| a' A|' B'| a =, \o.\ alan, ararur wr in ae ‘ lAIAAA\ | \f | AA alala : AARNA AMAR ae Reena AAP Ap AAR ‘i Oy nnontgaronmnneat RAMAN NARA, NAAAAD CAR AARARRS WAN ARERR RAAAAARAR lain A AARAAAAAAA AA A gal ala | \ \ \ a a alg \, Walaa al RAY NAVAS A -AAAAAAs mnnnnnnn RARRAAAA A> | ( A RR RRGALELECOE wa A AAA RM AAARANAAAAND _o @& a . ai vy V¥Y Vw w wr PAA (PS WMS > ao >_ ie “ AEE > 2D) > > >» ) > > >» YD» >> » >> ») a | an! nAAAAAAAAAAAAA A 3 a} han WARARAAAM Anan ANAAAWAA AAA AAR: = ALA A ry | ‘a AVY V\ 2 Ap AAAAA P AP NAlA AARAA AR A AAAAAA AAA VV \f APA AAAS AAy Apr). IAAAAARAAAAAARRAAAAAAAA ioe wy LA NAANAAPYY lala AAA AA AAA aaa AAaaaaaaaa AA AAAAARAA ann AA ANA \ ~ i } | A AA A Maia rYV\.¥ aN tf | ) f - u | } \ \ eo i 5 } la AAAI AR A la WANA panaaanan Cara ARO AIAAAA AAA AAANAANAA A TAARe PERERA AAA BRERA EE al oe or) Rt ae senha ; Pe nnhih Iie, PRAY SY ue CHE ashe Oe yk my tay eti\. ences, Rag iui rf Seat f THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board Gary N. CALKINS, Columbia University E. E. JUST, Howard University E. G. CONKLIN, Princeton University FRANK R. LILLIE, University of Chicago E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden LEIGH HOADLEY, Harvard University T. H. MORGAN, California Institute of Technology M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University H. S. JENNINGS, Johns Hopkins University EDMUND B. WILSON, Columbia University ALFRED C. REDFIELD, Harvard University Managing Editor VOLUME LXXIII AUGUST TO DECEMBER, 1937 ‘Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA. li THE BIOLOGICAL BULLETIN is issued six timesa year. Single numbers, $1.75. Subscription per volume (3 numbers), $4.50. Subscriptions and other matter should be addressed to the Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa. Agent for Great Britain: Wheldon & Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W.C. 2. Communications relative to manuscripts should be sent to the Managing Editor, Marine Biological Laboratory, Woods Hole, Mass., between June 1 and October 1 and to the Institute of Biology, Divinity Avenue, Cambridge, Mass., during the re- mainder of the year. Entered October 10, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894. LANCASTER PRESS, INC., LANCASTER, PA. CONTENTS No. 1. AvucGust, 1937 . PAGE THIRTY-NINTH REPORT OF THE MARINE BIOLOGICAL LABORATORY 1 SAYLES, LEONARD P., AND S. G. HERSHKOWITZ (ERI sais Sig ty A eS RD Ba ae ee Ue tte a a aR sill Marza, V. D., EUGENIE V. MARZA AND Mary J. GUTHRIE Histochemistry of the Ovary of Fundulus heteroclitus with — Special Reference to the Differentiating Odcytes........... 67 MATTHEWS, SAMUEL A. The Development of the Pituitary Gland in Fundulus....... 93 Beams, H. W., AND R. L. KING we SUT INGT oe ove Gat Aten eNO satel Aa MMM aD oar weet cess, al cui nh 99 Miter, E. DEWrTT A Study of the Bacterial and Alleged Mitochondrial Content OhtheCells-of the Clover Nodulev 28) 3 sc dees 112 Mast, S. O., AND NATHAN STAHLER The Relation between Luminous Intensity, Adaptation to Light, and Rate of Locomotion in Amoeba proteus (Leidy).. 126 ABRAMOWITZ, A. A. The Réle of the Hypophyseal Melanophore Hormone in the Chromatic Physiolosy of Funduluse 2.2 see aee ee eee 134 BUTLER, MARGARET RUTH The Effect of its Nitrogen Content on the Decomposition of the Polysaccharide Extract of Chondrus crispus............ 143 PAYNE, NELLIE M. The Differential Effect of Environmental Factors upon Micro- bracon hebetor, Say (Hymenoptera: Braconidae) and its Host, Ephestia kiihniella Zeller (Lepidoptera: Pyralidae). TTI Te, GEO Scan aA Ch eB Ne aa EA alia hes ie i a a ea 147 HOADLEY, LEIGH AuLovomy mm: che brachyiran,|Uca pieiax.. 200.0...) ae 155 TYLER, ALBERT, AND HANS BAUER Polar Body Extrusion and Cleavage in Artificially Activated Bscrot Wnechis Cato sas. adn: eta sc eee es eee Me, ele 164 iv CONTENTS PAGE Boyp, WILLIAM C. Cross-reactivity of Various Hemocyanins with Special Refer- ence to the Blood Proteins of the Black Widow Spider...... 181 No. 2. OcToBErR, 1937 WELsH, J. H., F. A. CHAcz, Jr., AND R. F. NUNNEMACHER The Diurnal Migration of Deep Water Animals............ 185 COONFIELD, B. R., AND A. GOLDIN The Problem of a Physiological Gradient in Mnemiopsis During Nesenerationeee, tat eee ki) weer es fet 2 197 GLASER, OTTO, AND GEORGE P. CHILD ihe Hexoctalrednon and :Growthhy 4 32. ae ee ee ee 205 CARVER, GAIL L. Studies on Productivity and Fertility of Drosophila Mutants 214 BALL, Eric G., AND C. CHESTER STOCK The pH of Sea Water as Measured with the Glass Electrode 221 GOLDSMITH, E. D. The Relation of Endocrine Feeding to Regeneration, Growth, and Egg Capsule Production in Planaria maculata.......... 22 PROSSER, C. LADD, AND JOHN Z. YOUNG Responses of Muscles of the Squid to Repetitive Stimulation of the: GramtNerve: Fibers...) .h1 jus eee See gee eee 237 SPARROW, F. K., JR. The Occurrence of Saprophytic Fungi in Marine Muds...... 242 WHITAKER, D. M. Determination of Polarity by Centrifuging Eggs of Fucus LEBEN CUIS ty S302 le seam SOO MER Wert a ge age i 249 TYLER, ALBERT, AND W. D. HuMASON On the Energetics of Differentiation, VI. Comparison of the temperature coefficients of the respiratory rates of unfertilized andvok tentilized ie9es <5... occ Jae eee bee os le eee 261 KENK, ROMAN Sexual and Asexual Reproduction in Euplanaria tigris (Girard) 280 HORSTADIUS, SVEN Investigations as to the Localization of the Micromere-, Skeleton and Entoderm-forming Material in Unfertilized Egg (0) Svaicl OTC itn en MEISE ER Coo) ce 295 HORSTADIUS, SVEN Experiments on Determination in the Early Development of Gerebratulusilacteusan. ac occ eee taae eae ee 317 PROGRAM AND ABSTRACTS OF SCIENTIFIC MEETINGS, SUMMER OB DOS 7k tire aaa ate harness 508 ri oe a a SEs 343 CONTENTS Vv No. 3. DECEMBER, 1937 PAGE WATERMAN, A. J. Effect of Salts of Heavy Metals on Development of the Sea WicchiiwAnbacia pUNnctMlatay a! eat Me ee hha ieee 401 REDFIELD, ALFRED C., HOMER P. SMITH, AND BOSTWICK KETCHUM The Cycle of Organic Phosphorus in the Gulf of Maine...... 421 ANDERSON, BERTIL GOTTFRID, H. LUMER, AND L. J. ZUPANCIC, JR. Crowth and Vanability im Daphnia pulex:...-..0......... 444 CLARKE, GEORGE L., AND DONALD J. ZINN Seasonal Production of Zodplankton off Woods Hole with special reference to Calanus finmarchicus.................. 464. Lituick, Lots C. Seasonal Studies of the Phytoplankton off Woods Hole, Massa- . CIMMISSTNES Be: os Gk A Cae ete ne eR oe ee oa i ee 488 CAMPBELL, MILDRED L., AND ABBY H. TURNER Serum Protein Measurements in the Lower Vertebrates. I. The colloid osmotic pressure, nitrogen content, and refractive IMmdexoh turtle sexum- and body Mimiday sa. .456 eos. eo 8 504 TURNER, ABBY H. Serum Protein Measurements in the Lower Vertebrates. II. In marine teleosts and elasmobranchs...................-. Sulil Goopricu, H. B., AND Maurice A. SMITH Genetics and Histology of the Color Pattern in the Normal and Albino Paradise Fish, Macropodus opercularis L........ o2i Darton, H. CLark, AND H. B. GoopRIcH Chromatophore Reactions in the Normal and Albino Paradise IS Tereyi eg Mew Psa et eciaes. gS SMe ik Sl Aa RE eR eee, aie aaa wa 535 STURTEVANT, A. H. Autosomal Lethals in Wild Populations of Drosophila pseudo- lose uve: Ox sete aes i eeea sits IANS NAD UNA DADA O RID Se Dette eT 542 WHITAKER, D. M., anp C. M. CLANCY The Effect of Salinity upon the Growth of Eggs of Fucus fur- CSAC UIS) 103 Aare pte ees RUSS Reais Naas? A LPM eons NRE MER 6 552 HEILBRUNN, L. V., AND Kart M. WILBUR Stimulation and Nuclear Breakdown in the Nereis Egg...... Soil Fry, Henry J. Studies of the Mitotic Figure. VI. Mid-bodies and their sig- nificance for the central body problem..................-- 565 Riocu, Davin McK. A Physiological and Histological Study of the Frontal Cortex GS TOV S1E ea Si ea 2S oe ae rk i a RE Lf 591 "Volume LXXDT quill sy : Number 1 ia ~ as “4 A ” : ‘ } 2 5 J ' a . ' if s! Na f a N47; AS “SONAL EEE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Editorial Board GARY N. CALKINS, Columbia University E. E. JUST, Howard University E. G. CONKLIN, Princeton University FRANK R. LILLIE, University of Chicago E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden LEIGH HOADLEY, Harvard University T. H. MorRGAN, California Institute of Technology M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University H. 8. JENNINGS, Johns Hopkins University EDMUND B. WILSON, Columbia University ALFRED C. REDFIELD, Harvard University Managing Editor AUGUST, 1937 Printed and Issued by LANCASTER PRESS, Ine. PRINCE & LEMON STS. LANCASTER, PA. THE BIOLOGICAL BULLETIN Tue BIoLocicaL BULLETIN is issued six times a year. Single numbers, $1.75. Subscription per volume (3 numbers), $4.50. Subscriptions and other matter should be addressed to the Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa. Agent for Great Britain: Wheldon & Wesley, Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London, W.C. 2. : Communications relative to manuscripts should be sent to the Managing Editor, Marine Biological Laboratory, Woods Hole, Mass., between June 1 and October 1 and to the Institute of Biology, Divinity Avenue, Cambridge, Mass., during the remainder of the year. | : INSTRUCTIONS TO AUTHORS Preparation of Manuscript. In addition to the text matter, manuscripts should include a running page head of not more than thirty-five letters. Footnotes, tables, and legends for figures should be typed on separate sheets. Preparation of Figures. The dimensions of the printed page (43¢x7 inches) should be borne in mind in preparing figures for publication. Draw- ings and photographs, as well as any lettering upon them, should be large enough to remain clear and legible upon reduction to page size. Illustrations should be planned for sufficient reduction to-permit legends to be set below them. In so far as possible, explanatory matter should be included in the legends, not lettered on the figures. Statements of magnification should take into account the amount of reduction necessary. Figures will be reproduced as line cuts or’ halftones. Figures intended for reproduction as line cuts should be drawn in India ink on white paper or blue-lined coordinate paper. Blue ink will not show in reproduction, so that all guide lines, letters, etc. must be in India ink. Figures intended for reproduction as halftone plates should be grouped with as little waste space as possible. Drawings and lettering for halftone plates should be made directly on heavy Bristol board, not pasted on, as the outlines of pasted letters or drawings appear in the reproduction unless removed by an expensive process. Methods of repro- duction not regularly employed by the Biological Bulletin will be used only at the author’s expense. The originals of illustrations will not be returned except by special request. — Directions for Mailing. Manuscripts and illustrations should be packed flat between stiff cardboards. Large charts and graphs may be rolled and sent in a mailing tube. Reprints. Authors will be furnished, free of charge, one hundred re- prisits without covers. Additional copies may be obtained at cost. Proof. Page proof will be furnished only upon special request. When cross-references are made in the text, the material referred to should be marked clearly on the galley proof in order that the proper page numbers may be supplied. Bntered October 10, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894. ee Vol. LXXIII, No. 1 THE August, 1937 BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY rt VANE BIOLOGICAL LABORATORY Turrty-NintH Report, FOR THE YEAR 1936— Forty-NINTH YEAR I. Trustees AND ExecuTIvVE ComMITTEE (As oF Aucust II, HOR GC) ee ee tere ee aM UN Cleve, ok Ba lta 1 TEAR an © OVEN TD RE n 1s cic de apsaeinie ty eet atiencuabsreectn OS elcr enelb 3 ee NC REO HIN CORPORAI TON) (0s ents cusnay a chee ie eusoshecMicteyedeheueier <2 seve 3 wy eAWGuOn DEE | CORPORATION sets efile 5 a)2 Dinh arbre ile Bee « 4 IN eGIWE PORMOF DHE URDASURER ay shel ae te eek arene ele Gh S) Neier PORTION CMELE STB RATAN ch viet os ella crares Ge ain sale eg seals 10 Vie PORT) Ob VMETE GOMRECHOR! ass «6. sacs see ales wire nee els 11 SEAECMTE TUES AU ey nee cere Peele ree Altes te as alate. alin areas 11 Addenda: end Se RS teat S Oy anv rita akeer cc ck otk Pinata rates ote mua 16 2 Anvestivatons and Students, 1936) 2: 2a. 5a. 5 -). sae 18 Salalbularm Niewmolse AtreM@ance: |... vas sacs see 29 4. Subscribing and Codperating Institutions, 1936 .... 