), possibly because the lumen of the duct is narrower than the diameter of the eggs. They are compressed, as Conant (1896; page 82) described it, "from before backward" until their shape is cylindri- cal rather than spherical. When all the eggs have migrated, the duct is completely filled with eggs (Fig. 1H) so that, viewed from above, it has the appearance of a ladder, the boundaries between two adjacent eggs froming the rungs, and the walls of the duct and the outer boundaries of the eggs forming the uprights of the ladder. The immature eggs in the ovary are compressed against the gut (Figs. 1H and 2B). The animal will now lay its eggs within 10 or 15 minutes. In sections of individuals fixed at this stage there is no evidence of any duct other than the one occupied by the eggs (Figs. 1H and 2B). Animals which are almost ready to lay eggs often attach to the side wralls of the aquarium but they do not necessarily remain continuously attached till the laying is accomplished, but rather shift position every few minutes. Even animals whose eggs are migrating tend to move sporadically about, and we suggest that the muscular contractions of the body wall may aid in compressing the ovaries so that the eggs may more easily migrate from ovary to duct. Tlic duct In sections of inseminated animals whose eggs have already migrated through the germinal epithelium, sperm may be seen directly adjacent to the cell wall of the egg (Figs. 2B and C). In addition to the direct visual evidence that sperm are not separated from eggs when the latter emerge on the outer side of the crescent, at 1000 magnifications, even under low power sperm bulges can be seen where they are trapped betwreen two adjacent eggs or forced anteriorly into the blind end of the duct. Indeed, it is not unusual to see bundles of motile sperm extruded to the exterior (Fig. IF) as eggs migrate through the crescent. There is, therefore, no rvidence that eggs and sperm occupy different ducts. Tlic gonopore Observations of the gonopore, both in sections (Fig. 2B) and with the scanning electron microscope, indicate that it has only one opening, and that only one duct connects with it. Sections of animals in which eggs have migrated prior to being laid (Figs. 1H, 2A and B) clearly show the terminal egg in the cavity of a duct continuous with the opening of the gonopore, and where sperm are present (Fig. 2B) show them to occupy the same cavity. Expulsion of mature eggs Our observations indicate that the eggs are expelled by a sudden and intense contraction of the ovary and body walls. We are indebted to our colleague, M. A. Walter, for a photograph of a live animal (Fig. 2D ) in the process of expelling its eggs, in which the body wall contractions are clearly visible. Contractions of the ventral body-wall musculature in the region of the gonopores "(Fig. 2A) is pre- sumably also involved, by expanding the orifice of the gonopores so that the eggs EGG-LAYING IN SAGITTA 251 J ew- F FIGURE 2. A — eggs in process of being laid from gonopore ; B — eggs about to be laid adjacent to gonopore; C — sperm and eggs in duct; D — eggs being laid; E — ovary showing con- traction bands after egg-laying ; F — mucus-embedded eggs attached to grass blade. Figures A- C are from sections, D-F from live material. Diameter of eggs approx. 0.2 mm. e4 — eggs in process of being laid; e5 — attached (laid) eggs; gb — grass blade; m — muscle; mu — mucus, Other abbreviations as in Figure 1. M. R. REEVE AND B. LESTER do not have to deform their shape (Fig. 2A) as much as they did in order to pass through the crescent. The eggs issue in two streams from each gonopore (Fig. 2D). and not, as Conant (1896) reported, one from each gonopore, so that the egg mass ultimately has the appearance of four strands of heads. The eggs are laid about ten or fifteen minutes after the last one has migrated through the germinal epithelium (at 26° C). Following laying, the ovaries appear crumpled and contracted in an anterior- posterior plane; they are wide, filling the diameter of the body cavity but not even extending as far forward as the anterior of the posterior fin (Fig. 2F). They m i\ be distinguished from those of a less mature animal which has yet to lay eggs (see Reeve and Cosper, 1974), for in the latter the ovaries are narrow and smooth- walled. As noted elsewhere (Reeve and Copper. 1974) ,V()5) proposed the existence of two accessory fertilization cells, the outermost being inserted in the wall of the "oviduct" and the inner being inserted in the ovum. The sperm swims through a series of vacuoles in these cells from the "oviduct" to the ovum, Ghiradelli (1968) confirmed these observations in Sai/itla bipitnctata and Spadclla cephalaptcra. Our observations confirm those of Conant (1896) who suggested that the developing oocytes are attached to the ovary by a stalk (Fig. 1A) and contradict those of Stevens ( 1903 ) who maintained that the eggs develop free inside the ovary and then "each one becomes connected with two of the epithelial cells which are just lateral to the region" (page 232). This stalk later serves as a passage for the sperm to the egg as suggested by Stevens (1903), Buchner (1910), Vasiljev (1925) and Ghirardelli (1968). Following fertilization, and indeed, even in its absence (Reeve and Cosper, 1974) the mature eggs move from the ovary into the duct which leads to the gono- pore. Since they make this migration irrespective of whether copulation and fertilization has occurred, this process must be triggered by the state of maturation of the egg itself and not by its penetration by the sperm. In Sagitta the egg must pass through the germinal epithelium of the crescent into the duct by which it gains access to the exterior. Exactly how the egg moves, or is moved, is uncertain, although Ballot (1968) has reported seeing con- tractions of the ovary wall, confirming Conant's (1896) earlier observations. Ghirardelli (1968) and Stevens (1905, 1910), however, maintained that the egg makes its way into the duct by ameboid movements. In Sagitta hispida the move- ment from ovary to duct appears to be partly active and partly by pressure exerted by the ovary walls and possibly other tissues. In Spadella. Ghiradelli (1968) EGG-LAYING IN SAGITTA 253 indicated that the eggs moved posteriorly in the ovary before entering a very short terminal duct. Perhaps the greatest source of controversy concerns the structure of the female reproductive organs, and in particular the nature and position of the duct which the egg enters on leaving the ovary. Hertwig (1880) wrote that the ovaries con- sisted of the ovary or egg tube and the oviduct. Conant (1896), working with Sagitta hlspida, suggested that there was an oviduct (separate from the sperm duct) lying between the germinal epithelium and the epithelium of the sperm duct. According to his observations, after the eggs were fertilized, they moved through the germinal epithelium into the oviduct, which was a temporary structure. Stevens (1910) in studying ,V. bif>nnctata, stated that "both living material and sections show that the sperm duct and oviduct are entirely separate for their whole length, each having its own opening to the exterior" (page 284). She con- firmed Conant's (1896) observations concerning the temporary oviduct. Later authors fell into one or other of the two camps. Bordas (1920), while working on the same species studied by Stevens, stated that the single tube, although serving as a sperm duct, is principally an oviduct. Vasiljev (1925) referred to a sperm duct and mentioned elastic fibers running along the length of this duct, which, he said, serve to keep the sperm duct extended. He was unable to find anything resembling a provisional or temporary oviduct, and stated that in .Y/W<7/72a ) showed that insemina- tion confers fertility on one batch nf eggs only, subsequent eggs being infertile when laid, if the animal is prevented from being reinseminated. It would seem that a primary advantage of a separate sperm duct would be to store viable sperm for fertilization of ova which had yet to mature. Stevens (1910) suggested that eggs reached the exterior by their own active movements, crawling along the duct to the gonopore. Ameboid movements were also observed by John (1933) and Ghirardelli (1968). Conant (1896) described contractions of the ovary wall (which surrounds the developing ova, germinal epithelium and duct) as providing the force for the expulsion of eggs in Sagitta hispida and Dallot (1968) confirmed this for S. sctosa. Grass! (1883), Stevens (1910) and Vasiljev (1925) had noted the presence of elastic fibers. Our observa- tions showed that the eggs, which become cylindrically compressed in the duct, are extruded very rapidly by a combination of muscular forces, which includes the body wall musculature as well as the ovary itself. In Sagitta his pitta, we observed, as already noted, that eggs leave the body via the single ovarian duct through the gonopore. Regarding other members of the genus, Vasiljev (1925) believed that eggs left the temporary oviduct by being ex- truded through a rupture in the body wall. From the figures of Ghirardelli (1968) it appears that a temporary opening to the oviduct occurs adjacent to the opening of the sperm duct at the gonopore. In Spadclla, Ghirardelli (1968) indicated by description and a figure that eggs pass down to the posterior end of the ovary before entering a short temporary oviduct formed by a splitting of the two layers of the wall of the terminal portion of the sperm duct, and exit by a temporary opening to the exterior. John (1933) had previously indicated that the eggs penetrated into the seminal receptacle (the expanded portion of the sperm duct at the gonopore). Sagitta liispida is a neritic species (Owre, 1960) and is the only planktonic chaetognath which has adapted to the semi-estuarine conditions of Biscayne Bay. The behavioral pattern of attaching eggs presumably assists in maintaining the Biscayne Bay population, since that part of the life cycle during which either the mature adult or the egg is attached is less susceptible to expatriation (an inherent EGG-LAYING IN SAGITTA 255 danger of a planktonic existence) into the Florida Current adjacent to Biscayne Bay. This work was supported by National Science Foundation Grant number GA- 28522 (Oceanography Section). SUMMARY 1. There is considerable disagreement in the literature concerning the manner in which eggs are fertilized and reach the exterior in chaetognaths, even in studies on the same species. This studv makes use of observations on populations of living animals maintained in the laboratory, and of material fixed at known stages of the egg-laying process, and provides photographic evidence to illustrate the points enumerated below. 2. There is a single duct, serving both as an oviduct and sperm duct in the female reproductive system of the chaetognath Sagitta hispida, which runs from the gonopore anteriorly along the outer border of each ovary. 3. Following fertilization eggs slowly pass through the germinal epithelium (crescent) through pores much smaller than their own diameter into thi.s duct, and displace the sperm in it. 4. The eggs, which become cylindricallv compressed, are extruded very rapidly from the duct to the exterior by a combination of muscular forces, and deposited in a mucous matrix onto a surface to which the animal is attached. LITERATURE CITED BORDAS, M., 1920. Estudio de la ovogenesis de la Sagitta bipitncfata. Trab. Mas. Nac. Cicnc. Natr., Madrid, Scr. Zool., 42 : 5-119. BOVERI, T., 1890. Zcllenstudien III. Jena. Z. Nature-its., 24: 314-401. BUCHNER, P., 1910. Keimbahn und Ovogenese von Sagitta. Anal. Anz., 35: 433-443. BURFIELD, S. T., 1927. Sagitta. Liverpool Afar. Biol. Comm. Mew., 27: 1-104. COSPER, T. C., AND M. R. REEVE, 1970. Structural details of the mouthparts of a chaetognath, as revealed by scanning electron microscopy. Bull. Mar. Sci., 20: 441-445. CONANT, S., 1896. Notes on the Chaetognaths. Ann. Mag. Natur. Hist., 6: 201-204. DALLOT, S., 1967. La reproduction du Chateognathe planctonique Sagitta sctosa Miiller, en etc, dans la rade de Villefranche. C. R. Acad. Sci., Paris, 264: 972-974. DALLOTT, S., 1968. Observations preliminaries sur la reproduction en elevage du Chaetognathe planctonique Sagitta sctosa Miiller. Rapp. Comma. Int. Mer. Mcdit., 19: 521-523. GHIRARDELLT, E., 1968. Some aspects of the biology of the chaetognaths. Advan. Mar. Biol., 6: 271-375. GRASSI, G. B., 1883. I Chetognati. Fauna Flora Golf Neapcl, 5 : 1-116. HERTWIG, O., 1880. Uber die Entwicklungsgeschichte der Sagitten. Jena. Z Natiiriviss., 14 : 196-303. JOHN, C. C., 1933. Habits, structure and development of Spadclla ceplialoptcra. Quart. J. Microscop. Sci. 75 : 625-696. JOHN, C. C., 1943. Chaetognatha. Structure of the reproductive organs of Sagitta. Proc. Indian Sci. Congr., 30(Part III, Ser. VI, Zool.): 71. OWRE, H. B., 1960. Plankton of the Florida current. Part VI. The Chaetognatha. Bull. Mar. Sci. Gulf Carib., 10 : 255-322. REEVE, M. R., AND T. C. COSPER, 1974. Chaetognatha. In press in A. C. Giese and J. S. Pearse, Eds., Reproduction in Marine Invertebrates, Vol. 2. Academic Press, New York. 256 \l. R. REEVE AND B. LESTER REEVE, M. R., AND M. A. WALTER, 1972a. Observations and experiments on methods of fertilization in the Chaetognath Sagitta liispida. Biol. Bull, 143: 207-214. REEVE, M. R., AND M. A. WALTER, 1972b. Conditions of culture, food size selection and the effects of temperature and salinity on growth rate and generation time in Saijilta liispida Conant. /. Exp. Mar. Biol. Ecol, 9 : 191-200. STEVENS, N. M., 1903. On the ovogenesis and spermatogenesis of Sagitta bipunctata. Zool. Jahrb. (Anat.), 18: 227-240. STE\TA-. X. M., 1905. Further Indies on the ovogenesis of Sagitta. Zool. Jahrb. (Anat.}, 21 : 243-252. STEVENS, N. M., 1910. Further studies on reproduction in Sagitta. J. Morphol., 21: 273-319. VASILJEV, A., 1('25. La fecundation chez Spmh'lln ccplidloptcra LGRHS. ct 1'origine du corps determinant la voie germinative. Biologia Gen., 1 : 249-278. Continued from Cover Two 4. Literature Cited. The list of references should be headed LITERATURE CITED, should conform in punctuation and arrangement to the style of recent issues of THE BIOLOGICAL BULLETIN, and must be typed double-spaced on separate pages. Note that citations should include complete titles and inclusive pagination. Journal abbreviations 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 below. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS of those abstracted (most recent issue: November, 1972). Foreign authors, and others who are accustomed to use THE WORLD LIST OF SCIENTIFIC PERIODICALS, may find a booklet published by the Biological Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K. at £0.65 or $1.75) useful, since it sets out the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these differ. CHEMICAL AB- STRACTS 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 period, whole word components must not (not strictly as USASI usage, i.e. J. Cancer Res.) D. Space between all components (e.g. J. Cell. Comp. Physiol. not J.Cell.Comp. Physiol.) E. We strongly recommend that more unusual words in journal titles be spelled out in full, rather than employing lengthy, peculiar "abbreviations" or new abbreviations invented by the author. For example, use Rit Visindafjelags Islendinga without abbreviation. Even in more familiar languages, Z. Vererbungslehre is preferred to Z. VererbLehre (WORLD LIST) or Z. Verer- bungsl. (USASI). Accurate and complete communication of the reference is more important than minor savings in printing costs. F. All single word journal titles in full (e.g. Veliger, Ecology, Brain). G. The order of abbreviated components should be the same as the word order of the com- plete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. Spell out London, Tokyo, Paris, Edinburgh, Lisbon, etc. where part of journal title. I. Series letters etc. immediately before volume number. J. A few well-known international journals in their preferred forms rather than WORLD LIST or USASI usage (e.g. Nature, Science, Evolution NOT Nature, Land.; Science, N. Y.; Evolution, Lancaster, Pa.) K. The correct abbreviation for THE BIOLOGICAL BULLETIN is Biol. Bull. 5. Figures. The dimensions of the printed page, 5 by 7f inches, should be kept in mind in preparing figures for publication. Illustrations should be large enough so that all details will be clear after appropriate reduction. Explanatory matter should be included separately in legends as far as possible, although the axes should always be numbered and identified on the illustration itself. Figures should be prepared for reproduction as either line-cuts or halftones; no other methods will be used. Figures to be reproduced as line-cuts should be 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, and both designating numbers or letters and scale-bars should be affixed directly on the figures. We recommend that halftones submitted to us be mounted prints made at about If times the linear dimensions of the final printing desired (the actual best reductions are achieved from copy in the range from \\ to 2 times the linear dimen- sions). As regards line-blocks, originals can be designed for even greater reductions but are best in the range \\ to 3 times. All figures should be numbered in consecutive order, with no distinc- tion between text and plate-figures. The author's name should appear on the reverse side of all figures, and the inked originals for line-blocks must be submitted for block-making. 6. Mailing. Manuscripts should be packed flat. All illustrations larger than 8J by 11 inches must be accompanied by photographic reproductions or tracings that may be folded to page size. Reprints. Reprints may be obtained at cost; approximate prices will be furnished by the Managing Editor upon request. CONTENTS Annual Report of the Marine Biological Laboratory .................... 1 CALOW, P. AND J. B. JENNINGS Calorific values in the phylum Plalyhelminthes : the relationship between potential energy, mode of life and the evolution of ento- parasitism ................................................... 81 HEFFERNAN, J. MICHAEL AND STEPHEN A. WAINWRIGHT Locomotion of the holothurian Euapta lappa and redefinition of peristalsis ................................................... 95 PARDY, ROSEVELT L. Some factors affecting the growth and distribution of the algal endosymbionts of Hydra viridis ................................ 105 POSTLETHWAIT, JOHN H. Juvenile hormone and the adult development of Drosophila ....... PRICE, C. A., L. R. MENDIOLA-MORGENTHALER, M. GOLDSTEIN, E. N. BREDEN, AND R. R. L. GUILLARD Harvest of planktonic marine algae by centrifugation into gradients of silica in the CF-6 continuous-flow zonal rotor ................. 136 READ, CLARK P., GEORGE L. STEWART, AND PETER W. PAPPAS Glucose and sodium fluxes across the brush border of Hymenolepis diminuta (Cestoda) .......................................... 146 SlVASUERAMANIAN, P., S. FRIEDMAN AND G. FRAENKEL Nature and role of proteinaceous hormonal factors acting during puparium formation in flies .................................... 163 SMITH, SCOTT, JAMES OSHIDA AND HANS BODE Inhibition of nematocyst discharge in hydra fed to repletion ....... 186 STOFFEL, Lois A. AND JERRY H. HUBSCHMAN Limb loss and the molt cycle in the freshwater shrimp, Palaemo- netes kadiakensis ........................................... 203 SUGIMOTO, KEIJI AND MlTSUAKI NAKAUCHI Budding, sexual reproduction, and degeneration in the colonial ascidian, Symplegma reptans .................................. 213 WHITLATCH, ROBERT B. Food-resource partitioning in the deposit feeding polychaete Pectinaria gouldii ............................................ 227 YOUNG, L. G. AND L. NELSON The effects of heavy metal ions on the motility of sea urchin spermatozoa ................................................. 236 REEVE, M. R. AND B. LESTER The process of egg-laying in the chaetognath Sagitta hispida ...... 247 Volume 147 Number 2 THE BIOLOGICAL BULLETIN '- PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY 5 ' • Editorial Board JOHN M. ANDERSON, Cornell University F. H. RUDDLE, Yale University JOHN O. CORLISS, University of Maryland BERTA SCHARRER, Albert Einstein College of Medicine DONALD P. COSTELLO, University of North Carolina HOWARD A. SCHNEIDERMAN, University of California, Irvine PHILIP B. DURHAM, Syracuse University M£LVIN SpIEGELt Dartmoutn College CATHERINE HENLEY, University of GROVER C. STEPHENS, University of North Carolina California, Irvine GEORGE O. MACKIE, University of Victoria EDWARD O. WILSON, Harvard University W. D. RUSSELL-HUNTER, Syracuse University Managing Editor OCTOBER, 1974 Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA. 0 THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Pennsylvania. Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, Woods Hole, Massachusetts. Agent for Great Britain: Wheldon and Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2. Single numbers, $7.00. Subscription per volume (three issues), $18.00, (this is $36.00 per year for six issues). Communications relative to manuscripts should be sent to Dr. W. D. Russell-Hunter, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between May 23 and September 1, and to Dr. W. D. Russell-Hunter, P.O. Box 103, University Station, Syracuse, New York 13210, during the remainder of the year. Copyright © 1974, by the Marine Biological Laboratory Second-class-postage paid at Lancaster, Pa. INSTRUCTIONS TO AUTHORS %% THE BIOLOGICAL BULLETIN accepts original research reports of intermediate length on a variety of subjects of biological interest. In general, these papers are either of particular interest to workers at the Marine Biological Laboratory, or of outstanding general significance to a large number of biologists throughout the world. Normally, review papers (except those written at the specific invitation of the Editorial Board), very short papers (less than five printed pages), preliminary notes, and papers which describe only a new technique or method without presenting substantial quantities of data resulting from the use of the new method cannot be accepted for publication. A paper will usually appear within four months of the date of its acceptance. The Editorial Board requests that manuscripts conform to the requirements set below; those manuscripts which do not conform will be returned to authors for correction before review by the Board. 1. Manuscripts. Manuscripts must be typed in double spacing (including figure legends, foot-notes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 8$ by 11 inches. They should be carefully proof-read before being submitted and all typographical errors corrected legibly in black ink. Pages should be numbered. A left-hand margin of at least 1J inches should be allowed. 2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and placed in correct sequence in the text. Because of the high cost of setting such material in type, authors are earnestly requested to limit tabular material as much as possible. Similarly, foot- notes to tables should be avoided wherever possible. If they are essential, they should be indi- cated by asterisks, daggers, etc., rather than by numbers. Foot-notes are not normally permitted in the body of the text. Such material should be incorporated into the text where appropriate. Explanations of figures should be typed double-spaced and placed on separate sheets at the end of the paper. 3. A condensed title or running head of no more than 35 letters and spaces should be included. Continued on Cover Three Vol. 147, No. 2 October, 1974 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY ALLOMETRIC STUDIES ON ENERGY RELATIONSHIPS IN THE SPIDER CRAB LIBINIA EMARGINATA (LEACH)1 Reference: Biol. Bui!., 147: 257-273. (October, 1974) JOHN CARLSON ALDRICH 2 Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 Allometry, in the extended sense employed by Gould (1966), is the study of variables exhibiting differential growth. These variables may be morphological (e.g., width/length), physiological (e.g., size/oxygen consumption), or chemical (('.(/., size/lipid weight) ; and the relationships may be dynamic, as in successive measurements made on a growing animal, or they may be static, as in comparisons between non-growing adults. Changes in the rates of differential growth have been found to accompany matur- ity in both vertebrates and invertebrates (Teissier. 1931, 1960; Needham, 1942; Brody, 1945; von Bertalanffy, 1957). In decapods, developing maturity is manifested at the molts of prepuberty and puberty (Hartnoll, 1963). As an animal passes from an immature and actively growing stage to a mature, non-growing and reproducing stage, it is likely that its energy requirements change. Further, if an animal is found in different habitats correlated with suc- cessive stages of growth, it is likely that the energy available to it also changes. A succession of growth stages and their associated habitat changes have been termed growth stanzas (Parker and Larkin, 1959). It seemed likely that the energy requirements of a crab would be reflected in the differential growth of variables sensitive to energy requirements, with these differences especially marked when successive growth stanzas were compared. Now, allometry facilitates the comparison of greatly varying sizes by reducing constant relative growth rates to a power function. The relationship of Y to X is allometric when Y = bX" ; where a is the rate of change of Y with respect to X, and b is the intercept, or size of Y when X — 1 (Huxley, 1924). Such rela- tive growth rates appear on double-logarithmic plots as straight lines, with changes 1 This study formed part of a master's thesis presented at Boston University in 1972. Support was provided by Boston University in the form of teaching fellowships during 1970- 1972, and a summer stipend in 1971. 2 Present address : Department of Zoology and Comparative Physiolgy, Queen Mary Col- lege (University of London), Mile End Road, London El 4NS, England. 257 Copyright © 1974, by the Marine Biological Laboratory T iHrar^r f\f f^nncrrfcc f~"orr1 Mn JOHN CARLSON ALDRICH in rates shown by inflections or discontinuities (Teissier, 1960), as is the case with morphological allometry during the transition from immature to mature stages in Maiid crabs (Hartnoll, 1963). Hemmingsen ( 1950) suggested a theoretical limitation of metabolic rate based on the absorptive area of the digestive apparatus. For a crab such a relationship would be shown by similar constants for the slopes of oxygen consumption, stomach capacity, and hepatopancreas weight, this latter being the chief organ of digestion ( Youk. 1960). Such a comparison can be used for other purposes, especially if the sizes of the stomach (or its capacity), and the hepatopancreas (or its lipid content) can be shown to have a maximum likely value at a given size. Such maximum values would be represented by the upper bound of a scattergram of individual points. This is especially valid if the upper bound has the same slope as the overall average. Relative energy flow within a given growth stanza could then be shown by the average size of the organ measured, relative to the regres- sion of the maximum size. The organ in crabs that is apparently most sensitive to energy flow is the hepatopancreas, with lipids being the most labile constituent, probably reflecting overall energetic demands (Armitage, Buikema and Willems, 1972). This has been extensively studied in relation to the molt cycle by Drach (1939) and Renaud (1949), and has been looked at over growth stages in general in the crayfish Orconccfcs nais (Armitage ct al, 1972). This study was therefore designed to examine the allometric relationships of stomach capacity, the adequacy of natural feeding rates, and the size and lipid content of the hepatopancreas ; both as related to the state of maturity of the crab reflected by its energetic demands, and the capacity of its habitats to supply these demands. The work done is presented in two major parts. In the first part (morphological allometry) the growth stages are defined by their allometric con- stants of external morphology, and by their associated habitats. In the second part (physiological allometry), oxygen consumption, stomach capacity and the pro- portion used, hepatopancreas total and lipid weights, average caloric intake, the cycling rate of stomach contents, and the efficiency of assimilation are treated, and the relative proportions of usage or energy flow computed for the stages previously described. METHODS Crabs Specimens of Libinia ctnarginata (Leach), (Maiidae: Pisinae) the common spider crab of the east coast of the United States, were collected in the shallow waters within a radius of ten miles of Woods Hole, Massachusetts. Immature crabs, from 0.2 to 28 g were usually collected in masses of chlorophytous alga. The crabs clung to this coarse alga and covered themselves with finer filamentous chlorophytes and rhodophytes. Many of these crabs were found in a soft, newly-molted condition, indicating active growth in this habitat. Mature crabs were always collected on relatively barren mud or gravel bottoms, predominantly the former. These larger crabs ranged from 28 to 480 g. A naturalist's dredge was used to collect the immature crabs and an otter trawl for the mature ones. Specimens used for organ analyses were killed and examined within a few hours ALLOMETRY AND ENERGETICS OF LIBIA' 1. 1 259 of collection, others were kept in running seawater tanks and were used for feeding and respiration experiments. Throughout the testing and collecting period of July-September, 1971, the ambient seawater temperature remained bet ween 20-22° C. Allometry General. When plotted on double-logarithmic axes, an allometric relation is shown by a straight line having the equation; log Y - a log X + log b; where a is the slope of the allometric line and can be measured directly on the plot (Teissier, 1960). Such plots were used initially to identify relationships, but for purposes of statistical comparisons, least squares regressions were computed and F tests of slope (a) and level (b} made between lines. Strictly speaking, none of the regressions can be said to have an independent and a dependent variable, instead they covary with both variables probably under the same influence of differential growth. However, there is a much greater chance of error in measuring lipid weight or oxygen consumption, than there is in measuring total weight or carapace length. Therefore, the standard regression calculation has been made with the more erratic variable being the dependent one. This method has been used for morphological allometry as well, but here the very high correla- tion coefficients make it unlikely that any serious error is introduced by not using standard allometric methods (for review see Kidwell and Chase, 1967). In the case of morphological allometry, both a and b are useful in describing growth stages, and both have been reported for many Maiid crabs (e.g., Hartnoll, 1963). In the case of physiological allometry, the comparisons between variables are made on the basis of the slopes (o) ; testing the similarity of differential growth. The value of b has no meaning when compared here between different variables since it is based on different units and origins. But, when a single variable is compared throughout different growth stages, the value of b represents relative levels of usage in each stage. (Conversely, in this case a has little meaning. ) Morphological. Three identifiable growth stages exist in Maiid crabs ; im- mature, prepubescent, and mature (Hartnoll, 1963). These stages are identified by two important points during the growth of the animal (Teissier, 1935) : (1) the molt of prepuberty, where the immature proportions of the crab begin their trans- formation into those of the adult. This point is marked by an inflection in the allometric line, and is coincident with the first appearance of gonads. Vernet- Cornubert (1958), working with Pisa tetraodon, showed that this inflection appears at varying individual sizes. (2) The molt of puberty, or in Maiid crabs, the terminal molt (Carlisle, 1957) ; at this point the crab is fully mature and the plot of allometric lines exhibits a discontinuity representing the terminal changes in proportions. The most striking morphological allometries in crabs are those between the length of the male chelipeds (or propodus thereof) and the length of the carapace; and the width of the female abdomen, and the width of the carapace. These measurements were used after the fashion of Teissier (I960) and Hartnoll (1963) to establish the three stages in this crab. Size measurements were made to the nearest 0.5 mm with a caliper. 260 JOHN CARLSON ALDRICH 100 50 O) c 2 T3 O 0. O 0. b 10 10 O" Length 50 100 100 50 10 c (D E O TJ JQ O 10 9 Width Carapace 50 FIGURE 1. Morphological allometry, male propodus/carapace length and female abdo- men/carapace widths. Lines are: males, "a" immature, "b" prepubescent, "c" mature; female--, "d" immature-prepubescent, "e" mature. Small dots are individual values ; large dots are size class averages. In the case of the female abdomen/carapace allometry, the difference between molting and non-molting stages was apparent upon constructing a scattergram (Figure 1) and they could be collected into immature-prepubescent and mature stages for the computation of allometric constants. A preliminary computation based on size classes (shown in the figure) gave some hint of the prepubescent stage, but this was neither distinguishable in individual crabs nor statistically valid for the purposes of computing constants. The males presented a more difficult case, although the immature and prepubescent sizes were rapidly distinguishable in the scattergram, much of their size ranges intergraded imperceptibly. How- ever, the smallest prepubescent and the largest mature sizes were evident beyond the intergrading region, and the very smallest mature sizes were identified by their mature gonads and also by their having three layers of barnacles, an indica- ALLOMETRY AND ENERGETICS OF LIBINIA 261 tion that they had not molted for several years (Hartnoll, 1963). (The subtidal barnacle found on these crabs, Balanus ainphitritc niveus, sets only once each year according to Ray, 1959). These smallest and largest sizes were used as guides in drawing lines through the uncertain region, separating individual points into prepubescent and mature stages, with ambiguous points arbitrarily divided between the two lines. When all of the points had been assigned to their respective stages, final allometric constants were computed. Stomach capacity. The maximum stomach capacity was estimated by plotting whole stomach weights (cardiac and pyloric regions together) both as found in newly-collected specimens and in some fed to satiety. Feeding was necessary in large crabs because they were never collected with full stomachs. After weighing whole, the stomachs were flushed and blotted and the empty weight recorded. Separate regression lines were computed for the maxium points, and for the empty points, the difference between these lines at a given size being taken as the maximum capacity. For purposes of computation, the specific gravity of stomach contents was considered to be unity so that one gram was equivalent to one cc of contained food. O.vygcn consumption. This was measured in closed containers, the dissolved oxygen not being allowed to decrease to less than '} of the ambient level (Stroganov, 1964). Measurements were obtained for the entire size range of Libinia, from newly-released zoeae, to the largest (450 g) mature males (Rathbun, 1925). Such a millionfold span of weight decreases the erratic value for a that can be found when using small ranges of weight (Hemmingsen, 1960). Large crabs were measured singly, crabs less than 5 g were measured in groups, and about a thousand zoeae were measured simultaneously. The length of the experiments varied from half an hour in the largest crabs (measured in a sub- merged wash basin with a plexiglas lid), to several hours in the smallest crabs (measured in submerged plexiglas boxes). These measurements were made only upon freshly-collected, non-molting specimens (23), and were considered to repre- sent the routine rate (Bayne, 1973). Hepatopancrcas. This was weighed wet without draining, and portions of tissue frozen for later lipid analysis by the chloroform-methanol method of Folch, Lees and Sloane-Stanley (1957). Subsequent calculations were based on wet weight. Stomach contents. The contents removed in the stomach capacity measure- ments were recorded as rough percentages in four categories : ( 1 ) Detritus, the flocculent upper 2-3 cm layer of muddy bottoms (Sanders, 1960). Because of the variability in published values for the organic content of this material, two samples were ashed giving an average organic (burnable) content of 1.83%. Figuring 4.000 cal/g (carbohydrate) this amounts to 73 cal/cc. (2) Algae, mostly fine filamentous rhodophyta and cholorophyta : as determined by ashing, their organic content averaged 7.5% of the wet weight. At 4,000 cal/g (Odum, 1963) this is an average of 300 cal/cc. (3) Invertebrates, primarily pelecypods and marine worms, indicated by valves and setae. An average dry weight of 20% was assumed, and multiplied by 5,700 cal/g (Brody, 1945) as protein, gives 1,140 cal/cc. (4) Flesh, this last general type consisted of small squid and fish and was figured at the same caloric value as (3). These percentages, multiplied by the weight of the contents gave the caloric value for each stomach examined. 262 JOHN CARLSON ALDRICH The total weight of the contents was compared to the maximum capacity of the stomach, indicating the percentage of capacity used. Feeding experiments. Four sets of feeding experiments were made having durations of 9-13 days. Crabs previously starved for up to two weeks were fed each day, uneaten food from the previous day being removed and weighed. The maximum rates of consumption were usually realized during the first day of the test, consumption declining rather steadily during succeeding days. Because they occur in groups under natural conditions, the crabs were fed in groups and the rates obtained are averages. These averages are expressed as multiples of the maximum stomach capacity (stomachsfull) eaten per day. Assimilation efficiency. This was measured during the four sets of feeding experiments. Faeces were collected several times each day to prevent their pos- sible consumption by the crabs (never observed). The crabs defecated immediately upon their first feeding, therefore the faeces produced on a given day might be the result of an earlier feeding. Faeces output, like food consumption, declined from a maximum during the first days, and the two rates were in proportion, but with the faeces output lagging behind food consumption by 2-3 days. Because of this lag, overall assimilation efficiencies were computed for each experiment on the basis of the total dry weights of food consumed and faeces produced. TABLE I Allonietric constants for the equation: log Y - alog .Y + log b; and a = the slope of the line, b is the intercept when X ••= 1; R = correlation coefficient Stage R (a) Morphological allometry Males: Carapace /propodus lengths Immature 1.08 -0.56 0.9934 Known prepubescent All prepubescent Known mature 1.61 1.72 1.98 -1.36 -1.53 -1.89 0.9660 0.9915 0.9956 All mature 1.95 -1.85 0.9830 Females: Carapace /abdomen widths Immature 1.28 -0.86 0.9898 Prepubescent Mature 1.35 1.12 -0.95 -0.42 0.9854 0.9791 (b) Physiological allometry Stomach fullness/live weight All immature full 0.80 -1.37 0.7901 All immature empty 0.97 -2.00 0.9058 Mature female full 1.01 -1.71 0.8190 Mature female empty 1.31 -2.77 0.9234 Mature male full 0.51 -0.94 0.8108 Mature male empty 0.78 -1.87 0.9096 ALLOMETRY AND ENERGETICS OF LIBINIA 263 TABLE II F tests of regression lines showing; f values for a and b (slope and level), degrees of freedom, and level of significance (1%, 5%, X ••= not significant) Pair of lines tested a (slope) 6 (level) f d.f. Sig f d.f. Sig (Male propodus) Immature /known prepubescent 25.4 1/35 1 8.77 1/36 1 Known prepubescent/known mature 8.79 1/51 1 33.4 1/52 1 All prepubescent/all mature 24.5 1/183 1 237 1/184 1 (Female abdomen) Immature /prepubescent 0.646 1/24 X 0.891 1/25 X Prepubescent/mature 7.48 1/31 5 539 1/32 1 Immature-prepubescent/matun- 3.47 2/42 5 274 2/44 5 (Stomach weights) All maximum/all empty 0.491 1/11 X 2376 1/12 5 Immature as found/immature empty 0.543 1/22 X 6.00 1/23 5 Mature female as found/m.f. empty 5.03 1/36 5 72.9 1/37 5 Mature male as found/m.m. empty 1.00 1/18 X 84.3 1/19 5 (Hepatopancreas) Immature wet wt/all mature wet wt 0.039 1/56 X 12.7 1/57 5 Imm. lipid wt/all mature lipid wt 0.162 1/56 X 9.89 1/57 5 Routine Oz/imni. hepato. wet wt 1.06 1/44 X 153 1/45 7 Routine Oz/imm. lipid wt 0.267 1/47 X — — — Routine O2/all stomach empty wt 0.083 1/63 X _ _ RESULTS Morphological allomctry The plots of male propodus/carapace lengths, and female abdomen/carapace widths are given in Figure 1, where both individual points (small dots) and size class averages (large dots) are shown. For purposes of computation the males were divided into five categories: (1) immature; (2) known prepubescent (this excluded large doubtful cases); (3) known mature (small barnacled sizes and large crabs with extreme propodus lengths) ; (4) all possible prepubescent; (5) all possible mature. The allometric constants (Table la) were compared by F tests (Table II). The lines including the doubtful points were indis- tinguishable from those formed only by known points, therefore these all inclusive lines were used. Male immature, prepubescent, and mature stages were all significantly different in slope and level. The inflection at the molt of prepuberty occurs at the average carapace length of 31.5 mm, and is evident in the figure from the difference in slope between lines "a" (immature) and "b" (prepubescent). The mature crabs (line "c" ) range in length from 47.0 to 106.0 mm, their individual points making a line distinct from that of the prepubescent crabs, which are found up to 87.0 mm long. Female crabs were divided into three categories for computation : ( 1 ) im- mature; (2) prepubescent; (3) mature. The molt of prepuberty was not significant (Table II), and immature and prepubescent females were treated together. This combined stage (line "d") was found up to the carapace width of 43.0 mm. Ma- ture females were found over the width range of 37.0 to 62.0 mm (line "e"), and 264 JOHN CARLSON ALDRICH were significantly different from the immature line. Expressed as carapace lengths for comparison to the males, these figures are 52.5 mm for line "d," and 42.0 to 77.5 mm for line "e." The most distinct differences in the physiological constants were subsequently found between the molting (immature and prepubescent) and non-molting (ma- ture) crabs, therefore immature and prepubescent crabs of both sexes are treated together in the following comparisons, with mature females and mature males treated separately for a total of three stages, equivalent to two major growth stanzas within each sex. The first stanza is immature, actively growing and found in an abundance of algal food. The second is mature, reproducing, non-growing, and found on relatively barren mud bottoms. (This second stanza might be termed a "life" stanza because there is no growth.) The planktonic larvae of a crab would form a third major growth stanza. / 'hysiological allometry Oxygen consumption measurements gave an average routine rate of 4.7 ml Oo/hr for a crab of 123 g fresh weight, similar to other published values. Vernberg (1956) gives 3.2 ml Oo/hr for this weight of L. emarginata, and Zeuthen (1953) gives a general value for Crustacea equivalent to 3.7 ml O2/hr at this weight. The values found in the present study may be termed a lower range of the excited rate due to measuring after handling (Aldrich, 1974a). Odd, high values, prob- ably representing elevated rates due to biological rhythms (Ansell, 1973; Aldrich, 1974b) were excluded from the calculation. The allometric constants for ml Oa/g/hr were: a ==0.81, b - 0.89, R == 0.9950. The slope (a) is in virtual agreement with the a of 0.80 found by Zeuthen (1953). The average daily caloric cost of maintenance was computed at 4.83 cal/ml O2 (Brody, 1945) for a basal rate (the very lowest range of points in the overall regression) 80% of the routine rate. Stomach weights are given for individual crabs in Figure 2. The regression of maximum full weights has the constants: a == 0.80, b - -1.15, R == 0.9950; and the regression of all empty stomachs has the constants: a -- 0.80, b - -1.89, R : 0.9566. Thus the full and empty stomach weights bear a constant relationship to one another over a hundredfold range of body weight. Therefore the capacity of the stomach has a constant allometric relationship to the weight of the crab. The slope of this capacity regression is virtually the same as that of the routine metabolism (an F test showed no significant difference). Although this does not prove that metabolic rate is dependent upon the capacity of the stomach, it does indicate that these relationships are similarly constant with increasing weight, suggesting a functional correlation. The percentage of the maximum stomach capacity used is not the same in the three stages (Figure 2). Within each stage there is a significant difference between the levels of "as found" and empty stomach weights (Table II, b values). Immature-prepubescent, and mature females appear to use almost the maximum capacity, but large mature males seem to use a decreasing proportion as they increase in size. This is reflected in the significantly different slopes of the male "as found" and empty regressions (Table II), a difference not found in other stages and implying the constancy of empty weights, but the relative decrease of ALLOMETRY AND ENERGETICS OF LIBINIA 265 I "5 u ra r* C O .05 ,01 1*TFed ad libitum • As found o Empty 10 50 IOO Live Wt (g) 500 FIGURE 2. Stomach capacity and weight of contents found in collected crabs with maxi- mum capacity = Max. line — Min. line at any given crab weight. Size ranges are : immature crabs to the left of the heavy dashed line ; largest mature males to the right of the light dashed line; mature females and small mature males between the two dashed lines. Individual values both as found and when emptied are given. The regression of oxygen consumption is shown for comparison of slope, and is parallel to the regression of stomach capacity. fullness with increasing size. Note the different empty stomach slopes (a in Table Ib) for the three stages and how their overall combination makes a "regular" allometric line parallel to that of the routine oxygen consumption. The relationships of hepatopancreas total and lipid weights to total live weight are shown in Figure 3. Here immature-prepubescent crabs, as well as very large male crabs that have recently molted to maturity (Points "N") fall on the same regression line of hepatopancreas wet weight. The constants of this line are: a == 0.82, b- -0.84, R == 0.9280 ; therefore it is parallel to the regression of routine metabolism ( an F test showed no significant difference in slopes, see Table II). The combined regression of mature male and female 266 JOHN CARLSON ALDRICH hepatopancreas wet weight has the constants: a = 0.86, b = — 1.16, R = 0.8150 and is not significantly different in slope from the previous regression, but is different in level. Thus the weight of the hepatopancreas follows the same allometric formula (slope of maximum weight) as the stomach capacity and routine metabolism, and again larger (mature) crabs have lower levels (b) of utilization of a physiological capacity. (That this drop in proportion is not due simply to the larger sizes of mature crabs is shown by the points "N" for very large newly-molted crabs. ) The total weight of lipid contained in the hepatopancreas follows an analogous pattern. Here the constants are : Immature-prepubescent, a -- 0.98, b - —2.08, R;= 0.644; all mature, a ^1.16, b -- 2.98, R - 0.5758. F tests showed that these two lines are insignificantly different in slope, but again significantly different in level (Table II). Due to the great variability in individual values, MI F test showed no difference between the immature-prepubescent slope and that of the routine metabolism. The lower level of utilization of physiological capacity in mature crabs is again suggested by the caloric value of stomach contents (Figure 4). Here the caloric value decreases relative to the cost of maintenance as the size of the crab increases, O 30r 10 5 u c rt a. o •M rt o. .5 .3 • Wet Wt A Lipid Wt 10 50 100 Live Wt (g) 500 FIGURE 3. Hepatopancreas wet (total) weight (upper solid lines), and lipid weight (lower dashed lines). Data presented as size classes with standard deviations shown by the vertical bars. Crabs to the left of the vertical dashed line are immature-prepubescent, those to the right are mature (both males and females), the points "N" are recently molted males. ALLOMETRY AND ENERGETICS OF LIBINIA 267 2000 1000 500 .? 100 U 50 10 5 o 9 cT Averages as found Averages x 2 Level of basal maintenance .5 I 5 10 50 100 Live Wt fg) 500 FIGURE 4. Caloric value of stomach contents found in collected crabs, contrasted with the caloric cost of basal maintenance. Crabs to the left of the vertical dashed line are immature- prepubescent, those to the right are mature. Data presented as size class averages (dots) and means (small symbols) for immature-prepubescent, mature females (mean only), and mature males. The effect of the cycling rate is shown by doubling the caloric values (large symbols and dashed regression lines). especially in mature males. Again, immature crabs maintain a balance close to the maintenance cost, the regression of caloric values following the same slope as metabolism. Mature females also have a balance close to the maintenance cost. (The cost of monthly reproduction is less than that of rapid immature growth, Aldrich, 1972.) These trends suggested by the stomach contents are not as clear as the preceding ones due to the extreme variability in both the amount and caloric value of the contents (Table III), and the presentation is made without statistical significance (according to correlation coefficients). However, the trends agree with those previously found, the low caloric values for the mature males are especially marked, suggesting a real underutilization of stomach capacity. The stomach content findings are complicated by the cycling rate, or multiples of stomach capacity (stomachsfull) that can be processed in a 24 hour period. In four sets of feeding experiments the maximum consumption in 24 hours averaged 1.95 stomachsfull (Table, IV). This means that the estimated maxi- mum stomach capacity could be processed twice a day, or the cycling rate is such that any given contents could be digested within 12 hours. These figures were obtained with cleaned fish, mussels, and algae, all readily digestible. An earlier experiment (No. 5 in Table IV) where the crabs fed on starfish gave much the same result with this relatively ttndigestible food. 268 JOHN CARLSON ALDRICH TABLE III Percentage fullness (% of max. capacity) and stomach contents of crabs examined within 2 hours of collection Size class (grams) No. Ave. % fullness % Fullness ace. to food types Detritus Alga Inverts. Flesh Immature males and females 0.00-0.33 3 100 100 0.34-0.67 2 50 50 0.68-1.00 2 100 100 1.1-3.3 8 56 50 6 3.4-6.7 8 75 65 10 6.8-10 4 70 70 11-33 1 100 100 Mature females 34-122 22 38 11 Prepubescent males 14-74 46 23 19 Mature males 33-67 7 44 15 12 7 10 68-100 7 51 20 23 8 101-200 7 19 7 7 5 201-500 5 14 4 10 The assimilation efficiency averaged 96% (Table IV), so high that it was considered unity for purposes of calculations. There was no real difference in efficiencies whether the crabs were digesting animal or vegetable food. Such TABLE IV Feeding experiments showing average weight and number of crabs, food given (cleaned fish, mussel flesh, algae, and whole starfish), multiples of stomach capacity (stomachsfull) processed within 24 hours, and overall assimilation efficiencies Exp. No. Ave. Wt. ± S.D. No. Crabs Food Multiple Assim. Efficiency 1 5.8±3.41 5 Fish, Stenotomus 1.53 96.5% 2 37.2±6.59 10 Mussel, Mytilus 1.90 96.7% 3 45.0±13.5 25 Fish, Stenotomus 2.05 99.3% 4 58.7±10.1 8 Alga, Codiuni 2.31 95.0% (5) 144 ±169 16 Starfish, Aster ias (1.65) ALLOMETRY AND ENERGETICS OF LIBINI. I 269 high efficiencies are not uncommon in arthropods consuming animal food (Law- ton, 1970, 1971). DISCUSSION As indicated by the scattergram (Figure 1), there is a considerable range of mature sizes in both males and females. For males, this range requires three molts at an average increase factor of 1.26 (Dyar's Law; Teissier, 1960). Two of these molts are optional in the sense that some prepubescent crabs exist in ever increasing sizes but still have not reached maturity as defined by the inability to undergo further molts. Because some of these larger prepubescent crabs appear to have mature gonads, "prepuberty" for them may include animals sexually mature before reaching their terminal anecdysis (Hartnoll, 1963). Some very large pre- pubescent males were kept in the laboratory tanks and they all molted in the autumn, whereas no mature crabs of any size ever molted. The consequence of the optical molts is a 16-fold weight range for mature, non-molting male crabs. This weight range is similar to that found for many other Maiid crabs (Hartnoll, 1963; Vernet-Cornubert, 1958), but was not anticipated since it is "common knowledge" that only the largest males are the mature sizes. Some of the smallest mature males were kept together with mature females and they actively pursued and mated with them, giving a behavioral confirmation of the status suggested by their allometry. The lesser range of female mature sizes requires two molts, of which one would be optional. Smaller mature females were often collected bearing eggs, as well as the commoner ovigerous middle range, but none of the very largest mature females were collected with eggs. The reproductive status of these largest females crabs is therefore unclear. Thus the morphological allometry made possible the recognition of this great range of mature sizes, leading apparently to quite different energetic requirements within the mature stage. Small crabs appear to live for several years when mature, as evidenced by their several layers of barnacles. Such crabs were usually collected in rocky habitats where food may have been more abundant than on the mud flats where the largest mature crabs were found. Large crabs were never found with more than a few scattered barnacles, an indication that they may not have lived as long as the small males. Two reproductive strategies may be available to Maiid crabs then; small size (and small numbers of larvae in the case of females) coupled with several reproductive years, and large size coupled with a shorter reproductive period. Although females will accept the smaller males as mates, the larger males have a distinct competitive advantage, when kept together in tanks during these experiments, they merely picked up and cast aside their smaller rivals. Three major points recur in the physiological variables. First, the maximum capacity of the stomach, the maximum hepatopancreas weight, and the metabolic rate all follow the same slope (a) thus exhibiting similar physiological relation- ships at the maximum values. Secondly, regularly molting crabs exhibit high proportional usage of digestive capacity and hepatopancreas storage capacity as shown by their relatively high values of b (level). Thirdly, the mature stages (non-molting) are functioning at less than the maximum physiological capacity 270 JOHN CARLSON ALDRICH for these factors. The weight and lipid content of the hepatopancreas, as well as the portion of stomach capacity used are lower than in the molting stage. This lower level of usage is reflected in their b values. The calculated daily levels of caloric intake based on stomach contents could be doubled by assuming that these levels were processed twice daily, and that there was no correction for undigested food. This is probably an overestimate for the larger crabs since there would be no advantage in consuming less than the maximum possible in the food-poor habitat of the muddy bottoms. In this case the stomach contents may represent the actual daily consumption. However, doubling the caloric value of the stomach contents may be a good estimate for the growing stage. The great rate of growth (up to monthly doubling in weight) may require as much food as they can process and their algal habitat would supply it without limit. Whatever the rate of processing stomach contents may have to do with their average "as found" level, the mature males appear to use less of their capacity as their size increases. Even doubling the caloric level of the contents leaves the largest males with an intake less than the minimum required for basal me- tabolism, as is shown in Figure 4 by the dashed lines through the average caloric values X 2. Further, mature females found in the same habitat are relatively better fed than the mature males. In the summer, females must bear the cost of monthly reproduction (Hinsch, 1968, 1972) and require more food than males, yet some specimens had up to three layers of barnacles, apparently indicating- several years in the mature stage. Although mature females may be relatively better feeders than males, it could be that mature males do not have an "inclination" to eat at a sufficient rate for long survival. Perhaps such large crabs, those that have passed through the optional molts, represent a specialized reproductive stage like spawning male salmon that never feed. This idea is strengthened by the large crabs not being found with layers of barnacles, suggesting that they do not live long. There may be a general reduction in feeding levels in mature crabs, at least during certain stages of reproduction. Pearson (1908) mentions that berried Cancer payurus "feed very little," and Carlisle (1957) described the "heaps" formed by molting and mating M\a squinado, one persisting for two months. Presumably the crabs did not feed while in this stationary heap. (In this context, the author remembers as a little boy finding a heap of Libinia, a discovery rapidly abandoned. ) The very small non-molting mature sizes were not examined separately and it would be instructive to see if there is a marked drop in their b values cor- responding to that found in large mature sizes. This would show if there is a size effect within the mature stage or whether b values simply drop with the cessation of growth. In all, the differences in b values between immature-prepubescent crabs (the first growth stanza) and mature ones (the second stanza) are significant and consistent throughout the variables compared. Because Maiid crabs do not molt when mature, this stanza is quite distinct and the size of the hepatopancreas may be a better indicator of relative energy demand than in brachyrhynchous crabs where molting continues through most of the reproductive life. One such brachy- rhynchous crab, Carcinus macnas, exhibits seasonal fluctuations in metabolites stored in the hepatopancreas, but has no obvious storage of material in prepara- ALLOMETRY AND ENERGETICS OF LI BIN I A 271 tiou for ecdysis, there being a marked variation between individuals in the later molt stages (Heath and Barnes, 1970). Seasonal fluctuations were not measured in Libinia and their effects may be important, but the molting cycle will not complicate the interpretation as it would in another brachyrhynchous crab, Cancer pagurus (Renuad, 1949). Whether or not the hepatopancreas does indicate relative energy demands, the level of stomach contents must indicate the rate of energy intake and this level is also shown to decrease in mature Libinia. The habitats defining the immature-prepubescent, and mature growth stanzas are different, and the rate of energy supply in the habitat of the mature crabs must be quite limited. As the points "N" in Figure 4 imply, the condition of newly- molted but very large mature crabs appears to be similar to a normal immature growing crab. However, the reserves are not maintained during subsequent life in the mature stanza. The capacity of the crab as a system probably remains unaltered, as is suggested by the similar a's for maximum stomach capacity and hepatopancreas weight. Thus the mature stage of this crab appears to be an underutilized end point of a system designed for the growth of the immature stage. Many thanks are due to Prof. Arthur G. Humes, Director of the Boston University Marine Program, under whose supervision this study was performed, and to Drs. William Stewart and Melbourne Carriker for their assistance. Dr. Gertrude Hinsch kindly allowed the measurement of the crabs in her labora- tory, and Dr. Richard Newell and Mrs. Islay Marsden of Queen Mary College gave advice and assistance in the preparation of the manuscript. Most of all, thanks are due to my wife Judith who acted as crew on our trawling expeditions and as technician in the laboratory. SUMMARY 1. Allometric plots were used to separate Libinia emarginata into three male growth stages ; immature, prepubescent, and mature ; and two female stages ; immature-prepubescent, and mature. 2. The existence of two optional prepubescent molts in males allows a 2.75-fold variation in mature (non-molting) length ; and one optional molt in females allows a 1.75-fold variation in mature length. 3. Because of the very great size range in mature crabs (especially males), there are apparently two reproductive strategies ; small mature size and several reproductive years, and large size with one reproductive year. 4. The regression of routine oxygen consumption, measured over the entire size range of this crab, exhibits virtually the same slope (0.81) as found by Zeuthen ( 1953) for Crustacea in general. 5. The maximum capacity of the stomach, and the maximum hepatopancreas wet weight follow the same allometric relation (slope or a) as the oxygen consumption, in agreement with the idea that metabolic rate is reflected in the size of the digestive apparatus (Hemmingsen. 1950). 6. The actual weight of stomach contents, and the wet and lipid weights of the hepatopancreas (level or b] are not fixed and reflect relative energetic demands or availability. 272 JOHN CARLSON ALDRICH 7. During the stages undergoing regular molts (immature and prepubescent) these variables exhibit high level implying the maximal use of physiological capacity. 8. At the stage of maturity, these levels decrease relative to the growing stages, implying the underutilization of this capacity. LITERATURE CITED ALDRICH, J. C., 1972. On the biology and energetics of Libinia cmarijinata, an omnivorous decapod. Master's thesis, Boston University, Graduate School of Arts and Sciences, 145 pp. ALDRICH, J. C., 1974a. On the oxygen consumption of the crabs Cancer pagunis and Maia squinado. Conip. Biochcin. Physiol., in press. ALDRICH, J. C., 1974b. Individual variability in oxygen consumption rates of fed and starved Cancer pagurus and Maia squinado. Comp. Biochcin. Physiol., in press. ANSELL, A. D., 1973. Changes in oxygen consumption, heart rate and ventilation accompany- ing starvation in the decapod crustacean Cancer pagurus. Netli. J. Sea Res., 7 : 455-475. ARMITAGE, K. B., A. L. BI/IKEMA JR. AND N. J. WILLEMS, 1972. Organic consituents in the annual cycle of the crayfish Orconectes nais (Faxon). Comp. Biochem. Physiol., A41 : 825-842. BAYNE, B. L., 1973. Physiological changes in Mytilns editlis L. induced by temperature and nutritive stress. /. Mar. Biol. Ass. U.K., 53 : 39-58. vox BERTALANFFY, L.. 1957. Quantitative laws in metabolism and growth. Quart. Rev. Biol,, 32: 217-231. BRODY, S., 1945. Biocncrgctics and Growth. Reinhold Pub. Corp., New York, 1023 pp. CARLISLE, D. B., 1957. On the hormonal inhibition of moulting in decapod Crustacea. II. The terminal anecdysis in crabs. /. Mar. Biol. Ass. U.K., 36: 291-307. DRACH, P., 1939. Mue et cycle d'intermue chez les crustaces decapodes. Ann. Inst. Oceanog. (Paris), 19: 103-391. FOLCH, J., M. LEES AND G. H. SLOAXE-STANLEY, 1957. A simple method for the isolation and purification of total lipides from animal tissues. /. Biol. Chcni. 226 : 497-509. GOULD, S. J., 1966. Allometry and size in ontogeny and phylogeny. Biol. Rev., 41 : 587-640. HARTNOLL, R. G., 1963. The biology of Manx spider crabs. Proc. Zool. Soc. London, 141 : 423-496. HEATH, J. R., AND H. BARNES, 1970. Some changes in biochemical composition with sea- son and moulting cycle of the common shore crab Carcinus maenas (L.). /. Exp. Mar. Biol. Ecol.. 5 : 199-233. HEMMINGSEN, A. M., 1950. The relation of standard (basal) metabolism to total fresh weight of living organisms. Rep. Steno Memorial Hospital ( Copenhagen), 4: 7-58. HEMMINGSEN, A. M., 1960. Energy metabolism as related to body size and respiratory sur- faces and its evolution. Rep. Steno Memorial Hospital (Copenhagen), 9: 7-110. HINSCH, G. W., 1968. Reproductive behavior in the spider crab, Libinia emarginata L. Biol Bull., 135 : 273-278. HINSCH, G. W., 1972. Some factors controlling reproduction in the spider crab, Libinia emarginata. Biol. Bull., 143 : 358-366. HUXLEY, J. S., 1924. Constant differential growth-ratios and their significance. Nature, 114: 895-896. KIDWELL, J. F., AND H. B. CHASE, 1967. Fitting the allometric equation — a comparison of ten methods by computer simulation. Growth, 31 : 165-179. LAWTON, J. H., 1970. Feeding and food energy assimilation in larvae of the damselfly Pyrrhosoma nymphnla (Sulz.) (Odonata: Zygoptera). /. Anim. Ecol., 39: 669-689. LAWTON", J. H., 1971. Ecological energetics studies on larvae of the damselfly Pyrrhosoma nymphula (Sulzer) (Odonata: Zygoptera). /. Anim. Ecol., 40: 385-419. NEEDHAM, J., 1942. Biochemistry and Morphogenesis. Cambridge University Press, London, 785 pp. ODUM, E. P., 1963. Ecology. Holt, Rinehart and Winston, New York, 152 pp. ALLOMETRY AND ENERGETICS OF LIBINIA 273 PARKER, R. R., AND P. A. LARKIN, 1959. A concept of growth in fishes. /. Fish. Res. Board Can., 16 : 721-745. RATHBUN, M. J., 1925. The spider crabs of America. U. S. Nat. Mus. Bull, 129: 1-613. RAY, D. L., 1959. Marine Boring and Fouling Organisms. University of Washington Press, Seattle. RENAUD, L., 1949. Le cycle des reserves organiques chez les crustaces decapodes. Ann. Inst. Oceanog. (Paris), 24: 259-357. SANDERS, H. L., 1960. Benthic studies in Buzzards Bay. III. The structure of the soft- bottom community. Litnnol. Oceanog., 5 : 138-153. STROGANOV, N. S., 1964. Methods of study of respiration in fish. Pages 27-79 in E. N. Pavlovskii, Ed., Techniques for the Investigation of Fish Physiology. Israel Program for Scientific Translations, Jerusalem. TEISSIER, G., 1931. Recherches morphologiques et physiologiques sur le croissance des insectes. Trav. Sta. Biol. Rose off, 9 : 27-238. TEISSIER, G., 1935. Croissance des variants sexuelles chez Maia sqitinado. Trav. Sta. Biol. Roscoff, 13 : 93-130. TEISSIER, G., 1960. Relative growth. Pages 537-560 in T. H. Waterman, Ed., The Physiology of Crustacea, Vol. I. Academic Press, New York and London. VERNET-CORNUBERT, G., 1958. Biologic generale de Pisa tetraodon (Pennant). Bull. last. Oceanog. (Monaco), 1113: 1-52. VONK, H. J., 1960. Digestion and Metabolism. Pages 291-316 in T. H. Waterman, Ed., The Physiology of Crustacea, Vol. I. Academic Press, New York and London. ZEUTHEN, E., 1953. Oxygen uptake as related to body size in organisms. Quart. Rev. Biol., 28: 1-12. Reference: Biol. Bull., 147: 274-293. (October, 1974) ROLES OF OXYGEN AND CARBON DIOXIDE IN THE CONTROL OF SPIRACULAR FUNCTION IN CECROPIA PUPAE BARBARA N. BURKETT AND HOWARD A. SCHNEIDERMAN Department of Biology, University of Miami. Coral Gables, Florida 33124 and Center for Pathobiology, University of California. Irvine, California 92664 In most insects spiracular valve movements are controlled by oxygen and carbon dioxide (Beckel and Schneiderman, 1957; Schneiderman, 1960; Levy and Schneiderman, 1966a, 1966b). When the intratracheal concentration of carbon dioxide is high or that of oxygen is low, the spiracular valves open ; when the intratracheal concentration of oxygen is high, they close. However, the sites at which these gases act have never been demonstrated. The central nervous system is surely involved (Miller, 1966) as are the spiracles themselves. This report identifies the targets of oxygen and carbon dioxide in diapausing pupae of the Cecropia moth, discusses the interaction of oxygen and carbon dioxide in con- trolling spiracular behavior, and analyzes several aspects of the nervous control of the spiracular closer muscle. MATERIALS AND METHODS Diapausing pupae of the silkworm Hyalophora cecropia were used. They were stored in their cocoons at 2° C and 90% R.H. until they were prepared for experiments as follows : their abdomens were immobilized with strips of paraffin (Schneiderman and Williams, 1955) to prevent internal movements from damaging the ganglion preparation, and external telescoping movements from obscuring the spiracles of movable segments. The brain was removed to insure permanent diapause (Williams. 1946). Pupae then were allowed to recover for at least two weeks at 25° C and 50^ R. H. Two days before an experiment was begun the pupa was provided with an "inlet" and an "outlet" cannula (Fig. 1) which enabled us to perfuse the entire tracheal system, including the tracheae serving the spiracular muscles ("spiracular' flow). Cannulae were 8 cm lengths of polyethylene tubing (0.97 o.d., 0.58 mm i.d. (PE-50) : or 0.61 mm o.d.. 0.28 mm i.d' (PE-10)) (Tntramedic tubing, Clay-Adams Company, New York). Each cannula was inserted past the spiracular valve (Schneiderman and Schechter, 1966; Brockway and Schneiderman, 1967). The cannulae were checked for leaks as described by Brockway and Schneiderman (1967). The experimental gas was introduced into the tracheal system via the second or third right abdominal spiracle, and excess gas escaped via either the fifth or sixth left abdominal spiracle. The fourth abdominal ganglion was selected for neural and tracheal surgery since it is located conveniently in a non-collapsible segment, the spiracles of which are not obscured by overlying pupal wingpads. The spiracle on the right side of that segment was exposed and observed in all experiments. In addition 274 SPIRACLE CONTROL IN INSECTS 275 the third or fifth right abdominal spiracle or both were exposed and observed as control spiracles. The condition of the exposed spiracular valves was checked daily by probing the pupa gently on the cuticle in the area of the exposed spiracles ; this maneuver caused properly functioning valves to open. In air, valves of a normal pupa pulsate most of the time, that is, they move but do not open. At the beginning of each experiment moist air was perfused through the pupa's tracheal system ; if the valves failed to pulsate, the pupa was discarded. INLET CANNULA , TRACHEAL MANIFOLD •^^^^W^^^l^^^^^W^^'^''''^ INTEGUMENT - 3RAS FILTER APPARATUS 4RAS _ OUTLET CANNULA SPIRACULAR VALVE 5RAS- FIGURE 1. Arrangement of cannulae for spiracular H<>\v. The upper part of the diagram shows a pupa with the inlet and the outlet cannula in position ; the lower part of the figure shows a portion of the tracheal system prefused via this arrangement. The inlet cannula (2RAS, above 3RAS) was connected to the perfusion pump, and the outlet cannula (5LAS, opposite 5RAS) was open to the atmosphere. Gas entering the inlet cannula flowed throughout the tracheal system; excess gas escaped via the outlet cannula; 3RAS, 4RAS, 5RAS — third, fourth, fifth right abdominal spiracle; CNG — ganglion; TrT — transverse trachea; LnT- longitudinal trachea. Animals were mounted ventral side up in modeling clay for observations. Ex- periments were performed at 22-25° C and 40-50% R.H. During the recovery intervals between gas perfusions the pupa w^as kept in a water-saturated environ- ment. .Inatoniy of the spiracle The spiracular apparatus was described in detail by Beckel (1958) (Fig. 2). It consists of a membranous spiracular valve, a chitinous frame which acts as a 276 B. N. BURKETT AND H. A. SCHNEIDERMAN lever, and the spiracular closer muscle which is opposed by an elastic ligament and is innervated by a single spiracular nerve (Fig. 3). Recording spiracular valve movements The method described by Schneiderman (1956, 1960), in which the leading edge of the spiracular valve is followed in an ocular micrometer, was used for recording valve movements. A Grass FT 03 force displacement transducer and polygraph (Grass Instrument Company, Quincy, Massachusetts) were substituted for the lever-pen system. All except the most rapid and frequent valve move- ments could be followed and recorded almost as soon as they occurred. It was possible to record the movements of only one valve at a time. However, all spiracular valves engage in the same mode of behavior even though they are not synchronous (Van der Kloot, 1963; Brockway and Schneiderman, 1967), i.e., CUT STP FA STR MAN FIGURE 2. Frontal section through the spiracular region (redrawn from Beckel, 1958). The closing bars push the tracheal membrane against the closing bow, thus closing the spirac- ular opening as in this drawing ; ATR — atrium ; CUT — cuticle ; FA — filter apparatus ; HYP — hypodermis ; MAN — manifold; PTR — peritreme; STP — stigmal plate; STR — spiracular trachea ; X, Y — dorsal lateral and dorsal median closing bar, respectively. they constrict, flutter, or open fully. Hence the behavior of the observed valve is qualitively like that of the others. A dissecting microscope permitted viewing of several valves in turn at 30 X without disturbing the preparation. Consecutive recordings lasting 10 seconds to 3 minutes each were made when observations of two or more valves were required. Gas perfitsions Gas mixtures were made up in small cyclinders or directly in a 50 ml syringe from commercially available cylinders of oxygen, carbon dioxide, and nitrogen The syringe was equipped with a three-way Luer-Lok MS 02 valve which per- mitted direct filling of the syringe from cylinder lines at approximately atmospheric pressure, and minimized the possibility of contaminating the contents of the syringe with external air. The syringe could be refilled in less than one minute when long perfusions were necessary. Gas analyses accurate to ±1% were performed, using a Model E2 paramagnetic oxygen analyzer and a Model 215 infrared carbon SPIRACLE CONTROL IN INSECTS 277 dioxide analyzer (Beckman Instruments, Inc., Fullerton, California). As recom- mended by the manufacturer, neoprene tubing was used for cylinder lines and Viton-A vacuum tubing (Bearings, Inc., Cleveland, Ohio) connections were used with the analyzers. These materials are impermeable to the gases used in these experiments. A hypodermic needle was attachd to the syringe valve ; a 50 cm length of polyethylene tubing the same diameter as the inlet cannula of the pupa connected the needle with the cannula. The plunger of the syringe was driven at a contant predetermined rate by a motor. Two syringes were used when two gas mixtures were perfused simultaneously. M 15 ALN OPENER LIGAMENT VALVE CLOSING LEVER CLOSER MUSCLE A8 MLN CNG FIGURE 3. Innervation of an abdominal spiracular closer muscle (redrawn from Beckel, 1958). CNG — ganglion; ALN — anterior lateral nerve; MLN — mid-lateral nerve; A8 — brand-, of ALN innervating the muscle; MIS — branch of MLN innervating the muscle. Animals were perfused at a rate of 360 /xl/min (occasionally 1430 /A/m\u ) for at least 20 minutes or until a definite unchanging pattern of spiracular response had been established. A 20-minute recovery period was allowed between exposures to different gases although normal behavior of spiracular valves, assuming that behavior in ambient air is "normal," was resumed usually within 10 minutes after the gas flow had been terminated. As a routine check on the initial condition of the pupa, and to determine the pattern of spiracular valve activity, the following procedure was used : The tracheal system of each of 59 pupae was perfused before the experiment began with at least 5 of the following: 2, 3, 5, 10, 21 (air), or 50% O2 (balance N2 and CO2). Since the tracheal system normally contains at least 3% CO2 (Levy and Schneider- man, 1966b), 3 or 4% CO2 was added. Similar results were obtained whether the concentration of oxygen was selected at random or progressively increased 278 B. N. BURKETT AND H. A. SCHNEIDERMAN from 2 or 3% to 21 or 5Q%. Records of valve responses by each pupa to different gas mixtures were used as controls for subsequent experimental records of that same pupa. Perfusing the tracheae of an abdominal ganglion To provide the fourth abdominal ganglion with its own gas supply, separate from that of the rest of the pupa, the following procedures were used : An 8 mm square of integument overlying portions of the third, fourth, and fifth abdominal segments was removed. A few crystals of a 1:1:1 mixture TO SYRINGE GAS DELIVERY TUBING GAS ESCAPE TUBING BELL CNG — KEL-F OIL GANGLION SUPPORT TRACHEA FIGURE 4. Assembly for supporting the fourth abdominal ganglion and perfusing it with the ganglionic flow. The bowl of the ladle was filled with Kel-F oil to retard desiccation and to prevent the ganglion from sticking. The gas escape tubing permitted excess gas to escape and prevented the Kel-F oil from bubbling; CNG — ganglion. of phenylthiourea, streptomycin sulfate, and penicillin were placed in the wound. Any lost hemolymph was replaced with insect Ringer (Ephrussi and Beadle, 1936). The same gas mixtures that were used before surgery were perfused through the pupa's tracheal system immediately after surgery. Thus the spirac- ular valve responses of a pupa to various oxygen concentrations just before and just after injury could be compared. Following this second set of perfusions, all fat body was removed from the vicinity of the third and fourth abdominal ganglia. This procedure did not inter- rupt the tracheal supply to the spiracular closer muscle of the fourth spiracle or that of its adjacent control spiracles. The intact fourth abdominal ganglion was SPIRACLE CONTROL IN INSECTS 279 supported from its dorsal side in the bowl of a "ladle" which was inserted beneath the ganglion with a micromanipulator (Fig. 4). A second micromanipnlator posi- tioned the gas delivery bell directly over the ganglion. The bell completely en- closed the ganglion resting on its support but did not compress the nerves or tracheae supplying the ganglion. The entire assembly dipped into a drop of Kel-F high polymer oil (Minnesota Mining and Manufacturing Company, St. Paul, Minnesota) so that no external gases contaminated the experimental gas flow. Thus the fourth abdominal ganglion could receive its own separate, localized tr. ALN O.I mm FIGURE 5. Gross anatomy of the immediate tracheal supply to the fourth abdominal ganglion; ALN— anterior lateral nerve; MLN— mid-lateral nerve; PLN— posterior lateral nerve; CNG — ganglion; tri-s, tli_3 — tracheae. gas flow via this arrangement ("ganglionic" flow) while a second gas could be perfused simultaneously through the animal's tracheal system (spiracular flow). The third abdominal ganglion served as a control. At this point, the intact fourth abdominal ganglion of each of six pupae was perfused with its separate ganglionic flow, while the spiracles were perfused si- multaneously with the spiracular flow, which consisted of the same or a different gas mixture. The perfusions were begun within a minute after the ganglion had been exposed and supported. The effects on spiracular valve behavior of surgery itself and of gas flow over the surface of the intact ganglion could be compared. 280 B. N. BURKETT AND H. A. SCHNEIDERMAN After these perfusions the delivery bell was raised, and tracheae tri, tr2, tr3, tl], t!2, and t!3 (Fig. 5) were cut at their junction with the ganglion; thus the ganglion was "detracheated." This permitted gases flowing over the ganglion to enter the open ends of tracheae attached to it, and evenutally to reach the cells in the ganglion. In addition those tracheal branches passing along and attached to the ventral nerve cord near the ganglion were punctured with a fine glass needle to avoid damaging the nerve cord, and then the tracheal ends were teased apart gently. To prevent fluid from creeping into the cut ends of the tracheae attached to the ganglion, thus blocking the ganglion from the experimental gas flow, the tracheae were not severed until just before the bell was lowered over the ganglion and the gas perfusions were begun. During intervals between experi- ments, air was perfused over the ganglion to prevent hypoxia. To assess the long-term effects of detracheation, a plastic window could be sealed over the wound in the integument with melted paraffin. Pupae then were stored at room temperature as described above, and examined periodically. Each experiment reported in this paper was performed on at least two and as many as eight pupae. Departures from these general procedures as well as special methods will be explained in the appropriate sections below. RESULTS Effects of routine experimental procedures on spiracular valve behavior Four major procedures — cannulation, use of spiracular windows, tracheal per- fusion, and surgery — were followed routinely to prepare pupae for experiments. To evaluate the effects of each procedure, the normal behavior of the spiracular valves was used as the standard. Since valves of a pupa respiring in ambient air pulsate most of the time and occasionally flutter (Schneiderman, 1960), this pattern was chosen as "normal." Cannulation. Four different diameters of polyethylene tubing were tested as cannulae. Spiracular valve responses to air perfusions through cannulae of dif- ferent bores were the same. PE-10 tubing generally was used because it was the most convenient. Five pupae were cannulated, each with a different combination of spiracles as cannulation sites. Spiracular valves of these pupae responded in the same way to air perfusions. Thus cannulation itself, the cannula bore size, and the sites of cannulation did not affect the behavior of spiracular valves. Spiracular window. A plastic window was sealed over exposed valves to prevent the pupa from desiccating. The behavior of the spiracular valves of seven intubated pupae, first with and then without the window, was the same in response to air perfusions. Thus spiracular valve behavior was not visibly affected by the presence of a plastic window over the exposed spiracles. Tracheal perfusion. The object of perfusing the tracheal system was to control the intratracheal gas composition. To meet this objective we considered the structure of the tracheal system and the rate at which gases were perfused through it. In addition we determined the time needed for equilibration of intratracheal gas with the perfused gas, the time needed for recovery from a perfusion, and whether valve responses to a given gas mixture are reproducible. SPIRACLE CONTROL IN INSECTS 281 Gross dissection of a pupa reveals that tracheae form an interconnecting net- work. Large tracheae run longitudinally between spiracular manifolds from which numerous smaller transverse tracheae branch and rebranch repeatedly. There are no valves to cause unidirectional flow within pupal tracheae (Brockway and Schneiderman, 1967). Furthermore, when air is perfused through the tracheal system at a high rate, the pupa distends. The flow rates used in these experiments were a compromise between rates high enough to insure constant average intratracheal levels of oxygen and carbon dioxide and rates low enough to prevent the pupa from distending. It was deter- mined that a flow rate of 360 /ul/min does not itself visibly affect valve behavior yet still meets the oxygen requirements of a pupa, whereas flow rates greater than 1430 /xl/min cause the pupa to distend. After 2 to 5 minutes of perfusion, valve behavior stabilized and usually re- mained the same until perfusion ended, usually 20 minutes but occasionally up to 4 hours later. In only eight of over 600 perfusions did valve behavior after 20 minutes differ from that after 2 to 5 minutes of perfusion. Moreover, results were highly reproducible from day to day in any individual and among individuals. The resumption of normal valve behavior after perfusion had ended was interpreted as recovery and usually occurred within 5 to 10 minutes after perfusion ceases. The duration of perfusion had no visible effect on recovery time. Similar valve responses to a given oxygen concentration occurred not only in the different spiracles of the same pupa but also among different pupae. Devia- tions were in the degree of response, not in different modes of behavior. For example, valves of 50 pupae fluttered at amplitudes ranging from low to high in response to perfusions of \Sc/( OL. : there were no prolonged valve openings or valve constrictions in any pupa. Surgery. Injury increases the metabolism of brainless diapausing pupae sev- eralfold (Schneiderman and Williams, 1955). This increased metabolism is conspicuous about 24 hours after injury, becomes maximal about 4 days after injury, and persists from one to several weeks. It was important to know what effect surgery per se had on spiracular valve behavior, aside from any specific effect resulting from tracheal or neural surgery. Each of 14 pupae that had cyclical respiratory activity before integumentary surgery was examined immediately after surgery for burst cycles. Only one pupa continued to exhibit cyclical respiratory activity after injury; its cycles were about 50 minutes long after surgery as compared to about 80 minutes before surgery. Valves of the 13 other pupae fluttered at low amplitude immediately after injury, although several minutes of constriction with pulsations were common. The spiracular valve responses of injured pupae to different perfused gas mixtures were also examined, and compared to the responses of the same pupae to the same gas mixtures before injury. The valve responses of an individual pupa to a given perfused gas mixture were generally the same after injury as before. In five pupae, post-injury valve responses were recorded almost 72 hours after surgery, at a time when injury metabolism was near maximum. Changes in valve behavior after injury were limited to an increase in flutter amplitude, or to motionless, fully-open valves whereas before injury the fully-open valves occa- sionally closed partially. 282 B. N. BURKETT AND H. A. SCHNEIDERMAN Spiracular valve responses of intact intubated pupae to gas perfusions It was necessary to know how the spiracular valves of intubated pupae that had not undegone surgery responded to various mixtures of carbon dioxide and oxygen. Thus pupae were perfused with gas mixtures in which the concentration of oxygen varied between 2 and 50%, and that of carbon dioxide ranged from 0 to 7%. Records of valve behavior under these conditions were obtained for I .U., ..... I I. .^11 I. I Is FIGURE 6. Spiracular valve responses to simultaneous spiracular and ganglionic per- fusions following detracheation of the fourth abdominal ganglion. The large arrow at the beginning of each record indicates the amount that the pen was deflected when valves opened fully. The short arrows below the tracings indicate the beginning (upward-pointing arrows) and end (down-pointing arrows) of recording. The short breaks in recording represent the time required to bring the next valve in focus (chart speed — 0.5 mm/sec; each division on the chart represents 50 seconds) ; 3RAS, 4RAS, 5RAS — third, fourth, fifth right abdominal spiracle, respectively; (A.) Spiracular flow — 50% O» + 3% CO2; ganglionic flow— 3% Oa + 3% CO*; (B.) Spiracular flow— 3% O2 + 3% CO2; ganglionic flow— 3% O2 + 3% CO.; (C.) Spiracular flow— 21% Os + 3% CO.; ganglionic flow— 3% O, + 3% CO2; (D.) Spiracular flow— 3% O2 + 3% CO,; ganglionic flow— 21% O» + 3% CO.. each pupa, and served as a type of control record for subsequent experimental records for that same pupa. The following results were obtained : In 61 of 101 perfusions of 2 to 5% O2 and 0 to 4% CO2, valves opened; in the 50 other perfusions, valves fluttered at high amplitude. Since the concentra- tion of carbon dioxide in all perfusions was too low to cause opening (Levy and Schneiderman, 1966a), the wide opening of the valves was a response to low oxygen. When the concentration of carbon dioxide was 0 to 4%, while that of oxygen was 5 to 15%, valves fluttered in 143 of 151 perfusions. SPIRACLE CONTROL IN INSECTS 283 When the concentration of carbon dioxide was 3% and oxygen was 50%, valves were constricted and motionless. When the concentration of oxygen was too high to cause prolonged valve opening, i.e., above 5%, valves could be induced to remain open by perfusion mixtures containing 7% or more carbon dioxide. TABLE I Effect of carbon dioxide concentrations on the rali'e behavior of the fourth right abdominal spiracle Gas mixture Pupa no. Spiracular Ganglionic Valve response* 4RAS** %02 %CO, ',<> %C02 10-1 3 3 5 5 3 2 5 7 1 10-2 5 7 1 8 12 1 8 15 1 8 20 1 8 25 1 8 30 1 8 32 4 8 35 6 8 40 6 10-1 21 3 21 3 1 21 3 5 7 1 5 7 5 7 5 5 7 21 3 5 50 3 50 3 0 50 3 5 7 0 5 7 5 7 6 5 7 50 3 6 10-2 8 3 8 3 1 8 3 8 12 1 8 12 8 12 5 8 12 8 3 5 * 1 == closed, motionless; 2 = closed, pulsating; 3 == low amplitude flutter; 4 == medium amplitude flutter; 5 == fully open, moving; 6 = fully open, motionless. ** 4RAS = fourth right abdominal spiracle. In summary, intratracheal carbon dioxide concentrations of 7% or more were required to cause valves to open fully and remain open when the intratracheal oxygen concentration was 5 to 21%. When the carbon dioxide concentrations were too low to cause valve opening, the valves behaved as follows : they opened fully or fluttered at high amplitude in 2 to 5% oxygen; fluttered at low amplitude in 5 to 15% oxygen; constricted in 21% oxygen or higher. These results agree well with those obtained earlier by other methods (Schneiderman, 1960; Levy and Schneiderman, 1966a, 1966b). 284 B. N. BURKETT AND H. A. SCHNEIDERMAN TABLE II Effect of oxygen concentrations on the valve behavior of the fourth right abdominal spiracle Gas mixture Pupa no. Spiracular Ganglionic Valve response* 4RAS** %02 %C02 %o, %cos 10-1 3 3 3 3 5 3 3 21 3 1 21 3 21 3 1 21 3 23 3 3 5 3 "5 3 2 5 3 21 3 1 21 3 21 3 1 21 3 5 3 2 10-2 3 3 3 3 5 3 3 21 3 1 21 3 21 3 1 21 3 3 3 5 * 1 == closed, motionless; 2 = closed, pulsating; 3 == low amplitude nutter; 4 == medium amplitude flutter; 5 == fully open, moving; 6 = fully open, motionless. ** 4RAS = fourth right abdominal spiracle. Control of Spiracular valve behavior To determine whether the central nervous system responds to intratracheal oxygen or carbon dioxide, or to oxygen or carbon dioxide dissolved in the body fluids, the exposed fourth abdominal ganglion (with its tracheae intact) was per- fused about 6 hours after the square of integument had been removed. When both the ganglion and the tracheal system were perfused simultaneously with the same gas mixture, spiracles three, four, and five responded similarly. Moreover, all valves behaved alike even when the ganglion and the tracheal system were TABLE III Effect of oxygen concentration on the COz-trigger threshold for Spiracular valve opening Gas mixture Valve response* Pupa no. Spiracular Ganglionic %C>2 %co-. %0s %COi 4RAS .^RAS 5RAS** 10-1 8 7 21 3 1 3 3 5 7 21 3 5 5 5 5 7 50 3 6 6 6 10-2 5 7 21 3 6 6 6 * 1 == closed, motionless; 2 = closed, pulsating; 3 = low amplitude flutter; 4 == medium amplitude flutter; 5 = fully open, moving; 6 = fully open, motionless. ** 3RAS = third right abdominal spiracle; 4RAS = fourth right abdominal spiracle; 5RAS = fifth right abdominal spiracle. SPIRACLE CONTROL IN INSECTS 285 perfused simultaneously, each with a different gas mixture. Apparently tracheae are not very permeable to externally applied gases. It is the tracheal gas per sc that affects the ganglion, not gases dissolved in the hemolymph (or Kel-F oil) surrounding the ganglion. The key steps in these experiments were detracheating and perfusing the ganglion with ganglionic flows. Tables I-III include representative data taken from recordings of valve responses to various perfusions following detracheation. Figure 6 shows sample records. Table I shows the effects of carbon dioxide in the ganglionic flow on the valve behavior of the fourth spiracle. When the carbon dioxide concentration was 7 to 30% — concentrations that have been demonstrated repeatedly to cause valve opening when the spiracles are per- fused with them — the spiracle remained constricted and pulsating. The carbon dioxide concentration over the ganglion had to be above 30% (pupa 10-2) before the fourth spiracle could be induced to open. Physiological concentrations of carbon dioxide (3 to 7%) perfused through the ganglion had no effect on spiracular behavior. When the spiracles were perfused with 5% O2 + 7% CO2, all the spiracular valves opened fully although the ganglionic flow was 50% O2 + 3% CO2. On the other hand, 5% O2 + 7% CO2 perfused over the ganglion did not cause valve opening of the fourth spiracle. Similarly, a ganglionic flow of 8% Oo + 12% COo failed to evoke valve opening by the fourth spiracle, although all valves opened fully when the spiracles were perfused with the same mixture. These data show that the behavior of the spiracles is not affected when the carbon dioxide concentration of the ganglionic flow is in the physiological range, i.e., there is no central carbon dioxide effect. The primary target of carbon dioxide is outside the central nervous system, i.e., peripheral. The results of 30 of 33 simultaneous spiracular and ganglionic perfusions indicate that the ganglion responded to its own concentration of oxygen, not to that of the spiracular flow (Table II). When 21 % O2 + 3% CO2 was perfused through the ganglion, the valve of the fourth spiracle closed although the spiracular flow contained 3% Oa + 3% CO2 (pupa 10-2). The valve of the fourth spiracle of pupa 10-1 could be induced to flutter by a ganglionic flow of 5% O2 + 3% CO2; the valve could be closed again by a ganglionic flow of 21% O2 + 3% CO2. These data show that the ganglion responds to changes in the oxygen con- centration of its own gas supply, i.e., there is a central oxygen effect. The data also indicate that each ganglion serves its own spiracles. Table III shows the effect of the concentration of oxygen in the ganglionic flow on the CO2-trigger threshold for spiracular valve opening. As shown earlier, a spiracular flow of 5% O2 + 7% CO2 causes full opening of all valves. The fourth spiracular valve of pupa 10-1 continued to open fully in 5% O2 + 7% CO2 even when the concentration of oxygen in the ganglionic flow was increased to 50%. The few available data indicate that the concentration of oxygen in the ganglionic flow apparently has no effect on the CO2-trigger threshold for valve opening. Table III also shows one example of the effect of the spiracular oxygen con- centration on the CO2-trigger threshold for spiracular valve opening. When carbon dioxide was kept constant at 7% and the spiracular oxygen level in- 286 B. N. BURKETT AND H. A. SCHNEIDERMAN creased from 5 to 8%, valve responses of the control spiracles changed from fully open in 5% Oo to flutter in 8% O2. The fourth spiracular valve, which had been fully open in S% O2 + 7% CO2, constricted and pulsated when the spirac- ular flow was 8% O2 + 7% CO2. and thus responded differently from its control spiracles. This experiment demonstrated that the spiracular. not the ganglionic, oxygen concentration affects the CO2-trigger threshold for valve opening. It also pro- vided additional evidence that each ganglion serves its own spiracles. However, further experiments were performed to locate the primary site of nervous control of spiracular activity. The ventral nerve cord was cut either pos- terior to, anterior to, or posterior and anterior to the fourth abdominal ganglion. TABLE IV Effect of nerve cord transection on valve behavior Pupa no. Position of cut Gas mixture Valve response* Spiracular Ganglionic Before transection After transection %o> %C02 %02 %co= 4RAS 3RAS 5RAS 4RAS 3RAS SRAS** 7-1 Anterior to 10 3 21 0 2 4 4 6 3 3 4AG 21 0 10 3 4 2 ? 2 2 2 7-2 Anterior to 10 3 21 0 2 4 4 2 2 2 4AG 21 0 10 3 3 2 2 4 9 2 7-3 Posterior to 10 3 21 0 2 4 4 2 4 4 4AG 21 0 10 3 2 4 4 4 2 2 7-4 Anterior and 10 3 21 0 2 4 4 2 4 4 posterior to 4AG 10-2 Anterior to 5 3 3 3 5 2 2 0 0 (7 4AG 3 3 50 3 6 6 6 4 5 5 * 1 == closed, motionless; 2 = closed, pulsating; 3 == low amplitude flutter; 4 = medium amplitude flutter; 5 = fully open, moving; 6 = fully open, motionless. ** 3RAS = third right abdominal spiracle; 4RAS = fourth right abdominal spiracle; SRAS = fifth right abdominal spiracle. thus isolating the ganglion from nerve impulses arising in ganglia either posterior and/or anterior to it. Table IV contains data derived from recordings of valve responses of pupae that had undergone nerve cord transection. When 50% O2 + 3% COo was perfused through the ganglion, the fourth spiracle fluttered (pupa 10-2) ; its controls oscillated about the fully open position in response to a spiracular perfusion of 3% O2 + 3% CO2. In most perfusions the behavior of the fourth spiracle differed from that of its controls. Cutting the nerve cord and isolating the fourth abdominal ganglion did not appear to affect the behavior of the fourth spiracle which responded to the ganglionic flow. These results con- firm other evidence that a spiracle is served primarily by its own ganglion, not by the ganglion in an adjacent segment. To further elucidate the role of the nervous system in controlling spiracular behavior, the spiracular closer muscle was partially denervated by cutting the SPIRACLE CONTROL IN INSECTS 287 mid-lateral nerve (Fig. 3) where it leaves the ganglion. The integumentary wound was sealed with a plastic window. Spiracular gas perfusions and record- ings of the valve behavior of the fourth right abdominal spiracle and its contra- lateral and adjacent control spiracles were begun about 9 minutes later. Per- fusions and records of valve behavior were made at various times up to 30 days after nerve transection. The partially denervated spiracle remained open for 24 hours even in 50% O2. while its controls fluttered in 2\% O2 and closed in 50% O2. Beginning at about 72 hours and continuing as long as 20 days after denervation, the fourth spiracle closed in response to spiracnlar perfusions of 21 and 50% O2, but opened fully in W% O2. Thereafter the fourth spiracle re- mained closed but could be induced to open by perfusions of 100% CO2 for about one minute; the controls continued to flutter in 21 and 10% O2, and to con- strict in 50% O2. These results show that the partially denervated spiracular muscle, like the completely denervated spiracular muscle, is capable of responding to carbon dioxide and oxygen (Beckel and Schneiderman, 1957). However, partial denervation alters the degree of the muscle's responses to changes in the oxygen concentration. In addition the muscle's hypersensitivity to a concentration of oxygen that caused its normal controls to flutter was demonstrated. DISCUSSION Several procedures used in these investigations are either described for the first time or are modifications of techniques reported in other papers. Since interpreting the results depends on understanding the advantages and limitations of each method used, two key experimental approaches, intratracheal perfusion and detracheation of the ganglion, are discussed before the results are interpreted. It is important not only to know the tracheal gas composition but also to be able to control it. One way of doing this is opening the pupal tracheal system to ambient gases of known composition by inserting short tubes past the spiracular valves (Buck and Keister, 1955; Schneiderman, 1960). This approach has the advantage of not affecting the animal by mass internal flows. The tracheal gas composition is not known with certainty "but at least it can be kept constant and the oxygen and carbon dioxide tensions can be varied independently" (Schneider- man, I960, page 520). The behavior of spiracles when the pupal tracheal system was perfused with various gases was very similar to the behavior of spiracles when the insect was placed in the same gases. However, intratracheal perfusion makes it possible to control the average tracheal gas composition. Detracheating a ganglion cuts its normal oxygen supply route, thus making the ganglion hypoxic. One would expect the valves of the same segment to gape as they do initially after exposing a denervated spiracular muscle to low oxygen (Beckel and Schneiderman, 1957), and this happened. Detracheating a ganglion produces "functional denervation" of the pair of spiracular muscles it serves. The fact that the detracheated ganglion perfused with appropriate gas mixtures did not cause the spiracles to gape is evidence that the method of perfusing the ganglion worked. The interaction between carbon dioxide and oxygen in triggering bursts in diapausing pupae of Lepidoptera has been described (Schneiderman, 1960; Levy 288 B. N. BURKETT AND H. A. SCHNEIDERMAN and Schneiderman, 1966b). Their results, based both on observations of spiracular valve behavior and on analyses of tracheal gases, showed that the triggering carbon dioxide concentration varied with the tracheal oxygen concentration. The carbon dioxide concentration needed to cause prolonged valve opening increased as the intratracheal oxygen concentration increased. Results of the present experiments confirm these earlier reports. Moreover, our data also illuminate another aspect of oxygen-carbon dioxide interaction, the effect of the carbon dioxide concentra- LU o: LU X Q_ 00 O ^ h- LU o tr LU Q_ i CM Q? 50- • CONSTRICTED C FLUTTERING O OPENED 40- 30- 20- 10- PC02-PER CENT ATMOSPHERE FIGURE 7. Oxygen-carbon dioxide interaction curves for pupa 10-2. See text for details. SPIRACLE CONTROL IN INSECTS 289 tion on the Oo-flutter threshold of the valves (the concentrations of oxygen at which fluttering begins). If one plots the intratracheal oxygen-carbon dioxide concentrations at which flutter occurs, in addition to the concentrations of these gases that cause valve opening, the two resulting curves define valve behavior over a range of oxygen and carbon dioxide concentrations. Using data obtained from perfusion experi- ments, such dual O2-CO2 interaction curves were constructed (Fig. 7). The area between curves A and B includes all combinations at which flutter was ob- served ; the area between the ordinate and curve A includes all combinations at which pulsation or constriction occurred. The area delimited by curve B. the ordinate, and the abscissa includes all combinations for which valves opened. These curves show that when the level of carbon dioxide is sufficiently high, flutter occurs at oxygen concentrations to which valves usually respond by constricting. Thus flutter occurred in 21% O2 when the carbon dioxide concentration was about 14% ; the valves of this same pupa constricted and pulsated in 21% O2 when the carbon dioxide concentration was 3%. Levy and Schneiderman (1966a) acknowl- edged the possibility that carbon dioxide could influence the O2-flutter threshold but they stated that "until Pco2 rises above 6% it exerts no marked influence on the response of the spiracles to POO" (page 97). Results of the present experi- (In the normal respiratory activity of diapausing Cecropia pupae, flutter can and whereas marked effects are evident at carbon dioxide concentrations above 10%. ments confirm their conclusion, and reveal no conspicuous effects of carbon dioxide on the Oo-flutter threshold when the carbon dioxide concentration is below 7%, does occur when the intratracheal oxygen concentration is between 18 and 21%. This phenomenon will be discussed in a subsequent paper (Burkett and Schneider- man, 1974).) Further evidence that increasing the concentration of carbon dioxide affects the flutter threshold comes from re-examining data derived from tracheal gas analyses (Levy and Schneiderman, 1966b, pages 108, 109, Figures 2a, 3a). If these data are equated to corresponding valve behavior, they can be plotted to show the same relationships as in Figure 7, confirming qualitatively the inter- relationships of carbon dioxide and oxygen on the flutter threshold. The following experimental results indicate that in the diapausing Cecropia pupa, carbon dioxide affects spiracular behavior by acting peripherally on the spiracular mechanism and not on the central nervous system. The fourth ab- dominal spiracle did not respond when the fourth abdominal ganglion was per- fused with pure carbon dioxide. Neither the experimental spiracle nor its con- trols opened fully and remained motionless when the detracheated fourth abdominal ganglion was perfused with gases in which the carbon dioxide concentration ranged from 7 to 30%. The same gas mixtures perfused systemically caused full and prolonged opening of all valves. Hence in the normal respiratory cycle, as intratracheal carbon dioxide increases, the spiracular mechanism responds long before tracheal carbon dioxide gets high enough to affect the central nervous system. There is no evidence from these experiments that carbon dioxide concentra- tions below 30% affect spiracular behavior either directly or indirectly via the central nervous system. The possibility remains that a direct or indirect peripheral 290 B. N. BURKETT AND H. A. SCHNE1DERMAN response to carbon dioxide may be transmitted over sensory fibers to the central nervous system, which then alters the flow of information to the spiracular muscle. No specific receptors that may respond to changes in the concentration of carbon dioxide or oxygen have been found in the spiracular region although Beckel (1958) made an extensive histological search for such sensory elements. That the spiracular muscle can behave autonomously was shown by studies on the denervated muscle, which relaxed in high concentrations of carbon dioxide (Schneiderman, 1956; Beckel and Schneiderman, 1957) ; present results of partial denervation studies confirm their observations. Hoyle (1960) concluded that in the locust spiracular muscle the partial pres- sure of carbon dioxide in the tracheae near the spiracle triggers spiracular opening. The present study indicates that this is also true for Cecropia pupae. Van der Kloot (1963) passed W% CO2 + 90% O2 over the isolated spiracular nerve-muscle preparation of a Cecropia pupa ; he found that firing in the anterior lateral nerve ceased but some firing continued in the mid-lateral nerve. This indicates a central effect of carbon dioxide on nerves connected with the spiracular muscle. In contrast, our studies indicate that in the intact pupa carbon dioxide acts directly on the spiracular mechanism, not on the central nervous system. How- ever, the possibility exists that sensory information concerning the carbon dioxide concentration or degree of stretch in the spiracular muscle is transmitted to the central nervous system, and this sensory input may affect motor output to the muscle. Oxygen appears to act by a separate mechanism. In the present experiments the valve behavior of the fourth spiracle could be "dictated" by perfusing the detracheated ganglion with a gas of an appropriate concentration of oxygen. Hence oxygen appears to act via the central nervous system. The evidence for a peripheral response to oxygen is inconclusive but it appears that spiracular per- fusions in which the oxygen concentration is 21% or more can mask a central response to oxygen. When the concentration of oxygen in the spiracular per- fusion was above 21%, the fourth spiracle occasionally ceased fluttering even though the ganglionic oxygen concentrations used (5 to 10%) usually induced fluttering. This suggests some peripheral effect of oxygen at high concentrations. But in the normal physiological range of oxygen concentrations (5 to 21%) we have never observed a peripheral oxygen effect. With regard to the nervous control of spiracular valve behavior, the results of the present studies are best evaluated by first considering the results of earlier studies (Beckel, 1958; Van der Kloot, 1963). Electrophysiological studies on the spiracular nerve and closer muscle (Van der Kloot. 1963) indicated that nerve A8 which innervates the muscle (Beckel, 1958) is made up of at least four axons, of which two are from the mid-lateral nerve (Fig. 3). When the anterior lateral nerve \va.s stimulated, the muscle twitched, and Van der Kloot concluded that in the intact pupa motor output from the anterior lateral nerve stimulates the closer muscle to contract. He also noted that when the carbon dioxide concentration was increased, efferent input through anterior lateral nerve ceased. This observation has important implications, for Beckel (1958) stated that the anterior lateral nerve is the median nerve which "arises in the ganglion preceding and courses between the ensheathed connectives to arise from the SPIRACLE CONTROL IN INSECTS 291 ganglion succeeding" (page 91). When the mid-lateral nerve was stimulated, there was a delay before an action potential was observed in the anterior lateral nerve (Van der Kloot, 1963), and Van der Kloot concluded that fibers of the two synapse. Furthermore, nerve cord transection between the third thoracic and first abdominal ganglia eliminated the action potential in the anterior lateral nerve that resulted from stimulating the mid-lateral nerve. This result indicated that the synapse lay outside the ganglion of the stimulated mid-lateral nerve, and that the central nervous control of a spiracle lies in the ganglion of a different segment from that of the spiracle. Van der Kloot found that the contracting spiracular muscle also receives impulses from the mid-lateral nerve, although the role of these axons is not known. In addition he found that even when motor output over the anterior lateral nerve to the muscle ceased (as it did when carbon dioxide was blown over the prepara- tion), the muscle did not always relax since the muscle itself generated potentials. He drew no conclusions as to the exact roles of either nerve in the intact pupa but he suggested two possible functions for the spiracular nerve : The innervated muscle depends in part on excitatory output from the CNS to contract, and the CNS sends inhibitory impulses to the muscle when the carbon dioxide con- centration increases. The following results of the present experiments provide an alternative picture, and support the view that the ganglion and the spiracles which it primarily con- trols are located in the same segment and that the mid-lateral nerve plays a major role in triggering contraction of the spiracular muscle : ( 1 ) Variations in the local oxygen concentration to the fourth abdominal ganglion affect the fourth abdominal spiracle. (2) Detracheating the ganglion has the same effect as denervating the spiracle but does not affect adjacent spiracles. (3) Cutting the nerve cord between the third and fourth, and between the fourth and fifth abdominal ganglia does not change the responses of the fourth spiracle to spiracular or ganglionic perfusions of various gas mixtures. Nor does this operation affect the responses of the third and fifth spiracles although each was deprived of innervation from the fourth ganglion. (4) The behavior of the two spiracles in a segment is coordinated with each other but is not strictly synchronous with that of spiracles in other segments (Van der Kloot, 1963; Brockway and Schneiderman, 1967). Cecropia pupae exhibit discontinuous respiration during which oxygen uptake is continuous whereas carbon dioxide release occurs periodically in "bursts." A carbon dioxide burst begins whenever the intratracheal concentration of carbon dioxide reaches the triggering threshold. The emphasis in this paper on the regula- tion of spiracular behavior by carbon dioxide and oxygen should not obscure the fact that the adaptive significance of discontinuous respiration, whether in insects or in other terrestrial arthropods (Robinson and Paim, 1969) is water conservation (Buck, 1958). What happens to the Cecropia pupa when the temperature in nature falls too low for the neuromuscular activity associated with discontinuous respiration to continue is discussed in a subsequent paper (Burkett and Schneider- man, 1974). We thank Professor Peter Miller for his helpful comments on the typescript. This research was supported by a Biomedical Research Grant to the University of B. N. BURKETT AND H. A. SCHNEIDERMAN Miami and AI- 10527 to the University of California, Irvine from the National Institutes of Health. SUMMARY 1. A technique was developed for perfusing the entire tracheal system of an insect with a known gas mixture at a rate that insured constant average intra- tracheal concentrations of oxygen and carbon dioxide. In addition to this tech- nique, a method was devised to provide a ganglion with its own gas supply, separate from that of the rest of the insect. These techniques enabled us to show the following : 2. The behavior of the spiracles is unaffected when physiological concentrations of carbon dioxide are perfused through the ganglion. However, carbon dioxide perfused through the spiracles does affect valve behavior. Hence the primary response to carbon dioxide is peripheral. 3. The behavior of the spiracles is affected by varying the concentration of oxygen through the ganglion. Hence the primary response to oxygen is central. 4. Central nervous control of spiracular behavior resides primarily in the ganglion of the same segment in which spiracles are located. 5. Not only does oxygen affect the COo-trigger threshold but also carbon dioxide affects the CVflutter threshold. Moreover, the interaction between oxygen and carbon dioxide occurs peripherally. LITERATURE CITED BECKEL, W. E., 1958. The morphology, histology and physiology of the spiracular regulatory apparatus of Hvalophora cccropia (L.) (Lepidoptera). Proc. 10th Int. Congr. Entomol.,2: 87-115. BECKEL, W. E., AND H. A. SCHNEIDERMAN, 1957. Insect spiracle as an independent effector. Science, 126: 352-353. BROCKWAY, A. P., AND H. A. SCHNEIDERMAN, 1967. Strain-gauge transducer studies on intra- tracheal pressure and pupal length during discontinuous respiration in diapausing silkworm pupae. /. Insect Physiol, 13 : 1413-1451. BUCK, J. B., 1958. Cyclic CO2 release in insects. IV. A theory of mechanism. Biol. Bull.. 114: 118-140. BUCK, J., AND M. KEISTER, 1955. Cyclic CO2 release in diapausing Agapema pupae. Biol. Bull., 109: 144-163. BURKETT, B. N., AND H. A. SCHNEIDERMAN, 1974. Discontinuous respiration in insects at low temperatures : intratracheal pressure changes and spiracular valve behavior. Biol Bull, 147: 294-310. EPHRUSSI, B., AND G. W. BEADLE, 1936. A technique of transplantation for Drosophila. Amcr. Natur., 52 : 218-225. HOYLE, G., 1960. The action of carbon dioxide gas on an insect spiracular muscle. /. Insect Physiol, 4 : 63-79. LEVY, R. L, AND H. A. SCHNEIDERMAN, 1966a. Discontinuous respiration in insects — II. The direct measurement and significance of changes in tracheal gas composition during the respiratory cycle of silkworm pupae. /. Insect Physiol., 12 : 83-104. LEVY, R. I., AND H. A. SCHNEIDERMAN, 1966b. Discontinuous respiration in insects — III. The effect of temperature and ambient oxygen tension on the gaseous composition of the tracheal system of silkworm pupae. /. Insect Physio!., 12: 105-121. MILLER, P. L., 1966. The regulation of breathing in insects. Pages 279-354 in J. W. L. Beament, J. E. Treherne, and V. B. Wigglesworth, Eds., Advances in Insect Phys- iology, Vol. 3. Academic Press, New York. SPIRACLE CONTROL IN INSECTS 293 ROBINSON, G. L., AND U. PAIM, 1969. Regulation of external respiration by the book-lung spiracles of the spiders, Arancus diadcmatus Clerck and A. marmorcus Clerck. Can. J. Zool, 47 : 355-364. SCHNEIDERMAN, H. A., 1956. Spiracular control of discontinuous respiration in insects. Na- ture, 117: 1169-1171. SCHNEIDERMAN, H. A., 1960. Discontinuous respiration in insects: role of the spiracles. Biol. Bull., 119: 494-528. SCHNEIDERMAN, H. A., AND A. N. SCHECHTER, 1966. Discontinuous respiration in insects. V. Pressure and volume changes in the tracheal system of silkworm pupae. /. Insect Physiol., 12: 1143-1170. SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1955. An experimental analysis of the dis- continuous respiration of the Cecropia silkworm. Biol. Bull., 109 : 123-143. VAN DER KLOOT, W. G., 1963. The electrophysiology and the nervous control of the spiracular muscle of pupae of the giant silkmoths. Comp. Biochem. Physiol., 9 : 317-333. WILLIAMS, C. M., 1946. Physiology of insect diapause : the role of the brain in the produc- tion and termination of pupal dormancy in the giant silkworm Platvsamia cecropia. Biol. Bull., 90 : 234-243. Reference: Biol. Bull, 147: 294-310. (October, 1974) DISCONTINUOUS RESPIRATION IN INSECTS AT LOW TEMPERA- TURES: INTRATRACHEAL PRESSURE CHANGES AND SPIRACULAR VALVE BEHAVIOR BARBARA N. BURKETT AND HOWARD A. SCHNEIDERMAN Department of Biology, University of Miami, Coral Gables. Florida 33124 and Center for Pathobiology, University of California. Irvine, California 92664 Insects face the problem of conserving body water. This problem is especially acute for pupae of the wild silkworm Hyalophora cccropia which in the northern parts of their range overwinter in environments where the temperature may drop to • — 30° C and below. At these low temperatures the humidity is also low, and there is a tendency for the animals to desiccate. Yet this rarely happens although pupae neither feed nor drink. Their only source of water, in addition to the body water with which they enter the pupal stage, is metabolic water. Most pupal water loss occurs through the spiracles. When the spiracles open, air enters the tracheae and carbon dioxide and water vapor leave. Any mechanism that keeps the spiracles closed will enable the insect to retain water. Buck (1958) suggested that discontinuous respiration is such a mechanism. Discontinuous respiration occurs cyclically ; each cycle consists of a burst, a constriction period, and a flutter period which is followed by another burst, thus marking the beginning of another cycle. During the burst spiracular valves remain open for several minutes and then close for some time in the constriction period. Following the constriction period, valves begin opening and closing continuously or "fluttering," and this is the flutter period. Discontinuous respiration in Cecropia pupae is well-documented but evidence that these pupae exhibit discontinuous respiration at the low temperatures they experience is conflicting. Levy and Schneiderman (1966) reported cycles at 8.5° C, and Brockway (1964) recorded cycles at 0° C. However, Kanwisher (1966) concluded that there is no discon- tinuous respiration below 10° C and that pore diffusion accounts for gas exchange at low temperatures. Understanding how these pupae conserve water at low temperatures and low humidities depends mostly on understanding how the spiracles behave under such conditions. This report examines the effects of low temperature on cyclical respiratory activity and on the behavior of spiracular valves. MATERIALS AND METHODS Experimental animals Diapausing pupae of Hyalophora cccropia had their brains removed three months before the experiment began to insure permanent diapause (Williams, 1946). Pupae then were stored at 22-25° C and 80-90% R.H. until used. To record intratracheal pressure changes, the spiracles of pupae were cannulated 294 DISCONTINUOUS INSECT RESPIRATION 295 as described previously (Burkett and Schneiderman, 1974). The pupae then were placed in a water-saturated environment for two weeks to recover. Recording intratracheal pressure changes Intratracheal pressure changes were recorded continuously except during tem- perature equilibrations. The methods used were similar to those described in previous papers (Schneiderman and Schechter, 1966; Brockway and Schneiderman, 1967). Pressure changes as small as 0.025 mm Hg could be detected (Brockway and Schneiderman, 1967). As in these earlier papers it should be emphasized that the transducer repsonded to pressure changes in a system composed of the tracheal system, the cannula, and the transducer itself. Hence the recorded pressure changes were smaller than the actual intratracheal pressure changes that occurred in an intact pupa. Recording spiracular valve behavior Beckel (1958) described the anatomy of the spiracles of Cecropia in detail. The width of the spiracular opening is regulated by the spiracular valve which is moved by a closer muscle. In one set of experiments reported here spiracular valve movements and intratracheal pressure changes were recorded simultaneously. To observe the valves of the third, fourth, and fifth right abdominal spiracles, they were exposed by scraping away the filter apparatus and gently pushing aside the peritreme with a hot needle. To keep the valves moist, a single transparent plastic window was sealed in place over them with melted paraffin. The valves were observed daily to insure that no sticking or desiccation had occurred. The method described previously (Burkett and Schneiderman, 1974) for recording valve movements was used. Valve movements throughout at least one complete respiratory cycle were recorded at 20°, 15°, and 10° C. Below 10° C very long cycles precluded continuous direct observations and recordings of valve movements ; thus 20-minute records with 20-minute intervals between recordings were made during the flutter and constriction periods of the cycle, and usually throughout the entire burst period. At - - 10° to • - 20° C valves were observed for 20-minute periods once every two hours. Recording of intratracheal pressure changes at different temperatures Intratracheal pressure changes of two pupae were studied by immersing the pupa-transducer system in a water-ethylene glycol bath. This method worked well at temperatures above 0° C. But at 0° C and below, recording of intra- tracheal pressure is affected noticeably by changes in room temperature and pres- sure. For this reason, and to permit observation of the spiracular valves at low temperatures, a different approach was developed. A pupa and the pressure trans- ducer to which it was attached were placed in a small deep freezer fitted with a viewing window (Fig. 1). The pupa and transducer were equilibrated at each temperature for 24 hours before recording was begun. Temperatures were lowered successively from + 25° 296 B. N. BURKETT AND H. A. SCHNEIDERMAN TO POLYGRAPH PRESSURE TRANSDUCER TRANSDUCER CANNULA (FROM 3LAS) GAS MIXTURE FROM PERFUSION PUMP OUTLET CANNULA (FROM 6 LAS) COIL PLASTIC WINDOW (OVER 3RAS.4RAS, 5RAS) INLET CANNULA (TO 2RAS) FIGURE 1. This assembly for simultaneously recording intratracheal pressure and valve movements was placed in a freezer, the temperature of which could be held constant to within ±0.5° C at each experimental temperature between ambient and -20° C. An infusion pump forced the experimental gas, cooled in the coil to the temperature of the freezer, into the tracheal system via the "inlet" cannula in the second right abdominal spiracle (2RAS). The "outlet" cannula in the sixth left abdominal spiracle (6LAS) prevented pressure build-up as a result of intratracheal perfusion. The cannula in the third left abdominal spiracle (3LAS) connected the tracheal system to the pressure transducer. To establish the baseline of intra- tracheal pressure records when the temperature was changed, and to check the size of a developing intratracheal vacuum without opening the freezer, the tracheal system was opened to the atmosphere by using two-way normally closed solenoid control valves (Allied Control Company, Plantsville, Connecticut). When the control valve of the transducer (Vt) was energized, fluid would have been forced from the transducer dome into the pupa had a second valve (Vp) not been used to equilibrate pressure. Movements of the third, fourth, and fifth right spiracular valves (3RAS, 4RAS, 5RAS) were observed and recorded as described earlier (Burkett and Schneiderman, 1974). DISCONTINUOUS INSECT RESPIRATION 297 or +20° C to --20° C in 5° intervals, and they were increased similarly to + 20° C after recording at - 20° C. Intratracheal perjusions at lo^v temperatures In one series of experiments the responses of the spiracular valves to oxygen and carbon dioxide at low temperatures were examined. The composition of the intratracheal gas was regulated by perfusing known concentrations of oxygen and carbon dioxide through the tracheal system as described earlier (Burkett and Schneiderman, 1974). TABLE I Summary of data from records of intratracheal pressure and valve movements of pupa II at different temperatures Tem- pera- ture (°C) Xo. of complete cycles recorded Average length of cycle (min) Average length of burst (min) "Open" phase (min) "De- cline" phase (min) \\ ( -rage length of constric- tion (min) Average intratracheal vacuum developed during constriction Average length of flutter (min) Average number of valve move- ments/ (mm Hg) mm + 20 6 341 30 21 9 41 -1.99 270 32 (8.8)* (70) (30) (12.0) (79.2) + 15 6 365 72 59 13 56 -2.35 237 25 (19.8) (82) (18) (15.4) (64.8) + 10 3 706 85 45 40 104 -4.5 517 19 (12.0) (53) (47) (14.7) (71.8) +5 6 839 85 34 51 377 -5.24 377 10 (10) (40) (60) (45) (45) 0 1(11) 15l7f 18()f 80f lOOf 187(11) -3.13(11) 1229f 1 2(111) (11.9) (44.4) (55.6) (12.3) (80.9) -5 0 > 16,200 Data — — — — • 16,180 1/min- incomplete 1/hr -10 0 — — — • — — — • — 0 -15 0 — — • — — • — . — — 0 -20 0 — — — — — — — 0 * Numbers in parentheses indicate per cent of cycle occupied by a given phase. Those for open and decline phases of burst are per cent of burst. f Average of all such phases recorded from two pupae (II and III). The experimental gas mixture was cooled to the temperature of the freezer before the mixture entered the tracheal system. Ten minutes were allowed for complete tracheal perfusion and equilibration with the gas before valve responses to the mixture were recorded. At least two hours were allowed for recovery from one gas mixture before a different mixture was introduced into the tracheal system. Air was perfused through the tracheal system during the recovery period. Other procedures will be discussed in the appropriate sections below. At least three cycles were recorded for each pupa at 5° intervals from + 20° to + 5° C. To record three complete cycles at 0° C two pupae wre used. No complete cycles were recorded below 0° C. 298 B. N. BURKETT AND H. A. SCHNEIDERMAN RESULTS Effects of decreasing temperatures on cyclical respiratory activity Tables I, II, and III summarize data obtained from records of intratracheal pressure changes in one pupa. These data are representative of those obtained from all pupae used in these experiments. Figures 2, 3. and 4 show portions of the respiratory cycle at 0° and -- 5° C. The following changes in cyclical respiratory activity occurred as the ambient temperature decreased: (1) Discontinuous respiration persisted down to -5° but not at - - 10° C or below. (2) The duration of the respiratory cycle increased TABLE 1 1 Average length (seconds ± s. e.) of microcycles during different parts of the flutter period Temperature (°C) Part of flutter period Beginning Middle End 20 57.4* ± 41.03** 58.53 ± 32.71 43.2 ± 16.24 (10-160)t (20-134) (10-70) 15 54.83 ± 8.03 67.00 ± 27.07 47.4 ± 16.92 (10-120) (20-230) (10-80) 10 68.30 ± 47.84 76.33 ± 52.83 42.40 ± 24.77 (20-170) (20-200) (10-85) 5 202.97 229.73 ± 223.19 125.77 ± 109.39 (4-833) (20-820) (10-200) 0 452.43 ± 418.12 809.57 497.83 (20-1040) (20-6864) (30-3160) o 460 ± 423.72ft — — (60-7330) * Each average figure is based on a total of 30 microcycles (ten from each of three cycles) from records of one pupa. ** S. D. f Range. ft Based on all microcycles recorded. as the temperature decreased. Cycle length approximately doubled between 20° and 10° C, and between 10° and 0° C. (3) At temperatures above 0° C the pattern of the respiratory cycle was unchanged, that is, cycles were made up of a burst, during which valves remained fully open ; a constriction period, during which valves remained closed ; a pressure rise period following constriction, during which valves opened fully for several seconds and then closed again ; and a flutter period, during which valves continuously opened briefly and then closed for several seconds. At 0° C, however, bursts and flutters persisted but con- striction disappeared in some cycles. At --5° C constriction disappeared altogther. Changes in burst, constriction, and flutter The percentage of a total cycle occupied by the burst did not change system- atically as temperature decreased (Table I). The burst itself is made up of two distinct phases (Schneiderman, 1960) ; an "open" phase, during which DISCONTINUOUS INSECT RESPIRATION 299 TABLE III Average intratracheal vacuum (mm Hg ± s. e.) developed during microcycles in different parts of the flutter period Part of flutter period Temperature (°C) Beginning Middle End 20 -0.113* ± 0.067** -0.096 ± 0.047 -0.068 ±0.041 (-0.025 to -0.125)t (-0.025 to -0.125) (-0.025 to -0.050) 15 -0.089 ±0.071 -0.089 ±0.045 -0.080 (-0.025 to -0.125) (-0.025 to -0.150) (-0.025 to -0.050) 10 -0.188 ±0.180 -0.105 ± 0.075 -0.067 ±0.041 (-0.025 to -0.200) (-0.025 to -0.250) (-0.025 to -0.125) 5 -0.256 -0.204 -0.060 ± 0.037 (-0.075 to -0.300) (-0.150 to -0.250) (-0.025 to -0.150) 0 -0.260 ± 0.182 -0.353 -0.073 ± 0.049 (-0.025 to -0.200) (-0.150 to -0.375) (-0.025 to -0.150) -5 -0.149ft (-0.04 to -2.273) * Each average figure is based on a total of 30 microcycles (ten from each of 3 cycles) from records of one pupa. ** S. D. f Figures in parentheses indicate ranges of microcycles examined. If Based on all microcycles recorded. a (- ^_ 4 300 D 44000 1 7 - E p — _i_ /--- +Tl J-l j : / . / . J - i ' 'l - •- \+ L_ •4 t= T J s — j ; ; £ t 1 t 45000 46000 47DOO / — - fr^"i^ J u f^— -i. j ' i (- 1- - _r.: p^, — ^-^_L,r - -j" i 1 3 1 [ , i ^ -r^r- ^ h~i — — ' 1 E — i , i i i S ! j j : ~* t ' 46OOO FIGURE 2. Intratracheal pressure record of pupa II at 0° C. Since this particular cycle lasted 28 hours, only a portion of the flutter period immediately preceding the burst is shown. The chart speed in this and subsequent records was 0.25 mm/sec unless otherwise indicated. Thus one small division between two thin vertical lines on the chart represents 20 seconds, and one large division between two heavy vertical lines represents 100 seconds. Recordings in this and other intratracheal pressure records were made at a sensitivity of 0.01 mV/cm. 300 B. N. BURKETT AND H. A. SCHNEIDERMAN valves are fully open and generally motionless; and a "decline" phase, during which valves oscillate about the fully open position. The percentage of the burst occupied by the decline phase increased as the temperature dropped below 10° C ; 3= ' " ~r /' Mr-M' 56000 tl ^ — li / / ouuu EZ^: *F s* =5FJB= •^-^J* [ \*J^"' "*" J -v - n. ^^^ . . *i4 -— 4 -r^= 1 ' ' — 1 iilfligp i • - — i seooo 5?DOO ^g ^.