30 5a areanae ILeeamey MOTO aM athe y Go diols Golem © 46 clots 30 6. SMOcwsr Sem msneleeynatsy MEE 855. ay doe yo 5 a 4 O 31 7, General Semanune Wiseutae, WO 2. c..0sce0d8ecae 34 On Menibersrot the Corporation) se eno tae eal 38 1 IN USUaISS) EX OFFICIO Frank R. Lituir, President of the Corporation, The University of Chicago. MERKEL H. JAcogs, Director, University of Pennsylvania. Lawrason Ruccs, Jr., Treasurer, 120 Broadway, New York City. CHARLES PACKARD, Clerk of the Corporation, Columbia University. EMERITUS C. Bumpus, Brown University. G. ConKLin, Princeton University. R. Crane, New York City. H. Donatpson, Wistar Institute of Anatomy and es J. GrrenMAN, Wistar Institute of Anatomy and Biology A. Harper, Columbia University. . M. Metcatr, Waban, Mass. . H. Parker, Harvard University. 1 gece ie i) MARINE BIOLOGICAL LABORATORY W. B. Scort, Princeton University. W. M. WueEE ER, Harvard University. EavBe WILson, Columbia University. TO SERVE UNTIL 1940 H. B. Bicetow, Harvard University. R. Cuampers, Washington Square College, New York University. W. E. Garrey, Vanderbilt University Medical School. CASWELL GRAVE, Washington University. S. O. Mast, Johns Hopkins University. A. P. MatHeEws, University of Cincinnati. C. E. McCune, University of Pennsylvania. C. R. Srocxarp, Cornell University Medical College. TO SERVE UNTIL 1939 W. C. ALLeE, The University of Chicago. ry N. CaLxins, Columbia University. M. Ducear, University of Wisconsin. V. HEILBRUNN, University of Pennsylvania. . IrnvinG, University of Toronto. W. J. V. OsterHour, Member of the Rockefeller Institute for Medical Re- search. A. H. Sturtevant, California Institute of Technology. LorANDE L. WoopruFr, Yale University. G Be Ie 1D TO SERVE UNTIL 1938 E. R. Ciark, University of Pennsylvania. Otto C. Graser, Amherst College. Ross G. Harrison, Yale University. E. N. Harvey, Princeton University. H. S. JeEnnines, Johns Hopkins University. F. P. Knowtton, Syracuse University. FRANZ SCHRADER, Columbia University. B. H. Witter, University of Rochester. TO SERVE UNTIL 1937 W. R. AmBerson, University of Tennessee. H. B. Goopricu, Wesleyan University. I, F. Lewts, University of Virginia. R. S. Litire, The University of Chicago. T. H. Morean, California Institute of Technology. A. C. REDFIELD, Harvard University. C. C. SpEIDEL, University of Virginia. D. H. TENNENT, Bryn Mawr College. EXECUTIVE COMMITTEE OF THE BOARD OF TRUSTEES FRANK R. Lituit, Ex. Off. Chairman. MERKEL H. Jacoss, Ev. Off. LAwrason Ric6s, Jr., Ex. Off. F. P. KNow Ton, to serve until 1937. B. H. Writer, to serve until 1937. ACT OF INCORPORATION 3 E. R. Cxark, to serve until 1938. C. C. SPEIDEL, to serve until 1938. Tue Lisprary COMMITTEE E. G. ConxKLin, Chairman. WILLIAM R. AMBERSON. CaO wISELING Ll: C. C. SPEIDEL. A. H. STURTEVANT. WILLIAM R. TAYLOR. . Tue APPARATUS COMMITTEE L. V. HEILBRUNN, Chairman. W. R. AMBERSON. D. J. Epwarps. W. E. Garrey. E. N. Harvey. L. IRVING. M. H. Jacoss. B. Luck. II. ACT OF INCORPORATION No. 3170 CoMMONWEALTH OF MASSACHUSETTS Be It Known, That whereas Alpheus Hyatt, William Sanford Stevens, William T. Sedgwick, Edward G. Gardiner, Susan Minns, Charles Sedg- wick 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 scien- tific study and investigation, and a school for instruction in biology and natural history, and have complied with the provisions of the statutes of this Commonwealth in such case made and provided, as appears from the cer- tificate 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 Massachusetts, do hereby certify that said A. Hyatt, W. S. Stevens, W. T. Sedgwick, E. G. Gardiner, S. Minns, C. S. Minot, S. Wells, W. G. Farlow, A. D. Phillips, and B. H. Van Vleck, their associates and suc- cessors, are legally organized and established as, and are hereby made, an existing Corporation, under the name of the MARINE BIOLOGICAL LABORATORY, 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 Ejighty- Eight. [SEAL] HENRY OB] Pree: Secretary of the Commonwealth. 4 MARINE BIOLOGICAL LABORATORY III. BY-LAWS OF THE CORPORATION OF THE MARINE BIOLOGICAL) LABORATORY I. The annual meeting of the members shall be held on the second Tuesday in August, at the Laboratory, in Woods Hole, Mass., at 11.30 A.M., daylight saving time, in each year, and at such meeting the members shall choose by ballot a Treasurer and a Clerk to serve one year, and eight Trustees to serve four years. There shall be thirty-two Trustees thus chosen divided into four classes, each to serve four years, and in addition there shall be two groups of Trustees as follows: (a) Trustees ex officio, who shall be the President of the Corporation, the Director of the Laboratory, the Assogiate Director, the Treasurer and the Clerk; (b) Trustees Emeritus, who shall be elected from the Trustees by the Corporation. Any regular Trustee who has attained the age of seventy years shall continue to serve as Trustee until the next annual meeting of the Corporation, whereupon his office as regular Trustee shall become vacant and be filled by election by the Cor- poration and he shall become eligible for election as Trustee Emeritus for life. The Trustees ex officio and Emeritus shall have all rights of the Trustees except that Trustees Emeritus 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. II. Special meetings of the members may be called by the Trustees to be held in Boston or in Woods Hole at such time and place as may be designated. III. Inasmuch as the time and place of the Annual Meeting of Members is fixed by these By-laws, 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 said meeting, at least fifteen (15) days before such meeting, to each member at his or her address as shown on the records of the Corporation. IV. Twenty-five members shall constitute a quorum at any meeting. V. The Trustees shall have the control and management of the affairs of the Corporation; they shall present a report of its condition at every annual meeting; they shall elect one of their number President of the Cor- poration who shall also be Chairman of the Board of Trustees; they shall appoint a Director of the Laboratory; and they may choose such other officers and agents as they may think best; they may fix the compensation and define the duties of all the officers and agents; 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 offices. They shall from time to time elect members to the Corporation upon such terms and conditions as they may think best. VI. Meetings of the Trustees shall be called by the President, or by any two Trustees, and the Secretary shall give notice thereof by written or printed notice sent to each Trustee by mail, postpaid. Seven Trustees shall constitute a quorum for the transaction of business. The Board of Trustees shall have power to choose an Executive Committee from their own number, and to delegate to such Committee such of their own powers as they may deem expedient. AUG 21 1937 REPORT OF THE TREASURER 5 VII. The accounts of the Treasurer shall be audited annually by a certified public accountant. VIII. The consent of every Trustee shall be necessary to dissolution of the Marine Biological Laboratory. In case of dissolution, the property shall be disposed of in such manner and upon such terms as shall be de- termined by the affirmative vote of two-thirds of the Board of Trustees. IX. These By-laws may be altered at any meeting of the Trustees, pro- vided that the notice of such meeting shall state that an alteration of the By-laws will be acted upon. X. Any member in good standing may vote at any meeting, either in person or by proxy duly executed. iy vee REPORT Ob Gib KREASURER To THE TRUSTEES OF THE MarINE BIoLoGicAL LABORATORY : Gentlemen: Herewith is my report as Treasurer of the Marine Bio- logical Laboratory for the year 1936. The accounts have been audited by Messrs. Seamans, Stetson and Tuttle, certified public accountants. A copy of their report is on file at the Laboratory and is open to inspection by members of the Corporation. At the end of the year 1936, the book value of the Endowment Funds in the hands of the Central Hanover Bank and Trust Company as Trustee, was CT POCO ed SCCUTUIES 6. ory. Ges VR yA ORIG ieee S$ 910572.2 Grrl phe Stet Poa at. a Ne each co ah Ober eae eee ee 9,037.32 SEIS) TS a ails DS I SED TL a adie ERE gah 1,616.08 BODILY UAL ES CCUFULIESS § Aa) dls, eu cicase Anes oe Ne Seah of 172,261.84 (GEIS Ge nee MRR EMMA Dig Fo A Aas ON 21,584.91 $1,115,072.36 The income collected from these Funds was as follows: GRC RO UD TONUCTIE AIRS cokes 6 65 Soe, ss RE RE Be le $41,941.24 LOTUS MI RU LI0 Wn Stoel rae Vet OA A Sua Pkg oe a 29132 $50,192.56 an increase of more than $2,000 over the income from these Funds in L955: The income due from these Funds in arrears, some of which may never be collected, was on December 31, 1936 General Fund ....... 2 ita the WS nae ch are te is Un arr $12,605.25 [LAND SAE TET IEG IS SSS. St RE ag a Oe ee 5,050.00 $17,655.25 6 MARINE BIOLOGICAL LABORATORY The total amount in arrears was about $325 less than on December 31, 1935. The dividends from the General Biological Supply House have con- tinued—the total received for the year amounting to $12,700. Retirement Fund. A total of $4,060 in pensions was paid. The Fund at the end of the year consisted of securities of the book value (Gye arena chy a otio! Sear eee ane IO pe ha oe a egy $18,923.27 Gala Perea re ee ice Walesa ae 684.04 $19,607.31 Income in arrears on December 3lst was ................ $3125 Plant Assets. The land (exclusive of Gansett and Devil’s Lane tracts), buildings, equipment and library, represented an investment VIEW Stet, Sado 8 cy ce Aout ne $1,755,892.28 less reservenion depreciation ..-.......- 471,880.72 $1,284,011.56 Income and Expenses. Expenses including $41,782.21 depreciation exceeded income by $6,951.86. There was expended from current funds net $26,319.79 for plant account. During the year the Laboratory acquired by gift from Dr. Meigs Lot ““X,” Bay Shore property and the bathhouse on it and by purchase the Howes property, completing the frontage on Water Street, and the assets of the Bar Neck Corporation which included the leasehold and buildings of the Penzance Garage and the adjacent Spindell lot. At the end of the year the Laboratory owed $8,500 on mortgage on the Howes property and $8,500 in notes given for the acquisition of the Bar Neck assets. In addition it owed on notes.and accounts payable $10,855.71. It had accounts and notes receivable of $14,367.37 and $14,773.65 in cash and bank accounts in its current assets. Following is the balance sheet, the condensed statement of income and outgo and the surplus account all as set out by the accountants: EXHIBIT A MariINE BroLocicAL LABORATORY BALANCE SHEET, DECEMBER 31, 1936 Assets Endowment Assets and Equities : Securities and Cash in Hands of Central Hanover Bank and Trust Company, New York, Trus- tee—-Schedulesil-agand labs. Gems .n:\ Peripheral region Period Je Aca. eerie. Period 2 Bee ee ens es Internal region Period) 2 Alysia s seas Period: 2 Bi. vaca « ++ ++ Yolk Intravesicular............. Extravesicular............. 1+ I+ ++ Oécyte nucleus Ghromatin' 0) 9. ces ok — Niucleabie cry eae ated neta 3 + (?) _ | _ the incineration of sections according to the method of Policard the various parts of the ovary are recognizable. Of the mineral residues only the oxides of iron are colored orange or red; others are white or gray. The detection of iron and its localization within the cells are HISTOCHEMISTRY OF OOCYTES fall thus possible. Microincinerated sections were examined against a dark field with a Greenough microscope and, also, by means of a Reichert oblique illuminator. Inorganic iron is most abundant in the interstitial cells, some of which give a more interise reaction than others, and is rarely seen in the stroma. The nucleus of the growing oocyte does not contain iron at any stage. During the first part of the growth-period iron has not been demonstrated either in the cells of the follicle or in the cytoplasm of the oocyte. However, in the second period inorganic iron is some- times seen in the follicle cells and in the zona radiata but not through- out the entire circumference of any one egg. Within the oocyte at this stage a positive inorganic iron test of weak intensity is sometimes ob- tained in the peripheral layer of cytoplasm, as well as in the internal cytoplasm between the yolk vesicles. In the intravesicular yolk the in- organic iron test is sometimes positive, but in the extravesicular yolk inorganic iron has not been detected. After microincineration the fully differentiated yolk is seen to contain iron, which must have been in organic combination (Fig. 11). The summary of tests for iron is given in Table I. It should also be recorded that white ash (probably calcium) is abundant in the cytoplasm of odcytes in Period 1 (Fig. 8). These oocytes differ strikingly from those of the hen, which do not contain ash at a comparable stage. White ash also occurs in both the peripheral and internal cytoplasmic regions during Period 2 in oocytes of FPundulus (Fig. 9). This white ash may obscure the red where it is present in small amounts and render the negative microincineration results for iron unreliable for certain regions. As the intravesicular yolk appears a gray ash remains after microincineration (Fig. 10). The nucleoli of the fish egg are very rich in white ash, and the chromatin in gray ash. Inorganic iron apparently passes through the follicle and zona radiata of the odcyte during the period of yolk deposition and is found in the cytoplasm and in the yolk forming within the vesicles. During the final differentiation of yolk the accumulated iron must be combined in organic form, probably in the hematogen of ichthulin. The finding of iron in both inorganic and organic form may account for certain discrepancies in previous reports. QUANTITATIVE DETERMINATION OF PorassIUM The quantity of potassium exceeds the combined amounts of cal- cium, sodium, and magnesium in ripe fish eggs. Bialaszewicz (1929) has shown that high potassium content is characteristic of egg cytoplasm (2, V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE of both invertebrates and vertebrates. Seventy-four to 92 per cent of all the diffusible bases in egg cytoplasm is potassium. Chemically the potassium content of a number of ripe and fertilized fish eggs has been determined (Needham, 1931, p. 356), but the eggs of Fundulus have not been studied from this point of view. A histochemical study of potassium has not been made on any fish egg, nor have studies been made on any differentiating odcytes. The method used was that of Macallum (1905) as modified by Marza and Chiosa in 1934 and 1935. By means of a comparison eye- Fic. 2. Use of comparison eyepiece in quantitative histochemical observations. A, arrangement of microscopes and light; B, diagram of preparation of ovary (left, below), standard slide (right, below), and field of vision (center, above). piece and a series of standards made of agar-agar and containing con- centrations of cobalt sulphide calculated as potassium equivalents, it is possible to determine the quantity of potassium in specific regions of the ovaries (Fig. 2).2 Ovaries from twelve fish were fixed in 96 per cent alcohol and sectioned in paraffin at 10 microns. One hundred and thirty-six odcytes of Period 1, 63 of Period 2A, and 72 of Period 2B 2 Marza and Chiosa (1935) have given the technique for the preparation and use of the standard slides. HISTOCHEMISTRY OF OOCYTES (fy) have been studied. Within these stages of differentiation the odcytes observed have been sorted according to actual size attained. The smallest oocytes contain more potassium in their cytosomes than do larger ones (Figs. 3 and 12, and Table II). Since these cells, which are from 20 to 50 microns in diameter, do not have a continuous follicu- lar layer the concentration of potassium is not conditioned by selective permeability of the follicle (Fig. 1). The nucleoli of such young oOcytes are very rich in potassium but the chromatin contains little. TABLE II Results of tests for potassium, indicated as average percentages. Two hundred and seventy-one odcytes in 12 ovaries were examined. Renton Hercenteee Region Percentage Stroma Odcyte cytoplasm @ytoplasm..... 0. .s cys ss 0.065 RemOd lean eyes 0.160 INICLeUS eer eee ais aie: ene 0.130 Peripheral region eniod 2 Ate se saae 0.087 Theca Period 2 B......... 0.084 Cytoplasm Internal region Period 2 AandB..... 0.032 Reriod 20 Ay Siena. ve 0.087 Nucleus Period 2 B......... 0.109 Period 2 Aand B..... 0.087 Follicular epithelium Yolk Cytoplasm Intravesicular.......... 0.108 Beriod 2.0 ics ee aces — Extravesicular.......... 0.075 Beriod 295A. 335 rhea: 0.066 Reriod’ 2 Bosh ass 0.057 | Odcyte nucleus Nucleus Chromatin Reniodydi tia. sec wines: — Period islta sncducts ues 0.098 Period! 2A. soe 0.109 Period 2 A......... 0.049 Reriod’ 2B) ic fo sie cie,s 0.116 Period’ 2) Baka... 0.065 Nucleoli Zona radiata Period. sive a soe): 0.151 Reriodi2 vA ye. 0.054 Period 2A. oak 0.125 Period 2° Bi... 5:5). 0.057 Reriody2) Bian ou: 0.130 During the second period the theca is somewhat more conspicuous, and the concentration of potassium is not the same in it as in the cells of the stroma (Table II). In the follicle cells the nuclei are richer in potassium than are the cytosomes. No concentration of potassium oc- curs in the zona radiata. With the appearance of vesicles in the odcyte (Period 2 A,) the concentration of potassium in the cytoplasm con- tinues to decrease (Fig. 3). The penetration of potassium does not keep pace with the rate of cytoplasmic increase in the odcyte. This de- 74 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE crease is noticeable, also, in the chromatin and nucleoli. During the early stages of yolk deposition the percentage of potassium in the cyto- 0.2 0.15 \ G of ‘\ omg a a 0.1 a ae : uN 4 eS & bs &e. Seu Se Vea Ee 0.05 ee Cy oplasm ts Mfemenge Peripheral region iat ie Internal region 0 Diameter 0.05 01 0.3 0.6 08 10 in mm. A, A, B, Bo Fic. 3. Variation in percentage of potassium, shown on vertical axis, in cyto- plasm of odcytes during the growth-period. plasm begins to rise, especially in the internal region. Yolk contains a high percentage of potassium which again falls off as the oocyte grows HISTOCHEMISTRY OF OOCYTES 75 (Fig. 12). A similar shift in proportion of potassium during yolk- formation has been observed in the hen’s egg in the ground substance of both white and yellow yolk-spheres. The fact that the theca and follicular epithelium vary so little with respect to potassium content throughout the growth-period is different from the findings in the ovary of the hen (Marza and Chiosa, 1936). There, toward the end of yolk-formation, the percentage of potassium increases in the follicle cells. In the ovary of the fish there appears to be little increased concentration, or temporary storage, in the follicle cells of substances utilized in the formation of yolk. However, there is some indication that the follicle cells determine the rate of penetration of potassium which decreases in amount in the cytoplasm of the odcyte following the association of follicle cells. The follicular epithelium ap- parently undergoes changes in permeability. It seems to be more permeable to potassium during stages A, and B, and only slightly permeable at later stages. That the result may be conditioned by an altered rate of entrance of other materials is also a possibility. PLASMAL REACTION In 1924 two color reactions were reported for cells after the ap- plication of Schiff’s sulfurous-fuchsin reagent. The first or plasmal reaction was observed in the cytosome, and the nuclei were uncolored except when the reagent was applied for a long period (Feulgen and Voit, 1924). If the sections were hydrolyzed with hydrochloric acid before being treated with Schiff’s reagent, the color reaction was in the nucleus ; this was called the nucleal reaction (Feulgen and Rossenbeck, 1924). The names given to these reactions indicate that aldehydes in the cytoplasm or chromatin were believed to be responsible for the appear- ance of the color. Feulgen and Voit (1924) pointed out that the plasmal reaction was related to the presence of fat in the cytoplasm. If the tissue is dehydrated before sectioning the reaction is usually nega- tive. Verne (1929) demonstrated that only unsaturated fats in the process of oxidation gave a positive plasmal reaction. However, fatty substances are not the only ones to react positively. Feulgen and Voit (1924), Voss (1929), and Marza and Marza (1934) have reported an intense plasmal reaction in the elastic elements of large blood vessels, and it occurs in tissue dehydrated after fixation. Many who use this method of Feulgen consider it to be specific for aldehydes. Feulgen and co-workers pointed out that while Schiff’s reagent reacted with alkalis (a red color) and bromine they considered it to be specific for 76 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE aldehydes in cells. Lison (1932) found that the color of the reaction with various aldehydes varied from deep violet to blood red. He con- cluded, therefore, that similar color reactions by substances other than aldehydes should be considered as positive. With this criterion alkalis, aliphatic ketones, certain unsaturated compounds (such as oleic acid), TABLE III Results of plasmal and nucleal reactions, indicated as averages of color intensity; 0.3 is the lowest positive reading. Region Plasmal reaction Nucleal reaction Stroma C@ytoplasmise venenatis es es == = INUIGIENS HAR Rear e. Bite ci Sed — 1.25 Theca @Cytoplasmiaaenie ements oes = = ING CTe tishe caylee Bac ach ots ce — 1.25 Follicular epithelium CGyLOplasnie ne detsts eas ase 213) = = Chromatin is. 3 aes cle cca ees — 1.25 INUCIEOLI YE rae eee oie Gis eee _ 1.25 Nucleoplasm.................. — 0.3 LoOnairadiat ance mone: vain wake: 0.3 0.3 Oécyte cytoplasm Periogiilimee nse cere 0.3 0.4 Peripheral region Reriod) 2yAvand 1B 7.0... - 0.3 0.3 Internal region Period 2 AandB......... 0.3 0.3 Yolk Intravesicular................. 0.75 0.75 Extravesicular................ 0.6 0.5 OGcyte nucleus eri OG lye casa exe sig arate ae = — PEriOd) 2eAG Set es ane htass _ — weak salts of strong bases (such as acetates and phosphates), some amino oxides, and certain catalytic oxidizing systems gave positive color tests. Even though certain of these substances occur not at all or in very minute quantity in normal cells, the presence of alkalis, phosphates, ketones, and oxidases makes it clear that a positive plasmal reaction can- not be considered specific for aldehydes or fats. HISTOCHEMISTRY OF OOCYTES a Voss (1927) first studied the plasmal reaction during the growth- period of odcytes, using ovaries of amphibians. Hibbard (1928) and Brachet (1929) employed it with amphibian ovaries, and Marza and * Plasmal reaction = Nucleal reaction y) j 0.50 SMS UU TMM MMM G PR. LR. LY. Bey; Fic. 4. Plasmal and nucleal reactions in the odcytes. C, cytoplasm of oocytes of Period 1; P. R., peripheral region of cytoplasm of odcytes of Period 2; J. R., internal region of cytoplasm of o6dcytes of Period 2; J. Y., intravesicular yolk; and E. Y., extravesicular yolk. The numbers on the vertical axis refer to the comparison color scale. Marza (1934) with the hen’s ovary. For the reaction in Fundulus, twelve ovaries were fixed in sublimate-acetic as recommended by Feul- gen. Frozen sections were used in order not to lose the fatty sub- 78 V. D. MARZA, EV. MARZAVAND IM: J, GUDEHRIE stances during dehydration. The sulfurous-fuchsin reagent was pre- pared according to Wermel (1927),? and the short technique was fol- lowed: thirty minutes in the reagent, rapid washing in three jars of water turbid with sulfur dioxide, washing in distilled water, dehydra- tion, and mounting. In order to evaluate the reaction qualitatively a scale of color intensity was employed. An intensity designated as 0.3 indicates the least positive reaction and below that value the coloring is the result of adsorption of the reagent; 3.0 is an extremely intense posi- tive reaction. A slightly positive plasmal reaction may be obtained in the cyto- plasm of odcytes of Period 1 (Table III and Fig. 4). In this material the so-called yolk-nucleus, as described by Hibbard (1928) in Discoglos- sus and Marza and Marza (1934) in the hen, is not distinguishable. The reaction of the nucleus of the odcyte is negative at all times except for an occasional faint reaction in a nucleolus. No reaction is ob- served in the ovarian stroma, the theca, or the follicular epithelium at any stage. The interstitial cells give an intense positive reaction. During the second period the zona radiata varies in reaction, ap- parently depending on the material present in its canals at any particular time; it is positive in about half the cases. In the peripheral and in- ternal cytoplasm of the odcyte the reaction is variable with an average on the positive side. Yolk forming within the vesicles gives a more positive reaction than that formed directly in the cytoplasm. Since the reaction is the same in sections dehydrated in alcohol before being treated with the Schiff’s reagent, the substance responsible cannot be fat. The difference in degree of reaction in the intravesicular and extra- vesicular yolk parallels the difference in potassium content. It seems possible that the plasmal reaction in the yolk is conditioned by the pres- ence of the alkali. NucLEAL REACTION The nucleal