(^ ', — f«< | 1 "—Hz: g ~[.:':'.:^ t:; fr ; -P-4L -t , H - — [ '^T 66000 s 1"- 1 — i — - — i — . -i - - 3 1 E! i , i= = i — r^- -i — — ^p^-j-, Kr -:^2 ,^E^ nooo 73000 wxo * t! " ^~ ^-t , '-i M rnr , . T U4- TT 4iv n ' - ; ^^-Fj __J , ._. L i ; f M _j 1_ ^r ftt *j t i I 44-^ ^ t^4^i-T--i K. . -( 75COO T7000 7BCDO FIGURE 3. Intratracheal pressure record of pupa II during part of a burst at 0° C. A portion of the open phase (B0) (54,500 to 68,000 seconds) and the entire decline phase (Bd) (68,000 to 77,400 seconds) are shown. Note the microcycles in the decline phase, for example, the one beginning at 76,000 seconds and ending at 77,420 seconds. DISCONTINUOUS INSECT RESPIRATION 301 r :~ • 959000 FIGURE 4. Intratracheal pressure record of pupa VI at — 5° C. Since the cycle lasted at least 11 days, only part of the decline phase and a portion of the flutter period are shown. (Chart speed = 1 mm/sec.) at 15° C it was 18%, whereas at 5° C it was 60% of the burst. At 15° C and below, the decline phase was marked by microcycles like those occurring during flutter. The proportion of the cycle occupied by the constriction period did not change appreciably between 25° and 10° C, but at 5° C constriction was proportionally longer (45% of the cycle) than at other temperatures. The intratracheal vacuum that developed during constriction increased steadily from - 1.99 mm Hg at 20° C to -- 5.24 mm Hg at 5° C, and then decreased to -- 3.13 mm Hg at 0° C (Table I). Above 0° C the flutter period occupied a greater proportion of the cycle than either the constriction period or the burst, except at 5° C where the flutter and the constriction periods were about equally long. In two cycles at 0° C the decline phase of the burst was followed not by constriction but by the flutter period ; how- ever, flutter was interrupted by periodic bursts. In another pupa the cycle at 0° C consisted of the usual burst, constriction, and flutter periods. Although no com- plete cycle was recorded at - 5° C, recordings of intratracheal pressure and valve behavior were begun during the early decline phase; like most cycles at 0° C, no constriction period was observed. The decline phase was followed by flutter which persisted for at least 11 days when the experiment was ended. Changes in microcycles during the flutter period Although valves open and close through a flutter period, they are closed most of the time (Schneiderman, 1960). When the valves open, intratracheal pressure rises to near atmospheric; when they close, intratracheal pressure falls. These 302 B. N. BURKETT AND H. A. SCHNEIDERMAN fluctuations in intratracheal pressure are microcycles. At 20° C an average micro- cycle lasted about 53 seconds but the duration varied from 10 to 160 seconds (Table II). Table II shows that as the temperature decreased, the average duration of microcycles in the flutter period increased. There was wide variation in the duration of microcycles at a given temperature, particularly low temperatures. However, Table II shows that the longest microcycles occurred at the lowest experimental temperatures. For example, at - 5° C a microcycle lasting 7330 seconds (2 hours) was observed whereas at 20° C the longest microcycle was 160 seconds. On the other hand, relatively brief (60 seconds) microcycles occurred at -- 5° C as well as at higher temperatures. Microcycles occurring just before the onset of a burst appeared to be shorter and to result in a smaller intratracheal vacuum than those at the beginning or middle of the flutter period. To determine whether these differences were sta- tistically significant, the duration and the size of the intratracheal vacuums in dif- ferent parts of the flutter period were measured. The first ten microcycles at the beginning, the ten in the middle, and the last ten just before the onset of a burst were chosen in each of three cycles of one pupa at 5° intervals between 20° and 0° C. The results are shown in Tables II and III. At a given temperature the length of microcycles was not significantly dif- ferent (at the 0.05 level) in different parts of the flutter period. When the same parts of the flutter period were compared at different temperatures, there were no significant differences in the length of microcycles at 20° as compared with 15° C, or at 15° as compared with 10° C. However, below 10° C the duration of micro- cycles in a given part of the flutter period increased significantly between 10° and 5° C, and between 5° and 0° C. The following general observations were made regarding intratracheal vacuums during flutter: (1) At a given temperature the vacuums occurring just before bursts were smaller than those at the beginning or in the middle of the flutter period. However, vacuums occurring before a burst at 0° C were not significantly larger than those at 20° C. (2 ) Vacuums occurring in microcycles at the beginning and in the middle of the flutter period were significantly larger at 0° than at 20° C. (3) There was a wide range in the sizes of intratracheal vacuums occurring during the flutter period at a given temperature ; however, the largest vacuums during flutter occurred at low temperatures (Table III). Effects of low temperatures on ike behavior of spiracular valves To determine how low temperatures affect the behavior of the spiracular valves, a pupa was exposed to different temperatures, and records of intratracheal pressure and movements of the third, fourth, and fifth right abdominal spiracular valves were made. As the ambient temperature decreased, the average number of valve movements during the flutter period decreased from 32 per minute at 20° C to 1 per minute at 0° C (Table I). At - 5° C the number decreased even further, and some- times valve movements occurred only once an hour. However, even at low tem- peratures there were periods when valves moved more frequently than on the average. DISCONTINUOUS INSECT RESPIRATION 303 "Pressure chonges 10C Valve movements MjuJLA..^^ FIGURE 5. Simultaneous records of intratracheal pressure and behavior of 4RAS during the beginning of a pressure rise period at 10° C. Full opening of the valve is represented by a pen deflection of 5 divisions above baseline. (Chart speed = 1 mm/sec.) Valves were capable of opening fully even at - 5° C but at this temperature opening often occurred so slowly that the movement could not be detected as it occurred. In this case movements were recorded only after it was obvious that the valve had moved. Correlation of valve behavior with in tra trachea! pressure changes Most pressure rises were accompanied by at least halfway opening of the fourth right abdominal spiracle, the spiracle we usually observed in all these experi- ments, although not every wide valve opening led to a pressure rise. Records of flutter occasionally showed that the valve of the fourth spiracle frequently began opening before a pressure rise occurred. However, many pressure rises were accompanied by, rather than preceded by, valve opening of the fourth right spiracle. During the open phase of a burst at all temperatures, the valve of the fourth spiracle remained open except for occasional partial closures, so brief that the intratracheal pressure scarcely wavered from atmospheric. During the late portion of the decline phase of the burst, the intratracheal pressure records were very similar to those made during the flutter period; there were discrete microcycles, and the beginning of most microcycles was accompanied by wide opening of the valve of the fourth spiracle. However, not every wide opening was accompanied by a pressure rise. During the constriction period, the valve of the fourth spiracle remained closed and motionless ; at the same time the intratracheal pressure fell steadily. Con- striction was followed by a step-wise pressure rise period (Fig. 5). As this figure shows, seven of the eight increments in pressure rise were accompanied by opening of the valve at least halfway. Responses to oxygen and carbon dioxide To determine what the spiracular valves do at temperatures below - 5° C, the following experiments were performed : 304 B. N. BURKETT AND H. A. SCHNEIDERMAN Ten pupae (5 per group) were weighed individually and then were placed immediately into one of two 650 ml desiccators, each containing a desiccant, CaClo (Drierite) and a carbon dioxide absorbent, Ca(OH)2 (Ascarite). Group I was kept at -- 2° C and Group II at - - 16° C for one week. At - - 16° C the pupae were frozen and their blood was solid. Pupae then were removed individually from the desiccators and were reweighed. The average weight loss of pupae in Group I was 1.29 ± 0.56 (S.D.) % of the original weights; the average weight loss of pupae in Group II was 0.65 ± 0.29%, or about half that of pupae in Group I. Since the spiracles are the major site of water loss, the valves of pupae at -- 2° C must have been open more often than those of pupae at • • 16° C. The desiccator containing the Group II pupae then was flushed at room tem- perature with a mixture of 20% CO2 + 80% air to open all spiracular valves of the pupae. The desiccator, now containing 20% CO2, again was placed at - 16° C for one week. The average weight loss following this treatment was 1.44 ±0.65%, or 2.2X the average weight loss of the same pupae at - 16° C in air. When the spiracular valves were opened and the pupae subsequently frozen, pupal weight loss at low temperatures, i.e.. - 16° C, was comparable to pupal weight loss at higher temperatures, i.e., --2° C. The average weight loss of two other groups of pupae (III and IV) in air at - - 16° C was determined. Then, without permitting warming of either desic- cator, the one containing Group III was flushed for 5 minutes with air while the desiccator containing Group IV was flushed simultaneously with 20% CO2 + 80% air. Both desiccators were returned to - 16° C. When these pupae were reweighed at the end of one week, those in Group III had lost an average of 0.40 ±0.20%, and pupae in Group IV an average of 0.22 ±0.13% of their weight of the previous week. Pupal weight loss, even in an atmosphere in which the carbon dioxide concentration was high, was comparable to weight loss in air at the same temperature (0.40% and 0.37% for Groups II and IV, respec- tively). Thus it appears that the spiracular muscle does not respond to high carbon dioxide concentrations at low temperatures (-- 16° C). Possible explana- tions for failure of the valves to respond to carbon dioxide at low temperatures are considered in the Discussion. Direct observations of the spiracular valves and recordings of movements of the fourth spiracle revealed the following : Valve movements occurred at all experimental temperatures between ± 20° and - 5° C. No movements were observed at - - 10° C or below. When 20% CO2 4- 80% air was perfused through the tracheal system at temperatures between + 20° and 0°, the fourth spiracle and its adjacent controls opened fully within 5 minutes after perfusion was begun, and remained open and motionless for the duration of perfusion. At --5° C these valves were about three-fourths open after 5 minutes of perfusion, but full valve opening could not be elicited at — 10° C or below even when 50% CO2 + 50% air was perfused through the tracheal system for one hour. This result indicates that at -- 5° C or below the spiracular muscle does not respond to carbon dioxide. To determine whether the spiracular response at -5° C is solely to carbon dioxide or if the spiracular mechanism also responds to oxygen, 20% CO2 + 80% air was perfused through the tracheal system. Ten minutes after perfusion was begun, the valves were three-fourths open and motionless. At this point, with the valves still open, the temperature was lowered to • • 10° C, and an air per- DISCONTINUOUS INSECT RESPIRATION 305 fusion replaced the 20% CO2 + 80% air flow. The air flow was continued for 17 hours with no change in the position of the valves. Nor did a perfusion of 100% O2 for two hours elicit valve closure. Then the pupa, receiving an intra- tracheal air perfusion, was allowed to warm to • - 5° C; about seven hours later the valves closed. In a second approach to the same question, intratracheal perfusions of both 1 and 0.5% O2 (balance N2) elicited opening of the valves about 20 minutes after the flow was begun at — 5° C. The spiracular mechanism is capable of responding to intratracheal oxygen and carbon dioxide at - 5° but not at - 10° C or lower. DISCUSSION Results of the present studies show that cycles of discontinuous respiration in Cecropia pupae persist at —5° C but not at - 10° C. These results confirm and extend those of Brockway (1964) and Levy and Schneiderman (1966) but contradict those of Kanwisher (1966). Since discontinuous respiration depends on the responses of the spiracular nerve and muscle to oxygen and carbon dioxide, neuromuscular activity must also persist down to - 5° C. As expected, the length of the respiratory cycle at low temperatures was greater than at high temperatures. At - 5° C, for example, the length of the cycle was at least 50 times greater than at 20° C even though two factors, a decrease in the CO2-trigger threshold and an increase in the length of micro- cycles during the flutter period, tend to shorten the cycle. At low temperatures the COs-trigger threshold (the concentration of carbon dioxide that causes pro- longed opening of all spiracular valves) decreases (Levy and Schneiderman, 1966), and this means that less carbon dioxide than usual is required to initiate a burst. On the other hand, an increase in the length of microcycles during the flutter period means that the proportion of the flutter period during which some out-diffusion of carbon dioxide can occur, i.e., the pressure rise period of micro- cycles, is much less at 0° than 20° C. Brockway and Schneiderman (1967) estimated that if the average length of a microcycle were 25 seconds, the spiracular valves are open no more than 8% or, more likely, 5% of the flutter period. Since an average microcycle at 0° C may be 10 times as long as at 20° C, more carbon dioxide is retained at lower temperatures than at 20° C (cf. Schneiderman and Williams, 1955). Since the length of the respiratory cycle increased at low temperatures, the combined effects of a reduced CO2-trigger threshold and the increased length of microcycles clearly were offset by other factors. The key factor is that at low temperatures the metabolic rate of insects is lower than at high temperatures. Consequently, carbon dioxide production is depressed, and a longer time elapses before carbon dioxide reaches the CO2-trigger threshold, reduced though it may be. In addition, more carbon dioxide dissolves in tissue fluids at low than at high temperatures, and this also increases the time required for carbon dioxide to reach the triggering threshold. Significant changes occurred during the various phases of the cycle at low temperatures. One of the most noticeable changes occurred during the decline phase of the burst. Not only did the decline phase increase in length from an average of 9 minutes at 20° C to 51 minutes at 5° C (Table I) but also it 306 B. N. BURKETT AND H. A. SCHNEIDERMAN was detected from intratracheal pressure records at 15° C or below, but not at 25° or 20° C. A burst ends when the tracheal carbon dioxide falls to a minimum level (Schneiderman, 1960). Although tracheal oxygen quickly approaches ambient soon after a burst begins, the valves remain wide open for some time before they begin fluttering for an additional period (the decline phase), pre- sumably because the pupa requires time to "unload" its accumulated carbon dioxide. One possible explanation for prolonged valve openings during bursts at low temperatures is that the rate of carbon dioxide release is limited by the activity of carbonic anhydrase. Less than 15% of the carbon dioxide released during a burst by diapausing Agapcma pupae is supplied by carbon dioxide in the tracheae at the onset of a burst (Buck and Keister, 1955, 1958), and this release occurs instantaneously. The remaining carbon dioxide escapes relatively slowly from the tissues during the remainder of the burst. There is enough carbonic anhydrase in the tissues of Cecropia pupae to account for the volume of carbon dioxide released during a burst (Buck and Friedman, 1958), and at low temperatures there is a decrease in the rate at which the enzyme acts. Thus, a pupa requires many minutes to unload carbon dioxide since the animal releases 2 to 3 times as much carbon dioxide during a burst at low temperatures as at higher tempera- tures (Schneiderman and Williams. 1955). Both the slow rate at which carbon dioxide is released and the larger volume of carbon dioxide that is impounded at low temperatures cause the valves to remain open for long periods. There are three noteworthy points about the decline phase at low tempera- tures : ( 1 ) The initial part of the decline phase cannot be detected from intra- tracheal pressure records. (2) Fluttering and microcycles occur in later parts of the decline phase. (3) Most important, these flutters and microcycles, which constitute a second period of discontinuous respiration, persist even though the concentration of oxygen in the tracheae is close to ambient (20%). At first this is surprising since it is a lou* concentration of oxygen, i.e., 5%, that normally triggers fluttering (Schneiderman, 1960; Burkett and Schneiderman, 1974). ' Why do microcycles appear during the decline phase at a time when the tracheal concentration of oxygen is much higher than that which normally triggers fluttering? If carbon dioxide is indeed released slowly at low temperatures, the spiracles are still under the influence of carbon dioxide during the decline phase. The Oo-flutter threshold (the concentration of oxygen at which fluttering begins) can be raised by increasing the carbon dioxide concentration (Levy and Schneider- man, 1966; Burkett and Schneiderman, 1967, 1974). Thus fluttering can be induced even though the oxygen concentration in the tracheae is about 20% if the carbon dioxide concentration is about 7% (Burkett and Schneiderman, 1974), as it is during the decline phase (Levy and Schneiderman. 1966). Hence one expects fluttering during the decline phase. Although the constriction period was proportionally longer at 5° C than at other temperatures (Table I), it disappeared in some cycles at 0° C and never was observed at --5° C. Both the existence and the duration of the constriction period depend on the metabolic rate of the pupa. The size of the intratracheal vacuum, which was also greater during constriction at 5° C than at other tem- peratures, depends not only on the metabolic rate but also on changes in the volume of the tracheal system with intratracheal pressure changes, and on the DISCONTINUOUS INSECT RESPIRATION 307 rate at which air leaks into the tracheal system (Schneiderman, 1960; Brockway and Schneiderman, 1967). At 5° C the metabolic rate of pupae still was high enough to cause them to use oxygen more rapidly than air leaked into the tracheae ; as a result, a large intratracheal vacuum developed. In two cycles at 0° C the decline phase led directly into a flutter period instead of the usual constriction period. The reason was this : Since the metabolic rate of the pupa was quite low, the rate at which the animal used oxygen was only slightly greater than that at which air leaked into the tracheae. Thus a sig- nificant vacuum, i.e., one that would facilitate the mass transfer of air past the valves, could never develop. Eventually the intratracheal oxygen concentration dropped to the level that triggers fluttering and microcycles. The presence or absence of the constriction period at low temperatures depends on metabolic rate ; pupae with a sufficiently high metabolic rate will have a constriction period, and those that have a lower metabolic rate will not. The flutter period generally lengthened as the temperature decreased. This was as expected since the metabolic rate also decreased. However, at 5° C the flutter period and the constriction period were about the same length. A possible explanation for this is that the large intratracheal vacuums which developed during both the flutter and the constriction period at 5° C impeded out-diffusion of carbon dioxide ; consequently, more carbon dioxide than usual accumulated. Thus the time required to reach the CO2-trigger threshold decreased, and the flutter period was shortened. The flutter period is a series of microcycles which generally became longer as the ambient temperature decreased (Table II). This, too, was expected since the duration of microcycles, like that of the constriction period, depends on the metabolic rate. When a pupa's metabolic rate is low, the rate at which it uses oxygen decreases. Thus a longer time is required for the concentration of oxygen in the tracheae to drop to the Oo-flutter threshold than when the metabolic rate is high. In the present experiments the size of the intratracheal vacuum developed during flutter also increased except at - 5° C where it was less than at 0° C. During flutter at 25° and 20° C valves are constricted most of the time. At lower temperatures down to 0° C they are constricted even more since there are fewer openings per unit time than at higher temperatures. Consequently, larger vacuums can develop during the flutter period at lower than at higher tempera- tures. At - 5° C the average size of the intratracheal vacuum developed during flutter was less than at 0° C, presumably since the rate of oxygen uptake at - 5° C was so low that air leaked into the tracheae almost as rapidly as oxygen was removed. During microcycles near the end of the flutter period the size of the intra- tracheal vacuum was less than in other parts of the flutter period (Table III). Observations of the spiracular valves explain why (Schneiderman, 1960). Before a burst, valves begin opening and closing more often and opening more widely than in other parts of the flutter period. As a result, intratracheal pressure never has an opportunity to fall much below atmospheric. Most movements of the spiracular valves are not effective (Brockway and Schneiderman, 1967), that is, valves may move yet they may not open the tracheal system to the atmosphere or else they open it so briefly that no pressure 308 B. N. BURKETT AND H. A. SCHNEIDERMAN change occurs. The results of the present experiments confirm this. Changes in intratracheal pressure were accompanied by openings and closings of valves. However, most valve movements, especially at low temperatures, were pulsations- movements not strong enough to break the valve seal or to permit more than very small, very brief openings — and did not cause changes in intratracheal pressure. Two factors must be remembered when attempting to correlate valve move- ments with intratracheal pressure changes: (1) If one spiracle closes, this does not necessarily mean that intratracheal pressure will fall since other valves may be open. (2) The fourth spiracle, which we observed, was covered with a plastic window ; although the spiracle may have opened, it did not open the tracheal system to the atmosphere. The contralateral spiracle, which was not covered with a window, usually behaves in a similar manner but the precise movements of the two contralateral valves may not be exactly the same (Brockway and Schneider- man, 1967) ; in fact, about 20% of the time one of the two opens while the other may move slightly but does not open. Hence a brief opening in the fourth right spiracle may not have led to a synchronous pressure rise since the fourth left spiracle may not have opened at the same time. However, if the valve of the fourth right spiracle remained open for several seconds, other valves usually were also open. There are several possible reasons why valves apparently do not respond to oxygen and carbon dioxide at temperatures below -5° C: (1) The spiracular muscle itself may be insensitive to high carbon dioxide and to low oxygen concen- trations. (2) The central nervous system, the principal target of oxygen (Burkett and Schneiderman, 1967, 1974) may not function; hence nervous control of spiracular activity may be blocked. (3) The enzymes that release the energy needed by a contracting muscle may not function at a level consistent with the energy requirements of the spiracular muscle (cf. Richards, 1958). (4) The muscle may be too viscous to move (cf. Richards, 1958). (5) The fluid film that covers the valves may become viscous, causing them to "stick." (6) The muscle may be frozen. Which, if any, of these alternatives is responsible for the failure of the muscle to respond to oxygen and carbon dioxide is not known. However, the blood of pupae that were chilled to - 10° C under the conditions of our experiments (cooling rate ^0.5° C/min) was frozen (Burkett and Schneiderman, 1968). This may also occur in nature. The fact that discontinuous respiration and response of spiracles to oxygen and carbon dioxide were observed at - 5° C indicates that neuromuscular control of spiracular behavior persisted at this temperature. Although the spiracular muscle is innervated by a nerve from the central nervous system (Beckel, 1958), and is controlled in part by the nerve (Burkett and Schneiderman, 1974), Van der Kloot (1963) found that spontaneous potentials generated in the isolated spiracular muscle cause it to contract. He suggested that when the temperature is very low, it is these spontaneous potentials that keep the muscle contracted. Our results suggest that both the spiracular nerve and muscle are "turned off" at the same temperature or certainly within a few degrees of each other. Fluttering is a significant mechanism by which diapausing pupae conserve water at low temperatures. Although the absolute duration of bursts at low temperatures DISCONTINUOUS INSECT RESPIRATION 309 increases, there are fewer bursts than at high temperatures, and these two factors must offset each other in providing opportunities for pupae to lose water. In the present experiments not only did the flutter period lengthen as the temperature decreased, thus occupying most of the respiratory cycle, but also the number of valve movements during the flutter period decreased. These results support an earlier suggestion that "... the heart of the problem (water conservation) surely lies in the flutter period, where hour after hour the insect practices the 'trick' of filtering in oxygen, while retaining carbon dioxide and water" (Schneiderman, 1960, page 525). At some temperature (between -- 5° and - 10° C in the present experiments) discontinuous respiration ceases. At that point pupae must rely on other mecha- nisms of gas exchange and water conservation. Kamvisher (1966) concluded that at low temperatures, where discontinuous respiration no longer occurs, gas exchange occurs by pore diffusion. Since the pupa's respiratory demands are low (^0.1 mm3/g/hr at -12° C (Kamvisher, as quoted in Asahina, 1966)), pore diffusion probably provides adequate gas exchange. Although less than 3% of the gas exchange at 25° C may occur through the cuticle (Schneiderman and Williams, 1955), at very low temperatures the insect's low metabolic demands may be met partially by this mechanism. But whatever the mechanism, one fact is clear : At low temperatures valves are constricted most of the time, and conditions in nature probably insure that they freeze in the constricted position. Hence the valves continue to serve the insect, enabling it to conserve water, even though they cannot respond to oxygen and carbon dioxide. \Ye thank Professor Peter Miller for his helpful comments on the typescript. This research was supported by a Biomedical Research Grant to the University of Miami and AI-10527 to the University of California, Irvine from the National Institutes of Health. SUMMARY 1. Experiments were performed to examine the effects of temperatures down to --20° C on discontinuous respiration in diapausing Cecropia pupae. Methods were developed that permitted simultaneous recording of intratracheal pressure changes and spiracular valve movements at low temperatures. 2. Discontinuous respiration continued down to - 5° C but ceased at some temperature between -- 5 and - - 10° C. 3. At 0° C the constriction period of the cycle, during which valves remain closed for some time, generally was absent. The open phase of the burst was followed by a lengthy decline phase, during which valves close briefly from the fully open position, and then by the flutter period, during which valves con- tinuously open briefly and then close. The flutter period made up most of the cycle at all temperatures except at 5° C where the flutter and constriction periods were of about equal length. 4. Longer microcycles, relatively brief pressure fall periods caused by the closing of the spiracular valves, with correspondingly greater intratracheal vacuums than at high temperatures were observed in the decline phase of the burst and in the flutter period as the ambient temperature decreased to 0° C. At - 5° C 310 B. N. BURKETT AND H. A. SCHNEIDERMAN microcycles were shorter and resulted in less of an intratracheal vacuum than at 0° C, presumably because the metabolic rate of the pupa decreased. 5. The spiracular valves responded to oxygen and carbon dioxide at •- 5° C but not at - - 10° C or below. Thus neuromuscular coordination of spiracular func- tion persisted at - 5° C but ceased at some temperature between — 5° and - 10° C. 6. Studies of pupal weight loss indicated that spiracular valves remained con- stricted at low temperatures. Under the experimental conditions pupae froze at some temperature between --5° and - - 10° C. It is suggested that in nature the spiracular valves freeze in the closed position. LITERATURE CITED ASAHINA, E., 1966. Freezing and frost resistance in insects. Pages 451-484 in H. T. Meryman, Ed., Cryobiology. Academic Press, New York. BECKEL, W. E., 1958. The morphology, histology and physiology of the spiracular regulatory apparatus of Hyalophora cccropia (L.) (Lepidoptera). Proc. 10th Int. Coin/r. Enfomol., 2: 87-115. BROCKWAY, A. P., 1964. Some physical aspects of gas exchange in diapausing pupae of the Cecropia silkworm. Ph.D. tlicsis, JTestcm Reserve University. 150 pages. BROCKWAY, A. P., AND H. A. SCHNEIDERMAN, 1967. Strain-gauge transducer studies on intratracheal pressure and pupal length during discontinuous respiration in dia- pausing silkworm pupae. /. Insect Physiol.. 13: 1413-1451. BUCK, J. B., 1958. Cyclic CO» release in insects. IV. A theory of mechanism. Biol. Bull.. 114: 118-140. BUCK, J. B., AND M. KEISTER, 1955. Cyclic CO2 release in diapausing Ac/apcnia pupae. Biol. Bull., 109: 144-163. BUCK, J. B., AND M. KEISTER, 1958. Cyclic CO- release in diapausing pupae — II. Tracheal anatomy, volume and pCOj ; blood volume; interburst CO2 release rate. /. Insect Physiol., 1 : 327-340. BUCK, J. B., AND S. FRIEDMAN, 1958. Cyclic CO2 release in diapausing pupae — III. CO2 capacity of the blood : carbonic anhydrase. /. Insect Physiol., 2 : 52-60. BURKETT, B. N., AND H. A. SCHNEIDERMAN, 1967. Control of spiracles in silk moths by oxygen and carbon dioxide. Science, 156: 1604-1606. BURKETT, B. N., AND H. A. SCHNEIDERMAN, 1968. Co-ordinated neuromuscular activity in insect spiracles at sub-zero temperatures. Nature, 217 : 95-96. BURKETT, B. N., AND H. A. SCHNEIDERMAN, 1974. Roles of oxygen and carbon dioxide in the control of spiracular function in Cecropia pupae. Biol. Bull., 147: 274-293. KANWISHER, J. W., 1966. Tracheal gas dynamics in pupae of the Cecropia silkworm. Biol. Bull., 130 : 96-105. LEVY, R. I., AND H. A. SCHNEIDERMAN, 1966. Discontinuous respiration in insects — III. The effect of temperature and ambient oxygen tension on the gaseous composition of the tracheal system of silkworm pupae. /. Insect Physiol., 12: 105-121. RICHARDS, A. G., 1958. Temperature in relation to the activity of single and multiple physio- logical systems in insects. Proc. lOtli Int. Congr. Entoinol.. 2: 67-72. SCHNEIDERMAN, H. A., 1960. Discontinuous respiration in insects: role of the spiracles. Biol. Bull., 119: 494-528. SCHNEIDERMAN, H. A., AND A. N. SCHECHTER, 1966. Discontinuous respiration in insects — V. Pressure and volume changes in the tracheal system of silkworm pupae. /. Insect Physiol., 12: 1148-1170. SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1955. An experimental analysis of the dis- continuous respiration of the Cecropia silkworm. Biol. Bull., 109: 123-143. VAN DER KLOOT, W. G., 1963. The electrophysiology and the nervous control of the spiracular muscle of pupae of the giant silkmoths. Coinp. Biochetn. Physiol., 9: 317-333. WILLIAMS, C. M., 1946. Physiology of insect diapause. The role of the brain in the pro- duction and termination of pupal dormancy in the giant silkworm, Platysamia cccropia. Biol. Bull., 90 : 234-243. Reference: Biol Bull, 147: 311-320. (October, 1974) IDENTIFICATION AND CHARACTERIZATION OF LYSOZYME FROM THE HEMOLYMPH OF THE SOFT-SHELLED CLAM, MY A ARENARIA * THOMAS C. CHENG AND GARY E. RODRICK Institute for Pathobiology, Center for Health Sciences, Lehigh University, Bethlehem, Pennsylvania 18015 Reactions in pelecypod molluscs to experimentally or naturally introduced non- self materials, biotic or abiotic, is primarily cellular, i.e., if the foreign material is too large to be phagocytosed, it is encapsulated (see Feng, 1967; Cheng, 1967; and Cheng and Rifkin, 1970 for reviews). In the case of materials small enough to be phagocytosed, those which are digestible usually are degraded intracellularly (Tripp, 1958a, 1958b, 1960; Feng, 1959, 1965). 'That the vegetative cells of certain bacteria are reacted against in this manner has been documented by elec- tron microscopy (Cheng, Cali and Foley, 1974; Cheng and Cali, 1974). Con- sequently, it appeared to be of interest to determine whether lysozyme and other enzymes occur in the hemolymph of several species of pelecypods and if so, to ascertain their kinetic properties. Such studies directed at the identification and characterization of lysozyme have been carried out on the serum and leucocytes of the American oyster, Crassostrea I'iryinica, by Rodrick and Cheng (1974) and it has been shown that this bacteriolytic enzyme does indeed occur in both the serum and leucocytes of that mollusc. Thus, the earlier reports by McDade and Tripp (1967a, 1967b) have been confirmed and in addition, we have characterized the oyster lysozyme relative to its stability to heat, sensitivity to changes in ionic concentration, salt dependency, optimal pH range, and activity on selected species of bacteria. The second species of pelecypod that has been studied is the soft-shelled clam, Alya arenaria, and in this paper is reported our studies on the lysozyme of this mollusc. MATERIALS AND METHODS Collection of whole hemolymph The soft-shelled clams, M. arenaria, used in this study were collected from the vicinity of Sandy Hook, New Jersey, and were maintained for up to 40 days in recirculating seawater tanks with a salinity of 25%c. During this period, hemo- lymph samples were taken from the mantle cavity, heart, and pericardial sac by use of sterile Pasteur pipettes. Because all pelecypods possess an open cir- culatory system, all hemolymph samples taken from these sites are assumed to be identical relative to qualitative composition. 1 This research was supported by a grant (FD-00416-03) from the U. S. Public Health Service. 311 312 T. C. CHENG AND G. E. RODRICK Measurement of lysozyme activity Lysozyme (EC 3 -2- 1-17, N-acetylmuramide glycanohydrolase) activity in the hemolymph of M. arenaria was determined spectrophotometrically by a modi- fication of the method of Shugar (1952). The modification involved using 0.2 mg/ml of dried Micrococcus lysodeikticiis cells (Sigma, St. Louis, Missouri) and 0.1 M glycine buffer at pH 5.5. All reactions were initiated by the addition of hemolymph, i.e., enzyme, and measured by use of a Gilford 240 spectrophotometer equipped with a Model 6040-A heat writing recorder. Only the initial velocities were measured and in all cases the initial velocities doubled when the amount of enzyme (hemolymph) was doubled. The protein concentration of whole hemolymph was determined by the method of Lowry, Rosebrough, Farr and Randall (1951) and crystalline bovine serum albumin (Sigma) was used as the standard. Determination of pH optima In order to determine the pH optima in the presence of several buffers, assays for lysozyme activity were carried out at various pH's in the presence of saturating substrate levels of 0.2 mg/ml using the following four buffers: (1) 0.1 M gly- cylglycine at pH 4.0 to 9.0, (2) 0.1 M imidazole at pH 4.0 to 9.0, (3) 0.1 M fris-HCl at pH 4.0 to 9.0, and (4) 0.1 M phosphate at pH 4.0 to 9.0. The pH of the lysozyme reaction mixture was determined before, during, and after each enzyme reaction. Determination of end products The lysozyme reaction mixtures were analyzed qualitatively for the presence of amino sugars (Rondle and Morgan, 1955) and reducing sugars by use of Benedict's solution (McDade and Tripp, 1967b) before, during, and after enzyme reactions. Thermal stability of lysozyme To determine the effect of temperature on the enzyme, samples of fresh, whole hemolymph were incubated at temperatures from 10° C to 100° C at 10° intervals for 30 minutes before being assayed for lysozyme activity. Distribution of lysozyme Fresh, whole hemolymph was centrifuged at 4000 X g for 15 minutes and the resulting pellet was homogenized in a minimal amount of 0.25 M, 0.5 M, or 0.9 M sucrose, whole hemolymph, or sea water that had been passed through a 0.22 /mi millipore filter. The crude homogenate was subsequently recentrifuged for 10 minutes at 4000 X g. The supernatant was decanted off and recentrifuged at 10,000 X g for 30 minutes. This was done to pellet any organelles that had been released during homogenization. All of the supernatants and pellets were tested for lysozyme activity and the protein content of each constituent was determined. LYSOZYME OF MY A ARENARIA 313 Effect of lysozyme on bacteria To test the specificity of the lysozyme in M. arenaria hemolymph, seven species of bacteria in addition to Micrococcus lysodeitikus were tested against this enzyme. These bacteria were Bacillus megaterium, B. sub tills, Proteus vitlgaris, Salmonella pullorum, Shigclla sonnei, Escherichia coli, and Staphylococcus anreus. The pro- cedure followed is identical to that described for M. lysodeitikus, which involved determining alterations in optical density of the enzyme-bacteria mixtures. Reactivation of lysosyme b\> salts Freshly collected hemolymph from M. arenaria was dialyzed in 0.1 M gly- cylglycine buffer at pH 5.5 for 210 minutes after which the lysozyme activity was determined in the absence and presence of various concentrations of NaCl, KG, and MgClo. The sigmoid portions (first halves) of the salt reactivation curves were analyzed by use of a Hill plot to determine the minimal number of interacting binding sites for the salts during reactivation of the lysozyme. Using the Michaelis-Menten assumption of equilibrium kinetics, i.e., V (*) n V m V55' V0 = Km + (s)" where v0 is the initial velocity, s is the salt concentration, and n is the number of interacting binding sites, and by taking the logarithmic form of the equation given above and arranging the logarithmic expression for the equation to fit a straight line (y := mx + b), the following equation is obtained: log (v,,/Vra -- v0) == n log (s) -- log Km When the data are plotted using the coordinates log (v0/Vm -- v0) vs. log (s), the resulting Hill plot provides a n (= slope) equal to the minimum number of interacting binding sites (Atkinson, Hathaway and Smith, 1965). Effect of heavy metals on lysosyme In order to determine whether heavy metals inhibit the lysozyme from M. arenaria hemolymph, enzyme assays were carried out as described above but in the presence of 5 ^M of zinc acetate and 0.6 ^M of lead nitrate. In addition, since Smith and Stocker (1949) have reported that egg-white lysozyme is inhibited by sodium tartrate, we have tested the effect of 0.1 mM of sodium tartrate on the enzyme from the clam. RESULTS Our results indicate that the hemolymph of Mya arenaria includes lysozyme activity. This conclusion is based on the finding of (1) a reduction of turbidity measured at 540 m/j, when intact cell walls of several species of bacteria are placed in the hemolymph, and (2) the liberation of reducing and amine sugars by intact bacterial cell walls when placed in the hemolymph. 314 T. C. CHENG AND G. E. RODRICK FIGURE 1. Graph showing effect of pH on the activity of the lysozyme in \vhole hemo- lymph of Mya arcnaria using 0.1 M glycylglycine (black dots), 0.1 M Tris-HCl (circles), 0.1 M imidazole (squares), and 0.1 M phosphate (triangles) as buffers. The lysozyme activity is expressed as AOD54o/min X 10~2 at 25° C. The effect of pH on the lytic activity of whole hemolymph is shown in Figure 1. When 0.1 M glycylglycine, 0.1 M imidazole, or 0.1 M phosphate buffers are used, the optimal pH of the lysozyme has been determined to be 5.0. On the other hand, the optimal pH is 4.5 when a 0.1 M Tris-HCl buffer is employed. It is noted that a sharp peak is obtained at pH 5.0 and relatively high lysozyme activities have beeen recorded between pH 4.5 and 6.0 when 0.1 M glycylglycine and 0.1 M imidazole are used. However, low activities and a relatively broad and nonspecific pH effect is observed when 0.1 M Tris-HCl and 0.1 M phosphate buffers are employed. As indicated in Table I, the major portion of the lysozyme activity is asso- ciated with the 4000 X g supernatant while only a small amount of activity is associated with the 4000 X g pellet. Similarly, when the 4000 X g supernatant is recentrifuged at 10,000 X g, most, if not all. of the lysozyme activity is associated with the supernatant. TABLE I Distribution of lysozyme activity in the hemolymph of Mya arenaria Sample AOD640 min Specific activity Whole hemolymph 4,000 X g supernatant 4,000 X g pellet 10,000 X g supernatant 10,000 X g pellet 0.012 0.020 0.008 0.021 0.001 0.009 0.014 0.004 0.015 undctectable LYSOZYME OF MY A ARENARIA 315 TABLE II Lytic activity of lysozyme in whole hemolymph of Mya arenaria on several species of bacteria. The specific activities are reported at of protein at 25°C and pH 5.5 Bacteria AODwo/min Specific activity Micrococcus I \sodcitik us Bacillus megateriitni Bacillus subtil is Proteus vulgaris Salmonella pullorum Shigella sonnei Escherichia coli Staphylococcus a lire its 0.021 0.018 0.008 0.016 0.013 0.011 0.00') no activity 0.020 0.014 0.006 0.010 0.009 0.008 0.007 no activity Eight species of bacteria were tested against the lysozyme in M. arenaria hemolymph. As indicated in Table II, the enzyme is most active in the degrada- tion of Micrococcus lysodeitikus. Bacillus megaterium, and B. subtilis. There is a low level of activity against Proteus vnlgaris, Salmonella pullorutn, Shigella sonnei, and Escherichia coli and the lysozyme has no effect on Staphylococcus aureus. That the hemolymph lysozyme of M. arenaria is very sensitive to dialysis is demonstrated by the reduction of its activity by 50% after 30 minutes of dialysis and the total inactivation of the enzyme after 210 minues of dialysis in 0.1 M glycylglycine (Fig. 2). The addition of either NaCl or KC1 to hemolymph that had been inactivated by dialysis restores the lysozyme activity (Fig. 3). However, the addition of both MgCl? and FeCls has little effect on reactivation of the enzyme. It is noted that the highest lysozyme activity (AOD54o/min) is obtained with 100 mM of NaCl or KC1. Furthermore, the activation portion of the curve (first half) is sigmoidal while the deactivation portion (second half) is not sigmoidal. By plotting the data from the sigmodial activation portion as a Hill plot (Fig. 4), a straight line with a slope (n value) of 2.3 for NaCl and 2.8 for KC1 is obtained. These slopes (n values) are equal to the minimum number of interacting binding sites for NaCl and KC1. TABLE III The effects of sodium tartrate, zinc acetate, and lead nitrate on lysozyme activity in the hemolymph of Mya arenaria Sample AODs4o/min Specific activity % inhibition Fresh whole hemolymph 0.025 0.020 0 Fresh whole hemolymph +0.1 HIM sodium tartrate 0.008 0.006 68.0 Fresh whole hemolymph + 5 MM zinc acetate 0.011 0.008 56.0 Fresh whole hemolymph + 0.6 yuM lead nitrate 0.005 0.003 80.0 316 T. C. CHENG AND G. E. RODRICK 7i CN I O 5- .E 4H Q O 2- 1- 30 60 90 120 150 180 210 Dialysis time (min) FIGURE 2. Graph showing the effect of dialysis on the lytic activity of the lysozyme in the whole hemolymph of Mya arenaria using 0.1 M glycylglycine at pH 5.5. The lysozyme activity is expressed as AOD5io/min X 10~2 at 25° C. 7n 6- CM O — '5H X c E4H Ifi o Q 3 O 25 50 75 100 125 150 175 Salt cone. (mMxlO'3) 200 FIGURE 3. Graph showing the effect of various salt concentrations on the lytic activity of the lysozyme in the whole hemolymph of Mya arenaria using NaCl (triangles), KC1 ('black dots), and MgCl- (circles). The lysozyme activity is expressed as AOD3i0/min X 10-a at 25° C and pH 5.5. LYSOZYME OF MYA ARENARIA 317 10D-! 2 X 2:0- >° Eio > .9 .8 .7 .6 .5 .4 .3- .2- •NaCI n=2-3 >KCI rv2-8 .2 .3 .4 .5 .6.7.8.910 2JD Salt cone, i mM x 10"3) 10.0 FIGURE 4. Hill plot analysis of the effect of various concentrations of NaCl (triangles) and KCI (black dots) on the lytic activity of the lysozyme in the whole hemolymph of Mya areuaria. The slope (n) is equal to the minimum number of interacting binding sites. The lysozyme in the hemolymph of M. arenaria is extremely sensitive to small amounts of zinc acetate, lead nitrate, and sodium tartrate (Table III). Lead nitrate is approximately 10 times more effective than zinc acetate and considerably more effective than sodium tartrate in inhibiting lysozyme activity. DISCUSSION According to Jolles (1964), the following criteria must be met before an enzyme can be designated as lysozyme. Specifically, (1) it must catalyze the. release of re- ducing sugars from susceptible bacteria, (2) it must cause the liberation of amine sugars and or muramic acid from susceptible bacteria, and (3) it must cause the reduction in turbidity of intact bacterial cell walls. The enzyme that we have examined from the hemolymph of M. arenaria meets these criteria and therefore can be considered as lysozyme. The lysozyme from M. arenaria hemolymph shares several biochemical prop- erties with lysozymes from other sources. It is relatively heat stable, being able to withstand 75° C for 30 minutes; is salt dependent; and portrays sensitivity to alterations in salt concentration. Relative to the last characteristic, it is noted that in the absence or very low concentrations of salts, the M. arenaria lysozyme has essentially no effect on the cell wall of Micrococcus lysodcikticus as indicated by only minute amounts of reducing and amine sugars released and little, if any, decrease in turbidity of the bacterial reaction mixtures. Based on the 318 T. C. CHENG AND G. E. RODRICK available data, it would appear that the salt dependency of the lysozyme is specific for monovalent cations since both KC1 and NaCl are very effective in restoring enzyme activity while MgClo is rather ineffective. As reported, the activation of the molluscan lysozyme at low cation concen- trations is sigmoidal, thus indicating the possible occurrence of preferential inter- acting binding sites for NaCl and KC1. A Hill plot analysis of these data has revealed that a minimum of 2.0 binding sites occur and are required for optimal activity. On the other hand, the inhibitory segments of the two salt-dependency curves (Fig. 3) are not sigmoidal and hence probably do not involve multiple reaction sites. Therefore, it may be concluded that the activation and inhibition of lysozyme activity by NaCl or KC1 may represent two distinct kinetic processes. It is noted that Fitt, Dietz and Grunberg- Manage (1968), working with poly- nucleotide phosphorylase, have reported activation of this enzyme by low salt concentrations in a manner similar to that of the hemolymph lysozyme of M. arenaria. It is also noted that the hydrolysis of glycol chitin requires the presence of small amounts of salts (Rupley and Gates, 1967) and the interactions of lyso- zyme with chromatographic columns of neutral chitin are affected by changes in pH and ionic concentration of the medium (Davis, Heuberger and Wilson, 1969). Consequently, the overall conformation of lysozyme may require the presence of salts to affect lysis. Phillips (1967), based on x-ray crystallography data, has shown that the lysozyme molecule includes high concentrations of salts which may suggest a role for salts in the overall confirmation of the enzyme ; however, Praissman and Rupley (1968) have demonstrated that increases in ionic concentrations have only a minimal effect on the tritium-hydrogen exchange rates of lysozyme. This information could be interpreted to negate Phillips' hypothesis. Probably a more satisfactory explanation for decreased lytic activity of lysozyme in the absence of salts is that of Davis et al. (1969) who have postulated that the electrostatic bind- ing between the enzyme and the bacterial cell wall occurs in either a "productive" or "unproductive" fashion similar to that of oligomers of chitin (Rupley and Gates, 1967). It is noted that the levels of lysozyme activity are higher in both the 4000 X g and 10,000 X g supernatants than in the corresponding pellets. The latter is comprised primarily of cells and cellular constituents. Although evidence per- taining to molluscan hemolymph cells is yet unavailable, it is possible that this phenomenon reflects the release of lysozyme from cells into the serum as Wright and Malawista (1972) and Zurier, Hoff stein and Weissman (1973) have re- ported for mammalian leucocytes. If such is the case, then lysozyme, and pos- sibly other lysosomal enzymes, released into the serum in molluscs may also play a role in extracellular internal defense against invading bacteria. We wish to acknowledge Ms. Sherry Koehler for her technical assistance. SUMMARY Lysozyme activity has been demonstrated in the hemolymph of the soft-shelled clam, Mya arenaria. When whole hemolymph is centrifuged at 4000 and 10,000 X LYSOZYME OF MY A ARENARIA 319 g and each constituent is assayed, lysozyme activity is found to be greater in the two supernatants than in the corresponding pellets. The lysozyme from M. arenaria hemolymph is salt dependent, relatively heat stabile, very sensitive to alterations in ionic concentration and the presence of heavy metals, and has an optimal pH of 5.0 when 0.1 M glycylglycine, 0.1 M imidazole, or 0.1 M phosphate buffers are employed but an optimal pH of 4.5 when 0.1 M Tris-HCl buffer is used. A Hill plot of the data resulting from salt reactivation studies indicates that the lysozyme in M. arenaria hemolymph includes at least 2.0 interacting binding sites for NaCl and KC1. When tested against a number of bacteria, the lysozyme is most active against Micrococcus lysodcitikus and Bacillus megaterium. It is less active against Proteus vulgaris, Salmonella pullomm, Shlgella sonnei, Bacillus snbtilis, and Escherichia coll. It is not active against Staphyloccus aweus. It is suggested that the lysozyme in serum may be released from hemolymph cells. LITERATURE CITED ATKINSON, D. E., J. A. HATHAWAY AND E. C. SMITH, 1965. Kinetic order of the yeast diphosphopyridine nucleotide isocitrate dehydrogenase reaction and a model for the reaction. /. Biol. Chcm., 240 : 2682-2690. CHENG, T. C., 1967. Marine molluscs as hosts for symbioses. Ad-ran. Mar. Biol., 5 : 1-424. CHENG, T. C., AND A. CALI, 1974. An electron microscope study of the fate of bacteria phagocytized by granulocytes of Crassosti-ea rin/inica. Contcmp, Top. Immunobiol., 4: 25-35. CHENG, T. C., A. CALI AND D. A. FOLEY, 1974. Cellular reactions in marine pelecypods as a factor influencing endosymbioses. Pages 61-91 in W. B. Vernberg, Ed., Sym- biosis in the Sea. University of South Carolina Press, Columbia, South Carolina. CHENG, T. C., AND E. RIFKIN, 1970. Cellular reactions in marine molluscs in response to helminth parasitism. Pages 443-496 in S. F. Shieszko, Ed., A Symposium on Diseases of Fishes and Shellfishes. American Fisheries Society, Spec. Publ. No. 5, Washington, D. C. DAVIS, R. C., A HEUBERGER AND B. M. WILSON, 1969. The dependence of lysozyme activity on pH and ionic strength. Bwchiin. Biof>hys. ^Icta. 178: 294-305. FENG, S. Y., 1959. Defense mechanism of the oyster, Bull. X. J. Acad. Sci., 4: 17. FENG, S. Y., 1965. Pinocytosis of proteins by oyster leucocytes. Biol. Bull., 128: 95-105. FENG, S. Y., 1967. Responses of molluscs to foreign bodies, with special reference to the oyster. Fed. Proc., 26 : 1685-1692. FITT, P. S., F. W. DIETZ, JR. AND M. GRUNBERG-MANAGO, 1968. Activation by electrolytes and polylysine of poly (A) synthesis by Clostridium perfringens polynucleotide phos- phorylase. BiocJiim. Biopliys. Acta, 151 : 99-113. JOLLES, P., 1964. Recent developments in the study of lysozymes. ^Jngrcw. Client., 3 : 28-36. LOWRY, O. H., H. J. ROSEBROUGH, A. L. FARR AND R. I. RANDALL, 1951. Protein measure- ment with Folin phenol reagent. /. Biol. Chcm., 193 : 265-275. McDADE, J. E., AND M. R. TRIPP, 1967a. Lysozyme in the hemolymph of the oyster, Crassostrea I'irc/inica. J. Invert. Pathol., 9: 531-535. McDADE, J. E., AND M. R. TRIPP, 1967b. Lysozyme in oyster mantle mucus. /. Invert. Pathol.. 9: 581-582. PHILLIPS, D. C., 1967. The hen egg-white lysozyme molecule. Proc. Nat. Acad. Sci. U. S., 57: 484-495. PRAISSMAN, M., AND J. A. RUPLEY, 1968. Comparison of protein structure in crystal and in solution. III. Tritium-hydrogen exchange of lysozyme and a lysozyme-saccharide complex. Biochemistry, 7 : 2446-2450. 320 T. C. CHENG AND G. E. RODRICK RONDLE, C. J. M., AND W. T. J. MORGAN, 1955. Determination of glucosamide and galactosamine. Biochem. J., 61 : 586-589. RODRICK, G. E., AND T. C. CHENG, 1974. Kinetic properties of lysozyme from the hemolymph of Crassostrea z'irginica. J. Invert. Pathol., 24 : 41-48. RUPLEY, J. A., AND V. GATES, 1967. Studies on the enzymic activity of lysozyme. II. The hydrolysis and transfer reactions of N-acetylglucosamine oligosaccharides. Proc. Nat. Acad.Sci. U.S., 57: 496-510. SHUGAR, D., 1952. Measurement of lysozyme activity and the ultraviolet inactivation of lysozyme. Biochim. Biophys. Acta, 8: 302-308. SMITH, G. N., AND C. STOCKER, 1949. Inhibition of crystalline lysozyme. Arch. Biochem., 21: 383-394. TRIPP, M. R., 1958a. Disposal by the oyster of intracardially injected red blood cells of vertebrates. Proc. Nat. Shellfish. Ass.. 48 : 143-147. TRIPP, M. R., 1958b. Studies on the defense mechanism of the oyster. /. Parasitol., 44 (Sect. 2) : 35-36. TRIPP, M. R., 1960. Mechanisms of removal of injected microorganisms from the American oyster, Crassostrea virginica (Gmelin). Biol. Bull., 119: 210-223. WRIGHT, D. G., AND S. E. MALAWISTA, 1972. The mobilization and extracellular release of granular enzymes from human leukocytes during phagocytosis. /. Cell. Biol., 53 : 788-797. ZURIKR, R. B., S. HOFFSTEIN AND G. WEissxiAN, 1973. Mechanisms of lysosomal enzyme release from human leukocvtes. /. Cell Biol., 58: 27-41. Reference: Biol. Bull., 147: 321-332. (October, 1974) LARVAL DEVELOPMENT OF THE GIANT SCALLOP PLACOPECTEN MAGELLANICUS (GMELIN) * JOHN L. CULLINEY Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 The giant sea-scallop, Placopecten magellanicus (Gmelin) ranges along the east coast of North America from Labrador to Cape Hatteras (Abbott, 1954). Although primarily a continental shelf species, it may be found one meter below low tide in the Gulf of Maine (Read, 1967). At the southern end of its range, off North Carolina, P. magellanicus generally occurs in water over 150 feet (46 meters) deep (Porter, 1974). A regional fishery for P. magellanicus, chiefly off New England and eastern Canada, has stimulated some research on the biology of this scallop including studies of length-weight relationships and gonad development (Haynes, 1966) ; thermal tolerances and acclimation (Dickie, 1958) ; and growth rates in different geographical regions (Stevenson and Dickie, 1954 and Merrill, Posgay and Nichy, 1966). Up to the present, however, research on the giant scallop has been confined to the adult stage. The dearth of information concerning larval biology of P. magellanicus has formed a conspicuous gap in knowledge of the life history of this species. The purpose of this paper is to describe the complete larval development of P. magellanicus, including observations on larval salinity tolerance and settling behavior. Information derived from rearing larvae in the laboratory may help to confirm the identification of P. magellanicus larvae from the plankton and will aid future mariculture efforts with this valuable food species. MATERIALS AND METHODS Sexually mature scallops were collected by SCUBA diving at Isles of Shoals, New Hampshire on September 21, 1973. They were placed in sea water in a styrofoam chest and held for 8 hours at approximately 5° C until returned to the laboratory. I selected small adults ( 3"-5" diameter) to minimize stresses to the animals from confinement in limited volumes of water both during transport and in subsequent spawning containers in the laboratory. Males and females were separated for spawning, which was stimulated therm- ally by raising the temperature 3-5° C. Gametes and developing larvae were generally handled as recommended by Loosanoff and Davis (1963) and Culliney, Boyle and Turner, ( 1974) . Larvae were reared in cylindrical glass jars containing two liters of sea water filtered to remove particles larger than one micron in size. The sea water was changed every two days. Salinity during development was nearly constant at 1 This research was supported by ONR Contract Number NOOO 14-67A-0298-0027 with Harvard University. 321 JOHN L. CULLIXEY 32%c, and was measured with hydrometers. Dilutions for the experiment on salin- ity tolerance were made using filtered pond water. The naked flagellate, Isochrysis galbana Parke was used to feed larvae up to the settling stage. Spat have heen reared using a combination of /. galbana and Chroomonas salina Taylor. Concentrations of algal food ranged from approxi- mately 50,000 cells/ml in cultures containing the youngest veligers to 5,000 cells/ml as larvae approached the pediveliger stage. Young spat, however, were fed con- centrations approximating 10,000 cells/ml. Measurements of larvae were made using a calibrated ocular micrometer. Clean shells of pediveliger larvae were prepared by feeding live larvae to small sea anemones. After the empty shells were egested by the anemones, right and left valves were teased apart so that the hinge structure could be observed and measure- ments made of individual valves. Antibiotics were not used until larvae reached the pediveliger stage. Then 50 ppm Sulmet (sodium sulfamethazine) was added to all cultures to arrest an apparent necrotic disease. RESULTS Spawning One diver in our party observed spawning in the undisturbed population from which adults used in this study were taken ; he reported seeing a scallop suddenly emit a puff of milky material. On the day of collection, September 21, 1973, the temperature of the bottom water (about 50 feet deep) was 14° C; that of the surface was 16° C. Many captured scallops of various sizes (2"-8" diameter) spawned profusely in buckets of sea water within minutes after they were brought to the surface. In the laboratory, some individuals spawned readily within minutes after the temperature was raised from 5° C to 10° C. Others were held in large bowls of aerated filtered sea water at 12° C. A number of these spawned two days later when the water temperature was raised to 15° C. In all, two minor and two major spawnings were cultured from separate groups of individuals. The minor spawnings were not reared past the early veliger stage due to low numbers and deformities of the larvae. Each culture was produced by a different female and several males. Cultures from different spawnings were not mixed ; several populations of a convenient two-liter size were set up from each of the two major spawnings. Development of larvae The main features of development are summarized in Table I. Development to the earliest swimming stage occurred at 12° C in all populations and lasted between 30 and 40 hours. The embryos first became motile as ciliated gastrulae averaging 69 microns long and 63 microns across the greatest diameter. They appeared to elongate slightly into a trochophore-like stage, then became typical shelled straight-hinge veligers (Fig. 1) on the fourth day after spawning. Young straight-hinge larvae possessed a very short apical flagellum. The mean size of the earliest straight-hinge larva was 105 microns long by 82 microns high. The LARVAL STAGES OF SCALLOP 323 hinge line averaged 81 microns. (In the following description, shell dimensions will be given as length X height in consistent order.) Development from the ciliated gastrula to the straight-hinge stage occurred over the temperature range 12° C to 18° C. Such a thermal gradient, arranged vertically by setting culture jars in a shallow cold-water bath, allows the motile larvae to seek their optimum temperature during the thermally sensitive period of straight-hinge shell deposition (Culliney et al., 1974). Initially several populations of veligers were reared at two temperatures, 15° C and 19° C, but populations at 19° C did not complete development. Development was rapid with little apparent difference in growth rates at the two temperatures. TABLE I. Major features of larval development of Placopecten magellanicus at 15° C, 32%c • ,i • or distinctive feature Age Mean size (M-W Range (M-U) S.E. UM) X* Remarks Unfertili/ed egg 0 <>4 63-65 17 Color pink to brown Temp. 12° C. Earliest swimming stage (ga--lrula) 30-40 hours length: (,<) f.S 70 0.3.S 6 diain.: <>3 62-64 0.24 13 Temp. 12-18 °C. length: 105 99-107 0.24 27 Earliest veliger ( straight-hinge i 4 day- height : X2 77 84 0.24 27 hinge-line : XI 75-82 0.24 27 Temp. 15° C. "Los< of apical flagellum X days \ "ill 1 H > \ eligrl 13-2X days lengt h : I7.S .".1 61 Temp. 15° C. height : 155-232 61 Appearance of eyeshot 23 days length: >2<0 Bottom seeking behavior height : >210 I >evelopment of foot Adhesive tendency Pediveliget (sporadic crawling, w.il. 28 days length: 27') 206 292 0.94 46 Temp. 15° C. byssu:-, growth of gill rudiment) height : 242 23X 2.S4 L.64 46 depth : 1 27 1 111 142 1.40 20 Spat (extensive crawling, strong >35 days IPV--SUS. growth of dissoconch ^hell and gill) * N refers to the number of individuals measured. The tiny apical flagellum disappeared ; it could not be detected by the eighth day. The hinge line remained visible and of nearly constant length until about the 13th day when larvae began to exceed 175 X 150 microns in size (Fig. 1). Beginning on the 15th day, extensive mortalities were observed in populations at 19° C. By the 19th day, cultures at this temperature were nearly completely decimated. A fungus appeared associated with the mortality at 19° C, but was not seen attacking living larvae. Throughout this period, populations of larvae at 15° C appeared normal and showed almost no mortality. During larval development, shell length remained 25—30 microns greater than height. A regression of length versus height, calculated from over 100 measure- ments of veligers in all stages of development, illustrates the quantitative relation- ship between the two dimensions : height = 0.96 length - • 18.2. The extended velum in swimming larvae had a unique "keyhole" outline. This was most pronounced in half-grown veligers. The outline of the velum tended to become nearly oval as larvae approached the pediveliger stage (Fig. 2a). 324 JOHN L. CULLINEY X 105x82 157x131 210x185 256x230 273x246 FIGURE 1. Individual and group photographs of Placopectcn magellanicus larvae from the straight-hinge stage to the pediveliger (sizes in microns). The more pointed end, to the left in photos of individual larvae, is the anterior end. LARVAL STAGES OF SCALLOP 325 The shape of the larval shell, standing on edge (Fig. 2b) revealed a slightly longer taper in the anterior than in the posterior direction. These features may have some taxonomic value. On the 23rd day, typical bivalve veliger eyespots, such as occur in ostreid and mytilid larvae, were first noticed in larvae larger than 230 X 200 microns (Fig. 2c). By this time the larval foot had become prominent, although it was not yet observed extending outside the shell. In addition many larvae showed a change in swimming behavior. Prior to the 23rd day they were well dispersed in the water column, althought most swam in the upper third of the container. After the 23rd day increasing numbers of larvae were observed massed in dense swarms, swimming just above the bottom of the culture jar. As larvae approached the pediveliger stage, they often displayed the foot as they swam. The most conspicuous feature of the foot was a cluster of long active cilia at its tip (Fig. 2d) ; also a well-formed heel or byssal spur was present. About this time, the larvae began to show an adhesive tendency, causing them to TABLE II. Dimensions of left and right valves of Placopecten magellanicus pediveligers Left valve Right valve Mean (nM) Range (n\I) S.E. (M.U) N* Mean (M-!/) Range (nM) S.E. (MM) N* Length 280 264-292 1.64 21 276 257-290 1.76 20 Height 254 232-268 1.88 21 250 230-264 1.76 20 Depth 67 50-76 1.38 21 60 48-70 1.40 20 * N refers to the number of individual values measured. stick to each other and to dust particles, pseudofeces, and other debris. By the 28th day after spawning more than 50% of the larvae in all populations possessed the functional foot characteristic of the pediveliger stage. Pediveligers crawled readily, but for only short distances on glass slides and on the glass bottoms of culture containers. Crawling seemed to be initiated after larvae were disturbed. For example, brief episodes of crawling occurred when larvae were discharged from a pipette, or after they were sieved and placed in new culture containers. Pediveligers secreted a weak byssus which could not be seen readily, even when magnified at 100 X. A prominent gill rudiment was present in most pediveligers (Fig. 2c). Average measurements of pediveliger larvae were 279 X 242 microns with a depth of 127 microns. The hinge structure of P. magellanicus larvae was almost featureless. A series of minute taxodont teeth were present, but their actual number was not determined and good photographs of the hinge were not obtained. Pediveliger shells were difficult to separate, indicating the presence of a strong binding force, possibly a ligament, in the hinge area. Many shells cracked or fractured in the attempt to disarticulate them. A number of shells were separated into intact valve pairs, however, and measurements of these show that the left valve averages slightly larger than the right in length, height, and depth (Table II). The difference in depth between the two valves is statistically significant at the 95% level. 326 JOHN L. CULLINEY B LARVAL STAGES OF SCALLOP 327 Effects of lozv salinity A mixed population of umbo-stage larvae, originating from the two female parents, was tested for salinity tolerance over a 42 hour period. Twenty larvae were pipetted into each of five finger bowls containing 180 ml of water at salinities of 10.5, 16.9, 21.5, 26.2, and 30.0%0. (Salinity was measured after the experi- ment). The temperature was held at 15° C. No food was given during the experiment. The LD50 level for salinity could not be estimated for the conditions stated, since all the larvae survived the 42 hour experiment. However, the be- havior of the larvae was recorded. At W.S%C, larvae remained on the bottom of the finger bowl. After an initial shock produced immobility for about two hours, most larvae remained moving during the experiment. However, normal swimming was not observed. Most larvae lay on one valve and rotated in place ; they appeared incapable of retracting the velum which had a distended appearance. The color of the larvae remained normal. At 22 hours, six of the twenty larvae showed no movement when the distended velum was touched by a fine needle. But at 42 hours all larvae were moving vigorously. At 16.9%o, an initial shock produced immobility for about two hours. However, many individuals recovered and exhibited normal swimming and translational movement, but they remained close to, often touching, the bottom. At 2l.S%c, an initial shock was evident, but most larvae regained full mobility within two hours. Most larvae remained swimming close to the bottom. One individual was seen well above the bottom, however, 15 minutes after the beginning of the experiment, and two individuals swam near the surface at 42 hours. At 26.2 and 30.0/ro, no alteration of larval behavior was noted. Larvae swam throughout the water column, but most remained near the surface, throughout the experiment. Settling behavior The first spat was found 35 days after spawning. A conspicuous feature of the spat was the presence of a strong, thick, plainly visible byssus (Fig. 2e), contrasting with the "invisible" threads produced by pediveligers. Crawling by spat was vigorous and extensive, and was observed predominantly after the byssus was broken. At least some settling individuals cast off the velum in large pieces. On several occasions a nearly complete detached velum was seen. P. inagellanicus, however, appeared to settle in culture containers only re- luctantly. As pediveliger larvae accumulated and very few spat were found, it appeared that a delay of metamorphosis was occurring. Presumably this was due to the lack of an attractive substrate for settlement. Two experiments were set up to examine potential stimuli to settlement. These included presence of adult P. niagellaiiicns shell material and general thigmotatic effects of three- dimensional surfaces. The results of these experiments are shown in Figure 3. FIGURE 2. Special features of larval development of Placopecten magellanicus; (A) out- line of the velum of swimming larvae from early umbo stage to the pediveliger (X 120) ; (B) comparison of horizontal and vertical shell profiles in late umbo stage larvae (X 180) ; (C) gill rudiment and eyespot in pediveliger (X 140) ; (D) tip of pediveliger's foot, showing tuft of long cilia (X 580) ; (E) spat with well-formed byssus (X 110). 328 JOHN L. CULLINEY The experiments utilized pediveliger larvae in 4-inch finger bowls containing 200 ml of filtered sea water. Twenty larvae were exposed to each settling stimulus. Controls were simply held in finger bowls without an added stimulus. The temperature was 13°-15° C. Larvae were fed 5,000 cells/ml Isochrysis. Experimental and control populations were examined every 3-4 days, and fully metamorphosed and dead individuals were removed. Criteria for metamorphosis were attachment by a strong byssus and complete absence of the velum. Indi- viduals showing only one of these features were considered partially metamorphosed (not shown in Figure 3). 20T CO 6 z 20- 10- B SHELL - PEBBLES GLASS CONTROL 10 DAYS 20 FIGURE 3. Cumulative plots of numbers of Placopcctcn inagcllanicits settling in response to different stimuli. Experiment B followed the completion of A. The single vertical bar in each plot represents metamorphosed individuals in the control populations. Experiment A examined the reactions of larvae exposed to six fragments of fresh adult P. magellanicus shell. The fragments ranged from 3 mm to 2 cm long and were scattered on the bottom of the finger bowl. The results (Figure 3) show a strong settling response to the fragments. Thirteen larvae metamorphosed ; two were partially metamorphosed at the end of the experiment, and three were dead. Settled larvae nearly always attached directly to the shell fragments and predominantly tended to metamorphose on the undersides of the fragments. Of the thirteen fully metamorphosed individuals, all but two were attached under- neath the shell pieces, and only one did not attach to the shell. In the control, two larvae metamorphosed and one died. An additional set of larvae was exposed at this time to increased food concen- tration (25,000 cells/ml Isochrysis. None of these larvae metamorphosed, indicat- ing that increased food concentration is not an important settling stimulus. This result is omitted from Figure 3. Experiment B attempted to separate properties of Placopccten shell as stimuli from a more general thigmotactic effect. The response to scallop shell fragments was compared to that induced by small pebbles and glass fragments. All of these objects were visibly clean but not sterile. Numbers of larvae and background LARVAL STAGES OF SCALLOP 329 conditions were the same as in Experiment A. Experiment B followed the end of Experiment A. The results (Fig. 3) demonstrate that a strong stimulus to settling and meta- morphosis is provided by all classes of physical objects tested. In the case of shell fragments and pebbles, the results were nearly identical to those in Experi- ment A. There is, however, a slight indication that Experiment B larvae, two weeks older than those in Experiment A, were prone to settle more quickly on the stimulus objects. Mortality in the presence of shell fragments and pebbles was low and comparable to that in Experiment A. The population of larvae exposed to glass fragments, pieces of a microscope slide approximately the size of the shell fragments, suffered an unusually high mortality (nine dead). However, metamorphosed individuals accumulated on the glass fragmens at a rate that paralleled the results with shell and pebbles. Again in Experiment B, most metamorphosed individuals were found on the undersides of the objects tested. In the control population only two larvae settled. DISCUSSION The successful rearing of Placopecten iinit/ellanicus may reflect the fact that adults were obtained in prime condition for spawning. One previous attempt during the winter, to condition adults to develop gonads and spawn failed. This large active scallop may need more food than normally can be supplied under laboratory conditions. Comely (1972), working with the large European scallop, Pecten ma.vimus, also experienced failure in conditioning experiments. Previous attempts by other investigators to culture P. magellanicus larvae have also been unsuccessful. (P. Chanley, Shelter Island Oyster Co.; H. Hiclu, Univ. of Maine, personal communication). A number of purported difficulties at the time of spawning have been en- countered by investigators rearing other species of scallops. These problems have included poor vitality of embryos resulting from self-fertilization in Aequipecten irradians (Sastry, 1965) and Pecten maximus (Comely, 1972) ; abnormal de- velopment associated with polyspermy or supernumerary sperms (Comely, 1972 and Gruffycld and Beaumont, 1972) ; and mechanical damage to extremely fragile eggs by contact with a nylon sieve (Comely, 1972). Fortunately, none of these problems occurred during the rearing of P. magellanicus. Abnormal larvae did appear in the two minor spawnings whose volume of eggs was 1/10 or less than that of the major spawnings. Most of the larvae from these spawnings were conspicuously smaller than larvae from major spawn- ings (about 10 microns in length and height at the early straight-hinge stage). However, some deformities in shell formation were also observed. It seems likely these weak spawnings consisted of unripe or underdevelopd eggs. The mass mortality suffered by larval populations at 19° C may indicate that this temperature is close to the upper limit of thermal tolerance for larvae. Dickie (1958) found that lethal temperatures for adult P. magellanicus ranged from 21° C to 23.5° C, depending on acclimation. The true cause of death of the half- grown larvae at 19° C is unknown. LoosanofT and Davis (1963) observed that bivalve larvae reared at relatively high temperatures were more susceptible to 330 JOHN L. CULLINEY mortality from disease. Gruffydd and Beaumont (1972) found that temperatures of 18° C and higher favored bacterial and ciliate growth that decimated their Pccten maxinms larval cultures well before metamorphosis. The fungus observed attacking dead P. niayellanicns larvae in my cultures at 1Q° C may have started growing in living larvae but this is uncertain. The extremely fragile nature of larval shells of Pcctcn ina.vinnis reported by Comely (1972) was not apparent in P. mac/ellaniciis. Larvae were washed routinely in nylon sieves with no detectable damage to their shells from the earliest straight-hinge stage onward. One problem, common to bivalve larvae, noted in P. magellanicus, was a tendency of late stage larvae to stick to each other and to debris in the cultures. This tendency appeared suddenly, at the time larvae first descended to swim in large concentrations near the bottom. The stickiness may be associated with the first attempts at byssus formation by the larvae. This interpretation is also sug- gested by Gruffydd and Beaumont (1972), who describe the same phenomenon in Pecten nia.i-iinus. The problem can be controlled by drastically reducing the food allotment of the larvae when the eyespot is first detected, that is just before the appearance of bottom seeking behavior. The food-cell concentration should approximate 5 X 10:i to 1 X 104 cells/cc day or less, depending on larval popula- tion density. This eliminates much of the bottom debris caused by mass wastage of excess food. Bacterial and protozoan growth is minimized. Larvae still feed efficiently with the relatively large velum at this stage and remain largely free of detritus. Newly- settled spat which climb up off the bottom should be alloted more food than pediveligers, as the tiny gill at first appears to be an inefficient feeding mechanism. The differences in size of the right and left valves of pediveligers may have taxonomic importance. This observation, especially with respect to depth, seems to show a predisposition in larvae to the adult condition where the right valve is more nearly flat. The observations of behavior at different salinities show that larvae might sur- vive considerable incursion into estuaries. Also, a general lowering of salinities in coastal waters following heavy rains would not seriously affect them. Because of their bottom-seeking behavior at salinities near 20c/co and lower, larvae entering stratified estuaries might be transported some distance inland. It is also possible this behavior might indirectly cause mortalities by subjecting larvae to benthic predation or to entrapment in bottom debris before they are ready to settle. Pediveligers appeared capable of delaying metamorphosis for at least a month. In contrast to larvae of Mytihis ednlis, studied by Bayne (1965) Placopectcn pediveligers did not gradually lose the velum during the period of delay of metamorphosis. Unlike 1(1. cdulis, the scallop larvae retained their swimming ability. The experiments on settling behavior suggest a generalized thigmotactic re- sponse in pediveligers. This differs from the major trend of results in experiments on larval settlement in which highly specific biochemical or biophysical stimuli have been observed (Scheltema, 1961 ; Crisp, 1965, 1967; Wilson, 1968). Because conditions in my experiments were not sterile, there remains a possibility that larvae responded to microbial films associated with the stimulus objects. How- ever, such stimuli should also have been present in the sea water and on the LARVAL STAGES OF SCALLOP 331 glass surfaces of finger bowls. The extremely high mortality associated with fragments of glass microscope slides may have been caused by some toxic coating on the slides, which were fresh from a new box. Attachment of spat on the undersides of the stimulus objects may be a behavioral adaptation to escape certain types of epibenthic predators, for example crabs, which are known to take a heavy toll of bivalve spat (P. Chanley, Shelter Island Oyster Co., personal communication). A number of factors encountered in culturing larvae of P. niagcllanicits indicate that this would be a prime species for commercial mariculture. First the fact that the sexes are separate eliminates the problem of poor viability accompanying self-fertilization (Sastry, 1965; Comely, 1972). Larvae grew rapidly with little mortality at 15° C and thrived on a diet of the easily cultured Isochrysis galbama. Extreme precautions involving sterility of rearing equipment (Comely, 1972; Gruffydd and Beaumont, 1972) were not necessary with P. uiagellanicns, although an antibiotic such as Sulmet, used at 50 to 100 ppm would be recommended as larvae approach the pediveliger stage and aggregate near the bottom of the culture container. A reduction in food concentration would also be beneficial at this time. The natural tendency of pediveligers to settle beneath solid objects can be exploited. Spat are thus easily caught, manipulated, and transported on artificial substrates. Presently, the most serious obstacle to hatchery production of P. mayellanicus spat is that gametes cannot be obtained out of season. The potential for mari- culture of these scallops is also a function of growth of juveniles and adults under artificial and captive conditions. New approaches and techniques will be needed. At the present time, juvenile scallops, derived from my larval populations, are growing rapidly in a flowing laboratory sea water system at Woods Hole. I am grateful to L. Harris of the University of New Hampshire for providing the opportunity to collect scallops at Isles of Shoals. Thanks are due R. Turner, R. Scheltema, P. Chanley and P. Boyle for critically reading the manuscript. P. Boyle also helped to prepare the illustrations. SUMMARY (1) Sexually mature Placopecten magellanicus from Isles of Shoals, New Hampshire, were observed to spawn in nature at 14°— 16° C. In the laboratory, spawning occurred from 10°-15° C. (2) Average sizes of developmental stages were: eggs, 64 microns diameter; swimming gastrula, 69 microns long by 63 microns diameter ; earliest straight- hinge veliger, 105 X 82 microns, with a hinge line of 81 microns. The umbo stage began in larvae exceeding 175 X 155 microns. Pediveligers averaged 279 X 242 microns with a depth of 127 microns, and were inequivalved, the left valve being larger. (3) Development from the zygote to the swimming gastrula took 30-40 hours at 12° C. The straight-hinge veliger stage was reached in four days at tempera- tures between 12° and 18° C. 332 JOHN L. CULLINEY (4) Veligers reared at 15° C reached the pediveliger stage in 28 days, and the first spat was observed on the 35th day. Veligers reared at 19° C suffered a mass mortality when approximately half grown. (5) Larvae remained viable at salinities as low as 10.5^c>, and exhibited normal swimming from \6.9%c to 30.0^ c in a 42 hour test. (6) Larvae showed a thigmotactic settling response to shell fragments, small pebbles, and glass fragments. Predominant settling was on the undersides of these objects. Pediveligers appeared to delay metamorphosis until suitable physical substrates for settlement were encountered. LITERATURE CITED ABBOTT, R. T., 1954. American Seashells. D. Van Xostrand Co., Princeton, New Jersey. 541 pp. BAYNE, B. L., 1965. Growth and the delay of metamorphosis of larvae of Mytilus edulis (L.). Ophelia, 2 : 1-47. COMELY, C. A., 1972. Larval culture of the scallop, Pec ten ma.rimus (L.). /. Cons. Int. Explor. Mer,. 34 : 365-378. CRISP, D. J., 1965. Surface chemistry, a factor in the settlement of marine invertebrate larvae. Pages 51-65 in Botanica Gothoburgensia III: Proceedings of the Fifth Marine Bio- logical Symposium, Goteborg. CRISP, D. J., 1967. Chemical factors inducing settlement in Crassostrca viryinica (Gmelin). /. Anim. Ecol.. 36: 329-335. CULLINEY, J. L., P. J. BOYLE AND R. D. TURNER, 1974. New approaches and techniques for studying bivalve larvae. In press, W. Smith and M. Chanley, Eds., Culture of Marine Invertebrate Animals. Plenum Publishing Co., New York. DICKIE, L. M., 1958. Effects of high temperature on survival of the giant scallop. /. Fish. Res. Board Can.. 15(6) : 1189-1211. GRUFFYDD, LL. D., AND A. R. BEAUMONT, 1972. A method for rearing Pecten ma.rimus larvae in the laboratory. Mar. Biol. 15 : 350-355. HAYNES, E. B., 1966. Length-weight relation of the sea scallop, Placopecten magellanicus (Gmelin). International Commission of Northwest Atlantic Fisheries Res. Bull., No.. 3: 1-17. LOOSANOFF, V. L., AND H. C. DAVIS, 1963. Rearing of bivalve mollusks. Pages 1-136 in F. S. Russell, Ed., Advances in Marine Biology, Vol. 1. Academic Press, London. MERRILL, A. S., J. A. POSGAY AND F. E. NICHY, 1966. Annual marks on shell and ligament of the sea scallop, Placopecten magellanicus. U.S. Fish U'ildl. Serv. Fish. Bull., 65: 299-311. PORTER, H. J., 1974. The North Carolina Marine and Estuarine Mollusca, an Atlas of Occurrence. University of North Carolina, Institute of Marine Sciences, 351 p. READ, K. R. H., 1967. Thermal tolerance of the bivalve mollusc, Lima scabra Born, in rela- tion to environmental temperature. Proc. Malacol. Soc. London, 37: 233-241. SASTRY, A. N., 1965. The development and external morphology of pelagic larval and post- larval stages of the bay scallop, Aequipecten irradians concentricus Say, reared in the laboratory. Bull. Mar. Sci., 15(2) : 417-435. SCHELTEMA, R. S., 1961. Metamorphosis of the veliger larvae of Nassarius obsoletus (Gastro- poda) in response to bottom sediment. Biol. Bull. 120: 92-109. STEVENSON, J. A., AND L. M. DICKIE, 1954. Annual growth rings and rate of growth of the giant scallop, Placopecten magellanicus (Gmelin) in the Digby area of the Bay of Fundy. /. Fish. Res. Board Can., 11 : 660-671. WILSON, D. P., 1968. The settlement behavior of the larvae of Sabellaria alveolata. J. Mar. Biol. Ass. U.K., 48 : 387-435. Reference: Biol. Bull, 147: 333-351. (October, 1974) DIAPAUSE IN THE GEMMULES OF THE MARINE SPONGE. HALICLONA LOOSANOFFI, WITH A NOTE ON THE GEMMULES OF HALICLONA OCULATA 1 PAUL E. FELL Department of Zoology, Connecticut College, New Lodon, Connecticut 06320 Haliclona loosanoffi is one of several marine sponges that produce special struc- tures called gemmules (Hartman. 1958; Wells, Wells and Gray, 1964; Fell, 1974; and Simpson and Fell, 1974). Each gemmule consists of a mass of large granular cells enclosed within a collagenous capsule which may he fortified with spicules. In some cases such gemmules persist during certain unfavorable conditions and germinate, forming new sponges, when more favorable conditions recur. For example, in Southern New England, specimens of Haliclona loosanoffi degen- erate during the late summer and early fall, leaving exposed on the substrate gemmules that were produced at the bases of the sponges. During the winter when the water temperature may fall to below 0° C and small protected bodies of water may be covered by ice for several weeks, this sponge occurs only in the form of gemmules. These gemmules germinate in late spring when the water temperature rises to about 20° C (Hartman, 1958; and Fell, 1974 ). On the other hand, at Hat- teras Harbor, North Carolina Haliclona loosanoffi is abundant during the winter when the water temperature drops to at least 5° C. There this sponge is found ex- clusively in the form of gemmules during part of the warm summer (Wells et al, 1964). In view of the apparent relationship of gemmule germination to water tempera- ture, two types of studies were undertaken in order to examine this relationship in greater detail. First, a 3-year field study of Haliclona loosanoffi. was made to determine with greater precision the times at which various events in the life history of this sponge occur. Secondly, the germination of gemmules at different temperatures in the laboratory was studied with the hope of gaining some informa- tion concerning the regulation of germination. These studies provide evidence for winter diapause in New England populations of Haliclona loosanoffi. The occurrence of gemmules in a related species, Haliclona oculata, from Fishers Island Sound is also reported. Although many specimens of this sponge from the northeastern coast of the United States, including specimens from Long Island Sound, have been observed (see de Laubenfels, 1949; Hartman, 1958), there appears to be no previous record of gemmule production. However, Topsent (1888) has briefly described the gemmules of Haliclona (Chalina) oculata on the Channel Coast of France. MATERIALS AND METHODS Field studies A 3-year study of Haliclona loosanoffi was conducted in the Mystic Estuary (Connecticut) in the region north of Mason Island. Observations and collections 1 This study was supported by several Faculty Research Grants from Connecticut College. 333 334 PAUL E. FELL of specimens were made at approximately 2-week intervals during this period. The water temperature was also recorded. From March through December of the last 2 years water samples \vere taken for salinity determinations. The salinity of filtered samples (Whatman No. 1 filter paper) was measured to the nearest 0.5c/cc with a Goldberg refractometer (American Optical). The specimens were fixed in Benin's solution in sea water, and each was dis- sected and examined under a dissecting microscope for the presence of gemmules and embryos (and/or large oocytes). Small samples of the specimens were then embedded in paraffin and sectioned serially at 10 p.. The mounted, deparaffinized sections wrere stained with hematoxylin and eosin. Examination of such histo- logical sections revealed the stages of development of the gemmules and the pres- ence of small oocytes and spermatic cysts. Studies of Haliclona oculata in Fishers Island Sound (New York) were also made over a 3-year period. Specimens were collected from gravel bottom near the Dumpling Islands in about 30 to 40 feet of water, using 2-bushel oyster dredges. The specimens were preserved on board the boat immediately following their collection. The preservation and examination of the specimens was the same as for Haliclona loosanoffi. Experimental studies Gemmules of Haliclona loosanoffi., encrusting blades of eel grass and algae, were collected during the fall in the Mystic Estuary. Some of the gemmules were stored in the dark at 5° C in covered finger bowls containing sea water; and others were immediately prepared for culturing. The gemmules collected during 1970 were stored in Mystic Estuary sea water, while those collected during 1971 and 1972 were stored in Instant Ocean sea water (Aquarium Systems, Inc.). The Instant Ocean sea water with trace elements was made with glass-distilled water to produce a solution with a salinity of about 24% c. The sea water was changed in the storage vessels at approximately 2-week intervals. Small pieces of eel grass (ca 4x3 mm) bearing clusters of gemmules were cultured on sheets of lens paper in covered 4 inch finger bowls containing about 150 ml of Instant Ocean sea water (24f/(c). Four to 12 such pieces of eel grass were cultured in a single bowl ; and the sea water wras usually changed at approxi- mately 5-day intervals. In some cases the sea water was changed every other day once germination had begun. The cultures were maintained in the dark at a constant temperature (±1° C) in B.O.D. incubators. In most cases when the germination of gemmules under two different tempera- ture regemes was to be compared, blades of eel grass covered by gemmules were cut into pieces across their width ; and alternate pieces were put in one of 2 groups. One group of cultures was placed under one set of conditions, and the second group was put under another. In this way the 2 groups were as nearly identical as possible. The germination of the clusters of gemmules was scored according to 5 arbitrary stages. The characteristics of each of these stages were as follows : stage 1 — one or a few small masses of sponge tissue with no oscular tubes ; stage 2 — more extensive germination, but with less than half of the surface of the culture covered by sponge tissue, no oscular tubes; stage 3 — half or more of the surface of the DIAPAUSE IN A MARINE SPONGE 335 culture covered by sponge tissue with no oscular tubes ; stage A — less tban half of the surface of the culture covered by sponge tissue, but with one or more oscular tubes ; stage 5 — half or more of the surface of the culture covered by sponge tissue with one or more oscular tubes. The average stage attained by any particular group of cultures is called the germination index (GI). A "P" is used after the germination index to indicate that germination was largely restricted to the edges of the cultures. RESULTS Field observations In order to adequately evaluate experimental results, such as those presented below, detailed information about the occurrence of various events in the life history of Haliclona loosanoffi under natural conditions is needed. A population of this sponge was regularly sampled at approximately 2-week intervals for a period extending from March 1969 through February 1972. Other less regular observa- tions were made both before and subsequent to this 3-year period. A preliminary report on this study has already been made (Fell, 1974). Here primary con- sideration is given to the gemmules ; sexual reproduction will be the subject of another publication. The present study was carried out in the Mystic Estuary, in an area that extends between Mason Island and Pequotsepos Brook. The water in this part of the esutary is shallow, ranging from about 2 to 4 feet in depth. The water temperatures during the winter and summer may differ by more than 25° C. During the months of June through September, the water temperature ranged from 19.0° C to 28.5° C (mean 23.0 ± 2.5° C), while during the months of January and February, the water temperature was generally at or below 0° C (range - 2.0° C to 3.0° C, mean - 0.5° C). For a major portion of the cold period a layer of ice several inches thick covered the region. From March through May the water temperature rose steadily, and from October through December it sharply declined (see Table I). The salinity generally ranged betwen about 22% c and 32/^r and was highest during the late summer and early fall (also see Pearcy. 1962). Table I summarizes some of the observations made on the life history of Hali- clona loosanoffi. The earliest time of year that active specimens of this sponge have been found in the Mystic Estuary is late May. Such specimens and some ungerminated gemmules were collected on 27 May 1969 and 1 June 1970. In May of 1969 and June of 1969 and 1970 a total of 42 specimens of Haliclona loosanoffi, were found resting on the bottom. Nearly all of these specimens (39) enclosed empty gemmule capsules, and 9 of them were also associated with yet ungerminated gemmules. It is therefore evident that these specimens developed from gemmules. Pieces of eel grass bearing only ungerminated gemmules or gemmules in the initial stages of germination ( Fig. 1 ) were also observed. The latter were found on 9 June 1969 and 1 June 1970. Bottom specimens of Haliclona loosanoffi were not found until 15 July in 1971 and were not observed at all in 1972. However, such specimens were found on 12 June and 26 June 1973. The observed variation in the occurrence of specimens 336 PAUL E. FELL derived from gemmules is probably due in part to sampling problems and in part to differences in the production and survival of gemmules from year to year. Apparently sexual reproduction is initiated soon after the sponges develop from gemmules. The 4 bottom-specimens of Haliclona loosanoffi collected in late May 1969 and most (34/38) of the bottom-specimens taken in June 1969 and 1970 possessed oocytes and/or embryos or spermatic cysts. Specimens with sexual reproductive elements were found through July. Both bottom-specimens (many, at least, produced from gemmules) and speci- mens attached to living eel grass and algae (developed from sexually produced larvae) were found in the study area during most of the summer of some years (1969, 1970 and 1973). Gemmule formation was first observed in late June in TABLE I The occurrence of gemmules and active specimens of Haliclona loosanoffi in the Mystic Estuary Gemmules Month Water temp. °c* Active sponges Developing Formed Exposed Germinating Jan. -0.5 — — + + — Feb. 0 — — + + — March 5.5 — — + + — April 10 — — + + — May- 16 + — + + + ** lime 22 + + + + + July 24 + + + — ? Aug. 24 + + + ± — Sept. 21 + + + ± — Oct. 16 ± + + + — Nov. 7.5 ± — + + — Dec. 2.5 — — + + * Mean surface water temperature at low tide. ** Not actually observed; but since some sponges have been found in May, some germination must occur during this month. some of the bottom-specimens and in late July in the sexually produced specimens attached to eel grass. Thus this process may begin within about one month after the germination of the gemmules produced during the preceding year. The pro- duction of gemmules was found to continue into October, the substrates of the sponges being progressively covered with gemmules as the specimens grow. Regression began in some specimens of Haliclona loosanoffi by late August and early September ; and some gemmules were exposed at this time. By early October there were many pieces of eel grass and algae bearing exposed gemmules. Many of these gemmules were completely exposed, while others were partially covered by a latticework of bare parental skeleton. Living specimens of Haliclona loosanoffi were associated with some of these gemmules, but not with many others. Many of the specimens showed evidence of degeneration. In some cases the sponge tissue was retracted slightly away from the basal gemmules, and in other cases it was separated from them by a broad zone of vacated skeleton (Fig. 2). DIAPAUSE IN A MARINE SPONGE 337 Furthermore, many of the specimens found at this time of year were a purplish color instead of the usual rosy beige or tan. The latest that "active" specimens of Haliclona hosanoffi were observed in the Mystic Estuary was mid-November. A few small lavender-colored specimens were found on 14 November 1970, but no active specimens were observed during this month in other years. FIGURE 1. Living gemmules of Haliclona loosanoffi in an early stage of germination, col- lected on 9 June 1969 (scale bar — 1 mm). FIGURE 2. Preserved specimen of Haliclona loosanoffi collected on 11 October 1969. Note the exposed gemmules (g) and vacated skeleton (s) (scale bar — 10 mm). Gemmules were the only form in which Haliclona loosanoffi was found from November through late May. These gemmules encrusted pieces of dead eel grass and algae ; and in some cases they were covered by the tissue of another sponge, Halichondria boiverbanki, which is sometimes abundant during the winter. The structure of the gemmules of Haliclona loosanoffi has been described by Hartman (1958) and Simpson and Fell (1974). 338 PAUL E. FELL Germination of fall gcuimnles at 20° C Gemmules collected on 4 October 1971 were used to prepare 50 cultures which were divided into two equal groups (A-l and A-2). Both groups were initially placed at 20° C. After about one week most of the cultures showed germination of the gemmules along their edges, and many of the resulting sponges possessed 5 4 3 2 CD A-l, 25 CULTURES PERIPHERAL GERMINATION GENERAL GERMINATION DAYS 10 20 30 40 50 60 70 80 90 100 NO 120 TEMP 20°C 20°C X CD A-2, 25 CULTURES DAYS 10 20 30 40 50 60 70 TEMP^ 20°C FIGURE 3. Germination of gemmules, collected on 4 October 1971, at 20° C before and after a 4-week period at 5° C. oscular tubes (stage 4P). The germination index was 3.3P (see Fig. 3). After about 20 days many of the sponges showed signs of degeneration. The oscular tubes disappeared, and the sponge tissue began to gradually waste away. At this time all of the cultures still contained many ungerminated gemmules. The occurrence of germination along the edges of the cultures and especially along the cut edges was very striking (see Fig. 4). On the 9th day of culture 82% of the cut edges bore sponges compared with only 18% of the uncut edges DIAPAUSE IN A MARINE SPONGE 339 and 22% of the upper surfaces. This pattern of germination suggests that physical damage to some of the cells of the gemmules and/or to the gemmule capsules may in some way stimulate the gemmules to germinate. This suggestion is strengthened by the fact that cultures of gemmules. which have no cut edges, do not exhibit such preferential germination at their periphery (see below). On the 28th day one group of cultures (A-2) was placed at 5° C for 30 days. At the end of the cold treatment it was returned to 20° C. Seven days after the cultures were placed at the higher temperature, all of them had begun to germinate ; and by the 9th day all of them had reached stage 5 (see Figures 3 and 5). Germination was therefore essentially complete on the 67th day of total culture. In the other group of cultures (A-l), which was kept at 20° C, the first indica- tion of later germination was on day 59. Eventually 5 of the 25 cultures showed some germination (GI = 0.9), after which the sponge tissue began to regress. On day 77, which was 10 days after all of the A-2 cultures had reached stage 5, the A-l cultures were placed at 5° C for 33 days. Then they were returned to 20° C. By the 12th day at 20° C (the 122nd day of total culture), all 25 cultures were at stage 5 (see Fig. 3). In another experiment 17 cultures were prepared using large, uncut pieces of eel grass covered by gemmules (average size, ca 34 X 4 mm). The masses of gemmules, which were collected on 28 October 1971, were divided into 2 groups. One group of 9 cultures (B-l) was placed initially at 20° C; and the other group (B-2), which consisted of 8 cultures, was cultured at 5° C for one month before being placed at 20° C. The B-l cultures did not experience an early, peripheral germination. By day 29 only 2 cultures showed any signs of germination. Both of these cultures were at stage 1 (GI — 0.2). After 39 days the germination index was only 1.1; on day 46 it was 3.1; and on day 55 it was 3.7. In all of the cultures there were still many ungerminated gemmules. On the 55th day the cultures were placed at 5° C for 41 days. At the end of the cold treatment the cultures were returned to 20° C. By the llth day at 20° C (the 107th day of total culture) all of the cultures had reached stage 5. The B-2 cultures were transferred from 5° C to 20° C after 31 days of culture. By the 12th day at the higher temperature (the 43rd day of total culture) the germination index was 4.4. Five of the 8 cultures produced large sponges, but all of them still possessed many ungerminated gemmules. The results of these experiments suggest that gemmules collected during the early fall do not readily germinate at 20° C and that low temperature enhances germination of the gemmules. Additional support for these conclusions is given by the experiment described in the next section. Germination of fall gemmules at 20° C following a short period at 5° C Four groups of cultures were set up using gemmules collected on 4 October 1971. These groups consisted of gemmules which had been stored at 5° C for different lengths of time (1, 2, 5 and 9 weeks) before being cultured at 20° C. All of the cultures derived from the same mass of gemmules were placed in the same finger bowl so that differences between different masses of gemmules (at least 3 per group) could be detected. Table II summarizes the results of this experiment. 340 PAUL E. FELL Twenty eight cultures were set up using gemmules which had been kept at 5° C for one week. These were divided into 2 groups: group BA-1 containing 18 cultures and group BA-2 containing 10 cultures. When the cultures were placed at 20° C, they behaved like cultures which had received no cold treatment. After about one week there \vas extensive germination along the edges of the cultures (GI==3.6P), but all of the cultures still contained many ungerminated gemmules. Again most of the germination occurred along the cut edges of the cul- tures. On the 7th day 87% of the cut edges were covered by small sponges, while only 29% of the uncut edges and 25% of the upper surfaces of the cultures bore sponges. Over a period of several weeks, the sponges in these cultures gradually wasted away. On the 28th clay the BA-1 cultures were placed at 5° C for 28 days and then returned to 20° C. By the llth day at 20° C (the 67th day of total culture) most of the cultures exhibited extensive germination, and the germination index TABLE 1 1 The effect of a period (s) at 5° C on the stibxei/in-nt germination of gem HI ides at 20° C (all gemmules collected 4 October /''7/>; germination index: (A) initial germination at 20° C, (B) subsequent germination by end of temperature regime, (C) germination after a later 4-week cold treatment Germination index E\D rV i ). cult u if - Tcm pp. Tilt, u rp rcsiiris^1' J— "*-f « A B c A-2 25 4-4-1 (3.3P) 5.0 — A- 1 25 10 0.9 5.0 BA-1 BA-2 18 10 7-4-4-2 7-10 (3.61' 4.7 0.8 5.0 BB-la 10 2-1 / 5.0 — — BB-lb 10 2-4-3-2 ( 4.2 , - 5.0 — BB-2 12 2-9 \ 1.3 5.0 BC 33 5-2 4.0 — — BD 30 9-2 4.9 • — • — ' Number of \veeks at 20° C in regular type; number of weeks at 5° C in italics. was 4.7. At this time the germination index of the BA-2 cultures, which had been kept continuously at 20° C, was only 0.8. On day 71 the BA-2 cultures were placed at 5° C for 31 days. At the end of the cold treatment the cultures were returned to 20° C, and by the 7th day at the higher temperature all of them had reached stage 5 (see Table II). A group of 8 cultures (BA-3) was prepared at the same time as the other two, using gemmules kept at 5° C for one week. The capsules of many of the gemmules in this group were pierced with a sharp iridectomy knife. After about a week at 20° C, 7 of the cultures showed extensive germination ; and the germina- tion index was 4.75. Admittedly the number of cultures is small, but the results lend further support to the suggestion that damage to the gemmules may cause early germination. Although many of the gemmules in this group of cultures germinated within 2 weeks, many others did not germinate until after they had received a 31-day cold treatment. DIAPAUSE IN A MARINE SPONGE 341 Thirty two cultures were prepared from gemmules which had been stored at 5° C for 2 weeks. These were divided into one group of 20 cultures (BB-1) and another group of 12 cultures (BB-2). Both groups of cultures were initially placed at 20° C. By the 8th day 14 cultures were at stage 5, and the germination index was 4.2. Ten of the stage-5 cultures (part of group BB-1) were from a FIGURES 4 and 5. Cultures of gemmules collected on 4 October 1971 : 4. 8-day culture at 20° C showing germination along the cut edges of the culture ; 5. stage-5 culture on the 13th day at 20° C following a long period at 5° C. Note the oscular tube (o) (scale bar = 1 mm). FIGURE 6. Surface of a specimen of Haliclona oculata that was formerly in contact with the substrate. Note the gemmules. Specimen was collected on 6 June 1972 (scale bar = 2 mm). single blade of eel grass. Excluding these cultures, the germination index was 3.5 or about the same as that for the cultures kept at 5° C for one week prior to being placed at 20° C. However, in this set of cultures there was little pe- ripheral germination ; only 10 of the 32 cultures had sponge tissue restricted to their periphery. 342 PAUL E. FELL The 10 stage-5 cultures of group BB-1, which were from the same mass of gemmules, were discarded since most, if not all, of the gemmules had apparently germinated. After the remaining BB-1 cultures had been at 20° C for 30 days, they were placed at 5° C for 18 days. At the end of the cold treatment the cultures were transferred back to 20° C. By the 13th day at the higher tempera- ture (the 61st day of total culture) all of the cultures were at stage 5. On the other hand, the BB-2 cultures, which had not had a second exposure to low temperature, had a germination index of only 1.3. On the 64th day of culture the BB-2 cultures were placed at 5° C for 31 days, and then they were trans- ferred back to 20° C. After 9 days at 20° C all of these cultures had reached stage 5 (see Table II). Gemmules, which had been stored at 5° C for 5 weeks, were used to set up 33 cultures. After about 2 weeks at 20° C the cumulative germination index was 4. Although in this case the germination index was somewhat low, in other experi- ments a 4 to 6 week period at 5° C was sufficient to produce nearly complete germination. This was true in experiments A, BA and BB (see Table II). TABLE III The germination of cold-treated getnninlf* at various temperatures (all gemmules collected in October) Culture temp. °C No. cultures Weeks stored at 5° C Days to max. germination Germination index 10 34 43 — 0 15 42 30 & 3<> avg. 22 5.0 20 180 4-9 avg. 11 4.73 2n 146 30-41 avg. 10 4.96 25 35 41 12 5.0 30 69 31 & 35 avg. 10 3.96 34 18 38 • — • 0 Thirty cultures were set up using gemmules which had been stored at 5° C for 9 weeks. By the llth day at 20° C nearly all of the cultures were at stage 5, and the germination index was 4.9. Although most of the cultures produced large sponges, there were still a number of ungerminated gemmules in many of the cultures. Germination of fall gciuiniiles at various temperatures after a long period at 5° C Gemmules, which had been collected in October and stored at 5° C for from 30 to 43 weeks, were tested for their capacity to germinate at 10, 15, 20, 25, 30 and 34° C. In most experiments the germination of gemmules at 20° C was compared with that at some other temperature. The results of these experi- ments are summarized in Table III. Thirty four cultures, which were kept at 10° C for 30 days, showed no evi- dence of germination. That the gemmules were healthy and capable of germination under favorable conditions was demonstrated by their response when they were transferred to 20° C. By the 9th day at the higher temperature all of the cul- tures had reached stage 5. DIAPAUSE IN A MARINE SPONGE 343 A total of 42 cultures were kept at 15° C. In one experiment involving 30 cultures, the first evidence of germination was on the 15th day and it was not until the 27th day that all of the cultures had reached stage 5. In another experiment, 12 cultures kept at 15° C all reached stage 5 by day 17. While the time required for germination was less than in the first experiment, it was sub- stantially greater than that for 12 control cultures kept at 20° C. All of the latter cultures reached stage 5 on the 8th day of culture. As has already been indicated, gemmules, which have had at least a 4-week cold treatment, germinate within about one to 2 weeks at 20° C. In a number of experiments involving over 300 cultures, the maximal germination index was usually achieved within 8 to 12 days, and 50% or more of the cultures showed evidence of germination within 5 to 7 days (see Table III). Thirty five cultures kept at 25° C behaved essentially like cultures kept at 20° C, except that germination was accelerated by about 24 to 48 hrs. On the 5th day most of the cultures placed at 25° C were in early stages of germination, and one day later 28 of them were at stage 5. By comparison, only 4 of 35 control cultures kept at 20° C had begun to germinate by the 5th day, and none of them were at stage 5 on day 6. However, on the 7th and 8th days of culture the number of cultures to have reached stage 5 was 18 and 28 respectively. A total of 69 cultures were placed at 30° C. Many of these cultures reached stage 5, but their development was inferior to that of gemmules kept at 20 or 25° C. In one experiment, involving 38 cultures, about half of the cultures reached stage 5 ; and the germination index was 3.7. In a second experiment, using 31 cultures, the germination index was 4.3. Cotnrol cultures (30) kept at 20° C were delayed in their germination by about 24 hrs compared to the cultures kept at 30° C, but all of them reached stage 5 by the 8th day of culture. Not only did more cultures reach stage 5 at 20° C, but the cultures appeared to be more robust. Twelve cultures of the second experiment, which did not reach stage 5 (GI -- 3.2), were subsequently transferred to 20° C. After 9 days at the lower temperature the germination index of this group of cultures was 4.9. Twenty nine cultures were placed at a variable high temperature (31.0 to 34.5° C). By the 8th day of culture only 10 cultures (all from one group of gemmules) showed any evidence of germination. On the other hand, 29 control cultures kept at 20° C had all reached stage 5 by this time. Already by the 7th day of culture sponge tissue, which resulted from the limited germination at high temperature, was beginning to regress. On the 8th day the remaining sponge tissue was removed from the gemmules with a sable brush, and the cultures were placed at 20° C. By the 13th day at 20° C (the 21st day of total culture) 27/29 cultures had attained stage 5 (GI — 4.8). This experiment suggests that the germination of the gemmules of Haliclona loosanoffi may be reversibly inhibited by high, as well as by low, temperature. Finally, 18 cultures were kept at 34° C. These cultures showed no evidence of germination during 20 days at this temperature ; and when they were sub- sequently transferred to 20° C, there was no germination during a 16-day period. By comparison most of the 18 control cultures placed at 20° C were in early stages of germination on the 6th day and had reached stage 5 by day 12 (GI = 4.9). This result suggests that a prolonged exposure of the gemmules to 34° C is lethal. 344 PAUL E. FELL Additional observations In a number of cultures it was noted that little or no germination occurred among the gemmules situated on the lower surface of the eel grass. However, when the cultures were recultured in an inverted position, there was frequently extensive germination among these gemmules. This observation suggests that conditions are less favorable for germination on the undersurfaces of the cultures where the gemmules are in contact with the substrate. This may be due to a restriction of respiratory gas exchange and/or an accumulation of certain metabolic byproducts. It appears that the gemmules of Haliclona loosanoffi can be stored at 5° C under the conditions used in this study for only about one year. In all, 112 cul- tures were prepared from gemmules collected on 8 October 1970 and stored at 5° C for from 7 to 10 months. These cultures, which were kept at from 15 to 30° C, had a germination index of 4.7. The only cultures, which did not reach stage 5, were some of those kept at 30° C. On the other hand, 90 cultures derived from gemmules collected on the same date and stored at 5° C for from 16.5 to 19.5 months had a total germination index of only 0.1 when cultured at 20 or 25° C. Finally, many of the gemmules, kept at 10 to 20° C for long periods of time without germination, became covered by a thin, dark brown layer of what appeared to be algal growth. However, this layer did not seem to significantly hinder the subsequent germination of the gemmules. The gemmules of Haliclona oculata More than 300 specimens of Haliclona oculata were dredged from Fishers Island Sound in the vicinity of the Dumpling Islands. Collections were made at irregular intervals over a period extending from May 1971 to February 1974 and during every month except September. In all of the 18 collections, many of the sponges (40 to 100%, avg. 76%) possessed gemmules. These were situated at the base of the stalk, either in contact with the substrate or only a few milli- meters above it (Fig. 6). From March through June many of the specimens possessed oocytes, embroys, and/or larvae in progressively advanced stages of development. In some cases these occurred throughout the endosome from the fibrous stalk to the tip of the sponge. Some of the specimens, which lacked oocytes and embryos, were found to contain large numbers of spermatic cysts. In many of the collections, a few of the specimens showed evidence of basal degeneration. In some cases the stalk was devoid of living tissue, but most of the rest of the specimen appeared to be healthy. In other cases the dead stalk supported only a few small masses of sponge. Dead stalks alone were also found. What appeared to be living gemmules were observed at the base of some of the dead stalks. The number of gemmules produced by any specimen was relatively small. An exact enumeration of the gemmules was difficult, because they were frequently embedded in a dense network of spongin (collagen) fibers. However, it appeared that some of the specimens possessed fewer than 10 gemmules, while others con- tained as many as 70 or more. This is in general agreement with the observa- tions of Topsent (1888) ; however, the estimate given by him that 30 is the DIAPAUSE IN A MARINE SPONGE 345 maximal number of gemmules produced by a single specimen of this sponge is low. Although one specimen 57 mm in height was found to possess gemmules, it appears that these structures are generally restricted to larger specimens ( > ca 80 mm in height ) . The gemmules were generally from about 600 //. to more than 1000 //, in their greatest diameter. An attempt to bring about the germination of the gemmules of this sponge was unsuccessful. Fifteen groups of gemmules, collected on 4 May 1973, were placed in finger bowls in an aquarium (Dayno Aqua Lab) containing 20 gallons of freshly collected sea water (30/£c ) that was constantly circulated. One to 3 groups of gemmules were placed in each finger bowl. Four of the groups of gemmules were collected with dead stalks, and the rest were obtained from living specimens. The water in the aquarium was maintained at 8° C during the first 48 hours, after which it was allowed to come up to room temperature (ca 23° C). During 28 days of culture there was no evidence of germination, although the gemmules of Haliclona loosanoffi readily germinate within a few days under similar con- ditions (Olmstead and Fell, unpublished). DISCUSSION Haliclona loosanoffi and Haliclona oculata are among at least five species of Haliclona that are known to produce gemmules. The other species are Haliclona (Clialina) gracilcnta (Topsent, 1888); Haliclona ccbasis (Fell, 1970); and Haliclona. pennollis (David Elvin, Oregon State, personal communication). How- ever, not all haliclonids form gemmules. For example, in Long Island Sound Haliclona caiialictilata overwinters in a simplified form lacking flagellated chambers, but does not produce gemmules (Hartman, 1958). Other gemmuliferous marine sponges include Subcrites doiniinciila (Herlant-Meewis, 1948), Sitbcrites fi'~its (Topsent. 1888; Hartman, 1958), Prosuberitcs inicrosclcrns (Wells et al, 1964), La.vosubcritcs lacitstris (Annandale, 1915), Cliona rastifica (Topsent, 1888; Annandale, 1915), and Cliona tniitti (Wells ct al, 1964). A number of other sponges, including Haliclona heferofibrosa (Bergquist, Sinclair and Hogg, 1970), apparently produce gemmule-like masses which develop into free-swimming parenchymula larvae. Such "gemmules" usually are not in contact with the substrate of the sponge, and they do not possess a thick enveloping capsule. Although this form of asexual reproduction was first described by Wilson in 1891 and 1894, it remains poorly understood (see Fell, 1974). The gemmules of Haliclona loosanaffi are frequently present throughout most of the year in the Mystic Estuary. They are produced beginning in June or July and do not germinate until the following May or June. During the late summer and early fall the parent sponges degenerate, leaving the gemmules attached to the substrate. The gemmules are the only form in which this sponge exists during the colder months of the year, and consequently at this location they are an obligatory part of the life history. Evidently the gemmules of this sponge are special structures which permit it to survive low temperature and other adverse conditions (also see Hartman, 1958). In this study it has been shown that the gemmules of Haliclona loosanoffi may be stored at 5° C for up to 10 months without showing any signs of germina- tion. However, such gemmules readily germinate when they are subsequently 346 PAUL E. FELL cultured at 20° C. Gemmules collected in early October have germinated in the laboratory as early as December and as late as September. Germination also occurs at 15° C but not at 10° C. Well developed sponges usually develop in about one to 2 weeks at 20° C or in approximately 2 to 4 weeks at 15° C. The results of these laboratory studies are in good agreement with field studies which suggest that germination normally occurs as the water temperature is rising from about 15 to 25° C. It therefore appears that germination of the gemmules is prevented during the winter by low temperature. A similar situation has been found to exist for the gemmules of several fresh- water sponges (Rasmont, 1954, 1962 and 1963; Strekal and McDifTett, 1974). However, the gemmules of S pony ilia laciistris and TrocJwspongilla (Tnbclla) pennsylvanica (in Northern New England) normally germinate when the water temperature is still only 4 to 5° C (Simpson and Gilbert, 1973). The regulation of gemmule germination in Haliclona loosanoffi is not as simple as it first appears. The gemmules begin to form early in the summer, and the completed gemmules become exposed on the substrate at the end of the summer when the water temperature may be close to 20° C or about the same as that occurring in the spring when the gemmules normally germinate. Some mecha- nism (s) must therefore be operating to prevent the gemmules from germinating prior to the onset of winter conditions. Gemmules of Haliclona loosanoffi, collected in the early fall and cultured at 20° C without a period of storage at 5° C. do not germinate as readily as cold- treated gemmules. In a number of experiments, involving gemmules with either no or only a very short cold treatment, few of the gemmules germinated during approximately 2 months of culture at 20° C. On the other hand, cultures of gem- mules, which had been kept at 5° C for about 4 weeks before being placed at 20° C. underwent extensive germination within a period of 2 weeks. These experiments suggest that some process (es) in the germination of the gemmules requires a relatively long period of time and that this process is accelerated by, but is not totally dependent upon low temperature. Such a condition is known as diapause (Agrell, 1951; Rasmont, 1954). The diapause experienced by the gemmules of Haliclona loosanoffi appears to be very similar to that occurring in the gemmules of the fresh-water sponge, Ephydatia miillcri. Gemmules of the latter species, collected in July and main- tained for nearly 4 months at 3.5° C, showed 90% germination after 8 days at 20° C, while gemmules, kept at 14 to 16° C for the same period, exhibited only 1% germination after 8 days and 11.5% germination after 20 days at 20° C (Rasmont, 1954). The timing of certain events in the life histories of these sponges is also similar. Near Brussels, Belgium Ephydatia miillcri reproduces sexually in May, produces gemmules primarily during June and July, and regresses, except for gemmules, by early fall (Rasmont, 1962). The gemmules of Spongllla fragilis and the brown gemmules of Spongilla lacustris also undergo diapause, but in these cases the diapause appears to be less deep (Rasmont, 1954 and 1955). The exposure of the gemmules of Haliclona loosanoffi to 5° C for a period of one week has little effect on subsequent germination compared to that of unchilled gemmules. However, a cold treatment of only 2 weeks appears to result in a definite but variable enhancement of germination; and exposure to 5° C for 4 or more weeks usually leads to the germination of most of the gemmules when they DIAPAUSE IN A MARINE SPONGE 347 are then cultured at 20° C. Similarity, Rasmont (1954) found that 3-day ex- posure to 3.5° C had no effect on the subsequent germination of the gemmules of Ephydatia millleri, but that treatments of 12, 20 and 30 days resulted in progres- sively higher percentages of germination, 45%, 75% and 92% of the gemmules respectively. The optimal temperature for breaking diapause in the gemmules of Haliclona loosanoffi has not yet been determined. However, it has been shown that 3° C and 8° C are equally effective in bringing about germination of the gemmules of Ephydatia millleri (Rasmont, 1955). A temperature of 12° C was somewhat less effective than the lower temperatures but was substantially more effective than 18° C. Gemmules collected on 1 September and exposed to 3 or 8° C for one month subsequently showed 90% germination when they were cultured at 20° C. By comparison 60% of the gemmules exposed to 12° C and only 25% of those exposed to 18° C germinated when tested under the same conditions. However, it was found that after 2 to 3 months at 18° C, most of the gemmules were capable of germination. A similiar situation was shown to exist for the gemmules of Spongilla fragilis, but for this species 8° C was somewhat more effective than either 3 or 12° C in breaking diapause (Rasmont, 1955). If the gemmules of Haliclona loosanoffi are physiologically similar to those of these fresh-water sponges, 5° C should be close to the optimal temperature leading to the matura- tion of the gemmules. When and how the diapause is initiated are unknown. One would like to know whether it is initiated when the gemmules are formed or at some later time. Also nothing is presently known concerning the nature of the diapause itself. However, one observation on the germination of gemmules, which had received no cold treatment or only a very short one, is of interest in this con- nection. Frequently many of the gemmules situated along the cut edges of the cultures germinated within a few days, while most of the rest of the gemmules did not germinate even after a long period of time. This suggests that injury to the gemmule capsule and/or to the cells of the gemmule may stimulate germina- tion. Perhaps the gemmule capsule, like the coats of certain seeds, is important in maintaining dormancy either by restricting respiration or by retarding the loss and/or inactivation of germination inhibitors (Roberts, 1969). Ephydatia fluviatilis and Spongilla lacustris produce an inhibitor of gemmule germination (so-called gemmulostasin), but Ephydatia millleri does not (Ras- mont, 1965; Rosenfeld. 