reaction has been studied in the eggs of numerous in- vertebrates and some vertebrates. Contrary to the usual positive re- action in the nuclei of cells, the nucleal reaction becomes negative in the nuclei of odcytes early in the growth-period and remains negative until the period of diakinesis (Koch, 1925; Voss, 1927; Hibbard, 1928; Lud- ford, 1928; Brachet, 1929; Gresson, 1930; Mukerji, 1930; Bauer, 1932; 3 Dissolve 1 gram of basic fuchsin in 200 cc. of boiling distilled water and filter. (Griibler’s gelblich fuchsin was used.) Add 20 cc. of normal hydrochloric acid. When the solution is cool add 2 grams sodium bisulfite and stir for several minutes. After 2 hours add 0.2 cc. acetaldehyde; the solution becomes intensely red. After 45 minutes again add 20 cc. of normal hydrochloric acid and 1 gram of sodium bisulfite. Stir the solution for 15 minutes and set aside until the fuchsin is decolorized. The pale amber reagent should be kept dark and cool. HISTOCHEMISTRY OF OOCYTES 79 Marza and Marza, 1934). ‘This peculiar situation is characteristic of only female germ cells, and the nucleal reaction is positive during the entire course of spermatogenesis. Ovaries of Fundulus were fixed in sublimate-acetic, and the Feulgen- Rossenbeck technique was followed on paraffin sections. The color scale used for the plasmal reaction records was again employed. The cytoplasm of cells of the stroma and of the follicular epithelium has a negative nucleal reaction (Table III). The nuclear membrane and chromatin, but not the nucleoplasm, of cells of the stroma give posi- tive tests. In the nuclei of the follicular epithelium the reaction is negative in the membrane, faintly positive in the nucleoplasm, and equally strong in the chromatin and in the nucleoli. These reactions do not change conspicuously during the course of differentiation. There is a positive reaction in the cytoplasm of the oocyte during the first period of differentiation which is similar to the reaction without hydrolysis—the plasmal reaction (Fig. 4). It seems likely that this is to be explained on the basis of the potassium content. Even in the smallest oocytes the nucleal reaction is negative in all parts of the nucleus. During the second period of differentiation the zona radiata has a faintly positive reaction in stage B. The cytoplasm at the periphery of the oocyte and between the yolk globules gives a faintly positive reaction. As with the plasmal tests, the yolk gives a positive nucleal reaction of similar intensity. Since hydrolysis does not alter the reaction, it is ap- parent that it is not always conditioned by fat, and that it may be at- tributed to the potassium content. In the odcyte of the hen Marza and Marza (1934) made further tests that led them to the hypothesis that the positive plasmal and nucleal reactions arise from the presence of po- tassium. They found in the yolk of the hen’s egg that the plasmal and nucleal reactions were localized in the centers of. the globules. When the nucleoprotein test (method of Unna, 1921) was positive the locali- zation was peripheral. The reaction for oxidases was negative. Pro- longed extraction with ether and chloroform did not alter the reaction, which eliminated fats as the source of aldehyde. There was no aug- mentation of the reaction after treatment with alcohol; it had been sug- gested that alcohol might be adsorbed by the yolk and partially changed to aldehyde. The nucleal reaction is negative in nuclei of all odcytes during the second period in Fundulus, and no observations were made on the time of shift of the reaction before meiosis. There is no evidence upon which to offer an explanation of this negative test. Koch (1925) and Brachet (1933) attributed it to a chemical change in the nucleic acid 80 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE during the growth-period, while Feulgen (1926) and Ludford (1928) considered it to be the effect of dispersion or dilution of thymonucleic acid as the nucleus of the odcyte increased in volume. Acip PROTEINS In order to study changes in the proteins during differentiation of the oocyte the method of Unna (1921) for acid proteins was used. This procedure enables one to distinguish albumins, globulins, and nucleoproteins. It is based on differences in color reaction and solu- bility between the groups. ‘Tissues are fixed in 96 per cent alcohol, sectioned in paraffin, and stained for twenty minutes in the Pappenheim- Unna reagent.* Albumins and globulins are stained red by the pyronin, while the nucleoproteins are stained blue-green with the methyl green. To separate the albumins and globulins advantage is taken of their specific solubilities. Albumins and pseudoglobulins are soluble in dis- tilled water while globulins are not. Both albumins and globulins are soluble in salt solutions. By comparison of sections stained before and after chromolysis in water and in salt solution the location and relative proportions of acid proteins of the three groups can be determined. Sections of the ovary of Fundulus were divided into three series. One series was stained directly. In such sections all acid proteins were shown. A second series was placed in sterile distilled water in an oven at 39° C. for twenty-four hours before staining. This was the first chromolysis. The difference in reaction with pyronine between these sections and those of the first series was conditioned by the loss of albumins (cf. Figs. 16 and 18). A third series was placed in a sterile 2 per cent solution of sodium chloride in an oven at 39° C. for twenty- four to forty-eight hours before staining. This was the second chro- molysis which removed both albumins and globulins. If nucleoproteins were present in such sections they would then be stained with the methyl green (cf. Fig. 17 with Figs. 16 and 18). To obtain results subject to comparison, sections of the three series were handled simul- taneously by an arrangement of forceps during the staining and dehydra- tion. The latter process had to be completed rapidly. Results were read on a standard color scale in which 0.3 was the least positive reaction and 4.0 the most intense. Two hundred and sixteen oocytes have been studied in sections from twelve ovaries. Two varieties of interstitial cells are demonstrated with Unna’s technique. Those which are pigmented, occur in masses, and contain 4The Pappenheim-Unna reagent used consisted of 0.15 gram of Grubler’s methyl green (from which admixed methyl violet was removed by shaking the crystals in chloroform), 0.25 gram of pyronine, 2.5 cc. of 96 per cent alcohol, and 100 cc. of 0.5 per cent phenol. Se ee ee eee ee ee yo a gee pig rede HISTOCHEMISTRY OF OOCYTES 81 iron do not color with the stains used. Cells of the other type have variable shapes, with pseudopodia-like histiocytes or simple elongations like fibroblasts, or may be oval with eccentric nuclei somewhat similar 3 =.= Glebulins / Albumins yi SS 2.9 1.5 0.5 0 : - 0.3 05 08 1.0 1.6 Diameter in mm. A A, B, Bo Bs Fic. 5. Variation in acid proteins in the cytoplasm of follicle cells. The numbers on the vertical axis refer to the comparison color scale. to plasmocytes. Such cells have been described in the ovarian stroma of the hen by Goodale (1919), Nonidez (1921), Marza (1934), and Marza and Golaescu (1935); the analogies are not completely clear. 82 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE These cells in Pundulus are found in the endothelium of the blood vessels, isolated in the stroma, in the tunica albuginea, and, rarely, within the masses of pigmented cells. The cytoplasm of the non-pigmented interstitial cells stains intensely with the pyronine. 0.3 0.6 08 1.0 16 A, As B, Bo B, Fic. 6. Variation in acid proteins in the cytoplasm of oocytes in Period 1 and in the peripheral region of cytoplasm in Period 2. Designations as in Fig. 5. Diameter in mm. Only a small quantity of irregularly localized globulin is found in the cytoplasm of cells of the ovarian stroma, and neither globulins nor HISTOCHEMISTRY OF OOCYTES 83 albumins occur in the nuclei which are, however, rich in nucleoproteins (Table IV). Both globulins and albumins are found in the cytoplasm of thecal cells during Period 2A; the amount is somewhat reduced 2.0 ‘ \ 15 3 ‘ \ a ere 1.0 0.5 a) Diameter in mm. 0.6 0.8 1.0 A, B, Bo Fic. 7. Variations in acid proteins in the internal regions of cytoplasm in oocytes of Period 2. Designations as in Fig. 5. during Period 2 B. The nuclei of thecal cells are rich in nucleoproteins. Concentrated masses of globulins are seen localized along the length of the cytosomes of follicle cells during Period 1 (Fig. 5), but the ground 84 WS IDS IWR ZG, Ti) WG WIDER. JANIIID) IES jf, (EAU AD Teale cytoplasm of follicle cells gives only a slight reaction (0.5). The amount of globulins in the follicle is increased conspicuously during Period 2, and albumins appear in stage 2B. These albumins may occur because of the splitting of globulins in the follicle as penetration into the odcyte begins. Globulins appear in the chromatin during stage 2 B and are found, together with nucleoproteins, in the nucleoli through- out the growth-period. No acid proteins have been detected in the zona radiata at any time. EXPLANATION OF PLATE I Microphotographs Frc. 8. Small odcyte (Period 1) after microincineration, showing much white ash in cytosome and some in nucleus. X 150. Fic. 9. Odcytes of Period 2 A after microincineration. Ash is visible in the cytoplasm but not in the vesicles. X 30. Fic. 10. Odcyte of Period 2 B: after microincineration. White ash occurs in the region of the cytoplasm and gray ash in the intravesicular yolk which is pres- ent. The follicle appears two-layered because of the plane of the section. X 56. Fic. 11. Odcyte of Period 2 Bs after microincineration. The homogeneous central mass of yolk is somewhat shrunken from the peripheral cytoplasm and its ash content is visible. X 30. HISTOCHEMISTRY OF OOCYTES Aion aN; Results of tests for acid proteins, indicated as averages of color intensity; 0.3 is the lowest positive reading. Where a range is given it indicates a shift in content between the smaller and larger cells of a given period. 85 Regions Albumins Globulins Nucleoproteins Stroma Gy toplasiiews sees set Se seit — 0.8 — Ghromeatinee ss ee eke i ess ees = = 25 INFRUGISONI es aides toes 5 ctsleen sean eineemen — = 3.0 Theca Cytoplasm BETO GQ w Aten yee ees (see 0.4 0.6 — Period 2° Bes... 56 5.42 Steet — 0.6 = Chromatin Period 2 A ame 1B. sss sen oo — — 2.0 Follicular epithelium Cytoplasm Remiodelisess Us sstess) sted see — 1.1 = Period 2 Aj-As............. — 2.2-2.3 — enmiOdiwwe Bie cates a eae 0.6 2.9 — Period 2 Boa=B3.:........-.. 1.0-0.5 2.7-2.4 = Chromatin Re reOclbleamee see cl create tea — — 2.5 TRYSSEOyGI EAU Ge ae ee oe nee — — 3.0 Beriode2i Bie. ics eee cs — 3.0 3.0 Nucleoli Reriocdswliance?) sae ee ee -~ 3.0 3.0 Oécyte cytoplasm ReTiOd ee teins ee An eels — 2.0—4.0 — Peripheral region RETIOG eee AG tsk ete ater ae — 3.4 — PenodetAcie sack ho ee 0.4 2.4 — RETO CED Brits nee cs et nsteess 0.4 1.4 — Period 2 B:-B3............. 0.6-0.4 1.4-0.7 — Internal region Periodt2 Anew caer ken ene 0.7 2.4 -= Reriod=2"Bie <8 cc weak 0.5 1.1 — Reriode2 Bate iy ccs oor 0.6 1.1 — Yolk Intravesicular REGiOUE2R Bias, a: Sats eks oe — 0.75 0.5 Pernod 245 en nace sees — 0.5 0.5 Extravesicular PeniOdeZ Borgo ce can eee — — 0.5 Oécyte nucleus Chromatin Remodsp lean een ee 0.3 0.5 — Nucleoli IRtesrnoals il @inGl 7. 