1970). The first 2 species form gemmules late in the year and do not degenerate until the onset of winter conditions (Rasmont, 1962). The inhibitor produced by the parental tissue inhibits the germination of the gemmules until this time ; and the gemmules of Ephydatia fluviatilis and the green gemmules of Spongilla lacnstris, unlike those of Ephydatia millleri, do not undergo diapause (Rasmont, 1954 and 1962). The inhibitors of Ephydatia fluviatilis and Spongilla lacnstris are not species specific and inhibit the germina- tion of the gemmules of Ephydatia millleri. That of Ephydatia fluviatilis acts on an early phase of germination ; and once this phase is past, the inhibitor has no effect on the later development of the new sponge. As would be expected, the effect of the inhibitor is reversible (Rosenfeld, 1970). It is of interest that if the capsule of the gemmules is pierced, germination is not inhibited by gemmulo- stasin although it is slowed compared to that of untreated controls (Rosenfeld, 348 PAUL E. FELL 1971). From the preceding discussion one would predict that Haliclona loosanoffi does not produce such an inhibitor of gemmule germination. Certainly if it does, the inhibitor could play only a minor regulatory role. At Hatteras Harbor, North Carolina some specimens of Haliclona loosanoffi degenerate during the winter, but many others are present throughout this period when the water temperature falls to about 5° C (Wells ct al., 1964). Although the winter at Hatteras Harbor is apparently less severe than at Mystic, there presently seems to be no obvious explanation for the survival of Haliclona loosanoffi at the former location but not at the latter. In the Mystic Estuary the sponges degenerate in the late summer and early fall when the water is still warm. The gemmules exposed during the winter at Hatteras Harbor germinate in April when the water temperature reaches approximately 16° C, a situation similar to that in New England (Wells ct al., 1964). On the other hand, at Hatteras Harbor Haliclona loosanoffi is absent, except for gemmules, during the middle of the summer when the water temperature is about 30° C. The summer gemmules, which germinate is September, may be of 2 types : those produced during the fall and winter and exposed to winter tem- peratures while enclosed by parental sponge tissue and those produced during the late spring and early summer (Wells ct al., 1964). It would be of interest to know whether these 2 classes of gemmules differ in their capacity to germinate. One class of gemmules was chilled and the other was not. However, the chilling of gemmules within the parental sponge may not have the same effect as that of chilling exposed gemmules. The gemmules of Haliclona ocitlata appear to play a different biological role from that played by the gemmules of Haliclona loosanoffi. Specimens of the former sponge evidently remain active and gemmules apparently are present throughout the year. The gemmules may repopulate the substrate when speci- mens degenerate or when the sponges are torn loose during storms. Herlant- Meewis (1948) believes that the gemmules of Suberitcs doinuncnla play a similar role. Although the present report is primarly concerned with the gemmules of Haliclona loosanoffi and Haliclona ocitlata, some information concerning sexual reproduction is given. In the Mystic Estuary specimens of Haliclona loosanoffi with reproductive elements are found in May, June and July (also see Fell, 1974). Larvae may be released as early as the middle of June, and the peak of larval settlement appears to occur in July (Fell, in preparation). The timing of repro- duction is very different from that reported for this species at Milford, Connecticut (Hartman, 1958). At the latter location oocytes and embryos are present in specimens collected during late August and early September and the peak of larval settlement occurs during late September and early October. It is somewhat sur- prising that the reproductive periods of this sponge at 2 localities as near to each other as Mystic and Milford should be so different. At Hatteras Harbor, North Carolina Haliclona loosanoffi. has 2 periods of larval settlement, one during June and July and the other in October and November (Wells ct al., 1964). At Mystic, Connecticut and Hatteras Harbor, North Carolina (Wells ct al., 1964) sexual reproduction is initiated very soon after sponges develop from gem- mule germination. A similar situation has been found to exist for the fresh- DIAPAUSE IN A MARINE SPONGE 349 water sponges, Spongilla lacustris and Trochospongilla (Tiibella) Pennsylvania! (Simpson and Gilbert, 1973). Furthermore, in Spongilla lacustris oocytes begin to develop within one week after refrigerator-stored gemmules are inplanted back into the natural habitat of the sponge, even when this is done well after the normal period of sexual reproduction (Gilbert, 1974). These facts suggest that there may be a regulatory connection between sexual reproduction and gemmule germination. However, gemmulation does not appear to be an obligatory prere- quisite to sexual reproduction in these species. Larva-derived specimens of Spongilla lacnstris (Simpson and Gilbert, 1974) and HaHdona loosanoffi (Fell, in preparation) may exhibit low levels of gamete production. It is reported here that the reproductive period of Haliclona ocnlata in Fishers Island Sound extends from March through June. The only other reference to the reproductive period of this sponge is that of Hartman (1958). He reported the occurrence of reproductive specimens in Block Island Sound during |uly. Appreciation is expressed to Sibyl Hausman, Ann Huckle, and Sandra Smith for assistance with the histological work and to Dr. Frances Roach and Ruth Fell for their invaluable help with the field studies. The assistance of Capt. L. H. Malloy and Frank Malloy with dredging aboard the ANNE is also gratefully acknowledged. SUMMARY 1. The gemmules of Haliclona loosanoffi are present throughout the year in the Mystic Estuary, but active specimens of this sponge are found only during the period extending from late May to late October or early November. Germina- tion of the gemmules occurs primarily during May and June, and new gemmules are produced from late June through early October. The gemmules of this species evidently are a means for surviving adverse environmental conditions. 2. Both gemmules and active specimens of Haliclona ocnlata are found through- out the year in Fishers Island Sound. The gemmules of this species may re- populate the substrate when the parent sponges degenerate or are ripped loose during storms. 3. Gemmules of Haliclona loosanoffi collected in the fall do not germinate readily when they are cultured at 20° C. There is frequently only limited germina- tion after 1 to 2 months at this temperature. However, if the gemmules are first put at 5° C for 4 or more weeks, they usually germinate within a few days after being placed at 20° C. Thus low temperature appears to enhance germination of these gemmules which undergo diapause. 4. Although low temperature appears to accelerate maturation of the gemmules of Haliclona' loosanoffi, it also inhibits actual germination. Gemmules do not germinate at either 5 or 10° C. However, germination occurs within 2 to 4 weeks at 15° C and within 1 to 2 weeks at 20° C. At 25° C germination appears to be accelerated by approximately 24 to 48 hours compared to that occurring at 20° C. 5. The gemmules of Haliclona loosanoffi, may be stored in the dark at 5° C for at least 10 months without any detectable reduction in their capacity to germi- 350 PAUL E. FELL nate at higher temperatures. However, such gemmules can not be stored indef- inately. Gemmules kept at 5° C for from 16.5 to 19.5 months did not germinate when they were subsequently placed at either 20 or 25° C. 6. Gemmules of Haliclona loosanoffi germinate at 30° C, but this tempera- ture appears to be less favorable than 20 or 25° C. Slightly higher temperatures inhibit germination, and continuous exposure to 34° C apparently is lethal. 7. The reproductive period of Haliclona loosanoffi in the Mystic Estuary ex- tends from late May through July and that of Haliclona oculata in Fishers Island Sound extends from March through June. LITERATURE CITED AGRELL, I., 1951. The diapause problem. Annee Biol, 27: 287-295. ANNANDALE, N., 1915. Fauna of the Chilka Lake: Sponges. Mem. Indian Mits., Calcutta, 5: 23-54. BERGQUIST, P. R., M. E. SINCLAIR AND J. J. HOGG, 1970. Adaptation to intertidal existence : reproductive cycles and larval behavior in Demospongiae. Pages 247-271 in W. G. Fry, Ed., Symposium Zoological Society London, No. 25, Biology of the Porifcrn. Academic Press, New York. DE LAUBENFELS, M. W., 1949. The sponges of Woods Hole and adjacent waters. Bull. Mus. Comp. Zoo!., Harvard, 103 : 1-55. FELL, P. E., 1970. The natural history of Haliclona ccbasis de Laubenfels, a siliceous sponge of California. Pacific Sci., 24 : 381-386. FELL, P. E., 1974. Porifera. Pages 51-132 in A. C. Giese and J. S. Pearse, Eds., Reproduc- tion of Marine Invertebrates, Volume I. Academic Press, New York. GILBERT, J. J., 1974. Field experiments on sexuality in the freshwater sponge Spongilla lacustris. The control of oocyte production and the fate of unfertilized oocytes. J.Exf. Zoo!., 188: 165-178. HARTMAN, W. D., 1958. Natural history of the marine sponges of southern New England. Bull. Peabody Mus., Yale, 12 : 1-155. HERLANT-MEEWIS, H., 1948. La gemulation chez Snbcritcs domuncula (Olici) Nardo. Arch. Anat. Micro. Morphol. Ex p.. 37 : 289-322. PEARCY, W. G., 1962. Ecology of an estuarine population of winter flounder Pscudopleuro- ncctcs americanus (Walbaum) I. Hydrography of the Mystic River Estuary. Bull. Bingham Oceanog. Coll., 18: 5-15. RASMONT, R., 1954. La diapause chez les Spongillides. Bull. Acad. Rov. Belgium, Sci., 40: 288-304. RASMONT, R., 1955. La gemmulation des Spongillides II. Modalites de la diapause gem- mulaire. Bull. Acad. Roy. Belgium, Sci.. 41 : 214-223. RASMONT, R., 1962. The physiology of gemmulation in fresh-water sponges. Pages 3-25 in D. Rudnick, Ed., Twentieth Symposium Society for the Study of Development and Grozvth, Regeneration. The Ronald Press Co., New York. RASMONT, R., 1963. Le role de la taille et de la nutrition dans le determinisme de la gem- mulation chez les Spongillides. Develop. Biol., 8: 243-271. RASMONT, R., 1965. Existence d' une regulation biochimique de 1' eclosion des gemmules chez les Spongillides. Acad. Sci. Paris, Compt. Rend., 261 : 845-847. ROBERTS, E. H., 1969. Seed dormancy and oxidative processes. Pages 161-192 in Symposium Society Experimental Biology, No. 23, Dormancy and Survival. Academic Press, New York. ROSENFELD, F., 1970. Inhibition du developpement des gemmules de Spongillides : specificite et moment d' action de la gemmulostasine. Arch. Biol., 81 : 193-214. ROSENFELD, F., 1971. Effets de la perforation de la coque des gemmules d' Ephydatia fluviatilis (Spongillides) sur leur developpement ulterieur en presence de gemmulostasine. Arch. Biol, 82: 102-113. SIMPSON, T. L., AND J. J. GILBERT, 1973. Gemmulation, gemmule hatching, and sexual re- production in fresh-water sponges I. The life cycle of Spongilla lacustris and Tubella pennsylvanica. Trans. Amer. Microscop. Soc., 92 : 422-433. DIAPAUSE IN A MARINE SPONGE 351 SIMPSON", T. L., AND J. J. GILBERT, 1974. Gemmulation, gemmule hatching, and sexual re- production in fresh-water sponges II. Life cycle events in young, larva-produced sponges of Spongilla lacustris and an unidentified species. Trans. Amcr. Microscop. Soc., 93 : 39-45. ' SIMPSON, T. L., AND P. E. FELL, 1974. Dormancy among the Porifera : gemmule forma- tion and germination in fresh-water and marine sponges. In press in J. H. Bushnell, Ed., Perspectives on the Biology of Dormancy. Transactions American Micro- scopical Society, New York. STREKAL, T. A., AND W. F. McDiFFET, 1974. Factors affecting germination, growth, and distribution of the freshwater sponge, Spongilla jragilis Leidy (Porifera). Biol. Bull., 146: 267-278. TOPSENT, E., 1888. Notes sur les gemmules de quelques silicispongidae marines. Acad. Sci. Paris, Compt. Rend., 106 : 1298-1300. WELLS, H. V., M. J. WELLS AND I. E. GRAY, 1964. Ecology of sponges in Hatteras Harbor, North Carolina. Ecology, 45 : 752-767. WILSON', H. V., 1891. Notes on the development of some sponges. /. Morphol., 5: 511-519. WILSON, H. V., 1894. Observations on the gemmule and egg development of marine sponges. /. Morphol., 9 : 277-406. Reference: Biol. Bull., 147: 352-368. (October, 1974) HISTOCHEMICAL OBSERVATIONS OX THE LOCALIZATION OF SOME ENZYMES ASSOCIATED WITH DIGESTION IN FOIK SPECIES OF BRAZILIAN NEMERTEANS RAY GIBSON Dcpartamcnto de Zoologia, Unh-crsidade de Sao Paulo, Cai.ra Postal 8105, Sao Paulo, Brasil and Department of Biohnjy, Liverpool Polytechnic. Byrom Street, Liverpool L3 3AF, England The processes of digestion have been investigated histochemically for several nemertean species (Jennings, 1962a; Gibson and Jennings, 1969; Jennings and Gibson, 1969; Gibson, 1970), and results so far obtained suggest that although the fundamental digestive sequence is essentially similar for all the species, differences in details can be related either to their systematic position or mode of life. Irrespective of the nature of the food utilized, digestion occurs in two distinct phases. Initial extracellular digestion is accomplished in the intestinal lumen at an acidic pH and involves mainly proteolytic enzymes secreted by the gastrodermis. Food particles are subsequently engulfed by lamellar outgrowths of the distal ciliated cell walls (Jennings. 1969) and food vacuoles passed back into the gastro- dermis for the second, intracellular. stage in digestion. This involves exopeptidases (arylamidases), acid and alkaline phosphatases and. in some species at least, carbohydrases and lipases. The most uniform patterns of digestion are found in anoplan nemerteans (Jennings, 1962a; Jennings and Gibson, 1969). Acidophilic gland cells in the foregut, rich in carbonic anhyclrase, discharge their contents during ingestion which serve both to kill prey taken alive and provide the correct lumenar pH for extracellular digestion. Mucoid secretions, discharged from other foregut glands, lubricate the food as it is passed into the intestine. Food entering the intestinal lumen stimulates the gastrodermal gland cells to discharge endopeptidases. which are responsible for early proteolysis and function optimally under acidic conditions. Intracellular digestion initially involves endopeptidases phagocytosed along with food particles and is marked by a sharp increase in acid phosphatase activity in and around food vacuoles. Acid phosphatases may be concerned in some way with the maintenance of the correct pH for intralumenar endopeptic activity (Jennings and Gibson, 1969) or with food vacuole formation (Rosenbaum and Rolon, 1960), Slinger and Gibson (1974) demonstrating biochemically that these enzymes func- tion optimally in the range pH 4.1-5.0, depending upon the species. A few hours after the commencement of phagocytosis acid phosphatase and endopeptidase activity in the food vacuoles and surrounding cytoplasm declines, being replaced by alkaline phosphatases and exopeptidases. This final, alkaline, phase in digestion operates within the pH range 8.7-10.1 (Slinger and Gibson. 1974), and continues until digestion has been completed. Endopeptidase enzymes can be demonstrated in the gastrodermal gland cells at all times, irrespective of the nutritive state, but exopeptidase activity can only be visualized histochemically in gut cells at the appropriate stage in digestion. 352 NEMERTEAN GUT ENZYMES 353 This contrasts markedly with blood system exopeptidases, which can be detected at all times (Gibson and Jennings, 1967). Far greater variation in digestive physiology is found amongst the Enopla. At least two distinct types of foregut physiology have been demonstrated, one resembling the anoplan pattern and involving carbonic anhydrase production (Prostomd), the other achieving an intralumenar acidic pH via some other mechanism and not possessing demonstrable carbonic anhydrase (Amphiporus, Paranemertes, Tetrastemma) (Jennings and Gibson, 1969; Gibson, 1970). Differences between the two classes are also found in the gastrodermal physiol- ogy ; no enoplan species has yet been recorded with endopeptidase enzymes located in its gastrodermal gland cells, the enzymes instead being synthesized in and se- creted from spherical inclusions housed within the columnar cells. The nature of the enzymes secreted by hoplonemertean gastrodermal glands has not yet been determined, but it is supposed that since the animals are carnivorous the secre- tions are proteolytic in form. Other enzymes involved in digestion essentially follow the pattern outlined for the Anopla. A major departure from the usual type of digestive physiology is found in the bdellonemertean Malacobdella (Gibson and Jennings, 1969), where both mor- phological and physiological characters of the gut are considerably altered. These modifications, however, such as the replacement of endopeptidases by a-amylase- like carbohydrases as the principal enzymic group in digestion, and the total absence from the gastrodermis of demonstrable exopeptidases, can be entirely related to the species' atypical way of life and unselective microphagous feeding habits. In the present study species of nemerteans belonging to families not previously investigated have been examined to determine whether or not existing known pat- terns of digestive physiology are evident, or whether additional variations could be detected. MATERIALS AND METHODS The species of nemerteans investigated, listed systematically, were ANOPLA Order : HETERONEMERTEA Baseodiscus delineatns (Delle Chiaje) ENOPLA Order : HOPLONEMERTEA Ototyphlonemertes affinis Kirsteuer, Ms. name Ototyphlonemertes crncba Correa Ototyphlonemertes lactca Correa RAY GIBSON Ototyphlonemertes affinis is soon to be described as a new species by Dr. Ernst Kirsteuer, The American Museum of Natural History, New York (Morphology, taxonomy and ecology of the nemertean genus Ototyphlonemertes, with special reference to the American species — in preparation ) . Specimens of Baseodiscits were collected from beneath stones and boulders in the intertidal zones of shores at Ubatuba, Sao Sebastiao and Praia de Siriuba, on the coast of Sao Paulo State, Brazil. The three Ototyphlonemertes species, all psammobiontic or interstitial forms (Kirsteuer, 1967, 1971), were obtained from a sandy beach at Ilhabela on the Island of Sao Sebastiao, some 100 km east of Santos, Brazil. Samples of sand were "panned" in the manner employed by Correa (1958), fresh fish meat being used to bait the sand surface for about ten minutes before panning was carried out. The living worms were maintained in the laboratory in frequently changed, cool sea water. A selection of associated fauna and artificial foods was tested in attempts to set up a feeding series, but under laboratory conditions none of the species were ever observed to feed. Results, therefore, are based upon histolog- ical and histochemical evidence obtained from different animals fixed at progressive time intervals after collection. In the case of the Ototyphlonemertes species, although responding to the first bait and often showing signs of having fed on it, a definite feeding series could not be established because there was no way of knowing when the animals' last meal had been prior to accepting the bait. Histological observations were made on paraffin wax (56° C mp) sections of specimens fixed in marine Bouin and stained either by Mallory's trichrome or the \% aqueous Alcian blue methods. The nature and location of enzymes was investigated in animals fixed for 2—1- hr at 4° C in 10% buffered formalin, pH 7.0, washed in chilled distilled water and frozen-sectioned at 10-12 /j. on an International Equipment Co. Microtome-Cryostat Model CTF. Sections were air-dried on clean slides and rinsed in cold absolute acetone before incubation for enzyme visualization. The following methods were used to investigate enzymes present : the Hausler (1958) technique for carbonic anhydrase ; the indoxyl acetate (Holt, 1958) and a-naphthyl acetate (Gomori, 1952) methods for non-specific esterases ; the Bur- stone and Folk (1956) L-leucyl-/^-naphthylamide method for exopeptidases (arylamidases) ; the Burstone (1958) azo-dye method for acid phosphatases, with naphthyl AS-TR phosphate as substrate and Red-violet LB salt as simultaneous coupler; the Gomori (1952) Tween method for lipases ; and the Gomori (1939) calcium salt method for alkaline phosphatases, with sodium /}-glycerophosphate as substrate. Controls for these histochemical methods included the use of heat inactivated sections and incubation media from which the specific substrates were omitted. In addition, fat deposits were studied in paraffin wax sections of animals fixed in Flemming's osmium tetroxide fluid, and the occurrence of glycogen reserves investigated in sections after fixation in 909r alcohol containing 1 */r picric acid and stained bv the Best's carmine method. NEMERTEAN GUT ENZYMES 355 OBSERVATIONS ANOPLA Order : HETERONEMERTEA Baseodiscus delineatiis Structure of flic gut. The gut is divisible into two principal regions, the foregut and intestine, both of which are lined by an epithelium formed from glandular and ciliated columnar cells. By far the shorter of the twro alimentary regions, the foregut can be differentiated from the intestine by its mucus-secreting gland cells and shorter, more densely arranged, cilia. The mouth, only 1 mm or less in length, opens into a buccal cavity with walls which are deeply folded into longitudinal ridges. These ridges have been reported from other Baseodiscus and heteronemertean species and are interpreted as per- mitting the dilation of the buccal cavity for the ingestion of a large-sized meal (Jennings and Gibson, 1969; Gibson, 1974). The buccal and foregut epithelium are histologically identical but differ in their thickness. \Yhere buccal folding is most prominent the epithelium may be up to 300-350 p. tall whereas further back in the foregut it decreases to a height of only 200-250 p. Cilia of the columnar cells in both buccal cavity and foregut are densely arranged and 6-8 p, long. There appear to be several types of gland cells in the anterior gut epithelium, although two major varieties can be distinguished. These are: (1) Elongate or pyriform acidophilic glands filled with numerous closely packed minute spheres 1 p, or less in diameter. Other, similarly shaped, glands with more loosely arranged contents and less obvious acidophilic affinities, may represent gland cells of the same type but in a different physiological state. The glands are negative to Alcian blue; (2) Irregularly-shaped gland cells filled with a coarsely granular cytoplasm, approximately twice as numerous as the acidophils. Many of these glands stain with the Alcian blue method for mucopolysaccharides, the intensity of staining varying from faint to deep. Large numbers of gland cells of both groups can also be found in the paren- chyma underlying the foregut and buccal epithelium, discharging their contents through the gut wall into the lumen. A similar situation has been found in lineid heteronemerteans (Jennings, 1960, 1962a; Jennings and Gibson, 1969). The intestine is typically anoplan in form and for most of its length bears serially repeated lateral diverticula. The intestinal epithelium, or gastrodermis, is identical in both the main canal and the lateral diverticula. It consists of ciliated columnar cells, up to 150 p. tall and 6-8 p. wide and with sparsely distributed cilia 10-12 p. long, interspersed with rather smaller pyriform gland cells containing oval or spherical acidophilic globules which show a strong positive reaction to methods for the demonstration of esterases (Fig. 1). The gland cells lie proximally in the gastrodermis, tracts of discharging globules reaching up to the intestinal lumen between the columnar cells. Towards the posterior of the body there is a gradual reduction both in the size and number of lateral diverticula and in the gastrodermal height. Imme- diately before the anus the columnar cells are no more than about 50-60 p. tall. 356 RAY GIBSON The density of gastrodermal gland cells similarly decreases posteriorly and they are completely absent from the short "rectal" region, 3.5 mm long in an animal 27 cm in length. The anus opens at the extreme posterior tip of the hody. Enzymes of the gut. Carbonic anhydrase activity has been previously reported from the foregut acidophil glands of both palaeo- and heteronemerteans (Jen- nings, 1962a ; Jennings and Gibson. 1969), where it is believed to be concerned in the initiation of the correct intralumenar pH for the subsequent extracellular phases of proteolysis. No evidence of carbonic anhydrase activity could be demon- strated in the foregut or any other tissue in Bascodiscns del in eat us. The Hausler method does, however, stain many elongate ovoid bodies located in the epidermis. These bodies, which are equally strongly stained in heat- denatured control sections, are of similar size and shape to the epidermal rhabdite- like cells and the reaction in them may be due to some calcium or other salt component. Intense esterase activity, demonstrable with both the a-naphthyl acetate and indoxyl acetate techniques, was consistently found in some 10-12% of the buccal and foregut acidophilic gland cells (Fig. 2). Esterases have not previously been recorded from these sites in any nemertean species. In the intestine the pyriform gastrodermal glands at all times stain strongly for esterase activity (Fig. 1). Although it was not possible to utilize selective inhibitors and activators in conjunction with the indoxyl acetate method, as employed by Hess and Pearse (1958), it seems likely that the gland cell activity represents that of cathepsin C-type proteases, as recorded from other anoplan forms (Jennings, 1962a; Jennings and Gibson, 1969). There is no reason to suppose that intestinal endopeptidases are absent from this species, and both histochemical methods employed are known to react positively at sites of endopeptic activity in other nemerteans. Sections of animals prepared eight or more hours after collection showed that numerous food vacuoles and their surrounding columnar cell cytoplasm stained positively for esterase activity. The intensity of the colored reaction product varied between specimens ; in some examples only a weak activity could be seen, confined mainly to the proximal regions of the columnar cells, but in others the activity was distributed throughout the gastrodermis and appeared very much stronger. Arylamidases (exopeptidases) of the "leucine aminopeptidase" type were visual- ized in vacuolar and cytoplasmic regions of the gastrodermal columnar cells in only some of the specimens investigated (Fig. 3). When present the reaction, in general, appeared to be more intense in the posterior half of the intestine. Maximum arylamidase activity in food vacuoles occurred in animals with a weak vacuolar esterase reaction. Xemertean gastrodermal arylamidases are known to be histochemically demonstrable only during the appropriate phase in the digestive sequence and recent work by Slinger (1974) has shown that whole-animal aryl- amidase extracts function optimally on the alkaline side of neutrality. In the gut acid phosphatase activity was confined to the intestinal columnar cells, in and around food vacuoles (Fig. 4). Its degree of activity and intracellular distribution closely parallelled that shown by the gastrodermal esterases. NEMERTEAN GUT ENZYMES 357 Alkaline phosphatase activity in the intestine (Fig. 5) is similar to that of arylamidases, its maximum visualization in cytoplasm and food vacuoles occurring at the same time. The link between these two enzyme groups has already been clearly established in nemertean worms (Jennings, 1962a; Jennings and Gibson, 1969; Gibson, 1970), and both succeed esterase and acid phosphatase demonstra- tion in the digestive sequence. At times when arylamidase activity can not be shown in the gut, alkaline phos- phatases are confined to the distal border of the gastrodermis, forming a narrow but intensely staining zone of activity a few microns deep. The occurrence of alkaline phosphatases at this site, where they are believed to be independent of the nutritive state, has been found in other heteronemertean species (Jennings, 1962a ; Jennings and Gibson, 1969). The Tween method for the demonstration of lipases has not generally proved to be sensitive enough with nemertean tissues, unless the animals are maintained on a high fat content diet (Jennings, 1962a). In some of the Baseodiscus speci- mens studied a faint but definite brownish reaction, absent from control slides, could with prolonged incubation be distinguished in several food vacuoles. This activity, occurring at the same time as peak arylamidase visualization, may represent the occurrence of lipolytic enzymes. No other enzymes were demonstrated in any part of the alimentary system. Food reserves. Fat deposits are principally restricted to the distal half of the gastrodermal columnar cells, occurring as droplets of varying diameter from 5—6 /A downwards. Glycogen, appearing as small granules scattered throughout the gastro- dermis, can also be found in lesser quantities in freshly collected animals. Enzymes of other tissues. Several enzymes were demonstrated at sites of activ- ity other than the gut. Arylamidase activity was consistently found in the blood system endothelium, although the intensity of the staining reaction varied from weak to strongly posi- tive depending upon the part of the body and individual animal. The reaction was generally strongest in the lateral blood vessels where they ran alongside the in- testine. Blood system arylamidases are well known from nemerteans (Gibson and Jennings, 1967). Esterase activity was localized in many tissues of the body, including the epi- dermis (Fig. 6), ciliated cerebral canal and some parts of the body wall and proboscis musculature. At these sites the degree of activity was normally weak to moderate, although the proximal epidermal regions were often intensely stained. Consistently strong esterase activity was found in ganglionic cells of the principal nerve tracts, especially around the fibrous core of the cerebral ganglia and lateral nerve cords (Figs. 6 and 7). No instance was observed of any esterase or other enzymic activity in the fibrillar component of the nervous system, and it is prob- able that the esterases demonstrated are in fact of the cholinesterase type, as found in the nervous system of other nemertean species (Ling, 1969a, 1969b). Esterase activity could, in some examples, also be localized in certain gland cells of the proboscis epithelium. Alkaline phosphatases were irregularly found at several sites. Strong activity was seen in the nephridial ducts and cerebral canal, weak to moderate activity showed in the gland cell region of the cerebral organs. Other cells of the cerebral 358 RAY GIBSON FIGURE 1. Bascodiscus dclincatiis; Longitudinal section through part of the gastrodermis to show acidophilic gland cells staining intensely with the Holt indoxyl acetate method for esterases (proteases), scale = 105 /M. FIGURE 2. Bascodiscus dclincatiis; Longitudinal section through a portion of the foregut epithelium to show esterase activity localized in some of the acidophilic gland cells ; Gomori a-naphthyl acetate method, scale = 70 /*. FIGURE 3. Bascodiscus dclincatiis; Transverse section through a part of the gastrodermis, showing arylamidase activity (black) generally distributed throughout most of the columnar cells ; Burstone and Folk method, scale = 150 p. FIGURE 4. Baseodiscus dclincatiis; A part of the intestine showing acid phosphatase activ- ity distributed through the distal regions of the gastrodermis; Burstone azo-dye method, scale= 130 /j.. NEMERTEAN GUT ENZYMES 359 organs on occasion reacted faintly to the acid phosphatase technique. Traces of these enzymes were also found in the proximal regions of the epidermis, but the reaction was always extremely weak and, although absent from control slides, not positively identified. No other sites of enzymic activity were recorded in Baseodiscus delineatus, ENOPLA Order: HOPLONEMERTEA Ototyphlonemertes affinis, O. erneba and 0. lactca Structure of the gut. The gut structure of two of the three species has been described by Correa (1950, 1954) and all are extremely similar. In common with other monostiliferous hoplonemerteans, the rhynchocoel and buccal cavity of Ototyphlonemertes share a common anterior aperture, the rhynchodaeal pore. This opens into the rhynchodaeum from which the rhynchocoel leads dorsally and the oesophagus ventrally. Three regions of the gut can be recognized histologically, the oesophagus, the stomach and intestine. In the Brazilian Ototyphlonemertes the intestine lacks a caecum, the stomach therefore opening directly into it without the intervention of a pyloric tube such as is found in many hoplonemerteans. The oesophagus opens from the rhynchodaeum in front of the cerebral ganglia and extends posteriorly below the brain as a short straight ciliated tube devoid of gland cells. It opens directly into the stomach, which consists of a folded epithelium formed from densely ciliated columnar or cuboidal and large numbers of gland cells. The epithelial height differs between the species; in O. lactea the stomach lining is only 10-12 ^ tall compared with 25 ju, in O. affinis and 30-35 p in 0. erneba. Most of the gland cells contain a finely granular to fibrillar basophilic secretion, but isolated acidophilic glands filled with homogeneous contents are irreg- ularly distributed between the basophils. Stomach columnar cell cilia in all three species are densely arranged and 4—4.5 p long. Posteriorly the stomach opens directly into the intestine with only a slight narrowing of its lumen, unlike the situation depicted for 0. lactea by Correa (1954; Plate 7, fig. 30). The gastrodermis is very similar in structure to that described for other hop- lonemerteans, consisting of acidophilic gland cells interspersed with sparsely ciliated columnar cells. The distribution of gland cells is more or less uniform throughout the intestinal length in 0. affinis and 0. erneba, but in O. lactea the anterior half contains many more gland cells than the posterior, the ratio being approximately 3: 1. Correa (1954) notes the anterior aggregation of gland cells FIGURE 5. Baseodiscus delineatus; Longitudinal section through part of the intestine to show the cytoplasmic localization of alkaline phosphatase activity ; Gomori calcium salt method, scale = 83 ft. FIGURE 6. Baseodiscus delineatus; Longitudinal section to show intense esterase activity present in the proximal epidermal regions and ganglionic cells of a lateral nerve cord; Gomori a-naphthyl acetate method, scale = 270 n. FIGURE 7. Baseodiscus delineatus; Transverse section through a lateral nerve cord, showing esterase activity confined to the ganglionic nerve sheath ; Holt indoxyl acetate method, scale = 67 /j.. 360 RAY GIBSON in 0. lac tea, and in this and other species calls them erythrophil glandular cells. The glands are filled with acidophilic spheres 1 //. or less in diameter, similar sized or slightly larger globules also occurring in the columnar cells, particularly in the proximal half. Columnar cell inclusions are similar in appearance to those described from other hoplonemerteans (Jennings and Gibson, 1969; Gibson, 1970) which contain endopeptidases. The cilia of the gastrodermis are 7-8 /* long, the overall gastrodermal height some 45-50 ^ in all three species. Correa (1948) contrasts the low epithelial height of the stomach with the high gastrodermal development in other species (0. brevis and 0. evdinac) without mentioning cellular dimensions. In the animals used during the present investigation the con- trast between stomach and intestinal height depends upon the species concerned and is least in O. erneba. As Correa (1950, 1954) reported, the "rectal" region of the intestine is not dilated and can be distinguished from the more anterior intestinal regions by the absence of gland cells. The region also lacks diverticula. but throughout the in- testinal length these are at best only poorly developed. Enzymes of the gut. No evidence of carbonic anhydrase activity could be found in the gut or any other tissue of the body. Strong esterase activity was found in the gastrodermal acidophilic glands (Fig. 8), demonstrable by both histochemical methods employed. The individual globules filling the glands stained intensely in most of the animals studied, but in others could not be distinguished. Jennings (1962a) found that in the hetero- nemertean Linens rnber gastrodermal gland cells which had discharged their spheres shortly after feeding failed to react to techniques for the visualization of cathepsin C-type endopeptidases. It thus seems probable that in those Ototyphlo- nemertes specimens with negative gland cell reactions a similar situation is pre- vailing, an inference supported by the fact that this was only seen in animals fixed soon after collection. No other hoplonemertean species yet investigated has possessed histochemical ly demonstrable enzymes in its gastrodermal gland cells (Jennings and Gibson, 8 '** * * FIGURE 4. An electron micrograph of the nerve trunk surrounded by a discontinous coat of epineural muscle (e), amobocytes (a) and glial cells (g) embedded in collagen. The radular protractor muscle cells (r) are also indicated; scale = 2.0 microns. INNERVATION OF RADULAR PROTRACTOR MUSCLE 375 Nerves and neuromuscular junctions in the radular protractor muscles: A branch of nerve 7, along with a branch of the blood vessel enter the proximal por- tion (with regard to the CNS) of the radular protractor muscle which is attached to the odontophore cartilage. The main trunk of the branch of nerve 7 runs parallel to the central artery in the mid part of the muscle (Figure 2). As the nerve proceeds anteriorly towards the attachment of the muscle to the radular sac, it subdivides to form smaller trunks. These smaller trunks continue to subdivide until finally single axons are seen among the muscle cells, where neuromuscular junctions are formed. Occasionally, a single nerve ending is observed forming a junction with two adjacent cells (Fig. 3) (Graziadei 1966). The main branch of nerve 7 in the radular protractor muscle is surrounded by a discontinuous layer of amoebocytes, glial and epineural muscle cells (Figure 4). These cells are embedded in a thick collagen matrix. The collagen matrix appears to be divided into two unequal layers. The adjacent layer (about 2500 A thick) is composed of collagenous fibers which for the most part wind circumferentially around the nerve trunk. The outermost layer (about 8 microns in width) is com- posed of mostly collagenous fibers which are arranged parallel to the long axis of the nerve trunk and epineural muscle cells. The cells embedded in the collagenous neural sheath do not form any continuous barrier to separate the axons from the extracellular space of the radula muscle. The main nerve trunk is about 40 microns in diameter and contains several hundred axons ranging in size from less than a micron to about 8 microns in cross section. The axons contain various arravs of microtubules and neurofilaments. In transverse section, the j larger axons are rather irregular in outline and are generally found close to the surface of the trunk. The axons are associated with cells termed sheath cells (Rogers, 1968) or glial cells (Amoroso, Baxter, Chiquoine and Nisbet, 1964; Nicaise, Pavans de Ceccatty and Baleydier, 1968; McKenna and Rosenbluth, 1973). Hereafter in accordance with the nomenclature of Amoroso, Baxter, Chiquoine and Nisbet, 1964, these cells will be referred to as glial cells (Rogers, 1969). In the lamellar extensions of the glial cells are large membrane-bound inclusions which vary greatly in size, density and shape (Fig. 5). The inclusions ranged in diameter from 0.3 to 0.8 microns. These glial bodies are comparable to those described by Barrantes (1970) in the Agentinian slug (Vaginula soleiformis') . The lamellar extension of the glial cell with its characteristic large membrane-bound vesicles is frequently located near neuromuscular endings. As also noted by Rogers (1969), processes of glial cells not associated with axons are often encountered between the muscle cells (Fig. 4). Smaller subdivisions of the main nerve trunk are encountered away from the central portion of the muscle bundle (Fig. 5). These smaller trunks (less than 10 microns in diameter) are no longer surrounded by a coat of epineural muscle. The thickness of the collagenous coat and the number of axons in these smaller trunks are also reduced. The maximum diameter of the largest axon is also smaller ; generally none larger than one micron. Figure 5 is an illustration of a small nerve branch about 5 microns in diameter which contains about seventy axons. In large and small nerve bundles granular and agranular vesicles (similar to synaptic vesicles in neuromuscular endings) are observed in various axons (Figs. 5, 6). In these nerve bundles, an axon appears to contain only one type of vesicle. However, both granular and agranular synaptic vesicles are observed in the same 376 R. B. HILL AND J. W. SANGER FIGURE 5. A cross section of a smaller nerve trunk among the radular protractor muscle cells. Note the absence of the epineural muscle cells and the meandering process of a glial cell (g) ; scale = 0.5 micron. INNERVATION OF RADULAR PROTRACTOR MUSCLE 377 FIGURE 6. Many of the axons in this small nerve trunk contain uniform populations of synaptic vesicles. A nerve ending (arrow) is surrounded by a process of the muscle cell ; scale — 1.0 micron. R. B. HILL AND J. W. SANGER . •«•» •.' ,." -^ FIGURE 7. The initimate juxtaposition of the nerve ending and muscle cell is illustrated here. A sarcolemniic tubule is also clearly pictured (arrow) ; scale =1.0 micron. neuromuscular junction (Figs. 7, 8, 9). The muscle cell and nerve are separated by a gap of about 150-200 A. The muscle is intimately associated with the nerve ending but there is no specialized membrane involution in the junctional area such ,-r ky# J-v FIGURE 8. A nerve ending containing a mixture of dense and clear synaptic vesicles. The glial cell (g) process contains many large granules; scale = 1.0 micron. INNERVATION OF RADULAR PROTRACTOR MUSCLE 379 _ ^^^$k KMM ' *V. * !.-«W f'r^ . ' * • .." ;»v. --/ ' • . .--ww?».r3^.' ^j %* ^!v 1**^ © FIGURE 9. A cross section of a. nerve ending with a. mixture of dense and clear vesicles ; scale = 0.5 micron. as is observed in vertebrate fast twitch skeletal muscle. The association of nerve and muscle is comparable to that observed in vertebrate and molluscan smooth muscle (McKenna and Rosenbhith, 1973). Frequently, part of the nerve ending is ensheathed by a process of the muscle cell it is innervating- (Figure 6). There is a decided segregation in the size of the synaptic vesicles. The agranular vesicles ranged in size from 500 to 1000 A with a mean diameter of 630 A (based on the measurement of 400 vesicles). The granular vesicles ranged in size from 800 to 1350 A with a mean diameter of 970 A (based on a count of 200 vesicles). The granular vesicles varied in electron density. Some vesicles were completely filled with dense material while others were only partially filled. Nevertheless, both types of granular vesicles had about the same diameter, 800-1350 A. The two extremes of neuromuscular junctions are illustrated in Figures 10 and 11. Figure 10 indicates a nerve ending where less than ten per cent of the synaptic vesicles are granular while Figure 1 1 illustrates an ending where ninety per cent of the synaptic vesicles are granular. Nerve endings with intermediate proportions of granular and agranular vesicles are also observed. DISCUSSION A good deal of past attention has been devoted to the function of buccal muscles in the feeding cycle of prosobranch gastropods, and some heat has been generated by past controversies which perhaps did not take sufficient account of the extreme diversity to be found in the anatomy of these muscles in this group. A definitive 380 R. B. HILL AND J. W. SANGER FIGURE 10. A nerve ending where many of the vesicles are clear. Note the intimate association of the glial cell (g) and nerve ending; scale = 0.5 micron. account of the feeding cycle is available in Chapter 8 of British Prosobranch Molluscs (Fretter and Graham, 1962). There are apparent homologies among the muscles in a number of genera, but there are also profound modifications linked with modifications in use of the radula. The modus operandi of the whole odontophoral mechanism of Busy con was described by Herrick (1906). As we re-examined the mechanism we felt the need for a redescription of the innervation, which was only one facet of Herrick's work. This was mainly because the responses of the muscles to stimulation of the nerves did not altogether correspond to those to be expected from Herrick's description. We now feel the discrepancies may principally be attributed to the anastomoses described above. Otherwise we have only improved in some details on Herrick's description, but we have provided an integrated picture of the whole buccal innervation. Hoyle (1964) noted in a review some ten years ago the scarcity of work reported on the ultrastructure of molluscan nerve endings. In the ensuing decade much less than a score of papers have been devoted to this subject (see summary of references in Heyer, Kater and Karlsson, 1973). Even fewer of these studies have been done on systems which are also being investigatd by physiological and pharmacological experiments. The radular protractor muscle of Busycon canaliculatum is being investigated in three different ways : physiologically, phar- macologically and ultrastructurally (Hill, 1958; Hill, 1970; Hill, Greenberg, Irisawa and Nomura 1970; Sanger'and Hill, 1972; Sanger and Hill, 1973a, 1973b ; INNERVATION OF RADULAR PROTRACTOR MUSCLE 381 Sanger, 1973). Our intention has been to combine the results from these three avenues of experimentation to study regulation of the contractile properties of the smooth muscle at several levels. The radular protractor muscle is of particular interest because it can be induced to show rhythmicity in the presence of acetyl- choline and tryptamine (Hill, 1958) and perhaps could be used as a model of cardiac rhythmicity. The two extreme types of nerve endings observed in this present study appear very similar to serotoninergic nerve endings (containing mostly granular vesicles) and to cholinergic (containing mostly clear vesicles) nerve endings. The nerve endings containing mostly clear vesicles are like those of cholinergic nerve endings in vertebrate smooth muscle (Richardson, 1964). Even there, occasional dense synaptic vesicles were observed by Richardson among the many clear vesicles. These clear or agranular synaptic vesicles might be acetylcholine storage sites while the dense or granular vesicle may be 5-hydroxytryptamine storage sites. Welsh and Moorhead (1959), using a fluorescence assay method, identified the presence of 5-hydroxytryptamine in the radular muscles of Busycon canaliciilatum. The concentration was rather low (0.09 /^g/g) compared to that in the pooled ganglia (9.2 //.g/g), and the origin may well have been from vesicles in the nervous tissue of the radular muscles. This is, of course, speculation since the radular protractor muscle has not yet been studied by subcellular fractionation or by the use of the Falck method. Acetylcholine has not been identified chemically in radular muscles or nerves, although a number of authors have found acetyl- choline-equivalent by bioassay in gastropod cardiac muscle (summarized by Welsh, 1956). The immediate interest of the observation that the nerve endings may have cholinergic and aminergic vesicles lies in the fact that a mixture of opposing syn- thetic neurohumors, acetylcholine (ACh) and serotonin (5-HT), induce strong maintained rhythmicity in the radular protractor (Hill et al., 1970). Acetyl- choline depolarizes and induces a contracture, which is not a "catch" since the muscle relaxes along with the repolarization which follows washing out of the acetylcholine. However, if the muscle is treated with tryptamine while still de- polarized writh acetylcholine, repolarization is gradual and accompanied by a regular oscillation which induces mechanical rhythmicity. The phenomenon is. being investigated, with the objective of discovering the mechanism of the rhythmic- ity but it may have nothing to do with the natural rhythmic rasping function of the radula, since to all appearances the muscles of the buccal apparatus function in a phasic non-spontaneous motor unit organization (Herrick, 1906). However,, it remains possible that acetylcholine may be the excitatory neurotransmitter. Whert the radular protractor is driven by stimulation of its nerve (Hill, 1962) 10~8 M ACh may increase amplitude of twitches by one-third. Michael J. Greenbergv (Florida State University) and his students in the Experimental Invertebrate Zoology Course (MBL) recently found that ACh antagonist blocked twitches caused by stimuli to the nerve (M. J. Greenberg, personal communication). ACh: antagonists also block ACh-induced contraction (Hill, 1970). However, tryptamine or 5-HT also increases amplitude of the twitches when the muscle is driven by- stimulating its nerve, with a threshold around 10~8 M. Furthermore, high concen- trations of 5HT (10~3 M) predispose the muscle to rhythmicity, in the sense that 382 R. B. HILL AND J. W. SANGER •'•'• -"-•••'-••' ;";- '•'•'• --*;-:-:- •--'-'•' -'•'' . -—pry -J , F r */'. v^^'*^ .36 - .." -^ .••••-••=• •- •• • .• ^ -N) ' K Jp • . V- ^M» . * * , t \ - .^^^^^ FIGURE 11. Two nerve endings where most of the vesicles contain dense material. The scale in the large picture equals 1.0 micron. The insert is a higher magnification (scale = 0.5 INNERVATION OF RADULAR PROTRACTOR MUSCLE 383 subsequent stimulation of the nerve or directly of the muscle evokes oscillatory responses, and subsequent stimulation with acetylcholine or KC1 may evoke a contracture with superimposed rhythmicity. Thus if the vesicles do indeed contain ACh and 5HT it remains possible that there may be two types of excitatory synapses The variation in types of synaptic vesicles in nerve endings reported here in the radular protractor muscle, a phasic muscle, is similar to that reported in a "catch" muscle, the anterior byssus retractor muscle (ABRM) (McKenna and Rosenbluth, 1973). Several workers have proposed the existence of two types of nerve endings in that catch muscle. One ending is cholinergic and that nerve ending produces depolarization and contraction, by the release of acetylcholine, which leads to the "catch" state (i.e., the persistent contraction after the muscle has repolarized). A second type of ending is serotoninergic (containing dense vesicles) and by the release of serotonin immediately relaxes the muscle from the "catch" state (Twarog, 1967). In contrast, the radular protractor muscle does not demonstrate any "catch" phenomena — yet it has a similar display of nerve endings as in the "catch" muscle (McKenna and Rosenbluth, 1973). These observations demon- strate a similar ultrastructure of the neuromuscular endings of two dissimilar types of muscles, a "catch" muscle and a phasic muscle. As indicated by McKenna and Rosenbluth (1973) the mixing of both granular and agranular synaptic vesicles in nerve endings is observed rather frequently in invertebrates. The presence of both types of synaptic vesicles in the same nerve endings has been reported in several other molluscan species (Barrantes, 1970; Dougan and McClean, 1970; Heyer, Kater and Karlsson, 1973). This mixing of different types of vesicles in the same nerve ending raises further questions about their identification and function. Can nerve endings have more than one neuro- transmitter? Are the vesicles in the small axons identical to those in the nerve endings? Are some of the clear vesicles in the nerve endings just depleted dense vesicles ? Further work-involving serial sections of the nerve endings and small nerve trunks as well as the isolation and identification of the various granules- is certainly needed to elucidate these questions. We wish to particularly acknowledge collaboration by Mrs. Else Froberg and Dr. Kiyoaki Kuwasawa. The diagram of the innervation is based on dissections by R. B. Hill, D. Spring, L. Vargish, F. Froberg and K. Kuwasawa. The drawing by G. DeVry is based on sketches by E. Froberg, K. Kuwasawa and M. Parmenter. The investigation was supported by research grant NS 08352 from the National Institute of Neurological Diseases and Stroke and by the National Science Founda- tion (GB1001 and GB5598) (R.B.H.). Research for this publication was also supported by NIH Grant No HL 15835 to the Pennsylvania Muscle Institute and the National Science Foundation (IG-73-3) (J.W.S.). Please send reprint re- quests to J.W.S. micron) of the upper nerve ending. The change in orientation in the insert is indicated by the letter "L." 384 R. B. HILL AND J. W. SANGER SUMMARY 1. A detailed and integrated picture of the whole buccal inner vation of Busy con canaliculatuni is presented. 2. The results of our observations on the fine structure of the nerves and of the neuromuscular junctions in the radular protractor are reported and discussed. Each radular protractor muscle is innervated by a nerve arising from the cerebro- buccal connective. The nerve trunk enters the proximal end of the muscle and runs parallel to the long axis of the muscle bundle. A layer of connective tissue and epineural muscle cells surrounds the trunk. Subdivisions of the main nerve trunk branch laterally into the muscle bundle losing their epineural muscle coat. Further subdivisions of the branches produce single axons which can be observed among the muscle cells, but no specialized motor nerve endings were observed. Within the nerve endings are two types of synaptic vesicles: agranular (clear) and granular (dense). The granular vesicles are larger, ranging in diameter from 800 to 1350 A (mean 970 A). The clear vesicles vary in diameter from 500 to 1000 A (mean 630 A). The ratio of agranular to granular vesicles within a single nerve ending varies widely. When one type of vesicle predominates in an ending, then that ending comes to have a resembance to a cholinergic or to a serotoninergic nerve ending. LITERATURE CITED AMOROSO, E. C, M. I. BAXTER, A. D. J. CHIQUOINE AND R. H. NISBET, 1964. The fine struc- ture of neurones and other elements in the nervous system of Archachatina inarginata. Proc. Roy. Soc., London, Scries B, 160 : 167-180. BARRANTES, F. J., 1970. The neuromuscular junctions of a pulmonate mollusc I. Ultra- structural study. Z. Zellforsch, 104 : 205-212. DOUGAN, D. F. H., AND J. R. McCLEAN, 1970. Evidence for the presence of dopaminergic nerves and receptors in the intestine of a mollusc. Tapes ivatlingi. Com p. Gen. Pharmac., 1 : 33-46. ESTABLE-PUIG, J. F., W. C. BAUER AND J. M. BujMBERG, 1965. Paraplienylene-diamine stain- ing of osmium-fixed plastic embedded tissue for light and phase microscopy. /. Neuropath. Exp. N enrol., 24: 531-535. FRETTER, V., AND A. GRAHAM, 1962. British Prosobranch Molluscs. Their Functional Anat- omy and Ecology. Ray Society, Vol. 144. Bernard Quaritch Ltd., London. GRAZIADEL, P., 1966. The ultrastructure of the motor nerve endings in the muscles of cephalo- pods. /. Ultrastnic Res., 15 : 1-13. HERRICK, J. C., 1906. Mechanism of the odontophoral apparatus in Sycotypus canaliculatus. Amer. Natur., 40: 707-737. HEYER, C. B., S. B. KATER AND U. L. KARLSSON, 1973. Neuromuscular systems in molluscs. Amer. Zool, 13 : 247-270. HILL, R. B., 1958. The effects of certain neurohumors and other drugs on the ventricle and radula protractor of Busvcon canaliculatuni and on the ventrile of Strombus gigas. Biol.BulL, 115: 471-482. HILL, R. B., 1962. Pharmacology of the radular protractor of Busycon canaliculatuni. Bio!. Bull., 123 : 499. HILL, R. B., 1970. Effects of postulated neurohumoral transmitters on the isolated radular protractor of Busycon canaliculatum. Comp. Biochem. Physiol., 33 : 249-258. HILL, R. B., M. J. GREENBERG, H. IRISAWA AND H. NOMURA, 1970. Electromechanical coupling in a molluscan muscle, the radular protractor of Busycon canaliculatum. J.Exp. Zool, 174:331-348. HILL, R. B., E. MARANTZ, B. A. BEATTLE AND J. M. LOCKHART, 1968. Mechanical properties of the radular protractor of Busycon canaliculatiim. E.rperientia, 24: 91-92. INNERVATION OF RADULAR PROTRACTOR MUSCLE HOYLE, G., 1964. Muscle and neuromuscular physiology. Pages 313-351 in K. M. Wilbur and C. M. Yonge, Eds., Physiology of Mollusca, Vol. I. Academic Press, New York. McKENNA, O. C. AND J. ROSENBLUTH, 1973. Myoncural and intermuscular junctions in a molluscan smooth muscle. /. Ultrastruct. Res., 42 : 434-450. NICAISE, G., M. PAVANS DE CECCATTY AND C. BALEYDIER, 1968. Ultrastructure des connexions entre cellules nerveuses, musculaires et glio-interstitielles chez Glossodors. Z. Zcll- forsch.,8S: 470-486. RICHARDSON, K. C., 1964. The fine structure of the albino rabbit iris with special reference to the identification of adrenergic and cholinergic nerves and nerve endings in its intrinsic muscles. Amcr. J. Aunt., 114: 173-205. ROGERS, D. C., 1968. Fine structure of smooth muscle and neuromuscular junctions in the optic tentacles of Helix aspcrsa and Limax flai'its. Z. Zclljorsch., 89 : 80-94. ROGERS, D. C., 1969. Fine structure of smooth muscle and neuromuscular junctions in the foot of Helix aspcrsa. Z. Zclljorsch.. 99 : 315-335. SANGER, J. W., 1973. Demonstration of a sliding filament mechanism of contraction in some invertebrate smooth muscles. /. Cell Biol., 49 : 2()la. SANGER, J. W., AND R. B. HILL, 1972. Ultrastructure of the radular protractor of Busycon canaliculatum. Sarcolemmic tubules and sacroplasmic reticulum Z. Zcllforsch., 127 : 314-322. SANGER, J. W., AND R. B. HILL, 1973a. The contractile apparatus of the radular protractor muscle of Busycon canaliculatum. Proc. Malacol. Soc. London. 40: 335-342. SANGER, J. W., AND R. B. HILL, 1973b. A study of the innervation of the radular protractor muscle of Busycon canaliculatum. Biol. Bull., 145: 454. TWAROG, B. M., 1967. Factors influencing contraction and catch in Mvtilus smooth muscle. /. Physio!., 192 : 847-856. WELSH, J. H., 1956. Netirohormones of invertebrates. I. Cardioregulators of Cyprina and Buccimim. J. Mar. Biol Ass. U.K., 35 : 193-201. WELSH, J. H., AND M. MOORHEAD, 1959. Identification and assay of 5-hydroxytryptamine in molluscan tissues by fluorescence method. Science, 129 : 1491-1492. Reference: Biol. Bull., 147: 386-396. (October, 1974) THE DISTRIBUTION OF SIX SPECIES OF GASTROPOD MOLLUSCS IN A CALIFORNIA KELP FOREST LLOYD F. LOWRY i, ALFRED J. McELROY 2, AND JOHN S. PEARSE 1 Hopkins Marine Station, Pacific Grove, California 93950 Kelp forests are common features along temperate shores of much of the world. They are confined to shallow, rocky neritic zones where conditions of light, substrate and surge are suitable. Their productivity is very high, and much organic material flows through complex, but definable trophic structures (Miller, Mann and Scarratt, 1971; Mann, 1973). Spatial structures within these forests also are complex but definable, and many organisms occupy restricted habitats. Drach (1960) distinguished at least 8 types of rock habitats along the coast of Brittany, each with a discrete faunal assemblage. Within the kelp forests of California, vertical rock faces are commonly covered by sponges, anthozoans, byrozoans and ascidians, horizontal rock surfaces at similar depths are covered by algae and motile invertebrates such as decapods, gastropods and asteroids, while the under- sides of loose rocks are the habitat for several species of brachyurans, amphineurans and ophiuroids (McLean, 1962, and personal observations). Within the range of the sea otter, rocky crevices are the main habitat of abalones and sea urchins (Lowry and Pearse, 1973). Algae, in addition to rocks, provide habitats for many organisms. The hold- fasts of Macrocystis spp. and Ncrcocystis luctkeana support an abundant and di- verse fauna consisting primarily of polychaetes, decapods, amphipods, gastropods and ophiuroids (Andrews, 1925, 1945; Ghelardi, 1971). Andrews (1925) found that the stipes of Nereocystis luctkctuia supported few animals while decapods, amphipods, isopods, gastropods and bryozoans were all abundant in the laminae. Wing and Clendenning (1971) found large numbers of micro-invertebrates asso- ciated with Macrocystis blades, especially blades that were encrusted with the bryozoan Membranipora. Clarke (1971) reported that the mysid Acanthomysis sculpta is found almost exclusively in Macrocystis canopies and within kelp forests this also appears to be the case for the shrimp Hip poly fe calif orni crisis (L. Lowry, unpublished data). In the study reported here, we examined habitat specificity and differentiation among 6 species of closely related gastropods (family Trochidae) on 4 species of kelp forest algae. The snails were chosen because of their apparent ecological similarity (all predominantly subtidal browsing animals of similar sizes) and for ease of collection and identification. The algae were chosen because they provided the snails with a wide range of conditions. The giant kelp, Alacrocystis pyrifcra. is a perennial plant and the dominant alga in the kelp forest studied. Individual plants are anchored to the substrate by a large rhizomatous holdfast from which extend many fronds, each consisting of a central stipe bearing unilateral blades 1 Present address : Division of Natural Sciences, University of California, Santa Cruz, California 95064. 2 Present address : Department of Life Science, Sierra College, Rocklin, California 95677. 386 DISTRIBUTION OF KELP FOREST SNAILS 387 and pneumatocysts. The fronds extend through the entire water column and spread out on the surface to form a thick canopy. The brown alga, Cystoseira osmundacca, consists of small ( less than 1 meter high ) , perennial vegetative portions from which massive reproductive fronds rise to the surface of the water from depths of 10 in or more. The reproductive fronds begin growing in mid-winter and break up and disappear by late summer. The vegetative portion of C. osmundacca is the main understory plant in our study area (density 1.6 individuals/m2 ), while the repro- ductive portion adds seasonally to the canopy. The other two algae used in this study grow beneath the C. osmundacea vegetative fronds and are a major portion of the ground cover algae. These are the brown alga Dictyoneuropsis rcticnlata which has large blades (100 X 25 cm) that usually lie close to the bottom, and the red alga Gigartina corymbifera which is similar in growth form to D. reticulata, but with somewhat smaller blades (up to 50 cm long). METHODS AND MATERIALS The study site was located off the northeast side of Point Cabrillo at Hopkins Marine Station, Pacific Grove, California. The area is described in more detail in Lowry and Pearse (1973). Collections were made by divers using scuba equip- ment during August, 1971, at five locations (depth 6-9 in ) in the kelp forest. The plants were chosen arbitarily and all snails were removed from each plant chosen. Snails on Gigartina and Dictyoneuropsis were simply removed from the algae and put into plastic bottles. Snails from the vegetative and reproductive parts of Cystoseira were collected separately. For sampling Macrocystis, a line marked in meter intervals was used ; snails from each one meter interval up to five meters from the bottom and those from above five meters (including the canopy) were kept separated. Using these procedures, a total of 85 Macrocystis fronds (from 6 separate plants), 60 Cystoseira plants, 315 Dictyoneuropsis blades and 440 Gigartina blades were sampled. The snails were then taken to the lab where they were identified, and the maximum basal diameter of each specimen measured to the nearest millimeter. RESULTS Sizes and distributions of the species of snails Calliostoma annulatum (Lightfoot, 1786). The least number of the 6 snail species collected was C. annnlatum. They also had the smaller modal size (14 mm, see Fig. 1). Twenty-two of the 24 individuals collected were found on Cystoseira, 2\ of these 22 were on the reproductive portions of the plants. Two individuals were found on Macrocystis and none on the ground cover algae (Dictyoneuropsis and Gigartina} (Table I). Calliostoma canaliculatum (Lightfoot, 1786). This species was only slightly more common in our collections than C. annulatum and it had a very similar distribution. Five individuals were found on Macrocystis, but most (21 out of 27) were on the reproductive portions of Cystoseira and none were on the ground cover algae (Table I). The size-frequency distribution for this species (Fig. 1) shows 388 LOWRY, MCELROY AND PEARSE o z UJ 30 25 $20 r 15 0 5 o UJ IT CALLIOSTOMA ANNULATUM N =24 CALLIQSTQMA LIGATUM CALLIOSTOMA CANALICULATUM N = 27 30 10 15 20 25 30 TEGULA BRUNNEA N = 226 15 20 30 35 10 15 20 BASAL DIAMETER (mm) FIGURE 1. Size frequency distributions of the total collections of Calliostoina spp. and Tegula brunnea. The relative frequency is the per cent of the total collection for a given species that was in each 1 mm size class. a single mode at 16 mm, and a predominance of large individuals (20-21 mm). The largest snail found (34 mm") belonged to this species. Calliostoma ligatum (Gould, 1849). Twenty-nine out of 39 individuals of this species collected were found on Cystoseira, 13 of these were on the vegetative portions of the algae. Five individuals were found on Ifacrocystis and 5 on Gigartina (Table I). The size-frequency distribution (Fig. 1) shows a single mode at 17-18 mm. Tegula brunnea (Philippi, 1848). One hundred eighty-seven out of 226 T. brunnea collected were from Macrocystis. Eleven were found on Cystoseira, 7 on the vegetative and 4 on the reproductive portions of the algae. Twenty-three specimens were found on Dictyoneuropsis and 5 were on Gigartina (Table I). The size frequency distribution (Fig. 1) shows a single mode at 21-22 mm. TABLE I Per cent of the total collection of each species of snail which was found on a given species of algae. N is the total number of individuals of each snail species collected. Species Cystoseira vegetative Cystoseira repro- ductive Cystoseira total Macro- cystis Dictyoneu- ropsis Gigartina Total N C. annulatum 4 88 92 8 0 0 100 24 C. canal iculatum 4 77 81 19 0 0 100 27 C. ligatum 33 41 74 13 0 13 100 39 T. brunnea 3 2 5 83 10 2 100 226 T. montereyi 8 13 21 51 12 16 100 156 T. pulligo 12 19 31 39 21 9 100 404 DISTRIBUTION OF KELP FOREST SNAILS 389 TEGULA MONTEREYI 3 25 £20 15 10 5 25 20 15 10 5 TOTAL TEGULA PULLIGO ON CYSTOSEiRA N = 19 ON DICTYONEUROPSIS ON GIGARTINA 10 20 25 30 10 BASAL DIAMETER (mm) FIGURE 2. Size frequency distributions for the total collections and collections from each species of algae for Tct/ula ntontcrcyi and T. fulligo. The relative frequency is the per cent of the total collection or the collection from a given algal species that was in each 1 mm size class. Tegula montereyi (Kiener, 1850). About half (80 out of 156) of the snails of this species collected were found on Macrocystis. Thirty-three were found on Cystoseira, a few more on the reproductive than on the vegetative portions of the plants. Nineteen were found on Dictyoneuropsis and 24 on Gigartina (Table I). The size frequency distribution for the total collection of T. montereyi shows 3 modes, one at 13 mm, another at 18 mm and a third at 22-23 mm (Fig. 2). The 390 LOWRY, MCELROY AND PEARSE modes at 18 and 22-23 mm are accounted for by a predominance of individuals of these size classes on Macrocystis, while the mode at 13 mm represents smaller individuals which were found mostly on Cystoseira, Dictyoneuropsis and Gigartina. Tcgula pulligo (Gmelin, 1791). This species was the most frequently collected snail in the study and was found regularly on all algal species investigated. One hundred fifty-six out of 404 individuals collected were on Macrocystis, 124 on Cystoseira, 85 on Dictyoneiiropsis and 39 on Gigartina (Table I). The size frequency distribution for the total collection shows 2 distinct modes at 14 and 20 mm, respectively (Fig. 2). As with T. niontcreyi, larger snails were found on Macrocystis, while smaller sizes were found on Cystoseira, Dictyo- neiiropsis and Gigartina. Comparison of distributions among algal species The figures in Table I were used to compute indices of association (Whittaker, 1952 ) to facilitate comparison of the distributions of the six snail species. The results are shown in Table IT. The mean value for the similarity indices cal- culated for comparisons of species pairs within the genus Tcgula is 68.3, due mostly to their similar high relative abundance on Macrocystis. The mean value for com- parisons of species pairs within the genus C all lost oina is 66.6, due largely to their similar high relative abundance on Cystoseira. These mean values are not sig- nificantly different [t — 0.125 with 6 degrees of freedom (D.F.)] indicating that although the two genera are found on different algal species, the degree of dis- tributional association found among species of the genus Tegiila is similar to that found among species of the genus CaUiostouia. The mean value for comparisons of species pairs which include a species from each of the 2 genera is 32.3. This is significantly lower than comparisons within the genus Tcgula (t - ' 4.126, 6 D.F.) or within the genus CaUiostotna (t -- 3.504, 6 D.F.). Vertical distribution of Tegula sf>f>. on Macrocystis Figure 3 show's the relative abundance of the 3 species of Tegula at various heights on Macrocystis plants. At the base of the plants, T. pulligo is the most TABLE II Indices of association (Whittaker, 1952) calculated from the distribution of snails shown in Table 1. The index for a species pair is calculated as IA,B == ]d'=iB m^n (a>b) where a and b are the values given in Table 1 for species A and B in a given column and i represents the columns used. The vegetative and reproductive portions of Cystoseira were considered separately; the values for total Cystoseira were not used in the calculations. The index has a maximum value of 100 (exactly equivalent distributions) and a minimum value of 0 (never found on the same species of algae). C. annulatum C. canaliculatum C. ligatum T. brunnea T. montereyi T. pulligo 31 42 53 56 81 T. montereyi 25 36 47 68 T. brunnea 13 24 20 C. ligatum 53 58 C. canaliculatum 89 DISTRIBUTION OF KELP FOREST SNAILS 391 CANOPY r- Qd UJ h- UJ o 00 o (T u_ UJ I 4-5 3-4 2-3 1-2 0- I - X 10 20 30 40 50 60 70 80 RELATIVE FREQUENCY IN EACH ZONE (%) FIGURE 3. Vertical distribution of Tcgula spp. on Macrocystis. The relative frequency is the per cent of the total collection of TcIana.\-is and L. scutulata. Biol. Bull., 106: 185-197. PAINE, R. T., 1969. The Pisastcr-Tcgula interaction : prey patches, predator food preference, and intertidal community structure. Ecology, SO : 950-961. SMITH, A. G., AND M. GORDON, JR., 1948. The marine mollusks and branchiopods of Monterey Bay, California, and vicinity. Proc. Calif. Acad. Sci., Series 4, 26: 147-245. WHITTAKER, R. H., 1952. A study of summer foliage insect communities in the Great Smoky Mountains. Ecol. Monogr., 22 : 1-44. WING, B. L., AND K. A. CLENDENNING, 1971. Kelp surfaces and associated invertebrates. Pages 319-339 in W. J. North, Ed., The Biology of Giant Kelp Beds (Macrocystis) in California (Nova Hedivigia 32 Suppl.). J. Cramer, Lehre, Germany. Reference: Biol Bull., 147: 397-410. (October, 1974) THE CELLULAR ORIGIN OF BIOLUMINESCENCE IN THE COLONIAL HYDROID OBELIA JAMES G. MORIN AND GEORGE T. REYNOLDS Department of Biology, University of California, Los Angeles, California 90024; Joseph Henry Laboratories, Department of Physics, Princeton University Princeton, New Jersey OS540; and Tlie Marine Biological Laboratory, U'oods Hole, Massachusetts 02543 The ability to produce light is a phenomenon present in many marine organisms. Within the hydrozoan coelenterates the mechanisms that control the production of light are gradually coming to he known and understood. Using the colonial hydroid Obclia, Morin and Hastings (1971a, 1971b) presented biochemical information on light emission from a calcium activated photoprotein and a secondary emission from an associated green fluorescent protein via an energy transfer. Physiological mechanisms controlling the multiple flash response in Obclia have been examined by Morin and Cooke (1971b, 1971c). The responses were monitored by recording the electrical potentials of an excitation system and bv measuring the coupled luminescence of small spots within the animal. \\ liile these previous papers have examined the mechanisms of bioluminescence, little is known of the structure of the photogenic tissues in hydroids. In the present paper we describe some of the structural aspects of the luminescent effector cells in Obclia. For the work described in this paper, image intensification and fluo- resence techniques were used to show that luminescent and unique green fluorescent sites are identical ; and to determine the size, shape, distribution and localization of these luminescent sites. Preliminary reports of this work have been published (Morin, Reynolds and Hastings, 1968; Morin and Reynolds, 1969, 1970). Panceri's papers (1876, 1877) are the most recent accounts of the sources of bioluminescence in the hydroid form of the hydrozoa. He concluded that lumines- cence was located as discrete spots in the ectodermal tissues everywhere within Canipannlaria flc.viiosa. Davenport and Nicol (1955) investigated the sources of luminescence in several hydrozoan medusae and showed that the photogenic ma- terial was intracellular, with masses of several thousand cells lying just under the gastrodermis of the marginal canal. Image intensification has been used to locate the exact sources of luminescence in the firefly (Hanson, Miller and Reynolds, 1969), Renilla (Buck, Hanson and Reynolds, 1967), Noctilnca (Eckert and Reynolds, 1967), Pyrocystis (Swift and Reynolds, 1968), Gonyaiila.r (Reynolds, Hastings, Sato and Sweeney, 1966), and Obelia (Morin, Reynolds and Hastings, 1968). Fluoresence techniques have been used for the inspection of luminescent regions in Mnemiopsis (Harvey, 1925; Harvey and Marfey, 1958), certain annelids (Nicol, 1953, 1954), Noctilnca (Eckert and Reynolds, 1967), Acquorca and other hydromedusae (Davenport and Nicol, 1955), and in the pennatulids : Pcnnatula, Ptcroides, Veretillum (Tit- schack, 1964, 1966), Ptilosarcus, Renilla, Stylatula, Acanthoptihim and J^irgnlaria (Morin and Reynolds, 1970; Morin et al., in preparation), 397 398 J. G. MORIN AND G. T. REYNOLDS MATERIALS AND METHODS Obelia geniculata (L.) was obtained and cultured according to the method of Morin and Cooke (1971a). The luminescent sites were examined using three methods: (1) autophotography by means of image intensification, (2) fluorescence microscopy and (3) histology which consisted of fixation by means of freeze-drying with subsequent embedding and sectioning for examination with the fluorescence microscope. Autophotography Flashes of light for autophotography were evoked by applying 0.5 to 5 msec duration square pulses from a stimulator through a pair of fine, closely spaced silver wires placed across an upright of Obelia (Morin and Cooke, 197 Ib). The physical arrangement of the image intensifier consisted of a microscope (American Optical) supplemented with a beam director (Zeiss) such that the field could be directed to the oculars for a direct view, to a camera for direct photography, or to the photocathode of the image intensifier tube for intensifica- tion and (a) photography at a second camera or (b) recorded on magnetic tape with simultaneous television monitoring. Technical details are given elsewhere (Reynolds, 1972). Magnifications of 3.5 X, 10 X and 55 X which gave fields of 4 mm, 1.5 mm and 0.25 mm diameter respectively, were used. Rear illumination photographs were taken by light transmitted from below the microscope stage which was directed to the image tube cathode and recorded. Image intensification provided a direct means of photographing very low light levels. The luminescence was too weak to be recorded on film without such image intensification. The image intensifier tube gain was varied from approximately 104 to 106, depending on the magnification and light output of the specimen. Even with highest numerical aperture objectives used, only the order of a few per cent of the light was collected and transmitted through the microscope, so that in general, high image tube gains were required. Sources of image noise which produced small spots on the developed film were (1) thermal electrons from the cathode, (2) background grains in the film (fog) and (3) scatter and reflection within the specimens, chamber and optical system. The large difference in size between the luminescent spots and the small noise spots reduced this problem to a negligible level (Fig. 1). Fluorescence microscopy Fluoresence was photographed through the image tube by means of rear illumination combined with the proper filters. A blue (460 nm) interference excitation filter was placed between the light source and the specimen, and a green (507 nm) interference pass filter was placed between the specimen (after the optics) and the image tube. The resulting image with its fluorescent spots was photographed on the camera behind the image intensifier for comparison with the pictures of the luminescence taken of the same field. Still photographs of the image tube anode were taken using an f/1.9 lens and Polaroid film (ASA 10,000). With the room totally darkened, the shutter was opened, the specimen was stimulated and then the shutter was mechanically closed HYDROID BIOLUMINESCENCE 399 by the observer of the image tube anode, after the flash had been seen (usually 1-2 seconds). A Leitz ultraviolet microscope with 2.5 X, 10 X, 54 X (oil immersion) and 94 X (oil immersion) objectives was used for fluorescence microscopy without the image intensifier. Ultraviolet illumination from a 200 watt, high pressure mercury lamp was used with a dark field condenser in order to maximize the observed fluoresence. Exciting light passed through a heat absorbing filter (Leitz BG 38) and an ultraviolet pass filter (Leitz BG 12) with a peak transmittance at approximately 400 nm. A barrier filter (Leitz K510) removed wavelengths below 500 nm. Kodachrome Tri-X black and white film was used for photography. Freeze-drying methods The freeze-drying method was similar to that of Rude (1966). The specimens were pinned to a planchet in sea water, dipped into distilled water for about one second in order to remove external salts, drained briefly on filter paper, and then plunged into isopentane cooled by liquid nitrogen ( — 160° C). The frozen colonies were transferred to a freeze-dry appartus and dried for about three days at a pressure of 2 /x Hg and an outside temperature of -40° C. The speci- mens were then slowly brought to room temperature. The specimens were placed in Maraglas embedding medium in a vacuum desiccator for several hours ; they were then transferred to a 40° C oven for about seven hours until the Maraglas hardened. Seven /* serial cross sections were made on a Spencer A.O. micro- tome with metal knives. The serial sections were mounted in Entellan, a non- fluorescing mounting medium, on microscope slides and examined with the fluorescence microscope. RESULTS Visual appearance of the luminescence Stimulation of Obelia geniculata evoked light which emanated from small points within the colonies. Within a given microscope field the light showed distinct multiple flashes in response to individual stimuli. It was difficult to determine visually the precise source of the luminescence because of the flickering and rela- tively weak light. It was not possible by visual means to be certain if individual spots were flickering or if the flicker was a consequence of sequential luminescence of different spots, possibly along the length of the colony. Evidence, presented from photometric responses of single luminescent sites (shown by fluorescence), indicated that single spots did flash repetitively (Morin and Cooke, 1971b, 1971c). Autophotography and fluorescence microscopy pro- vided further evidence for repetitive flashing of single spots and information about the general location of the luminescent sites. Autophotography and fluorescence of the luminescent: sites General characteristics of the luminescent sites. The usual pattern of lumines- cence in Obelia geniculata, demonstrated by single frame autophotography, is shown in Figure 1. Figure 1A-C shows a 3.5 X field of an upright with (A) rear 400 J. G. MORIN AND G. T. REYNOLDS illumination, (B) luminescence and (C) both almost superimposed. An enlarged field (10 X) of the central portion of figure 1A-C is shown in Figure 1D-F with the same format. The photographs show that the luminescent sites are discrete spots of variable size. The spots are located in the uprights and pedicels and not in the hydranths ; the spots remain fixed in their spatial patterns, at least between successive photographs spaced several minutes apart. In addition to this constant FIGURE 1. Image intensifier photographs of Ohclia (./cuicitliita bioluminescence : (A-C). Low power (3.5X) photographs showing (A) rear illumination of upright, (B) autophoto- graph of upright luminescence and ( C ) superposition of rear illumination ( A ) and autophoto- graph (B). (D-F), same colony but at a higher magnification ( 10 X ) taken a few minutes after (A-C); same format. Bars indicate 1 mm. Scattered small spots in (I)) and (E) are image noise (see text for details). pattern of organization, other photographs showed that the stolons contain lumines- cent spots and that medusae within the gonangia show no luminescence upon stimulation of the colony, although mature medusae are capable of emitting light. Correspondence of fliiorescnce u'itli luminescence. Biochemical evidence indi- cated that the luminescent sites could be located and observed using a method which exploits the fluorescence charcteristics of the emission system when excited with blue light (Hasting and Morin, 1969a, 1969b ; Morin and Hastings, 1971b). Such a method involved excitation of the luminescent sites with 460 nm light and examination of the emitted light using a barrier filter to exclude wavelengths shorter HYDROID BIOLUMINESCENCE 401 than 500 nm. The sites revealed by this method fluoresced a bright green color (Amax -- 508 nm). Yellow-green fluorescence from small particles (a few microns) was also observed within the colonies, especially in well nourished ones. There was no correspondence between these latter fluorescent sites and the luminescent sites. Epiphytic diatoms attached to the perisarc of the colonies displayed the characteristic red color of chlorophyll fluorescence. The direct correspondence between luminescence and fluorescence was shown by successive photographs of the same field through the image tube. First, an autophotograph was taken of the luminescence (Fig. 2A) and then a photograph was taken of fluorescence excited by filtered rear illumination (Fig. 2B). In all cases there was an exact correspondence for each site. In the following sections, therefore, it is considered that observations of the fluorescent sites provide a description of the luminescent sites. FIGURE 2. Autophotographic (A), fluorescence (B) and rear illumination (C) pictures of an Obclia upright. Note the direct correspondence between the luminescent and fluorescent spots, (six in each). They directly superimpose. The scale bar indicates 200 /j.. The fluorescence method is extremely useful for characterizing the luminescent sites because the fluorescent emission is not intermittent, as is the luminescence, and the living material can be examined without any manipulation of the colonies, (i.e., stimulation, surgical or histological procedures). Dimensions and shape of the luminescent sites. The size of the luminescent sites as shown by fluorescence varies within the range of 5 to 30 /m with a usual size of about 10 X 20 p.. The sites shown by autophotography are slightly larger than fluorescence sites possibly because of overexposure. The nonfluorescent cells in the same regions as the fluorescent sites have dimensions similar to those of the fluorescent sites. This observation strongly suggests that the luminescent sites are single cells, and we will therefore refer to them as photocytcs. Histological evidence given below supports this conclusion. The photocytes possess a wide range of shapes (Fig. 3). They are best observed in very young, distal tissues or in the distal part of the pedicels where the perisarc is relatively transparent and not overgrown by epiphytes (Fig. 3B). A characteristic feature of these cells is the frequent occurrence of one or 402 J. G. MORIN AND G. T. REYNOLDS FIGURE 3. Fluorescence microscope photographs of living Obelia gcniculata. (A). Basal part of an upright showing three nodes ; each node and pedicel shows a cluster of photocytes. The skeletal outline of the hydroid has been dotted in white. Hydranths are indicated by an H. The dim fluorescent material around the photocytes are red fluorescing diatoms attached to the hydroid. (B). Distal part of an upright showing two nodes; note the more numerous photocytes than in (A) and the limited diatom growth. Again the outline has been dotted in. (C). Single photocyte in a pedicel; note the long projection. (D). Three photocytes with projections; note the dark presumed nucleus of the upper cell. Bar indicates 40 ^ in (A) and (B); 25 /A in (C) and (D). HYDROID BIOLUMINESCENCE 403 more cytoplasmic projections that may be up to 20 ^ in length (Fig. 3C, D). These projections have no apparent orientation to neighboring photocytes. The distance between photocytes varies from several mm, usually in the stolons or proximal uprights (Fig. 3A), to near contact, especially in the pedicels and distal parts of the uprights (Fig. 3B). The distribution of fluorescence within a photocyte including the projections usually appears homogeneous in a living whole mount except for a nonfluorescent 14 O 14 13 12 IO 9 8 Segment 7 No. FIGURE 4. The number of photocytes per "segment" (one "segment" as defined here includes the pedicel, node and proximal internode — see diagram, upper right) is plotted against the "segment" number (the oldest, most proximal "segment" is referred to here as the first "segment") for three different, randomly selected uprights. One (circle with a cross) shows the number when a small side branch was and was not considered in the count. inclusion (sometimes two) within each cell. The size and location of this inclu- sion is suggestive of a nucleus (Fig. 3D). Occasionally a compartmentalization of the fluorescence within the photocytes is indicated. Distribution and density of the photocytes. The general pattern of photocyte distribution and density is shown in the photographs of fluorescence in Figure 3 A, B. These photographs show that (1) there are more photocytes in the pedicels and nodes than in the internodes of the upright, (2) the photocytes are all of approximately the same size and (3) there is a greater concentration of photocytes 404 .1. / Hydra. Coral Gables, University of Miami Press. DAVENPORT, D., AND 1. A. C. NICOL, 1955. Luminescence in hydromedusae. Proe. Ko\. Soc. London, Series B, 144: 399-411. ECKERT, R., AND G. T. REYNOLDS, 1967. The subcelhilar origin of bioluminescence in Xoctiluca miliaris. J. Gen. Physiol., 50: 1429-1458. FREEMAN, G., AND G. T. REYNOLDS, 1973. The development of biohnninescencc in the cteno- phore Mnemiopsis Icidyi. Develop. Bid.. 31 : 61-100. HAN- sox, F. E., JR., J. MILLER AND G. T. REYNOLDS, 1969. Subunit coordination in the firefly light organ. Biol. Bull.. 137 : 447-464. HARVEY, E. N., 1925. Studies on bioluminescence. XML Fluorescence and inhibition of luminescence in ctenophores by ultra-violet light. /. (ien. Physiol., 7 : 331-33'). HARVEY, E. N., 1952. Bioluuiinescence. New York, Academic Press, n4l> pp. HARVEY, E. N., AND S. P. MARFEY, 1958. Fluorescence, phosphorescence and bioluminescence in the ctenophore Mnemiopsis leidyi. Biol. Bull.. 115: 336-337. HASTINGS, J. W., AND J. G. MORIN, 1969a. Comparative biochemistry of calcium-activated photoproteins from the ctenophore Mnemiopsis and the coelenterates Acquorea. Obelia, Pelagia and Rcnilla. Biol. Bull., 137 : 402. HASTINGS, J. W., AND J. G. MORIN, 1969b. Calcium-triggered light emission in Rcnilla. A unitary biochemical scheme for coelenterate bioluminescence. Bioehein. Biophvs. Res. COHIHI.. 37: 493-498. MORIN, J. G., AND I. M. COOKE, 1971a. Behavioural physiology of the colonial hydroid Obelia. I. Spontaneous movements and correlated electrical activity. /. E.rp. Biol., 54 : 689-706. MORIN, J. G., AND I. M. COOKE, 1971b. Behavioural physiology of the colonial hydroid Obelia. II. Stimulus-initiated electrical activity and bioluminescence. /. E.rf>. Biol., 54: 707-721. MORIN, J. G., AND I. M. COOKE, 1971c. Behavioural physiology of the colonial hydroid Obelia. III. Characteristics of the bioluminescent system. /. E.vp. Biol.. 54: 723-735. MORIN, J. G., AND J. W. HASTINGS, 1971a. Biochemistry of the bioluminescence of colonial hydroids and other coelenterates. /. Cell. Physiol., 77 : 305-312. MORIN, T. G., AND f. W. HASTINGS, 1971b. Energy transfer in a bioluminescent system. /. Cell. Physiol., 77 : 313-318. MORIN, J. G., AND G. T. REYNOLDS, 1969. Fluorescence and time distribution of photon emission of bioluminescent photocytes in Obelia geniculata. Biol. Bull., 137 : 410. MORIN, J. G., AND G. T. REYNOLDS, 1970. Luminescence and related fluorescence in coelenter- ates. Biol. Bull. 139 : 430-431. MORIN, J. G., G. T. REYNOLDS AND J. W. HASTINGS, 1968. Excitatory physiology and localiza- tion of bioluminescence in Obelia. Biol. Bull., 135 : 429-430. NICOL, J. A. C., 1953. Luminescence in polynoid worms. /. Mar. Biol. Ass. U.K., 32 : 65-84. NICOL, J. A. C., 1954. The nervous control of luminescent responses in polynoid worms. /. Mar. Biol. Ass. U.K., 33 : 225-255. 410 J. G. MORIN AND G. T. REYNOLDS PANCERI, P., 1876. Intorno alia sede del movimento luminoso nelle Campanularie. Rend. Accademia Delle Sciense Fisiche e Matematiche. 15: 193-195. PANCERI, P., 1877. Luminous Campanulariae. Nature, 16: 30. REYNOLDS, G. T., 1972. Image intensification applied to biological problems. Quart. Rev. Biophys., 5 : 295-347. REYNOLDS, G. T., J. W. HASTINGS, H. SATO AND A. R. SWEENEY, 1966. The identity and photon yield of scintillons of Gonyaulax polyedra. Biol. Bull., 131 : 403. RUDE, S., 1966. Monoamine-containing neurons in the nerve cord and body wall of Lumbricus tcrrestris. J. Comp. Neurol., 128 : 397-411. SHIBLEY, G. A., 1969. Gastrodermal contractions correlated with rhythmic potentials and pre- locomotor bursts in Hydra. Amer. Zool., 9 : 586. SWIFT, E., AND G. T. REYNOLDS, 1968. Localization of bioluminescence in the marine dino- flagellate Pyrocystis lunula Schutt, using an image intensifier. Biol. Bull., 135 : 439-440. TITSCHACK, H., 1964. Untersuchungen iiber das Leuchten der Seefeder Veretillum cyno- morium. Vie Milieu, 15 : 547-563. TITSCHACK, H., 1966. Uber die Lumineszenz und ihre lokalisation bei Seefedern. Zool. Anz. Suppl.,29: 120-131. WAMPLER, J. E., Y. D. KARKHANIS, J. G. MORIN AND M. J. CORMIER, 1973. Similarities in the bioluminescence from the Pennatulacea. Biochiin. Biophys. Ada., 314: 104—109. Reference: Biol. Bull., 147: 411-421. (October, 1974) ON THE OCCURRENCE OF COLONY SPECIFICITY IN SOME COMPOUND ASCIDIANS * HIDEO MUKAI AND HIROSHI WATANABE Biological Institute, Faculty of Education, Gunma University, Gunma, Japan and Shimoda Marine Biological Station, Tokyo Kyoiku University, Shisuoka, Japan In some colonial organisms, colony specificity, or histoincompatibility, has been watched recently by many authors. Colony specificity in some compound ascidians is manifested by the fusibility between colonies; two colonies fuse with each other to form a single mass in one case or do not fuse in the other. This problem was first taken up by Bancroft (1903). According to him, two colonies of different origin in Botryllus schlosseri do not fuse together after grafting. Fragments from any one single colony, however, easily fuse together. Sister and brother colonies developed from larvae released by one parental colony sometimes fuse and sometimes do not. Oka and Watanabe (1957, I960, 1967) encountered similar results in the Japanese ascidian Botryllus prhnigemts, and the genetic control of colony specificity has been further investigated. The colony specificity in B. schlosseri also has been investigated, to a lesser extent, by some authors (Sabbadin, 1962; Karakashian and Milkman, 1967). In Botrylloides gascoi and B. leachi, according to Bancroft (1903), fusion is invariably established between two colonies, regardless of their origin. Oka and Usui (1944) have reported that for Polycitor niutabilis, in which the zooid of a colony exists solitarily in common test, fusion is never seen when two colonies come into contact as a result of natural growth, but when cut surfaces of two pieces are placed in contact, they can be fused to form a single colony. Hence, in these species colony specificity seems to be absent. Besides ascidians, the occurrence of colony specificity has been known in some groups of coelenterates, e.g., a hydrozoan H \dractinia echinata (Hauenshild, 1954, 1956: Ivker, 1966, 1967, \972\ Toth, 1967) and two anthozoans Eunicella stricta and LopJwgorgia sarmetitosa (Theodor, 1970). In certain bryozoans (cf. Ryland, 1970, page 30) and freshwater sponges (Rasmont, 1970), the presence of colony specificity has also been suggested. Thus, the colony specificity seems to be a phenomenon widely distributed among colonial organisms. The present paper reveals the presence of colony specificity in some compound ascidians and some insight is given, from the comparative point of view, into the mechanism involved in the establishment of fusion or rejection and, also, into the evolutionary trend of the colony specificity. MATERIALS AND METHODS The materials used in the present experiments were living colonies of five compound ascidians, such as Botryllus primigenus, Botrylloides violaceus, Pero- 1 Contribution from the Shimoda Marine Biological Station No. 280. 411 412 H. MUKAI AND H. WATANABE phora oricntalis, Symplegma rep tans and Didemnum mosclcri. The first three species were collected in the vicinity of the Shimoda Marine Biological Station, Shizuoka Prefecture, the next two in the vicinity of the Usa Marine Biological Station, Kochi Prefecture. To facilitate handling of the colonies, they were fixed on glass plates. For Botryllus and Botrylloides, the colonies taken from nature were fastened to glass plates by the method described by Oka and Usui (1944). For Perophora, Sym- plegina and Didemnum, the colonies, with the substratum or after separation from it, were fastened to glass plates by tying down the edges with string. The plates with the colonies attached were then set in wooden frames and hung below the sur- face of the sea. The colonies were thus reared in their natural environment, except for transfer to the laboratory aquarium for performing fusion experiments. The fusibility of the colonies of Botryllns. Botrylloides and Sywiplegma, was tested by fusion experiments. Fusion experiments were routinely carried out by the following procedure : A piece of colony of about 1 sq cm in size was cut out from each of two colonies. The two pieces were placed in juxtaposition on a glass plate and these colony pieces made a contact with each other. The colonies attached successfully on the glass plates after being kept in a moisture chamber in one hour or so, were returned to the culture boxes in the bay as was just mentioned. For the Didemnum colonies, the above method, to attach them on a glass plate in one hour or so, proved to be a not very practical one. Hence, two colony-pieces with contact were fastened to a glass plate by tying them down with string. The plate was then hung in the bay. The string could be removed the following day. For the PcropJiora colonies, use of the following contrivance was made. That is, two stolon-pieces, each consisting of a stolon fragment one centimeter or more in length and some zooids with the actively pulsating heart, were cut out and placed on a filter-paper sheet in a Petri-dish. They were then forced to be in contact at their cut ends. About twenty minutes later filtered sea water was poured in the dish, when the stolon-pieces, now adhered together in a single stolon, floated on the surface. After removal of the filter paper, the stolon-pieces were submerged to the bottom and cultured in the dish. In each species the fusibility of colonies was examined under two conditions, i.e., by bringing either the cut surfaces or their growing edges into contact. The observation was made one or two days later under a binocular stereo- microscope. In the reaction resulting from contact of two colonies in respective species, three cases can be distinguished. In this paper, by fusion is meant the complete union of blood vessels and/or the test matrix of one colony with the same tissues of the other, in other words, the establishment of a common vascular system or a common test between the two. The case in which some active antago- nism is seen between the two colonies is referrd to as rejection to distinguish from indifference in which no particular reaction can be observed. RESULTS Botryllus primigenus Botryllus is a member of the subfamily Botryllinae (Berrill, 1950). In a •colony the individual blastozooids are grouped into star-shaped systems and are COLONY SPECIFICITY IN ASCIDIANS 413 connected with one another by the ramifying network of vascular vessels, which terminate in ampullae at the periphery of the colony. The processes of fusion and rejection have already been described and illustrated in detail by Oka and Watanabe (1967) and Oka (1970). Therefore, the brief account of the processes will be described below. If the cut surfaces of two colonies were brought into contact, either fusion or rejection occurred. In the case of fusion, both the test and the blood vessels of one colony were united with those of the other to form a single colony. In the case of rejection, after about 24 hours the test cells in the contact area became opaque and slightly brown. This change was easily recognizable with the naked eye as a white line in the contact area between the two colonies. When contact was made between the growing edges of two fusible colonies, ampullae of each colony mutually extended into the test of the facing colony. By 24 hours or so tip-to-side contacts occurred between facing extended ampullae and there fusion took place, i.e., the blood vessels of the two colonies became inter- connected. Both by multiplying the number of fused ampullae and by reducing their size to become internal vessels, finally the original two colonies were com- pletely united into a single colony. It is a singular fact, as has already been pointed out by Bancroft (1903), that the fusion was never seen between the tips of ampullae. It always took place be- tween the tip and the side of ampullae. When two non-fusible colonies came into contact at their growing edges, the vascular ampullae of each colony actively extended into each other just like in the case of two fusible colonies. Meanwhile, however, a sign of rejection always appeared at the contact area. The first change detectable was that the ampullae and the test cells in the contact area became deep brown and opaque. Then, about two days after contact, the ampullae detached from the body of the colony and finally disintegrated. Botrylloides violaceus Botrylloides is also a member of the Botryllinae and has a common vascular system similar to that of Botryllus. Bancroft (1903) could not find the evidence of colony specificity in B. gascoi and B. leachi, but in our species the presence of it has been confirmed. If two colonies were placed with their cut surfaces in contact, two fusible colonies easily fused together, but non-fusible ones always rejected each other. In the case of rejection, the blood vessels and the test cells of each colony in the contact area turned black in color as in Botryllus and disintegrated in about half a day after contact. When two fusible colonies were allowed to grow naturally towards each other, ampullae of either colony actively extended into the test of the facing colony until they came into contact with the blood vessels. It was often observed that the tips of the ampullae being in contact with the blood vessels became inflated before fusion took place, as if by contact with the blood vessels the normal growth of the ampullae were being obstructed. Fusion took place after the colonies had been in contact for about a day (Fig. 1A). Also in Botrylloides, fusion never occurred between the tips of ampullae in each colony. 414 H. MUKAI AND H. WATANABE In contrast to Botryllus, two non-fusible colonies took indifference when they came into contact at their growing edges. Ampullae of both colonies bent upwards and remained pushing against each other for several days, as if they were striving to overcome each other. Usually, however, neither the growth of one colony over the other nor the rejection was recognized. Thus, the edges being in contact grew thicker than usual and a deadlock ensued (Fig. IB). A colony could readily be pulled off by a pair of forceps from the other, showing that no union of the test-. FIGURE 1. Fusion and indifference in Botrylloidcs riola-ccus. A shows fusion and B shows indifference. In A, arrows indicate the places where the two colonies have fused. In B, two colonies are pushing against each other. of each colony was accomplished between the two colonies. When they were separated after a long-term contact, however, occasionally a sign of rejection was found in some test cells of the contact surfaces. Symplegma re plans Symplegma also belongs to the Botryllinae and has a common vascular system similar to that of Botryllus. Among the colonies of S. reptans, the presence of colony specificity has been revealed. When there was contact between the cut surfaces of colonies, either complete fusion or complete rejection took place. The processes of fusion and rejection were essentially similar to those observed in Botryllus and Botrylloides. In the case of rejection, the blood vessels and the test cells in the contact area gave a necrotic appearance. When the growing edges of two fusible colonies came into contact, they pushed against each other. In most cases, the ampullae of both colonies remained being in COLONY SPECIFICITY IN ASCIDIANS 415 contact at their tips for about a day (Fig. 2A) and finally fusion took place between them (Fig. 2B). As is clear in Figure 2B, in this species tip-to-tip fusion of ampullae was rather common, though of course tip-to-side fusion was also observed in some cases. In Svuiplcy-ina the ampullae seem to be associated with one another too closely to allow the invasion of the ampullae of another colony. While, in Botryllus and Botrylloides, in which tip-to-side fusion is usual, the association seems to be too loose to prevent the invasion of the ampullae of other colony. This will be the reason why the tip-to-tip fusion occurs only in Symplegma. When the growing edges of non-fusible colonies came into contact, usually the ampullae of both colonies were pressed tightly against each other as in the case of fusion. About one or two days later, however, the distal parts of ampullae became dark and opaque. Then, the ampullae retracted almost invariably away B 0.5m m FIGURE 2. Fusion and rejection in Symplegma rcptuns. In A, two colonies are in con- tact at their growing edges. B shows fusion; fused ampullae are indicated by arrows. C shows rejection. Distal parts of ampullae (densely dotted) have been retracted leaving the empty test (sparsely dotted). The boundary between the two colonies is clearly detected. from the contact area, leaving there the empty test in which they had formerly extended. In most cases, the boundary of the two colonies was clearly seen under a binocular stereomicroscope. A stage of this rejection is illustrated in Figure 2C. Before the sign of rejection appeared, each colony being in contact could be easily separated by a pair of forceps from the other. Once it was observed, how- ever, they were no longer separable at the boundary line, thus indicating that the retraction of ampullae was brought about after the establishment of some union between the test matrices of both colonies. In the process of rejection, the breakdown of ampullae subsequent to their penetration into the test matrix of the other colony, which was the case in Botr\Uits, never occurred in Symplegma. In the colonies in which rejection took place, new ampullae always budded at the basal part of the retracted ampullae and continued to grow over them. The process of rejection presented above is applicable to the combination of 416 H. MUKAI AND H. WAT AN ABE colonies of equal strength, armed with the same weapons. If the growing edges of non-fusible colonies, one begin considerably thicker and containing more ampullae than the other, came into contact, the thinner edge was pushed over by the thicker edge instead of being opposed to it. Thus, indifference ensued without giving any evidence of rejection between them. In general, new ampullae just regenerated from the cut ends of blood vessels were less vigorous than the old ones. D id en i n u m moseleyi Didemninn belongs to the family Didemnidae ; the colony has no common vascular system in contrast to the above three species. Each zooid is being embedded solitarily in common test and has some vascular processes with faint circulation of blood, which terminate in enlarged bulbs or ampullae. The test usually contains numerous bladder cells, being characteristic of this family. In Didemnum also, the presence of colony specificity has been revealed. If we cut out two pieces from a colony and placed them with their cut surfaces in contact, they fused completely to form a single colony. About 24 hours after contact, the boundary between the original two pieces could not be detected. On the other hand, rejection resulted from contact between the cut surfaces of two pieces derived from different colonies. The following day after operation, numerous bladder cells existing in the contact area showed an opaque appearance, and the vascular processes had retracted away from that area. Though a union had been established between the test matrices of the two pieces, the boundary line between them could still be detected. Then, at last, the contact area collapsed and disintegrated. Only five colonies were at our disposal, all of which were mutually non-fusible. When two pieces derived from a single colony came into contact as a result of natural growth, they extended their vascular processes into each other. Thus, at first the two pieces were connected with a thin sheet of test containing vascular processes and later zooicls appeared in that area. In the course of this fusion, union between vascular processes never took place. The pieces derived from different colonies rejected each other when also the con- tact was made between their normal edges. In the actively growing edges of the colony, vascular ampullae were crowded. \Yhen two such edges being non-fusible were placed in apposition, most of the ampullae of each colony always retracted away from the surface shortly before they came into contact, thus giving the im- pression that the ampullae deprived the colony of its vigor from that area. Some of the ampullae of each colony, however, continued to grow7 and extended towards each other. After contact, the ampullae and their surrounding test became opaque and finally disintegrated. Two stages of this rejection are illustrated in Figures 3 A and 3B. Pcrophora orientalis Perophora belongs to the family Perophoridae, in which respective zooids are connected with one another only by basal stolonic vessels from which they have arisen as buds. The stolons are sometimes branched, but not interconnecting the individuals in a complex vascular network as in Botryllus. COLONY SPECIFICITY IN ASCIDIANS 417 From contact between the normal surfaces of two stolons, indifference always ensued regardless of their origin. That is, they continued to grow, adjoining- together side by side or crossing one over the other. When two stolon-pieces cut out from a colony were forced to come into contact closely at their cut ends, they easily fused, i.e., the blood could be distinctly recognized passing from one piece to the other. The same occurred even when two pieces derived from different colonies wrere used. We used five colonies in all ; all of the five colonies were mutually fusible. Thus, two stolons, not only of the same colony but also of differnt colonies, could be easily fused together by graft- ing. From these results, the conclusion seems to be justified that in a Pcrophora colony specificity is absent. 0.5mm B FIGURE 3. Rejection in Didcmnum moscleyi. In A, two vascular processes of the left- hand colony are extended on the right-hand colony. In B, disintegration occurred in the vascular processes and their surrounding test. Hctei'oc/cnetic combinations Using respective colonies of three species, Botryllus, Botrylloides and Sym- pleyma belonging to the Botryllinae mentioned before, heterogenetic or interspecific combinations were prepared. When the cut surfaces of the two colonies of different species were brought into contact, no special phenomenon could be observed between them. In the same way, when the two colonies of different species were allowed to grow towards each other, neither fusion nor rejection was seen to occur between them. They simply competed with each other for the substratum to grow upon. Thus, contact between the two colonies of different species always resulted in indifference. DISCUSSION The results from the present experiments, together with those of Oka and Usui (1944) on Polycitor colonies, are summarized in Table I. In the reaction 418 H. MUKAI AND H. WATANABE resulting from contact between two colonies, fusion, rejection and indifference are distinguishable. Of these, fusion and rejection are essential ; indifference is simply a defect of the other two. Colony specificity can be considered as a type of allogeneic recognition and manifests itself among others as a hindrance to fusion between two colonies. As is clear from Table I, colony specificity is present in some species, such as Botryllns, BotryUoidcs, Symplegma and Didemnum, but not in Perophora and Polycitor. According to Bancroft (1903), BotryUoidcs gascol and B. leachi colony specificity is absent, i.e., a union of two colonies is always established either by grafting or by natural growth, regardless of their origin. Discrepancy between the results ob- tained by Bancroft and the present experiments will be due to the limited number of his observations rather than to variance of species. TABLE I Summarizing representation of the results affusion experiments Contact between Species cut surfaces growing edges Botryllus primigenus Fusion Fusion Rejection Rejection Botrylloides violaceus Fusion Fusion Rejection Indifference Symplegma reptans Fusion Fusion Rejection Rejection (Indifference)** D idem n u >n moseleyi Fusion Fusion Rejection Rejection Peroph o ra o r ie ntal is Fusion Indifference Polycitor mutabUis* Fusion Indifference Heterogenetic combinations Indifference Indifference * Oka and Usui (1944). * When one very vigorous and one rather weak edge come into contact, sometimes indifterence ensues. In those species in which the presence of colony specificity is assured, the grow- ing edges of the same colony fuse together with the complete union of test and/or blood vessels whenever they come into contact. On the other hand, in those species in which colony specificity is absent, when the growing edges or the test surfaces of the same colony are brought into contact, indifference can be recognized. In summary, the presence or absence of colony specificity may be correlated with the ability or inability of fusion between the growing edges of the same colony. Judging from these facts, colony specificity seems to be a feature of common occurrence in those ascidians in which fusion of test and/or blood vessels always occurs between the growing edges of its own. The validity of this assumption in other colonial organisms should be examined by future researches. When two colonies of different species are brought into either by natural growth or by experimental means, they show indifference to each other. Thus, as has been pointed out by Oka (1970), fusion or rejection recognizable in ascidians is a feature specific within the species, though in certain cnidarians, heterogenetic COLONY SPECIFICITY IN ASCIDIANS 419 rejection has been reported (Kato, Hirai and Kakinuma, 1967; Theodor, 1970). As has been suggested by Oka and Usui (1944) with Policitor colonies, the test of ascidians seems to consist at least of two layers, one external and the other internal, the former being thin and tongh as compared with the latter. Taking the process of fusion or rejection between the growing edges into consideration, at least two steps will be separated. The first step is the elimination of the external layer of the test, which will be attained enzymatically. The second step is the reaction leading to the completion of fusion or rejection, which itself may consist of a series of reactions. The first step is either possible or impossible, being alternative. In the second step, however, it seems that the two processes or mechanisms of fusion and rejection are not necessarily alternative ; in a sense the former will be more basic and may be overlapped by the latter. In the case of rejection at the growing edges of Botryllns, for instance, the two colonies being in contact extend their ampullae into each other as in the case of fusion ; before the union of vascular systems takes place, however, the ampullae begin to disintegrate. Judging from these facts, it will be the case that the process of fusion is being interrupted by that of rejection, if it is present. The former is completed only when the latter is absent. Provided that both the first and the second steps can proceed after contact of tissues, which seems to be the case at least in the botryllids, the following will be justified: In the contact between growing edges, if the first step is lacking the second step is not realized, when indifference ensues. On the contrary, in the contact between cut surfaces the first step is experimentally attained, when either fusion or rejection occurs. On the basis of the above discussion, the results obtained with respective species will be interpreted below. In Polycitor iinttabilis and Pcrophora oricntalls, in which colony specificity is absent, the mechanism of the elimination of the external layer is completely lacking. Accordingly, when there is contact between the test surfaces indifference ensues. When there is contact between the cut surfaces, however, fusion occurs, because they are naked, i.e., there is no layer between them. In Botr\llus primigenus the elimination of the external layer is always complete when two colonies come into contact at their growing edges, without reference to their relation, fusible or non-fusible. Therefore, in this species the first step is not colony specific. Colony specificity exists only in the second step, and either fusion or rejection is attained. The results obtained in Didemnum- uwselcyi are similar, seemingly at least, to those of B. primigenus. In Didcnmnin, however, most of the vascular ampullae of non-fusible colonies retract away from the surface shortly before they come into direct contact. Only some of them grow upon the other colony and undergo disintegration. These results may be taken to indicate that in this species the substances participating in rejection can pass, to some extent at least, out of the colony. No definite conclusion, however, can be drawn, until suitable studies are carried out. Botr\lloides violaccus will be worthy of particular notice. In this species rejec- tion occurs only when contact is made between the cut surfaces of non-fusible colonies ; indifference is always obtained when they come into contact at their 420 H. MUKAI AND H. WATANABE growing edges. In contrast to that, when fusible colonies are brought into contact either by means of natural growth or by grafting, fusion invariably takes place Thus, in this species both the first and the second step will be considered as colony specific. It should be noticed here that in B. violaceus the specificity of the first step is completely correlated with that of the second step, in other words, fusion and indifference between growing edges perfectly correspond to fusion and rejec- tion occurring between cut surfaces, respectively. How this correlation is governed is an open question. In Symplcgma rcptans rejection between growing edges takes place only when two colonies remain pushing against each other for a definite period. In this process neither the penetration of ampullae of one colony into the other nor the extinction of the boundary between them is observed. Once any sign of rejection is seen, however, the two colonies cannot be artificially separated from each other indicating the establishment of some union between the test matrices. Since observa- tions on this species are limited, we cannot say indiscreetly whether the elimination of the external layer of the test, i.e., the first step of rejection, is brought about by means of chemical action which may be the case with Botrylhts, or by means of physical action as a result of pushing against each other. Provided that the latter is the case, both the first and the second step will be colony specific. More recently, the significance of the colony specificity in relation to the evolution of adaptive immunity of vertebrates has been discussed by Burnet (1971). On the basis of the above discussion and from the view point of the evolution of colony specificity, the colonial organisms may be classified into the following four groups. (1) Colonies have neither the specificity nor the structural barrier to fusion. Some sponges may belong to this group. (2) An external layer which prevents fusion with one another is formed, but the specificity is still lacking. In this group, to which Pcrophora and Polycitor belong, the individuality is assured simply by the structural barrier. (3) Specificity in the second step of fusion is acquired. In those species in which the substances participating in rejection can pass through the external layer of the colony or in which no such a layer is dif- ferentiated, rejection may take place without colonies coming into direct contact. Such species also, if they exist, should be classified in this group. Thus, Botryllus and Didemnnm belong to this group. (4) Colony specificity is established in both the first and the second steps of fusion. To this group Botrylloidcs and Symplcyma (?) belong. It is our present duty to acknowledge our indebtedness to the staff of the Shimoda Marine Biological Station or the Usa Marine Biological Station. SUMMARY 1. The presence or absence of colony specificity, i.e., the recognition of self and not-self in colonial organisms, has been investigated with several species of compound Ascidians. If the reaction resulted from contact either between grow- ing edges or between cut surfaces of colonies, fusion, rejection and indifference have been distinguished. Of these three cases, indifference means simply a defect of the other two. Both fusion and rejection are specific within the species. 2. The presence of colony specificity has been demonstrated in Botryllus COLONY SPECIFICITY IN ASCIDIANS 421 primigenns, Botrylloides violaceus, Symplegma reptans and Didemnum ntoseleyi. But, in Pcroplwra oricntalis colony specificity is absent. 3. From the above facts, it has been suggested that the colony specificity may be a feature being common to those ascidians in which fusion of test and/or blood vessels always occurs between the growing edges of its own. 4. In the process of fusion or rejection at the growing edges, two steps have been distinguished. The first step is the elimination of the external layer of test, which is followed by the second step terminating in the completion of fusion or rejec- tion. In BotryUus and Didcinmun, only the second step is colony specific. In Botrylloides, however, both the first and the second step are colony specific. 5. A possible evolutionary trend of colonial organisms in relation to the colony specificity has been represented. LITERATURE CITED BANCROFT, F. W., 1903. Variation and fusion of colonies in compound ascidians. Proc. Calif. A cad. Sci., Scries 3, 3 : 137-186. BERRILL, N. J., 1950. The Titnicata with an Account of tJic British Species. Ray Society, London. BURNET, F. M., 1971. "Self-recognition" in colonial marine forms and flowering plants in relation to the evolution of immunity. Nature, 232 : 230-235. HAUENSHILD, C., 1954. Genetische und Entwicklungsphysiologische Untersuchungen viber Inter- sexualitat und Gewebevertraglichkeit bei Hydractinia cchiiiata (Flem.). Arch. Enhv. mcch. Org., 147: 1-14. HAVENSHILD, C., 1956. Uber die Vererbung einer Gewebevertraglichkeitseigenschaft bei dem Hydroidpolypen Hydractinia. Z. Natitrforscli., 11 : 132-138. IVKHK, F. S., 1966. Histoincompatibility and stolon overgrowth between interbreeding strains of Hydractinia cchinata. Biol. Bull., 131 : 393. IVKER, F. S., 1967. Localization of tissue incompatibility and specificity of the overgrowth reaction Hydractinia echinata. Biol. Bull., 133: 471-472. IVKER, F. S., 1972. A hierarchy of histo-incompatibility in Hvdractinia echinata. Biol. Bull., 143: 162-174. KARAKASHIAN, S., AND R. MILKMAN, 1967. Colony fusion compatibility types in BotryUus schloesseri. Biol. Bull., 133: 473. KATO, M., E. HIRAI AND Y. KAKINUMA, 1967. Experiments on the coaction among hydrozoan species in the colony formation. Sci. Rep. Tolwku Univ., Series 4, 33 : 359-373. OKA, H., 1970. Colony specificity in compound ascidians. The genetic control of fusibility. Pages 195-200 in H. Yukawa, Ed., Profiles of Japanese Science and Scientists. Kodansha, Tokyo. OKA, H., AND M. Usui, 1944. On the growth and propagation of the colonies in Polycitor i/iittabilis (Ascidiae compositae). Sci. Rep. Tokyo Bunrika Daigaku, Scries B, 23-53. OKA, H., AND H. WATANABE, 1957. Colony-specificity in compound ascidians as tested by fusion experiments (a preliminary report). Proc. Jap. Acad., 33: 657-659. OKA, H., AND H. WATANABE, 1960. Problems of colony-specificity in compound ascidians. Bull. Mar. Biol. Stat. Asannislii, 10: 153-155. OKA, H., AND H. WATANABE, 1967. Problems of colony specificity, with special reference to the fusibility of ascidians (in Japanese). Kagakn (Tokyo), 37: 307-313. RASMONT, R., 1970. Some new aspects of the physiology of freshwater sponges. Pages 415- 422 in W. G. Fry, Ed., The Biology of the Porifera. Academic Press, London. RYLAND, J. S., 1970. Bryosoans. Hutchinson University Library, London, 175 pp. SABBADIN, A., 1962. Le basi genetiche della capacita di fusione fra colonies in BotryUus schlosseri (Ascidiacea). Rend. Accad. Naz. Lined, Scries 8, 32: 1031-1035. THEODOR, J. L.. 1970. Distinction between "self" and "not-self" in lower invertebrates. Nature, 227: 690-692. TOTH, S. E., 1967. Tissue compatibility in regenerating explants from the colonial marine hydroid Hydractinia echinata (Flem.). /. Cell Pliysiol., 69: 125-132. Reference: Biol. Bui!., 147: 422-432. (October, 1974) LOCOMOTOR ACTIVITY RHYTHMS OF JUVENILE ATLANTIC SALMON (SALMO SALAR} IN VARIOUS LIGHT CONDITIONS NANCY E. RICHARDSON AND JAMES D. McCLEAVE Department of Zoology, University of Maine, Orono, Maine 04473 Endogenous circadian components in the activity rhythms of fishes have not been convincingly demonstrated. Various circadian oscillations, subject to phase setting primarily by natural or artificial light cycles, have been documented in several fishes. This synchronizing effect of light cycles on the locomotor activity patterns of salmonids, in particular, has been cited in several instances (Ali, 1964; Swift, 1964; Byrne, 1968; Varanelli and McCleave, 1974). Whereas circadian rhythms in certain mammals and birds have been known to free-run in constant conditions for months at a time (Aschoff, 1960, 1966), the rhythms termed "endogenous" in various fish studies have usually not been apparent for more than a few days. Activity rhythms have persisted for two or three days in constant conditions in juvenile Atlantic herring, Chipea harengus, (Stickney, 1972), juvenile sockeye salmon, Oncorhynchus nerka, (Byrne, 1968), juvenile Atlantic salmon, Salmo sal or, (Ali, 1964), and in a European minnow, Lencaspius delineatiis, (Seigmund and Wolff, 1973). The adult sea lamprey, Petromyson mar inns, maintained a circadian activity rhythm for five days in constant dim light at which time observations were discontinued (Kleerekoper, Taylor and Wilton, 1961). An exceptional case is the circadian rhythm of the swell shark, Cephaloscyllium ventriosum, which continued for about 15 days in both constant light and constant darkness (Nelson and Johnson, 1970). The same authors did not observe free- running rhythmicity in the horn shark, Heterodontus francisci. Circadian rhythms of swimming speed in the bluefish, Pomatomus saltatrix, have been reported to dissipate after two or three days in constant dim light, but later to become reestablished (Olla and Studholme, 1972). A similar reestablishment of rhythmicity has been suggested for Atlantic salmon (Ali, 1964). Gibson (1971) found that exposure to light-dark cycles was necessary for as long as two to four months before circadian rhythms which would persist in con- stant darkness could be entrained in blennies, BIcnniiis pholls. Activity rhythms with a 12 hr period were easily entrained in the same fish by changes in hydro- static pressure resulting from local tidal cycles. A circadian rhythm of electric organ discharge has been observed to free-run in constant dim light in the electric gymnotid, Gymnorhamphichthys hypostomous, (Lissmann and Schwassmann, 1965). The appearance of the corresponding loco- motor activity rhythm in these fish seems dependent upon the presence of a light- dark cycle and the natural environmental substrate. It was concluded that the activity rhythm is not a good indicator of the endogenous oscillation (Schwassmann, 1971). Locomotor activity patterns in juvenile Atlantic salmon were observed by 422 LOCOMOTOR ACTIVITY RHYTHMS OF SALMON 423 Varanelli and McCleave (1974). Photoperiod and temperature were altered in different experiments to approximate concurrent seasonal conditions. Activity rhythms became synchronized to the imposed light-dark cycle, but in most cases failed to persist in constant conditions. The present study was designed to gain more information about the effects of light regimes on the activity patterns of Atlantic salmon by subjecting larger numbers of fish to the same experimental conditions. About two thirds of the fish were simultaneously tested for their responses to weak extremely low fre- quency electric and magnetic fields. Since these fields were found to have no effect on activity (Richardson, McCleave and Albert, unpublished data), fish ex- posed to the fields are combined with control fish in this paper. MATERIALS AND METHODS Between June 1972 and June 1973, 16 experiments involving 192 fish were car- ried out. Usable records were obtained from 177 fish. In each experiment the locomotor activity of 12 fish was individually recorded for 10 days, though a few- fish records were less than 10 days due to equipment failures. The fish were exposed to a 12 hr light- 12 hr dark cycle ( LD 12: 12) in eight experiments, to constant light (LL) in four experiments, and to constant darkness (DD) in three. In one experiment, a 23 hr cycle containing 1 hr of light (LD 1:22) was imposed. Sea run Atlantic salmon parr between 13 and 20 cm total length were obtained from the Craig Brook National Fish Hatchery in East Orland, Maine. Fish were placed in individual activity chambers as soon as they arrived at the university and were allowed two days to acclimate to the apparatus. The apparatus and recording methods were similar to those described by Varanelli and McCleave (1974). The activity chambers were circular channels 10 cm wide by 27 cm deep, built from two concentric polyethylene cylinders, 47 cm and 27 cm in diameter. Air was bubbled into the center of the inner cylinder and diffused into the channel through holes in its inner side. Two activity chambers were housed in each of six light-tight water baths. The chambers were water-tight and opaque so that fish in the same water bath had no visual or chemical contact with one another. Water was circulated through the baths from a refrigeration unit which maintained the temperature at 15 ± 1° C. Temperature was continuously recorded in one of the chambers. Activity was recorded electromechamcally. Two plastic probes were sus- pended in opposite sides of the swim channel. The ends of the probes hung 4 cm off the bottom. The upper portion of each probe was attached to a piece of copper braid which hung through a carbon ring. When a fish moved the probe, contact was made with the carbon ring completing an electric circuit. This caused one count to be registered by an automatic counter. The two probes from each chamber were wired in parallel to one counter, which printed 10 times per hour. Counter tapes were removed once each day and the activity counts were coded directly for computer processing. Fish were not fed during the experiments to eliminate inducement of activity cvcles based on feeding. Experiments were carried out in a basement laboratory 424 X. E. RICHARDSON AND J. I). McCLEAVE which was entered only once each day at 1200 EOT. Efforts were made to isolate the fish from as many environmental disturbances as possible. Light was provided by either fluorescent or incandescent bulbs mounted on the lid of each water bath. Illumination ranged between 1 and 15 lux at the water surface depending on the area of the tank measured. In the LD 12: 12 experi- ments, automatic timers switched the lights on at 0600 EDT and off at 1800 EDT each day. Computer drawn plots of the hourly activity for each fish over the course of each day were obtained. Composite plots encompassing all 10 days were also TABLE I Summary of locomotor activity patterns and periodicities of Atlantic salmon parr in LD 12:12 Experiment date Xumher of salmon Range of period lengths Mean period length Diurnal Nocturnal Light change Aperiodic 6/22-7 / 2/72 11 0 1 0 23.8-24.7 24.1 7/ 7-7/17/72 5 0 6 0 23.9-24.0 24.0 7/21-7/31/72 6 0 4 0 24.0-24.2 24.1 8/ 4-8/14/72 8 1 0 0 24.0 24.0 2/15-2/25/73 1 8 3 0 24.0-24.5 24.1 3/ 3-3/13/73 3 4 4 1 24.0-24.5 24.0 4/15-4/25/73 1 5 5 0 23.9-24.6 24.1 5/16-5/26/73 0 2 9 0 23.9-24.0 24.0 produced for each fish. These plots were visually examined to determine the nature of individual activity patterns. Data from fish with similar patterns were combined to produce plots illustrating typical patterns. Activity records were analyzed for periodicity by the periodogram method of Enright ( 1965 ) . This method is appropriate for serially correlated data because it does not require random independent observations. The periodogram method calculates amplitudes for a series of possible period lengths within the range of interest. The amplitude is the standard deviation of the hourly means of activity. The hourly means used for each amplitude were drawn from arrangement of the data based on the assumption that a given period existed. If a true periodicity exists the amplitude produced by the proper assumed period length will be clearly larger than the other amplitudes. Computer calculated amplitudes were obtained for assumed period lengths between 3.0 and 33.0 hr in increments of 0.1 hr. LOCOMOTOR ACTIVITY RHYTHMS OF SALMON RESULTS 425 LD 12 : 12 Periodogram analysis showed that fish subjected to a 12:12 light-dark cycle possessed a rhythmicity in locomotor activity with a period length very close to 24.0 hr (23.8-24.7) (Table I, Fig. la). Only one fish in 88 was aperiodic. All periodic fish exhibited one of three typical patterns : diurnal, nocturnal, or active primarily at times of light-dark transitions. From all eight LD experiments 80 -. 70 - 60 - 50 - 110 - 30 - 20 - 10 H 0 80 -, 70 - 60 - 50 - 10 - 30 - 20 - 10 - 0 a: 80 -i 70 - 80 -i 70 - 60 - 50 - HO - 30 - 20 - 10 - fl -] — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i i i i r 11 6 8 10 12 11 16 18 20 22 21 26 28 30 32 B i — i — i — i — i — i — i — i — i — i — \ — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i i i i r 1 6 8 10 12 11 16 18 20 22 21 26 28 30 32 60 - 50 - 10 - 30 - 20 - 10 - n - u J^ ^ _-J« H- -H. :}T 1 6 8 10 12 11 16 18 20 22 21 26 28 30 32 D T — I — I — I — I — I — 1 — I — I — I — 1 — 1 — 1 — I — I — I — 1 — I — I — 1 — 1 — 1 — 1 — 1 — 1 — 1 — I — I — 1 — 1 — i 1 6 8 10 12 11 16 18 20 22 21 26 28 30 32 PERIOD LENGTH- HR FIGURE 1. Sample periodograms for (a) a diurnal Atlantic salmon in LD 12:12. (b) a light-change-active salmon in LD 12:12, (c) an aperiodic salmon in DD, (d) salmon in DD with 24.0 hr periodicity. (Periodograms were calculated in steps of 0.1 hr, hut only hourly points are plotted for clarity). 426 N. E. RICHARDSON AND J. D. McCLEAVE 35 fish were diurnal, 20 were nocturnal and 32 were active at light change (Fig. 2). Mean percentages of daily activity occurring when lights were on ranged from 53 to 94% for diurnal fish, when lights were off from 62 to 86% for nocturnal fish, and in the 3 hr following lights on and lights off from 28 to 55% for light- change-active fish (Fig. 3). Expected percentages, if activity were equally dis- tributed throughout the day, were 50%, 50%, and 25%, respectively, for diurnal fish, nocturnal fish, and light-change-active fish. In all three patterns the most abrupt increase in hourly activity occurred during 10- 8- 6- LJ cr 2H Li °- I— I CD u_ 8H CO LU <_D 4 £ 2H R or o_ 10- ^ 8: ID 6- ° 4- CE 2H LU i i i i i i i i i i ! i i r i i i i i i i i r 12 14 16 18 20 22 24 2T 4 6 8 10 B i i i i i i i i i i i i i i i i i i i i i i i r 12 14 16 18 20 22 24 2 4 6 8 10 c I I 1 I I I I 1 I I 1 1 I I I I I I I I 1 I I I 12 14 16 18 20 22 24 2 4 6 8 10 HOUR GF DRY FIGURE 2. Composite plots of mean hourly percentages of daily activity ± 1 standard error for Atlantic salmon in LD 12:12 showing the three activity patterns; (a) light-active fish, N = 35, (b) dark-active fish, N = 20, (c) light-change-active fish, N = 32. Dark bar indicates lights off. LOCOMOTOR ACTIVITY RHYTHMS OF SALMON 427 lOO-i 80- 60- 40- 20- 0- R I I I I I 1 I ; i 1 I I | 1 CJ 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 cr 100-1 80- cr^ 60H Q£ bJ LU O DC O 0- B I I I I 1 I I 1 I I I I 1 I I I ! 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Q_ O 80- ^ "— 60- cr CO or 40- 20- c i i i i i i i i i i i \ i i 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 FISH NUMBER FIGURE 3. Mean percentage of daily activity ±1 standard error in LD 12:12 (a) of diurnal Atlantic salmon when lights were on, (b) of nocturnal salmon when lights were off, (c) of light-change-active salmon in the 3 hr following lights on and lights off. Only those salmon for which a full 10 day record was available were used. the hour immediately following lights off (Fig. 2). Diurnal and light-change-active fish showed a second burst of activity in the hour following lights on. Nocturnal fish showed a less abrupt response to lights on in the form of decreasing activity over several hours. Most light-change-active fish ( 30 of 32 ) exhibited a pattern very similar to that depicted by the composite plot (Fig. 2c). The least amount of activity occurred during the 6 hr preceding lights off. An abrupt rise at lights off was followed by 428 N. E. RICHARDSON AND J. D. McCLEAVE an intermediate level of activity during the dark interval. A second burst at lights on was followed by a higher level of activity for the next 6 hr. The other two fish in this group had high levels of activity only at times of light change and were relatively inactive at other times. Two fish showed some evidence of anticipating lights on by an increase in activity during the 2 hr preceding the transition. The periodogram of light-change fish contained distinct secondary maxima at the period length of 12.0 hr (Fig. Ib). The periodograms of nocturnal or diurnal fish occasionally showed secondary peaks which were less distinct. Some diurnal fish showed recurring bursts of activity shortly after midnight which lasted for one or two hours. Four LD experiments were carried out during the summer of 1972. Most fish (30 of 42) were diurnal (Table I). Eleven fish were light-change-active and only one was nocturnal. In contrast, Atlantic salmon tested during the previous summer in LD cycles of length approximating the natural cycle were about equally distributed among the three activity patterns (Varanelli and McCleave, 1974). Four additional LD experiments of the present study took place during the winter and spring of 1973. In these experiments 19 of 46 fish were nocturnal, 21 were light-change-active and five were diurnal. The trend was from mostly nocturnal activity in the winter to more light-change activity in the spring. Constant conditions Fish subjected to LL or DD were similar in being generally aperiodic (Fig. Ic). Intermediate levels of activity were distributed fairly equally over the day in the form of sporadic bursts. Five fish (of 30) in DD produced periodograms which had peak amplitudes at period values near 24.0 hr (23.5-24.2 hr) (Fig. Id). In only two of these cases could a pattern be found by visual inspection of the daily plots. These two series of plots suggested recurring peaks near subjective evening. No fish in LL showed any evidence of periodicity. The mean of the total activity counts was computed for each group of experi- mental fish. The mean from the LL experiments (36,670) was about twice as large as the mean from DD (17,240). The mean from LL experiments was close to the LD mean (31,100). The highest mean total activity (45,000) came from fish subjected to 1 hr of light (LD 1:22). Short signal entrainment Twelve fish were exposed to a 1 hr light stimulus beginning every 23 hr for 10 days in an attempt to initiate a 23 hr rhythmicity of activity. Most fish (11 of 12) were aperiodic. In one case a maximum amplitude occurred at 24,0 hr, but no pattern was obvious from the plotted data. DISCUSSION Locomotor activity patterns of juvenile Atlantic salmon were entrained by an artificial light-dark cycle. The fact that about one third of the fish had activity induced at times of light change, while the other two thirds were divided between dark-active and light-active suggests that the transition stimulus rather than the LOCOMOTOR ACTIVITY RHYTHMS OF SALMON 429 light intensity may be of primary importance in synchronizing the activity rhythms of these fish. Fish in this study generally did not show an anticipatory change in activity prior to the onset of light or darkness. Byrne (1968) reported that sockeye salmon showed a "pre-dawn" increase and a "pre-dusk" decrease in activity. Pre-dark increases in activity have been observed in the nocturnal swell shark (Nelson and Johnson, 1970) and in juvenile Atlantic herring (Stickney, 1972). Atlantic salmon showed a consistently greater response following a light to dark transition than following a dark to light change (Fig. 2). This was true even of light-active fish. Maximum activity in juvenile herring occurs at sunset at a time of critical light intensity (Stickney, 1972). Sunset is the primary syn- chronizer of activity rhythms of two Arctic sculpins Cottus gobio and Coitus poecilopus (Andreasson, 1969, 1973). In contrast, maximum bursts of activity occur at sunrise in brown trout, Sahno tmtta, (Swift, 1964), the blenny, Coryphoblennhts galerita, (Gibson, 1970) and bluefish (Olla and Studholme, 1972). The nature of light responses in juvenile Pacific salmon is variable and complex (Hoar, Keenleyside and Goodall, 1957). Although different species exhibited preferences for either bright areas (Onchorhynchns kcta, O. gorbuscha) or dark- areas (0. kisittch, O. ncrka), all individuals continually ventured into both light and dark regions of the test apparatus. Juvenile Pacific salmon also change their light responses during development. Byrne (1968) found that sockeye salmon are nocturnal for the first two weeks after emergence and then become diurnal. Pink salmon fry, 0. gorbuscha, lose their photonegative behavior after two months (Hoar, Keenleyside and Goodall. 1957). Coho salmon fry, O. kisutch, are indifferent to light, but the smolts are photo- negative. Coho and sockeye salmon smolts are more light sensitive than the fry and seek out deeper darker areas (Hoar, Keenleyside and Goodall, 1957). Byrne (1968) found that LL facilitated the expression of a free-running rhythm in sockeye salmon whereas DD inhibited it. This is opposite to what AH (1964) found for Atlantic salmon. Gibson (1971) entrained rhythms in the blenny. Blcnnlns plwlis, that persisted in DD after several months of exposure to LD. Another blenny, Coryphoblcnnlns galerita, exhibited a 12 hr tidal activity rhythm in DD (Gibson, 1970). Nelson and Johnson (1970) found that neither LL nor DD inhibited the free-running rhythm of the swell shark for about 15 days. Several fish having indications of rhythmicity in DD were noted by us and by Varanelli and McCleave ( 1974). The majority of fish in some experiments was diurnal while in others it was nocturnal or light-change-active. Also, different individuals tested simultaneously in the same conditions showed different patterns of activity. Varanelli and McCleave (1974) found that slightly more Atlantic salmon tested during the sum- mer of 1971 were nocturnal than were diurnal or light-change-active. In contrast, we found that during the following summer most fish were diurnal. During the winter and spring of 1973 the majority of fish were nocturnal or light-change-active and very few were diurnal (Table I). Because Varanelli and McCleave (1974) used various photoperiods and temperatures to approximate environmental condi- tions, it is possible that some interaction among photoperiod, light intensity and temperature was responsible for the difference in results between the two summers. 430 N. E. RICHARDSON AND J. D. McCLEAVE This does not explain why different experiments with the same conditions produced different results or why fish tested at the same time showed different behavior patterns. Byrne (1968) found that a photoperiod-temperature interaction was respon- sible for changes in the activity patterns of juvenile sockeye salmon. While test- ing fish in several photoperiod-temperature combinations, he found that an upper temperature limit existed for each photoperiod. When this temperature was exceeded the fish changed from light-active to dark-active. High temperatures ( 10° C) coupled with short photoperiods (LD 8: 16) caused an increase in noc- turnal activity. Seasonal changes in activity pattern have been found in Arctic populations of Cottus gobio and C. poccllopus (Andreasson, 1973), brook trout, Sak'diniis foiitinalis (Eriksson, 1972), and burbot, Lota lota, (Miiller, 1973). The trout is day-active all year except for a desynchronized interval in summer. The sculpins and burbot are day-active in winter and night-active in summer (Andreasson, 1973 ). During the phase shift C. poccllopus passes through an interval when peak activity occurs at sunrise and sunset. Andreasson (1973) suggested that the activity rhythms of these fish are controlled by two separate oscillators, one light-active and one dark-active. A seasonal change in the phase angle between the two oscillators, caused by exogenous factors, is responsible for the change in activity pattern. Muller (1973) found that the duration of activity in burbot in the Arctic is controlled by day length in winter and by the night length in summer. The activity time cannot exceed 10-11 hr. When the day length or night length exceeds this limit, the fish switches to nocturnal or diurnal activity, respectively. The phase of a circadian rhythm can sometimes be shifted by a single perturba- tion in the light, temperature, or other regime. A single chemical stimulation via trout scent initiated a circadian activity rhythm in the adult sea lamprey which persisted in constant dim light until observations were stopped after five clays (Kleerekoper, Taylor and Wilton, 1961). Fish in the present study failed to become entrained by an hour long light signal recurring periodically every 23 hr. Neither did they show the abrupt increase in activity following the light to dark transition observed in the LD experiments. Maximum outbursts of activity did not coincide with the hour of light exposure. The hour of light exposure began in the evening and occurred an hour earlier each day over the course of the experiment (ten days). This signal may have been presented at a point in a rhythm of light sensitivity at which it could not be effective. Bruce (1960) has generalized that a light presented during an animal's subjective day may cause little or no phase shift. Because these fish have shown abrupt responses to LD transitions in other experiments, the synchronizing regime applied in this case was probably inappropriate. More information is needed about the variation in light responses and the limits of entrainability of salmonids to clarify the phase relationships between their activity rhythms and the oscillations of their environment. Indicator processes other than locomotor activity may be better suited for the demonstration of endogenous or non-visual exogenous com- ponents of circadian rhythms. LOCOMOTOR ACTIVITY RHYTHMS OF SALMON 431 We thank Mr. Enoch H. Albert and Dr. Sentiel A. Rommel, Jr. for technical assistance. The Craig Brook National Fish Hatchery provided the Atlantic sal- mon and the Computing and Data Processing Services of the University of Maine provided computer time. The project was supported by Office of Naval Research contract # NOOO 14-72- C-0 130 and NIH research grant # NS 11276 from the National Institute of Neurological Diseases and Stroke to T- D. McCleave. SUMMARY Atlantic salmon parr exposed to a 12 hr light- 12 hr dark cycle (LD 12: 12) for ten days were entrained to a 24.0 hr periodicity in locomotor activity. Thirty- five fish were light-active, 20 were dark-active and 32 were active primarily when lights were turned on or oft". Fish maintained in constant conditions (75) were generally aperiodic. Five fish (of 30) in constant darkness (DD) showed evidence of 24.0 hr periodicity. Twelve fish exposed to a light signal of 1 hr duration recur- ring every 23 hr failed to become entrained. Fish in constant light (LL) showed more activity than fish in DD. The results suggest ( 1 ) that light-dark transitions are important in synchronizing locomotor activity rhythms and (2) that locomotor activity is not a good indicator of possible circadian oscillations in this species. LITERATURE CITED ALT, M. A., 1964. Diurnal rhythms in the rates of oxygen consumption, locomotor and feeding activity of yearling Atlantic salmon (Salmo salar*) under various light conditions. Proc. Indian Acad. Sci. Section B, 60 : 249-263. ANDREASSOIST, S., 1969. Locomotory activity patterns of Coitus poccilopus Heckel and Coitus gobio L. (Pisces). Oikos, 20: 79-94. ANDREASSON, S., 1973. Seasonal changes in diel activity of Coitus poccilopus and Cottus gobio (Pisces) at the Arctic Circle. Oikos. 24: 16-23. ASCHOFF, J., 1960. Exogenous and endogenous components in circadian rhythms. Cold Spring Harbor Symp. Quant. Biol, 25 : 11-28. ASCHOFF, J., 1966. Circadian activity with t\vo peaks. Ecology. 47 : 657-672. BRUCE, V. G., 1960. Environmental entrainment of circadian rhythms. Cold Spring Harbor Symfi. Quant. Biol, 25 : 29-47. BYRNE, J. E., 1968. The effects of photoperiods and temperature on the daily patterns of locomotor activity in juvenile sockeye salmon (Oncorhyiiclius nerka). Ph.D. tlicsis, University of British Columbia. Vancouver, 125 p. ENRIGHT, J. T., 1965. Accurate geophysical rhythms and frequency analysis. Pages 31-42 in J. Aschoff, Ed., Circadian Clocks. Xorth Holland Publishing Co., Amsterdam. ERIKSSON, L., 1972. Die Jahresperiodik augen-und pinealorganloser Bachsaiblinge Sali'elinits fontinalis Mitchell. Aquilo Scr. Zoo/., 13 : 8-12. GIBSON", R. N., 1970. The tidal rhythm of activity of Coryphoblennius galcrita L. (Teleostei, Blennidae). Anim. Bchav.. 18: 539-543. GIBSON, R. N., 1971. Factors affecting the rhythmic activity of Blcnnius pholis L. (Teleostei). Anim. Behai'.. 19 : 336-343. HOAR, W. S., M. H. A. KEENLEYSIDE AND R. G. GOODALL, 1957. Reactions of juvenile Pacific salmon to light. /. Fish. Res. Board Can.. 14 : 815-830 KLEEREKOPER, H., G. TAYLOR AND R. WILTON, 1961. Diurnal periodicity in the activity of Petromvzon marinas and the effects of chemical stimulation. Trans. Amcr. Fish. Soc., 90 : 73-78. LISSMANN, H. W., AND H. O. SCHWASSMANN, 1965. Activity rhythm of an electric fish, Gymnorhampkichthys hypostomoiis Ellis. Z. Vcrgl. Physiol.. 51 : 153-171. MULLER, K., 1973. Seasonal phase shift and the duration of activity time in the burbot Lota lota. J. Coinp. Physio!.. 84 : 357-359. 432 N. E. RICHARDSON AND J. D. McCLEAVE NELSON", D. R., AND R. H. JOHNSON, 1970. Diel activity rhythms in the nocturnal bottom- dwelling sharks Hctcrodontus francisci and Cephaloscyllium rentriosum. Cofeia, 1970: 732-739. OLLA, B. L., AND A. L. STUDHOLME, 1972. Daily and seasonal rhythms of activity in the bluefish (Poinatoinus saltatrix). Pages 303-325 in H. E. Winn and B. L. Olla, Eds., Behavior of Marine Animals, Vol. 2, Vertebrates. Plenum Press, New York. SCHWASSMANN, H. O., 1971. Biological rhythms. Pages 371-416 in W. S. Hoar and D. J. Randall, Eds., Fish Physiology, Vol. 6. Academic Press, New York. SEIGMUND, R., AND D. L. WOLFF, 1973. Circadian-Rhythmik und Gruppenverhalten bei Lencaspuis ddincatus. Expcricntia, 29: 54-58. STICKNEY, A. P., 1972. The locomotor activity of juvenile herring (Clupca harcngits) in response to changes in illumination. Ecology, 53 : 438-445. SWIFT, D. R., 1964. Activity cycles in the brown trout (Sahno tnitta) 2. Fish artificially fed. /. Fish. Res. Board Can., 21 : 133-138. VARANELLI, C. C, AND J. D. MCCLEAVE, 1974. Locomotor activity of Atlantic salmon parr (Sahno salar) in various light conditions and in weak magnetic fields. Aniin. Bchav., 22: 178-186. ' Reference: UioL Bull., 147: 433-442. (October, 1974) DIETARY FACTORS STIMULATING OOGENESIS IN AEDES AEGYPTI ANDREW SPIELMAN AND JOANN WONG Department of Tropical Public Health, Harvard School of Public Health. 665 Huntington Avenue, Boston, Massachusetts 02115 Vertebrate blood provides hematophagous insects with a unique oogenic stim- ulus, a relationship that has long intrigued students of mosquito biology but has not been fully explained. One unresloved issue concerns the central question of the nature of the oogenic stimulus. A classical study ascribed this stimulus to the physical stretching of the midgut resulting from engorgement on vertebrate blood (Larsen and Bodenstein, 1959). However, a more recent report presents con- tradictory evidence and suggests that only the nutrient content of the blood-meal is crucial (Bellamy and Bracken, 1971). If this were true it would require the mosquito to assess the potential nutrient available in its midgut even before apparent digestion had begun, and this information would have to be transmitted to the brain within a few minutes of feeding (Clements, 1956). No mechanism for such a rapid assessment seems evident. Accordingly, we re-examined this basic problem. The objective of the present study was to compare the roles of various physical and chemical properties of the blood meal in stimulating Acdcs acg\pti to commence vitellogenesis. MATERIALS AND METHODS Mosquitoes were obtained from a colony of Aedcs acyypti isolated on Grand Bahama Island in 1972 and maintained at 24-26° C, 70% R.H. and with 16 hours of light per day. Larvae were reared on Purina guinea pig chow and pharate adults separated as to sex. Virgin, female mosquitoes were provided raisins as food and used in the experiments at 3-5 days after adult ecydsis. One day prior to the experi- ment, food and water were removed. In experiments requiring measurement of the quantity of blood ingested, non- anesthetized mosquitoes were transferred to a tared vial via an aspirator and weighed on a Sartorius semi-micro balance accurate to 0.01 mg. After weighing, mosquitoes were permitted to feed on a human host until suitably engorged. Imme- diately following feeding, each mosquito was re-weighed and transferred to indi- vidual holding chambers. Various solutions were introduced into the mid-guts of mosquitoes in two ways : injection via the anus, and artificial feeding. Injection via the anus Non-anesthetized mosquitoes were transferred to an immobilization chamber by means of an aspirator (Fig. 1). The outer (sleeve) portion of the chamber con- sisted of a plastic tube (2 cm long and 1 cm diameter) that was closed at one end 433 SLEEVE A. SPIELMAN AND J. WONG ASPIRATOR PISTON SEALING WAX GLASS SLIDE , AIR PRESSURE RELEASE CERCUS INOCULUM LATERAL VIEW POSTGENITAL PLATE FIGURE 1. Insertion of mosquito into sleeve portion of restraining device. FIGURE 2. Assembly of restraining device. FIGURE 3. Method of drawing abdomen of mosquito through mesh and extension of post- genital region. FIGURE 4. Method for anal injection. Note that malpighian tubules are visible beneath distended pleural membrane. by a nylon mesh (tulle). The sleeve was placed over a snug-fitting post so that a mosquito confined in the sleeve would be pressed against the mesh ( Fig. 2 ) . Suction was then applied to the mosquito in order to draw the abdomen through an opening in the mesh (Fig. 3). This resulted in eversion of the terminal abdominal segments and exposure of the anus. A finely drawn pipette was then inserted superficially into the hind gut (free-hand) and fluid expelled from the pipette by means of compressed air ( Fig. 4 ) . Resulting distension of the abdomen made it possible to visualize the malpighian tubules which became pressed against the abdominal wall. Occasionally, the gut ruptured and the hemocoele rather than the midgut, filled with inoculum. When this happened, the malpighian tubules floated freely and such mosquitoes were discarded. Artificial jccdint/ Solutions were placed in 2 ml watch glasses and covered with baudrouche mem- brane. The watch glasses were then warmed to 37° C and mosquitoes, confined FOOD STIMULI FOR OOGENESIS IN AEDES 435 above the membrane, were permitted to feed to repletion. The ATP (0.01 M) was added as a feeding stimulant to those solutions not containing red blood cells. Unless otherwise indicated, serum was prepared as follows : Horse serum, ob- tained locally, was lyophilized and re-dissolved to a desired concentration in distilled water. Resulting solutions were dialyzed for 24 hours against 0.85% NaCl (buf- fered to pH 7.1 with phosphate). After feeding or injection, mosquitoes were held above water-soaked paper in individual, guaze-covered vials for two days. The ovaries of each were then removed, disrupted with a vibrating needle and examined at 430 X with trans- mitted illumination. The following commercial materials were employed : albumin (crystalline, bovine), albumin (5 X crystallized, egg), and adenosine triphosphate (crystalline, disodium salt) from Nutritional Biochemicals (Cleveland, Ohio) ; hemoglobin TABU. I Stimulation of oogenesis in female A. aegypti ajtcr ingestion of various amounts of human blood Mg. blood ingested Xo. 99 ' , with developing oocytes 0.1-0.4 16 0 0.5-0.9 49 13 1.0-1.4 42 57 1.5-1.9 21 81 2.0-3.0 17 100 (2 X crystalline, bovine) and glutathion (reduced) from Sigma (St. Louis, Missouri); and globulin (human, Cohn Fraction IV) from Schwartz-Mann (Orangeburg, New York). RESULTS Partial feeding In the first experiment, pre- weighed mosquitoes were permitted to feed indi- vidually on a human host and feeding was interrupted before engorgement was complete. The weight of blood imbibed was recorded immediately upon removal from the host, and mosquitoes were sacrificed two days later. Ovaries were removed, the follicles separated, and degree of development determined micro- scopically. The weight of non-blood-fed mosquitoes varied between 1.9 and 3.3 mg, being influenced by the quantity of fluid in the abdomen. Weight of the blood- meal varied between 0.1 and 4.0 mg. Oogenesis was not initiated when mosquitoes imbibed less than 0.5 mg of blood, while all mosquitoes taking 2.0 mg or more had well-developed oocytes (Table I). Ovarian development was stimulated in about half of the mosquitoes that imbibed 1.0 to 1.5 mg of blood. Of those mosquitoes that failed to commence oogenesis, none had more than a few degenerate primary follicles, nor did secondary follicles develop when primary follicles were not stimulated. 436 A. SPIELMAN AND J. WONG Supplementation of partial feeding by injection In order to determine whether distension of the midgut is prerequisite to oogenesis, saline was injected via the anus of partially-fed mosquitoes and ovarian development recorded after two days. Mosquitoes were permitted to feed on a human host but were removed as soon as blood could be clearly seen through the abdominal pleura. Such mosquitoes generally contained betweeen 0.5 and 1.0 mg of blood. Immediately upon removal from the host, saline (0.85% NaCl) was injected via the anus until the abdomen appeared to be fully distended. When injection was successful, blood and saline became thoroughly mixed and the total weight of blood plus saline was about 4.0 mg. During uninterrupted blood-feeding, mosquitoes normally imbibed about 2.6 mg of blood. Of 235 partially-fed mosquitoes, 141 received saline injected via the anus (Table II). Almost half of these had activated primary ovarian follicles, including TABLE II Stimulation of oogenesis in female A. aegypti after ingestion of trace amounts of blood (0.5-1.0 mg) and supplementation with saline or air injected via the anus % females with stimulated oocytes -.. , . _ , . . Material injected ISJr* O O via anus Developing Degenerating 0.5-1.0 112 12 0 0.5-1.0 Saline 141 38 9 0 Saline 27 0 0 0.5-1.0 Air 9 11 0 0 Air 9 0 0 13 in which most primary follicles degenerated. In contrast, of those that did not receive supplemental fluid via the anus, less than \2% had developing primary fol- licles and none of the remainder had more than a few degenerating follicles. When saline was administered via the anus of non-blood-fed mosquitoes, ovaries remained undeveloped. Air was injected via the anus of other partially blood-fed-mosquitoes in order to distend the midgut without diluting blood already present there. However, this treatment appeared not to affect the developmental state of the ovary (Table II). Nor did injection of air stimulate oogenesis in non-fed mosquitoes. Retention of injected solutions We noted that the midgut contents of partially-blood-fed mosquitoes and of non-blood-fed mosquitoes receiving saline or air via the anus were generally expelled during the day following feeding or injection. This early evacuation of the gut rarely occurred following normal blood-feeding. Accordingly, we studied the relationship between serum concentration and retention of mid-gut contents. Serum was injected via the anus until abdomens were fully distended (about 4 mg). Unusual mortality was noted following injection of non-dialyzed, hypertonic serum. FOOD STIMULI FOR OOGENESIS IN AEDES 437 Of 69 mosquitoes receiving twice concentrated serum, 33 died within the day following injection. In contrast, about 5% of mosquitoes died after receiving isotonic or hypotonic (diluted to twice previous volume) solutions. Accordingly, in subsequent experiments all solutions were dialyzed for 24 hours against 0.85% saline (phosphate buffered). Mortality resulting from the injection of such dialyzed, twice-concentrated solutions was about 10% while less concentrated serum produced negligible mortality. Of those mosquitoes that received serum diluted one part in ten, more than half lost the midgut contents within one day of injection (Table III). On the other hand, virtually all mosquitoes that received undiluted or twice-concentrated serum retained the inoculum. In an attempt to prevent evacuation of the midgut, shellac \vas placed on the anuses of 37 saline injected and 24 air injected mosquitoes. Resulting mortality exceeded 50% during the next 2 days and the survivors were dissected at that time. Of the 12 surviving saline-injected mosquitoes, none had a distended TABLE III Retention of horse serum, variously diluted, during the 24 hr period after anal injection Concentration of serum injected via the anus No. 99 % retaining inoculum 0.1 X 35 43 0.25 X 26 65 0.5 X 168 89 l.OX 142 99 2. OX 133 98 midgut at 2 days after injection, nor were developing primary oocytes found. Instead, each mosquito had large quantities of fluid in the hemocoel and rectum. Malphigian tubules were grossly distended. Air was present in the midguts of each of the eleven surviving mosquitoes that were injected with air. It is interest- ing that the ovaries of six of these mosquitoes contained degenerating primary fol- licles. When the anus was not sealed, neither air nor saline was retained and mortality was nil ; nor did the ovaries appear to be stimulated. Effect on ovarian development of anal injection of serum Mosquitoes were injected via the anus with varying concentrations and varying volumes of serum and subsequent ovarian development noted. Volume of ma- terial introduced was determined by weighing before and after injection. Mortal- ity remained below 5% and, since 50% was the lowest serum concentration used, virtually all mosquitoes retained the inoculum. Those few that failed to do so were discarded. Regardless of the serum preparation used, a greater proportion of mosquitoes began oogenesis when 2.0 mg or more of solution was administered as compared to 1.0 to 1.5 mg (Table IV). No progressive increase was noted at volumes above 2.0 mg. It is interesting that ovarian follicles invariably matured (once stimulated) in mosquitoes receiving 1.0 to 1.9 mg of solution, while such follicles degenerated in 438 A. SPIELMAN AND J. WONG TABLE IV Stimulation of oogenesis in female A. aegypti injected via the anus with various amounts and concentrations of serum Concentration of serum injected Mg. serum injected 0.5X l.OX 2. OX No. 9 9 % Stimulated No. 9 9 % Stimulated No. 9 9 % Stimulated 1.0-1.9 15 33 10 30 9 33 2.0-2.9 53 51 38 66 37 76 3.0-3.9 40 55 37 73 41 85 4.0-4.9 31 52 38 63 30 77 5.0-6.0 11 55 17 59 13 77 12% of mosquitoes receiving more inoculum. No further pattern in ovarian degen- eration was evident. It is clear that the proportion of mosquitoes initiating oogenesis was correlated with the concentration of serum in the inoculum (Table IV). Differential increments between twice-diluted (0.5x), normally-concentrated (Ix), and twice-concentrated (2x) serum approximated 12 Stimulation of ovarian development by various blood components We then administered various components of human blood per os and per anus and compared subsequent ovarian development. Those mosquitoes that did not TABLE V Stimulation of oogenesis in female A. aegypti after ingestion of solutions containing various nutrients Solution ingested Xo. 9 9 % females with stimulated oocytes Developing Degenerating Fresh blood 10 100 0 Stored blood 6 67 0 Blood cells 6 67 0 Serum 4 100 0 Hemoglobin-50 ing/ml 9 0 0 Albumin 10 nig/ml 25 4 0 (Bovine) 50 mg/ml 36 22 0 100 mg/ml 24 38 0 200 mg/ml 12 67 0 Globulin 10 mg/ml IS 38 20 25 mg/ml 14 50 7 Glutathione 0.34 mg/ml 13 0 23 3.4 mg/ml 8 13 25 Saline 15 7 0 FOOD STIMULI FOR OOGENESIS IN AEDES 439 gorge fully were discarded. Whole defibrinated blood was highly stimulatory when ingested through a membrane although prior storage seemed to reduce this property (Table V). Similarly, oogenesis was stimulated by washed cellular components and by serum alone. ATP was added to serum preparations in order to stimulate feeding. We attempted to identify more precisely those blood components that stimulate ovarian activity. Surprisingly, ingested hemoglobin appeared not to stimulate ovarian development (Table V) ; on the other hand, both bovine albumin and globulin were highly stimulatory. Glutathione appeared to be slightly stimulatory. In all but one mosquito, saline was non-stimulatory. However, in that exceptional mosquito, oogenesis proceeded as after normal blood-feeding. Degeneration of primary ovarian follicles was rarely observed in this series of observations. De- generation occurred in few (six) mosquitoes fed globulin and a similar number (five) fed glutathione. TABLE VI Stimulation of oogenesis in female A. aegypti injected via the anus with solutions (greater than 0.002 ml) containing various nutrients (50 mg/ni!) Solution injected No. 9 9 % females with stimulated oocytes Developing Degenerating Globulin Hemoglobin Albumin (egg) 29 29 18 41 0 0 52 31 0 Finally, three potentially stimulatory solutions were injected via the anus and subsequent ovarian development observed. Mosquitoes were weighed both before and immediately after injection in order to insure that each received at least 2 mg of solution. All mosquitoes in this experiment retained the inoculum at one day after treatment and more than 80% survived at two days. Globulin solutions administered via the anus stimulated oogenesis in more than half of the mosquitoes treated (Table VI). Follicles degenerated in only a few (three) mosquitoes. In contrast, hemoglobin was less stimulatory; 9 of 29 treated mosquitoes had de- generating follicles and none proceeded to mature eggs. Egg albumin was appar- ently non-stimulatory. DISCUSSION Ovarian development in anautogenous mosquitoes is arrested at a specific developmental stage, a condition normally sustained until the female gorges on vertebrate blood. Our observations on A. acyypti confirm previous reports (Col- less and Chellapah, 1960; Roy, 1936; Volozina, 1967; Woke, Ally and Rosen- berger, 1956) that resumption of development requires ingestion of a certain threshold volume of blood. A similar threshold effect has been reported for Culex pif>cns (Hosoi, 1954; Kupriyanova, 1966). It is interesting that other mosquitoes may differ in this regard; female Culex tritaeniorhynchus, for example, produce a 440 A. SPIELMAN AND J. WONG few eggs when only trace amounts of blood are taken (Mogi, Wada and Omori, 1972). These observations suggest that female A. aegypti may assess the quantity of blood present in the midgut by means of some volumetric measure. Since the threshold value for oogenesis appears to be about 1.5 mg of blood per mosquito and these mosquitoes normally imbibe about 2.6 mg of blood, completion of oogenesis requires that the midgut be nearly fully engorged. Indeed, when a mosquito has taken 1.5 mg of blood, it seems to be well distended and ingested blood is readily visible through the stretched pleural membrane. The oogenic signal is generated only after the mosquito is nearly completely filled. Time relationships seem to confirm that a volumetric measure may contribute to the assessment of the nutrient content of a blood-meal. An ovary stimulating signal is generated within the head of female A. aegypti within 30 minutes of the completion of blood-feeding (Clements, 1956) and primary oocytes commence micro-pinocytosis and R.N.A. synthesis within an hour (Anderson and Spielman, 1971 ; Anderson and Spielman, 1973). At this time the mass of blood in the midgut appears to be virtually undigested indicating that an exclusively chemical- nutritional assessment is unlikely. Solely on the basis of exclusion, reception of a physical stimulus seems to be required. Since the oogenic signal is generated after injection of nutrient through the anus, any physical estimate of food volume must involve sensations of degree of distension or of pressure within the abdomen. An estimate based on the quantity of flow through the anterior gut would be excluded. Larsen and Bodenstein ( 1959 ) have suggested that the oogenic signal is based on an assessment of the degree of distension of the midgut. Our observations provide experimental confirmation that the volume of nutrient present in the midgut is crucial to the release of the oogenic stimulus. Eggs develop after saline supplementation of blood meals that would otherwise be too small to stimulate oogenesis. It is paradoxical that supplementation by anal injection of air fails to enhance the oogenic stimulus since, in contrast to injection of saline, injected air does not dilute the blood meal. In addition to demonstrating the importance of physically derived oogenic stimuli, our observations confirm Bellamy and Bracken's (1971) suggestion that female mosquitoes can assess the chemical nature of their midgut contents. Working with Cnle.r pipicns, these investigators found that eggs mature following repeated daily hemocoelic injection of a concentrated mixture of amino acids. Although this suggests that products of protein digestion, themselves, might trans- mit an oogenic signal from the midgut via the hemolymph, such non-physiologic manipulation does not constitute rigorous proof. One such digestion product, isoleucine, deserves special attention. This amino acid appears to be an essential dietary component required by A. aegypti for the production of eggs (Greenberg, 1951). Our observations confirm that proteins such as hemoglobulin which are poor in isoleucine generally fail to stimulate oogenesis. Since oogenesis requires a combination of stimuli that are quite specific for vertebrate blood we might speculate that the role of the ventral diverticulum may not be to prevent stimulation of oogenesis by a sugar meal (Larsen and Bodenstein, 1959). When sugar solutions are delivered to the midgut, even in great volume, FOOD STIMULI FOR OOGENESIS IN AEDES 441 the ovaries are not affected. Indeed, unless this food happens to be isotonic with its body fluids, the insect will rapidly die. One role of the diverticulum would be to protect the midgut epithelium from osmotic stress. The wall of this inert diverticulum is highly impermeable to water (Clay and Venard, 1972) and con- tained fluids are only slowly released to the midgut where digestion and absorp- tion take place. Supported in part by Public Health Service Grant AI- 10,274 from the National Institutes of Allergy and Infectious Diseases. U. S. Public Health Service. SUMMARY 1. Female Acdcs aeg\pti generally fail to commence oogensis unless they imbibe veterbrate blood nearly to repletion. However, oogenesis frequently proceeds if a partial blood meal is supplemented with injection of saline via the anus. Injection of saline or air alone generally does not stimulate the ovaries. 2. When the midgut is fully distended with serum the proportion of mosquitoes developing eggs is correlated with concentration of serum. Similarly, feeding on the cellular fractions of blood, on globulin and albumin fractions of serum, and to a lesser extent on glutathione, stimulates oogenesis. Hemoglobin, administered either per os or per anus is relatively non-stimulatory and egg albumin appears to be without oogenic effect. 3. These observations suggest that oogenesis depends upon distention of the midgut as well as on the presence of sufficient concentrations of specific chemical moieties. 4. A principle function of the ventral diverticulum may be to protect the midgut against osmotic stress rather than to prevent premature oogenesis. LITERATURE CITED ANDERSON, W., AND A. SPIELMAN, 1971. Permeability of the ovarian follicle of Acdcs aeyvpti mosquitoes. /. Cell Biol.. 50 : 201-221. ANDERSON, W. A., AND A. SPIELMAN, 1973. Incorporation of RNA and protein precursors by ovarian follicles of Acdcs acgypti mosquitoes. /. Submicro. Cytol., 5 : 181-198. BELLAMY, R. E., AND G. K. BRACKEN, 1971. Quantitative aspects of ovarian development in mosquitoes. Can. Entomol., 103: 763-773. CLAY, M. E., AND C. E. VENARD, 1972. The fine structure of the oesophageal diverticula in the mosquito Acdcs triscriatus. Ann. Entomol. Soc. Aincr., 65: 964-975. CLEMENTS, A. N., 1956. Hormonal control of ovarian development in mosquitoes. /. Exp. Biol., 33: 211-223. COLLESS, D. H., AND W. T. CnELLApAH, 1960. Effects of body weight and size of blood- meal upon egg production in Acdcs acgypti. Ann. Trap. Mcd. Parasitol., 54: 475-482. GREENBERG, J., 1951. Some nutritional requirements of adult mosquitoes (Aedes acgypti} for oviposition. /. Nutrit., 43 : 27-35. Hosor, T., 1954. Egg production in Culcx pipicns pollens III. Growth and degeneration of ovarian follicles. Jap. J. Mcd. Sci. Biol., 7 : 111-127. KUPKIYANOVA, E. S., 1966. On the gonotropic cycle in mosquitoes of the genus Culcx I. The influence of the amount of human blood ingested on the development of the eggs and fecundity of Culcx pipicns molcstus and Culcx pipicns pipicns. [In Russian] Mcd. Parasitol., 35 : 310-316. 442 A. SPIELMAN AND J. WONG LARSEN, J. R., AND D. BODENSTEIN, 1959. The humoral control of egg maturation in the mosquito. /. E.vp. Zoo/., 140 : 343-381. MOGI, M., Y. WADA AND N. OMORI, 1972. The follicular development of Culcx tntaenio- rhynclnis sitintiwrosus females after taking various amounts of blood in reference to feeding and oviposition activity. /. Trap. Mcd., 14 : 55-63. Rov, D. N., 1936. On the role of blood in ovulation in Aedcs acg\pti. Bull. Entomol. Res., 27 : 423-429. VOLOZINA, N. V., 1967. The effect of the amount of blood engorged and of supplementary carbohydrate feeding on the process of oogenesis in the females of blood-sucking mos- quitoes in the genus Acdes of different weights and ages. [In Russian] Entoinol. Obosr.,46: 49-59. WOKE, P. A., M. S. ALLY AND C. R. ROSENBERGER, JR., 1956. The numbers of eggs developed related to the quantities of human blood ingested in Aedcs aegypti. Ann. Entoinol. Soc. Amer., 49: 435-441. Reference : Biol Bull., 147 : 443-456. (October, 1974) CURRENT-INDUCED FLOW THROUGH THE SPONGE, HALICHONDRIA STEVEN VOGEL Department of Zoology, Duke University, Durham, North Carolina, and Marine Biological Laboratory, Woods Hole, Massachusetts The entire phylum Porifera consists of suspension-feeding organisms, simple by comparison with other multicellular animals, but nonetheless highly specialized for separating small organisms and other nutritive participate matter from the water passing through themselves. A sponge, in Bidder's (1923, page 312) felicitous phrases, is "a mere living screen between the used half of the universe and the unused half — a moment of active metabolism between the unknown future and the exhausted past." As Grant (1825) first showed, water passes unidirectionally through an elaborate system of pores, gates, and canals and, following filtration, is returned to the medium from which it was drawn. This flow of water through sponges has been regarded as being due entirely to the activity of certain peculiar flagellated cells, the choanocytes. Indeed, early workers such as Grant (1825) and Bower- bank ( 1864) devoted considerable ingenuity to demonstrations that muscles played no role. And the general design of sponges is consistent with the requirements for a substantial output — one of Bidder's (1937, pages 129) Leuconia aspcra Schmidt "threw, to a distance of 70 cm, 1000 times its own bulk of water per hour"-— from a pump consisting of uncoordinated cells. Flagellated chambers with a combined cross-sectional area many times that of the outer surface of the animal lead into a collecting manifold in which successive generations of channels are both fewer in numbers and of smaller total cross-section. Thus the velocity of the final excur- rent stream through the osculum may be several hundred times that at the level of the choanocytes. The reduction in cross-section of the final common canal at the osculum has been interpreted as a further device to increase the excurrent velocity, and the oscular chimneys have been regarded as a means of increasing the distance between excurrent and incurrent openings. Both features presumably reduce the likelihood of recycling previously filtered water (Bidder, 1923; Leigh, 1971). But this generally accepted picture may, in fact, overlook another important element propelling water through at least some sponges under certain circumstances. Previous observations of the pumping activities of sponges were made with the animals in still water (except for the field work of Reiswig, 1971). Yet sponges appear to require moving water in their normal habitats ; DeLaubenfels (1954) feels that velocities of two or three kilometers per hour are optimal. This requirement has been interpreted as insuring that silt and debris do not accumulate and as further insurance against reingestion of a sponge's output. But, whatever its other functions, the presence of a current in the medium around on attached sponge raises the possibility of flmv through the animal unthout active pumping by the choanocytes. Such induced flows have been termed "passive" 443 444 STEVEN VOGEL since they do not entail immediate metabolic cost (Vogel and Bretz, 1972). The present investigation seeks to determine whether passive flow might be an appreciable factor in the lives of sponges. It should be noted that passive flow, while it may avoid metabolic expenditure, involves no novel source of energy. A potential exists between any two points where the velocity of the medium is different ; where a potential exists, work may be extracted by an appropriate device. For an attached organism, the velocity at the point of attachment is, of course, zero ; and it can extract energy from the flow of fluid around it just as does a windmill. At least three physical mechanisms permit the induction of flow in a spongelike structure by motion of the external medium (Fig. 1). (A) If a small pipe con- nects two points in a larger channel, and if the ends of the small pipe are normal to the walls of the channel, then fluid will flow in the small pipe from the end where flow in the channel is slower to the end where flow is more rapid (la). By (a) (b) (c) FIGURE 1. Several arrangements whereby flow in a larger channel may induce unidirectional fluid movement in a small pipe. Bernoulli's principle, the increase in velocity in the channel is concomitant with a reduction in pressure in order that energy be conserved ; this pressure difference, then, induces flow in the small pipe. The direction of flow in the channel is, of course, inconsequential. With sponges, the medium should travel fastest as it crosses the highest points, which commonly bear the oscula. Thus the internal flow will be directed from any ostia not atop a major protrusion toward any terminal oscula on the protrusions. (B) If a small pipe terminates normal to the wall of a larger channel through which fluid is moving, then fluid will be drawn from the pipe into the channel by an additional agency, termed viscous entrainment or sucking, and caused by the resistance of real fluids to rapid shear rates. Faster movement of the fluid in the channel will produce greater rates of entrainment (la). In addition, the rate of entrainment will depend on the size of the aperture (Ib) ; it is to avoid errors due to entrainment that the static (normal) aperture on pitot tubes (as used for airspeed indication in small planes) must be kept small (Prandtl and Tietjens, 1934). In a finger-like or encrusting sponge, water will therefore be drawn out of a terminal osculum because, as the highest point on the structure the latter will INDUCED FLOW THROUGH HALICHONDRIA 445 be exposed to the greatest velocity, as well as because the oscula are larger holes than the ostia. (C) If a small pipe bent at a 90° angle is oriented in a moving stream so that one aperture is directed upstream and the other is normal to the stream, then fluid will enter the former and run out of the latter (Ic). The normal aperture "sees" only the static pressure of the fluid ; the one directed upstream is exposed to the sum of the static pressure and the additional dynamic pressure caused by the deceleration of the fluid in front of the aperture. In a finger-like sponge, an ostium directed upstream will thus be exposed to a higher pressure than will an osculum normal to the flow. If the external flow reverses direction, then ostia on the opposite side of the finger will be exposed to the higher pressure, and internal flow will still be from ostia to oscula. For a sponge with a terminal osculum, all three mechanisms predict the same direction of flow — from ostia to oscula. It has not yet proven feasible, either by theory or measurement, to apportion the passive flow among the different mecha- nisms which might be responsible. One can, however, state that size, shape, and location of the openings should all be relevant to determining the direction of passive flow. Consideration of these possible physical mechanisms suggests that a good excurrent opening should be large, at the terminus of a fairly sharp projection, and farthest from the point of attachment to the substratum. That a saving in the energy expended in filtering water might be important to sponges is evident in the calculations of J0rgensen (1966). His figures, based on filtration rates and oxygen consumption in still water, suggest that the nutritive value of the suspended material available may not greatly exceed the cost of filtration. Passive flow makes a significant contribution to the ventilation of the burrows of prairie-dogs (Vogel, Ellington and Kilgore, 1973) and mounds of certain African termites (Weir, 1973), and it may have been crucial to the success of archeocyathids in the Cambrian (Balsam and Vogel, 1973). Ideally, an evaluation of passive flow in sponges should be based on direct observations in nature and on a theoretical analysis of the hydrodynamics of the situation. Direct observation, however, has given ambiguous results clue to the complexity and variability of the currents around the sponges investigated. And theoretical analysis has been frustrated by the complexity of the physical mecha- nisms involved. An operational compromise, the present investigation divides into three experimental portions : first, observation and measurements of flow through living and freshly killed sponges exposed to moving sea-water in the laboratory ; secondly, measurements of flow rates through a brass model of a sponge in a flow tank ; and thirdly, observations on the influence of the geometry of apertures on the induction of flow. MATERIALS AND METHODS I'rcsJi material Sponges (Halichondria bowerbanki Burton: see TIartman, 1058) were col- lected by the Supply Department of the Marine biological Laboratory, the en- crusting form near \Yoods Hole, Massachusetts, the more erect and finger-like STEVEN VOGEL (3) T.irt FIGURE 2. Rhodamine B passing through a freshly killed Ualichondr'w at an external current of about 5 cm/sec. FIGURE 3. Rhodamine B passing through a plexiglas model (described by Vogel and Bretz, 1972) under the same conditions as in Figure 1. The model is 2.5 cm high. FIGURE 4. Brass model on which the data of Figure 8 were obtained. The smaller cylinder inserts into the larger through a hole in the bottom of the latter. Two of the small holes in the outer cylinder communicate with each chamber in the walls of the inner cylinder ; thus holes in outer and inner cylinders are not aligned. FIGURE 5. Apparatus for testing apertures in pairs. Each aperture is radially symmetrical and was oriented normal to the stream; dye was introduced through the small pipe, which also served as a support. Large collars on either end of the main tube reduced the sensitivity of flow through the tube to minor misalignments in the stream. form from the Cape Cod Canal. All were maintained in running sea water and used within a clay of collection. In the erect form, active pumping usually ceased within a few hours after the sponges were received, probably due to the warmer water at the laboratory. With the addition of curved corner fairings, a sea-table, 132 X 7(> X 10 cm inside, provided an adequate flow tank for tise with the fresh sponges. The supply hose, fixed to one wall and directed downstream, propelled water at up to 10 cm/ sec. Water left through a porous cylinder fixed to the drain, thereby providing a continuously changing medium in the tank and facilitating observations of the pat- terns of flow with dye markers. Of several colored materials used, a concentrated INDUCED FLOW THROUGH HALICHONDRIA 447 solution of rhodamine B in sea water was by far the most suitable for marking flows ; in practice it was introduced from a hypodermic syringe, and it stained neither plastic models nor sponges. "Fingers" of Halichondria were cut from colonies with a substantial portion of the basal portion of the colony left attached and without cutting into the axial spongocoel ; these were pinned to wax in a small finger bowl. Above the wax, a small amount of sand provided convenient fairing material to obscure the cut end of the sponge and to raise the substratum to the level of the rim of the bowl (Fig. 2). After introduction into the sea table, each specimen was checked with a dye marker for active pumping ; several either stopped pumping shortly thereafter or never showed evidence of pumping. Certain of the latter were used for measure- ments of passive flow. Active pumping, when occurring, was steady, with none of the polyrhythmic character of the ventilatory flow of, for example, burrowing polychaetes. No specimen which spontaneously stopped pumping was ever ob- served to start again. For most measurements of passive flow, sponges were deliberately "turned off" by immersing them in fresh water for about five minutes. Parker (1910) found that this treatment eliminated active pumping without causing constriction of either ostia or oscula in Stylotella. Fresh water produced no visible effects on the openings of Halichondria, and the performance of sponges which had spontaneously stopped pumping was indistinguishable from those treated with fresh water. For measurements of flow inside a specimen, a probe (see below) was lowered about 4 mm below the tip of the osculum ; the same probe was repositioned about 3 cm in front of the sponge to record the outside velocity. Traverses from the level of the osculum down to the sand fairing showed an essentially constant velocity —the use of a finger-bowl to hold specimens reduced the boundary-layer thickness to insignificance. Models Models were made from cylindrical metal or plexiglas stock on a combination lathe and milling machine. A linear flow tank filled with fresh water was employed for measurements on all models. This device consisted of a plywood channel, 244 X 23 X 17 cm, connecting two 200-liter, open cylindrical drums. A 1.5 HP centrifugal pump in the return circuit provided flow rates of about 7 I/sec or (depending on the depth of water in the channel) up to 40 cm/sec. An adjustable bvpass around the pump together with an adjustable gate at the downstream end of the channel controlled the flow rate and depth in the channel. Flow was smoothed by baffling in the fore-tank and by an array of small pipes acting as a straightener at the entrance to the channel. These devices proved adequate : a blob of dye remained visually discrete as it travelled the length of the channel. More water passed through the lower portion of the adjustable gate than through the upper portion, thus compensating for the retardation caused by the floor of the channel. As a result, the speed of flow was nearly uniform from within a few millimeters beneath the surface to about a centimeter above the bottom at the place, about half-wav down the channel, at which all measurements were made. 448 STEVEN VOGEL Flow-meter Flow rates were measured with a heated-bead thermistor flow-meter (Figure 6) whose accuracy was estimated (on the basis of repeated calibrations) as ±0.5 cm/sec between 0 and 30 cm/sec. Several limitations on the use of this instrument deserve comment. In the absence of a temperature-compensating circuit, measure- ments had to be made at water temperatures differing no more than 1° C from that of the calibration. The instrument had a substantially non-linear response, giving a greater change in output for a given change in flow rate at low velocities. The thermistor bead together with its glass envelope on the end of the probe was 2.0 mm in diameter while the diameter of the cylindrical shaft of the probe was 1.5 mm; thus the probe seriously occluded oscula less than about 3 mm across. 115 VAC POWER SUPPLY 24 VDC THERMISTOR: FENWAL GB31P2 100 /VW 1000 O-50/jA 100 R ECORDER 150 FIGURE 6. Circuit diagram of the thermistor flow-meter used for measurements in liquids at low speeds. Finally, the probe could not be used in confined spaces (such as a spongocoel or oscular channel ) at speeds below one cm/sec without inaccuracy resulting from local heating of the medium. The flow-meter was, however, not appreciably directional in its response ; and its response time of about one second was amply short for present purposes. For measurements on fresh material, the flow-meter was calibrated with the probe in a motor-driven revolving cylindrical bowl of sea water, 20 cm in diameter and 10 cm deep. In practice, the bowl was rotated until the water inside was moving at the same angular rate as the bowl itself ; then the probe was lowered into the water near the center and slowly moved radially outward while the output of the meter was recorded. If the probe remained in one position for more than a few seconds, local heating and/or retardation of the medium gave a spurious reduction in the output of the meter. For measurements on models in fresh water, the flow-meter was calibrated in the linear flow tank simply by measuring with a stop watch the time necessary for a dye-marker to travel 150 cm and comparing the result with the output of the meter. INDUCED FLOW THROUGH HALICHONDRIA RESULTS 449 FrcsJi material Most of the actively pumping sponges gradually reduced their activity during the course of the measurements in the flow tank ; therefore it was necessary to determine the pumping rate in still water after every few measurements of total flow. Data from animals whose rates declined were discarded on suspicions of morihundity. Figure 7 shows the combined active and passive flow for a sponge whose pumping rate in still water did not vary detectahly over the course of 6 - 5 - o z Ct. 10 EXTERNAL CURRENT, cm. sec-' FIGURE 7. Flow across and through specimens of Halichondria. Closed circles — active pumping ; open circles — specimens "turned off" by exposure to fresh water. The regression equation for the upper line is Y = 0.450 X +2.207 ; for the lower line it is Y = 0.390 X -0.124. the measurements. It continued this rate of pumping for an hour afterwards, suggesting that the data for total flow through the sponge may he representative of flow rates in the normal habitat. Other specimens gave similar, but less regular and reproducible results. Figure 7 also shows flow rates for sponges which, as determined by observations with dye markers, were not actively pumping. Although always less than the total flow through active sponges, these rates were substantial : at an external current of 7 cm/sec, the passive component was typically about one-half the total flow through an actively pumping sponge. Neither the x nor y intercepts of the line fitted to the data on passive flow differ significantly from zero : at least in this range of speeds there is no evidence 450 STEVEN VOGEL of either residual pumping or of a non-linear relationship between internal and external flows. Moreover, the slopes of the lines for total flow and for passive flow through the sponges do not differ significantly : the active component appears to be simply "added on" to the passive flow and its magnitude does not depend on the strength of the external current in this range of speeds. For the sponges for which data are presented here, mean internal oscular diameter was 4.9 mm (S.D. =: 1.5, n == 7). The individual fingers were 31.0 mm ( S.D. = 4.7) high and 12.1 mm (S.D. -- 3.5) in outside diameter half-way between base and tip. A plot of external current versus passive flow through a sponge gives a con- venient measure of the resistance of the animal to the induction of flow through it. The reciprocal of the slope provides a dimensionless "resistance index" which lumps all of the factors affecting the efficacy of passive flow ; for these sponges, the resistance index is 2.56. Comparisons with other species and specimens from other habitats should be of interest. Observations on the encrusting or "hard" variety of Halichondria were made with dye markers, since the oscula \vere too small to permit insertion of the thermistor probe. Qualitatively, results differed in no way from those obtained with erect, finger-like specimens. Both active and passive flow were clearly evident, the former in still water and the latter after the usual exposure to fresh water to inactivate the choanocytes. Model sponges The results obtained on fresh material, actively pumping and "turned off" imply- that the so-called passive component of the flow through a sponge is a purely physical phenomenon, independent of movement of any part of the animal. Further evidence that active processes need not be invoked emerges from observations and measurements on completely non-living, physical models of sponges. A visual comparison of a "turned off" fresh sponge and a plastic model is provided by Figures 2 and 3. In both cases, a solution of Rhodamine B was injected near the upstream side of the basal end of sponge or model a few seconds before the picture was taken. The similarity of the pattern in which the dye solution emerges from the oscula is clearly evident. A quantitative view of the flow induced in a model is provided by Figure 8. This model (Fig. 4) incorporates analogs of incurrent and excurrent canals as well as flagellated chambers, although it is still a very crude imitation of the array of channels in even a simple sponge. In particular, it proves impractical to re- produce the minute dimensions and enormous numbers of internal channels of a sponge, and it is convenient to construct a model somewhat larger than the size of the Halichondria fingers. This model is about twice the height of a real finger, and thus an equivalent flow is achieved when the external current passing the model is half that passing a real sponge. The behavior of model and real (turned off) sponges are. nevertheless, quite similar; even the resistance indices (2.30 and 2.56) do not differ much if the data obtained on the model with external currents below 1 1 cm/sec are considered. It appears that the larger size of the channels in the model roughly compensates for their fewer numbers. INDUCED FLOW THROUGH HALICHONDRIA 451 The principal difference is the lower slope of the curve for the model obtained at external currents above 1 1 cm/sec. Limitations of the sea-water flow tank precluded making equivalent measurements on fresh sponges, particularly since the appropriate velocities would have had to be above, not 11, but 22 cm/sec. However, the reduced slope is most likely to be an artifact of this particular model. When a very loosely-fitting and porous collar shielded the cylindrical portion of the model from an external current above 11 cm/sec, the flow through the model increased. At lower external currents, the collar was without effect. Perhaps Z 1 O „-- -O- o c ,--'' o o __„_-'• o O o i O o d J L i I I I I 1_ ._!_ 10 20 EXTERNAL CURRENT, CM. SEC 30 FIGURE 8. Flow across and through the model shown in Figure 4. Below an external current of 11 cm/sec the regression equation is Y = 0.436 X —0.647 ; above 11 cm/sec it is Y = 0.066 X +3.29. Flow was undetectable (less than one cm/sec) through the model when the "ostia" were blocked. the unrealistically large diameters of the "ostia" in the model permitted significant entrainment at the higher speeds, thus opposing the entrainment at the osculum. In addition to the model sponges shown in Figure 3 and 4, several others were constructed and tested with dye markers for passive flow, the latter by members of the Experimental Invertebrate Zoology Course (1972) at the Marine Biological Laboratory. Ms. Sarah E. Swank made a conical, crater-like model, 3 cm high, with a ring of large collecting channels directed upward and entering the common spongocoel just below the osculum. Mr. Duen Yen devised a vase-like model, 12 cm high, consisting of a wire framework wrapped with percale cloth, and pro- ducing more numerous and tinier pores than could be made of metal. Both of 452 STEVEN VOGEL these models permitted substantial induction of internal flow by movement of the medium. Indeed, our impression, based on a total of six "sponge" and two "archeo- cyathid" (Balsam and Vogel, 1973) models, is that effective designs may vary greatly in their details. The basic constraints of the scheme are little more than small input pores and a large output opening, the latter, at least, exposed to a transverse current. It may prove to be the case that sponges of a wide variety of shapes and sizes eventually show evidence of significant passive flow7. Model oscnla The oscula of sponges are not limited to sharp-edged, chimney-like holes at the ends of fingers, as in these Halichondria. Rather, they display considerable diversity, from rimless openings on the flat surfaces of some sponges, to small, crater-like elevations distributed over a branching array of cylinders, or holes on the inner surface of large, basket-shaped sponges (see, for example, DeLaubenfels, 1950). Measurements on models of the mounds of prairie-dogs (Vogel, ct I using the lactoperoxidase procedure in order to follow changes in membranes which accompany fertiliza- tion. This method allows labelling of surface proteins with little modification of structure such that eggs can be normally activated. The vitelline layer was first removed with 0.01 M dithiothreitol (pH 8.5). A maximum of 0.2 ml eggs were added to 2 ml MBL artificial sea watef containing 40 /*Ci 120I in 0.1 N NaOH and 0.2 MBL lactoperoxidase (Sigma). Ten p] aliquots of 0.06% HoO2 were added at 2 minute intervals for 10 minutes and the cells were incubated for an additional 10 minutes. Samples were counted using Bray's scintillation fluid in a Beckman 250 scintillation counter. Calculations showed that 5% of the total 125I added was bound to the eggs and that 5.5 X 105 molecules per egg were "bound. After iodination, eggs were washed and half were activated by the method of Steinhart and Epel with the calcium ionophore (A23187). The ionophore was used to avoid introduc- tion of sperm proteins. Activated eggs showed a 26% (8%) loss of label to the supernatant sea water as compared to unactivated control eggs, indicating that surface proteins are lost following fertilization. Iodinated eggs were fixed with paraformaldehyde-glutaraldehyde mixture, post-fixed with OsO4 and embedded in epon-araldite for autoradiography. 0.5 /J. sections were mounted on slides and dipped in Kodak NTB2 emulsion and developed after two days at 4° C. Auto- radiography of unactivated iodinated eggs showed a heavy concentration of label on the sur- face of the eggs. This iodination technique does label surface proteins of the egg without loss of viability of eggs and can be used for further biochemical analysis to determine the fate of the plasma membrane after activation. This study was performed in the Fertilization and Gamete Physiology Training Program at the MBL (NIH-3T01-HD00026-12S1.) Reflexes in the tailspine system of the horseshoe crab, Limnlus polyphemus. DOUGLAS A. EAGLES. The tailspine of the horseshoe crab can move freely within the cone delimited by the margins of the socket. During normal movements the tailspine rarely contacts the socket margins, its position actively controlled by the eight large muscles inserting at its base. One of these, the ventral flexor muscle, is innervated by two branches of hemal nerve 16, one of which appears to be exclusively motort the other exclusively sensory. The sensory nerve includes the axons of at least two separate populations of receptor neurons. Cells of one group are multipolar and their dendrites terminate near the insertions of the flexor muscle fibers onto tailspine apodemes. Some of these neurons fire as a result of the tension elicited by a minimal stimulus to the motor nerve. These cells also fire during lengthening and shortening of the passive muscle and during isometric active contraction. Anatomical and physiological considerations suggest that these neurons are series tension receptors. The other group of neurons is located more distally and consists of bipolar and tripolar cells located near the arthroidial membrane and within the distal nerve processes. These cells fire during various joint movements but not during isometric contraction. Tension receptor activity elicited by an isometric contraction results in a reduced spike frequency in the motor nerve, indicating central inhibiton of the exclusively excitatory tailspine efferents. It there- fore appears that the tension receptors inhibit homonymous motoneurons, as in vertebrates and PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 475 crustaceans. Because the joint receptors are heterogeous it has not been possible to define specific reflex effects. Supported by a Grass Foundation Fellowship in the summer of 1974. . Inaerobic synthesis of bacterial luciferase. ANATOL EBERHARD. The bioluminescent bacterium, Photobactcrium fischcri strain 8265, when grown aerobically in a rich medium, produces the enzyme luciferase and the cultures give off light. The iso- lated enzyme requires reduced flavin mononucleotide, long-chain aldehyde and oxygen for in vitro light production. When grown anaerobically in a highly buffered medium with glu- cose as the carbon source and disodium dithionite as a reducing agent, the cells gave off no light whatever but still produced at least as much luciferase as when they were grown aerobically. The presence of luciferase was demonstrated by aerating an anaerobic culture and measuring the subsequent light emission. After five transfers of 0.1 ml of culture to 10 ml of fresh anaerobic medium the final culture produced 500 times as much light upon aeration as did the original culture. Since the luciferase present in the original inoculum had been diluted by ten orders of magnitude by the transfers, this finding clearly indicates massive de noro anaerobic synthesis of luciferase. The anaerobic synthesis of luciferase in the total absence of light production suggests that the luciferase must have some other important cellu- lar function. Effects of inrivalcnt (Fab) antibody fragments on the fertilizing capacity of sperm of the sea urchin, Arbacia punctulata. W. R. ECKBERG AND C. B. MKTZ. Rabbit antibodies against sea urchin sperm, rendered non-agglutinating by papain digestion (Asp-Fab), prevent sperm from fertilizing eggs. The present experiments were designed to determine the step or steps in sperm-egg interaction which are blocked by antibody. Asp-Fab had no effect on sperm motility. Examined microscopically, sperm motility de- cayed at the same rate after treatment with Asp-Fab, non-immune rabbit gamma-globulin (CFab), or sea water, and sperm longevity was the same in all three cases. Asp-Fab-treated sperm failed to fertilize dejellied eggs (pH 5 sea water, 5 min) or vitelline layerless eggs (0.05% trypsin, 30 min to remove both jelly and vitelline layers). Evi- dently, antisperm antibody treatment prevents sperm from engaging in some essential reac- tion(s) at the egg cell membrane level. Sperm binding to the egg may be one such essential reaction. Sea water or C-Fab-treated sperm bound to normal eggs, dejellied eggs, and vitelline layerless eggs, although binding was somewhat less in vitelline layerless eggs. Asp-Fab, on the other hand, inhibited binding in each case, although it did not effect jelly penetration. Such inhibition of sperm attachment or binding to the egg cell membrane can account for the fertilization inhibiting action of anti- sperm antibodies. This study was performed in the Fertilization and Gamete Physiology Research Training Program at the Marine Biological Laboratory ( NTH grant 3T01-HD00026-12S1). Adaptation and spatial summation in rods from the toad retina. GORDON L. FAIN. Intracellular recordings from Bitfo niarinus were used to investigate light adaptation and spatial summation in "red" rods. The onset of diffuse background illumination produced an initial transient peak of hyperpolarization, followed by a fast and then a slow decay of po- tential to a steady plateau level. The threshold of single rods in the presence of the back- ground increased as the background intensity increased, the two being linearly related (fol- lowing the Weber-Fechner line) for 3 log units. At backgrounds of about 5 X 105 incident quanta/rod-sec, threshold began to deviate from the Weber-Fechner relation, and backgrounds one log unit brighter completely saturated the rods. Saturation of the rod response was not caused by inhibition from cones, since increment- threshold curves had the same shape for backgrounds at 505 and 605 nm. Nor was satura- 476 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY tion produced by suppression of the response due to the steady plateau potential of the back- ground, since this was always less than half the maximum hyperpolarization elicited from the dark-adapted receptor. Since even the brightest backgrounds bleached only a small fraction of the visual pigment, saturation may occur at some intermediate stage in transduction be- tween bleaching and the production of membrane voltage change. Spatial summation was measured by recording from rods stimulated with centered, con- centric spots of varying radii. Sensitivity (measured from intensity-response curves) in- creased as spot radius increased, for spots up to 400 /j.m in radius. Since sensitivity increased by more than a log unit as spot radius was increased from 25 to 400 urn, the photocurrent of a single rod is apparently responsible for less than 10% of its sensitivity. The receptive fields of rods were not changed by moderate background illumination, suggesting that increases in acuity during light-adaptation are not caused by a reduction of receptor spatial summation. Supported by the Grass Foundation and by an NIH Fellowship, #EY-54840. Glycine transport in the intestine of the toadfisli - effect of luminal application of ATP in vivo. A. FARMANFARMAIAN AND DEREK BARKALOW. Substances which reduce ATP production are known to inhibit active transport of organic solutes. In viro studies in this laboratory on the absorption of amino acids and sugars by intestine have not supported the hypothesis of indirect coupling of transport to ATP via Na+ or K+ gradients. The present experiments were designed to test the possibility of direct cou- pling of ATP to glycine transport through interaction with the "carrier." Glycine absorption in toadfish intestine shows saturation kinetics with Kt about 5 mM and Jmax about 30 jitmoles per g per hr at 20°. At glycine concentrations of 10 mM or 20 mM, uptake is linear with time for periods up to 60 min. Under these conditions addition of 10 mM ATP to the luminal fluid causes a 30% reduction in absorption rate. Statistical analysis showed significance at 0.05 > P > 0.02. These results suggest that ATP is capable of acting on the carrier at the luminal side in a manner like that postulated to occur in the cell. Such a mechanism might involve phosphorylation and conformational change of the carrier protein, leading to reduced substrate affinity. If this process occurs in the lumen when ATP is introduced, a reduction in lumen- to-cell transport would be predicted. Supported by funds from the Rutgers Marine Sciences Center, Rutgers University, New Brunswick, N. J. Intraccllnlar Ca*+ injection produces localized desensitisation of Limultis ventral photoreceptors. A. FEIN AND J. LISMAN. Photoreceptors were penetrated with a microelectrode containing 0.09 M CaOH, 0.1 M Tris, 0.1 M EGTA (100-150 meg). The cells were dark-adapted until quantal responses were observed. Two microspots (nominal diameter, 10 /*) were focused on the photoreceptor ; one centered on the electrode, the other 60-80 /j. away. Spot intensities were adjusted so that 10-30 msec flashes evoked 20 mV responses. Small Ca++ injections (1-2 nAmp, 10 sec) pro- duced a large (4-15 fold) attenuation of the response to the local spot while producing little (<1.5 fold) attenuation of the response to the distant spot. If the size of the injection was raised, the response to the distant spot was also attenuated. Following such injections, the response to the distant spot recovered faster (1-2 min) than the response to the local spot (4-5 min). These results were found in cells where local and distant regions were of nearly equal (within 0.4 log units) sensitivity before injection. In some cells impaled with calcium elec- trodes, the region near the electrode was several log units less sensitive than distant regions and local effects of Ca++ injection could not be demonstrated. In cells where a local effect was found, it was also possible to demonstrate that light- adaptation was local, as previously reported for this preparation by Fein. Lisman and Brown have hypothesized that a light-induced rise in intracellular Ca++ is important in regulating PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 477 adaptation. The demonstration that both adapting lights and calcium injection produce localized effects is consistent with this hypothesis. The results suggest that Ca++ can diffuse over neuronal dimensions, but that diffusion is sufficiently slow so as to produce intracellular gradients. This work was supported by NIH grants EY 01362-01 to A. Fein and EY 01496-01 to J. Lisman. The development of light production in the Annelid Chaetopterus pegamentaceus. GARY FREEMAN, GEO. T. REYNOLDS, AND ALAN WALTON. In Cluictoph-rus larvae which have been raised at 22-23 degrees C, light production is first detected with a photomultiplier after KCI stimulation at 29 hours of development. During the next 10 hours of development the amount of light produced by a larvae increases 10 fold ; the rate of increase in the amount of light produced as a function of time is uniform during this period. Image intensification studies show that the trochophore larvae has a mass of light producing cells on either side of its mouth. These cells are a part of the epithelium which covers the larvae. Cytological studies show that these cells contain many granules with a diameter of 2-3 microns ; they closely resemble the photogenic cells of the adult. The spectra produced by the light from larvae and adults is the same. If the AB and CD blastomers are separated at the 2 cell stage and raised in isolation the larvae derived from the CD blastomeres will produce light while those from the AB blasto- meres will not. When compression is applied prior to the first cleavage the cleavage is fre- quently equal and the polar lobe material is segregated to both blastomeres. If the blastomeres are separated in eggs in which the first cleavage is equal as a result of compression and raised in isolation, both of the larvae derived from a pair of blastomeres will frequently produce light. Supported by AEC grant DBER AT-11-1-3120 to (i. T. R. and NIH grant GM 20024 to G. F. Studies on the size distribution oj lorieac and the c/rou'tli status of Tintinnida in Eel Pond. KENNETH GOLD. The size distribution of loricae was used to identify the growth status of 3 species of Tintinnida in the plankton. Tintinnopsis acuininata. T. dadayi, and Tintinnidium fluriatilc were each found to produce loricae having progressively shorter mean lengths, the changes often being detectable within a day or less. The very sudden shifts in the length-frequency distribution from longer to shorter loricae were interpreted as signalling the beginning of a period of rapid cell division. T. acuminata persisted in the plankton long enough to be studied throughout what is thought to be its entire bloom cycle lasting 6-8 days. Percentages of loricae in size categories of 5 /*m intervals over the range of lengths were plotted daily, and the resultant family of curves then analyzed for events seen as changes in the slopes of the curves. Three phases were identified which have the following characteristics: (I.) An in- crease in the percentage of longer loricae in the population followed by their abrupt replace- ment with shorter individuals (1-2 days); (II.) the disappearance of the longer loricae from the plankton and an approximately linear increase in the percentage of shorter individuals ; (III.) reversal of the trend set in phase II on the 6th day of the cycle. Based on similar curves for the smaller laboratory-reared species T. fiarra (2 /um intervals for the entire length range), these phases correspond — in batch culture — to the end of lag, exponential growth, and either stationary or death phase, respectively. The two other species are thought to have been from populations in phase I. The following other Tintinnida were detected in numbers estimated as rare to abundant during June-August, 1974: Stctwscinclla stcini, S. olira, Tintinnofsis niinuta, T. rapa, T. baltica, T. platcnsis, Farclla chrcHbcn/ii, and Hclicostoinclla subulata. Supported by a contract with the U. S. Atomic Energy Commission, reference number COO-3390-15. 478 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY Genome organization in the American oyster, Crassostrea virginica. ROBERT B. GOLDBERG AND JOAN V. RUDERMAN. In deuterostomes a large fraction of the genome consists of 300 nucleotide (NT) long repetitive sequences interspersed with 1000 NT single copy sequences. A small fraction is made up of clustered repetitive sequences. The interspersed single copy length corresponds to that required to encode an average size protein. The interspersed repetitive sequences may play a role in the regulation of transcription of contiguous single copy structural genes (Davidson and Britten, 1973, Quart. A'rr. Bid., 48: 565). It was of considerable interest to determine whether this pattern of DNA sequence organization extends to the protostomes as well. Our studies show that the arrangement of repetitive and single copy sequences in the oyster genome is similar to the deuterostome pattern, although there are some differences in the relative amounts of these sequences. Oyster DNA was sheared to 300 NT and 2600 NT long fragments. The 300 NT fragments were denatured, incuhated to various C0T (moles NT X I"1 X sec) values and assayed for per cent reassociation on hydroxyapatite. Com- puter analysis of the reassociation curve indicates that the oyster genome consists of 70% repetitive (CoT- = 28, complexity = 0.18 X 108 NT pairs) and 30% single copy C0T; = 850, complexity = 2.9 X 10K NT pairs) sequences. The repetitive sequences are present, on the average, 30 times per genome. 2600 NT fragments similarly analysed show 95% reassocia- tion by C,,T 100, indicating that most 2600 NT fragments contain at least one repetitive ele- ment. Hyperchromicity measurements and S-l nuclease digestion of 2600 NT fragments re- associated to C,,T 100 show that these fragments contain single copy sequences interspersed with repetitive elements. The A-50 agarose size analysis of the S-l resistant regions demon- strate the presence of both 100-300 NT long repeated elements and clustered repeated ele- ments. In comparison to the deuterostome genome, the oyster genome contains a relatively smaller amount of the 100-300 NT long repetitive elements and a larger amount of clustered repeated sequences. This study was performed in the Embryology Course at the Marine Biological Laboratory. Increase in conduction velocity of lobster giant a.vons during grou'th. C. K. GOVIND, F. LANG AND J. W. BLOOM. Two pairs of giant interneurons may mediate the tail flip escape reflex in lobsters and crayfish. These are the lateral (LG) and medial (MG) giant axons running the length of the ventral nerve cord. In adult lobsters, conduction velocities of these giant axons range from 8-18 m/sec. We have measured conduction velocities of MG and LG during postnatal development in the lobster Homanis americanus ranging in size from 11 mm (3rd larval stage) to 220 mm (sexually mature adult). In this growth series the conduction velocity of MG increased from 2.5 to 10 m/sec and LG apparently increased from 1.7 to 8 m/sec. In the early stages, con- duction velocity of LG was not appreciably faster than the conduction velocity of other through conducting fibers in the ventral cord. Preliminary studies were made correlating this increase in conduction velocity with in- crease in axon diameter. Cross section of the ventral cord above the first abdominal ganglion in a 14 mm larval lobster revealed a clearly identifiable pair of MG with diameters of 17 ft and several smaller (6 M) axon profiles including one situated laterally. The conduction velocities at the corresponding stage is 3-5 m/sec for MG and 2 m/sec for smaller fibers. This clearly indicates that giant fibers grow substantially during postnatal development and that this growth is associated with a corresponding increase in conduction velocity. Supported by grants from the NRC of Canada (to C.K.G.) and NIH-NINDS (to F. L.) and a predoctoral fellowship from NRC of Canada (to J. W. B.). Formation of dogfish lens fibers from soluble crystaUins. G. GRIESS, S. ZIGMAN, AND T. YULO. The relationship between soluble lens crystallins and the fibers of lens cortical cells was investigated. Aqueous homogenates of dogfish (Mustclus canis) were divided into total PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 479 water-soluble (TSP) and water insoluble protein (TIP) by centrifugation at 3000 rpm. The TIP consists of fragments of individual fiber cells, as visualized in a light microscope (X400). The protein associated with the cell surface maintains the characteristic cell morphology. The protein content of TIP relative to total lens protein is relatively high in the dogfish compared to mammals, and the fraction increases with aging. Lens proteins compose a heterogeneous system, which was studied by combination of im- munochemical and electrophoretic techniques. Rabbit antiserum was prepared against lens TSP using Freund's adjuvant. Purified TSP and TIP were dissolved and sulfonated with Bailey's reagent and then dialyzed. Parts of these solutions were mixed with TSP antiserum, and the precipitate formed was washed and then solubilized in SM urea-tris-glycine buffer, pH 8.4. Analysis of the makeup of these samples was by polyacrylamide gel electrophoresis with urea, and sulfonated TSP and TIP were references. All components of the TIP could be identified with TSP components, as was further verified by the superposition of Ferguson plots (log mobility vs gel concentration) of these samples. A significant difference in molecular charge was found for only one of the com- ponents. When the TSP components were sorted by DEAE-cellulose chromatography and stepwise selective membrane ultrafiltration using Diaflo filters, a preponderance of low molecular weight protein (<30,000 daltons) was found. This situation differs from mammalian lens proteins, which exhibit species with predominantly higher molecular weights. This fact may reflect the earlier evolutionary origin of the dogfish, and seems to indicate more embryonic forms of lens proteins. The proteins associated with the fiber cell surface would affect the physical properties of the lens, such as deformability and dioptic power. The requirements for these properties in dogfish are obviously different from those in mammals, as is reflected by the lens protein pro- files. It appears advantageous for dogfish to make fibers from soluble crystallins of lower molecular weight. This work supported by PHS grant EY-00459 ; Monroe County Cancer and Leukemia Society. Variation in cornea! epithelial cell surfaces, as possibly related to cellular matura- tion and senescence (S.E.M. studies). C. HARDING, M. BAGCHI, S. SUSAN, AND H. JAMPEL. The corneal surfaces of marine fish have been previously found to have numerous micro- projections, which can best be seen with the scanning electron microscope. Five species of marine teleost were found to have a peculiar form of long curved ridges which are arranged concentrically. Two elasmobranchs, the dogfish and the skate, were found to have microprojec- tions somewhat similar to those seen in the mammalian cornea (Harding, Bagchi, Weinsieder, Peters, 1973, Biol. Bull., 145: 438). In the present study, cell-to-cell variation in the pat- tern of microprojections has been analyzed in the smooth dogfish (Musfcliis canis). Over the surface of an individual cell, the microprojections are typically uniform in appearance and distribution. From cell to cell, however, there may be significant differences in the micro- projection pattern. The pattern varies primarily in the length and number of microprojec- tions. Several distinct patterns can be discerned. For example, short numerous microvilli ; numerous, relatively long microvilli ; sparse, relatively long microvilli, as well as other pat- terns are found. These different types of cells can have an apparently random distribution. It is suggested that these different patterns reflect stages in the maturation and senescence of the surface epithelial cells. Supported by USAEC. contract AT(11-1)2401, and NIH. ( N. E. I.) special fellowship EY 55,624 to M. B. Theoretical analysis of amino acid uptake by toailfish liver. AUDREY E. V. HASCHEMEYER. Studies on the kinetics of various processes associated with protein metabolism in toadfish in TITO have necessitated evaluation of uptake rates for amino acids entering the liver from 480 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY the portal circulation. In the first few minutes after a pulse injection of an amino acid A (e.g., 14C-L-leucine) its movement may be described by a set of reversible monomolecular re- actions representing the transfer of A from extracellular hepatic space (Ae~) to intracellular space (<4iFree) with rate constants ki/k_i, and to protein via a k2 pathway, and the loss of A from hepatic space to the body and its return by a k3/k_3 pathway. Values for k:i and k_3 have been obtained from simultaneous measurements of disappearance of 3H-mannitol (Be) from liver after portal injection; k2 is given by the rate of llC-leucine incorporation into protein relative to AiFTee in 1-minute experiments. Saturation data for leucine uptake were used to obtain an estimate of the influx rate constant ki. Various values of k_i were then tested to determine the best fit of experimental data. The ratio of amino acid to marker substance in plasma draining from the liver (Ae/Br} after hepatic portal vein injection of tracer quantities was plotted vs. total marker recovery in liver at the time of sampling (Bv) and compared with theoretical curves. Good agreement was found for ratios of ki to k^i in the range of 4: 1. This result suggests that a concentrative uptake process occurs at normal plasma leucine concentrations. A facilitated diffusion model (ki/k_i=1.5) does not fit the experimental data. Supported by National Science Foundation grants GB 14570 and GB 42752. Spermio genesis in Crepidula. CATHERINE HENLEY. Mature spermatozoa of Crepidnla foniicatu and C. f>lana are normally polymorphic. The nucleated uniflagellated typical spermatozoon is apparently the functional male gamete. There are two size-classes of multirlagellated atypical sperm which, at the level of light microscopy, lack chromatin in the mature condition. One atypical type is much shorter and broader than the other. Spermatocytes from which typical spermatozoa develop are compact, with relatively little cytoplasm and with meiotic divisions of the usual type. Those giving rise to the two sizes of atypical sperm are larger and with disorganized meiotic figures. Atypical meiosis in- volves the loss of chromosomes from the spindle, after which they form cytoplasmic chro- mosome vesicles. These persist briefly in young atypical spermatids, in variable numbers, but eventually disappear. Spermiogenesis of typical sperm begins with the extrusion of flagella from the syncytial multinucleated mass of spermatids. The nuclei condense from a diffuse condition (as seen in Feulgen preparations) to a very dense one, and mitochondria aggregate around the growing flagella. The spermatids continue elongation and often retain the syncytial condition until they are nearly full-grown. Development of the two kinds of atypical spermatozoa begins in both cases from a rounded compact cell, from which variable numbers of free flagella are extruded. At the side of the non-syncytial spermatid opposite that where flagellar extrusion occurs, a slender process appears and grows in length. Concurrently the flagella continue elongation and eventually become completely incorporated within the cell. So far as can be determined from light microscopy of living material and Feulgen smear preparations of testis, the two classes of atypical spermatozoa differ mainly in length and breadth, and in the relative abundance and size of retractile granular inclusions. The long slender atypical spermatozoa have fewer and smaller inclusions than the second atypical gamete. Aided by a grant from the National Institutes of Health, GM 15311. Biolunnnescence in cell free extracts of the scale worm Harmothoe (Annelida; Polynoidae} ALBERT A. HERRERA, J. W. HASTINGS AND JAMES G. MORIN Cell free extracts in 10~2 M phosphate, pH 6.6, were prepared by homogenization of magnesium anesthetized Harmothoe iinhricata and c.rtciiuata collected locally. Classical 'luciferin — luciferase' tests were negative : cold and hot water extracts gave no luminescence upon mixing. However, luminescence lasting several minutes occurred upon addition of NADPH or NADH to cold water extracts. A factor augmenting this response occurs in hot water extracts. The NADPH response is sedimentable (50,000 X g, 1 hr.) while the stimulatory factor remains in the supernatant. Some alternative reductants (2-mercaptoe- thanol, ascorbate) were inactive, actually inhibiting the NADPH response. Reduced flavins were effective without exhibiting specificity, giving a flash lasting about 100 msec because of rapid autoxidation of the reductant. A similar but 100 X more intense response was obtained PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 481 with sodium dithionite, either by injection of oxygenated buffer to an extract previously re- duced by excess dithionite, or by injection of dithionite into an extract containing oxygen. Using this latter assay we determined some properties of this material. It is sedimentable but, unlike the NADPH response, no stimulatory factor occurs either in the supernatant or in heated fractions. The involvement of only a single factor was confirmed by showing that activity was directly proportional to dilution. Activity occurs over a relatively wide pH range, optimal at 6.6, and is destroyed by heat. About 50% is lost in 5 min at 30° C; with a high activation energy. In summary we have isolated a particulate bioluminescent system whose in vitro activity is stimulated by reductants. Both NADPH and dithionite responses are low in extracts of scales which have been induced to luminesce prior to extraction. Neither activity occurs in bodies of luminous polynoids or in whole nonluminous polynoids. Finally, the activity does not appear similar to that reported for luminous annelids from other families, including Chactoptcrus where the absence of polynoid-like activity was confirmed. A study of the vitellogenic protein in the sent in of estrogen-treated Ictalnrus nebu- losus. EILEEN D. HICKEY AND ROBIN A. WALLACE. Fish oocytes may synthesize much of their yolk reserves as well as sequestering a yolk precursor, vitellogenin, from the serum. The yolk proteins are also more heterogeneous than those of other vertebrates. Studies of catfish vitellogenin were initiated to characterize it, and to determine its contribution to the total yolk proteins of the oocyte, and to their hetero- geneity. Vitellogenin synthesis was induced by intraperitoneal injection of estrogen (1 mg/50 g body weight). Five days later the animals were injected with :12P sodium phosphate (50 /uCl/ 100 g body weight) and the blood taken by heart puncture after 24 hours. Serum was chromatographed on DEAE-Cellulose in a citric acid-2-amino-2-methyl-l-propanol gradient. Vitellogenin was identified as a 280 nm-absorbing peak associated with 32P activity and eluted at high ionic strength. This peak was absent in the serum of control male fish, and was present to only a slight extent in naturally vitellogenic females. Use of the proteolysis in- hibitor phenylmethylsulfonyl fluoride (PMSF) in the sample and all buffers was required to elute the vitellogenin in a single peak with which all :'"P was associated. An acetone pre- cipitate of this peak was analyzed by electrophoresis on 5% SDS-acrylamide gels. Both specific phosphoprotein and general protein staining revealed many bands. To determine whether this was the natural condition of fish vitellogenin or the result of handling, plasma samples from estrogen-treated and control male catfish were denatured and applied to gels immediately following collection in PMSF. No phosphoprotein bands appeared in the con- trols, and one major and a very weak minor band appeared in the sample from estrogen- treated animals, indicating that fish vitellogenin is probably a single high molecular weight protein as in other vertebrates. This study was performed in the Fertilization and Gamete Physiology Training Program at the Marine Biological Laboratory (NIH grant 3-T01-HD00026-12S1), and supported in part by a faculty grant from Russell Sage College. Genome sequence organisation in coelenterate. Aurelia, determined b\ DNA re- association kinetics. RATCHFORD C. HIGGINS. Several deuterostromes have been studied in terms of their genome sequence organization. In order to determine the generality of these recent studies in animals, I have selected a similar study of a coelenterate, Aurelia, a representative of a lower phylum. DNA was ob- tained from male jellyfish gonads and from embryos. The native DNA exhibits a hyper- chromicity of 28% and a t,,, of about 83° C. Two preparations were sheared in the Virtis homogenizer and sizes were determined by alkaline sucrose gradient sedimentation to be 250 and 2000 nucleotides (ntp) respectively. These DNA's were separately melted and reas- sociated under controlled conditions. Hydroxyapatite was used to measure extent of reas- sociation. The 250 ntp fragments demonstrated two kinetic components, one 53% repetitive and the other 36% unique. The repetitive component was estimated to be present 170 times 482 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY per haploid genome. The 2000 ntp fragments reassociated with kinetics similar to the repetitive component of the 250 ntp DNA. This suggests the unique sequences are adjacent to repetitive sequences. SI single strand specific nuclease was used to destroy all but duplexes formed by repetitive sequences in reassociated unsheared DNA. These duplexes were then sized on agarose A50 columns showing two major size classes. About 40% of these duplexes are greater or equal to 2000 ntp and the rest are about 300 ntp. These observations are in accord with the emerging picture of animal genome structure throughout the animal kingdom. Soinc aspects of transmission at synapses in the labyrinth oj the toad fish. S. M. HIGHSTEIN, J. KEETER, M. V. L. BENNETT. The posterior semicircular canal was removed from adult toad fish (Opsanus tan) and placed in a bath (167 HIM NaCl, 2.5 HIM KC1. 1.5 imi CaCU, 15 HIM Tris or Pipes pH 7.3). Paired pipettes (100 n tips) in the canal were used to pass current and record voltage across the sensory epithelium. Single afferent fibers were silent or spontaneously active at regular or irregular rates of 5-80/sec. Lumen positive (LP) transepithelial pulses which would de- polarize the presynaptic faces of the receptor cells and hyperpolarize the lumenal faces in- creased the frequency of firing. Lumen negative ( LN ) pulses, which would hyperpolarize the secretory faces " decreased spontaneous activity and caused off responses. Frequency-voltage curves were linear over a wide range but tailed off near zero frequency and reached a plateau of 60-300/sec at transepithelial voltages of approximately 200 mV. Larger voltages then decreased frequency. As LP pulses increased, latency of the first spike smoothly decreased to a minimum of 0.5-1 msec. Direct stimulation of the nerve near the sensory epithelium with wire electrodes evoked responses at latencies of less than 0.1 msec. Substituting Mg for Ca (1.5 HIM) in the saline' slowed or stopped spontaneous activity and shifted the frequency voltage relation towards higher voltages. We conclude from the irreducible latency and Mg inhibition that the hair cell synapse is chemically mediated. Transepithelial stimuli ap- parently act directly on presynaptic secretory membrane and in tonically active units ongoing transmitter release is increased or decreased by stimuli of appropriate polarity. Large LP pulses may be less effective because Ca entry into the hair cells is reduced. Bath applied Noradrenalin 0.1-1 nm slows tonic activity markedly and shifts frequency voltage curves toward higher voltages. Electron microscopy reveals terminals containing dense core vesicles as well as afferent and efferent synapses. These results suggest a sym- pathetic innervation of the labyrinth. Effects of external monovalcnt cations on ouabain inhibition rate oj sodium pump in squid giant a.ron. ANN S. HOBBS. Effects of external Na+, K\ and Cs* on the ouabain inhibition rate of sodium pump in giant axons from Lolii/o pcalei were determined by following sodium pump activity (as ~Na efflux) in the presence of various concentrations of these ions and 10"T M ouabain. Relative pump rate was also computed, as the ratio of total ouabain sensitive efflux in the experimental solution to total ouabain sensitive sodium efflux in Na, 10 K ASW. In sodium containing ASW, 5, 10, and 20 HIM K+ solutions gave relative pump rates of 0.59, 1.0, and 1.27 and rate constants (expressed as min"1) for ouabain inhibition of 0.0596, 0.0484, and 0.025. When sodium was replaced by choline, 0, 10, and 100 mM K+ solutions gave relative pump rates of 0.60, 1.6, and 2.04 and rate constants for inhibition of 0.0141, 0.0072. , and 0.0036. Raising [Cs+] in sodium ASW from 25 to 75 nm produced an increase in relative pump rate from 0.45 to 1.06, but only a slight drop in the ouabain inhibition rate constant: from 0.0517 to 0.0474. In choline zASW neither relative pump rate (1.65) nor ouabain inhibition rate constant (0.022) changed significantly when [Cs+] was raised from 25 to 150 m\i. In both sodium and choline ASW, K+ appears to slow the ouabain inhibition rate. Sodium, conversely, enhances the rate. For equivalent pumping rates, sodium produces about a four-fold increase in the ouabain inhibition rate. In the absence of Na+, K+ solutions have PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 483 less than half the rate of ouabain inhibition as Cs+ solutions showing the same relative pump rates. The fact that these differences are seen even at equivalent pump rates implies that this is more than a difference in the occupational state of pump stimulatory sites. It could imply that the sites affecting ouabain binding are physically separate from the pump sites, or that the transport mechanism is different according to what particular ion is present externally. Supported by the Grass Foundation. Analysis of fast a.vonal transport in the sonic motor nerve of the toad fish. P. N. HOFFMAN, J. L. BARKER, H. GAINER AND R. J. LASER. The fast component of axonal transport was studied in the sonic motor neurons of the toadfish. Axonally transported polypeptides were labeled by the local injection of microliter quantities of either 35S-methionine or a mixture of 3H-leucine and 3H-lysine into the region of the medulla containing the motor neuron cell bodies. In one series of experiments cells were labeled with either 3H-fucose alone or in combination with ^S-methionine in order to label glycoproteins. Immediately after labeling, one sonic motor nerve was ligated near the point at which it enters the swim bladder, and the other nerve was left unoperated. Animals were sacrificed at various times after injection, and the radioactivity in 3 mm segments of the sonic motor nerves determined. A distinct peak of radioctivity was found moving at a rate of 70 mm/day. Radioactivity was found to accumulate in the region of the nerve im- mediately proximal to the ligature at times which were consistant with the rate of movement of the labeled peak. The absence of radioactivity in regions of the nerve distal to the ligature indicates that the local incorporation of blood-born precursors does not significantly contribute to the labeling of the nerve. Analysis of the labeled polypeptides of the fast component using SDS-polyacrylamide electrophoresis revealed the presence of a complex pattern of labeled polypeptides containing at least five major species with approximate molecular weights of 130,000; 60,000; 30,000; 18,000 and 5000 daltons. Electrophoretic analysis of the fast com- ponent material simultaneously labeled with :!H-fucose and ^S-methionine indicates that these major polypeptides are also labeled with fucose, suggesting that they are glycoproteins. These results are consistent with those of previous studies of the fast component. Metabolism of cysteine in relation to tJie synthesis oj isetJiionate by squid nerve. FRANCIS C. G. HOSKIN, MICHAEL L. POLLACK, AND ROBERT D. PRUSCH. Isethionate, HO-CHiCH^SOa', is the major anion of cephalopod nerve, occurring in squid giant axons, for example, at about 2CO mM. A parallel has been noted in cephalopod nerve between the presence of isethionate and an unusual enzyme termed "squid nerve-type DFPase." Since DFP, (C3H7O)2-P(O)F, is clearly not the natural substrate, it has been speculated that this enzyme may play a role in the presumed metabolic pathway from cysteine to isethionate. In order to explore this relationship we have examined several aspects of the metabolism of 14C- labellfd cysteine by squid nerve. For this purpose an improved radiochemical assay has been developed which involves the gas transfer of "COs out of the alkali in which it was trapped in Warburg vessels and into a scintillation fluid capable of absorbing 1 millimole CO2 per ml (pre- blend 3A80, Research Products International). The results show that, of the cysteine entering squid axons (240% of equivalent distribu- tion in 1 hr from an external concentration of 10~5 M), 30% is decarboxylated and only 1% is further and completely degraded to CO?. Carbons 2 and 3 of the other 29% are retained or metabolized in a different manner. Pyruvate, a somewhat similar 3-carbon compound, is metabolized quite differently by squid nerve. First, it is taken up at a much lower rate than cysteine (25% of equivalent distribution). Secondly, of the pyruvate entering squid axons, 100% is decarboxylated. Thirdly, 75% is further and completely degraded to CO?. Although we estimate that 3 to 5% of the cysteine entering squid axons is metabolized to isethionate or isethionate intermediates, we have not yet been able to relate the so-called DFPase to isethionate synthesis. The techniques for evaluating these metabolic pathways should now be applicable to this problem. This work was supported by NIH grant NS-09090. 484 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY Preliminary studies on natural decomposition o\ Fucus in tzvo marine environments. R. DOUGLAS HUNTER. Studies on the change in dry weight, ash and organic carbon were carried out on de- composing Fucus I'csiculos-iis in summer 1974. Dry Fitcus in mesh bags (1 mm2 mesh size) was placed on the bottom of a rocky shore one meter under mean low water and another set of bags on the bottom of a salt-marsh tidepool. Bags were collected weekly from each en- vironment. Oxygen concentration was 5.2 ml O2/l at 2\° C at the rocky shore site and 0.14 ml O»/l at 25.6° C at the salt-marsh site at the depth of the mesh bags (4 cm under loose detritus). The bag contents at both locations rapidly decreased in dry weight and carbon so that after one week of decomposition 57% of the dry weight and 50% of the carbon remained. This decline continued at the rocky shore site but stopped at the salt-marsh site with about 40% of the dry weight and carbon remaining. Ash content of the rocky shore samples in- creased from 25% to 35% in the first 16 days. Carbon per mg ash-free dry weight changed only slightly at both sites over the course of the experiment (44 days), however the salt- marsh samples were consistently higher than those at the rocky shore, suggesting differences in the nature of the decomposer organisms at the two sites. In vitro experiments to measure short-term leaching of ash and carbon from dry Fucus showed that a rapid loss of both of these components occurred in the first 20 minutes fol- lowed by a gradual increase in ash and a slow decline in carbon. Nitrogen analyses are in progress, but these results already suggest fundamental differences between the rates and nature of decomposition in FHCIIS at a rocky shore environment as op- posed to a salt-marsh environment. Supported by a Grant-in-Aid of Research from Sigma Xi, a faculty research fellowship from Oakland University, and NSF Grant GB-36757 to W. D. Russell-Hunter. Mitosis and nuclear morphogenesis in Barbulanympha. SHINYA INOUE AND HOPE RlTTER. Utilizing Ritter's successful culture of symbiotic flagellates of the wood roach Cryptdccrcns fiDictulatus, we have followed single Barbulanympha cells through division by combined polar- rized light and differential interference microscopy. In Barbulanympha a persistent nuclear envelope assists in the interpretation of chromosome movement : as described by L. R. Cleveland ct al. (1934, Alcm. Amcr. Acad. Arts Sci., 17: 180), kinetochores embedded in the nuclear envelope attach to astral rays and link chromosomes to the extranuclear spindle. Shortening of the astral ''chromosomal" fibers: (1) draws the nucleus to the previously formed spindle, (2) assists in spindle envelopment by the nucleus and (3) results in kinetochore movement to the stationary centrosomes ( anaphase A ) . Subsequently, chromosome sets sepa- rate further with their respective centrosomes while the spindle elongates several fold (ana- phase B). During anaphase A the mid-portion of the approaching nucleus is blocked by the spindle as the outer margins are pulled around the spindle by shortening astral rays. The mar- gins of the nucleus meet anteriad to the spindle and gradually form a seam parallel with the spindle axis. The resultant nuclear tube keeps the elongating spindle from splaying in early ana- phaseB. The seam then opens as with parting lips at midpoint along the spindle ; the resulting cleft enlarges towards both poles as well as perpendicular to the spindle axis. As the spindle and nucleus elongate, the cleft expands laterally then posteriorly around the spindle to finally complete karyokinesis. For any cell type we believe this is the first description of the process of karyokinesis accomplished through nuclear morphogenesis. Supported by NSF. Grant #GB-31739X and NIH Grant #5-R01-CA-10171-09. Effects of subzero temperatures on eggs and embryos of Arbacia punctnlata. S. C. JACKOWSKI AND R. A. WALLACE. Eggs and early embryos of Arbacia punctnlata were exposed to subzero temperatures and assayed for normal development by examination under the light microscope. Samples were PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 485 suspended in sea water and cooled to -5.8° C, then seeded with ice. Ten minutes later samples were cooled to temperatures ranging from —10.5° to —27.0° C at a rate of 0.7° Q minute and were immediately warmed at a rate of 2.1° C/minute. Eggs were subsequently fertilized or embryos were cultured in sea water 22° C and scored for viability at the 4-cell or blastula stages, respectively. After cooling to -5.8° C, viability of eggs decreased 30%; cooling to -10.5° resulted in a 98% decrease. Viability of embryos cooled to —5.8° C five minutes to 114 minutes after fertilization decreased only 4%. Embryos exhibited a differential susceptibility to lower temperatures dependent upon cell size and state: viability of 15-minute 1-cell embryos, 2-cell embryos, and 4-cell embryos decreased 43%, 76% and 22%, respectively, after cooling to -15.0° C whereas viability of embryos involved in first cleavage division and second cleavage division decreased 85% and 90% respectively. Similar patterns of susceptibility were ob- tained after cooling to -10.5° C and -17.0° C. Viability of embryos cooled to —27.0° C, re- gardless of cell size and state, did not rise above 5%. These studies demonstrate that there is a cell cycle dependency in the susceptibility to freezing damage; in addition, they indicate that future research into the protection of embryos at subzero temperatures is feasible and should be initiated at the 4-cell stage. This study was performed in the Fertilization and Gamete Physiology Research Training Program at the Marine Biological Laboratory (NIH grant 3-T01-HD00026-12S1). Fertilisation-associated changes in the plasma membrane proteins of Arbacia punctulata eggs. JAMES D. JOHNSON, BONNIE S. DUNBAR, AND DAVID EPEL. Ultrastructural studies on fertilization indicate that the plasma membrane is altered after fertilization ; this change primarily results from the fusion of the membranes of the cortical granules with the plasma membrane. To study egg plasma membranes after fertilization, the surface proteins were enzymati- cally iodinated as described by Dunbar, Johnson and Epel (Biol. Bull, abstract this issue). Only the egg surface was labeled as determined by autoradiography and loss of label fol- lowing pronase treatment. The eggs were activated with the Ca++ ionophore A23187, and within 20 min of activation 27% of the labeled protein is released into the seawater. To further characterize the fate of the labeled membrane, homogenates of iodinated eggs were placed on discontinuous sucrose gradients. Only one radioactive peak was detected at 40% sucrose (w/v) in both activated and unactivated eggs. When eggs were activated with ionophore or fertilized sperm, (1) 27% of the labeled protein is released into the seawater, (2) 30% is found in the cell sap as a small molecular weight species (TCA soluble) and (3) ap- proximately 40% remains associated with the membrane. When eggs were activated with 10 HIM ammonia (no cortical granule breakdown) only the 27% loss to seawater was observed. Ten major radioactive species were shown by SDS gel electrophoresis of labeled mem- branes from unactivated eggs, and only one major labeled component from membranes after activation of eggs. These data reveal striking changes and reorganization of the plasma membrane accompany- ing fertilization. One change (27% loss to sea water) is associated with metabolic activation. Other changes are associated with cortical granule breakdown of membrane fusion during normal activation. These include reutilization of 30% of original labeled membrane proteins and possibly reorganization of the remaining protein components. This study was performed in the Fertilization and Gamete Physiology Research Training Program at the Marine Biological Laboratory (NIH grant 3-T01-HD00026-12S1). Fine structure and permeability studies of a rectifying clcctrotonic synapse ]. S. KEETER, M. DESCHENES, G. D. PAPPAS AND M. V. L. BENNETT. In the first three abdominal ganglia of the crayfish, both the lateral septate and medial giant fibers form electrotonic synapses with the giant motor fiber of the third root. Junc- tional membrances of these synapses rectify (Furshpan and Potter, 1959, /. PhysioL, 145: 289). When the potential in the lateral or medial giant fiber (presynaptic) is positive relative to 486 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY the potential in the motor fiber (postsynaptic), junctional resistance is low. At rest, and when the motor fiber is more positive, junctional resistance is high; thus orthodromic trans- mission is favored over antidromic. Pre- and postsynaptic membranes are closely apposed at small regions, in contrast to the wider separation at chemically transmitting synapses (Stirling, 1972, Z. Zclljorsch., 131: 31). Our studies on Procambanis clarkii show that the junctional separation is only 40-50 A, similar to that demonstrated at gap junctions in septa between segments of the lateral giant fibers Septal gap junctions appear symmetrical in that pleomorphic vesicles are present in the cytoplasm on both sides. The rectifying junction ap- pear asymmetrical, dense material is often associated with the postsynaptic junctonal membrane, and a few pleomorphic vesicles are often found presynaptically. We assign no function to the vesicles which are often found in the axonal side of non-rectifying axosomatic and axo- dendritic synapses. Intracellular injected fluorescein crosses septal gap junctions and a num- ber of other non-rectifying gap junctions ( Pappas and Bennett, 1966, Ann. Nezv York Acad. Sci., 137 : 495). We do not see fluorescein crossing the rectifying synapse in either direction, al- though electrical transmission remains intact. Relative impermeability to fluorescein is also found in some embryonic electrotonic junctions. The morpohological bases for these perme- bility differences remain illusive. Activation of sea urchin sperm chromatin template for DNA synthesis by Ca~+, Mg2+dependent endonuclease present in oocytes of Arbacia punctulata. S. S. KOIDE, L. BURZIO AND Y. T/ANIGAWA A Ca2+, Mg2+-dependent endonuclease was partially purified by homogenizing washed oocytes in 25 HIM Tris-HCl (pH 8.0), 2 mM MgCl with a Bounce homogenizer. The homogenate was centrifuged at IS.OOOr/ for 15 min. The supernatant was fractionated by pre- cipitation with (NH4)aSO4 between 0-50% and 50-85% saturation. The sediment was suspended in a medium containing 0.01 M Tris-HCl (pH 8.0), 30% glycerol. Chromatin was prepared from sperms washed three times with a solution of 75 mM NaCl, 24 niM EDTA (pH 8.0) and suspended in 0.01 M Tris-HCl (pH 8.0). The sperm suspension was homogenized in a Dounce homogenizer. The homogenate was centrifuged at 15,000<7 for 15 min. The supernatant was designated as sperm extract. The sediment was suspended in 0.01 M Tris-HCl ( pH 8.0) and the process repeated. The sediment was resuspended and layered over a medium containing 1.7 M sucrose, 0.01 M Tris-HCl (pH 8.0), 1 mM EDTA. The tubes were centrifuged in a SW 25.1 rotor at 22.000 rev/min at 2° for 4 hr. The sedi- ment was resuspended in a solution containing 40% glycerol, 0.01 M Tris-HCl ( pH 8.0). The endonuclease activity was determined by measuring the amount of radioactivity solubilized from [:;H]DNA gel. The template activity of chromatin DNA synthesis was as- sayed by a standard method using E. coli DNA polymerase. Oocytes contained high alkaline endonuclease activities. The activities in the fractions precipitated between 0-50% and 50-85% ammonium sulphate saturation were 3.64 X 10:1 units/ mg of protein and 11.25 X 103 units/mg respectively. On the other hand, seminal fluid and sperm extract (0.01 M Tris-HCl fraction) possessed very low enzymic activities . Sperm chromatin was a poor template for DNA synthesis. On incubation with partially purified oocyte endonuclease, the template activity was greatly enhanced. These preliminary results suggest that oocyte Ca2+, Mg"+-dependent endonuclease might play a role in the as- sociation of sperm and oocyte DNA subsequent to fertilization and might activate chromatin for DNA synthesis. This work was supported by NIH Grant P01HD05671. Fish lateral vestibular neurons: electrotonic transmission from primary vestibular afferents, electrotonic coupling between Z'estibitlo spinal neurons and identifica- tion of efferent cells to the labyrinth. H. KORN, C. SOTELO, N. KOTCHABHAKDI AND M. V. L. BENNETT. Field and intracellular potentials were recorded in lateral vestibular nuclei of the toad fish, Opsainis tan. Vestibule spinal neurons were identified by their antidromic spikes fol- PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 487 lowing spinal stimulation. In these cells, stimulation of the ipsilateral eighth nerve evokes two component EPSPs, the first of which is transmitted electrotonically from primary vestibular afferents as indicated by its short delay. The latency of this first component was 0.34 msec (SD = 0.06; n = 34); the time of arrival of the presynaptic volley was 0.28 msec (SD = 0.08; n = 18) as measured from the peak of the extracellular field potential, and 0.27 msec (SD = 0.08; n = 43) as measured in presynaptic fibers. Thus the computed synaptic delay was 0.06 msec, which is too short for chemical mediation. Spinal stimulation evokes graded depolarizations in vestibule spinal neurons. The short latency (<0.02 msec) of these potentials suggests electrotonic spread of antidromic activity in neighboring neurons. Similar graded depolarizations were also recorded in primary vestibular afferents, therefore coupling of vestibulo spinal neurons appears to be at least in part by way of these fibers. In confirmation electron microscopy studies reveal club endings with numerous gap junctions on LVN somata and proximal dendrites. In some cells simulation of the ipsilateral nerve evokes anti- dromic spikes. The presence of LVN neurons which send their axons to the labyrinth was confirmed by their heavy staining with Procion Yellow following axonal electrophoresis. Graded stimuli applied to the eighth nerve evoke graded short latency depolarizations as well as long latency EPSPs; the former could indicate electrotonic coupling of the efferent cells or electrotonic transmission from primary afferents resulting in a short latency feedback loop. Electrotonic coupling of LVN neurons should reduce the latency of the responses to physio- logical stimuli, and also, as in other systems, may cause increased synchronization of post- synaptic firing. The relationship of small mode miniature end plate potentials and quanta! content of evoked responses in the stimulated frog neuromuscular junction. M. E. KRIEBEL, G. D. PAPPAS AND S. ROSE. There are two classes of spontaneous miniature endplate potentials (MEPPs) which are readily observed in small edge fibers of the sartorins muscle ( Kriebel and Gross, 1974, /. Gen. i'hysiol.. 64: 85). In the unstimulated preparation, amplitude histograms show that about 4% of the MEPPs form a distinct mode about 1/7 that of the major mode (which may range from 2.4-4 mV). Prolonged nerve stimulation (10 Hz, 15-45 min) reduced most MEPPs to small mode size. During the course of stimulation, the evoked responses became smaller until they appeared to be multiples of the small mode MEPPs. Finally, EPP amplitude histograms had the same profiles as MEPPs. In unstressed preparations, we adjusted ex- tracellular Mg++ (up to 5 HIM) and decreased Ca++ (to i> ) until failure of evoked responses were observed with 1 Hz nerve stimulation. The affect of excess Mg++ was progressive on both the MEPP and EPP amplitudes. The proportion of the small mode MEPPs increased and the mean of the major mode decreased. When the percentage of MEPPs contributing to the major mode was relatively small, many stimuli failed to elicit EPPs and some EPPs were the same size as small mode MEPPs. It is not clear at this time what the relationship is between small mode MEPPs and EPPs in high Mg++ and low Ca"+. The effect of D20 on sodium efflux from the squid ijiant a.von. D. LANDOWNE AND V. SCRUGGS. Replacing 97% of the H2O in the bathing sea water rapidly and reversibly reduced the efflux of "Na from the squid giant axon by about 35%. The effect appeared to be linear with respect to D2O concentration. In potassium-free sea waters D2O reduced the sodium efflux by only 5-10%. This is not because D2O affects only the potassium-sensitive sodium efflux as is shown by injecting arginine into the axon, reducing the potassium sensitivity. In these axons D2O produced a larger drop in sodium efflux than potassium free sea water. D-O also reduced the efflux in potassium-free sea waters by about 40%. The effects of D2O were fairly well mimicked by lowering the temperature a few degrees. That is to say the sodium efflux into potassium-free sea water was less temperature sensitive than the efflux into normal sea water. This work was supported by a National Science Foundation Grant GB-36859. 488 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY Anchorage of cytoplasmic filaments to intramembrane particles in the plasma mem- brane and in the nuclear envelope: transditction of membrane receptor topog- raphy to the genome. D. S. LEE AND J. METUZALS. Small blocks of rabbit brain from the lateral vestibular nucleus were isolated in 0.25 M sucrose, standard salt solution (Ishikawa ct al., 1969, /. Cell Bio]., 43: 312), 20% glycerol and freeze-etched. In electron micrographs of fractures of Deiters' neurons, 60 A filaments extend from the cytoplasm into the A face of the plasma membrane fracture where the fila- ments anchor with 80-150 A intramembrane particles. A regular cross lattice pattern of ar- rangement of intramembrane particles with short 60 A fiilament tails attached can be observed in plasma membrane fractures of unidentified cellular processes. In the cytoplasm of such processes filaments are arranged in a lattice of the same orientation. In fractures of the nuclear envelopes, 60-150 A filaments extend from the nuclear pore complexes into the fracture faces of the nuclear envelope membranes and anchor to the intramembrane particles of these membranes. Intramembrane particles often border the nuclear pore complexes. Small blocks from the lateral vestibular nucleus were treated with glycerol, standard salt solution and HMM, fixed and embedded. In electron micrographs of thin sections of such blocks, numerous actin- like filaments, labelled by HMM, can be seen in unidentified cell bodies and processes. These observations extend the reported findings on neurofilamentous network in Deiters' neurons by Metuzals and Mushynski (1974, /. Cell Biol.. 61 : 701). The neurofilamentous network establishes polarity through the anchorage to the intra- membrane particles of the plasma membrane and of the nuclear envelope. Such a polarity may have a twofold significance. First, in maintenance at the outer surface of the neuron, of a specific receptor topography in relation to the spatial order of the genome in the nucleus. Secondly, in transducing changes in receptor topography in the plasma membrane to localized and specific transcriptional responses of DNA. An experimental approach to^vard understanding the role of mciofaiina in a detritus based marine food urb. JOHN J. LEE, KENNETH TENORE, JOHN H. TIETJEN, CARMINE MASTROPAOLO. The rate of detritus degradation was measured in model systems maintained in 3 chemo- stats. Each reaction vessel (surface area <~ 5 X 10~3 m2) contained — 160 g of sediment ( — 4% detritus) removed from the surface of a salt marsh at low tide. The sediment was carefully frozen and thawed 3 times to kill most of the indigenous organisms and was asepti- cally cut and placed in the bottom of the reaction vessel while frozen. Microflora dislodged from surface detritus in the marsh by agitation and filtered through 8 /UM membrane filters (Millipore SC) was inoculated into each reaction vessel along with 0.5g JIC labeled (0.05 yttCi/g) Zostera detritus (ground to 120/j. particle size). One reaction vessel served as con- trol, the 2 others were inoculated with either CapitcUa capitata (~ 10,000/nr) or Ncphtltys iucisa (~1000/nr). One reaction vessel also contained the following meiofauna : nematodes — Chromodorinc gcrmamca (~ 2.5 X 10r'/nr) , Mouhystcra dcnticitlata (3.0 X 10r'/nr ) , Rhab- ditis marina (2.0 X 10"/rn2) ; foraminifera — Quinqueloculina scininuluin (5XlOs/m2), Elplii- diuin inccrtnm (5 X 103/rn2), Allogromia luticollaris (1.0 X 104/nr) ; a ciliate, Euphtcs ranniis (1 X 10°/nr) ; and an harpacticoid copepod — Leptocheirus sp. (5 X 10'Ynr). The flow rate in the chemostats was adjusted so that the sea water in the reaction vessels was replaced daily. Aeration was vigorous. In addition to the radioactivity measured daily in the overflow vessel, "CO2 was trapped in an additional vessel with saturated KOH. Under the experimental condi- tions, it took ~ 6 days for the microbial population colonizing the detritus to reach stable levels (as judged by 14C turnover). After 6 days, under the experimental conditions tested (26° C) <~ 1/20 of the radioactivity was released from the detritus per day. Total com- munity respiration was 3 X higher in the system with both meiofauna and Capitclla than in bacterized controls; respiration in the system with CapitcUa alone was ~ 50 controls. Bacterial populations were highest in the system with CapitcUa alone and lowest in the sys- tem with the meiofauna. No statistically significant differences in the radioactivity of Capi- PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 489 tclla or Ncphthys were discovered between those foraging alone or with meiofauna. It is possible that the specific activity of the labeled detritus was too low to detect such differences. This research was supported by NSF grant GA 28454. Aeqnorin stndv of the suppression potential in the squid giant synapse. R. LLINAS AND C. NICHOLSON. Stellate ganglia of squid (Lolif/o pcalci) were isolated and perfused in oxygenated arti- ficial seawater (ASW). The pre- and post-fibers forming the giant synapse were impaled. Two electrodes were introduced into the pre-fiber at its terminal. One contained 100 mM TEA Br for current injection and the other aequorin in 200 mM KC1 to record potential. The post-fibers were impaled in the immediate vicinity of the synapse. Electrophoretic in- jection of TEA for three to four hours into the pre-fiber blocked delayed rectification. Aequorin was then pressure injected for a protracted time and the Ca++ induced light emission accompanying single prolonged action potentials in the pre-fiber was detected with a photo- multiplier. The light emitted increased with the duration of the spike and could be detected 10 to 15 msec after spike initiation. Following TTX-ASW (10'Bg/ml) graded synaptic potentials produced by increasing presynaptic depolarization could be correlated with the amplitude of the aequorin flash. With presynaptic depolarization, beyond 150 mV (on the average), a suppression of the synaptic release occurred which returned at the end of the depolarization. In a parallel manner, the aequorin response was at maximum during maximum transmitter release and was absent at the suppression potential level, reappearing at the end of the current pulse. These results demonstrate directly that synaptic transmitter release is accompanied by an intracellular increase of Ca++ concentration, and that the suppression po- tential is produced by a potential-dependent decrease of intracellular Ca++. Two mechanisms can explain our results, (a) the suppression potential represents the Ca++ equilibrium potential (b) Ca++ conductance is inactivated by large membrane depolarizations. We favor the first mechanism since no large resistance increase was seen at the suppression potential in the pre- fiber. Supported by NINDS grant NS-09916. Effects on Arbacia hatching enzyme of 5-hydroxytryptamine and N-benzyloxycar- bonyl-L-glutamate. GARY W. LOPEZ AND DENNIS BARRETT. Deeb has reported that the hatching of the Mediterranean Arbacia at hatching stage is specifically prevented by 5-hydroxytryptamine administered at fertilization to 8-cell stage. Our attempts to repeat the effect reveal that in the American Arbacia, 5-hydroxytryptamine exerts instead a general retarding effect on development. We observed the half-time for the hatching process, in embryos incubated at 200/ml in millipore-filtered sea water at 23.5° C, as 6.6 hr, to an error of 0.3 hr, both in controls and in embryos treated from 0.5 hr after fertilization with 20 to 200 /xg/ml of the inhibitor. Beginning at about hatching time, the inhibitor causes a general retardation, demonstrated by the extent of development at 30 hr of incubation. Controls were: full pluteus ; in 20 /xg/ml : early pluteus (with short arms); in 50 Mg/ml : early pluteus ; in 100 /ug/ml : late prism ; in 200 Mg/ml : ovoid gastrula to early prism. Some delay in the hatching of the last approximately 30% of embryos in 100 or 200 Mg/ml of inhibitor we attribute to this latter effect. Having carefully tried to duplicate the conditions of Deeb's experiments, we have no explanation to offer for our disparate results except species differences between the congeneric Arlnicia. The hypothesis that the hatching enzyme of Arbacia. like that of Strongylocentrotus pur- puratits, is an endopeptidase with affinity for glutamyl-bonds, was tested by using blocked glutamate as an inhibitor. Crude hatching enzyme was prepared by ultrafiltering the super- natant above hatched blastulae ; substrate was casein, with lysine residues e-dimethylated to reduce background at pH 8; release of N-terminals was monitored by fluorescence after re- action with fluorescamine in acetone. Arbacia pioictulata hatching enzyme digests modified casein with a KM of 0.4 mg/inl ; N-benzyloxycarbonyl-L-glutamate (carbobenzoxyglutamate) competitively inhibits the process with a Ki of 2.4 mM. Work was supported by training grant 2T01 GM 00265-16. 