2 ocobancco. — 3.5 — 86 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE EXPLANATION oF PLATE II Microphotographs Fic. 12. Group of odcytes, showing variation in potassium content; method of Macallum. X 41. Odcyte 1 is in Period 1 and has a high potassium content. Oécyte 2 is in Period 2 B:; 3 is in Period 2 B2; and 4 is in Period 2 B: (4 is about 1,200 microns in diameter). Note the decrease in potassium content in the yolk in oocytes 3 and 4. Fic. 13. Group of odcytes showing variation in total acid proteins; method of Unna, without chromolysis.. X 17. Odcyte 1 is in Period 2 A, and gives an intense reaction in the cytoplasm; odcyte 2 is in Period 2 Bi; 3 is in Period 2 B:; and 4 is in Period 2 Bs with a central homogeneous mass of yolk (4 is about 1,200 microns in diameter). Note shift in reaction in yolk. Fic. 14. Group of odcytes showing variation in total acid proteins; method of Unna, without chromolysis. X 41. Odcytes 1, 2, and 3 are in Period 2 A, with numerous vesicles; and 4 is in Period 2 B». Note shift in reaction in cytoplasm. Fic. 15. Portion of odcyte early in Period 2 Bz showing variation in acid proteins; method of Unna, without chromolysis. X 204. Note larger hetero- geneous vesicles containing some intravesicular yolk and the much smaller globules of extravesicular yolk. inn pr HISTOCHEMISTRY OF OOCYTES 87 Acid proteins with a globulin reaction are conspicuous in the cyto- plasm of the smallest oocytes and reach a very high concentration by the end of Period 1 (Fig. 6). They are uniformly distributed in the cytosome (Figs. 16 and 18). No albumins or nucleoproteins are ob- served at this stage (Table 1V). As vesicles appear and increase in number the peripheral layer of cytoplasm has a decreasing amount of globulins (Figs. 6, 13, 14, and 15). Again we see in the oocyte of Fundulus during the first period a conspicuous accumulation of sub- EXPLANATION OF PLateE III Microphotographs Fic. 16. Odcytes showing intense acid protein reaction; method of Unna, without chromolysis. X 140. Smaller odcyte is in Period 1, larger is in Period 2 A: Fic. 17. Oocytes in Period 2 A:, showing the absence of globulins; method of Unna, after chromolysis in salt solution. X 140. The failure to stain with methyl green indicates the absence of nucleoproteins. Fic. 18. Odcytes showing globulin content; method of Unna, after chromoly- sis in water. XX 140. Smaller odcyte, of which only a part is seen, is in Period 1; larger is in Period 2 Ax. stances before the actual synthesis of yolk begins. Albumins appear toward the end of stage 2 A and are found in small quantities until the end of yolk-formation (Fig. 6). In general there is less acid protein in the internal cytoplasm, but the albumins are somewhat more abundant (Fig. 7). It would appear as if proteins adjacent to the places of yolk deposition were chiefly in the form of simple albumins. However, in the yolk deposited within the vesicles some globulins are found but no albumins, while in extravesicular yolk neither can be demonstrated (Fig. 15). The synthesis of simple acid proteins into ichthulin and its combination with other groupings is apparently complete and not de- 88 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE stroyed by the method of treatment. This differs from the findings in the hen’s egg where Marza (1935) found that globulins and pseudo- globulins were detectable in the fully differentiated yolk. The yolk in Fundulus contains an appreciable amount of nucleoprotein, much more than the yolk of the hen’s egg. Konopacka (1935) has reported the presence of nucleoproteins in the yolk of two fresh-water fishes; she has also identified a mucoprotein in the vesicles of Period 2A. The presence of nucleoproteins in eggs has stimulated considerable discussion. From an analysis of the ichthulin of various fish eggs Walter (1891), Hammarsten (1905), and Linnert (1909) concluded that there was no nucleic acid present. However, Konig and Grossfeld (1913) and others have isolated small quantities of purine bases or nucleic acids from the whole eggs of several fishes. The eggs of animals which develop in an aquatic environment contain the constituents of nucleoproteins, while the eggs of terrestrial animals are very poor in them. Needham (1931) has considered these facts and their im- plications. The histochemical method of Unna (1921) reveals the pres- ence of nucleoprotein in the growing odcyte of Fundulus which con- firms the chemical findings in other teleosts. It may be that the positive Feulgen-Rossenbeck (1924) reaction in yolk is conditioned in part by nucleoprotein, but the presence of potassium in the same region confuses the interpretation. The nucleus of the odcyte does not give a nucleoprotein reaction at any time (Table IV). Both globulins and albumins occur in the chromatin masses. Nucleoli are rich in acid proteins, but only globulins are present. In the largest nucleoli the globulin reaction at the centers is 2.0 and at the peripheries, 4.0. Smaller nucleoli give a homogeneous reaction like the cortices of the larger ones. The absence of nucleo- protein has been considered in the discussion of the nucleal reaction. Unna’s method does not provide data on which a decision can be made concerning the reason for the absence of the nucleal reaction in the nucleus of growing oocytes. THe FoLiicuLar EPpitHELIUM AND ZONA RADIATA It is very striking that the concentrations of the substances recorded in this paper shift conspicuously in the odcytes during the growth-period. This might be a result of changing resources in the blood stream, or of a selective action of the follicle and zona, or, in some cases, of synthetic activity within the odcyte in which combination altered the quantity of a substance free to react with the testing reagent. HISTOCHEMISTRY OF OOCYTES 89 Since variations in height of epithelial cells have been taken in many cases to indicate degree of metabolic activity, measurements of the width of the follicle were made. At the same time the zona radiata 22 ie Zona Radiata 20 Follicular Epithelium / 18 16 14 12 10 0.1 02 03 04 05 06 07 08 09 10141 12 13 1.6 Diameter — in mm. A, A; B, Ba Bs Fic. 19. Variation in height of follicular epithelium and zona radiata of the oocyte during growth-period. The height in microns is indicated on the vertical axis. was measured (Fig. 19). The smallest odcytes, from 10 to 50 microns in diameter, do not have a continuous covering of follicle cells, al- though some squamous cells may be associated with them (Fig. 1). 90 V. D. MARZA, E. V. MARZA AND M. J. GUTHRIE When the follicle is complete, in odcytes from 100 to 150 microns in diameter, its cells do not exceed 1.5 microns in height. During this period the cells of the follicle exercise no inhibitory effect upon the entrance of substances into the oocyte. The quantity of potassium and acid proteins is relatively very great. Maximum growth and differentiation of the follicle occurs during Period 2 A, and the height of the cells rises to 11 microns in 2 B,. The changes in potassium and protein content have been noted. Obviously chemical changes are correlated with the differentiation of the follicle cells. During Period 2 A the large vesicles which are the earliest indi- cation of the unique differentiation of the odcytes appear, first at the periphery and then throughout the cytosome. One may assume that no longer are the constituents of cytoplasm being delivered in the same relative proportions. Certain substances are accumulating in excess of others. That this is conditioned by the composition of the follicle cells is a simple assumption suggested by the evidence at hand. When the deposition of yolk is occurring rapidly and directly in Period 2 B the follicle undergoes slight change. Some decrease in height is observed as the oocyte reaches a maximum diameter, and nuclear regression is apparent. The amount of albumins is lowered during this stage. In contrast to the follicular epithelium, the zona radiata continues to increase in thickness until the end of the growth-period. Some sub- stances can be identified in its canals, but obviously nothing is stored in it. The observations made during this investigation do not suggest that the zona radiata functions in the selection of materials entering the oocyte. SUMMARY The plasmal and nucleal tests and the histochemical methods for detection of iron, potassium, and acid proteins indicate that changes in quality, quantity, and localization of substances characterize the period of differentiation in the odcyte of Fundulus. The oocyte at the end of its growth-period is morphologically very different from the oogonium. Its structural changes are found to be correlated with the metabolic activities leading to food storage. Cytosomal differentiations are con- spicuously chemical phenomena in egg-cells. Die RAL URE Clip. Bauer, H., 1932. Die Feulgensche Nuklealfarbung in ihrer Anwendung auf cy- tologische Untersuchungen. Zeitschr. f. Zellf. u. mikr. Anat., 15: 225. BrataszEwicz, K., 1929. Recherches sur la répartition des électrolytes dans le protoplasme des cellules ovulaires. Protoplasma, 6: 1. HISTOCHEMISTRY OF OOCYTES 91 BracHeT, J., 1929. Recherches sur le comportement de l’acide thymonucléinique au cours de l’oogenése chez diverses espéces animales. Arch. Biol., 39: 677. BracHeET, J., 1933. Recherches sur la synthése de l’acide thymonucléique pendant le développement de l’oeuf d’oursin. Arch. Biol., 44: 519. FEuLceEn, R., 1926. Die Nuklealfarbung. Abderhalden’s Handb. der biol. Arbeits- methoden, 5 (2): 1055. FEuLGEN, R., AND H. RossEenBEcK, 1924. Mikroskopisch-chemischer Nachweis einer Nucleinsaure vom Typus der Thymonucleinsaure und die darauf beruhende elektive Farbung von Zellkernen in mikroskopischen Praparaten. Zettschr. f. physiol. Chem., 135: 203. FreuLcen, R., anp K. Vorrt, 1924. Uber einen weitverbreiteten festen Aldehyd. Pfiiiger’s Arch. f. d. ges. Physiol., 206: 389. GoopaLe, H. D., 1919. Interstitial cells in the gonads of domestic fowl. Anat. Rec., 16: 247. Gresson, R. A. R., 1930. Certain phenomena of Tenthredinid oogenesis as re- vealed mainly by Feulgen’s nuclear-reaction. Quart. Jour. Micr. Sct., 73: 617. Guturiz, Mary J., 1925. Cytoplasmic inclusions in cross-activated eggs of teleosts. Zeitschr. f. Zellf. u. mikr. Anat., 2: 347. GuturizE, Mary J., 1928. Further observations on the ovarian eggs of Fundulus. Anat. Rec., 41: 64. Guturiz, Mary J., 1929. Cytoplasmic inclusions in the oocytes of vertebrates. Anat. Rec., 44: 231. Hammnarsten, O., 1905. Zur Chemie des Fischeies. Skand. Arch. f. Physiol., 17: 113. Hispard, Hore, 1928. Contribution a l’étude de l’ovogenése, de la fécondation, et de l’histogenése chez Discoglossus pictus Otth. Arch. Biol., 38: 251. Kocu, A., 1925. Morphologie des Eiwachstums der Chilopoden. Zeitschr. f. Zellf. u. mikr. Anat., 2: 293. Konic, J., AND J. GrossFetp, 1913. Der Fischrogen als Nahrungsmittel fur den Menschen. Biochem. Zeitschr., 54: 351. Konopacka, B., 1935. Recherches histochimiques sur le développement des pois- sons. I. La vitellogénése chez le goujon (Gobio fluviatilis) et la carpe (Cyprinus carpio). Bull. Acad. Polonaise Sci. et Let. (sér. B, II), p. 163. Konopackt, M., anv B. Konopacxka, 1926. La micromorphologie du metabolisme dans les périodes initiales du développement de la grenouille (Rana fusca). Bull. Acad. Polonaise Sct. et Let. (sér. B.), p. 229. LiesecANc, R. E., 1923. Nachweis geringer Eisen- und Kupfermengen in Leinen, Papier oder tierischen Geweben. Zeitschr. f. wiss. Mikr. u. mikr. Tech., 40: 14. Linnert, K., 1909. Enthalt Kaviar (Stor- resp. Hauseneier) Purinbasen? Bio- chem. Zeitschr., 18: 209. Lison, L., 1932. Sur la spécificité du réactif de Schiff envers les aldéhydes. Bull. d’ Histol. Appl., 9: 177. Luprorp, R. J., 1928. Studies in the microchemistry of the cell. I. The chromatin content of normal and malignant cells, as demonstrated by Feulgen’s “nuclealreaktion.” Proc. Roy. Soc. London, B, 102: 397. Macatium, A. B., 1905. On the distribution of potassium in animal and vegetable cells. Jour. Physiol., 32: 95. Macatium, A. B., 1912. Die Methoden der biologischen Mikrochemie. heal LE tae eae 1,617 1,069 66 Vig Date Noy a ae ect Gas 1,973 1,164 59 UAE CO AUACAG es Gaga Cm are Ee ete angie 4,149 1,897 46 leone S< ISS col bobo oo bos 640 465 55 The significant feature of the results of Table II is the parallelism shown between the fertilities and the productivities of Table I. Thus the fertility of the Falmouth inbred stock was 59 per cent which was raised to 71 per cent and 73 per cent when these same females were outcrossed to vestigials or to heterozygotes. Similarly, the fertility of the vestigial females was raised from 41 per cent to 66 per cent and 59 per cent when outcrossed to Falmouth and to heterozygous males— the increase in fertility being similar in amount to the increases in pro- ductivity and presumably accounting for their occurrence. The fertility of the heterozygous females mated to wild males was markedly better (82) than that of the wild females (59) or vestigial females (66) mated to wild males. Finally, the fertilities of the F, extracted vestigials (55) was better than that of the inbred stock (46) and was matched by the increased productivity noted in Table I. Hence, it may be concluded that the primary genetic difference between the vestigial and wild stock (the vestigial gene) was responsible for the curtailment of productivity of vestigial females to approxi- PRODUCTIVITY AND FERTILITY DROSOPHILA MUTANTS 217 mately half that of wild through its determination of a phenotype not able to function for much more than half the normal span of life. The two original stocks differed in secondary gene pairs which (1) height- ened the productivity of crosses by improved viability of the hybrids in the egg stages, (2) heightened the productivity of