490 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY Purification and properties of an Arbacia sperm fertilisation antigen. UTPALENDU S. MAITRA AND CHARLES B. METZ. Univalent antisperm antibody (Fab fragment), made by papain digestion of rabbit anti Arbacia sperm antibody, inhibits fertilization. Fab fragments evidently inhibit fertilization by- binding to and blocking one or more sperm surface antigens essential for fertilization. Sperm extracts (prepared by repeated freeze-thawing and centrifuging at 105,000 X g for 60 min) neutralized the inhibiting action of Fab fragments (Cordel and Metz, 1973). The extracts evidently contain the essential "fertilization antigen (s)." This antigen was purified by fractionation on DE-52 column (0 to 1.4 M NaCl gradient in 0.01 M sodium phosphate pH 7.5). Five pools were made from the column fractions. Although pool II and III contains detectable antigens, they cannot neutralize the effect of Fab fragments. Pool V completely neutralized the action of univalent antibody, while pool IV did do so only partially, probably from cross contamination by pool V. Pool V was further purified by chromatography on DE-52 (0 to 0.7 M NaCl gradient). The protein thus prepared show's only one band upon polyacrylamide gel electrophoresis (7.5% native gel). The material stained with both PAS and alcian blue and hence is a glyco- protein. In a preliminary attempt at localization, purified antibody against this antigen was iso- lated by immunoprecipitation. When specific antibody treated sperm were challenged with fluorescine labelled goat anti-rabbit antibody, only the sperm head fluoresced suggesting that the antigen is located in the sperm head surface. The protein had no effect on the eggs as observed under light microscope. The purified protein was labelled with 12nl using chloramine T. Purified ^I-antigen bound to dejellied Arbacia eggs. The binding was partially blocked by Fab, while non-immune Fab had no effect. This study was performed in the Fertilization and Gamete Physiology Research Training Program at the Marine Biological Laboratory (NIH grant 3-T01-HD00026-12S1). Responses to low o.v\gen stress in relation to the eeoloc/v of littoral and siiblittoral snails. ROBERT F. MCMAHON AND W. D. RUSSELL-HUNTER. Six species were studied including the littoral species Littorina littorca. Littorina obtusatu and Littorina sa.ratilis and the sublittoral species Acmaea tcstitdinalis, Mitrclla lunata and Lacuna I'incta. Respiration rates in sea water with decreasing oxygen tension were recorded using oxygen electrodes before and for 25 hours after a five hour period of low oxygen stress. After such stress, A. testndinalis and L. sa.ratilis pay a large conventional oxygen debt on return to full saturation while /.. littorea pays a somewhat smaller one. There was no evi- dence of any regular oxygen debt for L. obtusata, M. lunata or L. rincta. After five hours of low oxygen stress L. littorca, L. obtusata, L. sa.ratilis, M. litnata and L. I'incta all show elevated oxygen uptake rates at all oxygen tensions and particularly at low oxygen tensions for at least 25 hours. However, after five hours low oxygen stress A. tcstitdinalis showed little change in oxygen uptake rate with decreasing oxygen tension. The degree of elevation of oxygen uptake rate after low oxygen stress varied among the five species, and was least in L. sa.ratilis and greatest in L. I'incta. Assessed as percentages of the prestressed rates at full saturation, the oxygen uptake rate at 10% oxygen tension (about 0.526 ml O2/l at 760 mm Hg and 20° C) before stress for L. sa.ratilis was 3.0% and for L. rincta was 8.5%. After 25 hours return to full saturation from low oxygen stress the corresponding rates were 31.8% for L. sa.ratilis and 222.1% for L. rincta. In these five species, this increased oxygen uptake rate (which also occurs in the freshwater snail Lacvapcx fusciis] after return to fully saturated sea water from low oxygen stress appears to be a short-term acclimation of respira- tory rate allowing the snails to maintain relatively high metabolic rates if low environ- mental oxygen tensions are encountered. Supported by Research Grant #15-635 from organized Research Funds of the University of Texas to R. F. M. and National Science Foundation Research Grant #GB-36757 to W. D. R-H. PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 491 Filamentous network of flic a.voplasin, as revealed b\ freeze-etching of the squid giant nerve fiber, in relation to actin, ttibnlin and niyosin components. J. MKTUZALS AND \V. E. MUSHYNSKI. (.-1} Electron microscopy of freeze-etched squid giant nerve fibers reveals in the axo- plasm curved, sinusoidal and ringshaped filaments of 20-200 A in diameter, which are as- sembled into a continuous space network formation. The individual filaments are continuous with irregular dense tangles of 20 A filaments and 50 A granules. (B) Actin-like filaments can be identified regularly in electron micrographs of axoplasm freshly extruded in standard salt solution containing 400 fj.g/m\ HMM and negatively stained with \% uranyl acetate, pH 4.4 HMM-decorated filaments can be seen in thin sections of axons treated with glycerol, stand- ard salt solution containing HMM, fixed and embedded. (O SDS polyacrylamide gel electro- phoresis was carried out on: (1) freshly extruded whole axoplasm; (2) axoplasm filaments segregated by uranyl acetate and extracted differentially in 2 M guanidine-HCl and in phenol ; (3) muscle actin and myosin as well as protein standards run in parallel. Proteins in the molecular weight range of actin and myosin are present in the axoplasm. Myosin-like pro- tein has been isolated from squid brain by the method of Pollard ct al. (1974, Anal. Biochcni., 60 : 258) and an antibody against this protein has been prepared. The ordered state of the filamentous network in the axoplasm may be maintained through interactions of its several protein components, such as filamentous protein (70,000 M. W.), actin, tubulin and myosin (215,000 M. W. ). The network could he in a state of equilibrium between assembled and disassembled monomers of these proteins. Probably, a number of factors, such as Ca++ ions, may cause a shift of such an equilibrium state between the mono- meric protein pool and the ordered state of the network. A photoreceptor sensitivity parado.r. B. MINKE, S. HOCHSTEIN AND P. HILLMAN. It is becoming apparent that the invertebrate visual pigment, in contrast to the vertebrate, has a dark-stable (long-lived) metarhodopsin state. Early Receptor Potential (ERP) mea- surements have shown that the visual pigment of the barnacle lateral ocellus has two dark- stable states, rhodopsin and metarhodopsin, with spectral sensitivities having maxima at 532 nm and 495 nm respectively. Because in this case rhodopsin is much more sensitive to red light than metarhodopsin, and less sensitive to blue, it is possible to shift substantial amounts of pigment from rhodopsin to metarhodopsin and vice versa. We have attempted to correlate the amount of pigment in the rhodopsin state, as judged from the ERP, with the sensitivity of the LRP. Intracellular recordings from Balanns rhunicits lateral ocellus photoreceptors were made. The cell was adapted to a saturating amount of blue light, and then dark-adapted for between 5 and 25 minutes. The threshold of the LRP was then tested with weak blue or red lights. The procedure was repeated fol- lowing saturating red light, the threshold to red and blue lights being tested after the same dark times. For each test flash (red or blue) and dark time, no differences in threshold between the blue and the red adaptation were found. Since the ERP measurements under similar adaptation conditions appeared to show changes of a factor of at least 30 in the population of the rhodopsin state this seems paradoxical. It unlikely that the ERP pigment is not that responsible for the LRP, because: (a) several investigators using different tech- niques showed that the LRP action spectrum matches well the action spectrum of rhodopsin measured by the ERP; (b) the induction of a prolonged depolarizing after-potential, which has the same ionic mechanism as the LRP, and its suppression or prevention, all almost certainly arise from the same pigment as the ERP. B. M. and P. H. are grateful for the use of the Rockefeller University laboratory at MBL. Interspersion of repetitive and non-repetitive sequence elements in the DNA of Spisula solidissima. GORDON PAUL MOORE AND WILLIAM GRAIN. It has previously been shown that the genome of a number of deuterostomes contains short [about 300 nucleotides (NT)] regions of repeated sequences interspersed with longer 492 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY (about 1000 NT) segments of unique sequences. Since none of the protostomes have been analyzed in such a fashion we have undertaken a study of Spisitla solidisshnct, the surf clam, and our data suggest that this genome is organized in a similar manner. DNA was ex- tracted, sheared and sized on alkaline sucrose gradients. Fragments 300 NT and 2000 NT were denatured and reassociated and the percent reassociation plotted against the C0T (DNA concentration times annealing time). Computer analysis of the data showed two components, one repetitive (43%) and one unique (38%) for the 300 NT fragments. The 2000 NT fragments reassociated much more rapidly and showed one component. The rate constants for the repetitive component of the 300 NT fragments and for the 2000 NT fragments were 0.117 and 0.083 respectively. These values are equivalent. We conclude that each 2000 NT fragment contains an interspersed unit of repetitive DNA as well as unique sequences. 2000 NT and 300 NT fragments were denatured, renatured to a C0T where only repeti- tive DNA could reanneal, then melted in the spectrophotometer with native DNA. The hy- perchromicity of the fragments was 3.9% and 8.8% respectively while that of native DNA was 26.8%. The fact that hyperchromicity of reassociated fragments decreases with increasing fragment size indicates that there are single strand regions present in reassociated DNA and that the percent of these regions is greater in larger fragments. This is evidence of intersperison of repeat sequences. The length of the interspersed repeat was estimated on the basis of hyperchromicity and percent of the associated molecule resistant to single strand specific SI nuclease. Our estimate for the length of the interspersed repeated unit is 200 ± 100 NT. Development of spontaneous hyper synchrony in flic hippocampal corte.v of the bullfrog, Rana catesbeiana. F. MORRELL AND N. TSURU. Bipolar stimulating electrodes were applied to the hippocampal cortex of one hemisphere in the unanesthetized. partially paralyzed bullfrog. Recording electrodes were placed ad- jacent to the stimulation pair and on the contralateral homotopic cortex. Electrical stimula- tion for 2 sec once per hour at current densities just sufficient to induce a brief after-discharge (AD) on first application led gradually to more prolonged AD involving, first, the side stimulated (1° focus) and, secondly, the opposite hemisphere (2° focus). After several hours it was noted that in the intervals between stimulations spontaneous, high voltage, paroxysmal electroencephalographic (EEG) spikes appeared in the 1° focus. Rarely at first and later quite frequently these spikes gave rise to evoked hypersynchonous potentials in the 2° cortex. Eventually the hypersynchronous discharges appeared independently in each hemisphere, a given spike being unrelated in time either to the electrical stimulus or to the occurrence of any other spike. There was a remarkable constancy of the wave-shape of the spontaneous hypersynchronous potentials which was specific to each animal. The wave-shape coherence presumably reflects a high degree of anatomical order and functional organization in the neuronal network giving rise to it. Once established the prop- erties of the network are largely intrinsic — the same electrical morphology as occurs "spon- taneously" being elicitable by transcommissural synaptic input to the region. Since hyper- synchronous potentials occurred as often as 5 per min and had amplitudes of 0.5-1 mV com- pared with normal background rhythms of 25-50 /uV, they indicate an enormous change in energy production and a major transformation of the functional properties of the tissue. This work was supported in part by USPHS Grant MH 24069-02. Elongation factor 1 of toad-fish liver. JENNIFER B. K. NIELSEN AND A. E. V. HASCHEMEYER A molecular basis is being sought for the changes in rate of polypeptide chain assembly in protein synthesis in vivo associated with temperature acclimation of toadfish. The present work confirms the elevation in liver aminoacyl transferase activity previously found in summer cold-acclimated fish, and establishes that the change occurs in elongation factor 1 (EFj), the protein responsible for binding aminoacyl tRNA at the codon recognition site. Similar high levels of EFi are found in winter toadfish acclimatized to 7° C. Identification and quantitation was made with two assay systems : the binding of 3H-phe tRNA to ribosomes in the PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 493 presence of poly U and GTP ; and the polymerization of 3H-phe tRNA to form polyphenyla- lanine in the presence of poly U, GTP, ribosomes and excess toadfish liver elongation factor 2. Highly purified ribosomes were obtained from rat and toadfish liver. EFi from other eukaryotic sources has been found to occur in vitro in heavy and light forms with different specific activities. To determine whether such forms may account for shifts in activity levels in vivo, the distribution of EFi actvity from warm- and cold-ac- climated fish has been examined. Purification has been carried out by pH 5 precipitation of the post-ribosomal supernatant, ammonium sulfate fractionation, and chromatography on Sepharose 6/3 and hydroxyapatite. Two active forms are eluted from hydroxyapatite at about 0.2 M and 0.3 M potassium phosphate, pH 6.8, at 30-fold and 60-fold purification, re- specively, relative to post-ribosomal supernatant. Similar distributions were obtained in prepara- tions from warm- and cold-acclimated fish. Supported by National Science Foundation grant GB 42752 and Public Health Service grant HD 04670. Association of divalent cations with membranes <>j st/uid giant a.von. JAMES L. OSCHMAN AND BETTY J. WALL. Electron microscopy reveals that opaque deposits are often associated with the cyto- plasmic faces of plasma membranes when calcium is present in the fixative. The phenomenon has been demonstrated in a variety of tissues by several different laboratories. Either glu- taraldehyde or osmium can be used as the primary fixative. Microprobe analysis has shown that the deposits are composed of calcium and phosphorus (Hillman and Llinas, 1974, /. Cell Biol. 61: 146; Oschman, ct <•//<; geniculata. It has been demonstrated at previous MBL General Meetings that the luminescent cells of hydroids and alcyonarians can be visualized by fluorescence microscopy as bright green photo- cytes. Using this technique, photocytes within individual uprights and adjacent stolons of each hydroid were quickly counted. Each piece of counted tissue was transferred to a scintillation vial, containing 1 ml sea water. One ml isosmolar KC1 was injected into the vial which had been placed over a photomultiplier. The integral of the total light emitted was recorded on a polygraph. Additional stimulation by KC1 did not elicit greater light output, therefore we assume our data represent the total emission capacity of photocytes at any given time. Results were calibrated against a known light standard. It was anticipated that a linear relationship should exist between the number of photo- cytes in a given tissue and its total light emitted. Our initial results confirmed this assump- tion. Therefore, using the slopes derived from a statistically significant linear regression of our data, we determined the photocyte output to be about 1-2 X 10s quanta/cell, with a range of extreme values of 0.2 — 6 X 10s quanta/cell, for both Clytia and Obclia. With these tech- niques, we found no statistically significant differences in light output from photocytes vari- ously localized within uprights and stolons. The reported values probably represent the lower limit of light emitted. This lower limit and the variability observed can be explained from the following: a) internal absorption by tissues, perisarc, and epiphytes, b) non-random light emission, c) biological rhythms, d) genetic variations among the sampled populations, e) nutritional and environmental differences, f) age of upright tissues, and g) prior stimulation during the various manipulations. Isolation of spore-forming microorganisms j'roin Little Sippewissett marsh. HOWARD J. SINGER AND E. R. LEADBETTER. The presence and relative numbers of aerobic spore-forming bacteria in different portions of this salt marsh were determined by quantitative dilution and plating on peptone-sea water (Pep-SW) and peptone-distilled water (Pep-DW) agar-solidified media before and after pasteurization (80° C, 10 min) of samples from sandy and mud-bottom ponds, running water, and submerged algal mats. Species of Bacillus were categorized initially as: (a) isolates unable to grow in the absence of Pep-SW, (b) isolates able to grow in Pep-DW, but with faster growth rates in Pep-SW, or (c) isolates able to grow in Pep-DW as well as or better than in Pep-SW. The one type "a" isolate examined was unable to grow in Pep-DW containing up to 3% (w/v) NaCl, although it would grow in Pep-DW containing 20% (v/v) (the lowest con- centration tested) sea-water or MBL Standard Artificial Seawater. No growth was detected on Pep-DW-agar when up to 107 viable cells were plated. Solidified media containing 80% (v/v) SW, 1% (w/v) NH4C1, and 1% (w/v) of either acetate, succinate, sucrose, starch, casamino acids, or glucose were employed in determining the ratio of spore-forming and non-spore-forming bacteria present in the samples and able to 500 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY grow on the given medium. Less than \% of the cultivable aerobic bacteria were spore- formers. Highest viable counts were obtained with casamino acids or succinate, while glu- cose and acetate were poorest in this regard. Of twelve isolates selected for further study, eight formed red, yellow, or orange colonies after prolonged incubation for four or more days at 25° C. The isolates were not readily equatahle with well-recognized Hacilhis species. Studies on protein metabolism and (/ro-^'tli in fish. MICHAEL A. K. SMITH AND A. E. V. HASCHEMEYER. Toadfish, mean weight 300 g, kept in running seawater at ambient summer temperature 22° C and fed killifish ad libitum, show a specific growth rate of about 0.4%/day. Studies were begun to assess protein synthetic activity in various tissues in relation to protein accumu- lation. Incorporation of "C-leucine into liver protein assayed at 6 hrs was similar for both hepatic portal vein and gill artery injection and for tracer or 100 nut leucine concentrations. Incorporation into gill protein per unit tissue weight was comparable to that of liver but that in muscle was only about l/30th as much. Distribution of protein radioactivity as % of dose per total tissue at t = 6 hr was as follows: liver 6.3%; muscle 5.4%; gill 1.8%. Free radioactivity in the tissues at this time was: liver 1.2%. Proportion of total body weight contributed by these tissues is : liver 2.5% ; muscle 60% ; gill 0.7%. Measurements of free leucine concentration in the various tissues by automatic analysis together with total (protein + free) radioactivities yielded estimates of leucine pool specific activities. These values combined with the protein incorporation data yield ratios of protein synthesis in muscle and gill (per 100 g body weight), relative to liver, of about 4.5 and 0.3, respectively. If it is as- sumed that the time course of utilization of 14C-leucine is similar in the three tissues, then true protein synthetic rate in muscle and gill may be calculated from the known synthesis rate in liver of 22°-acclimated toadfish (42 mg protein/100 g body wt/day) obtained in rapid pulse experiments. The results are 190 and 13 mg/100 g/day for muscle and gill respectively. Comparison with growth rate data indicates that the per cent of newly synthesized protein accumulated under these conditions is about 3% for liver and gill and 20% for muscle. Supported by grants from the Ford Foundation and the Lebanese Research Council. The life-cycle of the gasterostome trematodes, Rhipidocotyle transversale Chandler, 1935 and Rhipidocotyle lintoni Hopkins, 1954. HORACE W. STUNKARD. McCrady (1874) described a gasterostome cercaria from the American oyster, Ostrca I'irginica taken at Charleston, South Carolina, as a new species, Bucephalus cuculus. Linton (1905) described metacercariae encysted in Mcnidia menidia and adult worms from the gar, Tylosurus (= Strongylura marinas) taken at Beaufort, North Carolina, as Gasterostominn gracilesccns (Rudolphi, 1819). Tennent (1905, 1906, 1909) found the parasite of the oyster and reported the life-cycle. He postulated that the parasite from the American oyster is identical with the one from the European oyster described by Lacaze-Duthiers (1854), that it is identical with the metacercaria in M. menidia, and that it is the larva of G. gracilesccns. None of these postulates is correct. No life-cycle of an American marine gasterostome has previously been discovered. Chandler ( 1935 ) described encysted gasterostome metacercariae from M. menidia in Texas as Rhipidocotyle transversale n. sp. Hopkins (1954) noted that the metacercariae in M. menidia are identical with the adults in the gar and the species was identified as an R. transversale. Hopkins described three gasterostome species from -S". martinis. One was identified as R. transrcrsalc, one was described as a new species, Rhipidocotyle lintoni, and the third as a new species, Bucephaloides stronc/yliirae. Juveniles of both R. transversale and R. lintoni are common in M. menidia at Woods Hole. Both species may occur in the same fish and the intensity of infection varies from one to many worms ; usually the number is two to four. They are encysted in muscles throughout the body and have been taken from the surface of the brain. Menidia menidia and 5". marinas occur from Cape Cod to Texas and since gasterostomes utilize bivalve mollusks as first inter- mediate hosts, some species with the same distribution and from the same ecological milieu must be involved in the life-cycle. The number of species that meet these requirements is not PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 501 large and the host of the parasite has been discovered. It is Lyonsia hyalina. Developmental stages of the gasterostomes are under study. Energy transfer bctzcccn fluorescent probe molecules in ami across nerve membrane. I. TASAKI, A. WARASHINA AND H. PANT. Resonance transfer of energy was studied by incorporating two kinds of fluorescent probes into the membrane of crab nerve and squid giant axons. A solvent-polarity sensitive probe, p-Cl-2, 6-anilnonaphthalene-sulfonate, was used as the donor of energy; merocyanine-540 was employed as the acceptor. These probes are known to produce fluorescence signals in response to electric stimulation. Physico-chemical properties of these probes were examined in ritro as well as after incorporation into the nerve membrane. The transition moments of these molecules in the membrane were found to be oriented predominantly in the direction normal to the surface. The emission spectrum of the donor (410-570 nm) was shown to overlap the absorption spectrum of the acceptor. The absorption spectrum of the acceptor was com- plex indicating the existence of both strongly fluorescent monomers and weakly fluorescent climers (or aggregates) in the nerve membrane. Measurements of absorption changes as- sociated with action potentials showed that there is an increase in the number of monomer molecules (with an absorption maximum at about 580 nm) accompanied by a simultaneous decrease in climers. The nerve membrane stained the acceptor produced extremely weak fluorescence when optically excited at 365 nm (roughly corresponding to an absorption mini- mum). Addition of the donor to the nerve membrane brought about a distinct enhancement of the acceptor emission. This enhancement was observed when the donor and acceptor molecules were placed on the opposite side as well as on the same side of the membrane. The observed phenomena are interpreted on the basis of Forster's theory of resonance energy transfer. In vivo contractility of amoeba cytoplasm. DOUGLASS L. TAYLOR. The Theological and contractile properties of ectoplasm and endoplasm were studied in single specimens of Chaos carolincnsis. Cytoplasmic viscoelasticity was demonstrated by apply- ing quick stretches to microneedles inserted into various areas of the cell. The ectoplasm in both the tail and the advancing pseudopods were highly viscoelastic without any obvious gradient. In contrast, the endoplasm just behind the fountain zone was more viscoelastic than uroid endoplasm. Physiological buffers were micro-injected into intact cells in order to compare the chemi- cal control induced previously in isolated cytoplasm with the effects on endoplasm and ecto- plasm in I'iro. Small volumes (ca. 0.0001 n\) of a contraction solution containing the threshold calcium ion concentration caused local contraction of the endoplasm and ectoplasm at the site of injection in both the uroid and advancing pseudopods. However, larger volumes of this solution or higher calcium ion concentration caused large contractions of endoplasm and ectoplasm at the site of injection. Micro-injection of this solution into endoplasm just behind the fountain zone caused contraction of the ectoplasm and endoplasm which reversed the direction of cytoplasmic streaming. Micro-injection into the uroid increased the rate of streaming out of the uroid while producing a hyaline cap at the site of injection. Micro-injection of a relaxation solution into the uroids of advancing cells caused the local loss of ectoplasmic structure. Streaming continued out of the uroids until more than half the ectoplasm lost its visible structure. Relaxation solution induced the reversal of cytoplasmic streaming when injected into advancing pseudopods. These effects were completely reversible. These results indicate that here is an increasing gradient of endoplasmic viscoelasticity from the uroid to the tips of advancing pseudopods. In addition, endoplasm and ectoplasm are contractile at all points in the cell so that the site of calcium release or influx will identify the normal site of contraction. In addition, an efficient calcium sequestering mechanism must be present in the cell. 502 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY Induction of polvspennic fertilization of Arbacia eggs by specific protease inhibi- tors Icnpeptin and antipain. WALTER TROLL, HERBERT SCHUEL, AND WALTER L. WILSON. Cortical granules in sea urchin eggs contain a trypsin-like protease which is secreted into the ambient sea water during fertilization. Previous studies with soybean trypsin inhibitor suggested that the protease functions in the discharge of the cortical granules, elevation of the fertilization membrane, and establishment of the block to polyspermy. Leupeptin and antipain are small peptide aldehydes isolated from the fermentation product of actinomycetes by Umezawa and colleagues which are potent competitive inhibitors of proteolytic enzymes. Arbacia eggs fertilized in the presence of leupeptin and antipain (5X 10'" to 10~'' M) do not elevate normal fertilization membranes and became so heavily polyspermic that first cleavage was delayed or prevented entirely. Eggs were also initially inseminated in the presence of 5 X 10~* M antipain and leupeptin with a low concentration of sperm just sufficient to pro- duce simultaneous monospermic fertilization of all the eggs, and were then re-inseminated with a high concentration of sperm at later times. These experiments showed that poly- spermy resulted from a process of re-fertilization extending for a period of about 10 to 15 min post initial fertilization. Normal cleavage and development was obtained when eggs were placed in 10"" M leupeptin and antipain after the fertilization membrane had elevated. These data show that the cortical granule protease is inhibited by leupeptin and antipain, and con- firm the critical secretory function of this protease in establishing the block to polyspermy in sea urchin eggs. Supported by grants from NIH, the American Cancer Society, and the Population Council. Distribution of adenosine-5'-phosphosulfate (APS) and adcnosinc-3'-phosphatc-5'- phosphosnlfatc- (PAPS} sulfotransferases in assiinilatory sulfate reducers. MONICA LIK-SHING TSANG AND JEROME A. Sen IFF. A distinction between dissimilatory sulfate reducers which utilize sulfate as an electron acceptor in respiration and assiinilatory sulfate reducers which use sulfate to satisfy their requirements for reduced sulfur compounds for .growth is based on the type of activated sul- fate compound used as the immediate donor for reduction. Dissimilatory sulfate reducers have been shown to utilize APS and it was suggested that assiinilatory sulfate reducers use PAPS. With our findings that among assimilatory sulfate reducers, Clilorella pyrenoidosa uses APS as the substrate for reduction via APS-sulfotransferase while Escherichia coll uses PAPS via a PAPS-sulfotransferase, this correlation does not appear to hold, and we have undertaken a phylogenetic survey of APS and PAPS utilization by reducing systems to determine the significance of their distribution. We find that non-photosynthetic organisms such as E. coli among the procaryotes and yeast among the eucaryotes contain PAPS-sul- fotransferases while organisms which are photosynthetic such as the red algae, C hondrus crispus and Porphyridium aerugineum, the brown alga, FUCKS rcsicitlnsiis, the diatoms Cyclo- tclla and Chactoccrits psendocrinitis, the dinoflagellates Peridinium troclioidiuin and Gyinno- diniitni sp., the cryptophyte Chroomonas sp., the prasinophyte Platyinonas sp., and the blue- green algae Oscillatoria wononichmii and Synechococcus sp. as well as spinach chloroplasts all utilize APS rather than PAPS. This suggests that APS utilization for assimilatory reduc- tion is characteristic of chloroplasts and chloroplast-containing organisms while PAPS utiliza- tion for assimilatory reduction is characteristic of non-photosynthetic organisms. It is possible that the primitive case, a relict of the anaerobic phase of life, is represented by the dissimila- tory sulfate reducers. Assimilatory sulfate reduction may have evolved from this point, when oxygen entered the atmosphere, in two paths : the aerobic bacteria utilizing PAPS leading to other non-photosynthetic organisms and the blue-green algae utilizing APS leading to chloro- plasts and chloroplast-containing organisms. We thank Dr. Robert Guillard and Dr. Joe Rainus for supplying many of the organisms used. PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 503 Skin photoreceptors in the leopard frog. GEORGE WALD AND STEPHEN RAYPORT. In addition to old reports of behavioral responses (orientations, motions) excited by light striking the skin in frogs and tadpoles ( G. H. Parker, Am. J. Physio!., 10, 28, 1903; V. Obreshkove, /. Exp. Zoo!., 34, 235, 1921), there have been recent reports of both fast photo- voltages and slow electrophysiological responses from isolated pieces of frog skin (Becker and Cone, Science, 154, 1051, 1966; Becker and Goldsmith, Nature, 220, 1236, 1968). We, like the latter workers, have clamped pieces of skin in a modified Ussing cell, bathed on both sides with frog Ringer, with light-shielded Ag-AgCl electrodes dipping into the Ringer. We have been concerned only with the slow response (electrodermogram, EDG). It has a latency at 23° C of about 0.75 sec takes another second to peak, and about 2 sec more to return to the base-line. The skin has a resting potential in the dark of up to 100 mv, outer surface nega- tive; and this is increased (hyperpolarized) by light by as much as 0.8 mv. Over our work- ing range the amplitude of the EDG is proportional to the light intensity. The spectral sensitivity of photoexcitation ranges between 320 and 470 nm, and is maximal at about 395 nm. Repeated flashes in the near ultraviolet cause a large decrease in response (adaptation) that shows no recovery within 70 min in the dark at 23° C. Exposures to longer wavelengths, however, cause a strong photorecovery. The photorecovery spectrum overlaps with the photo- stimulation spectrum in the violet, but extends to about 560 nm, peaking near 500 nm. These observations indicate a photoreversible system involving two stable states, photostimulatory peaking at about 395 nm, and photorecovery peaking near 500 nm. Supported by funds from NIH, NSF and Harvard University. A general method for the introduction oj enzymes, l>\ means oj liposomes, into lysosomes of deficient cells. GERALD WEISSMANN, DAVID BLOOMGARDEN, ROBERTA KAPLAN, CHARLES COHEN, SYLVIA HOFFSTEIN, TUCKER COLLINS, AVRUM GOTTLIEB AND DAVID NAGLE. The phagocytes of the smooth dogfish (Mustelits cunis) contain no endogenous peroxidase within their lysosomes and therefore can be used as models for cells genetically deficient in lysosomal enzymes such as myeloperoxidase. Based upon our previous work showing ( 1 ) the possibility of trapping enzymes in the aqueous interspaces between the lipid layers of liposomes, and (2) the possibility of coating liposomes with immunoglobulins, we have man- aged to obtain incorporation of over 50% of exogenous horse-radish peroxidase by dogfish phagocytes, provided the enzyme is exhibited to cells after incorporation into liposomes (pre- pared with ovolecithin, dicetylphosphate and cholesterol in molar ratios of 7:2:1) and coating these with heat-aggregated IgM (62°, 10 min). Trapping of peroxidase by liposomes was established by chromatography of free and liposome-bound enzyme and markers of the aqueous space (glucose, chromate anion ) on Sephadex G-200, Sepharose 2B and Sepharose 4B : 0.67 (mean) micromoles of markers and 313 nanograms of peroxidase were trapped per micromole of phospholipid (n = 9). Enzyme and marker trapping were not electrostatic, varied with the molar ratio of charged component (dicetylphosphate), and was reversed after disruption of liposomes by Triton X-100 which showed these to be 98-100% "latent" within liposomes. Uptake (n = 12) of dogfish aggregated IgM-coated liposomes at 30° (9.7 ng/10" cells/hr) exceeded that of free enzyme (less than 1 ng), of uncoated liposomes (4.1 ng), of native IgM-coated liposomes (4.8 ng), of aggregated IgM-coated liposomes at 4° (5.7 ng), and of these treated with cytochalasin B to block phagocytosis (5.5 ng) or cytochalasin B- treated cells held at 4° (2.0 ng). Once taken up by phagocytes, over 50% of enzyme was latent within cells (as determined by lysis in distilled water) and sequestered exclusively within lysosomes (localized by ultrastructural histochemistry). This is a general method for the provision of exogenous enzymes to phagocytic cells genetically deficient in lysosomal hydrolases. Evidence for biased bipolar polymerization of actin filaments. D. T. WOODRUM, S. RICH AND T. D. POLLARD. Actin filaments polymerize in a biased bipolar manner judging from experiments where actin monomers are polymerized onto the ends of preformed actin filaments decorated with 504 PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY heavy meromyosin (HMM). Purified rabbit skeletal muscle actin was depolymerized by dialysis against 2 HIM tris, pH 7.8, 0.1 mM CaCl2, 0.2 mM ATP to yield monomeric G-ATP- actin. This monomeric actin was then passed through a Sephadex G-25 column equilibrated with 2 HIM tris, pH 7.8, 0.1 mM Cad-, 1 mM DTT to remove unbound ATP from the solution of G-ATP-actin. Samples of actin were polymerized by the addition of KC1 to 0.1 M and then decorated with a slightly substoichiometric concentration of HMM for use as nuclei in the following experiments. Addition of G-ATP-actin to the nuclei resulted in the growth of bare segments of actin from the ends of the decorated nuclei. The extent and the polarity of this nucleated poly- merization was assessed by electron microscopy of negatively stained specimens. When low concentrations of monomers (0.1 mg/ml or less) were added to nuclei, newly polymerized actin was found at the "barbed" end of 85 9£ of the nuclei, while none of the nuclei had bare additions at the pointed end. At higher concentrations of monomer (0.5 mg/ml), bare addi- tions were observed at both ends of the nuclei, although the additions were 4-8 times longer at the barbed end compared with the pointed end. Variation of the concentration of nuclei (0.05-0.5 mg/ml), the temperature (0-25° C) and the presence or absence of 1 ITIM Mg++ had no influence on the direction of nucleated actin polymerization. Hoii1 cell growth is inhibited b\ near IT light pJwtoproducts of tryptophan. S. ZIGMAN, T. YULO, G. GRIESS, B. ANTONELLIS, AND R. RUSTAD. Exposure of neutral pH, aqueous solutions of tryptophan (TRP) to near UV light (320 to 400 nm), termed NUVTRP, produces photoproducts with growth inhibitory properties. We report here that the growth of mouse embryonic fibroblasts (MEF), dogfish (Mitstclus canis) eye lens epithelial cells and sea urchin (Arhnchi punctulata) fertilized eggs is also strongly inhibited by dilute solutions of TRP (5 X 10"4 M) which were previously exposed to 3 mW/cnr of near UV light from several to 24 hrs. For MEF cells in Eagle's MEM, growth and synthesis of protein, RNA, and DNA were markedly inhibited (70 to 90%) by long term (3-day) exposure to NUVTRP. One-hr exposures to NUVTRP were also effective in appreciably inhibiting precursor incorporation into protein, RNA, and DNA. Gel filtration (Sephadex G 10) of whole NUVTRP yielded a yellow brown fraction (400 dalton Mol. Wt.) which had the greatest antigrowth activity. This was not N-formylkynurenine, a known photoproduct of TRP. The incorporation of amino acids, uridine and thymidine into eye lens epithelial cell pro- tein, RNA, and DNA was also markedly inhibited when NUVTRP was added at 5 X 10~* M to the elasmobranch Ringer's medium used for the incubations. For sea urchin eggs, Millipore filtered sea water plus TRP at the same level prevented structuralization of the mitotic apparatus prior to the first cleavage, approximately doubled division time, and led to abnormal multicelled stages. During the first hr postfertilization, protein, RNA, and DNA incorporated 34C and "H-amino acids, sH-uridine, and sH-thymidine at less than half the rate as the TRP controls. Exposure of sea water with the same level of TRP to window-glass-filtered sunlight for 8 hrs produced similar growth inhibition. The photoproduct fraction most active against the growth of MEF cells was also found most active in the sea urchin system. NUVTRP inhibits the growth of these cells by interfering with macromolecular synt|ieste. Abnormal mitotic apparatus formation may relate to deficient macromolecule supply. Inter- ference with polymerases and energy supplying enzymes may also be involved. Supported by PHS grant EY-00459; Monroe Cty. Cancer and Leukemia Society; ONR- CNA Univ. of Rochester Contract ; U.S.A. E.C. Studies on the establishment of a cytochalasin B sensitive rapid polyspermy block following fertilization of Spistila solidissima eggs. CAROL ANN ZIOMEK AND DAVID EPEL. The block to polyspermy has been best studied in the sea urchin eggs which appear to have two polyspermy blocks, a fast partial block and a slower complete block resulting from PAPERS PRESENTED AT MARINE BIOLOGICAL LABORATORY 505 the cortical reaction. The eggs of Spisnla snlidissiina also have a polyspermy block but no cortical reaction. Thus, a study of this organism may provide insight into the general blocks not involving the cortical granules. The timing of the block in Spisiila eggs was determined using the polyspermy assay of Rothschild and Swarm. One culture was lightly inseminated resulting in 100% monospermy. A second culture was heavily inseminated leading to 100% polyspermy. A third series of cultures were lightly inseminated and followed at various times by a heavy sperm dose. In Spisitla eggs, the half-time for completion of the block was 2.5 sec. The vitelline layer was removed from unfertilized Spisiila eggs by a 2 min incubation with 1 M /?-alanine containing 5 mM EDTA, pH 6.5, followed by washing two times with filtered seaweater. The timing and extent of the block to polyspermy was identical to un- treated eggs. Therefore it appears that the block to polyspermy involves changes in the plasma membrane. Spisiila eggs were then pretreated for 10 min with various concentrations of cytochalasin B (0.5, 1.0, and 5.0 jug/ml), washed and subjected to the polyspermy assay. In all cases, the eggs became polyspermic upon addition of heavy sperm doses. If the cytochalasin B was added after the block was established and then heavy sperm doses added, no polyspermy was detected. Therefore it appears that the establishment of the block is sensitive to cytochalasin B but insensitive once the block has been established. This study was performed in the Fertilization and Gamete Physiology Research Training Program at the Marine Biological Laboratory (NIH Grant 3-T01-HD00026-12SD. Continued from Cover Two 4. Literature Cited. The list of references should be headed LITERATURE CITED, should conform in punctuation and arrangement to the style of recent issues of THE BIOLOGICAL BULLETIN, and must be typed double-spaced on separate pages. Note that citations should include complete titles and inclusive pagination. Journal abbreviations 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 below. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS of those abstracted (most recent issue: November, 1972). Foreign authors, and others who are accustomed to use THE WORLD LIST OF SCIENTIFIC PERIODICALS, may find a booklet published by the Biological Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K. at £0.65 or $1.75) useful, since it sets out the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these^differ. CHEMICAL AB- STRACTS 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 components1 must be followed by a period, whole word components must not (not strictly as USASI usage, i.e. J. Cancer Res.) D. Space between all components (e.g. J. Cell. Comp. Physiol. not J.Cell.Comp.Physiol.) E. We strongly recommend that more unusual words in journal titles be spelled out in full, rather than employing lengthy, peculiar "abbreviations" or new abbreviations invented by the author. For example, use Rit Visindafjelags Islendinga without abbreviation. Even in more familiar languages, Z. Vererbungslehre is preferred to Z. VererbLehre (WORLD LIST) or Z. Verer- bungsl. (USASI). Accurate and complete communication of the reference is more important than minor savings in printing costs. F. All single word journal titles in full (e.g. Veliger, Ecology, Brain). G. The order of abbreviated components should be the same as the word order of the com- plete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. Spell out London, Tokyo, Paris, Edinburgh, Lisbon, etc. where part of journal title. I. Series letters etc. immediately before volume number. J. A few well-known international journals in their preferred forms rather than WORLD LIST or USASI usage (e.g. Nature, Science, Evolution NOT Nature, Lond.; Science, N. Y.; Evolution, Lancaster, Pa.) K. The correct abbreviation for THE BIOLOGICAL BULLETIN is Biol. Bull. 5. Figures. The dimensions of the printed page, 5 by 7f inches, should be kept in mind in preparing figures for publication. Illustrations should be large enough so that all details will be clear after appropriate reduction. Explanatory matter should be included separately in legends as far as possible, although the axes should always be numbered and identified on the illustration itself. Figures should be prepared for reproduction as either line-cuts or halftones; no other methods will be used. Figures to be reproduced as line-cuts should be 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, and both designating numbers or letters and scale-bars should be affixed directly on the figures. We recommend that halftones submitted to us be mounted prints made at about 1| times the linear dimensions of the final printing desired (the actual best reductions are achieved from copy in the range from 1| to 2 times the linear dimen- sions). As regards line-blocks, originals can be designed for even greater reductions but are best in the range 1| to 3 times. All figures should be numbered in consecutive order, with no distinc- tion between text and plate-figures. The author's name should appear on the reverse side of all figures, and the inked originals for line-blocks must be submitted for block-making. 6. Mailing. Manuscripts should be packed flat. All illustrations larger than 8£ by 1 1 inches must be accompanied by photographic reproductions or tracings that may be folded to page size. Reprints. Reprints may be obtained at cost; approximate prices will be furnished by the Managing Editor upon request. CONTENTS ALDRICH, JOHN CARLSON Allometric studies on energy relationships in the spider crab Libinia emarginata (Leach) 257 BURKETT, BARBARA N. AND HOWARD A. SCHNEIDERMAN Roles of oxygen and carbon dioxide in the control of spiracular function in Cecropia pupae 274 BURKETT, BARBARA N. AND HOWARD A. SCHNEIDERMAN Discontinuous respiration in insects at low temperatures : intra- tracheal pressure changes and spiracular valve behavior 294 CHENG, THOMAS C. AND GARY E. RODRICK Identification and characterization of lysozyme from the hemolymph of the soft-shelled clam, My a arenaria \ r .;,- U ; .j.v. 311 CULLINEY, JOHN L. Larval development of the giant scallop Placopecten magellanicus (Gmelin) WJjj* V^IpWi " ' ' ' ' ' ' '^ 321 FELL, PAUL E. Diapause in the gemmules of the marine sponge, Haliclona loosanoffi, with a note on the gemmules of Haliclona oculata 333 GIBSON, RAY Histochemical observations on the localization of some enzymes associated with digestion in four species of Brazilian nemerteans . . . 352 HILL,- ROBERT B. AND JOSEPH W. SANGER Anatomy of the innervation and neuromuscular junctions of the radular protractor muscle of the whelk, Busycon canaliculatum (L.) 369 LOWRY, LLOYD F., ALFRED J. MCELROY AND JOHN S. PEARSE The distribution of six species of gastropod molluscs in a California kelp forest Aj . . . T^C . . /. 386 MORIN, JAMES G. AND GEORGE T. REYNOLDS The cellular origin of bioluminescence in the colonial hydroid Obelia 397 MUKAI, HIDEO AND HlROSHI WATANABE On the occurrence of colony specificity in some compound ascidians 411 RICHARDSON, NANCY E. AND JAMES D. MCCLEAVE Locomotor activity rhythms of juvenile Atlantic salmon (Salmo salar) in various light conditions : . A . . . .-r.'^. ^1 422 SPIELMAN, ANDREW AND JOANN WONG Dietary factors stimulating oogenesis in Aedes aegypti 433 VOGEL, STEVEN Current-induced flow through the sponge, Halichondria 443 YINGST, DOUGLAS The vertical distribution and reproductive biology of Pelogobia longicirrata (Annelida) in the central Arctic Ocean 457 Abstracts of papers presented at the Marine Biological Laboratory 466 Volume 147 Number 3 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board Jomf M. ANDERSON, Cornell University F. H. RUDDLE, Yale University JOHN O. CORLISS, University of Maryland BERTA SCHARRER, Albert Einstein DONALD P. COSTELLO, University of HOWARD A. SCHNEIDERMAN, University of North Carolina California, Irvine PHIUP B. DUNHAM, Syracuse University MELVIN SPIEGEL, Dartmouth College CATHERINE HENLEY, University of GROVER C. STEPHENS, University of North Carolina California, Irvine GEORGE O. MACKIE, University of Victoria EDWARD O. WILSON, Harvard University W. D. RUSSELL-HUNTER, Syracuse University Managing Editor DECEMBER, 1974 Printed and Issued by LANCASTER PRESS, Inc. PRINCE £ LEMON STS. LANCASTER, PA. THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince ancf Lemon Streets, Lancaster, Pennsylvania. Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, Woods Hole, Massachusetts. Agent for Great Britain: Wheldon and Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2. Single numbers, $7.00. Subscription per volume (three issues), $18.00, (this is $36.00 per year for six issues). Communications relative to manuscripts should be sent to Dr. W. D. Russell-Hunter, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between May 23 and September 1, and to Dr. W. D. Russell-Hunter, P.O. Box 103, University Station, Syracuse, New York 13210, during the remainder of the year. Copyright © 1975, by the Marine Biological Laboratory Second-class-postage paid at Lancaster, Pa. INSTRUCTIONS TO AUTHORS THE BIOLOGICAL BULLETIN accepts original research reports of intermediate length on a variety of subjects of biological interest. In general, these papers are either of particular interest to workers at the Marine Biological Laboratory, or of outstanding general significance to a large number of biologists throughout the world. Normally, review papers (except those written at the specific invitation of the Editorial Board), very short papers (less than five printed pages), preliminary notes, and papers which describe only a new technique or method without presenting substantial quantities of data resulting from the use of the new method cannot be accepted for publication. A paper will usually appear within four months of the date of its acceptance. The Editorial Board requests that manuscripts conform to the requirements set below: those manuscripts which do not conform will be returned to authors for correction before review by the Board. 1. Manuscripts. Manuscripts must be typed in double spacing (including figure legends, foot-notes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 8} by 11 inches. They should be carefully proof-read before being submitted and all typographical errors corrected legibly in black ink. Pages should be numbered. A left-hand margin of at least li inches should be allowed. 2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and placed in correct sequence in the text. Because of the high cost of setting such material in type, authors are earnestly requested to limit tabular material as much as possible. Similarly, foot- notes to tables should be avoided wherever possible. If they are essential, they should be indi- cated by asterisks, daggers, etc., rather than by numbers. Foct-notes are not normally permitted in the body of the text. Such material should be incorporated into the text where appropriate. Explanations of figures should be typed double-spaced and placed on separate sheets at the end of the paper. 3. A condensed title or running head of no more than 35 letters and spaces should be included. Continued on Cover Three Vol. 147, No. 3 December, 1974 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY ULTRASTRUCTURE OF A CEPHALOPOD PHOTOPHORE. I. STRUCTURE OF THE PHOTOGENIC TISSUE Reference: Biol. Bull, 147: 507-521. (December, 1974) JOHN M. ARNOLD AND RICHARD E. YOUNG Pacific Biomcdical Research Center, University of Haivaii, Honolulu, Hawaii; Marine Biological Laboratory, Woods Hole, Massachusetts; and Department of Oceanography, University of Haivaii, Honolulu, Hawaii Of the many animal taxa which are capable of light production, the Cephalopoda present some of the most varied and elaborate photophores known. Photogenic organs are found within the mantle (Clarke 1965) in or by internal organs (e.g., ink sac; Boletzky, 1970) or on the siphon (Hoyle, 1904), on the skin of the mantle, head and/or tentacles, and on the surface of the eye (Berry, 1920a, 1920b). There are many different types of photogenic organs and several dif- O.lmm FIGURE 1. Light micrograph of a whole photophogenic tissue (pg) composed of an axial cone and spheroidal knob enclosed in a posterior cup (pc) made of regular iridophores. The anterior cap (ac) grades into a conical "plug" of iridophores in the center of the axial cone. Distal to the anterior cap is the clear lens (In) composed of iridophores with many iridosomes in each cell. Surrounding the axial cone is a conical layer of irregular iridophores (irr). The posterior cup and proximal side of the whole organ is covered with a layer of brownish pig- ment (pig). Muscle bands (mb) and blood vessels (bv) occasionally penetrate the posterior cup or irregular iridophores. 507 Copyright © 1974, by the Marine Biological Laboratory Library of Congress Card No. A38-518 508 J. M. ARNOLD AND R. E. YOUNG SQUID PHOTOPHORE ULTRASTRUCTURE 509 ferent types are often present simultaneously in the same individual and even on the same organ. The function of the photogenic organs is largely unknown, but it has been suggested that they are important in countershading (Young, 1972, 1973; Clarke, 1963), predation (Nicol, 1960), and possibly sexual interaction (Clarke, 1963, Young, 1974) or schooling behavior. Both bacterial and auto- photogenic organs have been reported in cephalopods (Nicol, 1960). The auto- photogenic photophore presents an interesting opportunity to study the production of light by a tissue whose function is the opposite of the retina but which still has many basic similarities to photoreceptive organs. By a careful fine structural analysis of these photogenic organs, insight might be gained into the basic struc- tural mechanism of interconversion of chemical and electrical energy into light which, in turn, might be compared in interesting ways to photoreception. The purpose of these two papers is to report the results of ultrastructual studies on one common type of photogenic organ in the squid Pterygioteuthis microlampas, a midwater, Pacific species which probably only encounters low levels of light, and to speculate on the possible function of these photophores. Although Hoyle (1904) has described the histological structure of a photophore of a closely related squid (Pterygioteuthis giardi), for purposes of convenience to the reader and because there are some slight differences in our observations, the following brief description of the whole photophore is provided. There are several photophores located on the ventral surface of each eye, and these are of at least three types. Eleven of these photophores have similar structure and have been the subject of this study. These photophores vary in size, depending on the in- dividual squid, but typical ones measure about 0.4 mm X 0.3 mm, are lentoid in shape, and are roughly radially symmetrical. Hoyle (1904) described a connective tissue cap which was reddish-brown ; a "posterior cup" of oval or circular fibrous "scales" ( = iridophores) in concentric layers; an "anterior cap" of "scales"; an "inner funnel" of rather coarse fibers (^irregular iridophores) ; and a central mass filled with almost structureless parenchymous tissue which is composed of a "spheroidal knob" and an "axial cone" (Fig. 1). In this paper this tissue will be referred to as the photogenic tissue and the presumed light producing cell will be called the photocyte. In addition there is also a clear layer of modified iridophores (=lens) distal to the anterior cap, which Hoyle did not include in his description. Nerves and blood vessels occasionally penetrate the sides or the base of the bulb region of the posterior capsule. Bands of muscle traverse the "inner funnel" re- gion and attach to those iridophores which make a conical proximal "plug" as a central continuation of the anterior capsule. Apparently, each of the three types of iridophores making up the posterior and anterior capsules, the "inner funnel" and clear "lens" have different functions, and this paper will be devoted to a description of their ultrastructure and speculation as to their functions. FIGURE 2. Photogenic tissue of the spheroidal knob and part of the axial cone. A blood vessel (bv) enters through the posterior cup. The photocytes (ph b) branch among the homogeneous packing cells (he). Branches of mitochondrial cells (me) are also evident. 510 J. M. ARNOLD AND R. E. YOUNG 10M SQUID PHOTOPHORE ULTRASTRUCTURE 511 MATERIALS AND METHODS The animals used in this study were taken at night by a closing trawl off the island of Oahu, Hawaiian archipelago. Once on shipboard they were fixed in a mixture of 6% glutaraldehyde buffered to pH 7.5 with 0.1 M collidine buffer and adjusted to approximately 1200 milliosmoles with sucrose as modified from Bell, Barnes and Anderson (1969). Fixation began at ambient temperature and then the vials containing the whole animal were chilled on ice and kept at 4° C until the ship returned the next morning. The photophores were then dissected off the surface of the eye, washed in cold collidine buffer for 1 hour and then post-fixed in 1% OsO4 in veronal-acetate buffer at pH 7.6 for one-half hour. The photophores were then dehydrated through an ethanol series and then embedded in Luft Epon for thin sectioning. These sections were stained with either uranyl acetate in methanol followed by lead citrate or in lead citrate alone. This fixation procedure seemed to be adequate despite the difficulties caused by working on shipboard, and the time delays involved. However, in some areas there are membrane whorls that are probably artifact. Despite slight variations in fixation procedures these whorls occasionally remained randomly associated with membranous structures in the cells. Because of the difficulty in obtained animals (one or two per night is an average catch), the expense of ship time, and the otherwise satisfactory fixation, the results are reported with no further apology. OBSERVATIONS Figure 2 shows a low power micrograph of the "spherical knob" and most of the "axial cone" region of the photogenic tissue. The photogenic tissues are surrounded by stalks of iridophores which are occasionally penetrated by blood vessels and nerves. These iridophores will be discussed in a separate paper. In addition to elements of circulatory system and the nervous tissue, these are four cell types: the photocytes; their associated sheath cells; the packing cells which occupy most of the volume of the photogenic tissue; and the mitochondrial cells. Each of these cell types will be discussed in turn. Although it has not been pos- sible to examine each of these cells in physiologically active state, by elimination and analogy it has been possible to assign these four cell types functions on the basis of their morphology. Photocytes The photocytes are branching, ramifying cells which are interspersed with no apparent pattern between the other cells of the photogenic tissues. They are characterized by an extensive body of microvillous protrusions which occupy most of the volume of the cell and are arranged about a central lumen filled with dense FIGURE 3. Mature photocyte (ph) in the nuclear region. The photocyte is enclosed in a sheath (sc) which separates it from the packing cells (he). The microvilli (mv) are quite evident. FIGURE 4. Higher magnification of the microvillous region of one branch of a photophore. The microvilli project into a central blood filled lumen (bd). The sheath can be seen to be of homogeneous density and lined by membranes of both sides. 512 J. M. ARNOLD AND R. E. YOUNG SQUID PHOTOPHORE ULTRASTRUCTURE 513 material (Fig-. 3). Surrounding each photocyte is a dense sheath of acellular homogeneous material which separates the cell from either the surrounding pack- ing cells or engulfing sheath cell (see below). In sections the nucleus of the photocyte appears more or less homogeneous when compared, for example, with the nuclei of the packing cells or the sheath cells. The microvillous bodies comprise a discrete part of each photocyte and when apparently mature, are easily distinguished by their electron density. The in- dividual microvilli are cylindrical and have a rounded end. They average 0.1 ju, in diameter and vary in length from 0.4/A to 0.6/x (Fig. 4). The microvilli project into a central lumen which apparently is filled with blood (Fig. 5). Cephalopod blood frequently has a characteristic paracrystalline appearance when fixed for electron microscopy (Barber and Graziadei, 1965) (Fig. 6). The hemocyanin molecules tend to form chains, each unit of which measures approximately 24 nm in diameter. Cut at various angles other characteristic spacing patterns are evi- dent (Fig. 5) so blood can often be distinguished from electron dense substances. However, the blood does not always fix in this paracrystalline pattern but in these instances other characteristics (basement membrane, homogeneous density) can be used to distinguish blood from other materials of similar density. The micro- villous body can often be seen to be continuous with blood vessels and in a few instances it appears that the microvilli are bathed by blood directly from the capil- laries (Fig. 6). The sheath which surrounds each photocyte often appears con- tinuous with the basement membrane of the blood vessel. In the region of the axial cone photocytes which appear to be in stages of development can frequently be found. These immature photocytes are charac- terized by complete enclosure in a sheath cell, few and less dense microvilli in poorly organized discontinuous microvillous bodies, and by areas of somewhat less dense cytoplasm between the microvilli and the sheath (Fig. 7). As the micro- villous body develops, this cytoplasmic area is mostly replaced by microvilli and what cytoplasm remains becomes increasingly dense (Fig. 8). Concommitant with this increase in density, the surrounding sheath cell apparently is retracted leaving the branches of the photocyte surrounded by a dense homogeneous sheath of variable thickness which is continuous with the basement membrane of the circulatory system. The microvilli apparently originate in conjunction with membranous components of the cytoplasm (possibly endoplasmic reticulum) and are frequently associated with infolding of the photocyte surface (Fig. 9). In some instances blood can be found contained within membranous networks within the cytoplasm which appear to be giving rise to microvilli. The origin of the intercellular mem- branous complexes containing this blood is uncertain but frequently endoplasmic reticulum and Golgi vesicles are seen in close proximity (Fig. 10). FIGURE 5. Central region of a photocyte showing blood (bd) in the lumen. The hemocyanin molecules align to form paracrystalline arrays, some of which have a chain-like appearance. Sectioned at different angles the pattern is still characteristic enough to make identification of blood possible. FIGURE 6. Connection between a photocyte and blood vessel. The paracrystalline pattern is evident (bd) in the vessel and the lumen of the microvillous region. The basement mem- brane (bm) is continuous with the sheath (sc) of the photocyte. The endothelial cell and pericyte are also evident. 514 J. M. ARNOLD AND R. E. YOUNG SQUID PHOTOPHORE ULTRASTRUCTURE 515 Sheath cells The sheath cells characteristically surround the whole developing photocytes encasing them in a layer of cytoplasm of varying thickness (Fig. 7). The nucleus of the sheath cell invariably occurs in proximity of the photocyte nucleus. The cytoplasm is considerably less dense than that of the photocyte and contains few scattered organelles. Mitochondria and membranous vesicles are occasionally encountered but mainly the cytoplasm is homogeneous with small aggregations of granular "background" material. Where the sheath cell contacts a blood vessel, vesicles suggestive of pinocytosis are apparent. The nuclei of the sheath cells are also less dense than those of the photocyte and appear to have fewer regional densities (Fig. 7). The sheath itself (Figs. 4, 8, 9) is homogeneous, amorphous, and electron dense. It appears to be entirely extracellular because it occurs outside the plasma membrane of both the sheath cell and the photocyte. It can be seen to be con- tinuous with the basement membrane of the blood vessels although it is usually of less uniform thickness. On mature photocytes, it is quite irregular in thickness and shape (Fig. 8). The sheath cells are quite similar in appearance to the peri- cytes of the blood vessels (Barber and Graziadei, 1965) and in some instances it is hard to separate them on morphological criteria other than position (Fig. 6). As the photocyte matures the association of the sheath cell and photocyte is ended apparently by retraction of the sheath cell so that the photocyte with its surround- ing sheath comes into direct contact with the surrounding packing cells (Fig. 8). The eventual fate of the sheath cell could not be determined from our micrographs. Packing cells The packing cells are characterized by being filled with a homogeneous granular matrix which occupies the position of the cytoplasm in a more typical cell (Fig. 11). The membranous components of the cell are in direct contact with this matrix and there is no evidence that the matrix is contained in a special vacuole. Oc- casionally mitochondria are encountered but they appear to be randomly distributed and the outer mitochondrial envelope is in direct contact with the matrix (Fig. 6). Golgi material and a few membranes suggestive of smooth endoplasmic reticulum are present infrequently (Fig. 6). The nuclei have large regions of dense ma- terial which primarily occupies the periphery of the nucleus. The packing cells tend to be globular in shape but assume the shape of the more highly structured elements around them. FIGURE 7. Developing photocyte within a sheath cell (sc). Note that the photocyte is completely enclosed within the sheath cell. FIGURE 8. Retraction of the sheath as the photocyte matures. The microvilli (mv) completely fill the mature portion of this photocyte branch and the sheath cell no longer surrounds this area. Where the microvilli are less dense and not as well differentiated the sheath cell enclosed the photocyte branch. 516 J. M. ARNOLD AND R. E. YOUNG •f£rS- ^- ,* ~m - ' 1 ' , ,» * ' . kMP .. * £> -i< «; • •• « • - SQUID PHOTOPHORE ULTRASTRUCTURE 517 Mitochondrial cells In contrast to the packing cells, the fourth cell type in the photogenic tissue, the mitochondrial cells, tend to have their cytoplasm occupied mainly by mito- chondria (Fig. 11). These cells tend to be concentrated in the axial cone and in rare cases form a more or less continuous layer. However, because they branch and ramify throughout the photogenic tissue they are present in the spherical knob and are seen in contact with all of the other cell types including the iridophores and no preferential association is obvious. The mitochondrial themselves have the "somewhat empty" appearance typical of cephalopod mitochondria but typically occupy better than 50% of the total cell volume. In contrast to the nuclear con- densation seen in the packing cells, the nuclei of the mitochondrial cells seem homogeneous although the cisternae of the nuclear envelope are infrequently ex- panded (Fig. 11). Vascular and neural elements Vascular elements are frequently encountered entering the photogenic tissue. They are easily identified by the characteristic basement membrane, endothelial cells, and pericyte (Barber and Graziadei, 1965). The hemocyanin frequently shows a paracrystalline pattern which also aids in identification (see above). Where vessels pass through the posterior cup, the iridophores are displaced or modified in shape to accommodate them. Inside the photogenic area the endothelial cells and pericytes are frequently reduced or even lacking (Fig. 9), thus allowing the blood to come into intimate contact with the cells of the photogenic tissue. The lumen of the photocytes is frequently seen filled with blood (Figs. 5 and 6) which is in direct contact with the microvilli and occasionally blood can be found within intracellular membranes in developing photocytes. Neural elements are represented primarily by single nerve processes which occur among the photogenic cells. These nerve fibers apparently form synapses with the photocytes and in favorable sections gaps in the sheath surrounding the photocytes can be found which suggest the possible transport of synaptic vesicles (Fig. 12). We have not encountered any synapses with other cell types of the photogenic tissue. DISCUSSION Although the inability to maintain living material in the laboratory imposes certain limitations on the interpretation of the observations presented here, it is possible to speculate on the function of the various cell types presented here on purely morphological grounds. The photocytes seem to be the only possible FIGURE 9. Developing microvilli and acellular region of blood vessel. The developing microvilli arise inside the photocyte but are in contact with the surface. A small acellular blood vessel is in contact with the sheath cell and developing photocyte. The sheath (sc) is continuous with the basement membrane (bm). Blood is evident inside the cell in the future lumen. FIGURE 10. Developing microvilli fmv). Note the membrane continuity between the future microvillous region and the surface (arrows). Blood is clearly evident. 518 J. M. ARNOLD AND R. E. YOUNG .& . - * X 'A ' '• •**••'*.' - .^ :, SQUID PHOTOPHORE ULTRASTRUCTURE 519 candidate for light production since they are in the proper position, are highly specialized, and have direct communication with both the nervous system and circulatory system. The microvillous areas of these cells would seem to be the most likely site of light production. In many invertebrate photoreceptors, light is converted to chemical energy in strikingly similar structures (Clark, 1967). The availability of nutrition and oxygen, innervation, and mitochondrial energy would strongly support this assumption. The possibility that the "photophore" might be some type of a secondary photoreceptor could be raised but in other genera organs with quite similar structure are undoubtedly photoproductive (Berry, 1920a, 1920b; Hoyle, 1902, 1904). The morphological similarities between organs of photoreception and photoproduction imply that interconversion of light and chemical energy may have a necessary basic subcellular mechanism. Although the observations on the development of the photocyte are not suf- ficient to reconstruct the complete pathway of development a few interesting points may be made. Apparently early in the life of a photocyte, it must be isolated from surrounding tissues; first by developing within another cell (the sheath cell) and later by being enclosed in a dense sheath continuous with the basement membrane of the circulatory system. This would imply that insulation from other cell types is necessary for either photocyte differentiation or function. It would be tempting to speculate that the sheath (and sheath cell) chemically or electrically insulate the photocyte but more likely is the possibility that the sheath functions in keeping the photocyte in intimate preferential contact with the circulatory system thereby in- suring an adequate nutrient and respiratory supply. The microvilli of the photo- cyte apparently arise from an intracellular membranous system that very early in development is in direct contact with blood. The retraction of the sheath cell as the photocyte presumably matures implies that the control of differentiation of the photocyte is somehow directed or enhanced by the sheath cell. The simple structure of the packing cells suggest that they primarily function in occupying space between more metabolically active tissues. In formalin preserved, unstained photophores, the packing cells are clear and would thereby not impede or modify light produced by the photocytes. It would seem most likely that the packing cells function in dispersal and separation of the photocytes and have a low metabolic level. Conversely, the mitochondrial cells, by the sheer volume of mitochondria seem to be very metabolically active. Because they branch and ramify throughout the photogenic tissue and make contact with many cells, they probably provide meta- bolic energy to the other cell types and in particular the photocytes. The structure of the circulatory system in the photogenic tissue is interesting because blood comes into direct contact with the photocytes and because in some areas vessels are apparently reduced to a basement membrane containing blood with neither pericytes nor endothelial cells present. Barber and Graziadei (1965) FIGURE 11. Packing cell and mitochondrial cell. Most of the cytoplasmic region of the packing cells (he) is occupied by a homogeneous material. In the mitochondrial cell (me) most of the cell's volume is devoted to mitochondria. FIGURE 12. Synapse with the photocyte. Note the sheath is broken in one region and the synapse appears to be in direct contact with the microvillous region (arrow). 520 J. M. ARNOLD AND R. E. YOUNG described four types of blood vessels in the circulatory system of Octopus and Sepia, two of which had the basement membrane exposed directly to the blood. A logical extension of this would be a completely "acellular vessel" composed of basement membrane only. Such a vessel is shown in Figure 9. These "acellular vessels" are frequently encountered and always are smaller than the cells they con- tact. However, since there always appears to be a basement membrane except where the blood directly contacts the photocytes this is not a true open circulatory system. In the case of the photocyte the basement membrane (=sheath) contains the whole cell suggesting the photocyte is intimately related to the circulatory system. It is possible to present an integrated speculative picture of the function of the cell types in the photogenic region of this photophore. Light is produced by energy conversion on the microvilli from precursors provided by the circulatory system and energy derived from the mitochondrial cells. The photocytes are probably stimulated to luminesce by the numerous synapses which penetrate the sheath. The photocytes are separated by packing cells and contained within the sheath. Although the above model of light production by one type of photophore of a deep sea cephalopod is hypothetical and theoretically tenuous, it is hoped this model can serve as a tentative structural basis for further comparative and physiological study. The authors would like to thank Dr. Richard M. Eakin for his helpful dis- cussions, Lois D. Williams-Arnold and T. Joiner Cartwright, Jr., for their technical assistance and Frances Horiuchi for preparing the manuscript. This paper was supported by NIH grant EY 00179 to the senior author. SUMMARY One type of photophore of the deep sea squid Pterygioteuthls microlampas was examined with the electron microscope and its fine structure described. The photo- genic tissue is composed of four cell types each with distinctive morphology which suggests their function. The photocytes branch and ramify throughout the central region of the photophore and have an extensive system of microvilli (the photo- genic organelle) which are arranged about a central blood filled lumen. The photocytes apparently develop inside a sheath cell and are surrounded by a sheath which is continuous with the basement membrane of the blood vessels. The photo- cytes and associated sheath cells are surrounded by packing cells whose cytoplasm is replaced with a homogeneous granular material. Finally, cells containing many mitochondria branch and ramify throughout the photogenic area. Apparently the circulatory system is in direct contact with the photocytes, and acellular blood vessels, composed only of basement membrane, are found throughout the photo- genic tissue. The similarity between photoproductive organelles and photoreceptive organelles is striking. SQUID PHOTOPHORE ULTRASTRUCTURE 521 LITERATURE CITED BARBER, V. C. AND P. GRAZIADEI, 1965. The fine structure of cephalopod blood vessels. I. Some smaller peripheral vessels. Z. Zcllforsch. 66: 765-781. BELL, A. L., S. N. BARNES, AND K. L. ANDERSON, 1969. A fixation technique for electron microscopy which provides uniformly good preservation of the tissues of a variety of marine invertebrates. Biol. Bull., 137 : 393. BERRY, S. S., 1920a. Light production in cephalopods. I. An introductory survey. Biol. Bull., 38 : 141-170. BERRY, S. S., 1920b. Light production in cephalopods. II. An introductory survey. Biol. Bull., 38: 171-195. BOLETZKY, S. V., 1970. Biological results of the University of Miami deep sea expeditions 54. On the presence of light organs in Semirossia Steenstrup 1887. (Mollusca, Cephalo- poda). Biol. Mar. Scl, 20 : 374-388. CLARK, A. W., 1967. The fine structure of the eye of the leech, Hclobdella stagnalis. J. Cell Scl, 2 : 341-348. CLARKE. M. R., 1965. Large light organs on the dorsal surface of the squids Ommastrephes pteropus, Symplectoteuthis oualaniensis , and Dosidicus gigas. Proc. Malacol. Soc. London, 36 : 319-321. CLARKE, W. D., 1963. Function of bioluminescence in mesopelagic organisms. Nature, 198 : 1244-1246. HOYLE, W. E., 1902. The luminous organs of Pterygioteuthis margaritifcra, a Mediterranean cephalopod. Manchester Mem., 46(16) : 1-14. HOYLE, W. E., 1904. Reports on the dredging operations off the west coast of Central Ameri- can by the Albatross, etc. Reports on the Cephalopoda. Bull. Mus. Comp. Zool., 43 : 51-64. NICOL, J. A. C., 1960. The Biology of Marine Animals. Interscience, New York, 707 pp. YOUNG, R. E., 1972. Function of extra-ocular photoreceptors in bathypelagic cephalopods. Deep-Sea Res. Oceanog. Abstr., 19: 651-660. YOUNG, R. E., 1973. Information feedback from photophores and ventral countershading in mid-water squid. Pac. Sci., 27 : 1-7. YOUNG, R. E., 1975. Leachia pacifica (Cephalopoda :Teuthidae) spawning, habitat and func- tion of brachial photophores. Pac. Sci., 29(1) : in press. Reference: Biol. Bull., 147: 522-534. (December, 1974) ULTRASTRUCTURE OF A CEPHALOPOD PHOTOPHORE. II. IRIDOPHORES AS REFLECTORS AND TRANSMITTERS JOHN M. ARNOLD, RICHARD E. YOUNG AND MAURICE V. KING Ke^valo Laboratory of the Pacific Biomcdical Research Center, University of Haivaii, Honolulu, Hazvaii; Marine Biological Laboratory, Woods Hole, Massachusetts; and Department of Oceanography, University of Hawaii, Honolulu, Hawaii, 2536 Olopita Street, Honolulu, Haivaii Animals that live in the deep sea confront predation pressures concerned with background illumination (Denton, 1970). When viewed from below a pelagic animal at moderate depth casts a shadow which makes it clearly visible to any potential predator below it. In many fishes the problem is, in part, solved by the evolutions of highly reflective sides which reflect light efficiently enough so that it matches the background illumination except when viewed from directly below. This latter problem is frequently diminished either by the fish becoming laterally compressed so it presents a minimum outline when viewed from below, or by the development of bioluminescent countershading which duplicates the background illumination. A third potential solution, that of evolving transparent tissues, is only partially successful because some tissues must be opaque to function. For example the eyes must receive light from only one direction or, in the cephalopods, the ink sac by its very nature cannot be transparent. Any organism which uses light as a countershading device has to match the physical characteristic of the background illumination. Daylight surrounding an animal swimming at a depth of a few hundred meters is limited in spectrum to the blue band because of the absorbtion characteristic of water. Furthermore, down- welling light is very directional and varies in intensity in relation to depth. For a bioluminescent countershading mechanism to be effective the photogenic organs must be able to simulate these light conditions. Recently Young (1972) has described mechanisms of countershading control involving feedback from extra- ocular organs of the cephalopods. This paper describes the iridophores of the photophores of Ptcrygiotcuthis microlampas and offers speculation of how they might function in effective control of bioluminescent countershading. METHODS AND OBSERVATIONS The techniques used in this study were described in our previous paper (Arnold and Young, 1974). The light microscope observations on freshly dissected photo- phores were made on shipboard in a darkroom. The iridophores in the photophore of Pterygioteuthls microlampas are of three major kinds with one additional kind occasionally encountered in the anterior capsule region. In addition, a reddish-brown pigment surrounds most of the posterior capsule and is evident even in 1 ^ thick sections. Each of these iridophore types is individually described below and the brown pigment is briefly included for the sake of completeness. Based on their morphology and location, the iridophores can be given descriptive names. The anterior and posterior cap iridophores are 522 SQUID IRIDOPHORE ULTRASTRUCTURE 523 FIGURE 1. Posterior cap regular iridophores showing the single iridosome within each iridophore. The iridophores are separated by a discontinuous layer of intercellular substance. Some of the brown pigment spicules are evident in one corner (pig). referred to here as "regular iridophores"; those of the inner funnel are called "irregular iridophores" (occasionally they also are found in the skin outside the photophore) ; and those in the clear lens are called "transparent iridophores" be- cause they transmit light. The other terminology used here is that of Arnold (1967). Regular iridophores The regular iridophores are characterized by a highly ordered layering of the individual iridosomal platelets which alternate with extracellular spaces. The platelets are very constant in thickness as is the space between them. The platelets average 75 nm in thickness (range 69 nm-90 nm) and the space between platelets averages 53 nm (range 34 nm-66 nm). The platelets run straight for long distances and follow the general shape of the cell, being slightly curved in the posterior cup or wavy in some regions of the anterior cap. There is one iridosome per iridophore, and it occupies almost the entire volume of the cell (Fig. 1). When viewed with a dissecting microscope individual iridocytes can be teased out of the posterior cup 524 ARNOLD, YOUNG AND KING SQUID IRIDOPHORE ULTRASTRUCTURE 525 and appear like silvery mirrors which reflect rather than transmit light. Each regular iridophore is joined to its neighbors by an interrupted dense intracellular substance. The spindle-shape cells overlap to form a continuous layer in the posterior capsule or a thick lentoid block in the anterior capsule. In the center of the axial cone, individual iridophores are stacked more or less above one another. Each of the axial cone iridophores terminates with a muscle junction which con- tinues across the axial cone through the inner funnel and attaches someplace out- side the posterior capsule (Fig. 2). Similar strands of muscle also are seen be- tween the iridophore and the rest of the anterior cap but not in the posterior cup region. The platelets are remarkably parallel and maintain constant spacing over the entire length of the iridosome. At their ends the platelets seem to end in an oblique line rather than all at right angles to their collective long axes, hence the iridosomes are trapezoidal or lentoid in shape (Fig. 1). Each platelet is membrane bounded and these membranes appear to arise by fusion of anastomosing vesicles which originate in the cytoplasm of the iridophores and are continuous with the extra- cellular space as described in Loligo iridocytes (Arnold, 1967). The platelets themselves apparently are formed by the coalescence of dense granular material first into droplets which fuse to a discontinuous reticulum, then into a solid platelet (Figs. 3, 4). The iridosome thus formed is more or less centered in the cell and the nucleus is displaced to one side. The iridophores of the anterior cap are fre- quently wavy and appear somewhat compressed but still maintain the regular packing between the platelets and intraplatelet space. These iridophores, and in particular, those in the "plug" in the axial cone are separated by a prominent flocculent intercellular materal (Fig. 5). There is a slight but consistent thickness and spacing difference between the iridosomal platelets of the posterior cup and the anterior cap with the iridophores of the posterior cup having an average thickness of 69 nm (range 69 nm-59 nm) and spacing of 50 nm (range 57 nm^4 nm) while the anterior cap platelets average 85 nm (range 90 nm-78 nm) and are spaced at an average of 57 nm (range 50 nm-66 nm). With respect to their electron density, development, and other morphological respects, the iridophores of the anterior cap and the posterior cup appear to be similar. Irregular iridophores In contrast to the regular iridophores, the irregular iridophores are extremely variable in their shape, platelet thickness and arrangement, and interplatelet spacing (Fig. 6). They occur in the inner funnel of the photophore and are occasionally found in the skin outside (usually below) the photophore. The platelets are extremely variable in thickness but generally are thicker than the regulars (average FIGURE 2. Muscle band (mb) traversing the axial cone and irregular iridophore layer. Note the muscle is attached to the regular iridophore of the "plug." The photogenic tissue is composed of several cell types but dominated by a homogeneous packing tissue (he). Dendritic processes of the photocytes (ph) are surrounded with a dense sheath or a sheath cell (sc). A mitochondrial cell (me) interdigitates between the packing cells. Within a sheath cell the nucleus (ph n) and one branch (ph b) of a developing photocyte can be seen. Several processes reminiscent of synapses are evident (sn). 526 ARNOLD, YOUNG AND KING *iV ' * '!^SB . ':•, ' -I" *^i * » > -wt* • ' , ' * ' •""•« " > :' • ' • . < • • -i f. , ' ^ .' . ' FIGURE 3. End of a regular iridosome showing the formation of the platelets (ip) by fusion of granular material (gm) and the origin of the interplatelet space (is) by fusion of vesicles (v) whose membranes persist as the platelet membrane. SQUID IRIDOPHORE ULTRASTRUCTURE 527 125 nm; range 100 nm-140 nm) and unevenly spaced (average 103 nm; range 58 nm-173 nm). In general the platelets run parallel and in approximately straight lines, but they are covered with irregular knobs which sometimes are hollow or vary in density (Fig. 7). The platelets end in cytoplasmic extensions which them- seives are extremely variable in thickness, orientation, and shape. Occasionally microtubules can be seen associated with the end of the platelets in a fashion similar to the development of Loligo iridophores (Arnold, 1967). These iridocytes inter- lock, overlap, and in general are not delineated from each other so that an exact determination of the boundaries of each cell can be made. In general they sur- round the axial cone to form another cone (inner funnel) which is much thicker at the distal end than the proximal. The general orientation of the platelets is somewhat perpendicular to the abutting concentric regular iridophores of the pos- terior cup and those of the distal anterior cap. The inner funnel is frequently traversed with muscle strands which connect to the innermost anterior cap irido- phores (Fig. 2). The platelets of the irregular iridophores appear to arise by fusion of granular material and interplatelet space seems continuous with a vesicular network in the cytoplasm of the iridocyte as well as the extracellular space (Fig. 6, 7). Transparent iridophores Distal to the anterior cap of the photophore there is a transparent flexible layer which will be referred to here as the lens. The outer surface is covered by an epithelial layer with a highly convoluted surface, a dense basement membrane, and a layer of muscle fibers. The inner surface is underlaid by a continuous surface of pavement epithelium. The iridophores themselves are unique because they con- tain multiple iridosomes in which the platelets are in almost exact parallel register. In cross section each iridophore contains 20 to 50 such iridosomes arranged through- out the cytoplasm (Fig. 8). In section iridophores may contain as many as 51 platelets which probably represent the maximum number. The platelets average 43 nm in thickness (range 39 nm-45 nm) and are spaced at an average of 82 nm (range 73 nm-90 nm). The platelets are membrane bounded and the spaces be- tween them are separated from the cytoplasm by a continuous plasma membrane. The platelets frequently have a discontinuous appearance and seem to be formed by coalescence of large droplets ; although no obvious developmental stages of these iridosomes have been found. Most frequently the platelets seem to end in direct contact with the surrounding cytoplasm but there are examples of one end of an individual platelet being capped by a membrane while the other end is continuous with the cytoplasm (Fig. 9). Although it has not been possible to trace the mem- brane bounded space to a point continuous with extracellular space, by analogy from the regular and irregular iridophores it would seem that such continuity is likely. Wide spaced iridophores In addition to the three commonly encountered iridophores mentioned above, a fourth type of iridophore is infrequently encountered in the photophore or surround- FIGURE 4. Glancing section of several iridophore platelets showing the reticulate nature of the platelets (ip) and the interplatelet space (is). SQUID IRIDOPHORE ULTRASTRUCTURE 529 ing skin (Fig. 11). These iridophores are typified by wide irregular spacing (average 51 nm; range 42 nm-59 nm) and by platelets that follow a somewhat erratic alignment (average 84 nm ; range 65 nm-120 nm. The single iridophore occupies a relatively small volume of the cell and the platelets appear to arise by fusion by dense granular material that becomes isolated between the fusing vesicles which form the interplatelet space (Fig. 11). These iridophores rarely occur between the transparent iridophores and the regular iridophores of the anterior cap and are occasionally encountered in the skin outside the photophore proper. Since they are so infrequently encountered, we have no data on their light reflecting characteristics. Pigmentation of the capsule In his description of photophores of Pterygiotciithis, Hoyle (1902 and 1904) mentioned a reddish-brown pigment layer which is found in the "connective tissue capsule" of the photophore. With the electron microscope, this pigment can be seen to be borne in small intracellular spicules which have a more or less random orientation with the cells (Fig. 12). The outer surface is bounded by a dense layer, but no internal structure is evident within the spicule proper. The pigment does not seem to be borne in any special vacuole or organelle although the cytoplasm between the clusters of spicules is less dense and lacks the granular nature of the rest of the general background cytoplasm. The pigment is dense enough to form a brown layer visible even in 1 ^ sections. With polarized light it shows a strong light-blue birefringence in Epon sections. Our micrographs provide no further information as to the origin or development of these brown pigment spicules. Freshly captured specimens produced a steady glow of blue light which was in a highly directional beam aimed anterioventrally. In freshly dissected photophores the posterior cup iridophores reflected blue to blue-green light normal to the platelets of the iridophore. The irregular iridophores have a "frosted-silver" ap- pearance when viewed at normal angles but at oblique angles are translucent. When illuminated normal to the platelet surface the anterior cap iridophores reflect yellow light and transmit blue and red light. It was extremely difficult to make reflectance observations on the transparent iridophores of the lens because in the intact photophore the reflection of the other iridophores caused confusion. When the lenses were dissected off the photophores they reflected yellow light and trans- mitted blue at angles normal to the platelet axis. DISCUSSION The major question to be considered in this paper is the possible function and significance of the various types of iridophores found in the photophore. Denton FIGURE 5. Iridophores of the "plug" region of the axial cone. Note the iridophores are separated by a dense floccular material and the iridosomes can be convoluted. FIGURE 6. Irregular iridophores in the layer outside the axial cone. The platelets tend to run in parallel arrays but vary in spacing and thickness and are covered with protrusions. FIGURE 7. Higher magnification of the irregular iridophore platelets. Microtubules (mt) can occasionally be seen in association with the forming ends of the platelet. The protrusions on the platelet frequently appear to be hollow and seem to be randomly spaced. Fine granular material (gm) appears to be fusing to form the platelets. 530 ARNOLD, YOUNG AND KING SQUID IRIDOPHORE ULTRASTRUCTURE 531 and Land (1971) have discussed in detail the mechanism of reflectance of irido- phores in fish and cephalopods. Huxley (1968) has provided a theoretical basis for the optical behavior of such reflectors, and Land (1972) has summarized both the physical and biological aspects of iridophore reflectance. In both ideal and non-ideal multilayer systems the first-order reflectance peak occurs at Amax = 2 (nada + nbdb) where na and nb are respectively the reflective indices of the optically light and dense layers in the stack, and where da and db are respectively the thick- nesses of these two layers (Land, 1972). We are somewhat hampered in the interpretation of the function of the iridophores because we do not have a direct measurement of the refractive index of either the platelets (optically dense layers) or the spaces (optically light layers) between them. However, Denton and Land (1971) have published values of 1.56 for the refractive indexes of the platelets from the "eyelid" of Loligo forbesi and Scf>ia elegans. Following Land (1972) we have assumed a refractive index of 1.33 for the optically light layers. Un- fortunately the calculated Amax values based on these indexes and measurements of thicknesses from photomicrographs do not agree with observed reflections from freshly dissected photophores. The most likely source of error is the measure- ments of the spaces between platelets which may suffer considerable shrinkage during fixation and embedding. Indeed by ignoring the measurements of the spaces and assuming the stacks represent ideal multilayers based on the mean thickness of the platelets, rather close agreement is achieved between observed and calculated reflectance in several cases. Calculations of reflectance for the regular iridophores in the posterior cup give a Amax of 43 nm which agrees with the subjectively determined blue reflection of freshly dissected iridophores. The orientation of these reflectors suggest that they function in redirecting light that would otherwise be lost. By selectively reflecting blue light these iridophores act as a color filter. Since another color filter is probably present in the anterior cap (see below) this system may seem redundant. However by reflecting only blue light, a light trap (the pigment screen surrounding the posterior cup) is provided for light reflected by the anterior cap iridophores. Calculations of reflectance for the anterior cap iridophores gives a Amax of 53 nm which agrees fairly well with the subjectively observed yellow reflection. Thus these iridophores will act as a color filter by passing blue light while reflecting probably yellow-green light back into the photophore. The iridophores of the "plug" region of the anterior cap have somewhat dif- ferent reflective characteristics. While measurements have not been made on the platelets, observations on fresh photophores indicate they reflect red light. It seems unlikely, however, that the photocytes produce light at these frequencies. These iridophores are attached to muscles which apparently have their origin out- FIGURE 8. Transparent iridophores of the lens. These iridophores contain many iridosomes, each of which has its platelets aligned with platelets in other iridosomes. Thus the whole lens is precisely oriented. The surface of the lens is covered with muscle processes and connective tissue. FIGURE 9. Higher magnification of a single transparent iridosome. Compare the inter- platelet spacing with that of the regular iridosomes. Note that some of the platelets are •continuous with the cytoplasm but that others end with a membrane cap (arrows). 532 ARNOLD, YOUNG AND KING ' '•? •»- • SQUID IRIDOPHORE ULTRASTRUCTURE 533 side the photophore proper. Tensional forces on the periphery of the iridophore from contraction of the muscles may result in reducing the spacing between the platelets thereby altering their transmittance and reflectance characteristics. This may provide a mechanism for regulating the intensity of emitted light perhaps by shifting the reflectance band into and out of the blue region. The irregular iridophores also offer some interesting ground for speculation. Since the spacing is so irregular and since platelets are covered with protrusions and knobs, they could not possibly function as ideal quarter A stacks. Their structure suggests they may diffusely reflect light back into the region of the axial cone. In this way light would be redirected but not directionalized as would be the case with the regular iridophores. This assumption agrees well with observa- tions on freshly dissected photophores. The lens iridophores are very different from the others. All stacks have similar alignment, and they have thin platelets and very thick spaces. These are clearly non-ideal multilayers. If the optical thicknesses of the platelets and spaces have a ratio of 1 : 3 (this assumes approximately the same degree of shrinkage as in the anterior cap and posterior cup iridophores), the calculated reflectance peak is at 54 nm or approximately the same as the anterior cap iridophores. This peak agrees reasonably well with the yellow reflection these iridophores seem to give in fresh photophores. Presumably only blue light reaches the lens iridophores be- cause of the filtering effects of the anterior cap iridophores. Blue light arriving at normal incidence to the lens iridosomes will pass through and out of the photo- phore. Blue light arriving at oblique angles may be reflected back into the photophores, however the transmittance and reflectance characteristics of these iridosomes require more careful scrutiny. Certainly the precise alignment of the iridosomes suggests they may act to collimate the light. It would seem, therefore, that the photophore of this midwater squid is a com- plex organ potentially capable of producing light of controlled wave length in a highly directional beam whose intensity can be regulated. The individual organs are all oriented in the same plane so that a beam of light could be directed vertically downward from the eyes. These organs, therefore, seem to meet all the require- ments necessary for ventral countershading. The authors would like to thank Dr. Richard M. Eakin for his helpful dis- cussions, Lois D. Williams-Arnold and T. Joiner Cartwright, Jr., for their techni- cal assistance and Frances Horiuchi for preparing the manuscript. This work was FIGURE 10. Wide spaced iridophore. This type of iridophore tends to be wavy and the platelets loosely parallel each other. The iridosome occupies a relatively small volume of the cell. This type of iridophore can be found in the skin or irregularly placed on the photophore. FIGURE 11. Higher magnification of the wide spaced iridophore. Except for the spacing the iridosome seems similar to the regular iridophore. FIGURE 12. Intracellular pigment spicules of the brown pigment layer surrounding the proximal posterior cup. The individual spicules more or less parallel the long axis of the tissue layer containing them. 534 ARNOLD, YOUNG AND KING supported by NIH Grant EY00179 to the senior author and by NSF Grant GA33659 to R. E. Young. SUMMARY The iridophores of one type of photophore of the deep sea squid, Pterygioteu- this microlampas were examined with the electron microscope and four different types were found. Three of these types have not been previously described. The regular iridophores of the posterior cup appear to be one-fourth wave length re- flectors and redirect the light produced by the photogenic tissue outward. The regular iridophores of the anterior cap have a different spacing and platelet thick- ness so they apparently pass blue light. The irregular iridophores form a cone around the photogenic tissue and probably randomly reflect light back into the photogenic tissue. The iridophores of the lens have many precisely aligned irido- somes with platelet spacing and thickness so that they appear to collimate light passing through them. It appears that these three types of iridophores reflect, transmit and collimate the light produced in the photophore to match the back- ground illumination hence making an efficient countershading mechanism. A fourth type of iridophore, the wide spaced iridophore, is rarely encountered and probably does not have a significant role in light attenuation in the photophore. LITERATURE CITED ARNOLD, J. M., 1967. Organellogenesis of the cephalopod iridophore : Cytomembranes in development. /. Ultrastruc. Res.. 20 : 410-420. ARNOLD, J. M., AND R. E. YOUNG, 1974. Ultrastructure of a cephalopod photophore : I. Structure of the photogenic tissue. Biol. Bui!., 146: 507-521. BAUMEISTER, P., AND G. PINCUS, 1970. Optical interference coatings. Sci. Amcr.. 223(6) : 58-75. DENTON, E. J., 1970. On the organization of reflecting surfaces in some marine animals. Phil. Trans. Roy. Soc. London Scries B, 258 : 285-313. DENTON, E. J., AND M. F. LAND, 1967. Optical properties of the lamellae causing interference colours in animal reflectors. /. Physiol. 191 : 23-24. DENTON, E. J., AND M. F. LAND, 1971. Mechanism of reflexion in silvery layers of fish and cephalopods. Proc. Roy. Acad. Sci. London Scries B, 178: 43-61. HOYLE, W. E., 1902. The luminous organs of Pterygiotcittliis margaritijera, a Mediterranean cephalopod. Manchester Mem, 46(16) : 1-14. HOYLE, W. E., 1904. Reports on the dredging operations off the west coast of Central America by the Albatross, etc. Reports on the Cephalopoda. Bull M*.is. Comp. Zool., 43: 51-64. HUXLEY, A. F., 1968. A theoretical treatment of the reflexion of light by multilayer structures. 7. Exp. Biol., 48 : 227-245. LAND, M. F., 1972. The physics and biology of animal reflectors. Prog. Biophys. MoL Biol.. 24 : 75-106. YOUNG, R. E.. 1972. Function of extra-ocular photoreceptors in bathypelagic cephalopods. Deep-Sea Res. Oceanog. Abstr.. 19 : 651-660. Reference: Biol. />»//., 147: 535-549. (December, 1974) CLINAL AND SIZE-DEPENDENT VARIATION AT THE LAP LOCUS IN MYTILUS EDULIS JOHN F. BOYER 1 Department of Zoology, University of loiva, loiva City, loiva 52240 and Marine Biological Laboratory, Woods Hole, Massachusetts 02543 The widespread use of enzyme electrophoresis has revealed that almost all popu- lations are characterized by a large amount of genetic variation (Lewontin and Hubby, 1966; Selander, Hunt and Yang, 1969; Ayala, Powell and Dobzhansky, 1971 ; Gooch and Schopf, 1972; Clegg and Allard, 1972) and the study of many enzymes has indicated that no locus can be predicted a priori as being monomorphic or polymorphic. To what extent these polymorphisms are adaptive has not been resolved, and this issue is central to our understanding of how genetic variability is maintained in natural populations. Evidence for selection may consist of (1) showing a convergence to a character- istic gene frequency after natural or artificial perturbation (Kojima and Yarbrough, 1967; Sved and Ayala, 1970; Berger, 1971), (2) demonstrating a correlation be- tween genie and environmental variation (Koehn, 1969; Schopf and Gooch, 1971; Hamrick and Allard, 1972; Powell, 1971), or (3) finding a progressive change in gene frequencies with increasing age (Koehn, Perry and Merritt, 1971 ; Tinkle and Selander, 1973; Fujino and Kang, 1968). Deviations from Hardy-Weinberg proportions or persistent linkage disequilibrium may also support a selectionist interpretation (Franklin and Lewontin, 1970; Charlesworth and Charlesworth, 1973; Allard, Babbel, Clegg, and Kahler, 1972; Clegg, Allard, and Kahler, 1972). This study of genie variation at the leucine-amino-peptidase locus in Mytilus edulis presents evidence of types (2) and (3). Mytilus edulis, the blue mussel, has a widespread range from North Carolina to Nova Scotia and is a prolific colonizer of a variety of habitats. Previous studies (Milkman and Beaty, 1970; Koehn and Mitton, 1972) have shown that the species is always polymorphic for three alleles at the leucine-amino-peptidase (LAP) locus and the frequency of these varies significantly from one locality to another. In addition to the three common alleles, [slow (S), medium (M). and fast (F), in order of increasing electrophoretic mobility], at least two other alleles, R and G, are found in most samples. Milkman (1971) and Milkman and Beaty (1970) have reported that Mytilus populations south of Cape Cod generally have a high frequency of the slow allele (around 50%) and those of Cape Cod Bay and north- wards have a frequency of 15-30%. Mussels from the Nissequoque River on Long Island have a slow frequency from 11% to 14% (Koehn and Mitton, 1072). Milkman (personal communication) has correlated these variations with differences in the timing of the tidal currents. The LAP genotypes of the Mytilus and Modiolus demissits populations at the four sites studied by Koehn and Mitton were signifi- 1 Present Address : Department of Biological Sciences, Union College, Schenectady, New York 12308. 535 536 JOHN F. BOYER cantly different between sites, but were similar for the two species at each site. Thus the large scale variation found by Milkman and the smaller scale variation in the Nissequoque River support the hypothesis of selective agency in maintaining this polymorphism. Mussels are dioecious and produce pelagic larvae which may disperse to habitats some distance from their parents. The larvae settle and after metamorphosis attach to a firm substrate and to each other by byssae, thus insuring an essentially sessile adult existence. For this reason, if a particular suite of environmental factors favors particular genotypes, we may anticipate a progressive change in gene frequency with increasing size. This study of Mytilus populations on the southern shore of Cape Cod Bay shows both clinal and temporal divergence in LAP allelic frequencies. METHODS AND MATERIALS Mussels were collected from two estuaries on the northern side of Cape Cod. The first locality, Sandwich Harbor/Mill Creek, is 1 km east of the northern terminus of the Cape Cod Canal, and the second, Scorton Creek, is 2.2 km east of Sandwich Harbor. Mussels are found in channels which drain extensive salt marshes and which are inundated by tidal sea water (mean tidal range is 3 meters). Mytilus is found in both the subtidal and intertidal zones from the entrance of the SANDWICH HARBOR MILL CREEK 600 500 700 I I /I FIGURE 1. Sandwich Harbor. The six sites in the estuary are indicated as filled circles, and the temporary colony from the high intertidal as an open circle. The distance from the entrance measured along mean tide line is given in meters. GENETIC VARIATION IN MYTILUS 2. Scorton Creek. Filled circles mark the three sites. Stone boulder jetty is indicated by hatching and the extensive mussel beds by stippling. channels into Cape Cod Bay to a point 800 meters upstream at Sandwich Harbor and 600 meters at Scorton Creek, but are not found higher than mean water. Mussels were sampled from 6 sites at Sandwich Harbor and 3 sites at Scorton Creek (see Figs 1 and 2). Collections were also made from the intertidal zone at the jetty on the east side of northern terminus of the Cape Cod Canal, the jetty at the Sandwich Harbor entrance, and five sites around the circumference of the very large bay and estuary which comprises Barnstable Harbor (12 km east of Scorton Creek). Mussels were maintained in the laboratory for one to three days in running sea water. Mortality was negligible prior to assay. The maximum length was meas- ured and then a portion of the liver was homogenized in a 1 : 1 dilution of running buffer (commercial Gelman tris-barbital, pH — 8.8). This extract was spotted on Gelman cellulose polyacetate strips and electrophoresis was carried out in Gelman Trays for 45 minutes at 250 V, 2 mAmp/strip. Strips were incubated for 6 minutes in 25 ml tris-maleate buffer (0.2 M) at a pH of 5.2 (adjusted with NaOH) con- taining 10 mg l-leucyl-/?-naphthalamide. They were stained for 6 minutes in 25 ml tris-maleate with 25 mg of Fast Black K-salt. 538 JOHN F. BOYER TABLE I Gene frequencies and sample sizes for all localities by size classes. Number of rare alleles, and observed and expected (Hardy-Weinberg) percentages of homozygotes (SS, MM, and FF) also given. Size range (in mm) Number in sample Frequency (in %) of: Number of rare alleles Per cent homozygotes: slow medium fast observed expected Sandwich Harbor/ Mill Creek Mill Creek Upstream 4-8 143 36 28 36 1 51 34 9-13 98 17 34 49 0 47 38 14-19 138 24 33 43 2 55 35 20-25 133 25 33 42 4 44 35 26-31 96 14 41 45 4 58 39 32-45 209 13 38 49 4 57 40 46-70 214 16 38 46 3 57 38 Mill Creek Subtidal 5-19 153 16 31 53 4 47 40 20-33 87 22 28 50 0 56 38 34-81 146 15 30 55 3 55 42 Station "C 20-25 30 25 34 41 1 26-31 53 22 32 46 2 47 36 32-56 63 35 27 38 1 45 34 Station "B? 26-31 32-44 30 66 30 26 32 34 37 40 1 1 60 34 Station "A1 12-25 26-39 74 22 24 49 33 32 43 19 1 1 59 35 Sandwich Harbor Entrance 9-13 49 24 36 40 1 52 35 14-19 69 18 42 40 4 55 37 20-25 47 53 22 25 1 65 39 26-31 75 54 19 27 1 43 40 32-53 69 47 27 26 3 30 36 GENETIC VARIATION IN MYTILUS 539 TABLE I (continued) Size range (in mm) Number in sample Frequency (in %) of: slow medium fast Number of rare alleles Per cent homozygotes: observed expected Scorton Creek Upstream 4-8 9-19 20-25 26-45 46-76 21 109 45 65 114 33 23 37 28 16 24 33 30 26 34 43 44 33 46 50 0 2 0 2 3 51 60 35 47 35 34 36 39 Intermediate 8-15 96 25 37 38 5 46 34 16-29 96 24 32 44 4 48 35 30-45 95 38 28 34 2 52 34 46-70 75 30 34 36 1 46 34 Entrance 12-20 89 29 32 39 1 50 34 21-30 20 30 37 33 0 — . — 31-40 35 47 23 30 0 46 36 Two Jetty Sites Cape Cod Canal 7-13 83 35 25 40 2 54 34 14-19 65 42 22 36 2 51 35 20-25 50 46 29 25 1 47 36 26-31 42 45 17 38 1 58 38 35-82 108 42 27 31 6 50 35 Sandwich Harbor 3-8 272 38 29 32 10 50 34 9-11 76 21 33 46 4 52 36 12-19 52 30 36 34 2 55 34 Barnstable Harbor/ Five Sites Mussel Point 14-29 30-40 79 63 44 32 23 26 32 42 0 1 48 52 35 35 540 JOHN I-'. BUYER TABLE I (continued) Size range (in mm) Number in sample Frequency (in %) of: Number of rare alleles Per cent homozygotes: slow medium fast observed expected Barnstable Harbor/ Five Sites Mill Way Bridge 18-25 31 39 22 39 1 . _ 26-32 19 40 34 26 0 - — — 33-45 50 53 24 23 1 45 39 46-72 50 56 22 22 1 43 41 Scudder's Lam: 20-25 33 33 27 40 0 . 26-31 15 47 20 33 0 — — 32-85 58 45 23 32 3 54 36 Rendezvous Lane 27-44 46-70 50 50 44 51 25 21 31 28 0 0 40 46 35 38 Barnstable Upstream 19-39 40-78 30 28 33 52 18 21 48 27 0 0 — — RESULTS The gene frequencies for the slow, medium and fast alleles for all localities are given in Table I. The number of the two rare alleles is also given; these alleles constitute only 1.1% of the total and show no pattern in their distribution over size classes or sites. Three mussels (in 4871) produced no discernible spots. The pattern of allelic variation From Table I we see that mussels in the different size classes from the Sand- wich Harbor and Scorton Creek localities exhibit a wide variation in the frequency of S (from 13% to 54%) and a relatively constant ratio of M to F (about 2:3). The smaller mussels from all sites at these two localities generally have intermediate S frequencies, although there are differences between some of the classes at a site (note the contrast among the smallest three classes at the Mill Creek Upstream site). The larger mussels (those greater than 25 mm) show the greatest differ- ences, and both Scorton Creek and Sandwich Harbor/Mill Creek have the same pattern of low frequencies of S at the upstream sites and high frequencies of S at GENETIC VARIATION IN MYTILUS 541 the entrance sites. This allelic variation is best summarized by the Chi-square tests of Table II, which compare the numbers of slow alleles among sites by com- bined size classes. For all mussels 25 mm and smaller, sites are essentially homogenous; the same test applied to the larger mussels reveals striking hetero- geneity. Comparison of the larger mussels from the two entrance sites alone shows no significant difference; the two upstream sites are significantly different (S is 15% versus 21%), but converge to virtually identical frequencies when only those mussels over 45 mm are compared. The mussels from all the intermediate sites are concordant with this clinal pattern in that there are similar slow frequencies in the smaller mussels and the larger mussels have intermediate frequencies. Size classes between different sites may not be strictly comparable because the growth rates depend on such environmental factors as exposure time and food supply ; however the class intervals were chosen a priori and not on the basis of tests of significance and therefore the tests and interpretations are conservative. There is no good topographical counterpart to the Sandwich Harbor and Scor- ton Creek estuaries at Barnstable Harbor. Three of the sites — Scudder's Lane, Rendezvous Lane, and Mussel Point — were rock/sand beaches on the perimeter of the open water of the Harbor ; in this they resembled the estuarine entrance sites, but differed in that they were far from the outlet of the Harbor to Cape Cod Bay. No mussels were found in the channels of the marsh except at the Barn- stable Upstream site where they were very- sparse ; very few mussels were found alive and none was smaller than 19 mm. The Mill Way site was an isolated colony attached to a stone bridge about 100 meters from the Harbor. The frequencies of the mussels from the Canal Jetty are essentially invariant throughout the size range. The Sandwich Jetty mussels \vere from a recent and temporary settlement, for no mussels were larger than 19 mm and most (87%) were less than 12 mm. Their growth rates were probably retarded since they were found high in the intertidal. Although the ratio of medium to fast frequencies is much less variable among samples than is slow allele frequency, the ratio of medium to fast is not independent of slow frequency ; the ratio is generally greater than 2:3 in those samples with a high slow frequency, and less than 2 : 3 where the slow frequency is low. This is true for Barnstable Harbor as well as Sandwich Harbor and Scorton Creek, and this observation may support the hypothesis that these alleles are not a neutral polymorphism. Deviations from Hardy-Weinberg frequencies Mussels in almost all size-classes from all localities show a significant excess of homozygote genotypes (SS, MM, and FF) over Hardy-Weinberg expectations. From Table I can be seen that for those samples of 35 or more individuals, only one (from the entrance to Sandwich Harbor) has fewer homozygotes than ex- pected, and the average excess is 14%. There are three general explanations for this which do not involve selection: (1) presence of a null (or silent) allele, (2) persistent inbreeding, or (3) incorporation of the progeny of different populations into one sample. 542 JOHN F. BOYER TABLE II Chi-square test for heterogeneity on allelic proportions for Sandwich Harbor /Mill Creek and Scorton Creek localities, by size; A = Mill Creek Upstream Intertidal; B = Sandwich Entrance; C = Scorton Creek Upstream; D = Scorton Creek Entrance (Jetty). Expected values in parentheses; nij = ni.n.j/ntotai. (1). Mussels Under 26 mm A B C D Sum Slow: 267 96 97 54 514 (262) (95) (102) (56) Fast + Medium : 628 228 251 137 1244 (633) (229) (246) (135) Sum : 895 324 348 191 1758 Chi-square = 0.58; 3 d.f. ; P > 0.90; 514/1758 = 29% (2). Afussels Over 25 mm A B C D Sum Slow: 150 144 74 42 410 (239) (66) (83) (22) Fast + Medium: 877 137 281 54 1349 (788) (215) (272) (74) Sum : 1027 281 355 96 1759 Chi-square = 190; 3 d.f.; P < 0.001 (Ja). Two Upstream Sites (A and C); Mussels Over 25 mm Mill Creek Scorton Creek Sum Slow : 150 (14.6%) 74 (20.8%) 224 (166) (58) Fast + Medium : 877 281 1158 (271) (297) Sum : 1027 355 1342 Chi-square = 7.56; 1 d.f.;P < 0.01 (3b) . Two Upstream Sites (A and C); Mussels Over 45 mm Slow: Fast + Medium: Sum : 68 (69) 357 (356) 425 (16.0%) 37 (16.6%) (36) 186 (187) 223 105 543 648 Chi-square = 0.04 ; 1 d.f. ; P > 0. 80 GENETIC VARIATION IN MYTILUS 543 TABLE II — (Continued) (4). Two Entrance Sites (B and D); Mussels Over 25 mm Sandwich Harbor Scorton Creek Sum Slow: 144 (51.2%) 42 (43.8%) 186 (139J (47) Fast + Medium : 137 54 191 (142) (49) Sum: 281 96 377 Chi-square = 1.61; 1 d.f. ; P < 0.20 A mussel heterozygous for a null allele will be scored as a homozygote, and if the null allele is lethal when homozygous, only test crosses will detect its presence. Under the hypothesis, the frequency of the null allele (i>) can be estimated as (1) (1 •- t)/(l + t) = j>, where t is the ratio of observed heterozygotes to ex- pected heterozygotes. Inbreeding without change in gene frequency will produce an excess of homo- zygotes with frequency (2) 2f(pi) + (1 -- f)pi2, and the total number of heterozygotes will be reduced by a factor / (the inbreeding coefficient, Crow and Kimura, 1970). Adjustment by this parameter reduces the disparity between expected and observed by a factor of about 10. Both the null allele and inbreeding models produce identical expected fre- quencies ; in fact we obtain (3) f = =f/(2«f) and therefore can not be distinguished by the data of a single locus. For most of the samples, estimates of v range from 7c/o to 16% (or / from 0.13 to 0.28). These are high values. The alternative explanation — that individuals in a sample originate from parental populations with disparate gene frequencies (the Wahlund effect) — seems more reasonable because (1) the extended pelagic existence favors dispersal and admixture, (2) striking differences in gene frequencies between adjacent size classes are observed, and (3) there are adjacent localities with characteristically high or low slow frequencies which are likely sources for the immigrants (Milk- man and Beaty, 1970; Milkman, 1971). In this model, we hypothesize that two parental populations produce progeny (with different Hardy-Weinberg proportions) which comprise a mixed population at the sampling site. The six genotypes in the mixed population may estimate five variables (the relative contribution of the parental populations and the frequencies of two alleles in each) and a minimum least-squares fit to the data obtained by 544 JOHN F. BOYER trial-and-error methods. However, once this is done, the system is still under- determined and even if parental populations are assumed to have slow frequencies at or beyond the observed extreme values for slow, this model does not account for the data as well as the inbreeding model. Further, this mixed population model would also predict a decrease in homozygote excess in those size-classes with extreme values of the slow allele. Figure 3 does not show (with one exception) this trend. A further objection is that all other populations of Mytilus — including those putative parental populations with a nearly constant high or low slow fre- quency— show a consistent homozygote excess (Koehn and Mitton, 1972; Milkman andBeaty, 1970). Environmental variation Since our hypothesis is that the divergence in slow allele frequency is brought about by selection, we would like to correlate such environmental factors as ex- posure time, temperature, salinity, current and wave action, substrate, and as- sociated biota with the pattern of genie variation. Environmental differences can be seen to affect the mussels even without electrophoretic assay, because their size range progressively increases as one goes upstream. Extensive search at both entrance sites did not produce any mussels larger than 53 mm at Sandwich Harbor and 40 mm at Scorton Creek. At the upstream sites there were many mussels per square meter larger than 60 mm. The period of exposure does not seem important because the Mill Creek mussels were taken from both the high intertidal ( exposure time in excess of five hours out of the twelve and one-half hour tidal cycle) and the subtidal zones, and the Scor- ton Creek upstream mussels were from the lowest intertidal (exposure time of about one hour). Mussels at both entrance sites were exposed for two to three hours. At high tide mussels at all localities were covered by full-strength sea water ( salinity of 32'/<( ) and at low water the subtidal mussels were in effluent of minimum salinity of 10/rV. The lowest values for sites A, B, and C were 12-14", and for the Sandwich Harbor entrance the minimum was 20/^r. The Mill Creek Upstream mussels were never exposed to water of less than 26% o. Although the ultimate sources of the channels are fed by fresh water, the extensive marsh stores sea water like a sponge, and therefore the lower salinity values (less than 25'/f ) are recorded in the stream only after many of the mussels have been exposed. The temperature range of the water follows the same pattern in that the temperature was between 19° and 21° except in the few tide pools where it reached 30°. and in the channel at low tide where a maximum reading of 24° was once obtained on a hot sunny day. Air temperature over exposed mussels was close to ambient. The chief difference between entrance and upstream sites is that mussels at the entrance were almost always attached singly to rocks buried in the sand, whereas upstream mussels were found in clumps or beds with direct attachment to one- another. At Mill Creek in particular, the beds were extensive and overlayed an ooze composed of fine silt, organic deposit, and broken mussel shells into which a collector would sink to his knees. At the Scorton Creek Upstream site the mussels were from mid-channel near low water, and had attached to large timbers which GENETIC VARIATION IN MYTILl'S + 20%- •" +15% 0 O O) o E o * w X + 10% + 5% 0% -5% • A A o.i 0.2 0.3 0.4 0.5 Frequency of Slow Allele FIGURE 3. Deviations from Hardy- Weinberg expectations of homozygote frequencies versus slow frequencies in different size-class samples. Some adjacent size-classes (of Table I) with very similar genotype frequencies have been combined to give larger samples ; N > 65 for all points except one (N = 48) from Scorton Creek (with star). Dotted lines delimit "intermediate" frequencies of slow ; Mill Creek/Sandwich Harbor filled circles ; Scorton Creek, filled triangles; Sandwich Canal Jetty, filled upside down triangle. presumably protected them from shifting sand and provided additional anchorage. To summarize, at the Scorton Creek Upstream and Intermediate sites, and at the two Mill Creek sites, and to a lesser extent Stations B and C, mussels were in clumps (several handfuls would fill a bucket), and at the Entrance sites (and Sta- tion A) they were collected individually. Although there was discernible water movement either upstream or downstream through the channels throughout most of the tidal cycles, swift roiling currents were observed only in the three hour interval around low tide, where speeds attain five knots. Only the subtidal and low intertidal mussels were subject to the strong currents; presumably only the entrance sites would be exposed to wave action during a storm. The associated biota seem to have little effect on the mussels themselves. No predators were seen except starfish and oyster drills at the Sandwich Canal Jetty ; the small crabs found throughout the marsh were common in the mussel beds. In some samples platyhelminths and nematodes were inadvertently collected — most commonly in the upstream samples and rarely from the sand/rock beaches of the entrance sites. Only at the Barnstable Upstream site was Modiolus found with Mytilus. 546 JOHN F. BOYER Soft substrate, higher density, larger size, and "upstreamness" generally seem to go together, but no single factor stands out as the major determinant of gene frequency — particularly in view of the fact that an equivalent range of variation for any one factor could probably be found at either Sandwich Jetty (where the mussels were as dense and as large as upstream) or Barnstable. The localities were studied only over the months of July and August, and it is likely that the greatest ecological contrasts may come in the winter months when productivity declines and freezing and ice impose a severe stress. A length of about 25 mm seems to mark the division of 1st and 2nd year animals (Harger, 1970; Milkman, 1971), and since allelic divergence begins at this size, a seasonal episode might be more important in selection than the differences which result from daily or tidal cycles. DISCUSSION The divergence of slow allele frequency with increasing size strongly argues for selection; moreover it suggests a form of balancing selection which would maintain an enzyme polymorphism. Therefore these results parallel and sub- stantiate the selectionist interpretation of Koehn and Mitton (1972) for the LAP locus in Mytilus and Koehn, Turano, and Mitton (1973) for the Tetrazolium oxidase (To) locus in Modiolus. There are however some important differences. First, the Mytilus populations described by Koehn and Mitton have a nearly con- stant S frequency (11%-14%) and the major component of the differences be- tween the sites is the degree of homozygote excess (the statistical tests were made on the genotypes of FF, FX, and XX, where X -= M or S). There is no evidence for a clinal pattern at the four sites; whether there is genetic heterogeneity be- tween size classes is not reported. Modiolus populations at both a high intertidal and low intertidal site have size- dependent changes in Tetrazolium oxidase frequencies (Koehn et al., 1973). These data are pertinent because these sites may be ecological counterparts of the up- stream and entrance sites of Mytilus, and Koehn and Mitton (1972) have shown that Modiolus has a pattern of genie variation similar to Mytilus in the Nissequoque River. The TO allele frequencies are essentially constant over size classes and be- tween sites, but at both sites there is an increase with age in the heterozygote rela- tive to the two homozygotes. In Mytilus \ve find — in general — variation in allele frequencies and no trend in Hardy-Weinberg deviations ; but for both species, there is an initial homozygotc excess. For Modiolus, Koehn, Turano, and Mitton (1973) suggest an alternative ex- planation for this excess, namely selection against heterozygotes at the larval stage prior to settlement. This seems unlikely in the case of the LAP locus in Mytilus because there are three heterozygote genotypes which would appear to be equally disadvantaged. The cause of the persistent Hardy-Weinberg deviations is important in that our conclusions as to what is being selected depend upon this issue. If the mussels at each site are immigrants from two (or more) parental populations, then all genes are linked (Milkman and Beaty. 1970) and the LAP locus is a marker of parental origins. Selection is acting in this case on an entire genome, and it is for this GENETIC VARIATION IN MYTILVS 547 reason we anticipate an increasingly better fit to Hardy-Weinberg proportions with age. If mating is non-random but essentially panmictic, then selection could be act- ing on the LAP locus and closely linked loci. If the populations are panmictic and all genes are in linkage equilibrium (and the deviations are due to a silent allele), then a strong case could be made for selection at the LAP locus itself. None of these alternatives is strongly supported by the evidence. The three mussels which produced no stain reaction support the null-allele theory, but they could have resulted from a physiological inactivation rather than a null genotype, and there is the further onus of justifying how this allele could persist in high frequency (around 12%). Inbreeding (or assortative mating) is very plausible for many animal populations, but the gametes of Mytilus are spawned into open water. Microgeographic variation, or more likely, temporal variation in spawning cor- related with genotype could be responsible for non-random mating. The difficulties with the mixed-population model have already been discussed; of course, to reject it as the explanation for homozygote excess is not to insist that all individuals at a site originate from one population. As in most studies of this kind, the mussels were collected over a period of time which was brief compared to the life-span of the animal, and the assumption has been made that the pattern we observe is persistent and not due to differential colonization in prior years. R. Milkman has studied a Mytilus population in the Cape Cod Canal for three successive years, and reports a consistent colonization pattern over this period. Further, differential colonization as a result of habitat selection would also imply an adaptive role in maintaining the polymorphism. Although the problem of excess homozygotes has not been resolved, the Mytilus data taken as a whole strongly implicate selection as the agent responsible for the pattern of genie variation. The clear divergence of LAP allele frequences with increasing size at two separate, but ecologically similar, localities is good evidence for an adaptive polymorphism in a natural population. I would like to thank Dr. Roger Milkman for suggesting this study and for his help on numerous occasions. The technical assistance of R. Zeitler and the use of some of the facilities of the Marine Ecology course at Marine Biological Laboratory are gratefully acknowledged. The work was supported in part by the Biological Sciences Development Grant, GM 2591, from the NSF to the University of Iowa and NIH GM 18967 to R. Milkman. SUMMARY 1. Samples of the mussel, Mytilus edulis, were taken from 16 sites on the northern shore of Cape Cod. The mussels were measured and their genotypes at the leucine-amino-peptidase (LAP) locus determined by electrophoresis. 2. At two separate estuarine localities, Sandwich Harbor and Scorton Creek, a pronounced cline in slow allele frequencies in the larger mussels was found. 548 JOHN F. BOYER with upstream sites showing characteristically low frequencies (about 15%) and downstream (entrance) sites having high frequencies (45% to 55%). Mussels smaller than 26 mm had intermediate (22% to 35%) slow frequencies. 3. This clinal divergence with increasing size strongly argues for an adaptive polymorphism. Mechanisms for the observed excess of homozygote genotypes in almost all samples are also discussed. LITERATURE CITED ALLARD, R. W., G. R. BABBEL, M. T. CLEGG, AND A. L. KAHLER, 1972. Evidence for co- adaptation in Az'cna barbata. Proc. Nat. Acad. Sci., 69 : 3043-3048. AYALA, F. J., J. R. POWELL, AND T. DOBZHANSKY, 1971. Polymorphisms in continental and island populations of Drosophila unllistoni. Proc. Nat. Acad. Sci., 68: 2480-2483. BERGER, E. M., 1971. A temporal survey of allelic variation in natural and laboratory popula- tions of Drosophila mclanogastcr. Genetics, 67 : 121-136. CHARLESWORTH, B., AND D. CHARLESWORTH, 1973. A study of linkage disequilibrium in popu- lations of Drosophila mclanogaster. Genetics, 73 : 351-359. CLEGG, M. T., AND R. W. ALLARD, 1972. Patterns of genetic variation in the slender wild oal species Avena barbata. Proc. Nat. Acad. Sci., 69 : 1820-1824. CLEGG, M. T., R. W. ALLARD, AND A. L. KAHLER, 1972. Is the gene the unit of selection? Evidence from two experimental plant populations. Proc. Nat. Acad. Sci., 69 : 2474- 2478. CROW, J., AND M. KIMURA, 1970. An Introduction to Population Genetics Theory. Harper and Row, New York, 324 pp. FRANKLIN, I. R., AND R. C. LEWONTIN, 1970. Is the gene the unit of selection? Genetics, 65: 707-734. FUJINO, K., AND T. KANG, 1968. Transferrin groups of tuna. Genetics, 59: 79-81. GOOCH, J. L., AND T. J. M. SCHOPF, 1972. Genetic variability in the deep sea : relation to environmental variability. Evolution 26 : 545-552. HAMRICK, J. L., AND R. W. ALLARD, 1972. Microgeographical variation in allozyme fre- quencies in Avcna barbata. Proc. Nat. Acad. Sci., 69: 2100-2104. HARGER, J. R. E., 1970. Comparison among growth characteristics of two species of sea-mussel, My tiles edulis and M. calif or nianus. Vcligcr, 13: 44-56. KOEHN, R. K., 1969. Esterase heterogeneity: dynamics of a polymorphism. Science, 163: 943-944. KOEHN, R. K. AND J. B. MITTON, 1972. Population genetics of marine pelecypods. I. Ecologi- cal heterogeneity and evolutionary strategy at an enzyme locus. Amcr. Natur., 106: 47-56. KOEHN, R. K., J. E. PERRY, AND R. B. MERRITT, 1971. Esterase enzyme function and genetical structure of populations of the freshwater fish, Notropsis strainincus. Amer. Natur., 105 : 51-69. KOEHN, R. K., F. J. TURANO, AND J. B. MITTON, 1973. Population genetics of marine pelecy- pods. II. Genetic differences in microhabitats of Modiolus demissus. Evolution, 27 : 100-105. KOJIMA, K., AND K. N. YARBROUGH, 1967. Frequency dependent selection at the Esterase — 6 locus in Drosophila melanog aster. Proc. Nat. Acad. Sci., 57: 645-649. LEWONTIN, R. C., AND J. L. HUBBY, 1966. A molecular approach to the study of genie heterozygosity in natural populations. II. Amount of variation and degree of hetero- zygosity in natural populations of Drosophila pscudoobscnra. Genetics, 54 : 595-609. MILKMAN, R., 1971. Genie polymorphism and population dynamics in Mytilus edulis. Biol. Bull, 141: 397. MILKMAN, R. AND L. D. BEATY, 1970. Large-scale electrophoretic studies of allelic variation in Mytilus edulis. Biol. Bull., 139 : 430. POWELL, J. R., 1971. Genetic polymorphisms in varied environments. Science, 174: 1035-1036. GENETIC VARIATION IN MYTILUS 549 SCHOPF, T. J. M., AND J. L. GOOCH, 1971. Gene frequencies in a marine ectoproct : a cline in natural populations related to sea temperature. Evolution, 25: 286-289. SELANDER, R. K., W. G. HUNT, AND S. Y. YANG, 1969. Protein polymorphism and genie heterozygosity in two European species of the house mouse. Evolution, 23: 379-390. SVED, J. A., AND F. J. AYALA, 1970. A population cage test for heterosis in Drosophila pscudoobscura. Genetics, 66: 97-113. TINKLE, D. W., AND R. K. SELANDER, 1973. Age-dependent allozymic variation in a natural population of lizards. Biochem. Genet., 3 : 231-238. Reference: Biol. Bui!., 147: 550-559. (December, 1974) FEEDING OF OVALIPES GUADULPENSIS (SAUSSURE) (DECA- PODA: BRACHYURA: PORTUNIDAE), AND MORPHOLOGICAL ADAPTATIONS TO A BURROWING EXISTENCE EDSEL A. CAINE *• Department of Biological Sciences, Florida State University, Tallahassee, Florida 32306 Ovalipes guadulpcnsis (Saussure) is a portunid commonly trawled from the sand bottoms along the northwestern Florida Gulf Coast. A sister species, 0. ocellatus (Herbst), is reported to bury itself in the sand bottoms off North Carolina. When buried the respiratory currents are reversed and the crab emerges from the sand and actively "hunts" for food during rising or falling tides (Pearse, Humm, and Wharton, 1942). The habitats of the two species are similar, except adult O. guadulpensis are usually not found near-shore (Williams, 1965). As no other information was available on the habits of O. guadulpensis, a study was undertaken to determine the food and feeding mechanisms of this crab and to assess the morphological adaptations to a burrowing existence. MATERIALS AND METHODS Specimens of 0. guadulpensis were trawled from sand bottoms in the Gulf of Mexico along northwestern Florida in waters from three to eight meters in depth. Some individuals were preserved upon collection in 70% ethanol and then re- frigerated. These crabs were used in stomach content analyses and morphological investigations. Other crabs wrere kept alive and returned to the laboratory and placed in running sea water aquaria. These were the subjects of observations on feeding, burrowing, and respiratory currents. Stomach contents and the gastric mill were investigated following the methods of Caine (1974). Motion pictures were made of both feeding and burrowing activity and the film was analyzed frame by frame. Analysis of the film was correlated with the morphology of the appendages and visually verified by repeated observations. Respiratory currents were tested using carmine to follow external currents and small quantities of methylene blue in a seawater solution to trace the internal water pathways. OBSERVATIONS AND RESULTS Laboratory observations indicated that 0. guadulpensis was a nocturnally active organism that was buried beneath the sand during periods of daylight. When buried the tips of the antennules may remain exposed, or all parts of the crab may be covered by a layer of sand up to 2 cm deep. When exposed (i.e., active), the crab walked with the dactyli of the walking legs penetrating the substrate and was extremely aggressive with conspecifics or other portunids (aggressive encounters 1 Current address : Department of Biology, University of Tampa, Tampa, Florida 33606. 550 ADAPTATIONS OF O. GUADULPENSIS 551 B FIGURE 1. Body postures of O. gitadulpcnsis. (A) at rest on the substrate; (B) while burrowing with the chelipeds flexed. probably do not occur with crabs of other families in nature and no aggressive behavior was directed toward members of other locally occurring brachyuran families when introduced in the laboratory). Burroiving When at rest on the substrate, the carapace of the crab was approximately parallel with the surface, the walking legs were at rest on the substrate, and the chelipeds were not tightly flexed against the carapace (Fig. 1A). The substrate was sandy, and the compactness of the sand was thought to dictate the type of burrowing activity. If the bottom was hard packed sand, the crab pushed its left second and fourth walking legs and the right first and third walking legs per- 552 EDSEL A. CAINE pendicularly into the substrate to the level of the distal portion of the merus. The carpus, propodus, and dactylus thus formed an unflexed "probe" into the sand. As the second and fourth left walking legs and the first and third right walking legs were withdrawn from the substrate, the first and third left walking legs and the second and fourth right walking legs were inserted into the sand in a similar manner. The flattened fourth walking legs pushed sand posteriorly as they were withdrawn from the substrate. This procedure was repeated two to four times in rapid sequence (film analyses indicate that a complete cycle for the insertion and withdrawal of the four pair of walking legs took approximately 0.5 seconds). Up to this point the chelipeds and the body of the crab remained relatively stationary. After the substrate had been sufficiently probed and made less compact (the cessation of probing activity is thought to be related to the proprioceptors in the walking legs and the extent of muscular activity needed to push the legs into the substrate) actual burrowing began. The walking legs continued to move in a manner similar to probing, except the dactylus of each leg was flexed approximately 30° upon full penetration of the substrate, and, instead of withdrawing the ap- pendages, the body of the crab was pulled downward by the same muscular activity. The angle of penetration of the body of the crab in relation to the substrate was about 60°. As the body of the crab began to be buried, the chelipeds were folded tightly against the subhepatic and pterygostomian portions of the body (Fig. IB). The alternate movements of the legs continued until the chelipeds touched the substrate — the laterally projecting merus, carpus, and propodus forming a barrier. With all four pair of walking legs extended into the substrate and the dactyli all flexed, the other joints of the walking legs straightened, and the anterior por- tion of the body of the crab was pushed 0.5-1.0 cm away from the substrate. As the body was elevated, the chelipeds scooped and pushed sand anteriorly from beneath the body of the crab and then returned to a tightly flexed position. The walking legs then contracted in unison and the water between the crab and the substrate was forced from the space, carrying sand with it. As the posterior por- tion of the crab remained buried and the chelipeds formed an anterior barrier, the main currents of water exit were between the chelipeds and the first pair of walking legs, between the first and second pairs of walking legs, and between the second and third pairs of walking legs. Due to the depression already formed by the burrowing activity, the water exited in an upward direction and, as the waterborne sand settled, the carapace of the crab was covered. The movements also forced the crab deeper into the substrate. This process was repeated until the crab was buried, with the extent of cheliped movement being an inverse function of the depth of burial. The setose ridges on the chelipeds and along the medio-lateral margins of the pterygostomian and subbranchial regions of the body functioned in blocking water from passing anteriorly, or into the branchial chambers, as the body of the crab was rapidly pulled downwards by the walking legs, thus insuring the movement of water in a lateral and upward direction. If the substrate was not compressed, the crab probed and dug with the second, third, and fourth legs of one side and then the other side. The mechanics of each leg was the same as previously described for burrowing. The first pair of walking ADAPTATIONS OF O. GUADULPENSIS 553 B FIGURE 2. Water filtration, (A) second antenna with associated setae, scale bar equals 2 mm; (B) antero-ventral view of the crab showing the placement of the appendages (setae not shown); (A) second antenna ; (M) third maxilliped. legs aided the chelipeds in removing sand from beneath the crab until the chela were partially covered, and then the first pair of walking legs functioned with the respective limbs of the right or left side of the crab. At this point the burrowing mechanisms became the same as for crabs on compressed substrates. Respiratory patlnvays A major problem encountered by burrowing decapods is that of respiration. The gills are encased by the carapace and, in most non-burrowing forms, the respiratory currents pass upwards between the walking legs or through the Milne- Edwards openings dorsal to the basis of the chelipeds and into the branchial cham- bers. The beat of the scaphognathite pulls water through the branchial chamber and out through the buccal area. Occasionally, short reversed currents occur in non-burrowing forms. Whether buried or unburied, Ovalipes guadulpensis primarily utilized a re- versed current with water entering around the scaphognathite, passing through the branchial chamber, and then exhaled antero-lateral to the fifth walking leg. The incorporation of reversed respiratory currents required little morphological modi- fication from the "normal" respiratory pathways of others non-burrowing crabs (described by Borradaile, 1922) : the recovery beat of the scaphognathite in normal respiration became the active, pumping beat in reversed currents, and the pumping action during normal flow became the recovery stroke. Water was then pushed, 554 EDSEL A. CAINE B D FIGURE 3. Oral appendages of O. guadulpcnsis, (A) third maxillipeds; (B) second maxillipeds ; (C) first maxillipeds; (D) second maxilla; (E) first maxilla; (F) manidible. Scale bar equals 2 mm. rather than pulled, through the branchial regions. Some morphological adapta- tions occurred in O. guadulpensis which helped to insure the exclusion of particu- late matter from entering the branchial chambers during reversed water currents. These included : ( 1 ) an anterior elongation of the distal portion of the merus of the third maxillipeds (Figs. 1A, IB, 2B, 3A) which are very setose, (2) the elongation of the distal portion of the endopodite of the first maxillipeds in a medial direction (Fig. 3C) and associated setae on the distal margins, (3) the fusion of the second and third segments of the second antenna, with the fused segments being very setose (Fig. 2A), (4) dense setae around the basis of the walking legs and chelipeds (Fig. 1A), and (5) a reduction in the size and an in- crease in the capability of closing the Milne-Edwards openings. ADAPTATIONS OF O. GUADULPENSIS 555 During anterior to posterior (reversed) respiratory ventilation, the setae on the distal portion of the first maxillipeds were held against the endostome, with the lateral margins of the appendage also in contact with the endostome. The medial setae of the paired appendages were in close proximity and effectively impeded participate matter from passing posteriorly. The third maxillipeds occasionally were slightly lowered (ca. 5°) to facilitate water entry (the space created by burrowing did not conform with all parts of the crab, a small space filled with interstitial water occurred in the buccal area and near the basis of the fifth pair of walking legs). Most water entered the buccal area around the fused segments of the second antenna (Fig. 2B). The water was thus filtered for sand and detritus at least twice, by the setae on the antennae and then the first maxillipeds, with water occasionally being filtered three times by passing through the setae on the distal portion of the merus of the third maxillipeds after being filtered by the antennal setae. As the scaphognathite beat posteriorly, the water within the branchial chamber was forced posteriorly with the water being exhaled from the chamber through the posterior openings. As the scaphognathite returned anteriorly, water was drawn through the buccal area and was filtered by the setae on the antenna and first maxillipeds. After passing the first maxillipeds, the water followed the endo- stome laterally and passed into the branchial chambers through respective openings in the antero-lateral portions of the endostome (one on either side). The openings were created as the anterior portion of the scaphognathite lost contact with the carapace. As the scaphognathite beat posteriorly, these holes were closed. Water entering the branchial chamber flowed over the dorsal surface of the gills with water flowing in two main pathways, one dorsally and the second being a broad current passing posteriorly and ventro-posteriorly (probably passing be- tween the gill lamellae). The gill lamellae were bathed and cleansed by the water. The maxillipeds were not moving during reversed ventilation, hence the epipodites on these appendages were relatively stationary and not moving along the gills. Normal respiratory currents (ventral to dorsal and then anterior), which would be reversed for this species, occurred occasionally while buried. When this oc- curred a stream of water was exhaled vertically from the buccal cavity (the direc- tion of the current due to the third maxillipeds), carrying the sand and detritus from the filtering setae. Both ventilatory pathways were utilized by non-buried specimens, but reversed currents still predominated. When not buried the chelipeds were not tightly flexed against the body of the crab (Fig. 1A), thus unblocking the Milne-Edwards openings, and the third maxillipeds were lowered ventrally approximately 15°. Water was not filtered as it entered the anterior branchial chambers from both dorsal and ventral openings, but the maxillipeds were often in motion indicating that the epipodites may have been cleansing the gills. Feeding Ovalipes guadulpensis was shown to be an opportunistic carnivore on macro- material by stomach content analyses. Examinations revealed that fish, ophuroids, crustaceans, and annelids composed the majority of the diet, but occasionally 556 EDSEL A. CAINE molluscs and algae were also found. It appeared that this crab captured unwary prey by rapidly attacking organisms which were detected while the crab was buried. The antennules were seldom buried and often the eyes were also exposed, therefore detection of prey would pose no problem. The presence of sedentary polychaetes and some molluscs also indicated that some prey were found while the prey, but not the crab, was buried. Detection was probably achieved with the dactyli of the walking legs which contacted the prey organisms while the crab was walking across the substrate. The chemosensitive properties of the dactyli of brachyurans has been documented by Case and Gwilliams (1961) and Case (1964). Whether the prey was detected while the crab was buried or not, the actual prey capture was accomplished while the crab was not buried. The prey was grasped by one or both of the chela and brought to the oral area. The oral ap- pendages were laterally spread as the prey was forced between the appendages by the chelipeds. All of the oral appendages except the mandibles articulate ventro- laterally and are capable of both medio-lateral and dorso-ventral flexture. The mandibles articulate dorso-laterally and are capable of similar appendage move- ment (Borradaile, 1922). The mandibles (Fig. 3F), which were extended ventrally, closed medially on the food material and the third maxillipeds (Fig. 3A), which were extended dorsally, also closed medially on the substance. The mandibles were typical for most decapods in that the molar process was not a masticating surface but more of a serrate cutting edge while the third maxillipeds were typical for most brachyura in that it functioned as an outer covering for the buccal area with the medial mar- gins of the ischium being serrate and forming the crista dentata. As the two pairs of appendages closed medially, the food material was clasped by the molar processes of the mandibles and the crista dentata of the third maxil- lipeds. The mandibles then contracted dorsally as the third maxillipeds contracted ventrally and the food material was usually cut or torn in the region of the mandibles. The crista dentata of O. guadulpensis was not exaggerated or strongly serrate, perhaps the function of water filtration with the dense setae on the medial margins of the third maxillipeds dictated a morphological reduction of the crista dentata. In any case, the ventral pull of the third maxillipeds were often supplemented by one or both of the chelipeds. The chelipeds, which were already holding the food material, merely pulled antero-ventrally. The food material still broke in the mandibular region. The endopodites of the second maxillipeds (Fig. 3B) aided in the movement of the nutritive substances through the buccal region while the endites of the first maxillipeds (Fig. 3C) aided in the retention of food materials in close proximity to the mandibles. The two pair of maxilla (Figs. 3D, E) also functioned in the retention of food material, covering the space ventral to the mandibles. The mandibular palp (Fig. 3F) aided in the movement of food into the mouth, as did the endopodites of the second maxillipeds (Fig. 3B), and the labrum (the upper lip). ADAPTATIONS OF O. GUADULPENSIS 557 A B FIGURE 4. Masticating portions of the ossicles of the gastric mill, (A) lateral tooth ; (B) median tooth. Once past the mouth the food passed through the esophagus by peristaltic waves and entered the cardiac portion of the stomach. The food of the Reptantia is masticated and the particles sorted by a collection of setae and ossicles known as the gastric mill (Patwardhan, 1935; Schaefer, 1970). Through muscular con- tractions, three toothed structures interdigitate and effectively break the food sub- stances into smaller pieces. These three structures, the left lateral tooth (Fig. 4A), the median tooth (Fig. 4B), and the right lateral tooth are interconnected so that upon contraction of certain muscles, the two lateral teeth move medially and the median tooth moves ventrally (Yonge, 1924). Each lateral tooth (Fig. 4A) was composed of a single, large anterior denticle and a series of three ventro-lateral teeth of progressively decreasing size from posterior to anterior. Dorsal to the ventro-lateral teeth was a series of six vertical ridges. The median tooth complex (Fig. 4B) was composed of a medio-ventral projection which was shaped like a pair of interconnected wings where the broadest portion was directly anteriorly. Lateral to the central projection were two enlarged, elevated ridges. The parts of these teeth that contacted each other were dependant on the extent of muscular contraction. A strong muscular contraction resulted in the greatest movement of the ossicles so that the anterior denticles of the lateral teeth were touching at the midline with the elevated central projection of the median tooth making simultaneous contact with the anterior denticles. As these structures .were the largest and the margins of these structures were the sharpest, it was thought that strong contractions oc- curred when masticating dense or hard material. The lateral teeth did not abut when the contraction of the stomach muscles were weaker. There was a space of approximately 0.5 mm separating the lateral teeth so that the lateral ridges of the median tooth passed over the ventro-lateral teeth 558 EDSEL A. CAINE of each lateral tooth. As the median tooth moved ventrally, the food material was cut by the ventro-lateral teeth and also pressed against and grated by the vertical ridges of the lateral teeth. DISCUSSION Ovalipes guadulpcnsis is a portunid adapted to an existence on hard packed sand, burrowing during periods of daylight and nocturnally becoming active and moving over the substrate. The burrowing mechanisms are advantageous for an existence only on a sand substrate; similar burrowing activity on a coarser sub- strate such as shell hash or a finer substrate such as silty mud would probably be ineffective. This conclusion is based on the initial probing behavior and the raising and lowering of the body of the crab which creates water currents that aid in bur- rowing. Experiments to verify the substrate specificity of the burrowing mecha- nisms were not conducted, though. Arudpragasm and Naylor (1966) related the predominance of a reversed res- piratory current with increased utilization of oxygen from oxygen poor waters. Interstitial water, which forms the bulk of the water circulated over the gills while buried, is probably less saturated with oxygen than open waters. Morphological adaptations to a burrowing existence are similar to those of Corystes cassivelaunus (Pennant) described by Hartnoll (1972), except the antennal flagellae are not setose and do not conduct filtered water to the buccal area. A compact sieve is formed dorsally from the fused second and third segments of the second antenna; ventrally from the merus of the third maxillipeds; and the lateral boundaries are extensions of the pterygostomial region of the carapace. The setose antenna of Corystes cassirclanus are probably analogous to the setose distal margin of the first maxillipeds of 0. guadulpensis. The range of food ingested by 0. guadulpcnsis indicates that the crab is an opportunistic carnivore and scavenger capable of detecting food materials while buried or while actively walking along the substrate. The feeding mechanisms are similar to those described for other carnivorous decapods, but the structure of the lateral teeth of the gastric mill indicate that larger material is more frequently in- gested than smaller material. While this study described various adaptations and habits of 0. guadulpensis. the factors causing adults to move off-shore were not found. LITERATURE CITED ARUDPRAGASM, K., AND E. NAYLOR, 1966. Patterns of gill ventilation in some decapod Crustacea. /. Zoo/., 150: 401-411. BORRADAILE, L. A., 1922. The mouthparts of the shore crab. /. Linn. Soc. (Zool.), 35: 115- 142. CAINE, E. A., 1974. Feeding and masticatory structures of six species of the crayfish genus Procambarus (Decapoda: Astacidae). Forma et Functio, in press. CASE, J., 1964. Properties of the dactyl chemoreceptors of Cancer anicrrarnis Stimpson and Cancer productus Randall. Biol. Bull. 127 : 428-446. CASE, J., AND G. F. GWILLIAMS, 1961. Amino acid sensitivity of the dactyl chemoreceptors of Carcinidcs macnas. Biol. Bull., 121 : 449-455. ADAPTATIONS OF O. GUADULPENSIS 559 HARTNOLL, R. G., 1972. The biology of the burrowing crab, Corystcs cassivelaunus. Bij- dragen tot dc Dicrkunde, 42 : 139-155. PATWARDHAN, S. S., 1935. On the structure and mechanism of the gastric mill in Decapoda. Proc. Indian Acad. Sci., 1 (B) : 183-196; 359-375; 405-413; 414-422; 693-704. PEARSE, A. S., H. J. HUMM, AND G. W. WHARTON, 1942. Ecology of sand beaches at Beau- fort, N.C. Ecol. Monogr., 12: 135-190. SCHAEFER, N., 1970. The functional morphology of the fore-gut of three species of decapod Crustacea : Cyclograspus punctatus Milne-Edwards, Diogenes brcvirostrus Stimpson, and Upogcbia africana (Ortmann). Zoo/. Ajricana, 5: 309-326. WILLIAMS, A. B., 1965. Marine decapod crustaceans of the Carolinas. U. S. Fish Wildl. Scrv., Fish. Bull., 65 : 1-298. YONGE, C. M., 1924. Studies on the comparative physiology of digestion. II. The mechanisms of feeding, digestion, and assimilation in Ncphrops norvcgicus. J. Exp. Biol., 1 : 343- 389. Reference: Biol. Bull., 147: 560-572. (December, 1974) BODY FLUID COMPOSITION AND AERIAL OXYGEN CONSUMP- TION IN THE FRESHWATER MUSSEL, LIGUMIA SUBRO- STRATA (SAY) : EFFECTS OF DEHYDRATION AND ANOXIC STRESS THOMAS H. DIETZ Department of Zoology and Physiology Louisiina Stiitc University, Baton Rouyc, Louisiana 70803 A number of bivalve molluscs are intermittently exposed to air (Bayne, 1973; Kuenzler, 1961; Lent, 1968; Moon and Pritchard, 1970). Usually the valves are closed and it is generally accepted that the animals are under anaerobic conditions (Dugal, 1939; Prosser, 1973). However, some intertidal mussels have been ob- served to partially gape and consume oxygen from the atmosphere (Coleman, 1973 ; Kuenzler, 1961 ; Lent, 1968; Boyden, 1972). The freshwater mussel, Liguinia subrostrata, lives in still water within 50 centimeters of the surface (Murray and Leonard, 1962). These animals will migrate with changes in the water surface level. Occasionally, receding flood wa- ters or drying of the pond leave L. subrostrata exposed to the atmosphere. During this time the valves are closed and some animals will survive for several weeks. Other bivalves have survived emersion over 12 months (Dance, 1958). Epithelial tissues used in gas exchange are permeable to water. When ex- posed to an atmosphere with high relative humidity, water loss is minimal. How- ever, prolonged exposure to a low relative humidity atmosphere results in sig- nificant water loss and concomitant changes in body fluid solute concentrations (Hiscock, 1953). This report examines the roles of aerial respiration, desiccation and anoxia on changes in body fluid composition and survival of L. subrostrata when removed from water. Evidence is presented indicating these animals are not anoxic when out of water but maintain aerobic metabolism. However, they can tolerate forced anoxia. METHODS Specimens of Ligumia subrostrata were obtained from a pond near Baton Rouge during July 1973 to March 1974. The animal shells were washed with tap water and most were used within one week of collection. Animals kept out of water were maintained at room temperature (22-25° C). Animals not im- mediately used were stored in tap water (3 HIM NaHCO3, 0.1 mM NaCl) or in artificial pond water (Prosser, 1973) at room temperature. The valves were separated by mechanical force and mantle cavity water drained. The posterior mantle tissue was separated from the shell at the pallial line and reflected anteriorly. The body fluid which accumulates in the exposed pallial space was withdrawn by a pipet. Frequently, the posterior adductor muscle was cut to facilitate fluid collection. The fluid was centrifuged (8000 X g] to 560 MUSSEL BLOOD AND AERIAL RESPIRATION 561 remove cells and total solute determined immediately. The remaining fluid was either diluted for other analyses or stored frozen (—20° C). Fluid analyses Total solute was determined on undiluted samples of body fluid using a Pre- cision Systems Osmette (0.2 ml) or a Hewlett-Packard vapor pressure osmometer (25 yul). Sodium and potassium were determined on diluted samples using a Coleman flame photometer. Chloride was estimated with a Buchler-Cotlove titrator. Body fluid was diluted with LaO3-HCl and calcium determined by atomic absorption (Perkin-Elmer) . Lactate and ninhydrin positive substances (NPS) were determined on super- natant of body fluid precipitated with equal volumes of cold 10% trichloroacetic acid and centrifuged at 8000 X g. Lactate was estimated by the Strom (1949) modified colorimetric method. NPS was measured by a colorimetric method (Rosen, 1957). Ammonium excreted by L. sitbrostrata to the bathing medium was measured by direct nesslerization. Equal volumes of bath samples and W% Nessler's reagent were mixed and nitrogen content determined by colorimetry. Rates of ammonium excretion were estimated from the appearance of nitrogen in the bath. Tissue glycogen and analyses Total body weight was determined by weighing the animals after draining the water from the mantle cavity. The soft tissue was dissected from the shell and both shell and tissue were dried to constant weight at 85° C. Shell weight was unchanged when dried further to 105° C, whereas tissue declined by 3.3 ± 0.2% (10). Total tissue carbohydrates (glycogen) was determined by a phenol-sulfunc acid colorimetric method (Montgomery, 1957). The soft tissue of the mussel was dried to constant weight (85° C) and digested with 20% KOH (100° C). Di- luted alkaline digest was analyzed directly for carbohydrate. The alcoholic pre- cipitation of glycogen was eliminated since direct analysis agreed with precipitate analysis (± 4%). Oxygen consumption Aerial respiration was determined using a Gilson respirometer. Animals were transferred to 100 ml flasks (Aminco) containing 200 [A water, to maintain water saturated air, and 200 pi 20% KOH as CO2 absorbant. Oxygen consumption was measured at 22° C for 1.5-3 hr after > 30 min equilibration period. The correc- tion factor for O2 consumption (Qo2) was (Pb X 273) / (760 X T), where Pb is atmospheric pressure and T is absolute temperature under experimental condi- tions. All Q0a values are expressed as /A O2/g dry tissue-hr at standard tem- perature-pressure. Sex and season may affect Q02 (Newell, 1970) but the data are incomplete to attempt corrections. Oxygen consumption was determined by a modified Winkler method (Strick- 562 THOMAS H. DIETZ land and Parsons, 1972). Animals were transferred to containers rilled with Oo saturated pond water and sealed for 1-2 hr at 21° C. At specific times, the con- tainers were opened and water carefully siphoned into 300 ml BOD bottles, with 150 ml overflow, for oxygen determination. The decrease in Oo content from the Oo measured initially in the water was used to estimate Qo2. Anoxia Animals were transferred to pond water in a desiccator jar and tubing from a No cylinder connected to an air stone in the water. The nitrogen was continuously flushed through the system and vented to the atmosphere. The pond water was changed daily with pregassed water (<54 pi Oo/l). Animals maintained in a N._, atmosphere were sealed in a similar container but were supported out of water. The atmosphere was replaced twice each day (5 min flushing with No). COo absorbant was also placed in the container. Data are expressed as the mean ± one standard error. Differences between means were analyzed by the student "t" test and considered significant if P < 0.05. Regression lines were estimated by the method of least squares. RESULTS Freshwater acclimated animals Lignmia subrostrata body fluid concentration is among the lowest concentration found in fresh water animals. Tissue and body fluid composition for tap water acclimated mussels is indicated in Table I. Total tissue water has not been parti- tioned into intra- and extra-cellular compartments. The measured ions account for 72% of the total solute. Bicarbonate is probably a major anion in the body fluid which was not measured (see Potts, 1954). TABLE I Tissue and body fluid composition of tap water acclimated L. subrostrata. Component Units Number of animals MeaniSEM Total weight g 26 23.7 ± 2.4 Shell g/lOg total 26 4.33 ± 0.10 Fresh tissue g/lOg total 26 5.67 ±'0.10 Mantle cavity water g/lOg total 10 1.2 ±0.1 Tissue water g/lOg fresh tissue 18 8.8 ± 0.0 Body fluid : Total solute mOsmoles/1 26 53.0 ± 0.6 Na mM/1 26 21.0 ±0.6 Ca mM/1 8 4.8 ± 0.2 K mM/I 10 0.4 ± 0.0 Cl mM/1 26 12.0 ±0.6 Lactate mM/1 15 0.14 ± 0.01 NFS* mM/1 20 1.28 ± 0.17 Xinhydrin positive substance. MUSSEL BLOOD AND AERIAL RESPIRATION 563 200r 100 50 20 10 i i 0.5 1.0 5.0 10 Dry weight (g) FIGURE 1. Aerial oxygen consumption in L. subrostrata as a function of dry tissue weight. The Q02 is in /xl O«/g dry tissue-hr. The slope of the regression line is — 0.51 ± 0.02 (35). Freshly collected animals excrete ammonia. The animals were placed in in- dividual containers with tap water and aliquots taken for nitrogen determination. The average net flux of ammonium was 1.1 ±0.1 /tm N/g dry tissue-hr (15) over a one hr interval. Oxygen consumption L. subrostrata transferred from water to a respirometer flash consumes oxygen from the air. The weight specific Qo2 is a logarithmic function of dry tissue weight (Fig 1). Although the weight range is narrow, the correlation (r = —0.68) is TABLE II L. subrostrata oxygen consumption from water saturated air. Values are listed as mean ± / SEM. A, in laboratory tap water less than 5 days; B, in laboratory tap water 12 weeks; C, in sealed container with water saturated air for 30 days. Condition N Tissue dry weight, g Qo2 A 20 1.38 ±0.21 59 ± 6 B 5 1.40 ± 0.11 42 ± 6 C 10 1.52 ±0.15 80 ± 8* * P < 0.05. 564 THOMAS H. DIETZ TABLE III L. subrostrata oxygen consumption from air saturated pond water. Values are listed as mean ± 1 SEM. A, in laboratory pond water less than 5 days; B, in laboratory pond water 12 weeks. Condition N Tissue dry weight, g Qo2 A 10 2.16 ± 0.30 287 ± 16 B 7 1.29 ± 0.06 185 ± 17* *P < 0.001. highly significant (P < 0.001). The aerial Qo2 is not significantly changed after several months laboratory storage in tap water (Table II). Variability in Qo2 was high, ranging from 18 to 150. Oxygen consumption in water is significantly higher than aerial respiration (Table III). The Qo2 from mussels stored in tap water several months was sig- nificantly (36%) less than recently collected animals, however, the difference in weight tends to minimize the reduction. Aquatic respiration exceeds aerial Qo2 even after prolonged storage in tap water. Part of the elevated Qo2 in water is due to activity ; siphoning water, valve and foot movement. There was no change in O().. when the stored animals were pretreated 24 hr with penicillin (800 u/ml ) in the" water : Qo, 160 ±37 (5). The weight specific aquatic Qo2 is a logarithmic function of dry tissue weight (Fig. 2). The slope of the regression line is significantly different from that noted for aerial respiration (P < 0.001). All animals used for the regression analysis 500 Q 200 100 70 0.5 1.0 5.0 Dry weight (g) 10 FIGURE 2. Aquatic oxygen consumption in L. subrostrata as a function of dry tissue weight. The Qo2 is in fj.1 O2/g dry tissue-hr. The slope of the regression line is — 0.40 ± 0.06 (39) and the correlation is highly significant (r = —0.75, P > 0.001). MUSSEL BLOOD AND AERIAL RESPIRATION 565 were stored in pond water less 7 days. By 14 days a detectable drop of Qo2 was noted which becomes pronounced with long term storage (cf. Table III). Effects of prolonged exposure to air Animals transferred to a sealed container, with water saturated air and CC>2 absorbant, moved intermittently during the first few hours. After 12 hr, move- ment ceased and the valves appeared closed. The animals continued to consume Oa from the atmosphere even after 30 days exposure. The Qo2 was higher than values observed for animals out of water less than 3 hr, however, part of the increase may be due to microbe contamination, which was not controlled. These animals lost 1.44 ±0.16 g/10 g fresh tissue weight over the 30 day interval. However, the body fluid composition was similar to that found in animals subjected to 2 days dehydration at lower humidity (see below). Total tissue glycogen in freshly collected mussels was 0.36 ± 0.02 g/g dry tissue (10). Tissue carbohydrate was unchanged after the animals were exposed to air (high humidity) for 40 days (36 ±1% of dry weight, N = = 10). This is 88-95% of tissue glycogen from fresh animals at that season. Survival of L. subrostrata is reduced to 6-10 days if they are exposed to a low relative humidity atmosphere (45-55%) at 25° C. Oxygen consumption from air is significantly increased after 4-5 days dehydration : QO2 100 ± 24 (7). Micro- bial contamination may contribute to the elevated Qo2- Particulate matter ac- cumulates in the mantle cavity near the siphons when dehydrated. These animals lost 2.40 ± 0.68 g/10 g fresh tissue weight. During dehydration, the rate of evaporative water loss is relatively constant over a 4 day period (30 /il/lO g fresh tissue-hr). The loss of water results in an increased body fluid total solute concentration (Table IV). The highest total solute measured was 160 mOsmoles/1. The initial changes in calcium are higher than expected and may be due to calcium dissolved from the shell to buffer ac- cumulation of metabolic products. Both lactate and NFS are elevated relative to TABLE IV Effect of dehydration on body fluid ion composition in L. subrostrata. Data are expressed as mean ± 1 SEM, number of animals in parenthesis. Animals were maintained in air at 25° C, 45 to 55% relative humidity. Days of dehydration Component Units 2 4 7 Total solute mOsm/1 68.8 ± 1.7 (8) 89.0 ± 2.7 (12) 91.9 ± 7.8 (14) Na mM/1 24.5 ± 0.8 (8) 32.9 ± 0.6 (12) 37.5 ± 1.2 (14) Cl mM/1 18.4 ± 0.6 (8) 29.4 ± 1.7 (12) 35.2 ± 2.0 (14) Ca mM/1 8.4 ± 0.6 (8) 8.0 ± 0.9 (9) — Lactate mM/1 0.3 ±0.1 (8) 0.8 ±0.2 (7) 0.8 ± 0.2 (7) NFS mM/1 3.6 ± 0.1 (8) 2.8 ± 0.2 (12) 3.8 ± 0.5 (7) Total H2O loss g/10g* 1.46 ± 0.22 (8) 2.85 ± 0.03 (8) 3.44 ± 0.14 (7) * Based on initial fresh tissue weight. 566 THOMAS H. DIETZ 90 80 70 60 to O 50 O 40 30 20 10 10 20 30 40 50 Deficit (ml/IOOg fresh tissue) 60 FIGURE 3. The effect of dehydration on solute concentration in the body fluids of L. subrostrata. The abscissa is the water lost during dehydration. The deficit is based on initial fresh tissue weight excluding the shell and mantle cavity water. The ordinate is the reciprocal of solute concentration (niM/1)"1 multiplied by 103. TS is the total solute (mOsm/l)'1. Number of animals used for determining the regression lines was 46. tap water controls. However, no major changes were noted with prolonged dehydration. It is possible to relate the body fluid solute concentration to the degree of de- hydration in an animal (Alvarado, 1972). Assuming the water in a compartment is available as solvent and, if there are no changes in the solute content in the com- MUSSEL BLOOD AND AERIAL RESPIRATION 567 partment, then it is possible to predict the body fluid concentration as a function of water deficit : C= (Co- V0)/(V0-D) (1) Where C0 and V0 are the solute concentration in the body fluids and total water content of a hydrated animal (excluding the shell), respectively; C is the body fluid solute concentration after dehydration and D is the water deficit. The recipro- cal of Equation 1 can be rearranged into an equation for a straight line : 1/C= l/Co-D/(C0- V0) (2) Figure 3 shows the data plotted using Equation 2. The correlation between water deficit and Cl, Na or total solute is highly significant (P < 0.001 ) with cor- relation coefficients (r) for each regression line: -0.90, -0.95 and -0.97, re- spectively. Linear regression estimates of the Y-intercept (1/C0) and the X-inter- cept (V0) are presented in Table V. The calculated and observed C0 values are in agreement. However, the V0 estimates from Cl and total solute are significantly different from the observed values. These data indicate a shift of Cl out of tissues into the body fluid. Effects of anoxia Exposure of L. snbrostrata to water saturated N2 atmosphere resulted in a significant behavioral change. Within 12 hr of exposure to anoxic conditions, the valves gaped and the foot was extended several centimeters. This is appar- ently an attempt to expose additional tissue surface area for oxygen absorption. Living animals responded to repeated mechanical stimulation by slowly with- drawing the foot but re-extend it within a few hr. These animals could survive 4-6 days anoxia. The magnitude of oxygen debt was noted when 7 animals were returned to air in a respirometer flask : Q02 249 ± 56 over a 1 hr interval. Total tissue glycogen was not changed from the tap water acclimated animals. Exposure of L. snbrostrata to a N2 atmosphere resulted in significant changes in the body fluid composition (Table VI). Total solute increased rapidly even though dehydration was minimized. The maximum total solute measured in a TABLE V Calculated and observed values of solute concentration and water content of L. subrostrata. All values are mean ± 1 SEM, number of animals in parentheses. Co(mM/l) V»(ml/100 g tissue) Calculated* Observed Calculated* Observed Na 21.7 21.0 ± 0.6 (26) 89.1 87.7 ± 0.4 (18) Cl 14.0 12.0 ± 0.6 (26) 62.6** 87.7 ±0.4 (18) Total solute 54.4 53.0 ± 0.6 (26) 79.9** 87.7 ±0.4 (18) * From least squares regression estimates, n = 46. **P < 0.001, 568 THOMAS H. DIETZ TABLE VI Effect of NI atmosphere on body fluid ion composition in L. subrostrata. Data are presented as mean ± 1 SEM, number of animals in parentheses. Animals were in sealed containers at 25° C, 100% R.H. gassed with NI twice daily. Days of anoxia Component Units 2 4 Total solute mOsm/1 107 ± 7 (14) 121 ±6(20) Na mM/1 20.0 ± 0.6 (14) 19.9 ± 0.6 (20) Cl mM/1 12.9 ±0.5 (14) 11.4 ±0.6(20) Ca mM/1 26.3 ± 4.0 (14) 28.0 ± 3.4 (6) Lactate mM/1 6.3 ± 0.9 (14) 6.7 ± 1.4 (10) NFS mM/1 4.1 ±0.4(8) 4.2 ± 0.4 (4) Weight loss g/10g* 0.65 ±0.15 (8) 0.69 ± 0.17 (10) * Based on initial total animal weight. surviving animal was 180 mOsmoles. Sodium and chloride do not change relative to tap water controls. The major ionic change is a 5 to 8 fold increase in calcium and the unidentified anion (see discussion). The highest measured calcium con- centration was 41 mM/1. The body fluid lactate concentration also increased 50 fold relative to tap water animals but the contribution to total solute is small. The NFS increased above tap water animals but the chemical nature is not known. The weight loss may be due to dissolved shell and subsequent evolution of CO2 rather than water loss since the Na and Cl concentrations were not changed. Although few animals survive longer than 4—6 days in a No atmosphere, sur- vival is extended if the animals are immersed in No gassed water (> 15 days). The animals did not extend the foot but all siphoned water. Apparently these animals do not accumulate an oxygen debt under these conditions. When returned to air the Q0o was 73.0 ± 14.6 (5) over a 1 hr interval. The body fluid composition is modified when exposed to Na gassed pond water (Table VII). The body fluid total solute is significantly higher than animals in aerated tap water but less than animals in No atmosphere. Calcium and lactate TABLE VII Effect of exposure to Ni gassed water on body fluid ion composition in L. subrostrata. Data are presented as mean ± / SEM. Days exposure* Component Units 4 15 Total solute mOsm/1 63 ±2 68 ± 1 Na mM/1 15.4 ±0.5 12.4 ±0.4 Cl mM/1 6.0 ± 0.3 4.1 ±0.2 Ca mM/1 14.4 ± 1.0 10.9 ± 0.6 Lactate mM/1 0.26 ± 0.05 0.23 ± 0.03 * N = 6 animals for each experiment. MUSSEL BLOOD AND AERIAL RESPIRATION 569 are elevated in response to anoxia but excess metabolic acids are apparently ex- creted. The measured net loss of calcium to the bathing medium was 2.0 ±0.1 ^m Ca/lOg fresh tissue-hr over 15 days exposure to N2 gassed water. Metabolic acid excretion was not monitored because of potential decomposition by microbes in the water. The reduction in body fluid sodium and chloride is significant but has not ^een pursued. DISCUSSION If water loss is retarded, L. subrostrata will survive over 40 days in air and other pelecypods exceed three weeks (Bayne, 1973; Dugal, 1939). Dance (1958) has reported a freshwater mussel survived 12 months out of water. Those animals which have been studied were not anoxic but consumed Oo from the atmosphere (Coleman, 1973; Kuenzler, 1961; Lent, 1968). The limiting factor for survival in air is primarily tolerance to desiccation. Although the body fluids of fresh- water mussels are dilute (Potts, 1954), L. subrostrata will tolerate a three fold increase in body fluid total solute, with an upper limit of 160 mOsmoles. Analyses of the body fluids indicate the higher osmolarity is due to water loss and concen- tration of solutes rather than a build up of metabolic products. It is noteworthy that these animals tolerate extensive changes in body fluid total solute and qualitative changes in ionic composition. They are able to survive a large increase in NaCl concentration during dehydration and an 8 fold increase in body fluid Ca during anoxia. The shift of Cl from the tissues to the body fluid during dehydration and the reduction of body fluid Na and Cl during anoxia in water are significant but the mechanism remains obscure. Apparently, the osmo- larity of body fluids is the primary factor limiting survival and the specific solutes are less important. A source of energy for survival out of water is glycogen (Stokes and Awapara 1968; De Zwaan and Zandee, 1972b). The quantity of total carbohydrate (36% of dry tissue weight) in L. subrostrata is similar to other pelecypods (Badman and Chin, 1973; De Zwaan and Zandee, 1972a, 1972b). Potential survival time, in air, for an animal of one gram dry tissue weight, using glycogen only, is about 200 days. Aerobic metabolism will also allow protein and lipid catabolism. Curiously, others have noted more rapid tissue glycogen loss (Badman and Chin, 1973; De Zwaan and Zandee, 1972b). In water, L. subrostrata excretes ammonia but out of water the nitrogen from protein catabolism may shift into less toxic urea pro- duction (Andrews and Reid, 1972). This may contribute to the elevated NPS production noted during dehydration. When pelecypods are returned to water, oxygen consumption is higher with a Qo about 3-4 times the aerial rate (Coleman, 1973; Kuenzler, 1961 ; Moon and Pritchard, 1970). This is probably reflecting the increased energy required in moving water to irrigate the mantle cavity. Animals recently collected have a greater Qo2 in water than animals stored in the laboratory for a period of time. Which value is "normal" cannot be established by these experiments. The higher Q0o may be due to greater foot movements, especially prominent during the first 2 weeks, or ventilation "searching" for food. Animals under both conditions have a high glycogen content so depletion of food stores is not the primary reason for 570 THOMAS H. D1ETZ low Qo2 'm laboratory adapted L. subrostrata. However, it has been suggested that starvation is the reason for the lower Q02 in some gastropods which have a lower carbohydrate reserve (von Brand, Baerstein and Mehlman, 1950). A transient higher Qo2 immediately after reimmersion in water (Moon and Pritchard, 1970) may be due to some oxygen debt or the energy required to remove trapped air and wastes accumulated in the mantle cavity during exposure to air (Boy den, 1972). Specimens of L. subrostrata were observed to open the valves allowing air to escape from ventral and posterior shell margins followed by forceful adduction of the shell which expelled additional air and participate matter from the siphon area. L. subrostrata is a facultative anaerobe. However, when exposed to anoxic conditions out of water, survival time is not longer than animals exposed to de- hydration in air (5-7 days). The body fluid total solute is elevated after pro- longed exposure to No but the maximum concentration tolerated is 180 mOsm. This is about the same for animals exposed to dehydration. However, the principal solutes contributing to the osmolarity in anoxic animals are calcium and its anion ; probably bicarbonate and succinate (Dugal, 1939; Potts, 1954; Stokes and Awa- para, 1968; De Zwaan and Zandee, 1972a, 1972b). The ability of L. subrostrata to tolerate prolonged anoxia in water (>15 days) is of ecologic significance. It is not unusual for water with a high organic content to drop to near zero oxygen for extended periods. Some gastropods will also tolerate anoxia in excess of two days, although there are considerable species dif- ferences (von Brand, et a/., 1950). The metabolic adaptations to facultative anaerobiosis in bivalves include pro- duction of the less toxic alanine and the less soluble succinate salts rather than lactate (Stokes and Awapara, 1968; De Zwaan and Zandee, 1972b). Recently, Hochachka and Mustafa (1973) have demonstrated that anaerobic metabolism in the Pacific oyster is adapted to generate additional substrate level phosphorylation as a-ketoglutarate is shunted into the Krebs cycle (Mustafa and Hochachka, 1973). The ability to excrete metabolic acids to the environment is energetically ex- pensive but prevents the build up of body fluid osmolarity which would be lethal. I wish to thank Ms. Gerry Bullock and Christine Angelloz for typing the manu- script, E. Stern for identifying the animals and M. Nowakowski for technical as- sistance. This work was supported, in part, by the Louisiana State University Graduate School Research Council. SUMMARY 1. Ligumia subrostrata, removed from water will survive >40 days if dehydra- tion is minimized by high relative humidity. They are not anoxic but consume oxygen from the air (Qo2 59 p\ O2/g dry tissue-hr). One source of energy is from the large glycogen stores (36% of dry tissue). 2. Animals removed from water and exposed to either low relative humidity (45_55%) or N2 atmosphere will survive about 5-7 days. The maximum total solute in the body fluids of surviving animals is 160-180 mOsmoles. MUSSEL BLOOD AND AERIAL RESPIRATION 571 3. Sodium, Cl, K, and Ca account for 72% of the total solute in the body fluids of fresh water acclimated animals. Dehydration increases the concentration of Na proportional to the amount of water lost. There is a significant shift of Cl from the tissues into the body fluids during dehydration. 4. When the animals are out of water, forced anoxia increases the body fluid total solute but the Na and Cl contribution is small. The major ions in the body fluids are calcium and bicarbonate or succinate ; reflecting a build up of metabolic acids. Lactate concentration in the body fluids is low. These animals experience a significant oxygen debt since oxygen consumption is elevated when returned to air (Qo2 249 (A O2/g dry tissue-hr) . 5. Mussels are facultative anaerobes and will survive for extended periods of time in No gassed water (<54 /A O^/l). The body fluids are slightly changed from normal suggesting the metabolic products are excreted. These animals do not suffer an oxygen debt. LITERATURE CITED ALVARADO, R. H., 1972. The effects of dehydration on water and electrolytes in Amhystoma tigriintin. Physiol. Zoo!., 45: 43-53. ANDREWS, T. R. AND R. G. B. REID, 1972. Ornithine cycle and uricolytic enzymes in four bivalve molluscs. Comp. Biochcm. Physiol., 42B : 475-491. BADMAN, D. G. AND S. L. CHIN, 1973. Metabolic responses of the fresh-water bivalve, Plcuro- bema cocciiiciim (Conrad) to anaerobic conditions. Comp. Biochcm. Physio!., 44B : 27-32. BAYNE, B., 1973. The response of three species of bivalve molluscs to declining oxygen ten- sion at reduced salinity. Comp. Biochcm. Physio!., 45A : 793-806. VON BRAND, T., H. D. BAERNSTEIN AND B. MEHLMAN, 1950. Studies on the anaerobic meta- bolism and the aerobic carbohydrate consumption of some fresh water snails. Biol. Bull., 98 : 266-276. BOYDEN, C. R., 1972. Aerial respiration of the cockle Cerastoderma cihilc in relation to temperature. Comp. Biochcm. Physiol., 43A : 697-712. COLEMAN, N., 1973. The oxygen consumption of Mytilns cdulis in air. Comp. Biochcm. Physio!., 45 A: 393-402. DANCE, S. P., 1958. Drought resistance in an African freshwater bivalve. /. CouchoL, 24: 281-283. DE ZWAAN, A. AND D. I. ZANDEE, 1972a. Body distribution and seasonal changes in the glycogen content of the common sea mussel, Mytilns cdulis. C/nnp. Biochcm. Physiol., 43A : 53-58. DE ZWAAN, A. AND D. I. ZANDEE, 1972b. The utilization of glycogen and accumulation of some intermediates during anaerobiosis in Mytihis cdulis L. Comp. Biochem. Ph\sioL, 43 B : 47-54. DUGAL, L. P., 1939. The use of calcareous shell to buffer the product of anaerobic glycolysis in Venus mcrccnaria. J. Cell. Comp. Physiol., 13: 235-251. HISCOCK, I. D.. 1953. Osmoregulation in Australian freshwater mussels ( Lamellibranchiata). Water and chloride ion exchange in Hydridclla aitstralia (Lam.). Aust. J. Mar. Freshivater Res., 4: 317-329. HOCHACHKA, P. W. AND T. MUSTAFA, 1973. Enzymes in facultative anaerobiosis of molluscs — 1. Malic enzyme of oyster adductor muscle. Comp. Biochcm. Physiol., 45B : 625- 637. KUENZLER, E. J., 1961. Structure and energy flow of a mussel population in a Georgia salt marsh. Limnol. Occanogr., 6: 191-204. LENT, C. M., 1968. Air-gaping by the ribbed mussel, Modiolus demissus (Dillwyn) : Effects and adaptive significance. Biol. Bull., 134: 60-73. MONTGOMERY, R., 1957. Determination of glycogen. Arch. Biochcm. Biophys., 67 : 378-386. 572 THOMAS H. DIETZ MOON, T. W. AND A. W. PRITCHARD, 1970. Metabolic adaptations in vertically-separated populations of Mytihts californianus Conrad. /. Exp. Biol. Ecol., 5: 35-46. MURRAY, H. D., AND A. B. LEONARD, 1962. Handbook of Unionid Mussels in Kansas. Mis- cellaneous Publications University Kansas Museum Natural History, Lawrence, 184 pp. MUSTAFA, T. AND P. W. HOCHACHKA, 1973. Enzymes in facultative anaerobiosis of molluscs —III. Phosphoenolpyruvate carboxykinase and its role in aerobic-anaerobic transition. Comp. Biochcm. Physiol, 45 B : 657-667. NEWELL, R. C., 1970. Biology of intcrtidal animals. American Elsevier, New York, 555 pp. POTTS, W. T. W., 1954. The inorganic composition of the blood of Mytilus cdulis and Anodonta cygnca. J. Exp. Biol., 31 : 376-385. PROSSER, C. L., 1973. Comparative Animal Physiology. Saunders, Philadelphia, 966 pp. ROSEN, H., 1957. A modified ninhydrin colorimetric analysis for amino acids. Arch. Biochem. Biophys., 67 : 10-15. STOKES, T. M. AND J. AWAPARA, 1968. Alanine and succinate as end-products of glucose de- gradation in the clam, Rangia cuncata. Comp. Biochcm. Physiol., 25 : 883-892. STRICKLAND, J. D. H. AND T. R. PARSONS, 1972. A Practical Handbook of Seau'ater Analysis. Fisheries Research Board of Canada, Ottawa, 310 pp. STROM, G., 1949. The influence of anoxia on lactate utilization in man after prolonged muscu- lar work. Ada Physiol. Scand., 17 : 440-451. Reference: Biol. Bull., 147: 573-585. (December, 1974) LARVAL SETTLEMENT OF A SYMBIOTIC HYDROID : SPECIFICITY AND NEMATOCYST RESPONSES IN PLANULAE OF PROBOSCIDACTYLA FLAVICIRRATA SYEN DONALDSON Friday Harbor Laboratories, Friday Harbor, U'asliington 98250 Planktonic larvae of many benthic invertebrates preferentially settle and meta- morphose on specific substrates (Williams, 1964; Newell, 1970). Frequently, the planktonic larvae of symbionts possess exceptionally well-developed physiological mechanisms for substrate selection. Such mechanisms presumably evolved be- cause, for these species, only a small proportion of the surfaces that a larva might contact are suitable for settlement (Davenport, 1955). I have studied planulae of the symbiotic hydroid Proboscidactyla flavicirrata (Brant). The structure of a planula larva is extremely simple, yet the planulae of Proboscidactyla settle with great specificity (Campbell, 1968). Sabellid polychaetes are the only known hosts of Proboscidactyla (Gosse, 1857; Uchida and Okuda, 1941; Hand, 1954; Brinkmann and Vannucci, 1965; Calder, 1970). Proboscidactyla colonies are found on the rims of sabellid tubes where their gastrozooids can contact the host's plume of tentacular cirri. According to Campbell (1968), Proboscidactyla planulae begin settlement by adhering with nematocysts to the pinnules that branch from the tentacular cirri. Later, when the sabellid retracts its tentacular cirri into its tube, the planulae somehow transfer to the rim of the tube where they metamorphose. In this report I confirm and extend Campbell's work. Two new findings are of particular importance. First, planulae use nematocysts both to attach to pinnules as Campbell (1968) suggested and to transfer to the tube rim. Between the time of attachment to a pinnule and transfer to the tube rim, the effective stimulus for nematocyst discharge shifts from pinnule to worm tube. Secondly, planulae settle and metamorphose on either of two sympatric species of sabellids, one of which does not support Proboscidactyla colonies. MATERIALS AND METHODS During June, July, and August of 1973, the medusae of Proboscidactyla were collected from surface waters, and the sabellids Schisobranchia insignis and Eudistylia vancouvcri from the undersides of mooring floats near the University of Washington Friday Harbor Laboratories. Nearly mature Proboscidatyla medusae sometimes ripened when fed Artemia larvae. The medusae, embryos, and planulae were kept in glass finger bowls or Syracuse dishes partially submerged in an open- circulation sea table and the water in the holding containers was changed at least every third day. The medusae shed their gametes at night and developing embryos were rou- tinely transferred to Syracuse dishes the next morning. Zygotes were sometimes 573 574 SVEN DONALDSON collected during the night to examine early development. Individual planulae, severed sabellid pinnules, and fragments of sabellid tubes were handled using tygon suction micropipettes (Josephson, 1965) held in micromanipulators. The micrographs illustrating this report were made with incident illumination using either a Wild M-5 stereo microscope, or a Zeiss photomicroscope equipped with Nomarski differential interference contrast optics. RESULTS Development Cleavage in Proboscidactyla eggs was holoblastic, unilateral, and equal or nearly equal. Cleavages 2 through 7 occurred at approximately 45 min intervals (14° C). A blastula had formed by 6 hr after fertilization. After about 10 hr individual blastomeres began to migrate inward to form the entoderm. In the majority of the embryos this migratory activity was basically unipolar, although small groups of cells sometimes ingressed away from the main site of gastrulation. In a few embryos, migratory cells were distributed almost uniformly over the inner surfaces of gastrulae. Nevertheless, as gastrulation progressed, the entoderm always thickened toward the pole that eventually became the posterior end of the swimming planula (see Fig. 1 ) . Cilia first appeared 14 hr after fertilization and at about 16 hr the planulae began to crawl and swim. At approximately the same time the planulae started to change shape, alternately becoming ovoid and nearly spherical. Planulae swam and crawled only when in the ovoid shape. Those crawling along the bottoms of dishes advanced without rotating, but swimming planulae always rotated slowly on their long axes in a clockwise direction when viewed from the anterior end. Accord- ing to Widerstein (1968) such clockwise rotation is a common feature in cnidarian larvae. The Cnidome Cnidocils began to protrude from the epidermis about 48 hr after fertilization. During the 24 hr that followed, their numbers gradually increased, presumably as more and more nematocysts matured or reached peripheral locations. During this period of development the planulae were periodically touched with severed Schizo- branchia pinnules to determine whether they had acquired the ability to attach. Prior to the appearance of cnidocils about 36 hr after fertilization, planulae never adhered to pinnules even momentarily. About 48 hr after fertilization, while the cnidocils were still comparatively sparse, planulae sometimes adhered but failed to remain attached for more than a few seconds. After 72 hr the density of cnidocils on the surface appeared to have reached its maximum. Planulae that touched Schizobranchia pinnules more than 72 hr after fertilization always attached and usually remained attached even if jostled vigorously. Peripheral nematocysts in fully developed planulae were distributed almost uni- formly over the ectoderm (Fig. 1). No concentration of nematocysts in the ectoderm of the anterior end comparable to that reported by Campbell (1968) was HOST-SPECIFIC PLANULA SETTLEMENT 575 FIGURE 1. Protruding cnidocils distributed fairly uniformly in the epidermis of a fully developed planula. The cavity within the planula appears as a lighter area centered nearer the anterior end (left). The entoderm is correspondingly thicker toward the posterior end; scale bar — 50 /A. FIGURE 2. Optical section of a planula showing a concentration of nematocyst capsules in the posterior entoderm (bottom third). (Fig. 2-5 are all to the same scale and the nematocyst capsules depicted in these figures all appear similar) ; scale bar = 25 //.. FIGURE 3. Epidermal distortion formed by pulling an adhering planula away from a frag- ment of sabellid tube ; scale bar — 25 ja. FIGURE 4. Nematocyst discharged into a fragment of sabellid tube. A portion of the em- bedded thread can be seen leading down and left from the capsule ; scale bar — 25 //.. FIGURE 5. Discharge epidermal nematocysts ; scale bar = 25 t*. 576 SVEN DONALDSON apparent. Nematocyst capsules were present in the entoderm as well, mainly toward the posterior end where the entoderm was thickest (Fig. 2). Mechanism of adhesion In nature, a Proboscidactyla planula contacts 2 different substrates during the course of settlement. Initially it adheres to a sabellid pinnule (attachment). While still attached to the pinnule it adheres to the rim of the sabellid's tube (the initial event in transfer). The simultaneous acquisition of cnidocils and attachment ca- pability suggested that nematocysts are responsible for adhesion to pinnules. This was comfirmed by pulling attached planulae off pinnules with micromanipulators. As planula and pinnule separated, areas of the planula epidermis stretched into cones attached to the pinnule at their apices (Fig. 3). At each apex was a dis- charged nematocyst with its thread embedded in the pinnule. Similar experiments performed with planulae that had transferred from pinnules to fragments of sabellid tube produced identical results. Nematocyst capsules left upon pinnules and tube fragments were also observed after planulae had been dislodged from their sites of adhesion. If the site was the edge of a thin piece of tube the embedded thread could sometimes be seen as well as the capsule (Fig. 4). Although each planula possessed over a hundred epidermal nematocysts, only a few (usually less than 5) were discharged during an adhesion. This suggested that nematocysts of planulae do not discharge unless the corresponding cnidocils contact sabellid sur- faces. Planulae released both everted and undischarged nematocysts when flattened between slide and coverslip. The nematocysts were of a single type which did not conform clearly to any described for the family Proboscidactylidae (Russell, 1938; Hand, 1954; Calder, 1970). They appeared to be basatrichous haplonemes (Fig. 5). The nematocyst capsules left on severed pinnules and sabellid tube fragments appeared identical to those released from flattened planulae. Behavior prior to attachment Planulae that contacted severed Schizobranchia pinnules attached to the pinnules immediately. However, they often swam or crawled along paths that were almost, but not quite, tangential to severed pinnules without altering their direction or velocity as they passed the pinnules. Campbell (1968) observed that in the vicinity of an intact, extended sabellid, the currents passing through the crown of tentacular cirri greatly exceed the swimming velocity of Proboscidactyla planulae. My ob- servations of planulae near severed pinnules and Campbell's result both sug- gest Proboscidactyla planulae take no active role in initiating contact with a suit- able host. Stimulus jor attachment To determine whether mechanical stimulation caused by contact with a ciliated surface might trigger nematocyst discharge and adhesion, a freshly excised portion of the branchial basket from a tunicate, Chelyosoma productnni, and a piece of the HOST-SPECIFIC PLANULA SETTLEMENT 577 lamellae from a scallop Chlamys hastata, were placed in each of 2 Syracuse dishes containing large numbers of planulae (about 250 in the first trial and 100 in the second). No planulae adhered to the ciliated ascidian or mollusc tissues. When severed Schizobranchia pinnules were added, planulae attached to the pinnules in large numbers. The presence of active cilia on a surface apparently does not, in itself, elicit nematocyst discharge. Unattached planulae in crowded Syracuse dishes also did not adhere to frag- ments of Schizobranchia tube, although they frequently encountered the fragments. Such planulae failed to adhere even when squeezed lightly between 2 fragments of tube or a fragment of tube and the bottom of this dish. On the other hand, pieces of the collar folds, short lengths of contiguous body segments, and isolated portions of the ventral shields of Schizobranchia placed one at a time in Syracuse dishes containing planulae appeared to stimulate attachments as effectively as pinnules. The great specificity of the attachment response suggest that a chemical stimulus found only on the body surfaces of sabellids is necessary to elicit nemato- cyst discharge in unattached Proboscidactyla planulae. I was unable to distinguish whether this chemical stimulus is an insoluble surface coating on the pinnules that the cnidoblasts detect only if they contact it directly, or a diffusable substance that triggers nematocyst discharge only in conjuction with a mechanical stimulus. Simple mechanical stimuli applied in the presence of mucus solutions or tissue "juices" from Schizobranchia failed to elicit a nematocyst response, but these solu- tions may have had toxic qualities that inhibited a nematocyst response. Attachment to pinnules of a non-host sabellid species Near Friday Harbor, Schisobranchia insignis is the most abundant sabellid species. Eudistylia vancouveri is morphologically similar to Schizobranchia, fairly common, and occurs in the same areas. However, Eudistylia tubes never bear Proboscidactyla colonies. Colonies can be successfully exchanged from one Schizobranchia specimen to another, but they regress when transplanted to Eudi- stylia (Strickland, 1971). Experiments comparing the tendencies of planulae to settle on Schizobranchia and Eudistylia pinnules revealed a clear preference for Schizobranchia although Eudistylia regularly stimulated attachment also (Fig. 6A and B). In A, the separate-dish design, 25 planulae and 10 Schizobranchia pin- nules were placed in each of 4 culture dishes 2 cm in diameter. Four similar dishes were prepared with Eudistylia, rather than Schizobranchia, pinnules. In B, the competitive design, 25 planulae, 5 Schizobranchia pinnules, and 5 Eudistylia pinnules were placed together in each of 3 dishes. In both experiments the number of planulae that had attached to the Schizobranchia pinnules by 1, 2, 4, and 6 hr was significantly higher at the 99 % confidence level (minimum value of t in Student test was 4.4 with 5 degrees of freedom). Nevertheless, in every trial or trial-pair some planulae attached to Eudistylia pinnules. Elongation following attachment Shortly after attaching to a pinnule, Proboscidactyla planulae lengthened along their longitudinal axes to a much greater extent than unattached planulae ever did 578 SVEN DONALDSON 41 - w 20 15 10 o 6 2 0 6A 6B 2 4 1 Ion is 0 2 4 1 lou rs 6 FIGURE 6. Experiments comparing planula settlement on pinnules from Schizobranchia and Edistylia; (A) separate dish design, (B) competitive design. Time elapsed following the addition of 25 planulae to each trial dish is plotted on the abscissas. The ordinates designate the mean number of planulae attached in all trials for each experimental design. (Fig. 7). During elongation the cavity within the planula narrowed and some- times nearly disappeared. Simultaneously the greatest diameter of the elongating planula shifted from anterior to posterior. The greater thickness of the posterior entoderm probably accounted for the posterior bulge of maximally elongated planulae (Fig. 7). The orientation of attachment (anterior, posterior, or lateral) had no visible effect on the shape of a planula upon elongation. Data were obtained on the time between attachment and complete elongation for 25 planulae. Elongation required from 1.3 to 11 min (x = 4 min, S.D. — 2.7). 8 FIGURE 7. Elongated planulae after 12 hr adhesion to the fragment of tentacular cirrus held by the suction pipette on the right. The planula on the left had been attached for 1 min and had not yet elongated when the photograph was made ; scale bar = 0.5 mm. FIGURE 8. Newly metamorphosed Proboscidactyla polyps on the rim of a Eudistylia tube ; scale bar = 0.5 mm. HOST-SPECIFIC PLANULA SETTLEMENT 579 The planulae that attached to severed Eudistylia pinnules during the inter-species comparison experiments (Fig. 6) elongated, but did so more slowly than planulae attached to Schizobranchia pinnules. Once elongated, attached planulae remained so in excess of 24 hr. By this time severed Schisobranchia pinnules had largely decomposed, although severed Eudistylia pinnules remained in good condition. Planulae attached to the intact pinnules of a small Schisobranchia specimen (removed from its tube to prevent transfers to the tube rim) remained elongated in excess of 48 hr. These planulae and those that remained attached to pinnules instead of transferring to tubes never metamorphosed. Controls which completed their transfers almost always began metamorphosis within 12 hr. Elongation ap- peared to be associated exclusively with attachment to pinnules. Following trans- fer, planulae always rounded up within a few minutes. Transfer appears to initiate planula metamorphosis. Desensitization to pinnules following attachment Planulae swept into the crowns of intact sabellids only occasionally attached to more than one pinnule, although the curling and writhing of the worm's crown frequently brought the newly-attached planula into contact with several pinnules. Fifty-two planulae that had been attached to severed pinnules for 10-12 hr all failed to adhere when maneuvered into contact with freshly severed tentacular cirri from Schizobranchia, although 40 of them later transferred to tube fragments (Table 1). On 5 occasions individual planulae were touched with a second pinnule at 0.5 min intervals after attachment. Two planulae failed to adhere to the second pinnule on the first trial at 0.5 min. Two others adhered at 0.5 min and 1.0 min, but not at 1.5 min. None of these planulae had elongated appreciably before becoming desensitized to pinnule contacts. (About 1.3 min was the minimum time required to achieve full elongation in any of the 25 planulae timed). The fifth planula adhered to a proffered pinnule at 2.0 min, but not at 2.5 min. It required even longer, 9 min, to elongate and did not elongate completely. Thus desensitization occurs before elongation and the minimum time to desensitization is approximately 0.5 min. TABLE I Comparison of Schizobranchia pinnules and tube surfaces as stimuli for the adhesion of attached planulae. Order of test surface application in each Number of planulae tested Xumber of planulae adhering to each test surface (Total 40) Number of planulae failing trial series series (total 52) pinnule inside mid-layer outside (total 12) pinnule, inside, outside 30 0 5 18 7 pinnule, outside, inside 9 0 0 7 2 pinnule, mid-layer, outside 3 0 0 2 1 pinnule, outside, mid-layer 10 0 1 7 2 580 SVEN DONALDSON Twenty-one planulae pulled off pinnules after 10-12 hr of attachment shortened up and resumed swimming after approximately 5 mm. Within 10 min all of these planulae had recovered the ability to attach to Schisobranchia pinnules. Pinnule contact apparently must be maintained to prevent planulae from discharging nemato- cysts in response to new pinnule contacts. Acquisition of a ncmatocyst response to worm tube After becoming desensitized to pinnule contacts, attached Proboscidactyla planulae eventually begin to adhere when brought into contact with sabellid tube material. This change in effective stimulus for nematocyst discharge will be termed respecification. Under natural conditions attached planulae can only contact worm tube when the worm withdraws its crown of tentacular cirri. To estimate the time required for a newly attached planula to become responsive to worm tube I touched individual planulae to fragments of worm tube at intervals ranging from 1 min to 45 min after attachment. Out of a total of 20 transfers, the 2 earliest occurred 4 and 6 min after attachment. An attempt to obtain a more accurate estimate of the time required for re- specification was made by touching individual planulae to worm tube fragments at 2 min intervals, starting 2 min after attachment. However, the planulae subjected to these manipulations usually came off their pinnules or failed to adhere to the tube fragment within 30 min. Table II displays data from the 6 trials in which planulae adhered to tube fragments. Some of the first adhesions were not tenacious enough to result in a transfer, but eventually 5 out of the 6 planulae transferred. The one exception was injured accidentally before its 10 min trial. None of these planulae discharged nematocysts until they had been attached for at least 6 min. All 6 planulae had been elongated for at least 3 min before first adhering to a worm tube. These and other observations suggested that elongation was associated with, and ordinarily preceded, respecification. However, 4 planulae that were sub- stantially less than fully elongated nevertheless transferred successfully. Re- specification and elongation may begin with the same stimulus or depend upon a common physiological event, but complete elongation is not a prerequisite for transfer. TABLE II Time requirement for respecification. Results of trials on tube at 2 min intervals Klapsed time between attachment (min after attachment) and elongation of each of 6 planulae (min) Adhesion without transfer Adhesion followed by transfer 2 6 8 2.5 8, 10, 12 18 11 14 3.5 S 1.5 12 2.5 6 HOST-SPECIFIC PLANULA SETTLEMENT 581 Orientation of attachment and transfer In many, but not all, of the observed attachments, a point at or to one side of the anterior end of the planula adhered to the pinnule. However, by using a manipulator to maneuver the severed pinnules posterior attachments were readily obtained. Promptly after contact, planulae attached by the pointed posterior end often toppled over and adhered by one side, which suggests that physical instability may account for the comparative rarity of posterior attachments. Planulae that remained attached by the posterior and elongated normally, the enlarged end being the posterior as usual. On 6 occasions I noticed elongated planulae that had at- tached by their posterior ends after being swept into the plumes of intact sabellids. Although many planulae, having attached by their anterior ends, adhered by their posterior ends during transfer, again this was not always the case. Often they adhered to worm tube fragments by one side. Four planulae, having attached to pinnules by their posterior ends, subsequently adhered to tube fragments by their anterior ends and transferred. The absence of a requirement for a particular orientation during either elongation or transfer suggests either that two popula- tions of nematocysts are intermingled, or that nematocysts from all parts of the epidermis are functionally identical. Stimulus for transfer The data presented in Table I suggest that the outside of Schizobranchia tubes stimulates transfer far more effectively than does the inside or a middle layer. In obtaining these results the responsiveness of each attached planula to pinnule contacts was first tested by touching it to a short section of isolated Schizobranchia tentacular cirrus. Having obtained a negative response, the planula was then touched to either the inner surface of a Schizobranchia tube fragment or a middle layer of a "peeled" tube fragment. If the planula failed to transfer it was touched to the outer surface of another fragment of the same tube. With other planulae the same stimuli was applied in reverse order. To determine whether attached and respecified Proboscidactyla planulae might adhere relatively non-selectively, 3 non-sabellid substrates from the mooring float community were tested: the tunic of CheHosoma production, test of the ectoproct Membranipora tuberculata, and an empty barnacle shell removed from the outside of a sabellid tube. None of the 5 planulae tested on each of the 3 non-sabellid substrates adhered, although all 15 subsequently transferred when touched to a fragment of Schiso- brancliia tube. The stimulus for transfer, whether a film of microorganisms or not, appeared to be exclusively a property of sabellid tube material. Individual planulae successfully transferred to fragments of Eudistylia tube on numerous occasions, sometimes after attachment to severed Eudistylia pinnules. The behavior of planulae that attached to the pinnules of small intact specimens of Eudistylia was indistinguishable from that of control planulae on Schizobranchia; they elongated, transferred, and metamorphosed into minute one-tentacled polyps (Fig. 8). Clearly Eudistylia, although not a host for Proboscidactyla, can induce 582 SVEN DONALDSON Proboscidactyla planulae to settle whereas other surfaces found in the same localities cannot. .1 DISCUSSION In comparative settling experiments, Schizobranchia pinnules were preferred to Eudistylia pinnules, hut Proboscidactyla planulae attached to pinnules of both sabellid species. Once attached, the responses of planulae to Eudistylia were qualitatively indistinguishable from the responses elicited by Schizobranchia. None of the other test substrates induced planulae to adhere even momentarily. Planulae metamorphosed on intact specimens of Eudistylia in the laboratory and there is no reason to suspect that they fail to do so in the held. Apparently the deficiencies of Eudistylia as a host for Proboscidactyla are not manifested until after metamor- phosis. Nematocyst responses play a central role in the settlement of Proboscidactyla planulae and constitute a major part of the behavioral repertoire of these larvae. Free-swimming planulae ordinarily discharge nematocysts only upon contact with the pinnules of a sabellid. Contact with non-sabellid, ciliated surfaces does not elicit nematocyst discharge, although the mechanical stimulus is probably com- parable to that provided by sabellid pinnules. However, contact with any body surface of Schizobranchia results in attachment. Eudistylia, although closely re- lated to Schizobranchia and morphologically very similar, is less effective in stimu- lating planulae to attach. Following attachment and respecification, planulae again respond specifically, this time discharging nematocysts upon contact with the outer surfaces of sabellid tubes. These results suggest that specific chemical stimuli are needed to elicit nematocyst discharge in Proboscidactyla planulae. Planulae discharge nematocysts individually at sites of contact rather than en masse. This indicates either that direct contact with a chemically stimulating surface is required to excite a cnidoblast, or that mechanical as well as chemical stimulation is needed. The first possibility has never been described, but the second, dual stimulation, has been demonstrated in some cnidarians by applying chemical and mechanical stimuli separately and in combination (Pantin, 1942; Jones, 1947; Burnett, Lentz, and Warren, 1960). I was unable to excite the nematocysts of Proboscidactyla planulae with simple mechanical stimuli in the presence of mucus solutions or tissue "juices" from Schisobranchia. Therefore it is conceivable that contact chemoreception alone mediates the nematocyst response in this case, although these negative results do not constitute a conclusive demon- stration. After about 30 sec of attachment to a pinnule, a Proboscidactyla planula be- comes desensitized to additional pinnule contacts. Subsequently, the nematocyst response is respecified, sabellid tube replacing pinnule as the effective stimulus. Intercellular controls of nematocyst function must be postulated to explain desensi- tization and respecification. There may be two functional classes of nematocysts, one responsive to pinnule and the other to worm tube. Because any part of a planula can adhere to either pinnule or worm tube, it follows that if two types of nematocysts exist, they must both be distributed uniformly over the epidermal surface. Prior to attachment, the pinnule nematocysts would be ready to discharge HOST-SPECIFIC PLANULA SETTLEMENT 583 while the tube nematocysts would be inhibited. After attachment, owing to changes in physiological state throughout the cnidoblast population, the tube nematocysts would become readied and the pinnule nematocysts inhibited. Al- ternatively, there may be only one type of nematocyst in Proboscidactyla planulae. Physiological events associated with attachments to a pinnule would respecify the remaining cnidoblasts to discharge only to worm tube. In the nematocyt literature the term "independent effector" has been used in different ways. Early authors used the term specifically to indicate that nemato- cysts are not directly excited by the nervous system (Parker and Van Alstyne, 1932; Pantin. 1942; Jones, 1947). More recently authors have described examples of endogenous nematocyst controls operating through unidentified physiological pathways as evidence against the independent effector hypothesis (Davenport, Ross, and Sutton, 1961; Ross and Sutton, 1964; Sandberg, Kanciruk, and Maris- cal, 1971; Oshida, 1972). It is possible that the nematocysts of Proboscidactyla planulae are independent effectors in the original restricted sense, although their function is clearly subject to endogenous controls. Although innervation of some cnidoblasts has been histologically demonstrated (Westfall, 1970; Westfall, 1971), the settlement behavior of Proboscidactyla planulae can be explained without assuming that such innervation exists. Desensi- tization probably requires a minimum of about 30 seconds to complete and some- times takes over 2 minutes. Respecification requires several minutes. Nervous (or epithelial) conduction would transmit the message of attachment from one end of the 250 JJL planula to the other in a few milliseconds. If nervous conduction is involved, there presumably are slow steps at the receptor sites, cnidoblasts, or both to account for the observed delays. On the other hand, slow conveyance of in- formation by a non-nervous process might largely account for the considerable time lag before desensitization and. in combination with a slow step in the cnido- blasts, for the still greater delay prior to respecification. Both desensitization and respecification are reversible changes maintained only by continuous contact with a sabellid pinnule. This suggests that information is continuously transferred from the attachment site. It is possible that a metabolite diffuses from the attachment site through the tissues of the planula and triggers the sequence of events that follow attachment. Planulae that attached and did not transfer remained elongated for extended periods of time, but never metamorphosed. After transferring to the rims of sabellid tubes, planulae that had previously been attached to pinnules promptly rounded up and subsequently commenced metamorphosis. These observations indicate that the stimulus for metamorphosis is not the same as the stimulus for attachment, desensitization, elongation, or respecification. The stimulus that initi- ates metamorphosis and the stimulus for adhesion and transfer to sabellid tube could be one and the same. Elongation, while not a prerequisite for transfer, may be a behavioral adaptation that increases the probability of transfer by increasing the likelihood of contact with the sabellid tube rim. 584 SVEN DONALDSON The facilities of Friday Harbor Laboratories were important to this study and I thank the director of the station, Dr. A. O. D. Willows, and its staff for their help. I also thank Drs. G. O. Mackie, M. Paul, L. Francis, and Ms. L. Atherton for reading the manuscript and making valuable suggestions. This research was supported by the National Research Council of Canada through operating grant A 1427 to Dr. Mackie and a Graduate Scholarship. SUMMARY Planulae of the symbiotic hydroid Proboscidactyla flavicirrata settle on the rims of sabellid tubes by adhering sequentially and specifically to two different sub- strates. For both steps, nematocysts are the agents of adhesion. Although the cnidoblasts of these planulae require local excitation, they are subject to endogenous controls. Upon being swept into the plumes of tentacular cirri that sabellids extend from their tubes, planulae attach to individual pinnules (branches of the tentacular cirri) by discharging nematocysts. Within 2 min following attachment, the planulae cease discharging nematocysts in response to additional pinnule contacts. Planulae transfer to the rims of sabellid tubes by adhering with nematocysts when the sabellids retract their plumes. Cnidoblasts of planulae only become responsive to contact with sabellid tube after 4 min or more of continuous attachment to pinnules. Transfer initiates metamorphosis. Planulae are capable of settling and metamorphosing on Eudistylia vancouveri, a sabellid species that is sympatric with the normal hosts of Proboscidactyla, but will not support colonies. Apparently the deficiencies of Eudistylia as a host are manifested only after planula metamorphosis. LITERATURE CITED BRINKMANN, A. AND M. VANNUCCI, 1965. On the life cycle of Proboscidactyla oninta ( Hy- dromedusae, Proboscidactylidae) . Pitbl. Sta. Zool. Napoli, 34 : 357-365. BURNETT, A. L., T. LENTZ, AND M. WARREN, 1960. The nematocyst of hydra (part 1). The question of control of the nematocyst discharge reaction by fully fed hydra. Ann. Soc. Roy. Zool. Bclg., 90: 247-267. CALDER, D. R., 1970. North American record of the hydroid Proboscidactyla ornata (Hy- drozoa, Proboscidactylidae). Chesapeake Sci., 11 : 130-132. CAMPBELL, R. D., 1968. Host specificity, settling, and metamorphosis of the two-tentacled hydroid Proboscidactyla flavicirrata. Pac. Sci., 22 : 336-339. DAVENPORT, D., 1955. Specificity and behavior in symbioses. Quart. Rev. Biol., 30: 29-46. DAVENPORT, D., D. M. Ross, AND L. SUTTON, 1961. The remote control of nematocyst dis- charge in the attachment of Calliactis parasitica to shells of hermit crabs. Vic Milieu, 12 : 197-209. GOSSE, P. H., 1857. On a new form of corynoid polypes. Trans. Linn. Soc., 22 : 113-116. HAND, C, 1954. Three Pacific species of "Lar" (including a new species), their hosts, medusae, and relationships. (Coelenterata, Hydrozoa). Par. Sci., 8: 51-67. JONES, C. S., 1947. The control and discharge of nematocysts in hydra. /. Exp. Zool., 105 : 25-60. JOSEPHSON, R. K., 1965. Three parallel conducting systems in the stalk of a hydroid. /. Exp. Biol, 42: 139-152. NEWELL, R. C., 1970. Biology of Intcrtidal Animals. American Elsevier Publishing Co., New York. HOST-SPECIFIC PLANULA SETTLEMENT 585 OSHIDA, J. W., 1972. Control of stenotele discharge in hydra as demonstrated through normal feeding activity. /. Undergrad. Res. in Biol. Sci. Univ. Calif. Irvine, 2 : 161- 173. PANTIN, C. F. S., 1942. The excitation of nematocysts. /. Exp. Biol., 19: 294-310. PARKER, G. H. AND M. A. VAN ALSTYNE, 1932. The control and discharge of nematocysts, especially in Metridium and Physalia. J. £.r/>. Zoo/., 63: 329-344. Ross, D. M. AND L. SUTTON, 1964. Inhibition of the swimming response by food and of nematocyst discharge during swimming in the sea anemone Stomphia coccinca. J. Exp. Biol., 41 : 751-757. RUSSELL, F. S., 1938. On the nematocysts of hydromedusae. /. Mar. Biol. Ass. U. K. 23 : 145-165. SANDBERG, D. M., P. KANCIRUK, AND R. N. MARISCAL, 1971. Inhibition of nematocyst dis- charge correlated with feeding in a sea anemone, Calliactis tricolor (Leseur). Nature, 232 : 263-264. STRICKLAND, D. L., 1971. Differentiation and commensalism in the hydroid Proboscidactyla flavicirrata. Pac. Sci., 25 : 88-90. UCHIDA, T. AND S. OKUDA, 1941. The hydroid Lar and the medusa Proboscidactyla. J. Fac. Sci. Hokkaido Univ., Scr. VI Zoo/., 7: 431-440. WESTFALL, J. A., 1970. Ultrastructure of synapses in a primitive coelenterate. /. Ultrastruct. Res., 32 : 237-246. WESTFALL, J. A., 1971. Ultrastructural evidence of polarized synapses in the nerve net of Hydra. /. Cell Biol., 51 : 318-323. WIDERSTEIN, B., 1968. On the morphology and development in some cnidarian larvae. Zoo/. Bidr. Uppsala, 37 : 139-185. WILLIAMS, G. B., 1964. The effect of extracts of Fuctis scrnitus in promoting the settlement of larvae of Spirorbis borcalis (Polychaeta). J. Mar. Biol. Ass. U. K.. 44: 397-414. NOTE ADDED IX I'KOOF The identification of the planula nematocysts given in this report is in error. They are microbasic euryteles not basitrichs. Reference: />;W. />'////., 147: 586-593. (December, l'»74 CYTOPLASMIC DNA IN SEA URCHIN OOGENESIS STUDIED BY 3H-ACTINOMYCIN D BINDING AND RADIOAUTOGRAPHY H. ESPER ENESCO AND K. H. MAN Department of Biological Sciences, Sir George Williams University, Montreal. Quebec 43G IMS Canada This study is concerned with the time at which cytoplasmic DNA of the sea urchin oocyte replicates in relation to the stage of oocyte development. Oocytes at many different states of development, as well as mature eggs, are found in mature sea urchin ovaries. They range in size from the smallest primary oocyte to the largest terminal oocytes. Since all different stages of oocyte development are present in the ovary, the events of oogenesis may he studied in a sequential manner. The method chosen to detect the presence of cytoplasmic DNA in relation to the stage of oocyte development was 3H-actinomycin D incubation of sectioned ma- terial, followed by radioautography. It is well established that actinomycin D binds specifically to DNA (Muller and Crothers, 1968) and that this binding occurs on tissue sections (Ebstein, 1967, 1969). Once the 3H-actinomycin D binds specifi- cally to DNA on the section, the location of the 3H-actinomycin D-DNA complex can he detected by radioautography. We thus have a method for localizing all nuclear and cytoplasmic DNA in tissue sections in relation to cell structures. This method allows us to compare the amount of cytoplasmic DNA detected by 3H- actinomycin D binding at progressive stages of oocyte development. MATERIALS AND METHODS Living supplies of the purple sea urchin, Strongyhcentrotits pnrpnratus, were obtained from Pacific Biomarine Company, Venice, California. Ovaries of mature females were fixed in Carnoy's solution (ethanol : acetic acid, 3:1) for two hours. They were passed through graded alcohols, cleared in toluene and embedded in paraffin. Sections were cut at 5 microns, deparaffmized in xylol and hydrated through a graded alcohol series. Control enzyme and TCA extraction oj /AY. / and h'NA Controls were necessary to check the specificity of the 3H-actinomycin D bind- ing under the experimental conditions used. Controls consisted of slides from which all DNA had been extracted prior to incubation with 3H-actinomycin D, either by treatment of the slides with DNase, using the method of Deitch (1966) or by extraction with hot trichloroacetic acid (TCA), using the method of Pearse (1961). As an additional control, RNase extractions, using the method of Amano (1962) were performed on another series of slides. The specificity of the above 586 CYTOPLASMIC DNA IN OOGENESIS 587 extraction procedures was confirmed by Azure B staining for DNA and RNA using the cytochemical staining method of Flax and Himes (1952). Incubation zuith 3H-actinomycin D To determine the location of DNA in the oocytes, slides were incubated in 3H-actinomycin D. The method of Ebstein ( 1967, 1969) as modified by Aczel and Enesco (1973) was used. This method consists of incubating the slides with 3H-actinomycin D (5 /iCi/ml) for two hours in a moist chamber in the dark, then rinsing the slides in distilled water and in a graded alcohol series to 70% ethanol to remove any 3H-actinomycin D which is not specifically bound to DNA. The 3H-actinomycin D (specific activity 3.7 Ci/niM ) was obtained from Schwartz- Mann Company, Orangeburg, New York. Radioautography and staining Slides were air dried following 3H-actinomycin D incubation to prepare them for radioautography. The slides were then dipped in Kodak NTB2 liquid nuclear track emulsion, using the method of Kopriwa and Leblond (1962). After sufficient exposure (10 days) they were developed in D170 developer (Young and Kopriwa, 1964). The slides were then stained following radioautographic development with either cresyl violet or toluidine blue. The staining procedures followed were those selected by Thurston and Jofts (1963) for their compatibility with the radioautographic emulsion. After staining, the slides were dehydrated and mounted in Permount. Staging of oocytes The dimensions of the oocyte nucleus and cytoplasm as sectioned through their largest diameter were measured with the aid of a calibrated ocular micrometer. On the basis of these measurements, oocytes were classified into ten numerical stages, as described in the results section. Numerical stages 1 and 2 correspond to primary oocytes, stages 3 to 6 are growing oocytes and stages 7 to 10 are terminal oocytes using the classification defined by Cowden (1962) and by Esper (1965). At least 10 measurements were made to define each stage. Histonietrics and grain counts Oocytes sectioned through their largest diameter were selected for this study. The nuclear and cytoplasmic diameter of these oocytes wras measured, and the nu- clear and cytoplasmic area was calculated. Using the measured values of nuclear and cytoplasmic diameter, nuclear volume and total cell volume were calculated. Cyto- plasmic volume was then obtained by subtracting nuclear volume from total cell volume. The number of grains per lOOOfi2 cytoplasmic area was determined on the same sections used for histometrics. The grain counts, originally expressed in relation to area, were then transformed to express the total number of grains ex- pected in relation to total cytoplasmic volume. 588 H. E. ENESCO AND K. H. MAN TABLE 1 Grain counts per 1000 i^- over nucleus and cytoplasm at 3 stages of oocyte development, corrected by subtraction of background fog. The amount of radioactivity in non-extracted, RNase, DNase and TCA extracted tissue is presented. The data for stages 4 to 10 is essentially the same. Stage No extraction RNase extraction DNase extraction TCA extraction X iicleus 1 266 282 8 2.9 2 200 203 5.9 5.4 3 130 153 0 0.5 Cytoplasm 1 66 77 4.5 1.7 2 60 78 3.2 5.7 3 43 48 2.8 1.4 RESULTS It was first necessary to establish that 3H-actinomycin D was binding specifi- cally to DNA under our experimental conditions. As would be expected, radio- autographic examination showed that :;H-actinomycin D labelling was heavy over nuclei at all stages of oocyte development. Label appeared over the cytoplasm as well. Evidence of specific 3H-actinomycin D binding to DNA in both nucleus and cytoplasm of the sea urchin oocytes is presented in Table I. This table presents the number of silver grains per 1000 p- over nuclei and over cytoplasm in the radioautographs, which in turn represents the amount of 3H-actinomycin D binding to DNA. As control data, grain counts from slides extracted with RNase, DNase, and hot TCA are presented. The grain count data for the RNase ex- tracted slides is not significantly different from the non-extracted slides. Thus, RNase extraction does not influence 3H-actinomycin D binding. In contrast, on slides where all DNA has been removed by extraction with DNase or with hot TCA, the grain counts were reduced to very low levels, not significantly above background. This data clearly indicates that there is no 3H-actinomycin D bind- ing in the absence of DNA. The effectiveness of the RNase, DNase and TCA extractions was further confirmed by the absence of selective staining when slides treated by these extraction methods were stained with Azure B bromide, which stains DNA and RNA selectively. The pattern described above shows that no 3H-actinomycin D binding takes place when all DNA has been removed from the sections. Table I presents representative grain counts from stages 1 to 3. The data for stages 4 to 9 follow exactly the same pattern. Table II presents the direct measurements of nuclear and cytoplasmic area used to define the 10 stages of oocyte growth. From this data, the total cytoplasmic volume at each stage of oocyte growth was calculated. The data on Table II shows that there is a 26 fold increase in cytoplasmic volume between stages 1 and 10. We now turn to the 3H-actinomycin D binding data for the cytoplasm, the primary concern of this study. Table II presents the grain counts, representing 3H-actinomycin D binding to cytoplasmic DNA, in a standard unit measure of CYTOPLASMIC DNA IN OOGENESIS 589 grains per 1000 //,-' cytoplasmic area. This value decreases markedly from stage 1 to 5, suggesting a dispersal of cytoplasmic DNA correlated with the marked in- crease in size of the oocyte between these two stages. In contrast, from stage 6 to stage 10 the grain counts per 1000 p.- remains almost constant, despite the con- tinued growth of the oocyte. Since the oocyte is a three dimensional sphere, we will now use the above data to obtain an estimate of the amount of 3H-actinomycin D binding to cytoplasmic DNA in the total cytoplasmic volume. Table II shows the total cytoplasmic volume at each stage of oocyte growth. Table II also shows the amount of 3H-actino- mycin D binding calculated to occur in relation to total cytoplasmic volume. This latter figure provides a comparative estimate of the relative amounts of cytoplasmic DNA present at various stages. The data show's that there is a 4-fold increase in 3H-actinomycin D binding to total cytoplasmic DNA during oocyte development. At each stage of oocyte development we have now determined two variables : cytoplasmic volume, and total grain counts calculated in total cytoplasmic volume Figure 1 expresses each of these variables in relation to stage of oocyte develop- ment. The two ordinates of Figure 1 are drawn to separate and arbitrary scales, for the purpose of comparing the shape of the two curves. This manner of pre- sentation emphasizes the points of similarity between the two curves. Figure 1 shows that the most rapid increase in the amount of cytoplasmic DNA (grain counts) occurs in the primary and young growing oocyte (stages 1 to 5). There is a stabilization in the amount of cytoplasmic DNA at stage 5, but no marked increase between stages 5 to 10. When the volume increase of the oocyte is plotted on the same graph for comparison, we see that the increase in cytoplasmic volume starts slowly from stages 1 to 3, then rises sharply from stages 3 to 5, plateaus briefly at the stages 6 and 7, and then increases steadily from stages 7 to 10. During the first stages of oocyte growth, the plots for increase in cytoplasmic volume and for increase in cytoplasmic DNA appear quite similar. However, after TABLE II The numerical stages of oocyte growth are defined in relation to the size of the nucleus and cytoplasm. Grain count data is expressed in relation to unit cytoplasmic area, total cytoplasmic area and total cytoplasmic volume. Stage Nuclear area M2 Cytoplasmic area M- Grain count per 1000 M2 cytoplasmic area Cytoplasmic volume M3 Number of grains calculated per total cytoplasmic volume 1 110 900 266 24,300 362 2 135 1080 200 32,000 451 3 500 1250 130 49,100 652 4 850 5000 49 326,000 1096 5 900 6000 35 421,200 1208 6 1 000 6500 29 483,800 1221 7 1100 6500 27 483,700 1253 8 1150 7200 25 573,200 1306 9 1200 9000 25 743,700 1289 10 1250 9500 25 867,900 1325 5<>0 H. E. ENESCO AND K. H. MAN O u _ Q- O > u o I— z z 13 O u z O 1800 1600 - 1400 - 1200 - 1000 _ 800 • 600 ~ 400 - 200 - — 900 — 800 — 700 — 600 n o _ 500 — 400 — 300 O u — Q_ O — 200 u — 100 123456 89 10 STAGE OF OOCYTE DEVELOPMENT FIGURE 1. Grain counts on radioautographs, which represent the amount of aH-actinomycin D binding to cytoplasmic DNA at successive stages of sea urchin oocyte development are plotted on the left ordinate. Increase in cytoplasmic volume is plotted on the right ordinate. Arbitrary scales were chosen for the purpose of comparing the shape of the two curves. This comparison shows that increase in mitochondrial DNA parallels increase in cytoplasmic volume only in the early and growing oocyte stages. stage 7 all similarity disappears. The oocyte continues to increase in volume, but there is no further increase in cytoplasmic DNA. DISCUSSION No controlled in vitro studies of 3H-actinomycin D binding to sea urchin oocytes had been carried out prior to the present study. In vivo incubation studies CYTOPLASMIC DNA IN OOGENESIS 591 by Greenhouse, Hynes and Gross (1971) have shown that actinomycin I) inhibits RNA synthesis in developing sea urchin embryos, and have established that 3H-actinomycin D readily penetrates into living embryos. Radioautographic localization of the 3H-actinomycin D binding sites showed that silver grains were distributed over the cytoplasm as well as over the nuclei of the early cleavage embryos. The authors interpret the cytoplasmic label as cytoplasmic DNA. Their experimental design did not permit confirmation with DNase extracted controls. The amount of 3H-actinomycin D binding which takes place in any experimental system is a function of both of the concentration of the 3H-actinomycin D used and of the incubation time used (Simard, 1967). While 3H-actinomycin D bind- ing does not quantitatively reflect the absolute amount of cytoplasmic DNA present in the oocytes, we can use the actinomycin D binding data as a comparative estimate of the relative amount of cytoplasmic DNA available for the 3H-actino- mycin D binding at the various stages of oocyte development. The lack of 3H- actinomycin D binding after DNase extraction in this study excludes any label being non-specifically bound to either nucleus or cytoplasm. Binding of 3H- actinomycin D to cytoplasmic DNA has been reported in only one previous study (Aczel and Enesco, 1973), although several investigators have reported on 3H-actinomycin D binding to nucleolar DNA (Simard, 1967, Camargo and Plant, 1967). It was first reported by Hoff-J0rgenson (1954) that sea urchin eggs contained large amounts of cytoplasmic DNA. This cytoplasmic DNA was first shown to be mitochondria! DNA by Piko, Tyler and Vinograd (1967), who also reported that one third of the cytoplasmic DNA in the sea urchin egg was associated with the yolk spherule fraction. These findings were subsequently revised by Berger (1968) who showed that DNA becomes associated with the yolk fraction in sea urchin eggs only as a result of contamination by the mitochondrial fraction. The results of Berger (1968), thus show that the "cytoplasmic" DNA of the sea urchin egg is entirely mitochondrial DNA. Dawid (1972) also cites evidence that the cytoplasmic DNA reported to occur in eggs of various species is entirely mitochondrial DNA. This is true not only for sea urchin eggs, but for frog eggs (Dawid, 1966) and for eggs of Urcchis caiipo as well (Dawid and Brown, 1970). Mitochondrial DNA from mature sea urchin eggs has been isolated and char- acterized by Piko and associates (Piko ct al, 1967, Pilo, Blain, Tyler and Vinograd, 1968). The replication of mitochondrial DNA during sea urchin oogenesis fol- lows the same molecular replication pattern generally reported for mitochondrial DNA (Matsumoto, Kasamatsu and Piko, 1973). In the present study, the fixation and staining methods necessary for radio- autography are not suitable for morphological observation of mitochondria. Even though we cannot correlate silver grains directly with mitochondria, it is clear from the work of Berger (1968) that the mitochondria are the only DNA con- taining organelles in the cytoplasm. The results presented here show that increases in the amount of cytoplasmic DNA, as detected by 3H -actinomycin D binding, are confined to the preliminary and growing oocyte stages. No increase in cytoplasmic DNA takes place in the 592 H. E. ENESCO AND K. H. MAN terminal oocyte stages. If we now interpret cytoplasmic DNA as mitochondrial DNA, how do these results correlate with our knowledge of mitochondria in oocytes ? We do know that the number of mitochondria in the sea urchin oocyte in- ceases markedly during oogenesis. Verhey and Moyer, (1967) report that there is an average of 393 mitochondria in a small oogonium, as compared to an average of 19,827 mitochondria in a mature oocyte. These figures are based on calcula- tions from electron microscope data, and represent a 50 fold increase in the num- ber of mitochondria from the earliest to the latest stage of oocyte development. This data does not tell us the time at which mitochondrial replication takes place, and provides no information about mitochondrial DNA. These combined results of this and the other studies cited show that the syn- thetic activities of the sea urchin oocyte are sequentially arranged into specific stages of oocyte differentiation. The primary and growing oocytes are characterized by intensive RNA syn- thesis and production of ribosomes (Verhey and Moyer, 1967) as well as by the increase in cytoplasmic DNA reported in this study. Intensive yolk synthesis takes place at later stages (Esper, 1962, 1965). Dawid (1972) suggests that large numbers of mitochondria may accumulate in the egg as a storage product to support embryonic development during the rapid cleavage stages. Gustafson (1965) reports that the number of mitochondria re- main constant in the sea urchin from the mature oocyte stage through gastrulation. It is clearly the oocyte mitochondria which support the early development and differentiation of the sea urchin embryo. The oocyte mitochondria, containing mitochondrial DNA are thus stored in the oocyte for distribution among the cells of the growing embryo. These results show that mitochondrial replication takes place primarily in the early oocyte stages. SUMMARY The relative amount of cytoplasmic DNA in sea urchin oocytes was studied at ten successive stages of oocyte differentiation in the sea urchin, Strongylo- ccntrotus [>-nrpnratus. The cytoplasmic DNA was selectively localized by means of 3H-actinomycin D binding on 5 ^ sections of the ovaries. The specificity of the 3H-actinomycin D binding to DNA was established using DNase extracted con- trol slides. The results show that cytoplasmic DNA increases continuously during the primary and growing oocyte stages. Although oocytes at later stages of dif- ferentiation continue to increase in volume, there is no increase in cytoplasmic DNA during these later stages. The cytoplasmic DNA observed in this study is interpreted as mitochondrial DNA. LITERATURE CITED ACZEL, J. AND H. E. ENESCO, 1973. DNA content of normal and tumor cells in mice. /. Cell Biol., 59 : 2a. AMANO, M., 1962. Improved techniques for the enzymatic extraction of nucleic acids from tissue sections. /. Histochcm. Cytochcm., 10 : 204-212. CYTOPLASMIC DNA IN OOGENESIS 593 BERGER, E. R., 1968. A quantitative study on sea urchin egg mitochondria in relation to their DNA content. /. Cell Biol, 39 : 12a-13a. CAMARGO, P. E. AND W. PLAUT, 1967. The radioautographic detection of DNA with tritiated actinomycin D. /. Cell Biol., 35 : 713-716. COWDEN, R. R., 1962. RNA and yolk synthesis in growing oocytes of the sea urchin Lytechinus vcrigatus. Exp. Cell Res., 28 : 600-604. DAWID, I. B., 1966. Evidence for the mitochondria! origin of frog egg cytoplasmic DNA. Proc. Nat. A cad. Sci. U.S.A., 56 : 269-276. DAWID, 1. B., 1972. Cytoplasmic DNA. Pages 215-226 in J. D. Biggers and A. W. Schuetz, Eds., Oogcnesis. University Park Press, Baltimore. DAWID, I. B. AND D. D. BROWN, 1970. The mitochondrial and ribosomal DNA components of oocytes of Urechis caupo. Develop. Biol., 22 : 1-14. DEITCH, A. D., 1966. Cytophotometry of nucleic acids. Pages 327-354 in G. L. Wied, Ed., Introduction to Quantitative Cytochemistry. Academic Press, New York. EBSTEIN, B. S., 1967. Tritiated actinomycin D as a cytochemical label for small amounts of DNA. /. Cell Biol., 35 : 709-712." EBSTEIN, B. S., 1969. The distribution of DNA within the nucleoli of the amphibian oocyte as demonstrated by tritiated actinomycin D radioautography. /. Cell Sci., 5 : 27-44. ESPER, H., 1962. Incorporation of 14C-glucose into oocytes of Arbacia punctulata. Biol. Bull., 123 : 246. ESPER, H., 1965. Studies on the nucleolar vacuole in the oogenesis of Arbacia punctulata. Exp. Cell. Res., 28 : 85-96. FLAX, M. H., AND M. H. HIMES, 1952. Microspectrophotometric analysis of metachromatic staining of nucleic acids. Physiol. Zool., 25 : 297-311. GREENHOUSE, G. A., R. O. HYNES, AND P. R. GROSS, 1971. Sea urchin embryos are permeable to actinomycin. Science, 171 : 686-689. GUSTAFSON, T., 1965. Morphogenetic significance of biochemical patterns in sea urchin em- bryos. Pages 139-292 in R. Weber, Ed., Biochemistry of Animil Development, Volume 1. Academic Press, New York. HoFF-JjziRGENSEN, E., 1954. Deoxyribonucleic acid in some gametes and embryos. Pages 79-90 in J. A. Kitching, Ed., Recent Developments in Cell Physiology. Colston Papers, Volume 7. KOPRIWA, B. AND C. P. LEBLOND, 1962. Improvements in the coating technique of radio- autography. /. Histochcm. Cytochcm., 10: 269-284. AlATSUMOxo, L., H. KASAMATSU, AND L. PIKO, 1973. Mitochondrial DNA synthesis in sea urchin oocytes. /. Cell Biol., 59 : 221. MULLER, W. AND D. M. CROTHERS, 1968. Studies of the binding of actinomycin D and re- lated compounds to DNA. /. Mol. Biol., 35: 251-290. PEARSE, A. G. E., 1961. Histocliemistrv: Theoretical and Applied. [2nd. Ed.] London, J. & A. Churchill, 998 p. PIKO, L., A. TYLER AND J. VINOGRAD, 1967. Amount, location, priming capacity, circularity and other properties of cytoplasmic DNA in sea urchin eggs. Biol. Bull., 132 : 68-90. PIKO, L., BLAIR, D. G., TYLER, A. AND J. VINOGRAD, 1968. Cytoplasmic DNA in the un- fertilized sea urchin egg : physical properties of circular mitochondrial DNA and the occurrence of catenated forms. Proc. Anat. Acad. Sci. U.S.A., 59: 838-845. SIMARD, R., 1967. The binding of 3H-actinomycin D to heterochromatin as studied by quantitative high resolution radioautography. /. Cell. Biol., 35 : 716-722. THURSTON, J. M. AND D. L. JOFTES, 1963. Stains compatible with dipping radioautography. Stain Techno!,, 28 : 231-235. VERHEY, C. A. AND F. H. MOYER, 1967. Fine structural changes during sea urchin oogenesis. /. Exp. Zool., 164: 195-226. YOUNG, B. A. AND B. M. KOPRIWA, 1964. The use of the Gevaert NUC-307 nuclear emulsion for radioautography at the electron microscope level. /. Histochcm. Cytochcm., 12 : 438-441. Reference: Biol. Bull., 147: 594-607. (December, 1974) FACTORS AFFECTING MUSCLE ACTIVATION IN THE HYDROID TUBULARIA ROBERT K. JOSEPHSON School of Biological Sciences, University of California, Irrine, California 92664 The control of muscle contraction in coelenterates is complex and not well understood. For example, some muscles of sea anemones are activated by a diffuse, all-or-nothing conducting system, probably the nerve net of the column and mesenteries (Pantin, 1935a, 1935b, 1952; Pickens, 1969; Robson and Joseph- son, 1969). There are requirements for facilitation such that a single impulse in the conducting system initiates little or no mechanical response while a pair of impulses at a suitable interval evokes a moderately rapid twitch. During repetitive stimulation the mechanical response both facilitates and sums. The site of facilita- tion appears to be between the conducting system and the muscle membrane, probably at the neuromuscular synapse (Pantin, 1935a; Josephson, 1966; Robson and Josephson, 1969). While these features seem conventional, other properties of the muscles are not. The same muscles in anemones which give rapid con- tractions can also contract very slowly. The slow contractions frequently occur spontaneously and can be slowly propagated along the muscle sheet (Batham and Pantin, 1954; Ewer, 1960; Pantin, 1965a). Conducting systems have been identi- fied which initiate and inhibit slow contractions (McFarlane, 1974) but in general the mechanisms coordinating slow contractions are unclear. In scyphomedusae, bell contractions are initiated by impulses in an all-or-nothing conducting system, here almost certainly a nerve net (Horridge, 1954). Facilita- tion is less pronounced in medusae and a long refractory period of the muscle prevents appreciable summation (Bullock, 1943; Horridge, 1955). Twitches of the bell musculature can be augmented by contemporary activity in a second con- ducting system (Horridge, 1956), but the mechanisms of this augmentation are unknown. Muscle contraction in hydrozoans has been less studied than in scyphozoans or anthozoans, largely because the small size of most hydrozoans is not conducive to tension measurement. In hydra, where tension recordings are available, longitudinal contraction of the column is peculiarly biphasic (Josephson, 1967). Excitation in a column conducting system initiates a rapid tension increase followed by a slower tension rise. Whether both tension components are generated by the same con- tractile cells or how either component is initiated is unknown. Thus coelenterate muscle contraction varies with the pattern of impulses arriving at the muscle ; the contraction may be influenced by activity in more than one conducting system ; and muscle sheets, possibly even individual muscle cells, may produce both fast and slow contractions (rf. Horridge, 1956; Pantin, 1952. 1965b). .Details about the mechanisms of muscle activation are obscure. Some of the uncertainties about the control of muscle contraction could be resolved with intracellular recording but the small size and contractility of coelenterate epithelio- 594 MUSCLE ACTIVATION IN A HYDROID 595 muscular cells make intracellular recording a formidable task. One is left then with less direct measures, developed tension or extracellularly-recorded muscle action potentials, to use in analyzing the mechanisms controlling muscle contrac- tion. Electrical events which are probably muscle action potentials can be easily re- corded from the hydroid Tubular ia with surface suction electrodes. These po- tentials can be used to quantify muscle activation without having to directly record muscle tension. This paper considers some of the factors modifying the amplitude of these potentials and hence, by inference, the intensity of muscle activation. Some background information on the morphological and functional organization of Tubularia would be appropriate at this point. Tubularia is a moderately large hydrozoan polyp with two sets of tentacles, a proximal set at the base of the hydranth and a distal set which encircles the mouth. The hydranth is borne on a long stalk. In mature polyps gonophores hang from stalks which arise just distal to the proximal tentacles. The spontaneous behavior of Tubularia results from the presence of a number of interacting pacemaker systems, the two most im- portant of which are the neck pulse (NP) system in the distal stalk and the hydranth pulse (HP) system in the hydranth (Josephson and Mackie, 1965). The NP system produces electrical pulses (neck pulses -- NP's) which typically occur as a sequence of single pulses interrupted periodically by a burst of pulses. Potentials from the HP system (hydranth pulses -- HP's) occur as single pulses and pulse bursts. When the HP system fires it initiates coordinated closure of the proximal tentacles and weaker, synchronized closure of the distal tentacles. Three conducting systems have been identified in Tubularia (Josephson, 1965) of which one, the distal opener system (DOS), is of interest here. The DOS courses through the stalk and hydranth. Activating the DOS initiates contraction of the aboral longitudinal musculature of the distal tentacles resulting in outward flaring (opening) of the distal tentacles. Activity in the DOS also inhibits the HP system and other pacemaker systems in the hydranth (Josephson and Uhrich, 1969; Josephson and Rushforth, 1973). Electrical potentials from the DOS can be recorded with surface suction electrodes or extracellular microelectrodes. In the stalk and most of the hydranth, DOS pulses (DOSP's) are small (50-200 /xV), short (9-15 msec), all-or-nothing electrical events. In the vicinity of muscles activated by the DOS a second component appears in the DOSP. With an electrode on a distal tentacle or at the base of the distal tentacles the short, all-or- nothing pulse is followed by a slow, graded potential. This second DOSP com- ponent is thought to be a muscle action potential because its amplitude varies monotonically with that of observed tentacle flaring. Both the slow potential and tentacle opening defacilitate to the same degree during repetitive stimulation and both decline in parallel during anesthetization with excess magnesium (Josephson, 1965). The initial DOSP component is unchanged during repetitive stimulation (Fig! 1A) and its amplitude remains constant during anesthetization until it abruptly ceases to appear. Thus the initial DOSP component appears to be gen- erated by the conducting system, the second component by the responding muscula- ture. The amplitude of the second DOSP component is here used as a measure of the intensity of muscle activation, 596 ROBERT K. JOSEPHSON MATERIALS AND METHODS Tubularia were collected from the Woods Hole area. Animals were obtained from different places at different times of the year and, judging by the information provided by Miller (1969), included three species; T. crocca from the jetty at New Bedford, T. spectabilis from Woods Hole and the Cape Cod Canal, and T. larynx from the Cape Cod Canal. For the physiological parameters considered in this study, animals thought to be T. larynx and T. spectabilis were indistinguishable. T. crocea differs from the other two in having a longer, more active proboscis, in producing generally larger DOSP's and in other particulars described below. The stimulating and recording methods are described in detail elsewhere (Josephson and Uhrich, 1969; Josephson and Rushforth, 1973). Briefly, animals were stapled to the bottom of a dish of sea water cooled to 16-18° C. A suction electrode on a gonophore or gonophore stalk was used to activate the DOS. The stimuli were 1 msec voltage pulses. Another suction electrode on the base of a distal tentacle recorded DOSP's and HP's. Recorded activity was displayed on a pen-writer or oscilloscope. RESULTS Defacilitation during repetitive DOS activation The most obvious factor affecting DOSP amplitude is preceding DOS activity; the slow component of DOSP's shows marked defacilitation (Fig. 1A; Josephson, 1965). The time course of defacilitation was examined by stimulating the DOS at a set of interstimulus intervals ranging from 2 to 300 seconds. In each trial the intervals of the set were presented in random order. Results from four ani- mals, each subjected to 20 sets of interstimulus intervals, are shown in Figure 2. In one of these animals the DOSP's reached maximum size with interstimulus B FIGURE 1. Depression of DOSP's during repetitive stimulation and following HP activity; (A.) defacilitation of the second component during stimulation at five-second intervals. The largest response is to the first stimulus which came after a long rest ; subsequent responses became progressively smaller. (B.) DOSP's during and after an HP burst; the DOS was being regularly stimulated at 10-second intervals. The first DOSP, the one with the greatly reduced second component, occurred in the burst ; the next three with progressively increasing amplitude came 7, 17, and 27 seconds after the burst. The first DOSP was followed by an HP (the potential halfway through the trace) which is poorly resolved at the high amplification used. Positive potentials are upward deflections. Note the constant amplitude of the first component. MUSCLE ACTIVATION IN A HYDROID 597 > E CL 2 < O Q 0 o 7; crocea • T. spectabilis 10 100 INTERSTIMULUS INTERVAL, SEC FIGURE 2. Amplitude of the DOSP second component as a function of interstimulus in- terval. Each point is the average of 20 determinations ; standard errors for the first and last points are shown to the left and right of the curves. See text for further information. intervals of ten seconds or more; in the other three, DOSP's continued to in- crease with increasing interval to the longest used. In the four animals the DOSP amplitude after a 300-second interval was 2.3 to 3.1 times greater than that with an interstimulus interval of 2 seconds. The pattern of DOSP amplitude change during bursts of stimuli varies with the stimulus frequency. With interstimulus intervals of about five seconds or more, DOSP's decline monotonically during the burst until they reach an approximate steady state after three or four stimuli. At shorter intervals the DOSP depression sometimes overshoots and the response to the second stimulus may be the smallest of the series, subsequent DOSP's growing slightly until a steady state is reached (Fig. 3). It might be noted that defacilitating effector responses are unusual in coelenter- ates; facilitating responses being more typical (e.g., Pantin, 1935a; Bullock, 1943; Morin and Cooke, 1971). Another example of defacilitating muscle contraction, again from a hydroid, occurs in the lashing of dactylozooids from Hydractinia (Stokes, 1974). 