hybrids (heterosis effect) and which (3) were able to redistribute with respect to the vestigial pair and bring about the improved fertility and consequent improved productivity of the F, extracted vestigials. FERTILITY OF THE HIGH-PRODUCING “ Harry” STRAIN Unusually high production had been noted for a few strains, notably for the heterozygotes of hairy. A study of the fertility of hairy (Table III) showed that the outcross of hairy by Falmouth gave an increased fertility of 64 per cent as compared with the 59 per cent of Fal. X Fal. and the 60 per cent of hairy < hairy. The hybrids them- selves gave an even higher fertility of 73 per cent corresponding to the unusually high productivity originally observed. ‘Cassie JOM Fertility of hairy and wild ©) 3s Ct Eggs Adults Fertility : per cent Epa all gees she Peat 3,505 2,080 59 IR eo Eva eee ut ta a ili a 1,157 743 64 J? SC lnk en cer eRe Bee ee ae ane On 1,706 1,032 60 States ose a) Osea evens 679 494 73 TG OS ONL ay erry perder 4,527 2,747 61 Delpy OV Sa) saa feo rath ee ere coe 2,625 1,459 56 Tue FERTILITY RELATIONS OF SINGED AND SINGED? Among the mutant strains at Columbia there were three which showed complete or nearly complete sterility of the females but for which there existed allelomorphic strains similar in phenotypes but with approximately normal fertility of the females. One such set of allelomorphs was singed and singed’, sex-linked recessive found by Mohr. In both mutants, the bristles and hairs are tightly curled or crumpled, but singed* has normal fertility while singed females lay few eggs all abnormal in shape and filaments. As shown in Table IV, the abnormal eggs laid by young singed females averaged only about 7 per day as compared with 20 per day 218 GAIL L. CARVER for singed* females and 25 per day for their hybrids. None of the eggs of the singed females gave offspring while the fertility of the eggs of singed* females was 33 per cent with brothers and 53 per cent in crosses to singed males. Fertility of the hybrids was 38 per cent and 37 per cent when mated to singed and singed* males—better than that of either pure race. TABLE IV Fertility of singed and singed 3 >< Cf Eggs Eggs Daily Adults Fertility per cent STU SGESTI Gn eer Medan te 754 6.6 0 0 SI MUSTO ea sian he nees 1,023 7.8 0 0 STIG OC STION EEE ettn aa: 3,272 20.4 1,085 33 SNUG ISH eae eee 3,269 19.2 1,718 53 SYST SK Stso esol sace 2,903 24.4 1,093 38 sn/sn? X sn?.......... 3,034 25.9 iL il?) 37 FertTItiry RELATIONS OF REDUCED AND SCRAGGLY The second—chromosome-recessive allelomorphs “reduced” (rd) and “scraggly”’ (rds) both found by Bridges, are characterized by irregular and variable reduction in number and size of the bristles. Scraggly is of normal fertility but females of reduced lay only a few eggs which almost never give offspring though they are normal in appearance. AV ABER OVS Fertility of reduced and scraggly SS Ct Eggs : Adults Fertility per cent TiC Clase ats evil Role a eat ge 460 0 0 TROL CEG Ey 6S) a ee ee 540 4 1 == /fidle> ss. eae 0 0 9 SSLHA N23 ce nM PER ERTE cooh Ba Asc ie 0 0 * The animals were decapitated on 4/23. They continued to produce capsules to 5/12. Comparison of the extremes of the series, the liver-fed with the starved and the extract-treated forms, leads one to believe that low cap- sule production may be due to a food deficiency. Of the starved and extract-treated forms which were becoming smaller, only Liver Aqueous Extract L was productive—a single capsule was produced. ‘The animals in this set were slightly larger than the starved and other extract-treated animals, and smaller than any of those which were fed fresh glands. The thyroid-fed animals, which were least productive of the gland- ENDOCRINE FEEDING AND REGENERATION 235 fed animals, were in good condition and increasing in size but were smaller than those on the liver diet. Although the planarians fed an- terior pituitary and anterior pituitary-thyroid glands approximated the liver-fed ones in all dimensions, they produced a smaller number of egg capsules than the liver-fed individuals. It is realized that the evidence is too scanty to permit a definite con- clusion. It is interesting to note that Greenberg and Schmidt (1936) described an ether-soluble factor in liver which acts as a growth- promoting agent for Planaria maculata. Smith and Seegers (1934) found a principle in liver which acts as a growth-promoting agent and which is, in some way, concerned with the typical functioning of the reproductive mechanism in the albino rat. The writer is indebted to Professor Robert Chambers for his criti- cism of the manuscript and to Professor F. Gudernatsch for his counsel throughout the course of this work. SUMMARY 1. Individuals of Planaria maculata were fed on abundant, exclusive diets of beef liver, anterior lobe of the pituitary and thyroid glands. Others were given aqueous extracts of the glands or were completely starved. 2. No significant differences were noted in the head regeneration time of the gland-fed animals following each of five amputations of the anterior region. 3. Decapitated animals which were starved and those which were kept on the aqueous extract diet and which showed starvation symptoms regenerated more slowly than those which were fed the fresh glands. There appeared to be a correlation between the rate of regeneration and the initial size of the animals. 4. Thyroid-fed individuals increased in size but to a lesser extent than the liver and pituitary-fed forms. All were in excellent condition following four decapitations. Individuals newly emerged from capsules also increased in size when fed thyroid gland exclusively. 5. Liver-fed animals produced a greater number of egg capsules than any of the others. The starved and extract-treated forms, with the exception of a single capsule by the Liver L Aqueous Extract set, pro- duced no capsules. 236 E. D. GOLDSMITH BIBLIOGRAPHY CastLeE, W. A., 1928. An experimental and histological study of the life-cycle of Planaria velata. Jour. Exper. Zool., 51: 417. GotpsmitH, E. D., 1935. Regeneration and growth in Planaria maculata under the influence of endocrine feeding. Anat. Rec., 64 (Supp. No. 1): 78. GREENBERG, L. D., ann C. L. A. Scamp, 1936. Studies on the properties of a growth-promoting factor for Planaria maculata. Jour. Exper. Zool., 73: 375. Morcutts, S., 1923. Fasting and Undernutrition. E. P. Dutton Co., N. Y. SmitH, H. G., ann W. H. Seecers, 1934. The nutritive value of animal tissues in growth, reproduction, and lactation. J. Alcohol extracted beef liver. Jour. Nutrition, 7: 195. II. The presence of a new dietary principle in liver. Jour. Nutrition, 7: 209. Stevens, N. M., 1901. Notes on regeneration in Planaria lugubris. Arch. f. Entw.-mech., 13: 396. Wut1zen, R., 1923. A study in the nutrition of an invertebrate, Planaria maculata. Univ. Calif. Pub. Physiol., 5: 175. RESON Ss OF MUSCLES OF THE SOUID TO REPETITIVE STIMULATION OF THE GIANT NERVE FIBERS CHEADD PRO SSE Re AND) a OLIN 2. YOUNG (From the Marine Biological Laboratory, Woods Hole, Mass.) INTRODUCTION Investigation of the response of muscles to repetitive stimulation of their nerve fibers has shown that there are great variations among dif- ferent muscles and different animals in the possibilities of facilitation at the neuromuscular junction. Although the response of an intact muscle fiber, normally activated, is probably always maximal (all-or- nothing), yet in some muscles, for instance those of Crustacea, increase in the frequency of stimulation often increases the tension developed on account of the fact that, some muscle fibers are not activated by single, or even by few impulses. On the other hand, in vertebrate striped mus- cle, unless drugged or fatigued, a single nerve impulse excites all the muscle fibers which it reaches (Lucas, 1909; Adrian and Lucas, 1912). The range of muscles which have been investigated from this point of view is still small, especially among invertebrates, and we have accord- ingly investigated the muscular response to repetitive stimulation of the giant axons in the stellar nerves of the squid, Loligo pealii. These fibers have been shown to innervate the circular muscle fibers of the mantle, and a single condenser discharge, unless of great intensity or duration, sets up in the axon a single impulse which is capable of ac- tivating all of the muscle fibers which it reaches (Young, 1937). The great stellar nerve, containing a single giant axon, was prepared for stimulation in the manner described elsewhere (Young, 1937), and the contractions of a portion of the muscles which it innervates were re- corded by attachment to a semi-isometric lever. Repetitive stimulation was applied by condenser discharges through a thyratron circuit. With the intensity constant and supraliminal the frequency was varied up to approximately seventy stimuli per second. Fatigue set in very rapidly at higher frequencies. Sixteen experiments in which the giant fiber of the great stellar nerve was stimulated yielded consistent results. 1 Assisted by a Fellowship of the Rockefeller Foundation. 237 238 C. LADD PROSSER AND JOHN Z. YOUNG RESULTS A typical experiment is shown in Fig. 1. Development of tension is registered by a downward deflection in these records. The first stim- ulation was at 13 stimuli per second and the second at 24 per second. Incomplete relaxation between twitches occurred in both, and the max- imum tension at 24 per second was very slightly greater than at 13. GE - am NM Ol (e%) Ol |! im | a | 5 (e) Fic. 1. Record of an experiment in which the giant fiber of the great stellar nerve was stimulated repetitively and the contraction of the mantle muscle was recorded semi-isometrically. Records in order in which they were taken. Num- bers indicate frequencies of stimulation. With stimulation at 4 per second there was complete relaxation between stimuli. In general, complete relaxation between twitches accompanies frequencies of stimulation up to approximately 8 per second. The prep- aration was then stimulated at 20 and at 25 per second, and showed very slightly increased tension. At 35 stimulations per second fusion of contraction was complete. At 50 per second, as at 35, there was no in- crease in maximum tension above that at 25 but at 50 the tension de- clined rapidly, indicating fatigue. The effects of this fatigue were shown when stimulation was returned to 20 per second, where a lower tension was recorded. Thereafter (6 and 40 stimuli per second) the tension varied with the frequency. Figure 2 shows similar results in three other preparations. In each of these experiments, as in most of the others, there was a slight increase RESPONSES SQUID MUSCLE TO STIMULATION 239 in tension (5-10 per cent) as the frequency increased during the range of incomplete relaxation, an effect which may be ascribed to the mechani- cal properties of the muscle. 