598 ROBERT K. JOSEPHSON 1/10 sees r\ 1/5 sees l/2.5secs FIGURE 3. Defacilitation of the DOSP during repetitive stimulation, T. crocca. The short, upward deflections preceding the DOSP's are stimulus artifacts. In this and the other pen-writer records (Figs. 4, 6) positive potentials are downward deflections. Depression following pacemaker activity DOSP's closely following HP's tend to be smaller than those coming long after HP's (Fig. 4). As with defacilitation it is only the second component of the DOSP which is depressed (Fig. IB). The time course of this depression was examined by stimulating the DOS at a regular frequency so that defacilitation reached a steady state and then determining the DOSP amplitude as a function of the interval between the DOSP and preceding HP activity. Rather large sample sizes and therefore long recording sessions were required for reasonable temporal resolution. For example, the data of Figure 5A repre- sents over 4 hours of continuous stimulation and recording. During long sessions the amplitude of recorded DOSP's often declines. This may be due in part to fatigue but probably more largely to slippage of the recording electrode or tissue damage beneath the electrode. To compensate for this drift the amplitude of a DOSP following HP firing was expressed as a percent of the DOSP size im- mediately before the HP activity. The stimulus frequency was one per 10 seconds and the stimuli were initially about twice DOS threshold. During long stimulation periods the DOS occasionally failed to fire following a stimulus. Each time there was such a miss the stimulus intensity was increased. If this eventually did not give faithful following, the stimulating electrode was repositioned and the run continued. Because of partial reduction in defacilitation, the DOSP was larger than normal following the first and sometimes the second successful stimulus after an unsuccessful one. For this reason only those instances were considered in which the control DOSP, that immediately before the HP activity, was preceded MUSCLE ACTIVATION IN A HYDROID 599 by two consecutive successful stimuli. Similarly if there was a miss following the HP activity, subsequent DOSP's in the post-HP interval were not included in the tabulated results. Another restriction was that the control DOSP had to be separated from the HP preceding it by more than 20 seconds, the major part of the HP effect on DOSP's being over by this time. Results from the animals giving the two longest series are shown in Figure 5. It is apparent that DOSP depression by HP's is greater following long HP bursts (4 or more pulses) than after short bursts (2 or 3 pulses) and slightly greater after short bursts than following single HP's. Recovery from depression by HP's is at first rapid but there is some indication that recovery can continue for as long as 40 to 60 seconds. At long post-HP intervals the DOSP's are, on the average, larger than controls. This is probably because the DOSP's in these bins occur at post-HP intervals which are longer than is true for the average control DOSP. HP depression of DOSP's is much more erratic in T. crocca than in the other species. In only a few of the more than 20 specimens of T. crocca examined was there clear depression and in these the variability in the extent of depression was too great to allow mapping its time course. The depression could be quite obvious. For example, in one T. crocca, stimulated at one shock each five seconds, the first DOSP following a single HP averaged only 61% of the amplitude of the DOSP preceding the HP (s.e. — 5%, n = 28). The probability of this reduction occur- ring by chance is extremely low (P < 0.005). But this was an exceptional case and more usually depression was absent or too small to lie readily detectable. Depression following spontaneous tentacle actii'itv Distal tentacles show several kinds of spontaneous activity including oral or aboral movements of one or a few tentacles, simultaneous outward flaring of all FIGURE 4. Depression of DOSP's following HP bursts. The upper and lower traces are from different parts of a long record during which the DOS was stimulated once each 10 seconds. Stimuli are marked by S in the figure. One HP burst occurs in the top trace, two in the lower trace. Note that DOSP's are depressed following these bursts. The records shown are a portion of those used to construct Figure 5A. 600 ROBERT K. JOSEPHSON tentacles, and coordinated outward flaring beginning at one point on the tentacle crown and progressing around unidirectionally. The latter will be called a ineta- chronal tentacle flare. Simultaneous outward flaring of all distal tentacles appears to be due to spontaneous firing of the DOS ; it is associated with electrical potentials identical to evoked DOSP's. Evoked DOSP's following a spontaneous one are depressed, presumably because of the same defacilitation seen during series of evoked DOSP's. There are electrical events associated with the other spontaneous tentacle move- ments as well. Activity in one or a few tentacles is recorded as small, irregular potential deflections ; metachronal tentacle flares are recorded as large potentials, a bit smaller and slower than DOSP's and lacking the first DOSP component. The spontaneous tentacle movements and associated potentials are best seen in T. A. single HP (n=80) 4 or more HP's (n=6IO) 4 or more HP's (n=!33) 20 40 60 0 20 40 60 0 20 40 60 Time after HP firing, sees FIGURE 5. DOSP amplitude following HP activity. A is from the animal for which the longest series of records was collected. This animal produced only single HP's and long HP bursts. B is from another animal, one which produced bursts with 2 or 3 pulses as well as longer bursts and single HP's. The intervals between HP activity in these animals ranged from less than 5 seconds to more than a minute. Therefore the number of DOSP's in a given time block decreases with increasing time after HP activity; the longer the post-HP time the fewer the number of HP intervals long enough to contain that time. The bins in the histograms of A have been increased in width at longer post-HP times so that each bin contains approximately the same number of entries. Further explanation is given in the text. MUSCLE ACTIVATION IN A HYDROID 601 1 mV 5 sec FIGURE 6. DOSP's following spontaneous distal tentacle activity. The events marked m are electrical potentials associated with sequential distal tentacle flaring in a metachronal wave. The dotted lines in A and C connect the DOSP peaks before and two after the spontaneous activity to emphasize the reduction in the DOSP immediately following the metachronal flare. crocea which has an active proboscis and in which the electrical events are gen- erally larger than in the other species. DOSP's are depressed following aboral movements of scattered tentacles and particularly following metachronal tentacle flaring (Fig. 6). In the animal of Figure 6 metachronal flares occurred 13 times during periods of regular DOS stimulation at one shock each 10 seconds. The amplitudes of DOSP's immediately following metachronal flares averaged 91% of the DOSP's preceding the flares (s.e. = 2.5%:, P < 0.01). Because spontaneous distal tentacle activity is of erratic occurrence and variable amplitude, no attempts have been made to quantify the time course of DOSP depression following this activity. The most complete DOSP depression encountered occurred during defecation. In defecation the mouth slowly opens widely and rolls backwards over the proboscis, turning the proboscis inside out. In three animals, two specimens of T. crocea and one T. spectabilis, defecation occurred during a recording session without dis- placing the recording electrode. During the period when the mouth is open and inflected back over the proboscis, the second component of the DOSP is com- pletely suppressed ; DOSP's occur as only apparently normal first components. DISCUSSION The two components of DOSP's recorded from distal tentacles have different properties and are probably generated by different kinds of cells. The initial com- ponent is similar in shape and amplitude to DOSP's recorded elsewhere in the polyp. This component reflects activity in the conducting system, either electro- 602 ROBERT K. JOSEPHSON genesis by the conducting cells themselves or responses triggered in other cells by the conducted activity (see Ball and Case, 1973, for an example of separable con- duction and electrogenesis). For reasons given above the graded second component of the DOSP is considered to be a muscle action potential, presumably generated by membranes of the epitheliomuscular cells which form the opening musculature of the distal tentacles. It was earlier suggested that the DOS is a neuronal con- ducting system (Josephson, 1965). The clear distinction between the first and second components of the DOSP ( e.g.. Fig. 1 ) supports this view, at least it in- dicates that the DOS is not part of the same epithelial sheet which, in the vicinity of the distal tentacles, forms the opening musculature. Were the DOS an epithelial conducting system one would expect the recorded DOSP to gradually change when it entered the muscle area, perhaps becoming larger as the expanded mem- brane around the contractile processes became involved. Instead the DOSP is quite similar to that elsewhere but with a new component added after a distinct delay. While certainly not proving the identity of the DOS, the form of the recorded potentials is consistent with the idea that the initial DOS component is neuronal, the slower component from epitheliomuscular cells, and the delay between the two due to transmission processes between the conducting system and the muscle. In fact the delay between the first and second DOSP components is 9 to 20 msec which is sufficiently long that there may be several steps involved in the coupling between the two. This delay is longer than that between nerve net potentials and muscle action potentials in anemones (0-8 msec ; Robson and Joseph- son, 1969; Pickens, 1969) ; about the same as the delay between conducting sys- tem pulses and evoked luminescence in the hydroid Obelia ( Morin and Cooke, 1971) and considerably shorter than the delay between the onset of electrical activity and that of column contraction in hydra (100—150 msec; Josephson, 1967). Several factors have been identified which selectively depress the second com- ponent of DOSP's : preceding DOS activity, HP firing, spontaneous distal tentacle activity including but not restricted to spontaneous DOSP's, defecation, and partial anesthetization with magnesium (fur the last see Josephson, 1965). The recovery of tentacle responsiveness is slow, taking tens of seconds after HP firing and minutes after DOS activity. The constancy of the first DOSP component and the lability of the second indicates that factors affecting DOSP's act on the link be- tween the two components, i.e.. they affect transmission processes between the conducting system and the muscle cells. There are several possible sites at which the DOSP reduction could be being effected ; the membranes of the muscle cells (post-junctional depression), the terminals of the conducting system (pre-junc- tional depression) or unidentified elements interposed between the conducting sys- tem and the muscle cells. Some consequences of the first two possibilities as sites of depression will now be considered. Depression of conducting system efficacy Depression by excess magnesium and defacilitation can be readily interpreted as due to pre-junctional mechanisms. Excess magnesium characteristically blocks transmission at chemical synapses by reducing the amount of transmitter released by presynaptic terminals {e.g., del Castillo and Engbaek, 1954), in turn resulting MUSCLE ACTIVATION IN A HYDROID 603 OPENER MUSCLES DOS INPUT FIGURE 7. A scheme which could account for DOSP depression during repetitive DOS stimulation and following spontaneous distal tentacle activity on the basis of reduced efficacy of shared final terminals. The DOS synchronously activates a set of interconnected elements through polarized junctions. Spontaneous activity spreads laterally through the interconnected elements without exciting the main conducting pathway of the DOS. in a smaller post-synaptic response. Although there may be both pre-junctional and post-junctional components in synaptic depression, a major factor is often depletion of available transmitter stores (e.g., Thies, 1965). The obvious de- facilitation of DOSP's could similarly be due to a reduced pool of available trans- mitter and hence less transmitter released to later stimuli of a series. The HP effects of DOSP's could also be due to reductions in the amount of transmitter released. The HP depression of DOSP's would then be an example of pre-synaptic inhibition. It is tempting to regard the DOSP reduction following spontaneous tentacle activity as due to the same defacilitation seen during repetitive DOS activation. If these are both pre-junctional effects it would require that spon- taneous activity and DOS activation excite the opener muscles through the same final terminals. When activated by the DOS the tentacle opening appears syn- chronous ; during spontaneous activity the tentacles can open sequentially, indicat- ing rather slow lateral propagation. A scheme which would allow synchronous tentacle activation by the DOS and sequential activation during spontaneous activity is shown in Figure 7. Depressed responsiveness of post- functional membrane Many of the features of DOSP depression could also be explained as due to the muscle membrane being partially refractory following activity. The time course of DOSP recovery would then reflect the time course of returning muscle re- sponsiveness. Refractoriness following activity would readily account for the de- cline of DOSP's during repetitive stimulation and following spontaneous activity, but not for the reduction of DOSP's during magnesium anesthetization since this occurs without muscle activity. There are at least two ways that HP firing could depress muscle responsiveness. The HP system may directly inhibit the muscle membrane by classical chemical inhibition. Alternatively the depression could again be related to post-firing 604 ROBERT K. JOSEPHSOX DISTAL TENTACLE OPENERS co LU I- co CO 1 CO DISTAL TENTACLE CLOSERS GONOPHORES PROXIMAL TENTACLE CLOSERS NECK MUSCULATURE CONDUCTING SYSTEM PACEMAKER SYSTEM MUSCLE GROUP EXCITATORY INPUT INHIBITORY INPUT FIGURE 8. The interactions between conducting systems, pacemaker systems and effectors in a Tnbularia polyp. This figure is based on one published earlier (Josephson and Uhrich, 1969) with the addition of gonophore interactions from unpublished observations of N. B. Rushforth and the inhibition of distal tentacle openers discussed in the text. refractoriness. HP firing initiates weak tentacle closing", but this does not pre- clude the possibility that the HP system simultaneously activates the opener muscles and the closing seen is a balance between fully active closer musculature and weaker or only partially-activated opener musculature. It should be stated that there is no evidence for opener muscle activation during HP firing. DOSP's are not recorded from the polyp or stalk during HP firing, indicating that the MUSCLE ACTIVATION IN A HYDRO1D 605 conducting system is not excited. The muscle action potentials could be obscured by the usually larger HP's, but in some recordings from T. crocca the second DOSP component was larger than HP's and should have been seen during HP firing were it present. So little is known about the mechanisms of mouth opening and proboscis eversion which are part of defecation that it seems fruitless to try to assign a pre- junctional or a post-junctional site to the DOSP depression during defecation. Certainly the circular muscles around the mouth must relax ; if the longitudinal muscles of the distal tentacles are involved in mouth opening or proboscis eversion is just not known. The above discussion is not meant to imply that all factors affecting DOSP amplitude operate on the conducting system nor that all operate post-junctionally on the muscle membrane. These speculations do emphasize that there are a num- ber of mechanisms which could account for the observed results and they indicate that with available knowledge it is not possible to distinguish which of the many possibilities are correct. Reciprocal inhibition in Tubularia From a behavioral viewpoint perhaps the most interesting of the interactions considered is the depression of DOSP's following HP firing. No matter what the mechanism, the net effect is inhibition of tentacle flaring by the HP system. The HP system and the DOS are antagonistic in their effects on the distal tentacles; the for- mer initiating closing and the latter opening of the tentacles. The two systems are also mutually inhibitory although in an unusual way. Activity in the DOS inhibits pacemaker firing by the HP system (Josephson and Uhrich, 1969; Josephson and Rushforth, 1973), and HP activity inhibits the tentacle movements which are a consequence of the DOS activity. These interactions and others previously identified in Tubularia are summarized in Figure 8. This diagram emphasizes the complexity of behavioral coordination in Tubularia, a complexity which extends to the activa- tion of individual muscle groups. This work was supported by U.S.P.H.S. grant NS-10336. I would like to thank J. Rosen for assistance with the experimental work and G. O. Mackie, D. R. Stokes and N. B. Rushforth for helpful discussions and useful comments on the manuscript. SUMMARY • 1. Activating the distal opener system (DOS), one of the conducting systems of the hydroid Tubularia, causes synchronous opening of the distal tentacles. Simul- taneously recorded potentials (DOS pulses =: DOSP's) from the vicinity of the responding muscles have two components: (a) an initial, short, all-or-nothing pulse reflecting activity in the conducting system, and, after a brief delay; (b) a 606 ROBERT K. JOSEPHSON second, slower potential thought to be a muscle action potential because its ampli- tude varies monotonically with that of tentacle movement. The amplitude of the second component is here used as a measure of the intensity of muscle activation. 2. The second DOSP component is depressed by preceding DOS activity (de- facilitation) ; by spontaneous tentacle movement including but not restricted to spontaneous DOS firing; by firing of the HP sytsem, one of the pacemaker sys- tems in the polyp ; and during the mouth opening associated with defecation. In all instances the first DOSP component is unchanged, indicating that the depression is of transmission processes between the conducting system generating the first component and the muscle membrane producing the second component. The re- covery from depression is slow, taking tens of seconds following HP activity and minutes after DOS firing. 3. The HP system and the DOS are mutually inhibitory. Previous work has shown that activity in the DOS inhibits HP firing; here it was found that HP firing inhibits the tentacle response which is a consequence of DOS activity. The depression following spontaneous tentacle movements suggests that spontaneous activity and DOS firing may excite the muscles through a common final pathway. LITERATURE CITED BALL, E. E. AND J. F. CASE, 1973. Electrical activity and behavior in the solitary hydroid Corymorpha palina. II. Conducting systems. Biol. Bull.. 145:243-264. BATHAM, E. J. AND C. F. A. PANTIN, 1954. Slow contraction and its relation to spontaneous activity in the sea-anemone Mctridiuin senile (L.). /. Exp. Biol., 31: 84-103. BULLOCK, T. H., 1943. Neuromuscular facilitation in scyphomedusae. /. Cell. Comp. Phvsiol., 22 : 251-272. DEL CASTILLO, J. AND L. ENGBAEK, 1954. The nature of the neuromuscular block produced by magnesium. /. PhysioL, 124: 370-384. EWER, D. W., 1960. Inhibition and rhythmic activity of the circular muscles of Calliactis parasitica (Couch). /. Exp. Biol., 37 : 812-831. HORRIDGE, G. A., 1954. The nerves and muscles of medusae. I. Conduction in the nervous system of Aurcllia aurita Lamarck. /. Exp. Bio!.. 31 : 594-600. HORRIDGE, G. A., 1955. The nerves and muscles of medusae. III. A decrease in the refractory period following repeated stimulation of the muscle of Rhizostoma pulmo. J. Exp. Biol., 32 : 636-641. HORRIDGE, G. A., 1956. The nerves and muscles of medusae. V. Double innervation in Scyphozoa. /. Exp. Biol., 33 : 366-383. JOSEPHSON, R. K., 1965. Three parallel conducting systems in the stalk of a hydroid. /. E.vp. Biol., 42 : 139-152. JOSEPHSON, R. K., 1966. Neuromuscular transmission in a sea anemone. /. Exp. Biol., 45 : 305-319. JOSEPHSON, R. K., 1967. Conduction and contraction in the column of Hvdru. J. E.rp. Biol., 47 : 179-190. JOSEPHSON, R. K. AND G. O. MACKIE, 1965. Multiple pacemakers and the behaviour of the hydroid Tubnlaria. J. Ex p. Biol., 43 : 293-332. JOSEPHSON, R. K. AND N. B. RUSHFORTH, 1973. The time course of pacemaker inhibition in the hydroid Tubnlaria. J. Exp. Biol., 59 : 305-314. JOSEPHSON, R. K. AND J. UHRICH, 1969. Inhibition of pacemaker systems in the hydroid Tubnlaria. J. Exp. Biol.. 50 : 1-14. MCFARLANE, I. D., 1974. Excitatory and inhibitory control of inherent contractions in the sea anemone Calliactis parasitica. J. Exp. Biol., 60 : 397-422. MILLER, R. L., 1969. Identification of Woods Hole species of Tubnlaria. Biol. Bull.. 137 : 409. MORIN, J. G. AND I. M. COOKE, 1971. Behavioural physiology of the colonial hydroid Obclia. MUSCLE ACTIVATION IN A HVDKOID 607 II. Stimulus-initiated electrical activity and bioluminescence. /. Exp. Biol., 54 : 707- 721. PANTIN, C. F. A., 1935a. The nerve net of the Actinozoa. I. Facilitation. /. Exp. Biol., 12 : 119-138. PANTIN, C. F. A., 1935b. The nerve net of the Actinozoa. II. Plan of the nerve net. /. £.r/>. Biol.. 12: 139-155. PANTIN, C. F. A., 1952. The elementary nervous system. Proc. Roy. Soc. London, Series B, 140: 147-168. PANTIN, C. F. A., 1965a. Capabilities of the coelenterate behavior machine. Amcr. Zoo!., 5 : 581-589. PANTIN, C. F. A., 1965b. A problem in muscular excitation in the Anthozoa. Pages 224-232 in J. W. S. Pringle, Ed., Essa\s on Physiological Evolution. Pergamon Press, Oxford. PICKENS, P. E., 1969. Rapid contractions and associated potentials in a sand-dwelling anemone. /. Exp. Biol., 51 : 513-528. ROBSON, E. A. AND R. K. JOSEPH SON, 1969. Neuromuscular properties of mesenteries from the sea-anemone Metridium. J. Exp. Biol., 50: 151-168. STOKES, D. R., 1974. Structural and functional polymorphism in the colonial hydroid Hydrac- tinia echinata. I. Polyp specialization. /. Exp. Zoo/., 190 : 1-18. THIES, R. E., 1965. Neuromuscular depression and the apparent depletion of transmitter in mammalian muscle. /. Newophysiol., 28: 427-442. KetVmice: liiol. Hull., 147: 608-617. (December, 1974) WATER TRANSPORT RATES OF THE TUNICATE CIONA INTESTINALIS1 KENNETH KUSTIN, KAYE V. LADD, GUY C. McLEOD AND DAVID L. TOPPEN Department of Chemistry, Brandcis University, Waltham, Massachusetts 02154 and the Neiv England Aquarium, Central Wharf, Boston, Massachusetts 02110 The interaction of suspension feeders with the environment depends primarily on the ability of these animals to transport sea water through their bodies. Al- though the determination of transport rates has been actively pursued by many investigators (JoYgensen, 1966), controversy exists over the accuracy and relevancy of the value produced by the use of a given method (Edmondson, 1966). The extreme rates, or the approximate average rate, are difficult to establish satis- factorily for tunicates due to experimental problems in handling the biological ma- terials (e.g., Hecht, 1916; Hoyle, 1953; Carlisle, 1966). Water transport rates of the tunicates Styela clava and Ascidiella aspersa were determined by different methods in static and in running sea water (Holmes, 1973). Even though the measured rates span a wide range, experiments with static sea water appear to yield lower transport rates than do those with running water. To establish sea water transport values for Ciona intestinalis, we are reporting transport rate studies using a variety of techniques. The average rates, or the extreme rates, are useful for determining the characteristics of sea water transport in suspension feeders and the relation of transport to tunicate physiology. MATERIALS The holding tanks of the New England Aquarium provide abundant specimens of a local species of tunicate, Ciona intestinalis. The freshly filtered sea water in which the experimental materials were maintained had a temperature of 16.0 ± 0.2° C, an average salinity of 30.0 ± 0.5/{f (—0.51 M NaCl), and an average pH of 7.8 ± 0.3 at the time of experimentation. All specimens were kept in flow- through aquaria. The 2-Methylquinoline (Quinaldine, Eastman Organic, practical grade) was used as anaesthetic to facilitate attachment of the tunicates to Pyrex tubes ( < 6 mm O.D. ) (Kustin, Ladd and McLeod, in preparation). The dye selected was a blue or green food color (Durkee's Food Color). The solvents for this dye, water and propylene glycol, were removed by freeze drying. The dye was then redissolved in sea water. At dye concentrations of approximately 0.01 ml original dye per 25.0 ml sea water, and/or high transport rates, Ciona was observed to react to the dye by squirting and ceasing to transport sea water. In some experiments sea water-dye solution was supplied to the specimen with a Sage Variable Speed Tubing Pump, Model 375A. The Sage pump was cali- brated by measuring the volume delivered to a clean dry graduated cylinder in a known time interval. 1 Supported by National Science Foundation Grant #GB-33617. 608 TRANSPORT RATES OF CIONA INTESTINALIS 609 . AB FIGURE 1. Overhead views of systems for determining the rate of sea water transport by the tunicate using the transfer of dyed sea water solution from compartment to compartment. The schematically rendered tunicate in the diagram shows the position of the attached specimen with respect to each compartment, dona which were used for experimental purposes lay horizontally, supported by the bottom of the container. Left shows dona attached by both siphons to Pyrex tubes with tygon tubing. Compartment A is attached to incurrent siphon ; compartment B is attached to excurrent siphon ; compartment C contains the body ; tube D maintains equal levels. Right shows Ciona attached by incurrent siphon only to compartment AB. Compartment C contains the body. Plume of dye emitted by tunicate is schematically rendered. Return tube prevents the mixing of AB and C. Dunaliclla euchlora was grown at room temperature under fluorescent light banks in one gallon bottles in supplemented sea water type f (Guillard and Ryther, 1962). Stock cultures were similarly maintained sterilely at 19° C in 125 ml Erlen- meyer flasks. Algae were concentrated by centrifugation at 2500 rpm and resus- pended in fresb sea water. METHODS AND RESULTS Volume methods Specimens were attached to a partitioned chamber (Fie. 1, left, with con- necting tube D absent). A typical rate of transport for a 4.0 g wet weight specimen was 180 ml hr'1 (or 45 ml/hr/g wet weight) determined by measuring the volume of overflow from Chamber B in 30 minutes. Careful observation showed that even under these conditions surface tension effects generated a pressure head approxi- mately 1 mm between Chambers A and B. 610 KUSTIN, LADD, McLEOD AND TOPPEN Connecting tube D (Fig. 1, left) was added to eliminate the pressure head and a blue non-toxic dye was introduced into Chamber A. The rate was determined by monitoring the change in the absorbance of the solutions in Chambers A and B. One milliliter samples were removed simultaneously from the chambers at approxi- mately six-hour intervals. Sea water was then carefully removed from Chamber C to keep all levels equal. Since the solution was pumped circularly the following treatment was used to calculate the flow rate produced by the tunicate. In Chamber A, HIA is the moles of dye, C± is the concentration of dye and VA is the volume of sea water; similar terms, but with subscript B, refer to Cham- ber B. If the volume of solution transferred between chambers is AF, then the instantaneous rate of change of dye in Chamber A is dinx/dt= (d(\V}/df} (CB-CA) (1) assuming mixing in Chamber A. The total number of moles, M, of dye is con- served ; therefore, (2) Also, the absorbance. A, of the solution in Chamber A is given by (3) where a is the molar absorptivity coefficient, and / is the pathlength through the dye solution. Insertion of (2) and (3) into (1), rearrangement, and integration over the limits inA -- inA° at time, t = 0 and WA -- ?HA at time, t = t assuming con- stant flow so that d(\V)/dt-- it -- constant, and FA -- FB yields (4) The solutions in Chambers A and B were gently stirred to assure uniformity as infrequently as possible, since the tunicates often responded to stirring by diminish- ing transport. In a typical experiment, a 2.3 g specimen had an average rate of (5 ± 0.5) ml/hr/g wet weight, which appeared to be a rather low value. Stress on the tunicate is relieved by reducing the number of attachments. Ac- cordingly, only the incur rent siphon was connected (Fig. 1, right), the partition between Chambers A and B was removed, and the specimen transported sea water from Chamber AB into C. A return tube between AB and C kept the system at zero hydrostatic head. The samples for absorbance readings were taken approxi- mately every 30 minutes for up to four hours, and analyzed according to equation (4) . The results are presented in Table I. In a further refinement, two specimens of dona intestinalis of approximately the same weight were attached to plastic bags partially filled with sea water. The bags were filled so as to exclude air bubbles, stoppered, weighed and allowed to equilibrate in a sea water bath. Under these conditions the pressure inside the bag should equal that outside. The specimens of C'wna intestinalis were anaesthe- tized and a Pyrex tube inserted in one siphon. They were then transferred to the sea water bath, allowed to revive and carefully connected to the bags so as not to TRANSPORT RATES OF CIONA INTESTINALIS 611 TABLE I Transport rates for Cioiia intestinalis determined by volume flow. Column headings refer to the following: m, wet mass Ciona ; FAB, volume of sea water in compartment AB; Vc, volume of sea water in compartment C; n, transport rate; V, (Vc + FAB)/ Fc FAB. m KAB Vc M/ V u u/m g ml ml hr~> ml /»-' ml hr~lg'1 1.3 78 1035 0.138 ± 0.006 10.0 it 0.4 7.7 ± 0.3 3.0* 260 890 0.32 ± 0.01 65 ± 2 21.7 ± 0.8 3.0* 260 890 0.742 ± 0.006 149 ± 1 49.8 ± 0.4 3.0* 260 890 0.418 ± 0.006 84 ± 1 28.0 ± 0.4 3.0** 260 890 0.67 ± 0.34 134 ±73 45 ± 24 2.5 260 1000 0.262 ± 0.006 54 ± 1 21.6 ± 0.5 2.1 78 1035 0.108 ± 0.009 7.8 ± 0.7 3.7 ± 0.3 2.1 260 1000 0.047 ± 0.004 9.6 ± 0.8 4.6 ± 0.4 * Same Ciona measurements over different 2 hour time periods in same day. ** Ciona as in * hut measurement over 4 hours on next day. change the mass of the system. One Ciona was connected via the incurrent siphon, the other via the excurrent siphon. After six hours the bags were carefully de- tached, stoppered and weighed. In this experiment a 2.2 g specimen transported in 71 g of sea water in 6 hours and a 2.5 g specimen transported out 64 g in the same period, yielding an average of 5 ml/hr/g wet weight. Volume methods ap- pear to give low rates. Possible causes could be cilia impairment due to insertion of the tubes and/or stress. Therefore, the clearance technique was also used (Fox, Sverdrup and Cunningham, 1937). Clearance method Specimens were fed cultures of the green marine flagellate algae Dnnaliella cuchlora. The average size of Dnnaliella was 4.6 ± 0.9 X 10 > 4 cm as determined by microscopical sizing. Twenty-four hours prior to experimentation, the tuni- cates were placed in aerated beakers containing known volumes of sea water. A 0.1-1.0 ml sample of Dnnaliella. cell concentration 1-5 X 10s cells ml'1, was then added. Aeration provided stirring, which, in addition to the motion of the algae, TABLE II Effect of time on cell counts with mixing by aeration. Organism is Dunaliella euchlora in 100 ml sea water. Column headings refer to the following: t, time; N, cell counts. t III IH 0 20 40 80 mean (ell ml~l X 10-= 1.14 1.3 1.20 1.15 1.20 it 0.05 ±0.8 ± 0.03 ± 0.07 ± 0.07 612 KUSTIN, LADD, McLEOD AND TOPPEN keeps their distribution uniform and counteracts the action of gravity. A control experiment, consisting of an aerated beaker containing only sea water and Duna- liclla, was sampled and cell counts taken to see whether gravitationally induced sedimentation was a factor that had to be taken into account. The data shown in Table II show that a uniform distribution of cells is maintained, within experi- mental error. Samples were removed for counting at periodic intervals over a span of two hours to determine the rate of cell removal. Cell counts were made on a Wild microscope at 10 X 40 power using an American Optical Neubauer hematocrit. 20 40 60 Time (minutes) 100 FIGURE 2. A semilog plot showing the relationship between cell counts per unit volume and time used in determining the rate of transport by particle clearance ; experimental data : mass dona, 1.6 g; volume sea water, 100 ml; initial cell/ml C2.47 ± 0.04) X 10s; u/V, 1.32 ± 0.12 hr1; u, 132 ± 12 ml/hr. TRANSPORT RATES OF CIONA INTESTINALIS 613 TABLE III Transport rates for Ciona intestinalis determined by clearance. Column headings refer to the following: m, wet mass Ciona ; N, initial cell/ml; V, volume sea water; u, transport rate. Samples with same mass are same Ciona at different times. Day m AT V u/V u u/m g cell ml'1 X 10~* ml hr-i ml hr~^ ml hr~lg~l 0 3.0 14.9 ± 0.8 250 2.2 ± 0.3 555 ± 70 185 ±25 1.5 13.7 ±0.7 100 1.6 ±0.2 160 ± 17 106 ± 18 1.1 13.1 ±0.4 100 1.02 ± 0.06 102 ±6 92 ±13 7 1.6 1.42 ± 0.04 100 1.3 ± 0.1 132 ± 12 82 ± 12 1.6 2.47 ± 0.04 100 1.3 ±0.1 132 ± 12 82 ± 12 1.6 97 ±3 100 0.82 ± 0.05 82 ± 5 51 ±6 2.4 2.5 ±0.1 100 0.16 ± 0.01 16 ± 1 6.6 ± 0.7 2.4 2.52 ± 0.06 100 0.11 ±0.02 11 ± 2 5 ± 1 9 1.6 1.44 ± 0.07 100 0.11 ±0.04 11 ±4 7 ±2 1.6 23.2 ± 0.4 100 0.11 ± 0.02 11 ± 2 7 ± 1 2.4 1.32 ± 0.09 100 0.78 ± 0.12 78 ± 12 32 ±5 2.4 22.5 ± 0.05 100 0.50 ± 0.04 50 ±36 21 ± 15 2.0 1.8 ±0.1 100 0.28 ± 0.04 28 ± 6 14 ±3 2.0 25 ± 1 100 0.14 ± 0.04 14 ± 4 7 ±2 15 1.6 10.9 ± 0.5 500 0.46 ± 0.10 231 ± 51 144 ± 32 1.6 6.5 ± 0.2 500 0.40 ± 0.05 201 ± 27 125 ± 17 2.4 10.7 ± 0.4 500 0.37 ± 0.04 183 ± 21 91 ± 10 The rate of removal was first-order (Fig. 2). The following first-order differ- ential equation allows the transport rate to be calculated, assuming all cells passing into the tunicate are removed from the sea water. -dC/dt= (n/V}C (5) In equation (5) C is cell concentration (cells/ml), u is the transport rate (ml/hr), and V is the volume (ml) of sea water in the beaker. The time constant u/V was determined by non-linear least squares analysis of equation (6) (Moore and Ziegler, I960), the integrated form of equation (5), evaluated using the boundary condition C = Co at t = 0 C = (6) Transport rates based on this assumption are at least minimum values, since not all particles may be removed. To determine if the system was affected by cell concentration or volumes in which the specimens were kept, experiments were performed at low and high cell concentration (and volumes) on the same specimen. In only one experiment (1.4 X 106 cells ml'1) was any effect of cell concentra- tion noted. This experiment terminated when the tunicate ejected a large mass of cells in a mucus-like matrix in the same manner as that previously described (JpYgensen and Goldberg, 1953) with respect to graphite uptake. In our case, however, the ejection was from the excurrent siphon. Otherwise the results (Table III) were independent of cell concentration. In a similar fashion, the volumes of sea water in which the animals were kept did not affect the rate of transport. 614 KUSTIN, LADD, McLEOD AND TOPPEN TABLE IV Transport rates for Ciona intestinalis determined by direct methods. 1'aliies of -H represent •instantaneous maxima. Column headings refer to the following: m, wet mass of Ciona; H, transport rate. m ti u/m g ml hr-i ml hr~lg~l 1.8 95 53 1.6 70 44 1.8 60-90 33-50 1.5 20 15 1.8 140 78 2.0 110 55 1.8 70 39 1.5 125 83 1.6 5 3 2.4 5 2 2.0 5 2.5 1.6 21 13 Direct method A direct method (Hamwi and Haskins, 1969) was modified to permit monitor- ing of the flow of a dyed sea water solution through a tunicate positioned in an upright manner. A variable flow Sage pump presented the solution at a right angle to the orifice of the incurrent siphon at a uniform rate (± 1%). The rate of flow of the dye solution was increased using the variable speed motor until it barely exceeded the rate at which the tunicate could remove it, whereupon the flow rate (assumed to be equal to that produced by the specimen) was recorded. The specimens responded to the dye concentration. At high concentrations they would transport sea water until the solution reached the stomach; they then reacted by violently squirting out the dye. Consequently, the lowest possible dye concentration consistent with visibility (A ^ 0.3 at 635 nm) was used. With this technique, great variability in rates for a single specimen was noted. There were periods when the tunicates would not transport sea water ; yet they appeared relaxed and normal. The values in Table IV are therefore instantaneous maxima and do not reflect prolonged transport. DISCUSSION Sea water transport in the tunicate serves the dual purposes of respiration and feeding. Neither of these functions requires continuous action. In fact, studies on the factors influencing rates of respiration have shown that, under stress, marine invertebrates may cease to transport sea water and respire anaerobically (Newell, 1973). Moreover, shutdown may follow exposure to low levels of oxygen (Holmes, 1973; Mangum and Van Winkle, 1973). Variability in transport has also been observed in our laboratory in connection with feeding and, of course, with squirting. Similar observations have been made for other ascidians; e.g., TRANSPORT RATES OF CIONA INTESTINALIS 615 Phallitsia mamillata C. (Hoyle, 1953). Thus, intermittant transport and vari- ability in the rate of transport mean that no single flow velocity is characteristic of a given species. Nor should a given method be expected to produce a single rate value. The average value does not represent a true instantaneous rate, but is useful for a number of applications. The maximum rate found with acceptable volume methods, namely 28 ml/hr/g wet weight, is less than the previously reported value of 80 ml/hr/g wet weight (Goldberg, McBlair and Taylor, 1951). We experienced difficulty in attaching Ciona, necessitating the use of the anaesthetic quinaldine. No mention of this problem is made in the previous report ; indeed, the specimen is shown in a figure therein pumping out of the incurrent siphon. Perhaps a larger specimen was at- tached in the earlier study, with less impairment to the ciliary action. Our experience indicates, however, that the adventitious occurrence of pressure heads is rarely avoided in a "constant level" device such as that previously described (Gold- berg, et al., 1951). We conclude that the earlier value represents a hydrostatically assisted transport rate. Clearance rate studies would appear to be a more reliable indicator of realistic rates, although this method also has drawbacks (Hamwi and Haskins, 1969; Holmes, 1973). The maximum value in the range of rates determined in this study, 185 ml/hr/g wet weight, is less than the previously reported value of 230 ml/hr/g wet weight (JoYgensen, 1966), although it is greater than that recorded in the volume method studies. The average value is 62 ml/hr/g wet weight. The range of values determined by the direct method overlaps with both the volume methods (lower limit) and clearance (the maximum directly determined value being 83 ml/hr/g wet weight). The average value of clearance and direct methods is 50 ml/hr/g wet weight, which represents a realistic reference value for com- putational purposes. It is interesting to compare this result with studies on the rate of vanadium isotope exchange in the tunicate Ciona intestinalis (Kustin, et al., in preparation). For, despite the intermittancy in transport and the variability in rate, the intrinsic time constant for vanadium assimilation from sea water is constant; i.e., it is independent of tunicate mass, for individual tunicates or groups of experimental materials. Suspension feeding and respiration are subject to control, accomplished through a variation in the rate of sea water transport. Nevertheless, even at its lowest level, the lowest transport time constant is greater than the vanadium assimilation time constant. In terms of the average value we find 50 ml/hr/g wet weight to be much larger than 6.5 X 10~2 ml/hr/g wet weight (Kustin, et al., in preparation). Hence, water transport is the more rapid of the two consecutive processes. Once the sea water source of vanadium is forced into the tunicate's body, an essentially chemical extraction process, subject to little or no control, takes place, and does not show variability. In addition to the above application, another example of the use that can be made of the average water transport rate data, is to estimate the physiological demand this activity places on the organism. Our visual inspection of the dyed sea water as it passes through the specimen shows a smoothly flowing column of fluid, which suggests that the Stokes-Hagen-Poiseuille laminar flow equation might 616 KUSTIN, LADD, McLEOD AND TOPPEN be appropriate (Cole, 1962). This observation is supported by physiological studies of mucociliary action (Schlesinger, 1973), which suggest that the ciliary stroke produces a wave-like motion inducing smooth flow, and by a Reynolds number calculation on our data for dona (see below). By using fluid mechanics (Shames, 1962), the amount of mechanical, i.e., pV (p is pressure), work done by the transport process can be related to the flow volume and the dimensions of the tube through which the volume of fluid flows. We made a transparent photograph of Ciona transporting dyed sea water. The liquid column within the tunicate is visible through the translucent body. The photo was placed in an enlarger and brought to life size ; the approximate dimen- sions of the column of dyed sea water were then measured. A 3.0 g specimen had dye solution in a U-shaped tube approximately 10 cm in length of narrowest diame- ter 0.08 cm and widest diameter 0.2 cm. (The plume of ejected fluid, clearly visible in the experimental aquarium and the photo, was approximately 0.15 cm wide and maintained this width for most of its length of 6-10 cm before dispersing). A diameter of 0.15 cm is therefore reasonable for the purpose of this estimation. At an average rate of 50 ml/hr/g wet weight a 3.0 g specimen achieves a flow volume of 150 ml hr'1 or 4.2 X 10"2 cm3 sec'1. Sea water flowing at this rate in a 0.15 cm diameter pipe has a Reynolds number of -—'36 and would therefore exhibit viscous laminar flow (Shames, 1962). The internal pressure generated by the ciliary "pump" is equal to the viscous loss in the tube through which the fluid passes. (The losses at the tube ends are negligible in comparison). The internal pressure, A/1, is given by p = 128 qLn/irD* (7) where q is flow volume (4.2 X 10'- cm3 sec"1), L is tube length exclusive of the "pump" (6.8 cm), /A is viscosity (1 X 10~2 dyne sec cm'2 for sea water at 20°), D is diameter (0.15 cm). The calculated internal pressure is 230 dyne cm~2 or <-~'2 mm H2O, which may be compared to values of 3-4 mm H2O for sponges (Parker, 1914; Bidder, 1923), and 2 mm H2O for Ascidia air a (Hecht, 1923). The energy lost at the end of the excurrent siphon (i.e., at the end of the tube) is 2.8 dyne cm~2, which is negligible compared with losses elsewhere. Flow volumes in excess of the maximum value reported here lead to significantly higher, physio- logically unreasonable internal pressures based on viscous losses. Moreover, the value of A/> <—> 2 mm H2O shows the considerable influence of even the smallest external pressure. SUMMARY Sea water transport rates of the tunicate Ciona intcstinalis were determined by measuring the volume of sea water transported through the specimen, and mea- suring the number of particles cleared by the specimen in a given time interval. The rate was also determined directly by matching the flow produced by the tuni- cate to that produced by a calibrated pump. Ciona transports sea water at variable rates; at times, it does not transport at all. The rate limits covering all tech- niques are : lower limit, 2.5 ml/hr/g wet weight and upper limit, 185 ml/hr/g wet TRANSPORT RATES OF CIONA INTESTINALIS 617 weight ; the average value based on clearance and direct measurements is 50 ml/ hr/g wet weight. Even at the lowest rate found, transport is rapid enough to en- sure complete mixing between sea water and reaction or absorption sites in the pharyngeal chamber, alimentary tract or atrial chamber. We conclude that the rate controlling process for absorption of oxygen, vanadate ions, micro-organisms or organic detritus is not the rate of passage of the feeding current, but rather the rate of the intrinsic absorption process such as complex formation, ion exchange or adsorption. LITERATURE CITED BIDDER, G. P., 1923. The relation of the form of a sponge to its currents. Quart. J. Microscop. Set., 67 : 293-323. CARLISLE, D. B., 1966. The ciliary current of Phalhisia [Ascidiacea] and the squirting of sea squirts. /. Mar. Biol. Ass. U.K., 46 : 125-127. COLE, G. H. A., 1962. Fluid Dynamics. Methuen, London, 238 pages. EDMONDSON, W. T., Ed., 1966. Marine Biology, Vol. 3, New York Academy of Science, New York, New York, 313 pages. Fox, D. L., H. U. SVERDRUP, AND J. P. CUNNINGHAM, 1937. The rate of water propulsion by the California mussel. Biol. Bull, 72 : 417-438. GOLDBERG, E. D., W. MCBLAIR, AND K. M. TAYLOR, 1951. The uptake of vanadium by tuni- cates. Biol. Bull, 101 : 84-94. GUILLARD, R. R. L. AND J. H. RvTHER, 1962. Studies on marine planktonic diatoms. Can. J. Microbiol, 8 : 229-239. HAMWI, A. AND H. H. HASKINS, 1969. Oxygen consumption and pumping rates in the hard clam Mercenaria mercenaria: A direct method. Science, 163: 823-824. HECHT, S., 1916. The water current produced by Ascidia atra L. /. Exp. Zoo/., 20: 429-434. HOLMES, N., 1973. Water transport in the ascidians Sty da clava Herdman and Ascidiclla aspcrsa (Muller). /. Exp. Mar. Biol. Ecol, 11 : 1-13. HOYLE, G., 1953. Spontaneous squirting of an ascidian, Phallusia mamillata Cuvier. /. Mar. Biol. Ass. U.K., 31 : 541-562. J0RGENSEN, C. B., 1952. On the relation between water transport and food requirements in some marine filter feeding invertebrates. Biol. Bull., 103: 356-363. J0RGENSEN, C. B., 1955. Quantitative aspects of filter feeding in invertebrates. Biol. Rev., 30 : 391-455. TO'RGENSEN, C. B. AND E. D. GOLDBERG, 1953. Particle filtration in some ascidians and lamelli- branchs. Biol. Bull., 105: 477-489. MANGUM, C. AND W. VAN WINKLE, 1973. Responses of aquatic invertebrates to declining oxygen conditions. Amer. ZooL, 13 : 529-540. MOORE, R. H. AND R. K. ZIEGLER, 1960. The solution of the general least squares problem with special reference to high speed computers. Report L. A. 2367 Los Alamos Scientific Laboratory, Los Alamos, Neil' Mexico. National Technical Information Service, 5285 Port Royal Road, Springfield, Virginia 22151. NEWELL, R. C., 1973. Factors affecting the respiration of intertidal invertebrates. Amer. ZooL, 13: 513-528. PARKER, G. H., 1914. On the strength and the volume of the water currents produced by sponges. /. Exp. ZooL, 16 : 443-446. SCHLESINGER, R. B., 1973. Mucociliary interaction in the tracheobronchial tree and environ- mental pollution. Biosciencc, 23 : 567-573. SHAMES, I. H., 1962. Mechanics of Fluids. McGraw-Hill, New York, New York, 555 pages. Reference: Biol. Bull., 147: 618-629. (December, 1974) THE ACCESSORY CELL AND YOLK HALO OF THE OOCYTE OF THE FRESHWATER TURBELLARIAN HYDRO LI MAX GRISEA (PLATYHELMIXTHES; PLAGIOSTOMIDAE) l W. DONALD XEWTON 2 Department of Zoology, University of North Carolina, Chapel Hill, Xorth Carolina 27514 Developing eggs are often associated with cells which, when numerous, form cell layers around the ovum, or, when single or few in number, appear to attach to the ovum in some way (Wilson, 1925). It is generally assumed that cells associated with eggs provide nutriment or act as nutrimental intermediaries to the growing oocyte (Raven, 1961). In the turbellarian Hydroliina.v grisea a single cell becomes intimately associated with the egg: this accessory cell completely surrounds the growing oocyte and remains with the developing egg throughout its growth period (Newton, 1970). At the time of fertilization, the accessory cell remains in place around the egg — the sperm must penetrate the accessory cell be- fore reaching the egg (Newton, 1970). Surrounding the oocyte and its investing accessory cell in Hydrolimax is the yolk halo (Hyman, 1938; Newton, 1970). In histological preparations, the yolk halo consists of two layers, the inner of fine radial fibers and the outer of spherules. Hyman (1938; page 16) concluded the yolk halo to be "... a non-cellular halo presumably of nutritive nature." There is no evidence of utilization of the yolk halo material by the growing oocyte or by the accessory cell (Newton, 1970). Professor Ulric Dahlgren, of Princeton University, first observed the accessory cell and yolk halo of the oocyte of Hydrolimax (Hyman, 1938). Opportunity for the present author to study some of Dahlgren's research notes and histological preparations of Hydrolimax grisea was made possible by a loan of the material from Dr. Ernst Kirsteur of the Department of Living Invertebrates of the Ameri- can Museum of Natural History. The observations of Dahlgren (unpublished), Hyman (1938) and Newton (1970) are supplemented in the present report by phase-contrast and electron-microscopic studies of the accessory cell and yolk halo. MATERIALS AND METHODS The material for this study was obtained from sources previously reported (Newton, 1970). Wet-mount squash preparations of live material were studied and photographed with a Zeiss Photomicroscope using phase optics. For electron microscopy, specimens of Hydrolimax were fixed whole, initially at room temperature or on ice, and subsequently diced with a razor blade in fixative on ice. The material was fixed for two hours in 3% glutaraldehyde in 0.1 M 1 Research performed in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Zoology, University of North Carolina at Chapel Hill, under the direction of Drs. Donald P. Costello and Catherine Henley. 2 Present address : Department of Biology, Colby College, Waterville, Maine 04901. 618 ACCESSORY CELL AND YOLK HALO 619 sodium cacodylate to which a trace of calcium chloride had been added, pH 7.1 (adjusted with HC1), followed by post-fixation in 2% osmium tetroxide, according to a procedure devised by D. P. Costello (personal communication). The material was dehydrated in a graded series of ethanol, passed through propylene oxide and embedded in Epon-Araldite. Sections were cut on a Sorvall Porter-Blum MT-2 ultra-microtome with glass or diamond knives and stained with 3% uranyl acetate (aqueous) alone or in combination with 0.5% lead citrate. The sectioned material was studied with a Zeiss 9S2 electron microscope. OBSERVATIONS In the life cycle of Hydrolimax, female germ cells first appear in the parenchyma of worms collected in early fall of each year. Oogonia occur singly, or, rarely, in groups of three or more, not united by cell connectives or processes ; they are free of association with other cells of the turbellarian parenchyma. Later many small, young oocytes are found free of cell associations in squash preparations of living material. The growing oocyte becomes closely associated with cells of the parenchyma — the accessory parenchymal cells (Newton, 1970; his Figs. 3, 6). One of the ac- cessory parenchymal cells becomes the accessory cell. Newton's Figure 7 shows an accessory cell fixed in the process of enveloping an oocyte. By the mid-Fall of each year, the growing oocytes of H. grisea have each been enveloped by an ac- cessory cell. The chronology of the association of oocyte and accessory cell can be correlated with cytological events of oogenesis : the accessory cell completely surrounds the young oocyte after the dissolution of the chromosome bouquet (diplotene) prior to the extensive growth of the oocyte. Figure 1 is a phase-contrast micrograph of an oocyte and its accessory cell. Although the phase halo in this micrograph makes it difficult to discern, the ac- cessory cell has completely enveloped the oocyte. Careful study through many planes of focus showed that the cytoplasm of the accessory cell is very thin on the side diametrically opposite the accessory cell nucleus. It may be significant to note that in squash preparations of living material accessory cells have not been observed in the process of enveloping oocytes. The process of envelopment may be fairly rapid. The squashing of living turbellarians disrupts the parenchyma and apparently prevents normal accessory cell-oocyte contact. The oocyte in Figure 2 is enveloped for the most part by a thin layer of ac- cessory cell cytoplasm. In the vicinity of the accessory cell nucleus, the cytoplasm is thicker. Apparently the enveloping layer of accessory cell cytoplasm becomes attenuated by the relatively rapid growth of the oocyte. Eventually, however, the accessory cell grows, for in later stages the layer of accessory cell cytoplasm sur- rounding the growing oocyte thickens (Figs. 3, 4) (cf. Newton, 1970). In squash preparations of live turbellarians collected in the late fall, the ac- cessory cell, with its enveloped oocyte, is surrounded by many accessory parenchy- mal cells (Fig. 3). The number of accessory parenchymal cells increases. By mid-Winter the accessory parenchymal cells have apparently been replaced by the spherules of the yolk halo (Fig. 4) (cf. Newton, 1970). The accessory parenchy- 620 W. DONALD NEWTON S^ ,V^9 *r ~ '<%** * tf £ '-*^.*1 k - • >*j <*. & '•\, . f. . • *•"* /^^ sr _^ on . ddc oo , QV ACCESSORY CELL AND YOLK HALO 621 mal cells, however, become transformed into the spherules of the outer portion of the yolk halo (Fig. 4) . The accessory cell is in close apposition to the oocyte (Figs. 1-5). In some specimens fixed, sectioned and stained for light microscopy, a wide space exists between the inner cell boundary of the accessory cell and the surface of the oocyte (Newton, 1970; his Figs. 8, 12). The space is an artifact of fixation caused by shrinkage of the oocyte, for no space is seen between accessory cell and oocyte in living material studied by phase optics. A shrinkage space exists between oocyte and accessory cell in some specimens fixed and sectioned for electron microscopy. Generally, however, electron micrographs show the accessory cell to be separated from the oocyte surface by a distance no greater than 0.2 /*, (Fig. 5). Cytoplasmic processes from the inner surface of the accessory cell penetrate the oocyte no deeper than 0.8 /* (Figs. 5, 6, 7). The cytoplasm of the accessory cell and the cytoplasm of the oocyte do not appear to intermix. The membranes of the two cells remain intact. The tips of the accessory-cell projections often touch the surface of the oocyte (Fig. 6) and prehaps fuse with the cell membrane of the oocyte, as indicated in Figure 6. The space between oocyte and accessory cell is empty in specimens prepared for light microscopy. In material prepared for electron microscopy amorphous particles of medium electron density are found between oocyte and accessory cell (Figs. 5, 6, 7). Similar amorphous particles are located within membrane-bound vesicles of the accessory cell (Figs. 5, 6, 7). An open vesicle in Figure 6 suggests an interchange of amorphous particles between accessory cell and the intercellular space. In sections stained with uranyl acetate or with uranyl acetate in combination with lead citrate, electron-dense bodies of un- known significance are located on or immediately beneath the plasma membrane (Figs. 5, 7, and especially 6) of the oocyte. Sections through accessory cells show that the cytoplasm immediately beneath the outer and inner membranes is uniformly compact and generally devoid of organelles and inclusions (Figs. 6, 7). The layers of relatively homogeneous cytoplasm at the surfaces of the accessory cell are here called the outer cortex and the inner cortex. The cortices are approximately 730 A thick. The cortical cyto- plasms are granular, consisting of dark particles, approximately 100 A or less, in a ground substance of medium electron density (Figs. 6, 7). Between the cortices are all manner of membranous elements and some granular components (Figs. 5- 9). Mitochondria are the only distinctive organelles. Whorls of membrane, mem- branous cisternae of irregular shape containing wisps of electron-dense material, and endoplasmic reticulum are the major components of inter-cortical cytoplasm. Microvilli arise from the outer surface of the accessory cell (Figs. 5, 7) and often project 3 to 4 /*. into cells of the turbellarian parenchyma (Figs. 5, 8). The FIGURES 1-4. The events depicted in this plate occur over a five- to six-month period. 7he young oocyte is surrounded by an accessory cell (Fig. 1) before oocyte growth (Fig. 2). Accessory parenchymal cells become associated with the outer surface of the accessory cell (Fig. 3) and transform into the spherules of the outer portion of the yolk halo (Fig. 4) ; phase- contrast micrographs; Figures 1, 2, 4, 784 X ; Figure 3, 980 X ; ac, accessory cell; an, ac- cessory cell nucleus; ap, accessory parenchymal cell; gv, germinal vesicle (nucleus) of oocyte; oo, oocyte ; yh, yolk halo ; yi, inner portion of yolk halo ; ys, spherules of outer portion of yolk halo; scale = 10 /*. 622 W. DONALD NEWTON £% «fc' v.-vjf.* - •• * : * ••-;•- • - . • .. : • ;" ?•>' • ••*•*. ',- ?•&• ' ' '' • ••-*'- '••' ^?r-.>-VJ •• it . -^ *•• • • ^ --•:,. 6J" •%r ....;. .,-*; . . •am' 4- -!.' .--.*•. Jfe ^' .*' .- •- .' '•:' ' '• . * --. .. - amv • '•-• >-,r. .;.;.: -...^L, I'jt ••!»»' '. i -' -•"•'"-•,'•.•''. -&•<' ;>V~ '••>.',-•" •.-""• •;. , • ^..v<, • ;i.r.,:..- '. l^ •'-- ->. • • ACCESSORY CELL AND YOLK HALO 623 microvilli of the accessory cell do not fuse with adjacent cells but occupy mem- brane-delimited canals within the cytoplasm of adjacent cells (Figs. 5, 8). The appearance by light microscopy of the inner portion of the yolk halo, consisting of fine radial fibers (Fig. 4) (Newton, 1970; his Fig. 17), is explained by the numer- ous microvilli which extend from the accessory cell. Electron-dense filaments are oriented longitudinally within microvilli (Figs. 7, 9) ; no filaments are observed in the projections from the inner surface of the accessory cell into the oocyte (rf. Figs. 5, 6, 7). Four to eight filaments are observed in transverse profile of micro- villi (Fig. 9). Evidence for the transformation of accessory parenchymal cells into the outer layer of the yolk halo was first obtained by the present author from some of the slides of Dahlgren. In Dahlgren's material, fixed with hot Flemming's fluid and stained with an undesignated hematoxylin, a basophilic material is present in the spherules of the outer layer of the yolk halo ; this material is not bounded by a nuclear membrane. Dahlgren suggested in his research notes that the nucleus of the accessory parenchymal cell may be degenerating in the process of transforma- tion by the accessory parenchymal cell into a yolk-halo spherule. In material pre- pared for light microscopy in Newton's (1970) study, evidence for nuclear de- generation is lacking: no basophilia is seen in the yolk spherules. Newton (1970; his Fig. 12) suggested that the yolk halo may be secreted by the accessory par- enchymal cell. The discrepancy between Dahlgren's observations and those of Newton's study is probably a result of the use of different methods of fixation and staining. Evidence that Dahlgren's interpretation is correct comes from electron micrographs of the outer layer of the yolk halo (Figs. 8, 10). The spherules of the yolk halo are composed of organelles, primarily vesicles, but also endoplasmic reticulum, a few scattered mitochondria, some microtubules, all bounded by a cell membrane, and without nuclei. After Feulgen-Fast Green the yolk halo spherules are Feulgen-negative. The vesicles (Fig. 10) are conspicuous because of the homogeneous, electron-dense material unevenly distributed within them. Trans- formation of accessory parenchymal cells into the spherules of the outer portion of the yolk halo remains an enigma : the function of the spherules has not been ascertained. Evidence for the utilization of the yolk-halo spherules by the oocyte is lacking. DISCUSSION According to the system of classification of eggs proposed by Korschelt and Heider (1902; cited in Wilson, 1925; and Raven, 1961), the egg of Hydrolimax is of the nutrimental type and its accessory cell is a nurse cell. The accessory cell FIGURE 5. The relationship between accessory cell and oocyte is marked by the cyto- plasmic processes extending from the inner surface of the accessory cell into the developing egg. Microvilli extend from the outer surface of accessory cell into membrane delimited canals in cells of the turbellarian parenchyma; uranyl acetate; 26,871 X; am, amorphous particles ; ci, swollen cisternae of endoplasmic reticulum ; cp, cytoplasmic process ; er, endo- plasmic reticulum ; ic, inner cortex ; mw, membranous whorl ; mv, microvillus ; oc, outer cortex ; scale = 1 p. 624 W. DONALD NEWTON •T * •> T-. *«'•" :•» • *-"ti^^A»« aft^s?^^^?/^ ;^«*-'* "• ' >'»-'-' fe*?i^5M^ ACCESSORY CELL AND YOLK HALO 625 of Hydrolimax is unique among the plagiostomids (Hyman, 1938) and apparently unique among the animal phyla. The relationship between accessory cell and oocyte in Hydrolimax is one of close apposition of the cells involved, and of cytoplasmic projections into the ooplasm from the inner surface of the accessory cell. There is no evidence that the cytoplasms of the accessory cell and of the oocyte intermingle. Gondos (1970) observed close apposition of cell membranes of germ cells and granulosa cells, and the presence of interdigitating cytoplasmic projections from granulosa cells into oocytes, of young rabbits. In the human, "localized projections from both the oocyte and the follicle cells interdigitate in the intercellular space and appear to reflect a constant active interchange between the oocyte and its follicle wall" (Hertig and Adams, 1967; page 668). The extension of accessory cell processes into the oocyte in Hydrolimax may likewise reflect an interchange between the two cells. Microvilli extend from the outer surface of the accessory cell, into extra- cellular spaces or into membrane-delimited canals of adjacent parenchymal or oviducal cells. Since microvilli are generally present on cells which have absorption as a primary function (Bloom and Fawcett, 1968), the function of the outer sur- face of the accessory cell may be to absorb nutrient material for transport across the accessory cell to the growing oocyte. The speculation is amenable to simple experi- mental analysis, as the oocyte and its surrounding accessory cell are easily isolated from the parenchyma of young turbellarians. The author is grateful to Dr. Donald P. Costello and Dr. Catherine Henley for advice during the course of this study and for helpful criticisms during the prepara- tion of the manuscript. Mrs. Wilma Hanton is very much appreciated for her patient advice and help with electron microscopy. The work was aided by a grant to Dr. Costello and Dr. Henley from the National Institutes of Health, GM 15311. SUMMARY 1. A single cell, the accessory cell, completely surrounds the growing oocyte of H. grisea. 2. Cytoplasmic processes from the inner surface of the accessory cell extend into the oocyte. The cytoplasms of the accessory cell and of the oocyte do not intermix. 3. Microvilli extend from the outer surface of the accessory cell. 4. It is suggested that the accessory cell is a nurse cell which transports nutrient material to the growing oocyte. FIGURE 6. Small arrows indicate apparent contact of tips of cytoplasmic process with oocyte membrane. Large arrow points to opening of a cytoplasmic vesicle of accessory cell into intercellular space; uranyl acetate, 91,140X ; cv, cytoplasmic vesicle; db, dense body; mi, mitochondrion ; scale = 1 /*. £ Li^Wwmw '^P:F P > :V-ra* j»S*K> £tb ' "'" fi .' A\f" * **?. •{ , Pi s •; ? • ^v:)fc^:^^Mim - ,4, ^^fiLiii - '-* , ' .*&rj ^* . .»•' *,' r FIGURE 7. The ultrastructure shown in this electron micrograph quite clearly suggests the function of the accessory cell : absorption of nutriment for transport to the growing oocyte. Microvilli absorb nutrients which are then transported or stored until required in the voluminous cisternae. Amorphous particles may represent processed material to be utilized by the egg; uranyl acetate, lead citrate: 91,140X ; scale = 1 /j.. •? »$ |/.. • - . 9 • te * * FIGURES 8 and 9. Many microvilli extend into parenchymal cells and remnants of ac- cessory parenchymal cells, now changed to yolk spherules (Fig. 8). The microvilli in Figure 9 are free of enveloping cytoplasm. Arrows point to transverse sections which show distinct profiles of microvillar filaments; uranyl acetate; Figure 8, 5468X ; Figure 9, 58,800 X ; scale = 1 n. FIGURE 10. A single yolk spherule is composed of cellular elements without nucleus. Linear densities at surface suggest that the yolk spherule is membrane bound; uranyl acetate; 26,871 X; me, membrane; mt, microtuble; sv, spherule vesicle; scale = 1 ft. ACCESSORY CELL AND YOLK HALO 629 5. The yolk halo consists of two layers, each of different origin. The inner layer is composed of the numerous microvilli which extend from the accessory cell into membrane-delimited canals within cells of the turbellarian parenchyma. The spherules of the outer portion of the yolk halo are derived from accessory par- enchymal cells. The function of the yolk-halo spherules is unknown. LITERATURE CITED BLOOM, W., AND D. W. FAWCETT, 1968. A Textbook of Histology. [Ninth Edition] W. B. Saunders Co., Philadelphia, 858 pp. GONDOS, B., 1970. Granulosa cell-germ cell relationship in the developing rabbit ovary. /. Embryol. Exp. Morphol., 23 : 419-426. HERTIG, A. T., AND E. C. ADAMS, 1967. Studies on the human oo'cyte and its follicle. I. Ultrastructural and histochemical observations on the primordial follicle stage. /. Cell Biol, 34 : 647-675. HYMAN, L. H., 1938. North American Rhabdocoela and Alloeocoela. II. Rediscovery of Hydroliinax grisea Haldeman. Amcr. MHS. Nor. 1004: 1-19. KORSCHELT, E., AND K. HfiiDER, 1902. Lchrbuch dcr vcrgleichcndcn Entwicklungsgeschichte der wirbellosen Thicre. Gustav Fischer, Jena, 750 pp. NEWTON, W. D., 1970. Oo'genesis of the freshwater turbellarian Hydrolimax grisea (Platy- helminthes; Plagiostomidae) with special reference to the history of the supernumer- ary asters and central bodies. /. Morphol., 132 : 27-46. RAVEN, C. P., 1961. Oogenesis: The Storage of Developmental Information. New York, Pergamon Press, 274 pp. WILSON, E. B., 1925. The Cell in Development and Heredity. New York, Macmillan Co., 1232 pp. Reference: Biol. Bull, 147: 630-640. (December, 1974) MODIFICATION OF SEA ANEMONE BEHAVIOR BY SYMBIOTIC ZOOXANTHELLAE : PHOTOTAXIS VICKI BUCHSBAUM PEARSE 1 Hopkins Marine Station, Pacific Grove, California 93950 Symbiosis between an animal host and a photosynthetic symbiont is unlikely unless the animal spends considerable time in the sun. So it is not surprising that a variety of invertebrates with unicellular endosymbiotic algae display positive phototaxis. Probably the first report of phototaxis in an algae-bearing inverte- brate is Trembley's (1744) observation that green hydras gather on the lighted side of a glass vessel. Later investigations show, however, that other, non-sym- biotic hydra species (Wilson, 1891) and green hydras which have lost their symbionts (Whitney, 1907) also respond positively to light. The acoel flat worms Convoluta rose off cnsis and C. convoluta show phototactic behavior correlated with their usual habitats (Keeble, 1910). Again, the symbiotic algae are not essential to the response. The action spectrum for phototaxis does not correspond to that of chlorophyll, and light apparently is received by the carotenoid-containing eyespots of the animal. Even worms just a few hours old and still white respond phototactically. Vandermeulen, Davis, and Muscatine (1972) mention that Placobranchus ianthobapsus, a saccoglossan opisthobranch gastropod with func- tional chloroplasts derived from siphonaceous green algae, displays positive photo- taxis even when chloroplast photosynthesis is effectively eliminated by chemical inhibition. In other cases, however, phototaxis appears to depend on the intact, functional symbiosis. The ciliate Paramecium biirsaria with zoochlorellae shows positive phototaxis, and the action spectrum of the response corresponds to the absorption spectrum of chlorophyll (Engelmann, 1882). P. bursaria without algal symbionts does not show positive phototaxis (Siegel, 1960), nor do any of the other, non-symbiotic species of Paramecium, show directed light responses (Gelber, 1956; Halldal, 1964). Besides the green hydras, there are several other examples of tactic responses to light among algae-bearing cnidarians. Coral planulas containing zooxanthellae swim toward light ; those without zooxanthellae display no light response (Kawa- guti, 1941, 1944; Atoda, 1953). Zahl and McLaughlin (1959) report that Condylactis sp., a Caribbean sea anemone with zooxanthellae, moves out of direct sunlight into shade, but they did not establish whether the anemones respond positively to light of lower intensity. Specimens of Condylactis that are kept in the dark lose their symbionts and also lose their phototactic response. Cotte (1921, 1922) describes inconsistent phototactic behavior in Anemonia snlcata, a European sea anemone with zooxanthellae. Of the few other sea anemones in which phototaxis has been tested, all non-symbiotic species, most appear either indifferent or negative to light (e.g., Hargitt, 1907; Parker, 1917; Ottaway, 1973). 1 Present address : Division of Natural Sciences, University of California, Santa Cruz, California 95064. 630 ANEMONE PHOTOTAXIS AND SYMBIOSIS 631 Most littoral anemones live in shaded habitats. However, the two algae- bearing sea anemones of the west coast of North America, Anthopleura elegan- tissima (Brandt, 1835) and A. xanthogramnrica (Brandt, 1835), are unusual in distribution ; most occupy rocks and tidepools in full sun. This unusual habit, apparently associated with the symbiotic relationship, led me to svispect that these anemones might show responses to light which were different from those of most actinians. Anthopleura elegantissima is particularly suitable for studying possible rela- tionships between behavior and symbiosis because the animals are hardy and abundant ; exposed for collection at almost any low tide ; mostly conveniently small (although solitary individuals may be as large as many large A. .rantho- graniinica} ; and reproduce asexually by dividing, forming distinct clonal beds with tens to hundreds of genetically identical individuals (see Francis, 1973). Clones can be found in light conditions grading from direct sunlight to deep shade, with the number of algal symbionts correspondingly graded. It is possible to introduce zooxanthellae into anemones that lack them (Trench. 1971c); and individuals with algae can be made to lose their symbionts, yet remain apparently active and healthy (see Methods). As part of a more extensive study on the symbiosis of A. elegantissima with zooxanthellae (Buchsbaum, 1968; Pearse, 1974), I have tested anemones of this species, with and without symbiotic algae, for evidence of phototactic behavior and have tried to discover whether their behavior and sym- biosis might be related. METHODS Specimens of the sea anemone Anthopleura elegantissima with abundant sym- biotic algae were collected from two rocky intertidal areas in central California. Specimens with few and no zooxanthellae were collected from intertidal rocks or cement pillars beneath the cannery buildings, Cannery Row, Monterey. Only golden-brown dinoflagellate zooxanthellae were present as algal symbionts in the anemones. Tests for phototaxis The sea anemones were tested for phototaxis in shallow seawater tables. Eight 112-cm Sylvania Lifeline fluorescent tubes were suspended 30 cm above the table; they were left on continuously during tests. Incident illumination at the center of the table was 700 foot-candles, as measured with a Photovolt 200 photometer, also used for intertidal light measurements. A continuous flow of sea water along the table maintained the temperature at 14° C ± 1° C. The water was approxi- mately 4 cm deep and covered the anemones. A wooden board supported just above the surface of the water shadowed half of the experimental area. The total area was 22 cm across by 125 cm long. At the beginning of each of the 19 experiments, 25-70 anemones were placed in the table under uniform illumination. After the anemones had all attached to the surface of the table, their positions were recorded, and the wooden board was arranged. The positions of the anemones, along the table as well as with 632 VICKI BUCHSBAUM PEARSE respect to the lighted and shadowed halves, were then recorded at intervals, usually daily. The day after the wooden board was set in place was counted as the first day. The small number of animals occasionally found on the borderline be- tween lighted and shadowed areas was divided equally between the two scores. The anemones were not fed. Elimination of zooxanthellae Maintenance in the dark, used successfully to eliminate symbiotic algae from various invertebrate hosts (see, e.g., Zahl and McLaughlin, 1959), was not very satisfactory for A. elegantissima because of the long time required. The anemones were maintained in a darkroom in running sea water and exposed to dim light for only a few minutes every few weeks, in order to take sample bits of tissue, cut from the tentacles and sides of the column. Only after 12 weeks in the dark did most of the tissue samples clearly have fewer zooxanthellae than at the beginning ; after nearly three times this long (34 weeks), a few zooxanthellae were still found. When the treatment was terminated at 48 weeks, no zooxanthellae were found in the tissue samples, nor did zooxanthellae return when these animals were subsequently kept in the light. The most satisfactory method found for ridding the anemones of their symbiotic zooxanthellae was exposure to elevated temperatures. The anemones were kept in warm sea water for one to several days, then put back into running sea water at 13°-15° C for 2-3 weeks. During this time they egested mucus- wrapped pellets of zooxanthellae until no algae were found in their tissues. The anemones appeared healthy and normal throughout the treatment, as long as the heated sea water was aerated continuously and changed frequently. The effectiveness of exposure to temperatures from 25° -37° C, for varying periods of time, was tested (see Buchsbaum. 1968 for details). The time and temperature ranges that were both effective and well-tolerated by the anemones were very narrow ; 30°- 32° C for 48 hours was most successful. Such heat-treated anemones have been kept in the light for more than a year, never regaining any zooxanthellae, and this appears to be a convenient and reliable method of obtaining anemones without zooxanthellae for experimental purposes. RESULTS Sea anemones with zooxanthellae Sea anemones with zooxanthellae, gathered from sunny intertidal habitats, always distributed themselves in the lighted half of the experimental area, whether the animals were initially equally distributed between the lighted and shadowed halves (Figure 1A) or placed entirely in the shadowed area (Figure 2A). The time course of the response varied in different experiments. The animals repre- sented in Figure 1A were the slowest observed; 95% were found in the lighted area after 13 days. In another, similar experiment 93% were found in the lighted area after only 4 days. In the experiment represented in Figure 2A, although all ANEMONE PHOTOTAXIS AND SYMBIOSIS 633 90 80 70 .= 50 «-» c o> u 40 J 30 20 . 10 \ 8 Days 10 12 14 FIGURE 1. Percentage of anemones in lighted half of seawater table, after being initially randomly distributed. At 2 weeks: (A.) Anemones with zooxanthellae, from a sunny habitat, showed a significant positive light response (x2:=30.4, P < 0.001, N = 38). (B.) Anemones without zooxanthellae showed no significant light response (x2 — 0, P > 0.99, N = 51); zoo- xanthellae were eliminated from these animals after exposure to elevated temperature. (C. ) Anemones with zooxanthellae, from a shaded habitat (about 9% of light intensity measured in open intertidal), showed a significant negative light response (x2— 16.0, P c 0.001, N — 68). of the animals were initially placed in the shadowed half, 92% had moved to the lighted area after only 5 days. Sea anemones -without zooxanthellae Sea anemones without zooxanthellae distributed themselves randomly with respect to the lighted and shadowed areas. When equal numbers were initially placed in each half of the experimental area, their distribution remained random with respect to light (Figure IB ). When all were initially placed in the shadowed half, they slowly emerged into the lighted area until equally distributed between the lighted and shadowed portions (Figure 2B, C). Again, the time course varied in different experiments. Sea anemones with few zoo.vanthcllae Sea anemones gathered from partially shaded intertidal habitats, with incident light down to about 10% of that measured in open intertidal areas, were found to harbor limited numbers of zooxanthellae. Such animals, when tested for photo- 634 VICKI BUCHSBAUM PEARSE A D 10 12 14 Days FIGURE 2. Percentage of anemones in lighted half of seawater table, after being initially distributed all in the shadowed half; (A-C) anemones with zooxanthellae when collected, from a sunny habitat, all from the same clone. (A.) Anemones with zooxanthellae showed a sig- nificant positive light response (at 5 days: x" = 37.2, P < 0.001, N = 52). (B.) Anemones having lost zooxanthellae after exposure to elevated temperature showed no significant light response (at 12 days: X2=1.2, P — 0.27, N = 41). (C.) Anemones having lost zooxanthellae after maintenance in darkness showed no significant light response (at 14 days: x2— 1.1, P — 0.30, N = 34). (D.) Anemones with zooxanthellae, from a shaded habitat (about 9% of light intensity measured in open intertidal), showed a significant negative light response (at 13 days: xs = 25.0, P < 0.001, N = 25). taxis as above, always occupied the shadowed half of the experimental area. If they were initially equally distributed between lighted and shadowed halves, they moved into the shadowed portion (Figure 1C). If initially placed entirely in the shadowed area, they remained there, none emerging into the light (Figure 2D). Light intensity The plasticity of behavior indicated by these results suggested the possibility that the anemones with zooxanthellae which I collected from shaded habitats, and which reacted negatively to the light intensity regularly used in the experiments (Figures 1C, 2D), might respond positively to a lower light intensity. To test this possibility, I selected anemones from a shaded habitat where midafternoon light readings were 200-250 foot-candles. Distributed randomly in a seawater table, these animals showed a negative response to the usual light intensity of 700 foot-candles, moving into the shadowed half of the table (Figure 3, left). After their distribution had remained stable for 2 weeks, the light intensity was reduced ANEMONE PHOTOTAXIS AND SYMBIOSIS 635 to 250 foot-candles; and within a few days the anemones began to move out into the lighted area, reversing from a negative to a positive response (Figure 3, right). Elimination of zooxanthellae All of the results just described suggested that phototaxis in these anemones depends directly on the presence of zooxanthellae. However, since naturally occurring anemones with and without zooxanthellae were collected from different habitats and belonged to different clones, several other hypotheses seemed possible. The response might depend on previous exposure to appropriate light intensities ; pigments produced by the anemones only in the light might be involved ( see Buchsbaum, 1968). The absence of phototaxis in animals from dark habitats might then reflect only their lack of previous exposure to light and not their lack of zooxanthellae. Alternatively, phototactic behavior might be developed only after the acquisition of zooxanthellae and then retained. In this case, animals without algae might display phototaxis if they had previously harbored zooxan- thellae. Finally, phototaxis might have a genetic basis, since an individual that actively orients to light, either as a settling planula or attached adult anemone, is 80 70 60- • 50 H 4-» c 09 0) CL Light intensity 700 f.c. Light intensity reduced to 250 f.c. \ 40 T\ 30- 20 9 8 10 12 14 28 30 32 34 36 38 40 42 44 Days FIGURE 3. Percentage of anemones in lighted half of seavvater table, after being initially randomly distributed. Anemones with zooxanthellae, from a shaded habitat ("about 47% of light measured in open intertidal), showed a significant negative light response at 700 foot- candles (at 14 days : x2 = 12.2, P < 0.001, N = 51) and later a significant positive light response at 250 foot-candles (after 16 days: x2 = 14.2, P < 0.001, N = 51). 636 VICKI BUCHSBAUM PEARSE 01 c o -C u Anemones with zooxanthellae Q — — Q Anemones without zooxanthellae 12 FIGURE 4. Movement of anemones up-current in the seawater tables. Data from 4 experiments with 4 different clones of animals (N = 34, 30, 38, 32). Changes in average position along the table (sum of anemone positions/no, of anemones) were taken as an index of movement, since individual anemones could not be conveniently followed. more likely to become host to light-requiring symbionts than an animal that is indifferent to light. To test these possibilities, I collected sea anemones from a single asexually produced clone, all with abundant zooxanthellae. Some were kept as controls in ANEMONE PHOTOTAXIS AND SYMBIOSIS 637 natural light in running sea water at ambient temperature. Others were induced to lose their zooxanthellae by subjecting them briefly to heat or darkness (see Methods). After these treatments, tissue samples examined by light microscopy revealed no zooxanthellae; nor did these anemones ever re-establish symbiosis with algae, although subsequently kept in the light. All of the anemones — presumably genetically identical, collected from the same small intertidal area, and originally harboring many zooxanthellae — were tested for their response to light. The control individuals with zooxanthellae showed a positive phototactic response (Figure 2A), while those which had lost their zooxanthellae distributed themselves equally between lighted and shadowed areas (Figures IB, 2B, C). I found no differences in phototactic behavior be- tween anemones occurring naturally without zooxanthellae, collected from dark habitats, and anemones in which I induced loss of zooxanthellae by exposure to heat or darkness. Activity controls Another hypothesis considered was that the apparent failure of anemones with- out zooxanthellae to respond to light might simply reflect inactivity or slowness of movement. However, in experiments in which all animals were initially placed in the shadowed area, anemones with zooxanthellae moved into the light at a rate fast enough to reveal any light response within the duration of the experiment (Figure 2B, C) but with a pattern quite distinct from the positive and negative responses of the anemones with zooxanthellae (Figure 2A, D). Moreover, in all experiments, regardless of light-directed movements across the seawater table, the anemones moved along the table, against the current of sea water. This movement was followed by noting the position of each anemone along the table each time its position with respect to light was recorded. Changes in the positions of the anemones along the table in 4 experiments are represented in Figure 4. Although rates of movement varied in different experiments, there appeared to be no dif- ference in the rates of anemones with and without zooxanthellae. DISCUSSION Under the conditions of these experiments, sea anemones with zooxanthellae always displayed phototaxis, either positive or negative depending on the light intensity. Anemones without zooxanthellae — even those that had previously harbored zooxanthellae and that were genetically identical clone-mates of photo- tactic individuals — never displayed phototaxis, appearing completely indifferent to light and shade. These observations indicate that phototactic behavior in the sea anemone Anthopleura elegantissima is not a fixed species character, but rather depends directly on the presence of endosymbiotic zooxanthellae. The mechanism by which the symbionts influence their anemone hosts is un- known, but possibly involves some metabolite consumed or produced in photo- synthesis by the algae. Production of oxygen or any organic compound, removal of carbon dioxide or other metabolic by-products, or release of any stimulant or 638 VICKI BUCHSBAUM PEARSE inhibitor by the algae could all work in the same way. The difference in light intensity on two sides of the animal would result in different degrees of activity by the algae. The quantities of both animal and algal pigments vary with the light intensities under which the organisms live (Buchsbaum, 1968), and light intensi- ties either higher or lower than the accustomed range might be expected to reduce algal photosynthesis. The gradient thus established within the animal would direct its movement. Determination of the action spectrum for phototaxis would provide evidence for or against some such photosynthetic mechanism. In the ciliate Parameciuin bursaria, which responds phototactically only under conditions of limited oxygen, the evidence suggests a mechanism involving oxygen production by the symbiotic algae (Jennings, 1915; Stanier and Cohen-Bazire, 1957). In this context it would be particularly interesting to know the phototactic behavior of specimens of A. clcgantisshna from the northwest coast of North America which are symbiotic with a green unicellular alga. Unlike zooxanthellae, which release approximately 50 % of their photosynthetically-fixed carbon to the anemone tissues in a variety of organic compounds (Trench, 1971a, 1971b), these green symbionts fix carbon photosynthetically at appreciable rates but re- lease less than 2% (Muscatine, 1971). Yet in symbiosis with both kinds of algae, the anemone tissues presumably receive oxygen and give up carbon dioxide and other inorganic nutrients to the algae. Tests for phototactic behavior in anemones with the green symbionts might therefore provide clues as to which algal activities influence phototaxis. The symbiosis between A. clegantissuna and zooxanthellae was the first in which transfer of photosynthetically-fixed carbon from an alga to its host was directly demonstrated (Muscatine and Hand, 1958). Studies since then have characterized the biochemistry of the algal contribution (Trench, 1971a, 1971b) and have shown that the zooxanthellae slow weight loss of the anemone during starvation in the light (Muscatine, 1961a; Buchsbaum, 1968). The physiological mechanism for a potential selective advantage in the symbiosis is thus well estab- lished. Phototaxis seems to be a behavioral mechanism ensuring the effectiveness and continued maintenance of the symbiosis. If anemones with zooxanthellae are placed in the dark, the number of algal cells and total chlorophyll per unit weight of anemone tissue decrease, eventually to undetectable levels (Muscatine, 1961b; Buchsbaum, 1968). Intertidal anemones in a sunny habitat, if suddenly shaded (for example, by growing seaweeds or by debris), may avoid such reduction in symbionts by positive phototaxis. On the other hand, anemones with zooxanthellae adapted to partially shaded habitats, if suddenly exposed to increased light intensity, also suffer a reduction in numbers of algae and total chlorophyll, the tentacles and oral disk becoming especially pale. After a few days in full sun, microscopic examination of tentacle tissue reveals only rare, bleached algal cells (Buchsbaum, 1968). Shade-adapted anemones may avoid such loss of algae by negative phototaxis. How much these anemones actually move under normal conditions in the intertidal is not known, but their phototactic behavior in laboratory experiments suggests that they may effectively control their light environment in this way. ANEMONE PHOTOTAXIS AND SYMBIOSIS 639 This study was supported by a predoctoral fellowship from the National Institutes of Health, carried out at the Hopkins Marine Station, Pacific Grove, California, and submitted to Stanford University as part of a doctoral dissertation. I am grateful to the faculty, staff, and students of the Marine Station, especially my advisory committee, Drs. J. H. Phillips, D. P. Abbott, and W. L. Lee; to Dr. J. S. Pearse for reading the manuscript; and to the Division of Natural Sciences, University of California, Santa Cruz for library and secretarial assistance. SUMMARY The sea anemone Anthopleura elegantissima, with and without endosymbiotic zooxanthellae, was tested for evidence of phototactic behavior. Anemones with zooxanthellae always displayed phototaxis, either positive or negative depending on the experimental light intensity and the light intensity of the habitat from which the animals were taken. Anemones without zooxanthellae — even those that had previously harbored zooxanthellae and that were genetically identical clone- mates of phototactic individuals — never displayed phototaxis, appearing completely indifferent to light and shade. The results indicate that phototaxis in this sea anemone depends directly on the presence of its symbiotic algae. It is suggested that the flexible phototactic behavior of the anemone may play an important role in favorably regulating the amount of light to which the zooxanthellae are exposed. LITERATURE CITED ATODA, K., 1953. The larval and postlarval development of the reef-building corals. Sci. Rep. Tohoku Univ., Series 4, Biol., 20 : 105-121. BUCHSBAUM, V. M., 1968. Behavioral and physiological responses to light by the sea anemone Anthopleura elegantissima as related to its algal endosymbionts. Doctoral disserta- tion, Stanford University, 123 pp. COTTE, J., 1921. Sur le phototropisme des Actinies. C. R. Soc. Biol. Paris, 85 : 188-190. COTTE, J., 1922. fitudes sur les comportements et les reactions des Actinies. Bull. Inst. Oceanogr., Monaco, 410: 1-44. ENGELMANN, T.W., 1882. Ueber Licht- und Farbenperception niederster Organismen. Pfliig. Arch. Ges. Physio I. , 29: 387-400. FRANCIS, L., 1973. Clone specific segregation in the sea anemone Anthopleura elegantissima. Biol, Bull, 144: 64-72. GELBER, B., 1956. Investigations on the behavior of Paraincciitin aiirclia: The effect of the presence and absence of light on the occurrence of a response. /. Genet. PsychoL, 88: 31-36. HALLDAL, P., 1964. Phototaxis in Protozoa. Pages 277-296 in S. H. Hutner, Ed., Bio- chemistry and Physiology of Protozoa. Academic Press, New York. HARGITT, C. W., 1907. Behavior of sea-anemones. Biol. Bull, 12 : 274-284. JENNINGS, H. S., 1915. Behavior of the Loit'cr Organisms. Columbia University Press, New York, 366 pp. KAWAGUTI, S., 1941. On the physiology of reef corals. V. Tropisms of coral planulae, con- sidered as a factor of distribution of the reefs. Palao Trap. Biol. Stat. Stud., 2 : 319-328. KAWAGUTI, S., 1944. Zooxanthellae as a factor of positive phototropism in those animals con- taining them. Palao. Trap. Biol. Stat. Stud., 2: 681-682. KEEBLE, F., 1910. Plant- Animals: A Study in Symbiosis. Cambridge University Press, 163 pp. 640 VICKI BUCHSBAUM PEARSE MUSCATINE, L., 1961a. Symbiosis in marine and fresh-water coelenterates. Pages 255-264 in H. M. Lenhoff and W. F. Loomis, Eds., The Biology of Hydra. University of Miami Press, Coral Gables, Florida. MUSCATINE, L., 1961b. Some aspects of the relationship between a sea anemone and its symbiotic algae. Doctoral dissertation, University of California, Berkeley, 100 pp. MUSCATINE, L., 1971. Experiments on green algae coexistent with zooxanthellae in sea anemones. Pac. Sri., 25 : 13-21. MUSCATINE, L., AND C. HAND, 1958. Direct evidence for transfer of materials from symbiotic algae to the tissues of a coelenterate. Proc. Nat. Acad. Sci., 44: 1259-1263. OTTAWAV, J. R., 1973. Some effects of temperature, desiccation, and light on the intertidal anemone Actinia tcncbrosa Farquhar (Cnidaria: Anthozoa). Aust. J. Mar. Fresh- water Res., 24: 103-126. PARKER, G. H., 1917. Pedal locomotion in actinians. /. Exp. Zoo!., 22: 111-124. PEARSE, V. B., 1974. Modification of sea anemone behavior by symbiotic zooxanthellae : Expansion and contraction. Biol. Bull., 147: 641-651. SIEGEL, R. W., 1960. Hereditary endosymbiosis in Paramccium bursaria. Exp. Cell Res., 19: 239-252. STANIER, R. Y., AND G. COHEN-BAZIRE, 1957. The role of light in the microbial world: some facts and speculations. Symp. Soc. Gen. Microbiol,, 7 : 56-89. TREMBLEY, A., 1744. Memoires pour scri'ir a I'histoire d'tin genre dc polypes d'cau douce, a bras en fonne dc corncs. Durand, Paris, 661 pp. TRENCH, R. K., 1971a. The physiology and biochemistry of zooxanthellae symbiotic with marine coelenterates. I. The assimilation of photosynthetic products of zooxanthellae by two marine coelenterates. Proc. Roy. Soc. London Series B, 177 : 225-235. TRENCH, R. K., 197lb. The physiology and biochemistry of zooxanthellae symbiotic with marine coelenterates. II. Liberation of fixed 14C by zooxanthellae in vitro. Proc. Roy. Soc. London Scries B,177: 237-250. TRENCH, R. K., 1971c. The physiology and biochemistry of zooxanthellae symbiotic with marine coelenterates. III. The effect of homogenates of host tissues on the excretion of photosynthetic products in vitro by zooxanthellae from two marine coelenterates. Proc. Roy. Soc. London Series B, 177 : 251-264. VANDERMEULEN, J. H., N. D. DAVIS, AND L. MUSCATINE, 1972. The effect of inhibitors of photosynthesis on zooxanthellae in corals and other marine invertebrates. Mar. Biol., 16: 185-191. WHITNEY, D. D., 1907. Artificial removal of the green bodies of Hvdra (Chlorohydra) viridis. Biol. Bull., 13: 291-299. WILSON, E. B., 1891. The heliotropism of Hydra. Amcr. Natur., 25: 413-433. ZAHL, P. A., AND J. J. A. MCLAUGHLIN, 1959. Studies in marine biology. IV. On the role of algal cells in the tissues of marine invertebrates. /. Protosool., 6 : 344-352. Reference: Biol. Bull., 147: 641-651. (December, 1974) MODIFICATION OF SEA ANEMONE BEHAVIOR BY SYMBIOTIC ZOOXANTHELLAE : EXPANSION AND CONTRACTION VICKI BUCHSBAUM PEARSE 1 Hopkins Marine Station, Pacific Grove, California 93950 No specialized photoreceptors have been identified in sea anemones, but it has been demonstrated repeatedly that at least some anemones react to visible light (e.g., Hargitt, 1907; Parker, 1917; Cotte, 1921, 1922; Batham and Pantin, 1950; North and Pantin, 1958; di Milia and Geppetti, 1964; Clark and Kimeldorf, 1970; Ottaway, 1973). And the most frequently reported reaction is contraction. Con- traction is not a specific response to light, however. It is provoked by almost any strong stimulus that might be considered unfavorable : vigorous mechanical or electrical stimulation, a variety of chemicals, exposure to air, extremes of tem- perature, osmotic shock, X-radiation, etc. (e.g., Torrey, 1904; Pieron, 1906; Fleure and Walton, 1907; Parker, 1917; Hall and Pantin, 1937; di Milia and Geppetti, 1964; Kimeldorf and Fortner, 1971 ). A large number of anemones expand in twilight or darkness and contract in daylight. Parker (1917) reviews reports of a dozen genera, some with several species that have been observed to behave in this way ; and 3 or 4 genera are reported to be indifferent to light. In contrast, 1 have found only a few reports of anemones that regularly expand by day. Bohn (1907, 1908a, 1910) affirms that specimens of Actinia eqitina from certain habitats expand in daylight and contract at night, while those from other habitats show exactly opposite behavior. Di Milia and Geppetti (1964) found that this species consistently expands in darkness and contracts in light. Anemonia sulcata, an anemone that harbors symbiotic zooxan- thellae, is reported to expand in the light and contract in the dark (Gosse, 1860; Bohn, 1906; Smith, 1939). Finally, Gee (1913) observed regular expansion in light and contraction in darkness by specimens of Anthopleura clegantissinia with zooxanthellae. There are similar reports for polyps of several gorgonians with zooxanthellae (Wainwright, 1967; Chapman and Theodor, 1969). Various authors have propounded diverse views of the factors which influence expansion and contraction in sea anemones. The difficulties encountered may be illustrated by comparing Bonn's (1908b) listing of 36 environmental factors which affect expansion and contraction with Batham and Pantin's (1950) finding that expansion and contraction continue in the apparent absence of any environmental stimuli. I have made no attempt to resolve the hypothetical effects of all these factors. However, as part of a more extensive study on the symbiosis of Antho- pleura elegantissima with zooxanthellae (Buchsbaum, 1968; Pearse, 1974), I have observed differences in the expansion of specimens with and without the endo- symbiotic algae and have tried to evaluate the possibility that this behavior is related to the symbiosis. 1 Present address : Division of Natural Sciences, University of California, Santa Cruz, California 95064. 641 642 VICKI BUCHSBAUM PEARSE METHODS Specimens of the sea anemone Anthopleura elcgantissinia (Brandt, 1835) with and without zooxanthellae were collected from rocky intertidal areas in central California, and zooxanthellae were eliminated from some specimens in the laboratory by exposure to elevated temperature, as detailed previously (Buchs- baum, 1968;Pearse, 1974). Recording of expansion and contraction was simplified by the fact that the large majority of sea anemones were usually either fully expanded or fully contracted under experimental conditions. Arbitrary decisions were necessary only occasion- ally. The anemones were not fed. Anemones were observed in the laboratory both in standing sea water and in running sea water. Light was provided by a bank of fluorescent tubes operated automatically on an alternating cycle (12 hours light: 12 hours dark). Observa- tions during the dark period were by dim white or green (540 nm Klett filter) incandescent light ; the period of exposure to these dim light sources was never more than a few seconds, and the anemones were not observed to react to them. Standing sea ivater Single animals were placed and allowed to attach in open glass jars 5 cm in diameter, containing about 100 ml sea water, in a temperature-controlled room at 12° C. The light intensity was about 450 foot-candles, as measured with a Photo- volt 200 photometer, approximating midafternoon light readings of the intertidal habitat where the anemones with zooxanthellae were collected. Records of ex- pansion and contraction of individual anemones were kept. Oxygen determina- tions by the Winkler method (Strickland and Parsons, 1965) were made using single animals in similar, sealed jars, under the same conditions of temperature and light (see Figure 1 ). Running sea water Groups of 30 animals each were placed and allowed to attach in open plastic boxes (20 X 28 cm, 10 cm deep), and the total number of animals expanded or contracted was recorded at intervals (see Table I, Figures 2 and 3). A con- tinuous flow of sea water at 14° C ± 1° was maintained through the boxes. The light intensity was approximately 250 foot-candles, chosen as intermediate in the habitat range of the anemones in this experiment. RESULTS AND OBSERVATIONS Anemones in standing sea ivater In observations over 8 days, the anemones without zooxanthellae, collected from a shaded intertidal habitat (midafternoon light intensity about 2 foot-candles), did not expand or contract in response to a daily cycle of light and darkness. At the beginning of the experiment, when supplied with fresh sea water, they stayed mostly expanded in light and dark. Left for the rest of the time in stagnant sea ANEMONE EXPANSION AND SYMBIOSIS 643 water, they contracted and remained so. In contrast, the anemones with zooxan- thellae showed a very regular pattern of expansion in light and contraction in darkness throughout the experiment. The animals usually expanded within about 10 minutes after the light was turned on. Two anemones from which all zooxan- thellae were eliminated in the laboratory were also observed; these animals be- haved as did the anemones naturally occurring without zooxanthellae, remaining contracted throughout most of the experiment. Anemones without zooxanthellae, even those that had been contracted for several days, expanded within minutes after the sea water was renewed or thoroughly stirred, or after air or oxygen was bubbled through. Bubbling nitro- gen through the sea water did not stimulate expansion of contracted anemones. The oxygen content of the sea water was measured in similar but sealed jars containing single anemones with and without zooxanthellae under the same condi- tions of light and temperature as in the behavior experiments. The oxygen data for experiments of 12 hours in light or dark (Figure 1) show that in the light, the oxygen content of the sea water increased substantially in jars containing sea anemones with zooxanthellae. Addition of light and dark values for each in- dividual anemone with zooxanthellae gives mixed positive and negative results; 6 - E 4 0) » x o 2 - 03 0 O) c 2 - -4 O O O O o o o o o Anemones with zooxanthellae Anemones without zooxanthellae FIGURE 1. Changes in oxygen content of sea water surrounding 5 anemones with zooxan- thellae and 4 anemones without zooxanthellae, in light (open circles) and darkness (solid cir- cles), after 12 hours. The oxygen content of the sea water at the start of each experiment averaged 5.66 ml O2/l of sea water and decreased a maximum of 0.15 ml O«/l in control vessels containing no anemones. Values in the figure were corrected for changes in control vessels. 644 VICKI BUCHSBAUM PEARSE TABLE I Expansion of anemones in light and dark in running sea water. Light, intensities of the habitats from which the anemones were collected were measured under clear skies in midafternoon. The rallies for the percentage of anemones expanded are means and standard deviations of data for 10 days. Light intensity of habitat (foot-candles) % Expanded Light Dark Clone from a deeply shaded habitat Group 1, without zooxanthellae 2 100 95.7 ± 6.0 Clone from a habitat spanning a range of light intensities Group 2, without zooxanthellae Group 3, with few zooxanthellae Group 4, with zooxanthellae 1 40 210 100 100 99.6 ± 1.1 91.2 ± 7.6 49.4 ± 13.0 27.5 ± 9.9 Clone from an open habitat Group 5, with zooxanthellae Group 6, without zooxanthellae (all eliminated in the laboratory) 450 450 99.4 ± 1.5 98.9 ± 2.0 23.8 ± 10.2 38.9 ±11.8 but experiments run continuously for 24 hours (12 hours light + 12 hours dark) consistently yielded negative oxygen values, suggesting microbial growth. Anemones in running sea water Expansion and contraction data from observations of anemones in running sea water are presented in Table I. Virtually all of the anemones, both with and without zooxanthellae, expanded in the light. Anemones without zooxanthellae, collected from shaded intertidal habitats (Groups 1 and 2), remained expanded in the dark also. In contrast, the large majority of anemones with abundant zooxan- thellae (Groups 4 and 5) contracted in the dark, just as they did in standing sea water experiments. The anemones from an intermediate habitat with dim light, containing few zooxanthellae (Group 3), showed intermediate behavior; in the dark, about half remained expanded and half contracted. The anemones that had previously harbored zooxanthellae, but had been caused to eliminate the algae in the laboratory (Group 6), did not remain con- tinuously expanded as did the anemones naturally occurring without zooxanthellae. Rather, like the anemones with zooxanthellae, the majority of anemones that had lost their zooxanthellae expanded in the light and contracted in the dark. The time course of expansion and contraction is illustrated in Figures 2 and 3. A majority of the anemones responded within 10 minutes after the lights were turned on or off, and by 20-30 minutes, all had usually responded. However, during the 12 hours of continuous light or darkness that followed, some variability in behavior often appeared, especially in the dark. This variability is evident in the observations summarized in Table I, which were always made toward the ANEMONE EXPANSION AND SYMBIOSIS 045 middle or end of the light and dark periods. This procedure was chosen in order to record a sustained level of response rather than the more dramatic Q% or 100% expansion that was usually observed immediately after a change in light phase. If the lights were turned on or off in the middle of a 12-hour dark or light period, both the time course and percentage of anemones responding were the same as when the lights went on or off after the usual 12-hour interval. Field observations Field observations were limited to anemones with zooxanthellae found in tide- pools at low water, since animals exposed to air by the receding tide almost in- variably contract, and no anemones without zooxanthellae were found in suitable pools. In pools viewed at dawn, all of the anemones were contracted. As the daylight gradually increased, they suddenly began to expand, until all were fully expanded after 20-30 minutes. The time course of dawn expansion in the field under a gradually increasing light intensity was thus the same as the time course in the laboratory, where the lights were turned on suddenly (Figure 2). On slightly overcast days, the anemones remained fully expanded. However, on clear days, the animals contracted in midday sunlight. At such times, the distribution of expanded and contracted anemones in the pools corresponded exactly to the pat- tern of shadow and sunlight, respectively. No temperature differences in the sea TOO "g so •o c 60 I 40 20 DARK LIGHT « 10 15 Minutes 20 25 30 FIGURE 2. Time course of expansion by sea anemones in running sea water. The data pre- sented are from a group of anemones with zooxanthellae, but the time course was the same for anemones that had lost their zooxanthellae. 646 VICKI BUCHSBAUM PEARSE 10 20 Minutes 40 FIGURE 3. Time course of contraction by sea anemones in running sea water. The data presented are from a group of anemones with zooxanthellae, but the time course was the same for anemones that had lost their zooxanthellae. water could be found to account for the distribution; however, internal tempera- tures of the anemones were not determined. Observations on anemones in outdoor tables and tanks with running sea water supplemented intertidal observations. Anemones with zooxanthellae ex- panded in diffuse daylight but contracted in direct sunlight ; they contracted on moonless or cloudy nights, but expanded fully in bright moonlight. Anemones from which zooxanthellae had been eliminated followed the same pattern. The behavior of anemones without zooxanthellae originating from dark habitats was erratic. DISCUSSION Observations on the sea anemone Anthopleura elegantissinia, in both standing and running sea water, have shown differences in the behavior of individuals with and without zooxanthellae. Anemones without zooxanthellae, originating from dark habitats, do not regularly expand or contract with changes in light. In con- trast, anemones with zooxanthellae expand in moderate light and contract in intense light or darkness, with striking uniformity. I have found also that in- dividuals without zooxanthellae do not display phototaxis, while those with zooxan- thellae show poMtive or negative phototaxis depending on the light intensity ANEMONE EXPANSION AND SYMBIOSIS 647 (Pearse, 1974). The pattern of expansion and contraction in this sea anemone is thus similar to that of its phototactic behavior in that both are modified by symbiosis with zooxanthellae. In both kinds of behavior, anemones without zooxanthellae appear basically indifferent to light while anemones with zooxan- thellae engage in active, light-related responses. Behavior in both standing and running sea water suggests that the anemones without zooxanthellae generally expand in well-oxygenated sea water and con- tract when oxygen levels fall. A direct response to oxygen is further implicated by the rapid expansion of contracted individuals in stagnant sea water after re- newal or stirring of the sea water or gassing with air or oxygen, and by the lack of expansion after gassing with nitrogen, which provided similar mechanical stimulation. Anemones with zooxanthellae may also be responding to changes in oxygen concentration, rather than directly to light. The oxygen data in Figure 1 indicate that in the light photosynthesis by zooxanthellae met the respiratory requirements of both algae and anemone and raised the oxygen content of the sea water, as has been found in other cnidarians with zooxanthellae (e.g., Yonge, Yonge, and Nicholls, 1932; Smith, 1939; Kanwisher and Wainwright. 1967; Roffman, 1968; Chapman and Theodor, 1969). In the dark, with the zooxanthellae no longer producing oxygen, its concentration fell. However, most of the anemones were contracted after only 10 minutes of darkness (see Figure 3), and it seems un- likely that oxygen content was much reduced in that short time. In running sea water, where anemones without zooxanthellae hardly ever contracted, it is especi- ally difficult to explain contraction of anemones with zooxanthellae by a hypothesis of oxygen deprivation, unless they require much more oxygen. However, they may respond simply to a sudden reduction of available oxygen, regardless of absolute concentration. In the dark, the lower oxygen values recorded for anem- ones with zooxanthellae (Figure 1 ) probably reflect only the larger size of these particular individuals, compared with those lacking zooxanthellae ; the respiratory burden added by the zooxanthellae was not determined. Use of specific photo- synthetic inhibitors (see, e.g., Vandermeulen, Davis, and Muscatine, 1972) or determination of the action spectrum of the response (see, e.g., Clark and Kimel- dorf, 1970) may help to establish whether expansion and contraction are direct responses to light or to some consequence of photosynthesis. In standing sea water, accumulation of metabolic by-products excreted by the anemones might also lead to contraction. There is evidence that other cnidarians excrete less phosphorus (Yonge and Nicholls, 1931a, 1931b; Smith, 1939; Yama- zato, 1966) and less nitrogen (Kawaguti, 1953; Muscatine, in press) when zooxanthellae are present, and that at least some respiratory carbon dioxide is fixed by zooxanthellae (Pearse, 1970). In the light, individuals of A. clcgantis- siwa with zooxanthellae excrete into the sea water only about half as much am- monia as do those lacking zooxanthellae, nitrogen being recycled between host and algal cells (L. Muscatine and C. D'Elia, University of California, Los Angeles, personal communication), and it seems likely that excretion of phosphorus com- pounds and carbon dioxide is also reduced. Accumulation of excretory products might also influence contraction indirectly through enhancement of microbial 648 VICKI BUCHSBAUM PEARSE growth in the jars of standing sea water. Comparison of oxygen data from 12- and 24-hour experiments suggests that microbial respiration was considerable; and I observed that the sea water in the jars containing anemones without zooxanthellae became cloudy after several days, while that in the jars containing anemones with zooxanthellae remained clear. However, if a direct response to change in available oxygen or any other consequence of photosynthesis by zooxanthellae were the only factor regulating expansion and contraction, the anemones from which zooxanthellae had been eliminated would be expected to behave exactly like anemones naturally occurring without zooxanthellae. This was indeed observed in standing sea water, where oxygen depletion or accumulation of excretory products could rapidly have be- come a determining factor. But in running sea water, like anemones with zooxan- thellae, individuals that had lost their zooxanthellae expanded in light and con- tracted in darkness. This suggests the possibility that the anemones' behavior was conditioned by their previous symbiosis. Also suggestive of conditioning were the apparent changes in the behavior of anemones naturally occurring without zooxanthellae, which took place after prolonged exposure, not to endosymbiotic algae, but to seaweeds in intertidal pools and diatoms in laboratory vessels (see Buchsbaum, 1968 for details). I did not carry out any experiments specifically designed to test for evidence of conditioned behavior in these animals. However, in view of the number of recent suggestions that cnidarians are capable of simple forms of learning (see review by Rushforth, 1973), the expansion and contrac- tion of these sea anemones seems a promising experimental system to investigate further. How contracting in the dark might be of special selective advantage to a sea anemone with zooxanthellae is obscure. Decreased oxygen consumption has been reported in contracted sea anemones (Pieron, 1908; Shoup, 1932; Smith, 1939), including Anthopleura elcgantissima (see Buchsbaum, 1968 for details). An intertidal anemone such as A. elegantissiaia, which often produces dense populations in isolated tidepools, may be subjected to low oxygen conditions during low tides, especially at night. Since oxygen consumption by at least some anemones (Sassa- man and Mangum, 1972), including A. elcgantissima (]. ]. Childress, University of California, Santa Barbara, personal communication), is proportional to the oxygen concentration of the surrounding sea water, it might be to the anemone's advantage to reduce its rate of oxygen consumption by contracting when the oxygen concentration first begins to drop. However, these anemones appear to survive low oxygen conditions for long periods, and contraction may simply con- serve energy. Possible selective advantages of the anemones' behavior in light are easier to defend. Under moderate light intensities, expansion exposes the abundant zooxan- thellae in the tentacles and oral disk to maximum illumination and thus pre- sumably favors maximum photosynthesis. Oxygen data from gorgonians suggest that less light reaches the zooxanthellae in retracted polyps than in extended ones, due to shading by ectoderm and spicules, especially in strongly pigmented species (Kanwisher and Wainwright, 1967 ; Chapman and Theodor, 1969). In intense light, however, maximum exposure of zooxanthellae may instead reduce photo- ANEMONE EXPANSION AND SYMBIOSIS 649 synthesis. Roffman (1968) suggests that in scleractinian corals with zooxan- thellae, midday decreases in photosynthesis are related to extended exposure to high light intensity. Thus, contraction by A. elcgantissima in bright sunlight may serve a significant function in shielding the zooxanthellae in the endoderm beneath the heavily green-pigmented ectoderm of the column of the animal. Ectodermal pigments are produced by these anemones under the influence of bright light, and poorly pigmented individuals from partially shaded habitats suffer substantial reduction in number of zooxanthellae and total chlorophyll when suddenly ex- posed to increased light intensities (Buchsbaum, 1968). The sea anemone Acti- niogeton scscrc, which occurs in shallow waters of Hawaiian reef flats and harbors zooxanthellae, produces extremely heavy concentrations of a similar green pigment in the tentacles and oral disk; these anemones remain expanded in full sunlight (Buchsbaum, 1968). Development of pigmentation (Buchsbaum, 1968) and phototaxis (Pearse, 1974) may represent two mechanisms by which the sea anemone Anthopleura elegantisshna favorably regulates the quantity of light to which its symbiotic zoo- xanthellae are exposed. These are both relatively slow responses and probably serve principally to permit the animal to select and adapt to habitats in a wide range of light intensities. Expansion and contraction provide a more rapid and flexible means for regulating the light reaching the zooxanthellae as light levels rise and fall from minute to minute throughout the day. This study was supported by a predoctoral fellowship from the National Institutes of Health, carried out at the Hopkins Marine Station, Pacific Grove, California, and submitted to Stanford University as part of a doctoral dissertation. I am grateful to the faculty, staff, and students of the Marine Station, especially my advisory committee, Drs. J. H. Phillips, D. P. Abbott, and W. L. Lee ; to Dr. T. S. Pearse for reading the manuscript; and to the Division of Natural Sciences, University of California, Santa Cruz for library and secretarial assistance. SUMMARY The pattern of expansion and contraction by the sea anemone Anthoplcura elegantisshna differs in individuals with or without endosymbiotic zooxanthellae. Anemones without zooxanthellae, found in dark habitats, do not regularly expand or contract under changes in light. Anemones with zooxanthellae expand in moderate light and contract in intense light or in darkness, with striking uniformity. However, this behavior does not always depend directly on the presence of zooxanthellae. Anemones that have previously had endosymbiotic zooxanthellae subsequently expand and contract with changes in light in the absence of these algae. Thus, conditioned responses may be involved. It is suggested that expansion and contraction of the anemones may play an important role in favorably regulating the amount of light to which their zooxan- thellae are exposed. 650 VICKI BUCHSBAUM PEARSE LITERATURE CITED BATHAM, E., AND C. F. A. PANTIN, 1950. Phases of activity in the sea-anemone Mctridium senile (L.), and their relation to external stimuli. /. Exp. Biol., 27: 377-399. BOHN, G., 1906. Mouvements en relation avec 1'assimilation pigmentaire chez les animaux. C. R. Soc. Biol. Paris. 61 : 527-528. BOHN, G., 1907. Le rythme nycthemeral chez les Actinies. C. R. Soc. Biol. Paris, 62 : 473- 476. BOHN, G., 1908a. De 1'influence de 1'oxygene dissous sur les reactions des Actinies. C. R. Soc. Biol. Paris, 64 : 1087-1089. BOHN, G., 1908b. Les facteurs de la retraction et de 1'epanouissement des Actinies. C. R. Soc. Biol. Paris, 64: 1163-1166. BOHN, G., 1910. Comparison entre les reactions des Actinies de la Mediterranee et celles de la Manche. C. R. Soc. Biol, Paris. 68 : 253-255. BUCHSBAUM, V. M., 1968. Behavioral and physiological responses to light by the sea anemone Anthoplcura clcgantissima as related to its algal endosymbionts. Doctoral dissertation, Stanford University, 123 pp. CHAPMAN, G., AND J. THEODOR, 1969. L'influence de la lumiere sur la consommation d'O2 chez Eunicclla stricta (Gorgone a Zooxanthelles symbiotiques) et chez Paramuricca clavata. Vic ct Milieu, 20 : 483-490. CLARK, E. D., AND D. J. KIMELDORF, 1970. Tentacle responses of the sea anemone Antho- plcura xanthogrammica to ultraviolet and visible radiations. Nature, 227 : 856-857. COTTE, J., 1921. Sur le phototropisme des Actinies. C. R. Soc. Biol. Paris, 85 : 188-190. COTTE, J., 1922. fitudes sur les cornportements et les reactions des Actinies. Bull. Inst. Oceanogr., Monaco, 410: 1-44. FLEURE, H. J., AND C. L. WALTON, 1907. Notes on the habits of some sea anemones. Zoo/. Anz., 31: 212-220. GEE, W., 1913. Modifiability in the behavior of the California shore-anemone Cribrina xantho- grammica Brandt. /. Anim. Bchav., 3 : 305-328. GOSSE, P. H., 1860. A History of the British Sea-Anemones and Corals. Van Voorst, London, 362 pp. HALL, D. M., AND C. F. A. PANTIN, 1937. The nerve net of the Actinozoa. V. Temperature and facilitation in Mctridium senile. J. Exp. Biol., 14: 71-78. HARGITT, C. W., 1907. Behavior of sea-anemones. Biol. Bull., 12 : 274-284. KANWISHER, J. W., AND S. A. WAINWRIGHT, 1967. Oxygen balance in some reef corals. Biol. Bull.. 133 : 378-390. KAWAGUTI, S., 1953. Ammonium metabolism of the reef corals. Biol. J. Okayama Unir., 1 : 171-176. KIMELDORF, D. J., AND R. W. FORTNER, 1971. The prompt detection of ionizing radiations by a marine coelenterate. Radiat. Res., 46 : 52-63. MILIA, A. DI, AND L. GEPPETTI, 1964. On the expansion contraction rhythm of the sea anemone, Actinia cquina L. Expcricntia, 20 : 571-572. MUSCATINE, L., 1974. Nitrogen metabolism of reef corals. Proc. Second Int. Syinp. Coral Reefs, in press. NORTH, W. J., AND C. F. A. PANTIN, 1958. Sensitivity to light in the sea-anemone Mctridium senile (L) : adaptation and action spectra. Proc. Roy. Soc. London Scries B, 148: 385-396. OTTAWAY, J. R., 1973. Some effects of temperature, desiccation, and light on the intertidal anemone Actinia tcncbrosa Farquhar ( Cnidaria : Anthozoa). Aust. J. Mar. Freshwater Res., 24 : 103-126. PARKER, G. H., 1917. Actinian behavior. /. Exp. Zoo/., 22 : 193-230. PEARSE, V. B., 1970. Incorporation of metabolic CO2 into coral skeleton. Nature, 228: 383. PEARSE, V. B., 1974. Modification of sea anemone behavior by symbiotic zooxanthellae : Phototaxis. Biol. Bull., 147 : 630-640. PIERON, H., 1906. La reaction aux marees par anticipation reflexe. C. R. Soc. Biol. Paris, 61 : 658-660. PIERON, H., 1908. De 1'influence de 1'oxygene dissous sur le comportement des invertebres ANEMONE EXPANSION AND SYMBIOSIS 651 marins. III. Des rythmes engendres par une variation periodique de la teneur en oxygene. C. R. Soc. Biol. Paris, 64 : 1020-1022. ROFFMAN, B., 1968. Patterns of oxygen exchange in some Pacific corals. Comp. Biochcm. Physiol, 27 : 405^18. RUSHFORTH, N. B., 1973. Behavioral modifications in coelenterates. Pages 123-169 in W. C. Corning, J. A. Dyal, and A. O. D. Willows, Eds., Invertebrate Learning, Vol. 1. Protozoans Through Annelids. Plenum Press, New York. SASSAMAN, C., AND C. P. MANGUM, 1972. Adaptations to environmental oxygen levels in infaunal and epifaunal sea anemones. Biol. Bull., 143 : 657-678. SHOUP, C. S., 1932. Salinity of the medium and its effect on respiration in the sea-anemone. Ecology, 13: 81-85. SMITH, H. G., 1939. Relationship between Actinians and Zooxanthellae. /. Exp. Biol., 16: 334-345. STRICKLAND, J. D. H., AND T. R. PARSONS, 1965. A Manual of Sea Water Analysis. [2nd ed.] Fisheries Research Board of Canada, Bull. Number 125, Ottawa, Canada. TORREY, H. B., 1904. On the habits and reactions of Sagartia davisi. Biol. Bull., 6: 203-216. VANDERMEULEN, J. H., N. D. DAVIS, AND L. MUSCATINE, 1972. The effect of inhibitors of photosynthesis on zooxanthellae in corals and other marine invertebrates. Mar. Biol., 16: 185-191. WAINWRIGHT, S. A., 1967. Diurnal activity of hermatypic gorgonians. Nature. 216: 1041. YAMAZATO, K., 1966. Calcification in a solitary coral, Fungia scutaria Lamarck in relation to environmental factors. Doctoral dissertation, University of Haivaii, 130 pp. YONGE, C. M., AND A. G. NICHOLLS, 1931a. Studies on the physiology of corals. IV. The structure, distribution and physiology of the zooxanthellae. Gt. Barrier Reef Exped. Sci. Rpts., 1 : 135-176. YONGE, C. M., AND A. G. NICHOLLS, 1931b. Studies on the physiology of corals. V. The effect of starvation in light and in darkness on the relationship between corals and zooxanthellae. Gt. Barrier Reef Expcd. Sci. Rpts., 1: 177-211. YONGE, C. M., M. J. YONGE, AND A. G. NICHOLLS, 1932. Studies on the physiology of corals. VI. The relationship between respiration in corals and the production of oxygen by their zooxanthellae. Gt. Barrier Reef Exped. Sci. Rpts., 1 : 213-251. Reference: Bid. Bull., 147: 652-660. (December, 1974) SOME CONSEQUENCES OF SEXUAL DIMORPHISM: FEEDING IN MALE AND FEMALE FIDDLER CRABS, UCA PUGNAX (SMITH) IVAN VALIELA, DANIEL F. BABIEC, WILLIAM ATHERTON,1 SYBIL SEITZINGER2 AND CHARLES KREBS 3 Boston University Maritie Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 The males of Uca pugnax (Smith), a fiddler crab species found along the coast of the northwest Atlantic Ocean, have one large chela used for mating dis- plays, sound production, and aggression (Altevogt, 1955; Crane, 1941, 1943, 1967; Pearse, 1912; Salmon, 1967), and one small feeding chela. The female fiddler crab has two small feeding chelae. The crabs feed by picking up bits of mud and detritus from their habitat using the feeding chelae (Miller, 1961). Kingsley (1888), Hyman (1920), Swartz and Safir (1915), Pearse (1912), Altevogt (1955) and Ono (1968) observed that females show more feeding motions than males, presumably because both feeding claws can operate alternately with almost continuous delivery of food particles to the mouth. If metabolic demands of both sexes are similar, the sexual dimorphism in the feeding claws must have forced the males to evolve some compensatory mechanisms such that food assimila- tion equals that of females. Male fiddler crabs could increase food ingestion by having faster feeding motions, by feeding for a longer period of time than females, by feeding on larger mouthfuls or by having a higher assimilation efficiency of the food they do ingest. This paper first attempts to determine if there are any differences in metabolic rates, food intake, fecal production or digestive efficiency between male and female U. [>n(/nu.v and then ascertains whether differences in duration of feeding or morphological dimorphisms involved in feeding exist between the two sexes. METHODS All crabs used in this study were collected at Great Sippewissett marsh in Falmouth, Massachusetts, in early fall, were brought to the laboratory and kept in seawater tables at room temperature. Activity cycles in fiddlers may last as long as one week away from water (Bennett, Shriner and Brown, 1967). All experi- ments were conducted at times corresponding to low tide, the time of maximum activity in the field. Respiratory rates were measured with an oxygen electrode and chart recorder connected to a respiration chamber consisting of a 250 ml plastic cylinder stoppered at both ends. Demeusy (1957), Teal (1959). Vernberg (1959), Vernberg and 1 Present address: Department of Biology, Dalhousie University, Halifax, N.S., Canada. 2 Present address : Graduate School of Oceanography, University of Rhode Island, Narra- gansett, Rhode Island. 3 Present addrcs: Science Division, St. Mary's College of Maryland, St. Mary's City, Maryland. 652 FEEDING IX UCA FUGNAX 653 Vernberg (1966) and Teal and Carey (1967) have previously measured respira- tion in Uca. Crabs in the present study were grouped by sex and size. Two crabs of the same sex and differing by no more than 0.25 mm in carapace width were placed in the seawater-nlled chamber. A perforated stopper was pressed down into the chamber until all bubbles were excluded. The electrode was then inserted through the perforation. The water was stirred to avoid saturation of the elec- trode (Kanwisher, 1962). Temperature was measured before and after the 15 min. period during which the crabs were held in the chamber for respiration mea- surements. Dry weight of males and females were obtained. To ensure reasonable com- parisons the weight of the sclerotized parts of one of the chelae was subtracted from total weight. This eliminated the bias due to the enlarged male claw. The sclerotized portion of the chelae was measured by carefully removing muscle tissues and weighing the dried hard parts. Size-matched crabs were starved two days and placed in individual fingerbowls to measure the amount of sediment ingested by male and female crabs of similar size. Three grams of marsh mud resuspended in water were placed with the crabs in 1 cm of seawater. The crabs were left undisturbed for six hours after which they were removed, along with any fecal pellets deposited. The remaining mud was redried and weighed. The number of fecal pellets produced by crabs per unit time and the amount of organic matter in the pellets were measured to provide some idea of the amount of food processed and the efficiency with which organic matter contained in ingested mud was assimilated by males and females. About 20 g of wet marsh mud were placed in a fingerbowl containing— seawater and offered to 3 male and female crabs of three size classes. The crabs were allowed to feed for five days under a regime of 12 hrs of dark and 12 hrs light. Fecal pellets were collected every three hours. The volumes of the fecal pellets as well as their dry weight were obtained. Whole fecal pellets were then combusted in a muffle furnace at 500° C for 24 hrs to allow the calculation of percent organic matter. The organic content of the mud on which the crabs fed was also measured. Further measurements of the number of pellets were obtained by cutting 4 cm deep, 20 cm diameter cores of marsh turf, placing them in fingerbowls, and allow- ing crabs to feed on the surface sediments. Field measurements of pellet produc- tion were carried out by first locating single-entrance burrows by pumping sea- water through the entrance and observing if water flowed out of any neighboring burrows. Once single entrance burrows were found, 6 inch aluminum flashing was used to construct a circular enclosure 20 cm in diameter. The marsh vegeta- tion was removed from the enclosure and all fecal pellets were removed from the sediment prior to starting. The number of pellets was counted at feasible intervals over a 24 hr period. The number of feeding motions of male and female crabs was recorded by observing individuals feeding on marsh sediment on a shallow seawater table for two minutes. Wooden barriers with narrow viewing slots were erected around the tables to prevent disturbing the crabs. Feeding motions consisted of a single chela moving from the sediment to the buccal cavity. 654 VALIELA ET AL. TABLE I Regression lines fitted to values of partial pressure of oxygen (mm Hg) in the respiration chambers during a 15 min period for male and female crabs of various size classes. Initial temperature was 20° C, final temperature 21° C. Mean weights were calculated as dry weight minus 88.5% of the larger claw. Carapace width cla 0.05, Table I). The very slight differences in weight between males and females were not significant. The amount of mud ingested by starved male and female crabs did not differ significantly when tested with a paired t test (Table II). The number of fecal pellets produced by male and female crabs (Table III) was variable but a two-way analysis of variance of the data from the laboratory TABLE II Amount of sediment ingested by starved U. pugnax during a six-hour feeding period in the laboratory. Carapace width g of sediment ingested (mm) Female crabs Male crabs 18.1 0.47 0.35 12.6 0.50 0.47 22.7 0.19 0.37 x ± s.e. : 0.39 ± 0.10 0.40 ± 0.04 FEEDING IN UCA PUGNAX 655 TABLE III Number of fecal pellets produced by individual crabs per 24 hrs, feeding on loose mud obtained from the surface of Great Sippewissett marsh, in the surface of a core of marsh sediments taken into the laboratory and within field enclosures. The values in parenthesis are the number of individual crabs involved. Size of crabs expressed as carapace width (cm) No. Mean of fecal pellets/24 hrs ± st. error Female crabs Male crabs Crabs feeding on loose detritus in lab: 0.8-0.9 1.2-1.3 1.6-1.7 172 ±58 53 ± 6 47 ± 13 (2) (2) (2) 119 ± 11 (2) 123 ±31 (2) 98 ± 33 (2) Crabs feeding on surface of marsh core in the lab : 1.2 1.3 1.4 129.5 ± 2.5 (1) 155.5 ± 23.5 (1) 174.5 ± 4.5 (1) 117.6 ± 3.5 (1) Crabs feeding on marsh surface within field enclosures : 0.6 1.2-1.3 1.6 1.9-2.1 176 143.5 ± 37.5 142 (1) (2) (1) 157.2 ± 15.2 (4) 177.5 ± 3.5 (2) 110.8 ± 11.0 (4) crabs fed on loose detritus showed that neither sex nor size of crabs significantly affected the number of fecal pellets obtained within 24 hrs. No differences are seen in the values obtained for the field measurments or the laboratory crabs feeding on the surface of a marsh core (Table III ) . The volume and weight of fecal pellets (Table IV) were affected significantly by size of crabs (F ~ 29.9 and 4.15 respectively, both with 2 and 6 d.f.) but not by sex (F = 0.8 and 4.5 respectively, with 1 and 6 d.f.). The amount of organic matter (Table IV) in the pellets is not significantly related to either size or sex of crabs. There was no significant interaction between sex and size of fiddler crabs in any of these experiments. These results suggest that male and female fiddlers did not have different metabolic demands and that the amount of bulk food processed by each sex is TABLE IV Volumes, weights and percent organic matter of fecal pellets for fiddler crabs. Six to 15 pellets were used from each of two crabs in each treatment combination. Tabled values are mean ± standard errors. Size of crabs expressed as carapace width Female crabs Male crabs Vol/pellet (10-5 cm3) Wt/pellet (10-5 g) % organic matter Vol/pellet (10-s cm3) Wt/pellet (10-sg) % organic matter 0.8-0.9 mm 1.2-1.3 mm 1.6-1.7 mm 56 ± 15 172 ± 22 364.5 ± 40.5 33.1 ± 3.1 77.3 ± 23.3 112.8 ± 1.2 31 ±0.2 34.5 ± 0.9 35.1 ± 0.7 83 ± 15 222.5 ± 30.5 349.5 ± 33.5 46.3 ± 2.8 78.9 ± 7.4 150.8 ± 1.8 34.6 ±1.1 29.2 ± 4.1 32.8 ± 0.1 656 VALIELA ET AL. TABLE V Xitniber of feeding motions per 2 min observation period. Tabled values are mean plus or minus standard errors. Carapace width No. of individuals observed No. feeding motions per 2 min Females Males 13-14 mm 14-15 mm 16 males, 16 females 15 males, 15 females 64.6 ± 2.0 68.9 ± 2.2 33.0 ± 1.1 38.5 ± 1.2 similar. Further, neither sex was more efficient at removing organic matter out of the ingested sediment. The organic content of the sediment on which the crabs fed was 23.7%, considerably lower than that of the pellets. This probably indi- cates selective feeding on particles of high organic content. There is a possibility that male crabs are more selective than females, choosing only the richest particles, but we have no evidence for this. Feeding belwior Table V shows that female U. pugnax carry out almost twice as many feed- ing motions as males during comparable spans of time. This was due to almost simultaneous feeding with two feeding appendages in the female. Males did not therefore use faster feeding motions to compensate for using only one feeding- claw. Although there was considerable variation among individual crab, males did feed roughly twice as long as females, since male crabs scored about twice the feeding units awarded to females (Table VI). Feeding niorpJioIogv The length of the feeding chela of males and females do not differ for specimens of comparable carapace width (Fig. 1). However, the width of the small flat ven- tral area on the tip of the dactyl of feeding claws is larger in males than in females (Fig. 2). The slopes of the regression lines are similar but the values for male U. pitgna.v lie above those for females for any comparable carapace width. This TABLE VI .\ umber of feeding units recorded for a period of 75 minutes and three different sea-water tables. Six male and female crabs were used in each table. See text for scoring of feeding units. Table no. Female crabs Male crabs Ratio male/female feeding units Range of carapace widths (mm) No. of feeding units Range of carapace widths (mm) No. of feeding units 1 2 3 15.1-17.4 14.2-18.3 14.0-17.1 6.8 ± 2.7 7.7 ± 2.3 3.7 ± 0.6 15.7-20.9 14.8-17.2 14.7-19.5 13.0 ± 2.7 16.8 ± 4.6 7.8 ± 1.1 1.9 2.2 2.1 FEEDING IN UCA PUGNAX 657 8 x i- LL! • Males ° Females 17 19 21 LU O Q UJ LU LJL 8 CARAPACE WIDTH (MM) FIGURE 1. Length of feeding claws versus carapace \vidtli in male and female Uca piigtnix. The lines show the calculated regression line for the points. might make it possible for males to obtain larger moutbfuls per feeding motion than females. DISCUSSION Our results suggest that male and female fiddler crabs have similar nutritional demands, food-processing abilities, and digestive efficiencies. Since males have only one functional feeding claw, the number of feeding motions is half that ob- served in female U. pugna.v. Males seem to compensate by extending the duration of feeding behavior. In the field feeding occurs mainly at low tide during daylight 658 VALIELA ET AL. (Ono, 1968, and our own observations). We have observed males feeding under- water in Massachusetts and Teal (1958) saw crabs feeding during high tides in Georgia. In the marshes of Buzzards Bay along the coast of Massachusetts, mud flats and grass-covered habitats in which Uca feed are exposed for about four hours at low tide. We have observed that as high tide approaches only males are still feeding. This agrees well with our results of prolonged feeding with males. The two-fold increase in feeding duration would seem to be enough compensa- tion for lack of a feeding appendage. However, the morphology of the feeding claw is such that males may be able to grasp and feed on larger mouthfuls per feeding motion. Longer feeding periods may expose males to greater predatory mortality. However, in a sample of 713 specimens of U. pugna.v from Great Sippewisset Marsh there were more males than females present (42.9% females). This sex ratio was very similar to that found by Shanholzer (1973) in a salt marsh in Georgia and is not unusual among marine invertebrates (Wenner, 1972) . Data now in preparation from Great Sippewissett show greater mortality of the smaller females than males. The need to repeatedly expose eggs to the flow of well- oxygenated water may increase the exposure of berried females to predatory mortality by forcing females to remain in the relatively unprotected creek banks at high tide. The egg masses also may impede locomotion of females and increase I h- Q 0.9 0.8 0.7 ^ 0.6 \- o Q 0.5 Males y = 0.03 + 0.44X,r -0.89 Females y = -0.02 + 0.40X, r = 0.93 12 13 14 15 16 17 18 CARAPACE WIDTH (MM) 19 20 FIGURE 2. Width of the tip of the feeding dactyl versus carapace width in male and female Uca pugnax. The lines show the calculated regression line for the points. FEEDING IN UCA PUGNAX 659 their vulnerability to predators (Hyman, 1920). It may be, however, that dif- ferential mortality between the sexes does not affect the sex ratios (Leigh, 1970) so that sexual dimorphism does not directly lead to different survivorship rates. There are no doubt seasonal fluctuations on the metabolic expenditures of male and female crabs related to sexual dimorphism. Males show great activity during breeding displays (Crane, 1958, 1966; Salmon, 1965), while berried females not only produce eggs from late May to early July but must, as already mentioned, care for the egg masses. To fully develop the consequences of sexual dimorphism these changes over time must be considered. This paper merely illustrates that at any one time there are mechanisms operating by which males compensate for the lack of one feeding appendage, and the results also evidence the complex con- sequences of apparently simple morphological and behavioral differences. SUMMARY 1. There were no differences in the respiratory rates of male and female Uca pitgna.v of comparable sizes. 2. The amount of salt marsh sediment ingested by starved male and female crabs was similar. 3. The number and weight of fecal pellets produced by male and female crabs were similar, as was the organic matter content. 4. The above suggests that there are minimal differences in food demands and digestive efficiencies between the sexes, yet the enlarged claw of the fiddler crabs cannot be used for feeding. This requires some compensatory mechanism in male crabs. 5. Male fiddlers do show about half the feeding motions per unit time compared to females, but they compensate by feeding about twice as long. This is corro- borated by field observations. 6. Further compensation, if needed, could be achieved by the slightly larger holding surface of the feeding claw in males, perhaps allowing the grasping of larger fragments of marsh sediment. LITERATURE CITED ALTEVOGT, R., 1955. Some studies on two species of Indian fiddler crabs, Uca marionis nitidns and Uca anallipes. J. Bombay Natur. Hist. Soc., 52 : 702-716. BENNETT, M. F., J. SHRINER, AND R. A. BROWX, 1967. Persistent tidal cycles of spontaneous motor activity in the fiddler crab, Uca pugna.v. Biol. Bull.. 112: 267-275. CRANE, I., 1941. Crabs of the genus Uca from the west coast of Central America. Zooloqica, 26 : 145-208. CRANE, J., 1943. Display, breeding and relationships of fiddler crabs (genus Uca~) in the field. Zoologica, 28: 217-223. CRANE, J., 1966. Combat, display, and ritualization in fiddler crabs (' Oxypodidae) , genus Uca. Phil. Trans. Roy. Soc. London Scries B.. 251 : 459-472. CRANE, J., 1967. Combat and its ritualization in fiddler crabs (Oxypodidae) with special reference to Uca rapax. Zoologica, 52 : 49-77. DEMEUSY, N., 1957. Respiratory metabolism of the fiddler crabs Uca pugilator from two different latitudinal populations. Biol. Bull. 113 : 245-253. HYMAN, O., 1920. Adventures in the life of a fiddler crab. Ann. Rep. Smith. Inst., 1920: 433-460. 660 VALIELA ET AL. KANWISHER, J., 1962. Oxygen and carbon dioxide instrumentation. Mar. Sci. Inst., 1 : 334- 339. KINGSLEV, S., 1888. Something about crabs. Amcr. Natiir.. 22 : 888-896. LEIGH, E. G., JR., 1970. Sex ratio and differential mortality between the sexes. Amcr. Natur., 104: 205-210. MILLER, D., 1961. The feeding mechanism of fiddler crabs, with ecological considerations of feeding adaptations. Zoologica, 46 : 89-100. ONO, Y., 1968. On the ecological distribution of ocypoid crabs in the estuary. Mem. Fac. Sci. Kyushu Univ. Ser. E. Biol., 4: 1-60. PEARSE, A. S.," 1912. The habits of fiddler crabs. Philliff- J- Sci. 21)., 7: 113-133. SALMON, M. 1967. Coastal distribution display and sound production by Florida fiddler crabs (genus Uca). Anim. Bchav.. 15: 449-459. SALMON, M., 1965. Waving display and sound production in the courtship behavior of Uca putjUator with comparisons to U. utina.r and U. pugnax. Zoologica. 50: 123-148. SHANHOLTZER, S. F., 1973. Energy flow, food habits and population dynamics of Uca pugnax in a salt marsh system. Ph.D. thesis, L'uii'crsity of Georgia. 91 pp. SWARTZ, B. AND S. SAFiR, 1915. Natural history and behaviour of the fiddler crab. Cold Spring Harbor Monogr., 8 : 3-23. TEAL, J. M., 1958. Distribution of fiddler crabs in Georgia salt marshes. Ecoloqv, 39: 185- 193. TEAL, J. M., 1959. Respiration of crabs in Georgia salt marshes in its relation to their ecology. Physiol. Zoo!.. 32 : 1-14. TEAL, J. M., AND F. G. CAREY, 1967. The metabolism of marsh under conditions of reduced oxygen pressure. Physiol. Zoo/., 40: 83-91. VERNBERG, F., 1959. Studies on the physiological variation between tropical and temperate zone fiddler crabs of the genus Uca. II. Oxygen consumption of the whole organism. Biol. Bull., 117: 163-184. VERNBERG, F. AND W. VERNBERG, 1966. Studies on the physiological variation between tropical and temperate zone fiddler crabs of the genus Uca. VII. Metabolic-temperature re- sponses in southern hemisphere crabs. Cow/', Biochcin. Physiol., 19: 487-524. WENNER, A. M., 1972. Sex ratio as a function of size in marine Crustacea. Amcr. Natur., 106: 321-350. Reference: Biol. Bull, 147: 661-677. (December, 1974) STUDIES ON THE REPRODUCTIVE SYSTEMS OF SEA-STARS. I. THE MORPHOLOGY AND HISTOLOGY OF THE GONAD OF ASTERIAS VULGARIS 1 CHARLES WAYNE WALKER Division of Biological Sciences, Cornell University, Ithaca, Neiv York 14850 and Shoals Marine Laboratory, Isles of Shoals, Maine Although the gonads of Asterias vulgaris are part of one of the major organ systems in this sea-star they have not been the subject of a basic morphological or histological study other than the preliminary account published by Field in 1892. The structure of the reproductive system has been described with varying degrees of completeness in a wide variety of other sea-stars (Asterias rub ens, Hoffmann, 1872; Asterias rubens and Asterina gibbosa, Ludwig, 1877, 1878; Asterias rub ens, Hamann, 1885; Asterina gibbosa, Cuenot, 1887; Solaster endeca and Asterias rubens, Gemmill, 1911, 1912 and 1914; Asterina batheri, Hayashi, 1935; Cos- cinasterias tenuispina, Cognetti and Delavault, 1958; Echinaster sepositus, Dela- vault, 1960; Marthasterias glacialis, Delavault and Cognetti, 1961; Odontaster validus, Pearse, 1965; Pisastcr ochraceus, Mauzey, 1966; Asterina gibbosa and Echinaster sepositus, Tangapregassom and Delavault, 1967; Leptasterias hexactis, Chia, 1968; Asterina gibbosa, Brusle, 1969; Henricia sanguinolenta, Chia, 1970; and Astropecten annatus, Henricia leviuscula, Patiria miniata, Pisaster ochraceus, Davis, 1971), but no comprehensive, detailed, and accurate description has been published for any sea-star which discusses the structure of the gonads based on observations with both the light and electron microscopes. In the present study, observations have been made on both sexes of Asterias vulgaris, involving principally the morphology and histology of the gonads; notes are included on the ultrastructure of the tissues of the gonad and also on modifica- tions occurring in the structure of the gonad during a single annual reproductive cycle. The information provided here should be useful to students of the physiology of reproduction in asteroids and will also provide a basis for comparative studies on the reproductive system in other asteroids. MATERIALS AND METHODS Collection and identification of specimens All specimens of Asterias vulgaris used in this study were collected at the Isles of Shoals, a group of islands in the Gulf of Maine about 10 miles southeast of Portsmouth, New Hampshire. The principal collection site for this work was the leeward face of a breakwater stretching in a north-south direction from Cedar Island to Star Island. A series of four collections of sea-stars (referred to as "major collections") was made at this site in October and December, 1970 and in 1 This paper represents part of a thesis submitted to the Graduate School of Cornell Uni- versity in partial fulfillment of the requirements for the degree of Master of Science. 661 662 CHARLES WAYNE WALKER March and July, 1971. Information gathered by Sherman (1966) about the reproductive cycle of Asterias vulgaris on the north side of Cape Cod was used as a guide in selecting the dates of these collections. At her study site, Asterias vulgaris showed a poorly defined yearly reproductive cycle, including a period of gametogenesis from October until March when gametogenesis was complete. Spawning followed during July and continued until August; the resulting spent gonads were then quiescent in terms of gamete production again until October. The "major collections" were made corresponding to Sherman's data ; in this way specimens with gonads demonstrating (1) active gametogenesis, (2) com- pleted gametogenesis, and (3) no gametogenesis (spent gonads) were obtained. Such specimens were used to observe changes in the basic histology of the gonad resulting from the activity of the germinal epithelium. In order to determine that the species used in this study was indeed Asterias vulgaris and not Asterias forbesi, the set of criteria outlined by Aldrich (1956) (amended by some personal observations) was used. Altogether, in the four col- lections, a total of 60 specimens was collected; thirty-five of these, which proved to be Asterias vulgaris, were used. Dissection of specimens Specimens were separately soaked in MgCl2 solution (8% in distilled water) until flaccid. This process relaxes the animal and prevents autotomy during dis- section. Two standard measurements commonly used in sea-star taxonomy, R and r, were recorded for each specimen. R is the length from the center of the disc along the aboral surface of the ray to the tip ; r represents the distance from the center to the edge of the disc in an interradial angle. Each animal was then transferred to fresh sea water and dissected under a Ban sen and Lomb Stereo- Zoom microscope in the following fashion. Two major cuts were made along each ray, one aborally through the carinal series of ossicles from the tip of the ray to the center of the disc, and another similar longitudinal cut orally through the center of the roof of the ambulacral groove. This procedure separated the animal into five pieces, each consisting of halves of two adjacent rays, joined together at the interradial angle. All extraneous tissue was cut away from each of these pieces, leaving five interradii, each containing two complete units of the reproductive sys- tem. Four interradii were then fixed in Bourn's fluid; the fifth was preserved in 70% ethanol. Procedures for major collections Specimens were allowed to stand in Bouin's fluid for periods of up to one month, which thoroughly decalcified them. The five interradii representing each animal were separated into two groups and processed as follows: Group 1. Two entire interradii were dehydrated in ethanol, embedded in paraffin in a vacuum oven, and sectioned serially either longitudinally or trans- versely to the axis of the rachis of the gonad (see Observations) at 7 /z to 10/*. One set of sections from each specimen was routinely stained with Mallory's Phosphotungstic Acid Hematoxylin (PTAH) (Lillie, 1965) for general histologi- THE GONAD OF ASTERIAS J'ULGARIS 663 cal observations and to reveal details of muscular and connective tissues. Selected regions of the second set of sections were stained with the following techniques: (a) Gomori's silver stain (Gomori, 1947) — to differentiate "reticular" collagen fibers in various connective tissue components of the reproductive system; (b) Aldehyde Fuchsin (Cameron and Steele, 1959) counterstained wTith Halmi's stain (1952) — to differentiate between elastic connective tissue and muscle and other connective tissue; and (c) Periodic Acid Schiff (PAS) (Lillie, 1965) — to demonstrate the distribution and abundance of the haemal fluid in the haemal sinus (see Observations). The presence of glycogen and other polysaccharide complexes is indicated by this technique. Group 2. Gonads in the remaining interradii were used for general observa- tions of shape and form of the gonad. The interradius which had been preserved in 70% ethanol was macerated in KOH solution (2% in distilled water) for periods of 2 to 20 minutes, depending on the toughness of the tissues of the ray, until the ossicles could be seen clearly. This preparation was preserved and used for observations to determine the point of suspension of the gonad. Electron microscopy In July, 1972, 6 specimens of Aster ias vulgar is were collected on the leeward side of the breakwater at the Isles of Shoals. These animals consisted of 2 females and 4 males. The specimens were fixed under field conditions in the following way. The gonads of the 6 animals were removed after each specimen had been relaxed separately in MgClo solution. Several blocks of tissue, 3-4 mm thick, taken from various parts of each gonad, were fixed at room temperature for 20-30 minutes in 3% paraformaldehyde-glutaraldehyde (Longo and Anderson, 1969) in sea water. The blocks were then diced into 1 mm cubes of tissue, and fixation was continued in a small refrigerator at 4° C. for 1-2 hours. The tissues were rinsed in 3 20-minute changes of seawater and post-fixed for 2 hours in \% OsC>4 in seawater, followed by 4 10-minute seawater rinses. Dehydration of the tissues was carried out in ethanol (10 minutes each in 25%, 50%, 70%, ^5%. 100% and 100% solutions). The tissues were then transferred to a 10-minute rinse in propylene oxide: 100% ethanol, 1:1, followed by 2 10-min- ute rinses in 100% propylene oxide, and embedded in Epon 812/Araldite 506 mixture (Mollenhauer, 1963). Silver-gray sections were cut on a Porter-Blum MT-2B ultramicrotome using glass knives and subsequently picked up on 0.25% formvar-coated 200 mesh grids. All sections were stained with both 2% uranyl acetate (30 minutes) and lead citrate (15 minutes) (Reynolds, 1963) and examined on a Philips EM 300 operated at 80 kV. The results from all these procedures have been combined to provide a de- tailed account of the morphology and histology of the gonad. OBSERVATIONS General morphology of the gonad In either sex of a normal five-rayed specimen of Asterias vulgaris, the repro- ductive system comprises 10 separate units, two in each ray. Each unit is attached 664 CHARLES WAYNE WALKER EE 1 FIGURE 1. Diagrammatic representation of the inner lateral wall of the ray (on the leftside) at the point of attachment of one unit of the reproductive system, showing the relationships of various layers of tissue which make up the gonad with those in the wall of the ray. Symbols used are : C, coelom ; D, dermis ; EE, external epithelium ; EG, epithelium of the gonoduct ; G, gonoduct; GCS, genital coelomic (perihaemal) sinus; GCT, gonoduct connective tissue; GHS, genital haemal strand; IS, inner sac; L, lumen of the gonad; M, major acinus; MA, minor acinus; OS, outer sac; PP, parietal peritoneum; SCT, subperitoneal connective tissue; VP, visceral peritoneum. proximally near the disc on the inner face of the lateral wall of the ray, below the pyloric caecum, and the principal parts of the unit, the gonad, the gonoduct, and the genital haemal strand-genital coelomic (perihaemal) sinus complex, ex- tend to other parts of the ray. The major features of a typical unit are shown diagrammatically in Figure 1. The gonad is large and obvious and projects freely into the coelomic cavity along the length of the ray ; the gonoduct and genital THE GONAD OF ASTERIAS VULGARIS 665 FIGURE 2. Diagrammatic representation of the typical arrangement and appearance of the tissues in the wall of the gonad. Symbols used are: BM, basement membrane; CF, collagen fibers ; CTF, connective tissue fibers ; CTL, connective tissue layer ; EP, epithelium ; F, flagel- lum; FM, fibrous matrix; GE, germinal epithelium; GCS, genital coelomic (perihaemal) sinus; HC, cell in the haemal sinus; HS, haemal sinus; HSS, haemal sinus space; IMF, inner longitudinal muscle fibers; IW, inner wall of haemal sinus; L, lumen of the gonad; MV, microvilli : OMF, outer circular muscle fibers; OW, outer wall of haemal sinus; SMF, sub- peritoneal muscle fibers ; VP, visceral peritoneum. haemal strand (and its associated coelom) are microscopic, the former penetrating the wall of the ray and opening aborally by several gonopores in the epithelium of the interradius, the latter connecting with the aboral haemal and coelomic rings in the disc. In both sexes, the gonad is a single large bag consisting of a number of parts. A major piece, the rachis, forms the axis of the gonad; by this piece the gonad 666 CHARLES WAYNE WALKER -: c, A. „ .^, .003mm | •a_ /^-.^•' y, . r THE GONAD OF ASTERIAS I'ULGARIS 667 is suspended from the body wall through a single stalk of connective tissue above the supra-marginal ossicles of the ray. Many projections extend from this axial piece into the coelom. A varying number (10-30) of elongate tubules, the major acini, project from the medial margin of the rachis toward the ambulacral plates. The remaining surface of the rachis, and often much of the major acini, are marked by short rounded protuberances, the minor acini. Histology of the gonad The wall of the gonad is composed of several layers of tissue arranged in two groups, the outer and inner sacs, details of which are shown in Figure 2. Al- though these layers are always found in the same order wherever they occur, their relative dimensions and general appearance differ markedly depending on the part of the gonad they form and on the degree of gametogenic activity occurring in the germinal epithelium at the time. The outer sac consists of a single stratum of visceral peritoneal cells that rests on a layer of fibrous connective tissue to which muscle and epithelial cells are attached internally. The inner sac of the gonad is often very thin, complex, and difficult to interpret; full understanding of its structure necessitates the use of low magnification electron microscopy. It is composed of the haemal sinus and associated tissues. In this paper, "haemal sinus" refers to a genital branch of the aboral haemal ring found in association with the germinal epithelium of the gonad. The term is used collectively to include (1) the haemal sinus space, which may contain (2) granular PAS-positive haemal fluid and cells, enclosed by (3) two fibrous layers, called the outer and inner walls of the haemal sinus. The outer wall serves externally as a basement membrane for muscle and epithelial cells FIGURE 3. Electron micrograph of a section of a testis, showing the cells of the visceral peritoneum: Uranyl acetate and lead citrate. Symbols used are: BM, basement membrane; C, coelom, CF, collagen fibers ; CP, cellular process ; F, flagellum ; MV, microvilli ; N, nucleus ; NSG, neurosecretory granules; TJ, tight junction. FIGURE 4. Cross-sectional view of the wall of an ovary to show the general appearance of the tissues that compose it. Compare with Figures 2, 7, and 15. PTAH. Symbols used are: C, coelom; CTL, connective tissue layer: GE, germinal epithelium : GCS, genital coelomic (perihaemal) sinus; HS, haemal sinus; IMF, inner longitudinal muscle fibers; OMF, outer circular muscle fibers; SMF, subperitoneal muscle fibers; VP, visceral peritoneum. FIGURE 5. A section of the wall of an ovary at the base of a minor acinus where the darkly staining borders of the outer connective tissue layer are obvious ; muscle and epithelial cells lie within the folds of the inner border (arrows: lightly stained here) ; Aldehyde fuchsin, Halmi's stain. Symbols used are: C, coelom; GCS, genital coelomic (perihaemal) sinus; GE, germinal epithelium ; HS, haemal sinus. FIGURE 6. Section of the wall of an ovary at the base of a minor acinus, showing the distribution of the fibers of the outer connective tissue layer (arrows) and the connective tissue fibers of both walls of the haemal sinus near the germinal epithelium ; Gomori's silver stain. Symbols used are: C, coelom; GCS, genital coelomic (perihaemal) sinus; GE, germinal epithelium; HS, haemal sinus. FIGURE 7. Electron micrograph of a section of an ovary showing features of the con- nective tissue layer of the outer sac of the gonad. Notice, in particular, the wavy basement membranes of the visceral and genital-coelomic peritoneal layers ; the arrows indicate the fine fibers of the matrix. Compare with Figure 4 ; Uranyl acetate and lead citrate. Symbols used are: CF, collagen fibers; CP, cellular processes; N, nucleus of an epithelial cell; OMF outer circular muscle fibers. 668 CHARLES WAYNE WALKER IMF THE GONAD OF ASTERIAS VULGARIS 669 found in the genital coelomic (perihaemal) sinus; the inner wall serves internally as a basement membrane for the germinal epithelium. A distinct genital coelomic (perihaemal) sinus lies between the sacs which make up the wall of the gonad (Fig. 2, GCS). This sinus extends through the con- nective tissue stalk suspending the gonad and in the wall of the ray connects with a genital branch of the aboral coelomic (perihaemal) ring (Fig. 1, GHS). Such an attachment exists for each gonad, and thus all ten gonads are interconnected by way of this coelomic sinus. Both sacs are attached to the connective tissue of the gonoduct near its orifice within the gonad (Fig. 1). Elsewhere, although the two sacs may be in contact and nearly indistinguishable from each other at the limit of resolution of the light microscope, observations utilizing the electron microscope show clearly that they remain physically distinct and maintain their separate identities. In the living animal, a fluid is undoubtedly present in the genital coelomic (perihaemal) sinus, but no trace of it is evident in fixed, sectioned, and stained material. We now proceed to describe specific components of the wall of the gonad in greater detail. The visceral peritoneal cells of the gonad are directly continuous with those of the parietal peritoneum (Figs. 1, 2, and 4, VP). The concept of these cells which is formed from observations using the light microscope is relatively simple and incomplete, because the cells are quite small and are usually not well preserved. The epithelial cells are basically cuboidal ; however, their shape is quite variable and depending upon the degree of expansion of the gonad, the cells may be flat, cubical, or tall and columnar in appearance on various regions of the same gonad. In gonads that have completed gametogenesis, most cells are flat (squamous), while in spent gonads or those involved in gametogenesis. all three types of cells may be found. Observations with the electron microscope reveal additional details (Fig. 3). The cytoplasm of most cells contains few organelles, although a large FIGURE 8. Electron micrograph of a section of an ovary, showing muscle fibers of the outer and inner sacs of the gonad. Notice the firm attachment of the inner longitudinal muscle fibers to the outer wall of the haemal sinus farrows) ; Uranyl acetate and lead citrate. Symbols used are: EP, epithelium covering the outer wall of the haemal sinus; GCS, genital coelomic (perihaemal) sinus; IMF, inner longitudinal muscle fibers; N, nucleus of epithelial cell; OMF, outer circular muscle fibers; OW, outer wall of haemal sinus. FIGURE 9. Longitudinal section of one minor acinus from an ovary, showing the orienta- tion of the outer (OMF) and inner (IMF) muscle fibers; PTAH. Symbols used are: C, coelom; GCS, genital coelomic (perihaemal) sinus. FIGURE 10. Section of an ovary, showing adjacent regions of the haemal sinus which appear either as a jumble of connective tissue fibers (A) or as an expanded space (HS) ; the connective tissue fibers are embedded on the luminal side of both the outer and inner walls of the sinus ; Gomori's silver stain. Symbols used are : C, coelom ; GE, germinal epithelium. FIGURE 11. Several major acini sectioned longitudinally showing the general distribution of the haemal sinus and its PAS-positive contents; PAS. Symbols used are: C, coelom; HS, haemal sinus. FIGURE 12. Electron micrograph cf a section of an ovary, showing the haemal sinus and its components in a region where the walls of the sinus are close together. The haemal sinus space contains cells and not haemal fluid ; Uranyl acetate and lead citrate. Symbols used are: GE, germinal epithelium; HS, haemal sinus; IMF, inner longitudinal muscle fibers; IW, inner wall of haemal sinus ; N, nucleus of a cell in the haemal sinus ; OW, outer wall of haemal sinus. 670 CHARLES WAYNE WALKER THE GONAD OF ASTERIAS VULGARIS 671 nucleus, golgi apparatus, mitochondria, flagellar basal bodies, and granules are present. Microvilli surround the base of the single flagellum of each cell as a distinct circle and are also distributed randomly over the free surfaces of cells. The peritoneal cells are attached to each other by tight junctions, often through extremely attenuated processes. Membrane bound granules are found between and below the typical peritoneal cells (Fig. 3, NSG) and are similar to structures identified by Brusle (1969) and Davis (1971) as nerve cell processes. Fibers which stain yellow with Orange G and blue with PTAH but do not stain with silver are occasionally found in thick regions of the outer sac below the peritoneum surrounding the bases of the major and minor acini (Figs. 2 and 4, SMF). Such fibers may form a network at the base of the peritoneum; prelimin- ary observations with the electron microscope show that they are muscle fibers. The outer connective tissue layer (Figs. 2 and 4, CTL), is continuous with a similar layer located subperitoneally in the wall of the ray (Fig. 1, SCT). This continuity is evident across the stalk suspending the gonad, which is composed exclusively of tissues corresponding to those of the outer sac of the gonad (Fig. 1 ). In the animals observed, the maximum thickness of this layer is 30 /A to 35 /A; it is thickest along the wall of the rachis especially around the bases of the major acini and near the point of suspension of the gonad. In other places, as on the distal tip of the gonad or on rounded surfaces of the acini, it may be 2 ^ or less in thickness. With the light microscope, the connective tissue layer appears to be condensed on both borders, and these regions stain dark blue with aldehyde fuchsin (Fig. 5). Sandwiched between these two condensed laminae is a thick zone of fibers which stain pink with PTAH (Fig. 4, CTL) and also stain with Gomori's silver tech- nique (Fig. 6, arrows). This layer is quite elastic and resilient even in preserved material. Observations of this connective tissue with the electron microscope (Fig. 7) support and extend the interpretation just presented. The entire layer is com- posed of a matrix of fine fibers which is often interrupted or replaced by collagen fibers of various sizes running in all directions and by occasional groups of cells or cellular processes. The outer and inner margins of this layer, seen as condensed FIGURE 13. Section of a testis showing the thin inner wall of the haemal sinus as seen hy light microscopy (arrows) and the indistinct nature of the outer wall of the sinus; PTAH. Symbols used are : C, coelom ; GE, germinal epithelium ; L, lumen of the gonad. FIGURE 14. Section of a minor acinus of an ovary showing the single cell layer covering the outer wall of the haemal sinus (arrows) (see Figure 8) and the imaginations projecting into the lumen of the gonad formed by the inner wall of the haemal sinus and the germinal epithelium; PTAH. Symbols used are: C, coelom; GCS, genital coelomic ( perihaemal) sinus; HS, haemal sinus ; L, lumen of the gonad. FIGURE 15. Electron micrograph of the wall of an ovary taken at low magnification. Using this figure, compare the resolution of details of the various tissue components of the wall of the gonad with that seen in light micrographs of similar subjects, in particular, Figure 4; Uranyl acetate and lead citrate. Symbols used are: C, coelom; CTL, connective tissue layer; GE, germinal epithelium; GCS, genital coelomic (perihaemal) sinus; HS, haemal sinus; IMF, inner longitudinal muscle fibers; INV, imagination of the inner wall of the haemal sinus and the germinal epithelium; OMF, outer circular muscle fibers; VP, visceral peritoneum. 672 CHARLES WAYNE WALKER borders with the light microscope, form basement membranes for the visceral peritoneum and muscle and epithelial cells of the genital coelomic (perihaemal) sinus, respectively. These basement membranes are usually wavy in appearance as seen with both the light and electron microscopes (Figs. 5 and 7) and are composed of fine fibers similar to those in the matrix with which they are inti- mately associated. Complex, sometimes large cells and attenuated cellular pro- cesses are found among the fibers of the connective tissue layer ; these are assumed to be associated with the formation of fibers. The muscle fibers attached to the inner basement membrane of the outer sac of the gonad (Figs. 2, 4, 7 and 8, OMF) stain yellow with Orange G, blue with PTAH, and not at all with Gomori's silver stain. These fibers run circularly around the axes of the rachis and acini of the gonad (Fig. 9). Viewed in sections, they often lie close against the connective tissue layer, within folds formed by this layer ( Figs. 2, 5 and 7). Observations with the electron microscope indicate that a discontinuous epi- thelium lines the genital coelomic (perihaemal) sinus throughout the gonad (Figs. 2, 7, and 8). The cells of this epithelium occur among muscle fibers of the outer and inner sacs ; they are often amoeboid and many are flagellated. Membrane bound granules associated with the epithelial cells of the genital coelomic (peri- haemal) sinus are similar to those seen in several sea-stars by Davis (1971). While all these tissues of the outer sac of the gonad may vary in thickness and shape, the tissues of the inner sac are even more variable ; their appearance de- pends upon the activity of the germinal epithelium. Muscle fibers are often numer- ous in association with the outer wall of the haemal sinus only (Figs. 2, 4, and 8, IMF); they are intimately connected to this wall externally (Fig. 8, arrows). Such fibers are oriented longitudinally along the axes of the rachis and acini form- ing a basket of muscle fibers (Fig. 9). Tissues associated with the haemal sinus of the gonad are similar to those of the genital haemal strand with which the sinus is continuous near the gonoduct (Fig. 1). Through the light microscope, the sinus appears either as a single layer or often in adjacent regions as two layers separated by the haemal sinus space. These layers are quite indistinct but correspond to the walls of the haemal sinus. Silver staining reveals distinct, thin, connective tissue fibers (clearly colla- gen fibers as seen with the electron microscope) running in random directions on the luminal side of both walls (Fig. 10). At certain times of the year (in this study, October and July, in particular) separate channels of the haemal sinus expand and become continuous ; the sinus space is filled with PAS-positive haemal fluid evident in sectioned and stained material as a reddish-pink coagulate (Fig. 11, HS). Details of the haemal sinus are resolvable with the electron microscope. Only with the high resolution provided by this instrument can it be appreciated that both walls of the haemal sinus are present throughout the gonad ; they are often wavy in appearance and may send projections into the lumen of the haemal sinus (Fig. 12). Each wall is composed of a fine fibrous material similar in ap- pearance to the matrix portion of the connective tissue layer of the outer sac. Except in very unusual cases, only the thin inner wall of the haemal sinus is dis- cernable with the light microscope (Fig. 13, arrows), although, when the sinus THE GONAD OF ASTER! AS I'ULGARIS 673 is maximally expanded, the existence of the outer wall is suggested by the single layer of epithelial cells which covers it externally (Figs. 2 and 8, EP and 14, arrows). Both walls of the sinus may stain brightly with PAS and aldehyde fuchsin. The inner wall of the haemal sinus and its associated germinal epithelium often form deep folds that project into the lumen of the gonad (Figs. 2, 4, 10, 11, 13, 14 and 15). In sections these may appear triangular or they may have various shapes; similar invaginations were noticed by Mauzey (1966) in Pisastcr and Chia (1968) in Leptasterias. The haemal sinus space generally contains granular fluid in which collagen and other fine fibers are often apparent ; the orientation of these fibers may be random or may parallel the walls of the sinus. Cells with vari- ous shapes and contents occur throughout the sinus space (Figs. 12 and 15) ; some are amoeboid with long cytoplasmic processes and others are contracted and filled with granules. In mature animals, the germinal epithelium forms a continuous lining of the lumen of the gonad; it connects with the epithelium of the gonoduct (Fig. 1, EG). The germinal epithelium is composed of follicular cells, oogonia, and oocytes in females and spermatogonia, spermatocytes, and maturing spermatozoa in males. Of all the tissues mentioned, seasonal changes are most dynamic in the germinal epithelium in connection with its yearly production of gametes. These changes have been described by a number of authors for several sea-stars and the essentially similar changes seen in Astcrias rulf/nris will not be discussed here. DISCUSSION The observations just concluded provide a detailed description of the gonad of Astcrias vnlgaris in morphological and histological terms ; morphologically the gonads of this species differ from those seen in most other sea-stars, histologically they appear to be quite similar to those in any asteroid. From the standpoint of microscopic anatomy it is fortunate that descriptive studies of various aspects of the histology of the gonads of asteroids are numerous. Collectively these provide a good outline which is applicable in general to almost any sea-star. As mentioned earlier, however, no studies are available which give a detailed description of the tissues that make up the wall of the gonad in both sexes of any species based on observations using both the light and electron microscopes. Many features of the general organization of various tissue layers of the gonad are obvious only with the light microscope, while details of the histology of the tissues are obvious only with low magnification electron microscopy. In the following section, the histology of the wall of the gonad of Asterias vulgaris will be compared with cor- responding features of the gonads of other sea-stars which have been previously described. The visceral peritoneum of the outer sac of the gonad has been described by some authors as a single layer of ciliated cuboidal cells (Field, Astcrias vulgaris, 1892; Gemmill, Solaster endcca. 1912; and Chia, Leptasterias hc.vactis and Henrida sangii'molenta, 1968, 1970). In contrast, some observers state that it consists of ciliated flat cells (Ludwig and Hamann, 1899, and Davis, 1971). From the present study, it is apparent that this layer is basically composed of simple flagel- 674 CHARLES WAYNE WALKER lated cuboidal cells like those which make up the peritoneum elsewhere. Local pressures and stresses result in their being stretched thin over protuberances or laterally compressed in folds and angles. Preliminary observations on the ultra- structure of the peritoneal cells in the present study are in general agreement with those of Tangapregassom and Delavault (1967) and Davis (1971). Brusle (1969) and Davis (1971) indicate that nerve cell processes containing granules that are possibly neurosecretory in nature are often seen below and between the peritoneal cells of the gonads of the sea-stars with which they worked. Granules were ob- served in the present study in a similar relationship with the visceral peritoneum, but the evidence is insufficient to confirm identification of such granules with nerv- ous tissue in Astcrias vidgaris. The muscle fibers seen in association with the peritoneum of the gonad in Asterias vulgaris have never been described in other species. Their function in the gonads of this species is unknown, although they might cause local deformations of the peritoneum that direct movement of coelomic fluid over the surface of the gonad. Hoffmann (1872). in his studies of Asterias mbens, was the first investigator to mention the connective tissue of the outer sac of the gonad, describing it as com- posed of delicate homogeneous connective tissue. More detail is presented by Gemmill (1914), who notes that in Asterias mbens the outer connective tissue layer may be subdivided into superficial and deeper sheets. Tangapregassom and Delavault (1967), Brusle (1969), and Davis (1971) state that it is composed of collagen fibers in which there are occasional pockets of cells ; other investigators mention that it exists but give no details. It is apparent from observations described earlier that this layer is extremely tough and elastic in Asterias vulgaris. Its borders serve as basement membranes, stain brightly with aldehyde fuchsin, are usually wavy in sections, and are very likely similar to the elastic tissue of vertebrates. The elastic basement membranes of this layer allow the gonad to expand while the abundant collagen fibers found between them maintain its basic shape. The circular muscle fibers facing the genital coelomic (perihaemal) sinus on the inside of the outer connective tissue layer are similar to the circular fibers described by Hoffmann (1872) for Asterias mbens and by Gemmill (1912) in the ovary of SoJaster endeea. Hayashi (1935), Chia (1968), Tangapregassom and Delavault (1967), Brusle (1969), and Davis (1971), all recognize muscle fibers in a similar location in the gonads of the various species with which they worked, but according to Hayashi (1935), quoting from the observations of Ohshima (1925), this layer is not present in the testes of Asterina batheri. Contraction of these muscle fibers is probably important during the shedding of gametes. The epithelial cells which line the genital coelomic (perihaemal) sinus of the gonad are very likely related to those in the wall of the ray which line the genital branch of the aboral coelomic (perihaemal) ring; they are flagellated and prob- ably keep the fluid contents of the genital coelomic (perihaemal) sinus in motion. Longitudinal muscle fibers were described by Gemmill (1912) on the inner sac of the gonad of Solasfer endeca. Hayashi (1935), Chia (1968), Tangapre- THE GONAD OF ASTERIAS VULGAR1S 675 gassom and Delavault (1967), Brusle (1969), and Davis (1971) also mention muscle fibers, similarly located, in the gonads of various species, without specifying their orientation. In Asterias I'ulgaris, as mentioned, longitudinal muscle fibers are attached only to the outer wall of the haemal sinus facing the genital coelomic (perihaemal) sinus. The function of these fibers remains unknown, although by their contraction they may exercise some control over the volume of fluid con- tained in the haemal sinus. Although most previous investigators have mentioned the presence of tissues associated with the haemal sinus, none has given a detailed description of the com- ponents of the sinus. Taken together, the works of Gemmill (1914) and Davis (1971) give the most complete account. Gemmill indicates that a branch of the genital haemal ring connects with the inner sac of the gonad near the gonoduct and then branches irregularly over the surface of this sac ; Davis describes the haemal sinus as a space filled with granular contents, cells, and collagen fibers, surrounded by two laminae. From the present study, it is clear that the sinus is an expansible bag with walls that are possibly elastic in character; the extreme variability in form and appearance of the haemal sinus is closely associated with changes in the condition of the germinal epithelium. In gonads which have com- pleted gametogenesis, the sinus is nearly empty, its walls are pressed together, and the haemal sinus space is occluded. Alternatively, in gonads which are either involved in gametogenesis or have spawned recently, the haemal sinus may be filled with fluid and cells and its walls are widely separated. Work by Mauzey (1966) and Chia (1968) has shown that haemal fluid is strongly PAS-positive. Mauzey (1966) finds that the haemal fluid in Pisaster ochracens does not contain glycogen since haemal fluid is PAS-positive before and after salivary amylase treatment. Chia concludes that this fluid may store and supply nutrients to the developing germinal cells, citing as evidence its location and the time during the reproductive cycle when it becomes conspicuous. In the present study, haemal fluid was present in greater volume in females than in males; it was found while gametes were maturing and was very abundant after gametes had been shed. No evidence is available on the function of this fluid in Asterias I'ulgaris. Through the aboral haemal ring and its genital branches, the haemal sinus of each gonad is connected with those found in all 9 other gonads and also indirectly to other organ systems in the body : the pyloric caeca, axial organ, radial nerves of the rays, etc. Such a system of channels between the reproductive organs and other organ systems is no doubt important in transmitting fluids and other ma- terials to and from the gonads, and also in coordinating the activities of all 10 gonads during the annual reproductive cycle. From every aspect, the studies reported here reveal the limited amount of de- tailed information upon which our understanding of structure and function in the eonads of asteroids is based. It is thus obvious that additional studies, utilizing as many and varied species as possible, are highly desirable. I wish to express my gratitude to Dr. John M. Anderson for his careful guid- ance throughout the course of this study and for his meticulous editorial assistance 676 CHARLES WAYNE WALKER in the preparation of this manuscript. I also thank my wife, Wilise, for her valuable assistance and encouragement. SUMMARY The results reported here provide a detailed account of the general morphology and histology of the male and female gonads of the sea-star Asterias vulgaris. The reproductive system of this sea-star (normal five-rayed specimens) consists of 10 separate units, each located proximally on the lateral wall of the ray, one on either side of the ray. Each unit is composed of a gonad, a gonoduct, and genital branches of the aboral haemal and coelomic rings. The gonad is a single hag-like structure with several protrusions (termed major and minor acini ) extending from its surface. Its wall is composed of two sacs, one inside the other, separated by the genital coelomic (perihaemal) sinus. The outer sac consists of visceral peritoneum, an elastic-collagenous connective tissue layer, and many epithelial cells and circular muscle fibers. The inner sac comprises epithelial cells and longitudinal muscle fibers, the haemal sinus and contents, and germinal epithelium. The haemal sinus includes the haemal sinus space, filled with granular haemal fluid, cells, and collagen and other fine fibers enclosed by two fibrous laminae. Significant modifica- tions in the form of the gonad and in the condition and relationships of the tissues which compose it occur during the annual reproductive cycle. Both sacs are stretched during growth of the gonad, the outer layers becoming attenuated and the inner layers being pressed against the outer, often obliterating the genital coelo- mic (perihaemal) sinus. The inner group of tissues is often extensively folded, pushing ridges formed from the inner wall of the haemal sinus and germinal epithelium into the lumen of the gonad. It is pointed out that previous studies on the gonads of asteroids have been relatively few, with no study for any species dealing comprehensively with morpho- logical and histological details of the gonad based on both light and electron microscopy. Comparison of the results of the present study with observations of previous investigators indicates that although significant differences occur (especi- ally in morphological terms), the general features of the histology of the wall of the gonad of many sea-stars are similar. In order to broaden our base for com- parative studies, and to pursue significant problems, morphological, histological, histochemical, and ultrastructural investigations should be extended to as many asteroid species as possible. LITERATURE CITED ALDRICH, F. A., 1956. A comparative study of the identification characters of Asterias forbesi and Asterias vulgaris. Notidac Natur. Acad. Natur. Sci. Philadelphia, 285: 1-3. RRUSLE, J., 1969. Aspects ultrastructuraux de 1'innervation des gonades chez 1'etoile de mer Asterina gibbosa P. Z. Zelljorsch. Mikrnsk. Anat., 98: 88-98. CAMERON, M. L., AND J. E. STEELE, 1959. Simplified Aldehyde-Fuchsin staining of neuro- secretory cells. Stain Technol., 34: 256-266. CHIA, F. S., 1968. Some observations on the development and cyclic changes of the oocytes in a brooding starfish, Lcptastcrias hcxactis. J. Zoo/., 154: 453-461. CHIA, F. S., 1970. Some observations on the histology of the ovary and RNA synthesis in the ovarian tissues of the starfish, Hcnricia sanguinohnta. J. Zool., 162: 287-291. THE GONAD OF ASTERIAS VULGARIS 677 COGNETTI, G., AND R. DELAVAULT, 1958. Aspetti istologici delle gonadi in Coscinasterias tenuispina Lmk. Monit. Zool. Italiano, 66 : 44-48. CUENOT, L., 1887. Contribution a 1'etude anatomique des Asterides. Arch. Zool. Esp. Gen., Ser. 2, 2 : 1-144. DAVIS, H., 1971. The gonad walls of echinodermata : a comparative study based on electron microscopy. Master's thesis, University of California, San Diego, 90 pp. DELAVAULT, R., 1960. La sexualite chez Echinaster sepositus Gray du Golfe de Naples. Pubbl. Stas. Zool. Napoli, 32 : 41-57. DELAVAULT, R., AND G. COGNETTI, 1961. L'hermaphroditisme chez Marthasterias glacialis. L.C.R. Acad. Sci. Paris, 253 : 2771-2772. FIELD, G. W., 1892. The larva of Asterias inilgaris. Quart. J. Microscop. Sci.. 34 : 105-128. GEMMILL, J. F., 1911. Adult anatomy of Solaster cndcca. Proc. Roy. Phvs. Soc. Edinburgh, 18: 174-191. GEMMILL, J. F., 1912. Development of Solastcr endcca. Trans. Zool. Soc. London, 20 : 1-71. GEMMILL, J. F., 1914. The development and certain points in the adult structure of Asterias rubens L. Phil. Trans. Roy. Soc. London, 205 : 213-294. GOMORI, G., 1947. Silver impregnation of reticulum. Amer. J. Pathol., 13 : 993-1001. HALMI, N. S., 1952. Differentiation of two types of basophils in the adenohypophysis of the rat and mouse. Stain Tcchnol., 27 : 61-64. HAMANN, O., 1885. Beitrdge cur Histologie dcr Echinodcnncn, Heft 2. Die Asteriden ana- tomisch und histologisch untersucht. Jena, Fisher. HAYASHI, R., 1935. Studies on the morphology of Japanese Sea-Stars, II Internal anatomy of two short-rayed sea-stars, Patiria pcctinifera (Midler and Troschel) and Astcrina batheri Goto. /. Fac. Sci., Hokkaido Imp. Univ., Ser. VI, Zool., 4 : 197-212. HOFFMANN, C. K., 1872. Zur Anatomic der Asteriden. Nicderland. Archiv. Zool.. 2 : 1-32. HYMAN, L. H., 1955. The Invertebrates. Volume IV, Echinodermata. New York, McGraw- Hill. LILLIE, R. D., 1965. Histopathologic Technic and Practical Histochemistry, [3rd. ed.] New York, McGraw-Hill, 715 pp. LONGO, F. J., AND E. ANDERSON, 1969. Sperm differentiation in the sea urchins Arbacia punctulata and Strongyloccntrotus purpuratus. J. Ultrastnict. Res., 27 : 486-509. LUDWIG, H., 1877. Beitrage zur Anatomic der Asteriden. Z. IViss. Zool., 30: 99-162. LUDWIG, H., 1878. Uber die Genitalorgane der Astcrina gibbosa. Z. Wiss. Zool., 32: 395-400. LUDWIG, H., AND O. HAMANN, 1899. Echinodermen, II Buch. Die Seesterne. Pages 591-604 in H. G. Bronn, Ed., Klasscn und Ordnnngen des Ticrreichs, Bd. 2, Abt. 3. Leipzig, Winter. MAUZEY, K. P., 1966. Feeding behavior and reproductive cycles in Pisaster ochraccus. Biol. Bull.. 131: 127-144. MOLLENHAUER, H. H., 1963. Plastic embedding mixtures for use in electron microscopy. Stain Tcchnol., 39: 111-144. OHSHIMA, H., 1925. Hermafrodita marstelo. Dobntsugaku Zasshi, 37 : 203-213. PEARSE, J. S., 1965. Reproductive periodicities in several contrasting populations of Odontastcr validus Koehler, a common Antarctic asteroid. Biology of the Antarctic Seas. II. Antarctic Res. Ser., 5 : 39-85. REYNOLDS, E. S., 1963. The use of lead citrate at high pH as an electron opaque stain in electron microscopy. /. Cell. Biol., 17 : 208-212. SHERMAN, V. T., 1966. The reproductive biology of selected echinoderms from Cape Cod, Massachusetts. Master's thesis. University of California. Los Angeles. 137 pp. TANGAPREGASSOM, A. M., AND R. DELAVAULT, 1967. Analyse, en microscopic photonique et electronique des structures peripheriques des gonades chez deux etoiles de mer Asterina gibbosa Pennant et Echinaster sepositus Gray. Cah. Biol. Mar., 8 : 153-159. Reference: Biol. Bull., 147: 678-689. (December, 1974) THE OCCURRENCE, DISTRIBUTION AND ATTACHMENT OF THE PEDUNCULATE BARNACLE OCTOLASMIS MVLLERI (COKER) ON THE GILLS OF CRABS, PARTICULARLY THE BLUE CRAB. CALLINECTES SAPIDUS RATHBUN 1 GRAHAM WALKER N.E.R.C. Unit of Invertebrate Biology, Marine Science Laboratories, University C Ullci/c of Xurtli H'tilcs. Mcnai Hridi/c, . ln//lcsc\\ l'.I\. During the summer of 1973 crabs were collected regularly from Beaufort Inlet, North Carolina and examined for the presence of the epizoic barnacle, Octolasmis iniillcri (Coker), within the branchial chambers. The blue crab, Callinectes sapid us Rathbun is the most common crab in Beaufort Inlet (Dudley and Judy, 1971) and was therefore chosen as the host for a detailed study of 0. iniillcri. The life-history and migrations of C. scipidus (Van Engel, 1958; Cargo, 1958; Costlow and Bookhout, 1959; Tagatz, 1968; Judy and Dudley, 1970; Dudley and Judy, 1971) have been studied because this crab is exploited commercially. The migrations of this crab may also influence the life-histories of epizoites. The present study investigates the occurrence, distribution and mode of attachment of the epizoic O. miilleri on the gills of C. sapidits and discusses the possible cor- relations with the life cycle and habits of the crab. MATERIALS AND METHODS When barnacles were found to be present in a blue crab, all the gills were re- moved and the number and position of the barnacles on each gill noted. During July several blue crabs were collected which had only the recently settled cypris larva of the barnacle present on the gills. These larvae, together with the portion of the gill to which they were attached, were fixed for a histological study. Adult specimens of O. miilleri were also fixed in Carnoy's and Bouin's fixatives for a histological study of the cement apparatus. Wax sections (5 /j. thick) were stained with Heidenhain's azan. The peduncles of both 0. iniillcri and the related Lepas anatifera were fixed in 1% OsO4 in veronal buffer, de- hydrated and embedded in Araldite. One ft thick transverse sections of the peduncles stained with toluidine blue in borax (1% aqueous solution) were examined with the light microscope. RESULTS The 4th gill of C. sapid us is drawn diagrammatically in Figure la. It is com- posed of many paired platelets (pi.) which have well defined spines and knobs around their outer margin (Fig. lb). The larger efferent blood vessel (e.v.) and the smaller afferent vessel (a.v.) are situated on the inner and outer side of the 1 This study was carried out at the Duke Marine Laboratory, Beaufort, N.C. 28516. 678 BARNACLE ON CRAB GILLS 679 FIGURE 1. (a) A drawing of the 4th gill of Callincctcs sapidus showing the paired platelets (pi.), small afferent vessel (a.v.) and larger efferent vessel (e.v.) ; (b) drawings of gill platelets taken from different regions (1, 2, 3, 4) of the same gill. The platelets are divided into afferent (aff.) and efferent (cff.) zones and show the marginal projections: spines a, b, d. g and knobs c and e; a.v., afferent vessel; e.v., efferent vessel; sp., spine. gill respectively. Most of the marginal projections of the platelets emerge at an angle to the surface of the platelet itself and some of these presumably act as "spacers" keeping the platelets separated and allowing water circulation. The spines a, b, d, f and g (Fig. Ib) can be picked out with the naked eye, but the more interior knobs c and e are hidden from view. In the proximal region of a gill the platelets possess all the spines and knobs previously mentioned, but spines g and knob c disappear towards the distal end. Seven of the eight pairs of gills in the crab lie close to one another so that platelets of adjacent gills touch. Each platelet can be divided equally into afferent and efferent zones (Fig. Ib). Examina- TABLE I Variation in the barnacles settled on each of 8 pairs of gills from 25 blue crabs. Source of variation Degrees of freedom Mean square F-ratio Between gill> Within gills Total 7 192 0.1961 0.00133 147 199 680 GRAHAM WALKER tion of 400 gills (25 blue crabs) revealed that 83.06% of the barnacle population occurred on the efferent side and 16.94% on the afferent side. The barnacle popu- lation on the efferent side of the gills was analsed in more detail. The mean number of 0. initllcri H,, n-2 . of 25 crabs and the proportion ns was assessed on each of the eight pairs of gills P* : '- ;^ « calculated for each gill pair. Analyses of variances was carried out on the ratios pi . . .ps for the 25 crabs. The results presented in Table I show that the variation between the mean proportionate infestation rate on the eight different pairs of gills exceeded significantly the variation between replicates (F = 147, P < 0.001). It •03- MEAN RATIO •02- •01- 0 _T -I- 1 2 posterior branchial chamber 4 GILL 5 PAIRS 8 anterior branchial chamber f No. of barnacles on gill pair "] r IGURE ^. Histogram showing the mean ratios of barnacles Total No. barnacles on crab J on the different gills of C. sapidus. level of significance. The vertical bars indicate the fiducial limits at the 95% BARNACLE ON CRAB GILLS 681 TABLE 1 1 List of crabs on which Octolasmis miilleri has been found. Crab Reference Habitat (from Williams, 1965) *Arenaeus ctibrarius Lamarck Calappa flammea Herbst Callinectes ornatus Ordway * Callinectes sapidus Rathbuii *Hef>atus epheliticus I. Libinia dubia H. Milne Edwards * Libinia einarginnta Leach *Menippc mcrccnaria Say *0valipes ocellatus Herbst Panopeus herbslii H. Milne Edwards *Persephona punctata aquilonaris Rathbun *Portunus spinimanus Latreille Uca minax Le Conte Humes, 1941. DeTurk, 1940; Humes, 1941. Coker, 1902; Pilsbry, 1907; Scarff, 1966; Scrocco & Fabianek, 1969. DeTurk, 1940; Humes, 1941. DeTurk, 1940. DeTurk, 1940; Coker, 1902. Coker, 1902; Humes, 1941; Pearse, 1947. Humes, 1941 DeTurk, 1940. Pearse, 1936. Ocean. Close to shore. Seldom enters estuaries. Shore — 37 fathoms. Ocean. Surface — 40 fathoms. Coastal. Often estuaries, sometimes to fresh water. Surface — 40 fathoms. Shallow ocean. Estuaries to fresh water. Surface — 20 fathoms. Ocean, Sometimes cstuario. 2-25 fathoms. Ocean. Shore — 25 fathoms. Ocean. Shore — 27 fathoms, occasionally 68 fathoms. Ocean and estuary. Surface — 28 fathoms. Ocean, close to shore. Surface — 18 fathoms. Coastal. Often estuaries. Surface — 12 fathoms. Ocean. 2-30 fathoms. Ocean. Surface — 50 fathoms. Marshes. Some distance from high salinity water. * Crabs examined in the present study. should be noted that the procedure adopted eliminated the influence of differences in individual infestation rate between crabs. Comparison between the mean ratios of the barnacles on the gills is shown graphically in Figure 2. It can be seen that the largest number of barnacles is found attached to gill 3, the numbers decreasing progressively on either side of gill 3. 682 GRAHAM WALKER , 4 , O-5mm. . BARNACLE ON CRAB GILLS 083 The efferent zone was split up into several regions (Fig. Ib) and the number of barnacles in these regions counted. Out of a total of 2,953 barnacles counted 1270 (43%) were found in the region between knob e and the efferent vessel, 701 (24%) were associated with spine d, 497 (17%) were associated with knob e, 344 (12%) were found in the region between knob e and spine d and 141 (5%) on the efferent vessel itself. Cyprids enter the branchial chamber in the inhalant respiratory current of the blue crab. They pass into the hypobranchial side of the branchial chamber. Since the gills are closely opposed to one another the cyprids are unable to pass through to the hyperbranchial side; the gills are therefore a filter to the cyprids. The cyprids must be able to adhere temporarily to the gills whilst searching for a suitable site of attachment, for periodically the respiratory current is reversed by the crab. O. miilleri not only settles on a wide variety of crabs (see Table II), but also on different sized crabs within each species (different age groups). In all juvenile crabs moulting is frequent and barnacles which have settled will be shed with the exuvia before reaching maturity; once outside of the host O. miilleri is thought to perish. The successful life cycle of O. miilleri therefore takes place either in crabs which will no longer moult i.e. mated female blue crabs, or on mature crabs which although able to moult still go through prolonged intermoult periods. A large barnacle population on any one crab is composed of individuals ranging from recently settled larvae through to mature adults, Figure 3, showing that the barnacle has a long settling season in comparison with its life span between metamorphosis and maturity. In addition to withstanding reversal of the respiratory current the cyprids must also survive the cleaning action of the epipodites of the second and third maxillipeds (see Pyle and Cronin, 1950). These epipodites have a fringing mass of setae which have back-curved hooks close to their tips (Fig. 4). The effectiveness of these cleaning appendages on exposed surfaces is reflected in the small numbers of O. miilleri which settle on the efferent vessel itself. Settlement on this vessel is restricted to the extreme tip and base of the gill, the areas not successfully cleaned by the epipodites. Judging by the high level of barnacle settlement another part of the platelets perhaps not effectively cleaned is that between the efferent vessel and knob e. When only a few barnacles were found in a crab they were always settled on the gills. However when a high infestation was found the barnacles were also settled on the wall of the branchial chamber, cleaning appendages and even on the outside of the crab on the basal parts of the maxillipeds. O. miilleri was also FIGURE 3. An excised gill of C. sapidus showing a large number of O. miilleri on the efferent side. The barnacles range from mature individuals (mat.) to small recently settled individuals (r.s.) ; ev, efferent vessel. FIGURE 4. The tips of some setae from the epipodite of the third maxilliped of C. sapidus, showing the back-curved hooks (arrows). FIGURE 5. Transverse sections through the peduncles of (a) Lcpas antifera and (b) Octolasmis miilleri. In Lcpas the outer cuticle (cut.) is thick and underlying longitudinal muscles extensive (1 m.) ; cement cells (c.) are also present in the section. In Octolasmis the outer cuticle is much thinner (cut.) and the longitudinal muscles are reduced to a single band (1 m.) ; cement cells (c.) are also present as is part of the ovary (ov.) 684 GRAHAM WALKER occasionally found settled either on the peduncle or capitulum of other settled individuals under severely crowded conditions. The movements and behavior of the cyprids in the branchial chambers prior to settlement are not known, hut since some site selection occurs active searching probably takes place. Such behavior is common in cirripedes at settlement (see Crisp, 1974). Although the majority of barnacles are found on the efferent side of the gills of blue crabs some barnacles were found on the afferent side and had presumably been taken into the hyperbranchial side of the branchial chamber as cyprids during the reversal of respiratory current by the crab. This reversal of respiratory cur- rent helps cleanse the branchial chamber of debris which has accumulated during the normal respiratory flow. It is possible that as the barnacle population on the efferent side of the gills increases, the cleaning appendages are unable to work effectively and debris accumulates. The host, therefore, reverses the respiratory current more frequently in order to remove the accumulated debris and in doing this takes more cyprids into the hyperbranchial side of the branchial chamber. It was interesting to find that all 0. iniilleri were settled on the afferent side of the gills in the branchial chambers of the crab Oralipes occllatns. The habits of O. occllatns are different from those of C. sap id its in that it spends a larger proportion of time buried in sand. The normal respiratory flow of O. occllatns is equivalent to the reversed respiratory flow of C. sapid us (see Pearse, Hunim and Wharton, 1942). Cyprids finally cement themselves to the gills a little way in from the margin of the platelets, both antennules being cemented to the same platelet by secretion from the cement glands within the cypris body (see Walker, 1971). There is apparently no preference shown by the cyprids for the upper or lower side of the platelets at settlement. Following permanent attachment the antennules are flexed thus drawing the body of the cyprid down onto the edges of the platelets. Metamorphosis then follows and the young adult 0. iniilleri emerges. The adult cement apparatus develops (see "Walker, 1973) and as the young barnacle grows "adult" cement is laid down. This will initially fill between the two platelets where the cyprid settled, but later as more cement is produced it flows over and between neighboring platelets. The adult stalk and cement apparatus of the sheltered O. iniilleri can be compared with that of the related exposed lepad. Lepas anatifera (Lacombe and Liguori, 1969). Differences in structure are revealed in traverse sections of the peduncles of these two barnacles (Fig. 5a, 5b). The peduncular cuticle of 0. iniilleri is much thinner than that of L. anatifera and the underlying musculature is also much reduced. In L. anatifera all the cement cells are aggregated close to the capitulum-peduncle junction and are interconnected by secondary cement ducts. These secondary ducts, which are not lined with chitin, join together eventually forming the two primary ducts (chitin lined), which lead down the peduncle and open out at the ends of the antennules. In O. tn^lleri the cement cells are found throughout the length of the peduncle \c.f. Pollicipcs (Koehler, 1889)], although the majority of the cells are found in the upper peduncular region. The two primary ducts extend up the peduncle for about two-thirds of its length BARNACLE ON CRAB GILLS 685 FIGURE 6. A cement cell of O. miillcri showing the connection with a collecting duct (d.) The dark mass is secretion which is aggregated around an intracellular canal; gr., granules; Azan. FIGURE 7. Oblique muscles (arrows) between the cement cells of Lcpas anutifcra; Diazo- coupling for tyrosine. and lead into the secondary ducts which connect with the cells (collecting ducts). The cement cells of 0. mi'tlleri are about 50 p. in diameter in an animal possess- ing a peduncle 2 mm long. Each cell contains a highly lobulated nucleus with a single nucleolus. There is an intracellular canal within each cell and this leads into the collecting duct. The secretion within the cell is aggregated close to the intracellular canal (Fig. 6) ; histochemical tests on this secretion confirm those of an earlier study on barnacle cement (Walker, 1970) which show it to be protein- aceous. Also in the cytoplasm there are distinct granules, I/A in diameter, which stain intensely red with azan. Similar granules have been found in the cement cells of Chchnibia tcstudhmria, the turtle barnacle, and are thought to be lipo- fuscin granules (Walker, unpublished). There are in some cement cells of 0. mi'tlleri clear vacuoles, up to 5 //. in diameter. The cement cells of L. anatifera are rounded in shape, 56-112 ^ in diameter (in a barnacle with a peduncle length of 4 mm), with spherical nuclei each containing 8-10 nucleoli. There is an ex- tensive intracellular canal within each cell and this leads into a collecting duct. The region of the cytoplasm surrounding the intracellular canal is filled with secretion which as for O. iniiHcri reacts positively for proteins. Between the 686 GRAHAM WALKER cement cells of L. anatifera there are many oblique muscle fibers (Fig. 7), which may aid the extrusion of secretion from the cells (see Lacombe, 1970). DISCUSSION 0. miilleri has been found on crabs which live in high salinity water, the crabs being either true oceanic crabs or spending part of their life cycle in the ocean. Pearse (1936), however, found an exception to this in Uca minax a crab which inhabits low salinity marshes. In the case of C. sapidus, the most common host of O. miilleri in Beaufort Inlet, the larval stages are spent in the ocean. As young crabs they migrate into low salinity water of estuaries, where they reach maturity after 1 year. Blue crabs mate in low salinity water, the mating taking place when the female is moulting for the final time (Judy and Dudley, 1970). The mated females then migrate back to high salinity water, where they spawn and where the larvae are released. Most males remain in the low salinity areas, only a few migrating back to the ocean with the females (Judy and Dudley, 1970). After the first spawn and hatch the females may re-enter the mouth of the estuary and spawn again, returning to the ocean for the second hatch. Finally, after spawning perhaps for a third time the females are thought to die in the ocean. Blue crabs rarely survive longer than 1 year after reaching maturity (Tagatz. 1968). In trawls taken at Core Creek, in the Newport River upstream of the Beaufort Inlet where the salinity is 30%e near the river bed only a single barnacle infested crab was taken, while at Adams Creek where the salinity close to the river bed was \6%c, 0. miilleri was absent from all the blue crabs caught. These preliminary field observations help confirm Scarff's (1966) observations in the laboratory of the salinity tolerance of adult O. miilleri. She found that in water of salinity 20-40?fo the barnacles, which were left attached to pieces of excised gill, had open valves with the cirri extended. At \Sc/,f the valves were closed, but when the barnacles were returned to high salinity water 80% recovered after 24 hrs. At Win valves were tightly closed and only 35% recovered after 24 hrs. At $%c the valves were tightly closed and none recovered. Salinity, therefore, appears to be a major factor limiting the distribution of 0. miilleri. DeTurk (1940) believed that the occurrence of O. miilleri was associated with the accumulation of debris in the branchial chambers of crabs, older blue crabs with large amounts of sand and other debris harboring the largest number of barnacles. He compared the infestation of dirty-gilled crabs, L. cmarginata, L. ditbia and C. sapid us with clean-gilled crabs, O. ocellatus, P. spinimanus and C. ornatus (see Table II), where few 0. miilleri occur. O. miilleri settles on most crabs close in- shore (Pilsbry, 1907) and therefore it is said to be a shallow water species (De- Turk, 1940; see Newman, 1967). Settlement is not host-specific and will take place on crabs of all age groups and sizes. However since juvenile crabs moult regularly, extruding the settled barnacles with the moult, the chance of observing O. miilleri on juvenile crabs is much less than on mature crabs. Although 0. miilleri was found on many different crabs in Beaufort Inlet, C. sapidus is the most important host because it is the most common crab in this area (Judy and Dudley, 1971). BARNACLE ON CRAB GILLS 687 The respiratory current of the spiny lobster, Pucniliis sewelli, was shown by Dinamani (1964) to affect the orientation of the cyprids of Octolasmis stella dur- ing initial settlement. They settled in such a position that following metamorphosis the young barnacle's cirral net was facing into the current. Specimens of O. milllcri were also found to be orientated to the respiratory current of the blue crab. Dinamani (1964) further explored the variation in certain features of the capitular valves of O. stella; he noted that the barnacles close to the outside of the branchial chambers of the lobster had armoured valves, whilst those barnacles better protected further in the branchial chamber had thinner, more delicate valves. Such differences due to situation were sought but not detected in O. miilleri. In the relationship between the host crab and the epizoic barnacle most ad- vantages lie with the barnacle. Once established in a branchial chamber it is pro- tected, such protection allowing the reduction of capitular shell peduncular cuticle and musculature (cf. Lepas anatifcra). A further advantage of being positioned in a branchial chamber is that there is a constant food supply brought in as sus- pended material in the crab's respiratory current. The presence of barnacles on the gills probably does not have any detrimental effects on the crab. However, bryozoans and nemerteans are regularly associated with barnacle infestations and all these epizoites would contribute greatly to the accumulation of debris on the gills and would consequently decrease the efficiency of the respiratory processes by im- pairing gill movements, reducing the amount of exposed gill surface and removing oxygen for their own needs. The crabs main protection is ecdysis. After breeding the female blue crab has fulfilled her role in propagating the species and survival becomes less important. Moulting is terminated and she gradually becomes debilitated under epizoite attack. I wish to thank Dr. J. D. Costlow and his staff for their hospitality and en- couragement during my stay at Beaufort, Prof. D. J. Crisp, Dr. LI. D. Gruffydd and Dr. J. A. Xott for reading and criticizing the manuscript, Mr. D. C. Williams for photography and the Office of Naval Research (Grant No. NR 104-194) for their financial support. I should also like to thank Dr. J. Miller for the photo- graphic work involved with Figure 3. SUMMARY 1. O. miilleri was present on the gills of most crab species in Beaufort Inlet but not on C. sapidus further upriver indicating that salinity is probably a factor controlling the incidence of the barnacle. 2. The distribution of the barnacle on the individual gills of C. sapidus has been analyzed and the factors affecting this distribution discussed. The main factors are the cleaning action of the epipodites and the respiratory flow of the crab. 3. The barnacle settlement stage larva (cyprid) attaches to blue crab gills a short distance in from the gill margin. The orientation of the larva at settlement 688 GRAHAM WALKER is a response to the respiratory flow of the crab resulting in the cirral net of the young barnacle facing into the current. 4. The cement apparatus and internal stalk structures of 0. miiUcri and Lepas anatifera are compared. LITERATURE CITED CARGO, D. G., 1958. The migration of adult female blue crabs, Callinectes sapidus Rathbun, in Chincoteague Bay and adjacent waters. J. Mar. Res., 16(3) : 180-191. COKER, R. E., 1902. Notes on a species of barnacle (Dichelaspis) parasitic on the gills of edible crabs. US. Fish Comm. Bull. 1901, 21 : 399-412. CRISP, D. J., 1974. Factors influencing the settlement of marine invertebrate larvae. Pages 177-265 in P.T. Grant, Ed., Chemoreception in Marine Organisms. London, Academic Press. COSTLOW, J. D., JR. AND C. G. BOOKHOUT, 1959. The larval development of Callinectes sapidus Rathbun reared in the laboratory. Biol. Bull., 116: 373-396. DE TURK, W. E., 1940. The parasites and commersals of some crabs of Beaufort, North Carolina. Ph.D. thesis, Duke University. DINAMANI, P., 1964. Variation in form, orientation and mode of attachment of the cirriped, Octolasmis miillcri, symbiotic on the gills of lobster. /. Aniin. EcoL, 33: 357-362. DUDLEY, D. L. AND M. H. JUDY, 1971. Occurrence of larval, juvenile and mature crabs in the vicinity of Beaufort Inlet, North Carolina. National Oceanic and Atmospheric Administration Technical Report, NMFS SSRF-637. [Available from NOAA Pub- lications Section, Rockville, Maryland 20852.] HUMES, A. G., 1941. Notes on Octalasmis miillcri (Coker), a barnacle commensal on crabs. Trans. Amcr. A'licroscop. Soc., 60(1) : 101-103. JUDY, M. H. AND D. L. DUDLEY, 1970. Movements of tagged blue crabs in North Carolina waters. Comm. Fish. Rev., 32(11) : 29-35. KOEHLER, R., 1889. Reserches sur 1'organization des Cirripedes. Arch. Biol., 9: 311-402. LACOMBE, D., 1970. A comparative study of the cement glands in some balanid barnacles (Cirripedia, Balanidae). Biol. Bull., 139: 164-179. LACOMBE, D. AND V. LIGUORI, 1969. Comparative histological studies of the cement apparatus of Lepas anatifera and Balanus tintinnabulum. Biol. Bull., 137: 170-180. NEWMAN, W. A., 1967. Shallow water versus deep-sea. Octolasmis (Cirripedia, Thoracica). Crustaceana, 12: 13-32. PEARSE, A. S., 1936. Estuarine animals at Beaufort, North Carolina. /. Elisha Mitchell Sci. Soc., 52(2) : 174-222. PEARSE, A. S., 1947. On the occurrence of ectoconsortes on marine animals at Beaufort, N. C. /. ParasitoL, 33 : 453-458. PEARSE, A. S., H. J. HUMM, AND G. W. WHARTON, 1942. Ecology of sand beaches at Beaufort, N. C. EcoL Monogr., 12(2) : 135-190. PILSBRY, H. A., 1907. The barnacles (Cirripedia) contained in the collections of the U.S. National Museum. U.S. Nat. Mus. Bull, 60: 1-114. PYLE, R. AND E. CRONIN, 1950. The general anatomy of the blue crab Callinectes sapidus Rathbun. Chesapeake Biol. Lab., Solomon's Island Maryland Publication, No. 87: 1- 40. SCARFF, J. M., 1966. A study of the commensal Octolasmis miillcri. Unpublished project, available from Duke Marine Laboratory, Beaufort, North Carolina. SCROCCO, V. M. AND J. FABIANEK, 1969. Symbiosis of Callinectes sapidus Rathbun with Carcinonemcrtcs Bryozoans and barnacles Fed. Proc., 28(2) : 526. TAGATZ, M. E., 1968. Biology of the blue crab, Callinectes sapidus Rathbun, in the St. John's River, Florida, US. Fish Wildlife Service Fish. Bull., 67(1) : 17-33. VAN ENGEL, W. A., 1958. The blue crab and its fishery in Chesapeake Bay. Part I. Repro- duction, early development, growth and migration. Comm. Fish. Rev., 20(6) : 6-17. BARNACLE ON CRAB GILLS 689 WALKER, G., 1970. The histology, histochemistry and ultrastructure of the cement apparatus of three adult sessile barnacles, Elminius modcstus, Balanus balanoides and Balanus hameri. Mar. Biol., 7(3) : 239-248. WALKER, G., 1971. A study of the cement apparatus of the cypris larva of the barnacle Balanus balanoides. Mar. Biol., 9(3) : 205-212. WALKER, G., 1973. The early development of the cement apparatus in the barnacle, Balanus balanoides (L.) (Crustacea: Cirripedia). /. Exp. Mar. Biol. Ecol., 12: 305-314. WILLIAMS, A. B., 1965. Marine decapod crustaceans of the Carolinas. Fish. Bull., 65(1) : 1-298. INDEX Absorption, squid axon, change in, 496 Accessory cell of Hydrolimax oocytes, 618 Acetylcholinesterase, effect on frog synaptic transmission, 494 Actin components of squid axon, 491 filaments, bipolar polymerization of, 503 Actinian behavior, 630 Action potential, change in squid axon absorp- tion during, 496 Adaptation in rods from the toad retina, 475 to burrowing in Ovalipcs, 550 Adenosine-3'-phosphate-5'-phosphosulfate in sulfate reducers, 502 Adenosine-5'-phosphosulfate in sulfate re- ducers, 502 Acdcs aegypti, dietary factors stimulating oogenesis in, 433 Aequorin study of squid suppression potential, 489 Aerial oxygen consumption in Ligumia, 560 AI.DRICH, J. C. Allometric studies on energy relationships in the spider crab Libinia cmarginata (Leach), 257 Algae, planktonic marine, harvesting of, 136 Algal endosymbionts of Hydra, growth and distribution of, 105 Allometry of energy relationships in Libinia, 257 ' Amino acid, uptake by toadfish liver, 479 toadfish liver, computer program for, 493 Amoeba cytoplasm, contractility of, 501 Anemone behavior, modification of, 630, 641 Annual Report of the Marine Biological Laboratory, 1 Anoxic stress in Li, in Ligumia, 560 Arbacia, eggs and embryos, effect of tempera- ture on, 484 sperm fertilization antigen, 490 ARNOLD, J. M., AND L. D. WILLIAMS-ARNOLD. Effects of a Ca++Mg++ ionophore (A23187) on cytokinesis in the squid, 466 AND R. E. YOUNG. Ultrastructure of a cephalopocl photophore. I. Structure of the photogenic tissue, 507 R. E. YOUNG AND M. V. KING. Ultrastruc- ture of cephalopod photophore. II. Irido- phores as reflectors and transmitters, 522 Ascidians, water transport in, 608 colony specificity in, 411 Astcrias rulyaris, gonad, morphology and histology of, 661 ATHERTON, W. See D. F. BABIEC, 652 ATLAS, S. J., J. D. JOHNSON, U. S. MAITRA AND R. A. WALLACE. Membrane-associated carbohydrate-containing macromolecules of Arbacia punctulata, 466 ATP in z'ivo, effect of luminal application, 476 Attachment of barnacles on crab gills, 678 Aurelia, genome sequence organization in, 481 Autoradiography of the plasma membrane of sea urchin eggs, 474 Axonal transport in sonic motor nerve of the toadfish, 483 Axoplasm, squid, filamentous network of, 491 BABIEC, D. F. See I. VALIELA, 652 BAGCHI, M., C. HARDING AND H. JAM PEL. Wound healing in the dogfish ocular lens, 467 See C. HARDING, 479 BARBER, M. L., AND M. EDIDIN. Changes in surface antigens and surface enzymes at fertilization in sea urchin eggs, 467 Barbulanympha, mitosis and morphogenesis in, 484 BARKALOW, D. See A. FARMANFARMAIAN, 476 BARKER, J. L. See P. N. HOFFMAN, 483 BARKER, L. A., M. J. DOWDALL, G. R. VICKERS AND T. W. MITTAG. High-affinity choline transport : uptake and metabolism of choline and pyrrol choline by synaptosomes from the optic lobe of squid (Loli(>o pcalci), 468 Barnacle lateral ocellus, photoreceptor sensi- tivity of, 491 Barnacles, pedunculate, on crab gills, 678 BARNETT, T., AND R. T. KOVACIC. Interspersion of unique sequence DNA with repetitive DNA elements in the genome of Ccrc- bratuhis lactcus, 468 BARRETT, D. See G. W. LOPEZ, 489 Bascodiscus delineates, gut morphology of, 352 BEALE, S. See J. RAMUS, 494 690 INDEX 691 BEGENISICH, T., AND M. CAHALAN. Internal potassium ions modify sodium channel selectivity, 469 Behavior, anemone, modification of, 630, 641 predatory, among the naticid gastropods, 469 BENNETT, M. V. L. See W. CLUSIN, 472 See S. M. HIGHSTEIN, 482 See J. S. KEETER, 485 See H. KORN, 486 BERG, C. J., JR., AND M. E. PORTER. A com- parison of predatory behavior among the naticid gastropods Lunatia lieros, Lunatia triseriata and Poliniccs duplicatus, 469 BlNGHAM, S. E., J. A. SCHIFF AND J. KAMI'S. Inhibition by molybdate of Prasiola cell attachment and completion of the Chlo- rclla life cycle, 470 Bioluminescence, in cell free extracts of the scale worm, 480 in Obelia, cellular origin of, 397 Biphasic response in sea urchins, 236 Bipolar polymerization of actin filaments, 503 BITTNER, G. D. See M. P. CHARLTON, 471 BLITZ, A. See A. POLITOFF, 494 BLOOM, J. W. See C. K. GOVIND, 478 BLOOMGARDEN, D. See G. WEISSMANN, 503 BODE, H. See S. SMITH, 186 Body fluid composition in Ligumia, 560 BORGESE, T. A., AND R. CARTicA. Differential labelling of Fundulus hemoglobins using different hemoglobin precursors, 470 BOYER, J. F. Clinal and size-dependent varia- tion at the LAP locus in l\fytiliis cdulis. 535 BREDEN, E. N. See C. A. PRICE, 136 BROWN, J. E. See J. A. CODES, 473 Bryozoan, gene frequencies in, 498 Buccal muscles of the whelk, 369 Budding in Symplcgina, 213 Bullfrog cortex, hypersynchrony in, 492 BURKETT, B. N., AND H. A. SCHNEIDERMAN. Roles of oxygen and carbon dioxide in the control of spiracular function in Cecropia pupae, 274 Discontinuous respiration in insects at low temperatures : intratracheal pressure changes and spiracular valve behavior, 294 Burrowing in Oralipcs, 550 BURZIO, L. See S. S. KOIDE, 486 Busy con canaliculatum } innervation of radular protractor muscle of, 369 CAHALAN, M. See T. BEGENISICH, 469 CAINE, E. A. Feeding of Ovalipes guadul- pensis (Saussure) (Decapoda: Brach- yura : Portunidae), and morphological adaptations to a burrowing existence, 550 Calcium, and desensitization of Limulns ven- tral photoreceptors, 476 current activation of a voltage-insensitive conductance, 472 Callincctcs sapidus, gills, barnacles on, 678 Calorific values in Platyhelminthes, 81 CALOW, P., AND J. B. JENNINGS. Calorific values in the phylum Platyhelminthes : the relationship between potential energy, mode of life and the evolution of ento- parasitism, 81 Carbobenzoxyglutamate, effect on sea urchin hatching enzyme, 489 Carbohydrate-containing macromolecules of sea urchin eggs, 466 Carbon dioxide, effect on spiracles of Cecropia pupae, 274 CARTICA, R. See T. A. BORGESE, 470 CASSEX, E. I. See R. C. RUSTAD, 497 Catfish oocytes, vitellogenic protein in the serum of, 481 Cations, requirements for activation, 471 effects of on ouabain inhibition rate, 482 Cecropia pupae, spiracular function in, 274 Cell growth inhibition by tryptophan photo- products, 504 Cellular maturation, relation to variation in cornea! epithelial cell surfaces, 479 Cellular origin of hydroid bioluminescence, 397 Cement apparatus of Octolasmis, 678 Centrifugation harvesting of planktonic marine algae, 136 Cephalopod photophore ultrastructure, 507, 522 Ccrcbratulits lacteus genome, unique sequence DNA in, 468 Chaetognath, egg-laying in, 247 Chaetoptcrus larvae, light production, 477 CHAMBERS, E. L. Effects of ionophores on marine eggs and cation requirements for activation, 471 CHARLTON, M. P., AND G. D. BITTNER. Fa- cilitation of transmitter release at the squid giant synapse, 471 CHENG, T. C., AND G. E. RODRICK. Identifica- tion and characterization of lysozyrne from the hemolymph of the soft-shelled clam, -l/vu arcnaria, 311 Chlorclla, completion of life cycle, 470 Choline uptake from squid optic lobe, 468 Ciona intcstinalis, water transport rates of, 608 Clam, soft-shelled, characterization of lysozyrne from hemolymph of, 311 Clinical variation at the LAP locus in Mytilus, 535 CLUSIN, W., D. SPRAY AND M. V. L. BEN- NETT. Activation of a voltage-insensitive conductance by inward calcium current, 472 692 INDEX Cnidarian behavior, 630, 641 CODES, J. A., AND J. E. BROWN. Effects of increased intracellular pH-buffering ca- pacity on the light response of Limulus ventral photoreceptor, 473 Coelenterate behavior, 630, 641 Coelenterate genome sequence organization, 481 COHEN, C. See G. WEISSMANN, 503 COHEN, I. S., AND W. G. VAN DER KLOOT. The effects of strontium on the timing of miniature end-plate potentials, 472 COHEN, L. B. See W. N. Ross, 496 COLLINS, T. See G. WEISSMANN, 503 Colony specificity in ascidians, 411 Computer program for toadfish liver amino acids, 493 Conditioned response in sea anemones, 641 Conductance, voltage-insensitive, activation of, 472 Conduction velocity of lobster giant axons, 478 Contractility of amoeba cytoplasm, 501 Copper, effect on sea urchin motility, 236 Cornea, dogfish, healing in, 467 epithelial cell surfaces, variation in, 479 Cortex, bullfrog, hypersynchrony in, 492 Countershading of a cephalopod photophore, 507, 522 Crab, spider, energy relationships in, 257 gills, barnacles on, 678 fiddler, feeding in, 652 GRAIN, W. See G. P. MOORE, 491 Crassostrea genome organization, 478 Crepidula, spermiogenesis, 480 CULLINEY, J. L. Larval development of the giant scallop Placopectcn magellanicus (Gmelin), 321 Culture of a colonial ascidian, 213 Current-induced flow through Halichondria, 443 Cytochalasin B, establishment of as polyspermy block in Spisula eggs, 504 Cytokinesis in squid embryos, 466 Cytoplasmic DNA in sea urchin oogenesis, 586 Cytoplasmic process in turbellarian oocytes, 618 D2O effect on sodium efflux from squid axon, 487 DAVILA, H. V. See W. N. Ross, 496 DE WEEK, P. Axoplasmic free magnesium levels, and magnesium ion transport in squid giant axon, 473 Decomposition of Fucus, 484 Degeneration in Symplcgma, 213 Dehydration in Ligumia, 560 Density-gradient-centrifugation with marine algae, 136 DESCHENES, M. See J. S. KEETER, 485 Desensitization of Limulus ventral photo- receptors, 476 Detritus based marine food web, meiofauna in, 488 Development, adult, of Drosophila, 119 D-glucose, conversion of to L-ascorbic acid, 497 Diapause in marine sponge gemmules, 333 Diatoms, conversion of D-glucose to L-ascor- bic acid in, 497 Dietary factors stimulating Acdcs oogenesis, 433 DIETZ, T. H. Body fluid composition and aerial oxygen consumption in the fresh- water mussel, Ligumia subrostrata (Say) : effects of dehydration and anoxic stress, 560 Digestion enzymes in nemerteans, 352 Distribution, vertical, of Pclogobia, 457 of algal endosymbionts of Hydra, 105 of barnacles on crab gills, 678 of gastropod molluscs in a kelp forest, 386 Divalent cations association with squid axon, 493 DNA, Spisula, sequence elements in, 491 unique sequence in Cerebratulns genome, 468 Dogfish, cells, lysosomes of, 503 cornea, healing in, 467 lens formation from soluble crystallins, 478 DONALDSON, S. Larval settlement of a sym- biotic hydroid : specificity and nematocyst responses in planulae of Proboscidactyla flavicirrata, 573 DOWDALL, M. J. See L. A. BARKER, 468 Drosophila, adult development of, 119 DUNBAR, B. S., J. JOHNSON AND D. EPEL. ^I lodination and autoradiography of the plasma membrane of Arbacia punctulata eggs, 474 See J. D. JOHNSON, 485 E EAGELS, D. A. Reflexes in the tailspine sys- tem of the horseshoe crab, Limulus poly- phemus, 474 EBERHARD, A. Anaerobic synthesis of bac- terial luciferase, 475 Ecdysis in shrimp, 203 Ecdysone action in flies, 163 ECKBERG, W. R., AND C. B. METZ. Effects of univalent (Fab) antibody fragments on the fertilizing capacity of sperm of the sea urchin, Arbacia punctulata, 475 Ecology of littoral and sublittoral snails, 490 EDIDIN, M. See M. L. BARBER, 467 INDEX 693 EDTA, effect on sea urchin motility, 236 Egg-laying in Sagitta, 247 Eggs, marine, effects of ionophores on, 471 sea urchin, macromolecules of, 466 sea urchin, surface antigens and enzymes in, 467 and embryos, Arbacia, effect of temperature on, 484 Electrotonic synapse, fine structure of, 485 Electrotonic transmission and coupling of fish neurons, 486 Elongation factor 1 of toadfish liver, 492 End-plate potentials, effects of strontium on the timing of, 472 in frog neuromuscular junction, 487 Energy, potential, in Platyhelminthes, 81 relationships in LibiniaJ 257 transfer in squid axons, 501 ENESCO, H. E., AND K. H. MAN. Cytoplasmic DNA in sea urchin oogenesis studied by 3H-actinomycin D binding and radio- autography, 586 Entoparasitism in Platyhelminthes, evolution of, 81 Enzymes, digestive, in nemerteans, 352 introduction into lysosomes of dogfish cells, 503 surface, in sea urchin eggs, 467 EPEL, D. See B. S. DUNBAR, 474 See J. D. JOHNSON, 485 See C. A. ZIOMEK, 504 Epineural muscles in the whelk, 369 Euapta lap pa, locomotion of, 95 Expansion in sea anemones, 641 F FAIN, G. L. Adaptation and spatial summa- tion in rods from the toad retina, 475 FARMANFARMAIAN, A., AND D. BARKALOW. Glycine transport in the intestine of the toadfish — effect of luminal application of ATP in vivo, 476 Feeding, in fiddler crabs, 652 of Ovalipcs, 550 FEIN, A., AND J. LISMAN. Intracellular Ca++ injection produces localized densensitiza- tion of Limulus ventral photoreceptors, 476 FELL, P. E. Diapause in the gemmules of the marine sponge, Haliclona loosanoffi, with a note on the gemmules of Haliclona ocu- lata, 333 Fertilization, antigen in Arbacia sperm, 490 in sea urchin eggs, 467 of Spisula eggs, 504 Fertilization-associated changes in sea urchin eggs, 485 Fertilizing capacity of sperm of the sea urchin, 475 Fine structure of electrotonic synapse, 485 Fish, protein metabolism and growth in, 500 variation in corneal epithelial cell surfaces, 479 nuerons, electrotonic transmission and cou- pling, 486 Files, hormonal factors in, 163 Fluorescent probes of squid energy transfer, 501 Food stimuli for oogenesis in Acdcs, 433 Food-resource partitioning in Pcctinaria, 227 FRAENKEL, G. See P. SIVASUBRAMANIAN, 136 FREEMAN, G., G. T. REYNOLDS AND A. WAL- TON. The development of light production in the annelid Chaetopterus peymncn- taccus, 477 FRIEDMAN, S. See P. SIVASUBRAMANIAN, 163 Frog, leopard, photoreceptors in, 503 neuromuscular junction, evoked responses in, 487 neuromuscular junction, miniature end- plate potential timing, 472 sartorius, synaptic transmission of, 494 sartorius neuromuscular junction, 495 Fucus, decomposition of, 484 Fimdulus hemoglobins, differential labelling, 470 Fusibility between ascidian colonies, 411 Fusion, rejection and indifference, in ascidians, 411 GAINER, H. See P. N. HOFFMAN, 483 Gasterostomes, life-cycle of, 500 Gastropod distribution in a kelp forest, 386 Gastropods, comparison of predatory behavior, 469 Gene frequencies in Schisoporella, 498 Genetic variation in Mytihis, 535 Genome, Cercbratuhts, unique sequence DNA in, 468 organization in Crassostrca, 478 sequence organization in coelenterate, 481 Germination in sponges, 333 GHARIB, S. D. See R. C. RUSTAD, 497 GIBSON, R. Histochemical observations on the localization of some enzymes associated with digestion in four species of Brazilian nemerteans, 352 Glucose transport in Hymenolepis, 146 Glycine transport in the intestine of the toadfish, 475 Glycogen in entosymbiotic species, 81 GOLD, K. Studies on the size distribution of loricae and the growth status of Tin- tinnida in Eel Pond, 477 694 INDEX GOLDBERG, R. B., AND J. V. RUDERMAN. Genome organization in the American oyster, Crassostrea virginica, 478 GOLDSTEIN, M. See C. A. PRICE, 136 Gonad, Astcrias, morphology and histology of, 661 formation in Symplcgina, 213 GOTTLIEB, A. See G. WEISSMANN, 503 GOVIND, C. K., F. LANG AND J. W. BLOOM. Increase in conduction velocity of lobster giant axons during growth, 478 GRIESS, G., S. ZIGMAN AND T. YULO. Forma- tion of dogfish lens fibers from soluble crystallins, 478 See S. ZIGMAN, 504 Growth, in fish, 500 of algal endosymbionts of Hydra, 105 of lobster giant axons, increase in conduc- tion velocity, 478 stages in Libinia, 257 status of Tintinnida in Eel Pond, 477 GUILLARD, R. R. L. See C. A. PRICE, 136 See J. A. SAUNDERS, 497 Gut morphology of nemerteans, 352 H 3H-actinomycin D binding study of sea urchin oogenesis, 586 Haemal sinus in sea stars, 661 Halichondria current-induced flow through, 443 Haliclona loosanoffi, diapause in gemmules of, 333 Haliclona oadata, gemmules of, 333 HARDING, C., M. BAGCHI, S. SUSAN AND H. JAMPEL. Variation in corneal epithelial cell surfaces, as possibly related to cellu- lar maturation and senescence (S.E.M. studies), 479 See M. BAGCHI, 467 IfannotJwc. bioluminescence in cell free ex- tracts of, 480 HASCHEMEYER, A. E. V. Theoretical analysis of amino acid uptake by toadfish liver, 479 See J. B. K. NIELSEN, 492 See M. A. K. SMITH, 500 HASTINGS, J. W. See A. A. HERRERA, 480 Healing in dogfish cornea, 467 HEFFERNAN, J. M., AND S. A. WAINWRIGHT. Locomotion of the holothurian Euapta lappa and redefinition of peristalsis, 95 HEMOGLOBINS, Fundnlus, differential labelling of, 470 Hemolymph, clam, lysozyme from, 311 HENLEY, C. Spermiogenesis in Crepidula, 480 HERRERA, A. A., J. W. HASTINGS AND J. G. MORIN. Bioluminescence in cell free ex- tracts of the scale worm llannothoc (Annelida; Polynoidae), 480 HICKEY, E. D., AND R. A. WALLACE. A study of the vitellogenic protein in the serum of estrogen-treated Ictalunis ncbulosus, 481 HIGGINS, R. C. Genome sequence organiza- tion in coelenterate, Aurclia, determined by DNA reassociation kinetics, 481 HIGHSTEIN, S. M., J. KEETER AND M. V. L. BENNETT. Some aspects of transmission at synapses in the labyrinth of the toad- fish," 482 HILL, R. B., AND J. W. SANGER. Anatomy of the innervation and neuromuscular junc- tions of the radular protractor muscle of the whelk, Busycon canaliculatum (L.), 369 HILLMAN, P. See B. MINKE, 491 Histology of Astcrias gonad, 661 HOBBS, A. S. Effects of external monovalent cations on ouabain inhibition rate of so- dium pump in squid giant axon, 482 HOCKSTEIN, S. See B. MINKE, 491 HOFFMAN, P. N., J. L. BARKER, H. GAINER AND R. J. LASER. Analysis of fast axonal transport in the sonic motor nerve of the toadfish, 483 HOFFSTEIN, S. See G. WEISSMANN, 503 Hormonal factors, proteinaceous, in flies, 163 HOSKIN, F. C. G., M. L. POLLACK AND R. D. PRUSCH. Metabolism of cysteine in rela- tion to the synthesis of isethionate by squid nerve, 483 Host-specific planula settlement, 573 HOWARD, K. See J. RAMUS, 494 HUBSCHMAN, J. H. See L. A. STOFFEL, 203 HUNTER, R. D. Preliminary studies on natural decomposition of FJICIIS in two marine environments, 484 H \alophora cccropia, respiration and behavior, 274, 294 Hydra, inhibition of nematocyst discharge in, 186 Hydra attcnuata , inhibition of nematocyst dis- charge in, 186 Hydra viridis, algal endosymbionts of, 105 Hydroid cellular origin of bioluminescence in, 397' muscle activation in, 594 symbiotic, larval settlement of, 573 colonial, light produced from, 499 Hydrolimax grisca, oocytes, accessory cell and yolk halo of, 618 5-hydroxytryptamine, effect on sea urchin hatching enzyme, 489 INDEX 695 Ilyincnolcpis diminuta, glucose and sodium fluxes in, 146 Hypersynchrony in bullfrog cortex, 492 Inhibition by molybdate of Prasiola, 470 Innervation of the buccal apparatus in Busy- con, 369 INOUE, S., AND H. RITTER. Mitosis and nuclear morphogenesis in Barbulanympha, 484 lodination of the plasma membrane of sea urchin eggs, 474 Ionic stimulation, prefertilization, in sea ur- chin eggs, 497 lonophores, effect on squid cytokinesis, 466 activation of starfish oocytes, 498 effects of on marine eggs, 471 Iridophores, cephalopod, 522 JACKOWSKI, S. C, AND R. A. WALLACE. Ef- fects of subzero temperatures on eggs and embryos of Arbacia punctulata, 484 JAM PEL, H. See M. BAGCHI, 467 See C. HARDING, 479 JENNINGS, J. B. See P. CALOW, 81 JOHNSON, J. D., B. S. DUNBAR AND D. EPEL. Fertilization-associated changes in the plasma membrane proteins of Arbacia punctulata eggs, 385 See S. J. ATLAS, 466 See B. S. DUNBAR, 474 JOSEPHSON, R. K. Factors affecting muscle activation in the hydroid Tidndaria, 594 Juvenile hormone and Drosophila development, 119 K KAPLAN, R. See G. WEISSMANN, 503 KEETER, J. S., M. DESCHENES, G. D. PAPPAS AND M. V. L. BENNETT. Fine structure and permeability studies of a rectifying electrotonic synapse, 385 See S. M. HIGHSTEIN, 482 Kelp forest, distribution of molluscs in, 386 KING, M. V. See J. M. ARNOLD, 522 KOIDE, S. S., L. BURZIO AND Y. TANIGAWA. Activation of sea urchin sperm chromatin template for DNA synthesis by Ca2+, Mg2+ dependent endonuclease present in oocytes of Arbacia punchdata, 486 KORN, H., C. SOTELO, N. KOTCHABHAKDI AND M. V. L. BENNETT. Fish lateral vestibular neurons : electrotonic transmission from primary vestibular afferents, electrotonic coupling between vestibulo spinal neurons and identification of efferent cells to the labyrinth, 486 KOTCHABHAKDI, N. See H. KORN, 486 KOVACIC, R. T. See T. BARNETT, 468 KREBS, C. See I. VALIELA, 652 KRIEBEL, M. E., G. D. PAPPAS AND S. ROSE. The relationship of small mode miniature endplate potentials and quantal content of evoked responses in the stimulated frog neuromuscular junction, 487 See S. ROSE, 495 KUSTIN, K., K. V. LADD, G. C. McLEOD AND D. L. TOPPEN. Water transport rates of the tunicate Ciona infcstinalis, 608 L-ascorbic acid, conversion from D-glucose, 497 Labelling of Fundulits hemoglobins, 470 LADD, K. V. See K. KUSTIN, 608 LAXDOWNE, D., AND V. SCRUGGS. The effect of D-O on sodium efflux from the squid giant axon, 487 LANG, F. See C. K. GOVIND, 478 LAP locus in Mytilus, 535 Larval development of Placopectcu. 321 Larval settlement of Proboscidactyla, 573 LASER, R. J. See P. N. HOFFMAN, 483 LEADBETTER, E. R. See H. J. SINGER, 499 LEE, D. S., AND J. METUZALS. Anchorage of cytoplasmic filaments to intramembrane particles in the plasma membrane and in the nuclear envelope : transduction of membrane receptor topography to the genome, 488 LKE, T. J., K. TENORE, J. H. TIETJEN AND C. MASTROPAOLO. An experimental approach toward understanding the role of meio- fauna in a detritus based marine food web, 488 LESTER, B. See M. R. REEVE, 247 Leupeptin and antipain, induction of poly- spermic fertilization by, 502 Lilnma emarginata, energy relationships in, 257 Life-cycle of gasterostomes, 500 Light, produced from colonial hydroids, 499 production in Chaftoptcrus larvae, 477 response of Limnhis ventral photoreceptor, 473 Ligitmia subrostrata, body fluid composition and aerial oxygen consumption in, 560 Limb loss in freshwater shrimp, 203 Liinulus, reflexes in the tailspine system of, 474 ventral photoreceptor, desensitization of, 476 ventral photoreceptor, light response of, 473 LTSMAN, J. See A. FEIN, 476 696 INDEX Littorina, respiration and vertical zonation in, 496 Littorina spp., ecology of, 490 LLINAS, R., AND C. NICHOLSON. Aequorin study of the suppression potential in the squid giant synapse, 489 Lobster giant axons, conduction velocity, 478 Locomotion of Euapta, 95 Locomotor activity rhythms of juvenile salmon, 422 LOEWUS, F. A. See J. A. SAUNDERS, 497 Loligo pcalci optic lobe, uptake from, 468 LOPEZ, G. W., AND D. BARRETT. Effects on Arbacia hatching enzyme of 5-hydroxy- tryptamine and N-benzyloxycarbonyl-L- glutamate, 489 Loricae in Eel Pond, size distribution, 477 LOWRY, L. F., A. J. MCELROY AND J. S. PEARSE. The distribution of six species of gastropod molluscs in a California kelp forest, 386 Luciferase, bacterial, anaerobic synthesis of, 475 Lunatia hcros, predatory behavior, 469 Limatia triscriata, predatory behavior, 469 Lysosomes of dogfish cells, 503 Lysozyme from hemolymph of Mya, character- ization of, 311 M MAC-DONALD, E. D. See H. Ripps, 495 Macromolecules of sea urchin eggs, 466 Magnesium ion transport in squid axon, 473 MAITRA, U. S., AND C. B. METZ. Purification and properties of an Arbacia sperm ferti- lization antigen, 490 See S. J. ATLAS, 466 MAN, K. H. See H. E. ENESCO, 586 Manganese effect on sea urchin motility, 236 Marine environments, decomposition of Fucus in, 484 Marine food web, role of meiofauna in, 488 MASTROPAOLO, C. See J. J. LEE, 488 MAUZERALL, D. See J. RAMUS, 494 MCCLEAVE, J. D. See N. E. RICHARDSON, 422 McELROY, A. J. See L. F. LOWRY, 386 McLEOD, G. C. See K. KUSTIN, 608 McMAHON, R. F., AND W. D. RUSSELL- HUNTER. Responses to low oxygen stress in relation to the ecology of littoral and sublittoral snails, 490 See W. D. RUSSELL-HUNTER, 496 Meiofauna, role in marine food web, 488 Membrane receptor topography, transduction of, 488 MENDIOLA-MORGENTHALER, L. R. See C. A. PRICE, 136 Mercury, effect on sea urchin motility, 236 Metabolism of cysteine, synthesis of isethionate by squid axon, 483 Metal ions, heavy, effect on sea urchin motility, 236 METUZALS, J., AND W. E. MUSHYNSKI. Fila- mentous network of the axoplasm, as re- vealed by freeze-etching of the squid giant nerve fiber, in relation to actin, tubu- lin and myosin components, 491 See D. S. LEE, 488 METZ, C. B. See W. R. ECKBERG, 475 See U. S. MAITRA, 490 Microorganisms, spore-forming, isolation of, 499 MlNKE, B., S. HOCKSTEIN AND P. HlLLMAN. A photoreceptor sensitivity paradox, 491 Mitosis in Barbulanympha, 484 Mitotic delay in sea urchin eggs, 497 MITTAG, T. W. See L. A. BARKER, 468 Mode of life of Platyhelmintb.es, 81 Molluscs, distribution in a kelp forest, 386 Molt cycle in freshwater shrimp, 203 Molt-acceleration factor in freshwater shrimp, 203 Molybdate inhibition of Prasiola, 470 MOORE, G. P., AND W. GRAIN. Interspersion of repetitive and non-repetitive sequence elements in the DNA of Spisula solidis- sima, 491 MORIN, J. G., AND G. T. REYNOLDS. The cellular origin of bioluminescence in the colonial hydroid Obelia, 397 See A. A. HERRERA, 480 See J. M. SHOREY, 499 Morphogenesis in Barbulanympha, 484 Morphological difference between oozooid and blastozooid in Symplegma, 213 Morphology of Astcrias gonads, 661 MORRELL, F., AND N. TSURU. Development of spontaneous hypersynchrony in the hip- pocampal cortex of the bullfrog, Rana catcsbciana, 492 Motility, sea urchin, effect of heavy metal ions on, 236 MUKAI, H., AND H. WATANABE. On the oc- currence of colony specificity in some com- pound ascidians, 411 Muscle activation in a hydroid, 594 MUSHYNSKI, W. E. See J. METUZALS, 491 Mustelus cams, formation of lens from soluble crystallins, 478 Mya arcnaria, characterization of lysozyme from hemolymph of, 311 Myosin components of squid axon, 491 Mytilus cdulis, genetic variation in, 535 N NAGLE, D. See G. WEISSMAN, 503 NAKAUCHI, M. See K. SUGIMOTO, 213 INDEX 697 Nannoplankton, harvesting of, 136 Near UV light inhibition of cell growth, 504 NELSON, L. See L. G. YOUNG, 236 Nematocyst, discharge inhibition in hydra, 186 responses in Proboscidactyla planulae, 573 Nemertean, food reserves, 352 digestion in, 352 Neurofilamentous network in rabbit brain, 488 Neuromuscular junction of frog sartorius, 495 Neurons, fish, electrotonic transmission and coupling, 486 NEWTON, W. D. The accessory cell and yolk halo of the oocyte of the freshwater turbellarian Hydrolimax grisca (Platy- helminthes; Plagiostomidae), 618 NICHOLSON, C. See R. LLINAS, 489 NIELSEN, J. B. K., AND A. E. V. HASCHE- MEYER. Elongation factor 1 of toadfish liver, 492 Non-digestive enzymes in nemertean tissues, 352 Obelia, bioluminescence in, 397 Occurrence of barnacles on crab gills, 678 Octolasmis mnllcri, occurrence on crab gills, 678 Oocyte growth and nutrition in Hydrolimax, 618 Oocytes of Hydrolimax, 618 Oogenesis, sea urchin, cytoplasmic DNA in, 586 Oogenesis in Acdes, dietary factors stimulat- ing, 433 Oogenesis in freshwater turbellarians, 618 Optic lobe, squid, uptake from, 468 Origin, cellular, of hydroid bioluminescence, 397 Origin of germ cells in Symplcgma, 213 OSCHMAN, J. L., AND B. J. WALL. AsSOCia- tion of divalent cations with membranes of squid giant axon, 493 OSHIDA, J. See S. SMITH, 186 Ototyphlonemertcs spp., gut morphology of, 352 Ouabain inhibition rate of sodium pump in squid axon, 482 Ovalipcs guadulpcnsis, feeding of, 550 Oxygen, effect on spiracles in Cecropia pupae, 274 Oxygen stress, low, in snails, 490 Oyster genome organization, 478 P Palacmonctcs kadiakcnsis, limb loss and molt cycle in, 203 PANT, H. See I. TASAKI, 501 PAPPAS, G. D. See J. S. KEETER, 485 See M. E. KRIEBEL, 487 See S. ROSE, 495 PAPPAS, P. W. See C. P. READ, 146 PARDY, R. L. Some factors affecting the growth and distribution of the algal endo- symbionts of Hydra viridis, 105 PEARSE, J. S. See L. F. LOWRY, 386 PEARSE, V. B. Modification of sea anemone behavior by symbiotic zooxanthellae : phototaxis, 630 Modification of sea anemone behavior by symbiotic zooxanthellae : expansion and contraction, 641 Pectinaria gouldii, food-resource partitioning in, 227 Pclogobia longicirrata, vertical distribution and reproductive biology of, 457 Peristalsis in Euapta, 95 Permeability of electrotonic synapse, 485 PERSELL, R. A computer program for evalua- tion of amino acid distribution in vivo, 493 Phasic muscle in the whelk, 369 pH-buffering capacity on the light response of Limn Ins ventral photoreceptor, 473 Photobactcrium fischcri, synthesis of luci- ferase, 475 Photocytes, colonial hydroids, light from, 499 Photophore, cephalopod, ultrastructure of, 507, 522 Photoreceptor sensitivity of barnacle lateral ocellus, 491 Photoreceptors, Limuhis ventral, desensitiza- tion of, 476 skate, synaptic vesicles in, 495 skin, in the leopard frog, 503 Photosynthetic capacity of algae, 494 Phototaxis in sea anemones, 630 Physiological capacity in spider crabs, 257 Phytoplankton, harvesting of, 136 Pigment content of algae, 494 Placopecten magellanicus, larval development of, 321 Plasma membrane of sea urchin eggs, 474 Platyhelminthes, calorific values in, 81 Poliniccs duplicatus, predatory behavior, 469 POLITOFF, A., A. BLITZ AND S. ROSE. Exo- genous acetylcholinesterase : effects on synaptic transmission at the neuromuscu- lar junction of the frog sartorius, 494 POLLACK, M. L. See F. C. G. HOSKIN, 483 POLLARD, T. D. See D. T. WOODRUM, 503 Polychaete, deposit feeding, food-resource partitioning in, 227 Polyspermic fertilization of sea urchin eggs, '502 PORTER, M. E. See C. J. BERG, JR.. 469 POSTLETHWAIT, J. H. Juvenile hormone and the adult development of Drosophila, 119 Potassium ions modify sodium channel selec- tivity, 469 Prasiola, inhibition of by molybdate, 470 698 INDEX PRICE, C. A., L. R. MENDIOLA-MORGENTHALER, M. GOLDSTEIN, E. N. BRED EN AND R. R. L. GUILLARD. Harvest of planktonic marine algae by centrifugation into gradients of silica in the CF-6 continuous- flow zonal rotor, 136 Proboscidactyla flavicirrata, larval settlement of, 573 Proboscis of the whelk, 369 Properties of Arbacia sperm fertilization anti- gen, 490 Protein, metabolism in fish, 500 of sea urchin eggs, 485 PRUSCH, R. D. See F. C. G. HOSKIN, 483 Puparium, formation in flies, 163 tanning factor in flies, 163 Purification of Arbacia sperm fertilization antigen, 490 R Rabbit brain, isolation of, 488 Radioautography study of sea urchin oogenesis, 586 Radular protractor muscle of Busycou, in- nervation of, 369 RAMTS, J., S. BEALE, D. MAUZERALL AND K. HOWARD. Correlation of changes in pig- ment content with photosynthetic capacity of seaweeds as a function of water depth, 494 See S. E. BINGHAM, 470 Rana catesbeiana, hypersynchrony in cortex of. 492 RAYPORT, S. See G. WALD, 503 READ, C. P., G. L. STEWART AND P. W. PAP- PAS. Glucose and sodium fluxes across the brush border of Hymenolepis diiirimita (Cestoda), 146 REEVE, M. R., AND B. LESTER. The process of egg-laying in the chaetognath Sagitta hispida, 247 Reflexes in the tailspine system of the horse- shoe crab, 474 Regeneration in freshwater shrimp, 203 Reproduction, in Libinia, 257 in Sai/itta, 247 Reproductive biology of Pelogobia, 457 Reproductive systems of sea-stars, 661 Respiration, discontinuous, in insects, 294 nervous control, in Cecropia pupae, 274 in littoral snails, 496 pathways in Ovalipcs, 550 Responses, evoked, in frog neurotnuscular junction, 487 REYNOLDS, G. T. See J. G. MORIN, 397 See G. FREEMAN, 477 Rhipidocotyle spp., life-cycle of, 500 RICH, S. See D. T. WOODRUM, 503 RICHARDSON, N. E., AND J. D. McCLEAVE. Locomotor activity rhythms of juvenile Atlantic salmon (Sahiw salar) in var- ious light conditions, 422 RIPPS, H., M. SHAKIB AND E. D. MACDONALD. Turnover of synaptic vesicles in photo- receptor terminals of the skate, 495 RITTER, H. See S. INOUE, 484 RODRICK, G. E. See T. C. CHENG, 311 ROSE, S., G. D. PAPPAS, M. KRIEBEL AND A. J. TOUSIMIS. Evidence for the synaptic vesicle calcium binding site at the neuro- muscular junction of the frog sartorius, 495 See M. E. KRIEBEL, 487 See A. POLITOFF, 494 Ross, W. N., B. M. SALZBERG, L. B. COHEN, H. V. DAVILA, A. S. WAGGONER AND C. H. WANG. A large change in axon ab- sorption during the action potential, 496 RUDERMAN, J. V. See R. B. GOLDBERG, 478 RUSSELL-HUNTER, W. D., AND R. F. Mc- MAHON. Patterns of aerial and aquatic respiration in relation to vertical zonation in four littoral snails, 496 See R. F. McMAHON, 490 RUSTAD, R. C., E. I. CASSEN AND S. D. GHARIB. Reduction of radiation-induced mitotic delay in sea urchin eggs following prefertilization ionic stimulation, 497 See S. ZIGMAN, 504 Sayitta liispida, egg-laying in, 247 Salmo salar, locomotor activity rhythms of, 422 Salmon, juvenile, locomotor activity rhythms of, 422 SALZBERG, B. M. See W. N. Ross, 496 SANGER, J. W. See R. B. HILL, 369 Sarcophaga bullata, hormonal factors acting during puparium formation in, 163 SAUNDERS, J. A., R. R. L. GUILLARD AND F. A. LOEWUS. Conversion of D-glucose to L-ascorbic acid in Cyclotclla cryptica Rei- mann, Lewin and Guillard, 497 Scale worm, bioluminescence in, 480 Scallop, giant, larval development of, 321 SC-HIFF, J. A. See M. L. TSANG, 502 See S. E. BINGHAM, 470 Schisoporclla, gene frequencies in, 498 SCHNEIDERMAN, H. A. See B. N. BURKETT, 274, 294 SCHOPF, T. J. M. Long-term (3 to 5 year") records of gene frequencies in natural populations of an abundant, sub-tidal species (the bryozoan Schizoporclla er- rata), 498 SCHUEL, H. See W. TROLL, 502 INDEX 699 SCHUETZ, A. W. Activation of immature star- fish oocytes by a divalent ionophore and sperm, 498 SCRUGGS, V. See D. LANDOWNE, 487 Sea urchin, fertilizing capacity of sperm, 475 hatching enzyme, effect of carbobenzoxy- glutamate on, 489 oogenesis, 586 sperm chromatin template, activation of, 486 spermatozoa, effect of heavy metal ions on motility of, 236 Sea urchin eggs, iodination and autoradiog- raphy of the plasma membrane of, 474 mitotic delay in, 497 polyspermic fertilization of, 502 proteins of, 485 Seaweeds, pigment content and photosynthetic capacity of, 494 SEITZINGER, S. See I. VALIELA, 652 Self and not-self in colonial organisms, 411 Senescence, relation to variation in corneal epithelial cell surfaces, 479 Sequence elements in Spisiila DXA, 491 Serum in catfish oocytes, 481 Sexual dimorphism in fiddler crabs, 652 reproduction in S\iiiplc(jnui. 213 SHAKIB, M. See H." RIPPS, 495 SHOREY, J. M., AND J. G. MORIN. Quantifica- tion of light produced from hydrozoan photocytes, 499 Shrimp, freshwater, limb loss and molt cycle in, 203 SINGER, H. J., AND E. R. LEADBETTER. Isola- tion of spore-forming microorganisms from Little Sippewissett marsh, 499 SlVASUBRAMANIAN, P., S. FRIEDMAN AND G. FRAENKEL. Nature and role of protein- aceous hormonal factors acting during puparium formation in flies, 136 Size-dependent variation at the LAP locus in Mytilus, 535 Size distribution of loricae in Eel Pond, 477 Skate, synaptic vesicles in photoreceptors of 495 electroreceptor epithelium, 472 SMITH, M. A. K., AND A. E. V. HASCHE- MEYER. Studies on protein metabolism and growth in fish, 500 SMITH, S., J. OSHIDA AND H. BODE. Inhibi- tion of nematocyst discharge in hydra fed to repletion, 186 Snails, littoral and sublittoral, ecology of, 490 Sodium channel selectivity modified by internal potassium ions, 469 efflux from squid axon, 487 fluxes in Hymenolcpis, 146 pump in squid axon, 482 Soluble crystallins formation of dogfish lens 478 Sonic motor nerve of the toadfish, 483 SOTELO, C. See H. KORN, 486 Spatial summation in rods from the toad retina, 475 Sperm, activation of immature starfish oocytes, 498 chromatin template, sea urchin, activation of, 486 fertilizing capacity of the sea urchin, 475 motility control, 236 sea urchin, effect of heavy metal ions on motility of, 236 Spermiogenesis in Crepidula, 480 SPIELMAN, A., AND J. WONG. Dietary factors stimulating oogenesis in Acdcs acg\pti 433 Spiracles, behavior of, 274, 294 Spisula, sequence elements in DNA of, 491 Spisula eggs, fertilization of, 504 Sponges, current-induced flow through, 443 marine, diapause in gemmules of, 333 Spore-forming microorganisms, isolation of, 499 STRAY, D. See W. CLUSIN, 472 Squid axon, association of divalent cations with, 493 effects of external cations on ouabain inhibi- tion rate of sodium pump in, 482 freeze-etching of, 491 magnesium ion transport, 473 sodium efflux from, 487 synthesis of isethionate, 483 Squid axon, absorption, change in, 496 energy transfer in, 501 sodium channel selectivity in, 469 Squid embryos, cytokinesis in, 466 Squid giant synapse, facilitation of transmitter release, 471 suppression potential in, 489 Squid optic lobe, uptake from, 468 Staining organic matter in sediments with histological stains, 227 Starfish oocytes, activation of, 498 Stellate ganglia, squid, transmitter release, 471 STEWART, G. L. See C. P. READ, 146 STOFFEL, L. A., AND J. H. HUBSCHMAN. Limb loss and the molt cycle in the freshwater shrimp, Palacnwiictcs kadiakcnsis. 203 Stomach capacity in spider crabs, 257 Strontium, effects of on timing of miniature end-plate potentials, 472 STRUNKARD, H. W. The life-cycle of the gasterostome trematodes, Rhipidocotylc transversals Chandler, 1935 and Rhipido- cotylc lintoni Hopkins, 1954, 500 SUGIMOTO, K., AND M. NAKAUci-ii. Budding, sexual reproduction, and degeneration in the colonial ascidian, Svinpleqnia reptans, 213 700 INDEX Sulfate reducers, sulfotransferases in, 502 Suppression potential, squid, aequorin study of, 489 SUSAN, S. See C. HARDING, 479 Symplcgma rcptans, budding, sexual repro- duction and degeneration in, 213 Synapses in the labyrinth of the toadfish, 482 Synaptic transmission, frog, effect of acetyl- cholinesterase on, 494 Synaptic vesicles, calcium binding site in frog sartorius, 495 in skate photoreceptors, 495 of radular protractor muscles, 369 Synthesis, anaerobic, of bacterial luciferase, 475 of isethionate by squid axon, 483 Tailspine system of the horseshoe crab, 474 TANIGAWA, Y. See S. S. KOIDE, 486 TASAKI, I., A. WARASHINA AND H. PANT. Energy transfer between fluorescent probe molecules in and across nerve membrane, 501 TAYLOR, D. L. In vivo contractility of amoeba cytoplasm, 501 Temperature influence on sponge gemmules, 333 low, effect on spiracles and respiration, 294 subzero, effect on Arbacia eggs, 484 TENORE, K. See J. J. LEE 488 TIETJEN, J. H. See J. J. LEE, 488 Tintinnida growth status in Eel Pond, 477 Tissue, photogenic, structure of cephalopods, 507 Toad retina, adaptation and spatial summation in rods, 475 Toadfish, analysis of fast axonal transport in the sonic motor nerve of, 483 glycine transport in the intestine, 476 transmission at synapses, 482 Toadfish liver, amino acids, computer program for, 493 elongation factor 1 of, 492 theoretical analysis of amino acid uptake, 479 TOPPEN, D. L. See K. KUSTIN, 608 TOUSIMIS, A. J. See S. ROSE, 495 Transmission at synapses in the labyrinth of the toadfish, 482 Transmitter release at the squid giant synapse, 471 Transport, glycine, in the intestine of the toadfish, 476 Transport in squid axon, 473 TROLL, W., H. SCHUEL AND W. L. WILSON. Induction of polyspermic fertilization of Arbacia eggs by specific protease in- hibitors leupeptin and antipain, 502 Tryptophan photoproducts, inhibition of cell growth by, 504 TSANG, M. L., AND J. A. SCHIFF. Distribution of adenosine-5'-phosphosulfate (APS) and adenosine-3'-phosphate-5'-phosphosul- fate (PAPS) sulfotransferases in as- similatory sulfate reducers, 502 TSURU, N. See F. MORRELL, 492 Tubularia, muscle activation in, 594 Tubulin components of squid axon, 491 Tunicate, water transport rates of, 608 Turbellarian, freshwater, oocytes of, 618 u Uca pugnax, feeding in, 652 Ultrastructure of a cephalopod photophore, 507, 522 Uptake and metabolism of choline from squid optic lobe, 468 VALIELA, I., D. F. BABIEC, W. ATHERTON, S. SEITZINGER AND C. KREBS. Some con- sequences of sexual dimorphism : feeding in male and female fiddler crab, Uca pugnax (Smith), 652 VAN DER KLOOT, W. G. See I. S. COHEN, 472 Vascular ampullae of colony, 411 Vertical zonation in Littorina, 496 VICKERS, G. R. See L. A. BARKER, 468 Vitellogenic protein in serum of catfish oocytes, 481 VOGEL, S. Current-induced flow through the sponge, Halichondria, 443 W WAGGONER, A. S. See W. N. Ross 496 WAINWRIGHT, S. A. See J. M. HEFFERNAN, 95 WALD, G., AND S. RAYPORT. Skin photorecep- tors in the leopard frog, 503 WALKER, C. W. Studies on the reproductive systems of sea-stars. I. The morphology and histology of the gonad of Astcrias vulgaris, 661 WALKER, G. The occurrence, distribution and attachment of the pedunculate barnacle Octolasmis mullcri (Coker) on the gills of crabs, particularly the blue crab, Cal- lincctcs sapidus Rathbun, 678 WALL, B. J. See J. L. OSCHMAN, 493 WALLACE, R. A. See S. J. ATLAS, 466 See E. D. HICKEY, 481 See S. C. JACKOWSKI, 484 WALTON, A. See G. FREEMAN, 477 WANG, C. H. See W. N. Ross, 496 WARASHINA, A. See L TASAKI, 501 Warts of apodous holothurians, 95 WATANABE, H. See H. MUKAI, 411 Water transport rates of dona, 608 INDEX 701 WEISSMANN, G., D. BLOOMGARUEN, R. KAP- LAN, C. COHEN, S. HOFFSTEIN, T. COL- LINS, A. GOTTLIEB AND D. NAGLE. A general method for the introduction of enzymes, by means of liposomes, into lysosomes of deficient cells, 503. WHITLATCH, R. B. Food-resource partitioning in the deposit feeding polychaete Pectin- aria gouldii, 227 WILLIAMS-ARNOLD, L. D. See J. M. ARNOLD, 466 WILSON, W. L. See W. TROLL, 502 WONG, J. See A. SPIELMAN, 433 WOODRUM, D. T., S. RICH AND T. D. POLLARD. Evidence for biased bipolar polymerization of actin filaments, 503 YINGST, D. The vertical distribution and re- production biology of Pclogobia longi- cirrata (Annelida) in the central Arctic Ocean, 457 Yolk halo of Hydrolimax oocytes, 618 YOUNG, L. G., AND L. NELSON. The effects of heavy metal ions on the motility of sea urchin spermatozoa, 236 YOUNG, R. E. See J. M. ARNOLD, 507, 522 YULO, T. See G. GRIESS, 478 See S. ZIGMAN, 504 ZIGMAN, S. See G. GRIESS, 478 T. YULO, G. GRIESS, B. ANTONELLIS AND R. RUSTAD. How cell growth is inhibited by near UV light photoproducts of tryp- tophan, 504 Zinc, effect on sea urchin motility, 236 ZIOMEK, C. A., AND D. EPEL. Studies on the establishment of a cytochalasin B sensitive rapid polyspermy block following fertiliza- tion of Spisula solidissiina eggs, 504 Zooxanthellae modification of sea anemone behavior, 630, 641 Continued from Cover Two 4. Literature Cited. The list of references should be headed LITERATURE CITED, should conform in punctuation and arrangement to the style of recent issues of THE BIOLOGICAL BULLETIN, and must be typed double- spaced on separate pages. Note that citations should include complete titles and inclusive pagination. Journal abbreviations 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 below. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS of those abstracted (most recent issue: November, 1972). Foreign authors, and others who are accustomed to use THE WORLD LIST OF SCIENTIFIC PERIODICALS, may find a booklet published by the Biological Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K. at £0.65 or $1.75) useful, since it sets put the WORLD LIST abbreviations for most biological journals with notes of the USASI abbreviations where these differ. CHEMICAL AB- STRACTS 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 period, whole word components must not (not strictly as USASI usage, i.e. J. Cancer Res.) D. Space between all components (e.g. J. Cell. Comp. Physiol. not J.Cell.Ccmp.Physiol.) E. We strongly recommend that more unusual words in journal titles be spelled out in full, rather than employing lengthy, peculiar "abbreviations" or new abbreviations invented by the author. For example, use Rit Visindafjelags Islendinga without abbreviation. Even in more familiar languages, Z. Vererbungslehre is preferred to Z. VererbLehre (WORLD LIST) or Z. Verer- bungsl. (USASI). Accurate and complete communication of the reference is more important than minor savings in printing costs. F. All single word journal titles in full (e.g. Veliger, Ecology, Brain). G. The order of abbreviated components should be the same as the word order of the com- plete title (i.e. Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. Spell out London, Tokyo, Paris, Edinburgh, Lisbon, etc. where part of journal title. I. Series letters etc. immediately before volume number. J. A few well-known international journals in their preferred forms rather than WORLD LIST or USASI usage (e.g. Nature, Science, Evolution NOT Nature, Land.; Science, N. Y.; Evolution, Lancaster, Pa.) K. The correct abbreviation for THE BIOLOGICAL BULLETIN is Biol. Bull. 5. Figures. The dimensions of the printed page, 5 by 7f inches, should be kept in mind in preparing figures for publication. Illustrations should be large enough so that all details will be clear after appropriate reduction. Explanatory matter should be included separately in legends as far as possible, although the axes should always be numbered and identified on the illustration itself. Figures should be prepared for reproduction as either line-cuts or halftones; no other methods will be used. Figures to be reproduced as line-cuts should be 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, and both designating numbers or letters and scale-bars should be affixed directly on the figures. We recommend that halftones submitted to us be mounted prints made at about lj times the linear dimensions of the final printing desired (the actual best reductions are achieved from copy in the range from \\ to 2 times the linear dimen- sions). As regards line-blocks, originals can be designed for even greater reductions but are best in the range If to 3 times. All figures should be numbered in consecutive order, with no distinc- tion between text and plate-figures. The author's name should appear on the reverse side of all figures, and the inked originals for line-blocks must be submitted for block-making. 6. Mailing. Manuscripts should be packed flat. All illustrations larger than 8^ by 11 inches must be accompanied by photographic reproductions or tracings that may be folded to page size. Reprints. Reprints may be obtained at cost; approximate prices will be furnished by the Managing Editor upon request. CONTENTS ARNOLD, JOHN M. AND RICHARD E. YOUNG Ultrastructure of a cephalopod photophore. I. Structure of the photogenic tissue 507 ARNOLD, JOHN M., RICHARD E. YOUNG AND MAURICE V. KING Ultrastructure of a cephalopod photophore. II. Iridophores as re- flectors and transmitters 522 BOYER, JOHN F. Clinal and size-dependent variation at the LAP locus in Mytilus edulis 535 CAINE, EDSEL A. Feeding of Ovalipes guadulpensis (Saussure) (Decapoda: Brachy- ura : Portunidae), and morphological adaptations to a burrowing existence 550 DIETZ, THOMAS H. Body fluid composition and aerial oxygen consumption in the fresh- water mussel, Ligumia subrostrata (Say) : effects of dehydration and anoxic stress 560 DONALDSON, SVEN Larval settlement of a symbiotic hydroid : specificity and nematocyst responses in planulae of Proboscidactyla flavicinata 573 ENESCO, H. ESPER AND K. H. MAN Cytoplasmic DNA in sea urchin oogenesis studied by 3H-actinomycin D binding and radioautography 586 JOSEPHSON, ROBERT K. Factors affecting muscle activation hi the hydroid Tubularia 594 KUSTIN, KENNETH, KAYE V. LADD, GUY C. MCLEOD AND DAVID L. TOPPEN Water transport rates of the tunicate dona intestinalis 608 NEWTON, W. DONALD The accessory cell and yolk halo of the ob'cyte of the freshwater turbellarian Hydrolimax grisea (Platyhelminthes ; Plagiostomidae). 618 PEARSE, VICKI BUCHSBAUM Modification of sea anemone behavior by symbiotic zooxanthellae : phototaxis 630 PEARSE, VICKI BUCHSBAUM Modification of sea anemone behavior by symbiotic zooxanthellae : expansion and contraction 641 VALIELA, IVAN, DANIEL F. BABIEC, WILLIAM ATHERTON, SYBIL SEITZINGER AND CHARLES KREBS Some consequences of sexual dimorphism: feeding hi male and female fiddler crabs, Uca pugnax (Smith) 652 WALKER, CHARLES WAYNE Studies on the reproductive systems of sea-stars. I. The morphology and histology of the gonad of Asterias vulgaris 661 WALKER, GRAHAM The occurrence, distribution and attachment of the pedunculate barnacle Octolasmis miilleri (Coker) on the gills of crabs, particularly the blue crab, Callinectes sapidus Rathbun. 678 MBL WHOI LIBRARY UH 1B1J