25 5O 15 Frequency Fic. 2. Plots of maximum tension developed by the mantle muscle against frequency of stimulation of the giant fiber in the great stellar nerve in three prep- arations. Stimulation proceeded from low to high frequencies. In C the lowest point (15 per second) represents response after fatiguing at highest frequency (75 per second). In Experiment 4, Fig. 2, the periods of stimulation were brief and there was no increase in tension with increasing frequency of stimula- tion; this is the general result when no fatigue occurs. Preparation B showed fatigue at 33 stimulations per second, and Preparation C at 75 per second. Thereafter the tension fell off and 240 €: LADD PROSSER AND JOHN,Z: YOUNG varied with the frequency of stimulation, the higher frequencies eliciting a greater response than the lower. The preparation is extremely sensitive to strong excitation and high frequency stimulation during one second causes irreversible fatigue. The failure of the response is parallelled by a growing opacity of the mantle. Discussion It is evident that there is normally no facilitation at the junction be- tween the endings of the giant fiber and the muscles. Increased fre- quency of stimulation produces no increase in tension, indicating that all the muscle fibers are activated by each single impulse which reaches them. Young (1937) observed similar results with increasing intensity of stimulation. In the state of fatigue, however, changes occur, prob- ably at some stage in the contractile mechanism, so that summation oc- curs and greater tensions are produced at the higher frequencies. This condition is closely similar to that found in the striped muscles of the frog (see Adrian and Lucas, 1912), but contrasts sharply with that of Crustacea, where a single impulse often elicits no mechanical re- sponse (Pantin, 1934, 1936). Thus Katz (1936) found that the tension produced by the flexor muscle of the claw of Maza increases nearly ten times when the frequency of stimulation is raised from 50-200 per sec- ond. Recent observations by Mr. Grossman in the Physiology Course at the Marine Biological Laboratory indicate that the tension developed by the claw of Limulus increases from 25 to 650 grams with a rise of frequency from 1-50 per second. The absence of such facilitation at the neuromuscular junctions of the giant nerve fiber system of the squid is correlated with the function which the system serves in the animal, namely to produce the contrac- tions by which a jet of water is expelled suddenly from the mantle. Once the contraction has occurred the mantle cavity must enlarge again before further work can be done, and there would be no use for sus- tained or gradually increasing tensions. The expulsion of each jet of water is a single unitary act, performed in an all-or-nothing manner, and any gradation in speed or distance of propulsion must be obtained by variation in the number of contractions set up. SUMMARY With increasing frequency of stimulation of a giant nerve fiber in the squid, Loligo pealti, the only increase in the tension developed by the circular muscle fibers of the mantle is a small amount (5 to 10 per cent) over the range of incomplete relaxation. The absence of any increased RESPONSES SQUID MUSCLE TO STIMULATION 241 response at higher frequencies shows that in the fresh muscle a single nerve impulse is capable of activating every muscle fiber which it reaches. However, the isolated muscle very readily becomes fatigued when stimulated at high frequency and thereafter greater tension is produced at the higher rates. In the normal animal there would be no use for peripheral facilitation and each contraction of the mantle is produced as an all-or-nothing response. REFERENCES ApriANn, E. D., ann K. Lucas, 1912. On the summation of propagated dis- turbances in nerve and muscle. Jour. Physiol., 44: 68. Katz, B., 1936. Neuro-muscular transmission in crabs. Jour. Physiol., 87: 199. Lucas, K., 1909. The “all-or none” contraction of the amphibian skeletal muscle fibre. Jour. Physiol., 38: 113. Pantin, C. F. A., 1934. On the excitation of crustacean muscle. I. Jour. Exper. Biola s ll Pantin, C. F. A., 1936. II. Neuromuscular facilitation. Jour. Exper. Biol., 13: Tae Pantin, C. F. A., 1936. III. Quick and slow responses. Jour. Exper. Biol., 13: 148. Youne, J. Z., 1937. The functioning of the giant nerve fibres of the squid. In press. THE OCCURRENCE OF SAPROPHYTIC FUNGI IN MARINE MUDS F. K. SPARROW, JR. (From the Woods Hole Oceanographic Institution1 and the Botany Depariment of the University of Michigan 2) INTRODUCTION Since it has been recognized for many years that fungi play an im- portant and significant role in the disintegration of organic materials in land soils, it is natural to suspect that they might perform a similar function in sea bottoms. Previous papers (Petersen, 1905; Sparrow, 1934, 1936) have shown that in the littoral of certain localities in north- ern Europe and eastern United States there are true marine fungi which are active in initiating the destruction of living, autophytic marine plants and in certain cases also in aiding in their disintegration. As no sys- tematic study of off-shore localities for the presence of wholly sapro- phytic fungi in the muds had been reported, the preliminary investigation described in this paper was undertaken during July-August, 1936, at the Woods Hole Oceanographic Institution. STATIONS The following are the locations of stations selected for study and from which the mud cores were obtained. All were in localities marked “sticky ” or “sand and mud” on the hydrographic charts. Station 1. One-half mile N.W. of Weepecket Rock Buoy, Buzzard’s Bay. Depth: 18.0 meters. Station 2. Western entrance to Vineyard Sound, 314 miles E. of Vine- yard Sound Whistling Buoy, 514 miles W. by S., % S. on Gay Head Light. Depth: 32.7 meters. Station 3. Gulf of Maine, 45°35’ N., 69°11’ W. (Atlantis Station No. 2640). Depth: 163.6 meters. Station 4. Gulf of Maine, 42°19’ N., 69°20’ W. Depth: 220 meters. Portions of two cores collected by Dr. Henry Stetson from the Gulf of Maine and labelled “ Canyon B” (depth 1127.2 meters; Station No. 5, Table I) and “ Canyon E”’ (depth 718.0 meters; Station No. 6, Table 1 Contribution No. 73. 2 Contribution No. 630. 242 SAPROPHYTIC FUNGI IN MARINE MUDS 243 I) were obtained from Dr. S. A. Waksman. Since these had previously been partly used for bacteriological purposes and had been stored for some days they were considered questionable sources of data. METHODS AND RESULTS In order to obtain any significant information in the limited time available, the methods outlined below were considered most practical even though they were subject to very definite limitations. Collection of Cores The apparatus commonly used in the collection of stratified mud cores for bacteriological purposes was employed. This is a modification of the instrument used by Moore and Neil (1930) in the Clyde Sea area. To eliminate as far as possible contamination during collection, all parts of the sampling apparatus in contact with the sterilized glass tube were thoroughly swabbed with 10 per cent formalin immediately before use. However, trials with swabbed and unswabbed apparatus showed little difference in the number of colonies of fungi obtained. All the usual precautions were employed to keep the cores free from contamination after collection. The lengths of the cores varied but averaged about 12 cm.; their diameter was 18 mm. Recovery of Fungi After the tubes containing the cores were brought into the laboratory, the upper cork was flamed, the free water removed, and the supernatant liquid immediately above the mud put in a sterile container for future use. The core was then “ blown” under sterile conditions into a steril- ized Petri dish and the outside surface and ends removed with a hot scalpel as an additional precaution against contamination during col- lection. Two methods were followed in recovering the fungi. The first, practised extensively in the study of fresh-water fungi, involves the use of water cultures. Five jars of sterilized sea water were prepared into each of which was placed a core from Station 2, a site at which cores were readily obtainable. The cores were broken up in the water, the mixture allowed to settle, and 20-30 dead Calanus finmarchicus and Saggita from Station 2 were added. Three jars were kept at room temperature (23° C.) in the light and two in the dark at about 6° C. Bits of “shiner ”’ (a small marine fish) removed aseptically were also dropped into the cultures. After 10 days no fungi were found on any of this “ bait.” 244 Bike SPARROW). All the significant data were obtained by the plate method. The medium used was the following: DexctioS@mrn tm ket ec een aiareee 10.0 grams PEpLOMeuee cmt tera Nace aise as ltrs 2.0 grams ANGHEN Er Altus Poayinaal Neca een St ea 15.0 grams SCAMVATC IUCR Tae ia oe teas isa 1000.0 cc. This gave a reaction of pH 6.1. In plating out, samples were taken from the supernatant water, the surface, the middle, and the bottom of the core. The water was dis- tributed, undiluted, among five dishes and the medium added. Five samples were taken from each of the three regions of the core. These were mixed with 1 cc. of sterile water and plated out in the usual man- ner. Each of the mud samples was about the size of a pea and weighed 200-250.0 mg. In many cases pieces of mud were also laid on the solidified medium, but these dishes merely told whether or not fungi were present on or near the face of the sample. Flasks of unsolidified media were also tried but these became too heavily overgrown with bacteria and protozoans to be of value. As controls for each core, five plates of media were used to each of which was added the contents of a 1 cc. sterile water blank. While these were the actual controls, it will be seen that the ten dishes containing material from the middle and bottom of the core also acted as checks against laboratory contamination. Table I gives the results of the plating out of the mud samples. The fungi recovered from the muds by the methods outlined all be- longed to genera commonly found in land soils and easily recovered in spore form from the atmosphere. A large majority were species of Penicillium, while others belonged to such genera as Cephalosporium, Trichoderma, Aspergillus, Chetomium, Alternaria, Cladosporium, and even Rhizopus. It was soon evident that no fungi which could be con- sidered characteristically marine were being recovered by these methods and interest in them, qualitatively at least, was greatly diminished. A preliminary experiment was also carried out to determine whether such fungi might be associated with decaying phytoplankton. Using a sterilized net, two sterilized jars were filled with a heavy concentration of phytoplankton from Vineyard Sound. ‘The material was distributed equally between two sterilized battery jars, both of which were then placed under running sea water in the laboratory. By allowing the water to flow very gently into one jar (““A”’) the mass of diatoms soon rested on the bottom. Since the control jar (“ B”’) was subjected to a stronger stream, the material was soon washed out. After four days, during which time the phytoplankton had gradually disintegrated, five 1 cc. samples were plated out from the bottom of each of the jars. Five 245 SAPROPHYTIC FUNGI IN MARINE MUDS ‘QUOTe PNUI 9deJINS JO sadaId asn 0} O—F SUOTIEIS WO] [PI1o}eUT JUSTOYJNS JOU SeM DIB, , ‘suOT]vUTUIe}UOD Ajqeqoid 919M PU YSIP 9Y} JO 9Spa ay} 7e 919M SaTUOTOD Vsaq} JO OMT ¢ *paiins00 pey UOT}eUTUTe}UOD }eY} AjTIqissod ay} Jo esnedaq PepAPOSIP SPA 9UO 4N Pe}II]JOO 919M S2I0I OMT ;z "JI BAOGK 10}VM UT SB [Jam se 1aAe] doy ay} Wor r8uny sjueseides A[qeqoid pue Appnul A1VA 1938 ; G sie) 14 sf) Vv s9 ev $9 LE 0¢ 6S 09 Gls Bieeee ee sjpeIOL 0 $ 0 s io ‘I C if es a 0 s 1 OST 9 0 $ 1 $ 0 a 0 S = rs 0 cl T C LCV “| 0 Ol ! Or 0 Or i! Or = —— @8 s raul! 0°0¢7 v 0 Or 0 Or 0 Ol CC Or 91 Or t Ly OT (4 Oreo £ I 07 i! 0¢ I! 0¢ 6S 0c 0c st a 07 v LCE (4 L ST i “il! 0 “il! 67 “i! i! ¢ (6 ST € 081 i S4a}aUL Isuny soysip Isuny so[dues Isuny sojdures Isunj sojdures Iduny sojdures Isuny so]dures $9100 jo jo jo jo jo jo jo jo jo jo jo jo jo yideq UO1}24S 2) 180) (0h) ‘ON SoTUO[OD ‘ON, iS) 1000) (o}@) “ON SOTUO[OD, ‘ON, SOIUO[OD ‘ON SoIuUO[OD, ‘ON ON soysIp yquelq O2H 99 T + yuelq OF 99 T + yuelq OFH 99 T + 2109 2109 sojdures jorqu08D 2109 jo u10}}0g 2109 JO 3[PplN 9109 JO 99ejINS jo 90RjInNS 2vAOge 19,2 JO UISIIO “padanorad wun, {0 Sa1U0709 [Oo LaQUinU aY] PUD UId1AO 41aY4 ‘pasn sajquips fo 4IQUnU AY] SD [Jam SD SUOYIDIS AY] Mos, UaYD] Sa409 fo Aaquinu puD YIdap ay] BUIMOYS I aIavy 246 F, K. SPARROW, JR. dishes of medium alone acted as controls. Sixteen colonies of fungi were recovered from jar “ A” which contained the phytoplankton, and none from jar “ B” which contained only the running water. The con- trol plates remained sterile. It was apparent in this case, at least, that there existed a definite relationship between the presence of disintegrat- ing phytoplankton and the moulds. Again, as in the muds, the fungi recovered were all common dust-borne species. DISCUSSION From the foregoing, which it must be emphasized is a very prelim- inary investigation intended more to stimulate further inquiry than to arrive at definite conclusions, certain points seem worthy of further consideration. If truly marine fungi of a filamentous type exist in the marine muds studied and if there is nothing radically wrong with the methods and type of medium employed, then these organisms must be extremely rare. The types of fungi recovered, i.e., species of Aspergillus, Penicillium, Trichoderma, etc., have cast doubts on their being concerned to any great extent in disintegrating processes in the sea. Proof of their presence in marine muds does not mean that they are in an active vegetative state and hence, a working factor in the cycle of decomposition. To the writer’s mind, large numbers of direct microscopic observations of my- celium in mud seem the only positive method of demonstrating this important point. The failure of the water cultures to produce sapro- phytic fungi is also in line with the negative aspects of the data. How- ever, the scanty evidence afforded by the plates of disintegrating phyto- plankton presents a more positive picture of relationship. Furthermore, in direct microscopic observations on this disintegrating algal material - where freedom from opaque, inorganic matter greatly facilitated the search, active septate mycelium was found in several instances. ! Since only land and dust-borne fungi have been recovered it is nat- ural to ask whether they are only contaminants or whether they are actu- ally in the mud. In the collection of the cores every precaution against outside contamination possible with the apparatus was employed and before using the cores the mud surfaces in contact with the tube and the water were cut away. Contamination during culture procedure does not seem likely since mud from the middle and bottom of the cores yielded under identical laboratory manipulation a total of eight colonies in 130 samples and the 65 control plates two. When we consider that 239 colonies developed in the 155 dishes containing samples of super- natant water or mud from the top layer of the core it would seem fairly conclusive that the fungi recovered were actually present in the mud. SAPROPHYTIC FUNGI IN MARINE MUDS 247 Two methods whereby the moulds may have reached the muds are suggested. Either they have been washed from land soils into the sea, or they have been blown, in spore form, from land onto the surface of the water where they have gradually sunk to the bottom. While the stations studied were not in the littoral zone, none was sufficiently off- shore to be free from the possibility of obtaining spores from sedimen- tary material of land origin. Spores are known to be everywhere in the lower atmosphere, and in the immediate vicinity of a continent would be particularly abundant. Many of these would eventually reach the surface of the sea where they might ultimately sink to the bottom. In this connection, qualitative studies (unpublished) by Miss Lois Lillick of the phytoplankton of Vineyard Sound and the Gulf of Maine show that in the former locality fungous spores occur generally throughout the year in the samples, and that they also occur in the shallower waters of the Gulf of Maine, particularly during April and May. It is entirely possible, therefore, that either or both methods of conveyance have con- tributed fungous material to the marine muds. Perhaps the most interesting feature of the present investigation has been the information added to our knowledge of the vertical dis- tribution of these ubiquitous moulds. Many types have been recovered from the atmosphere both over land masses and the ocean and even in the stratosphere. Land soils and fresh water have yielded fungi and now they have apparently been recovered in some viable form from the surface of the ocean floor at depths up to 220 meters. Such hardihood is not surprising to those who have observed the capacity of certain of these fungi to withstand adverse environmental circumstances and even to produce under these conditions active vegetative growth. Finally, it might be emphasized that while no typically marine fungi were found there is strong evidence for believing that certain soil and dust-borne fungi can exist in the surface muds although it has not been shown in what form they occur or that they take an active part there in the disintegration of organic materials. SUMMARY Stratified samples of marine bottom were collected under as sterile conditions as possible from four stations which varied from 18 to 220 meters in depth. These stations were located in Buzzard’s Bay, Vine- yard Sound, and the Gulf of Maine and varied considerably in their distances from land. Attempts were made to recover saprophytic fungi from these cores. Two methods were used: (1) water cultures “baited” with suitable material of marine origin; and (2) the plating out of samples from (a) the water immediately above the surface of 248 F. kK. SPARROW, JR. the mud core, (b) the surface of the core, (c) the middle, and (d) the bottom of the core. Suitable controls were maintained. No fungi were found in the water cultures. By the plate method using a nutrient medium made up in sea water, 239 colonies were formed in dishes con- taining water from just above the surface of the core and from the sur- face of the core itself, eight colonies, in dishes containing material from the middle and bottom of the core, and two colonies in the controls. A preliminary experiment to determine whether or not fungi were associated with decaying phytoplankton showed definitely that such was the case. The fungi obtained by the plate method were all common dust and wind-borne forms. Since the methods used in the recovery of fungi did not show in what form they existed on the sea-bottom and since no species which might be called typically marine were recovered, it is doubtful in the present state of our knowledge whether these organisms play an active part in the disintegration of organic materials present in the mud. BIBLIOGRAPHY Moorg, H. B., anv R. J. Netix, 1930. An instrument for sampling marine muds. Jour. Mar. Biol. Ass’n., 16: 589. Petersen, Hennine E., 1905. Contributions a la connaissance des Phycomycetes marins (Chytridineee Fischer). Oversigt K. danske vidensk. Selsk. For- handl., 1905 (5): 439. Sparrow, F. K., Jr. 1934. Observations on marine Phycomycetes collected in Denmark. Dansk. bot. Ark., 8 (6): 1. Sparrow, F. K., Jr., 1936. Biological observations on the marine fungi of Woods Hole Waters. Biol. Bull., 52: 236. DETERMINATION OF POLARITY BY CENTRIFUGING BEGCS, Oh PUCUS HUREADUS * D. M. WHITAKER (From the School of Biological Sciences, Stanford University, California) INTRODUCTION The visible inclusions in the protoplasm of many animal eggs have been segregated within the cell into zones or strata by means of the centrifuge. In smaller, less yolky eggs in which the protoplasm is quite fluid, the inclusions, such as yolk and oil droplets, pigment granules, mitochondria (Arbacia, Harvey, 1936) etc., segregate in the protoplasm in accordance with their relative densities. In other cases there is sometimes a rotation or dislocation of regions of the cytoplasm which move as a whole. The results of numerous earlier experiments on animal eggs are reviewed in such standard texts as Morgan’s “ Experimental Embryol- ogy” (1927) and Wilson’s “The Cell in Development and Heredity ” (1925), and therefore no exhaustive review of the literature will be undertaken here. It should be noted, however, that in general the mere displacement of the visible inclusions has had surprisingly little effect on the development of most of the eggs. The primary polar axis is not determined by the axis of stratification, although Runnstr6m (1927) has shown that dorso-ventrality may be determined in the sea urchin egg. The general conclusion has been that primary polarity depends on factors in the transparent hyaloplasm rather than upon particles large enough to be moved readily by the centrifugal forces employed.” By centrifuging unfertilized Urechis eggs at 4,800 X g. for very long periods (up to 18 hours), Taylor (1931) was able to determine or shift the primary polar axis considerably, as shown sta- tistically, but it did not come to coincide precisely with the axis of stratification in most of the eggs. Morgan and Tyler (1935) found that when fertilized Urechis eggs are centrifuged, the polar bodies may 1 This work has been supported in part by funds granted by the Rockefeller Foundation. 2 Particles moved by the ordinary type of centrifuge may have very pronounced physiological effects, however. Shapiro (1935) has shown that fragments of centrifuged Arbacia eggs which contain the heavier granules respire at nearly twice the rate of transparent fragments, and Navez and Harvey (1935) have found twice the indophenol oxidase activity in the fragments containing these granules. 249 250 D. M. WHITAKER be set free in regions determined by the axis of centrifugation, but in this case the developmental axes of the eggs do not shift with the polar bodies, but instead remain unaltered. Conklin (1931) found that a dislocation of organs takes place in ascidian eggs when whole regions of the cytoplasm have been displaced or dislocated by centrifuging. Other alterations of the normal devel- opment of animal eggs have been brought about by the mechanical consequences of centrifuging, such as twinning (Tyler, 1930; Harvey, 1933,0 9SaNetes): Eggs of the marine brown alge belonging to the Fucacee are more labile than most animal eggs. A number of environmental agents may determine polarity in the Fucus egg, including unilateral light (Rosen- vinge, 1889) (Kniep, 1907) (Hurd, 1920), especially in the short end of the visible spectrum (Hurd, 1920), direct electric current (Lund, 1923), the presence of neighboring eggs (Rosenvinge, 1889) (Kniep, 1907) (Hurd, 1920) (Whitaker, 1931), especially in acidified sea water (Whitaker, 1937), a pH gradient (Whitaker, 1935) and a temperature gradient (Lowrance, 1937). Knapp (1931) believes that the entrance point of the sperm determines the point of rhizoid origin and the polarity in the egg of Cystosira, but the effect of such environmental agents as directed light, if applied at an appropriate time, will supplant this determination and establish a new polarity. The effect of the entrance point of the sperm has not been tested in Fucus, but if it de- termines polarity, this is readily altered by such means as those just indicated. Knapp found further in Cystosira that rhizoids form cen- trifugally when the eggs are centrifuged either before or after fertiliza- tion. Schechter (1934) found that rhizoids develop toward the positive pole when pieces of the red alga Griffithsia are reared in a direct electric current. Lund (1923) earlier had found that Fucus eggs form rhizoids toward the positive pole. Schechter observed that chromatophores mi- grated toward the positive pole in the Griffithsia, and he tried moving the chromatophores by means of the centrifuge (1934, 1935). The chromatophores and other bodies were moved, and accumulated cen- trifugally under relatively low centrifugal force (150 X g.) so that the cell-materials stratified, but the place of rhizoid origin was unaffected. Polarity was altered, however, in that shoots formed in the regions to which heavier materials had been thrown. METHOD AND RESULTS Eggs of Fucus furcatus were obtained from the same locality and by methods which have been described previously (Whitaker, 1936). DETERMINATION OF POLARITY IN EGGS OF FUCUS 251 Experiments were performed in February and March, 1937. Since this species of Fucus is hermaphroditic, and sheds egg and antheridial cap- sules at the same time,’ fertilization takes place when the egg capsules break down. This can readily be observed, and eggs were used from capsules breaking down during a time-span of 20 minutes or less. The mid-point of this time-span was counted as the average time of fertili- zation. The eggs were centrifuged at 3,000 x g. An International electric centrifuge was used which tended to warm up during a 20- minute run, and therefore the sea water in the centrifuge tube and the water around the tube were cooled at the start. The temperature of the eggs during the centrifuging rose from about 8° or 10° to 20° C. At all other times the eggs were kept in a humid, dark, constant tem- perature room at 15+%4° C. After being centrifuged they were briefly exposed only to red light which does not affect the polarity. The pH of the sea water used in the experiments ranged from 7.9 to 8.0, as measured by means of a glass electrode. In order to rule out the effects of neighboring eggs on each other, no egg lying within 5 egg diameters of another was counted in the results (see Whitaker, 1937). THE First SERIES OF EXPERIMENTS In the first series of experiments, the eggs were centrifuged in sea water beginning from 12 to 26 minutes after the average time of fer- tilization, and were then reared in sea water in Petri dishes. Pre- liminary survey showed that after being centrifuged for 5 minutes the eggs were quite definitely stratified, but not so sharply so as after being centrifuged for 15 to 20 minutes (see Fig. 1, 4). It was also found, when the eggs were observed about 24 hours later, that the stratified material had redistributed much more markedly if the eggs were cen- trifuged for only 5 minutes. In all of the experiments to be cited, the eggs were centrifuged at 3,000 X g. for either 15 or 20 minutes (usu- ally 20). Most of the eggs remained spherical and nearly all of those that were distorted by neighbors in the centrifuge tube rounded up again in the Petri dish. Figure 1, A shows a typical egg 23 minutes after being centrifuged for 20 minutes. Three principal zones are sharply demarcated. At the centripetal end, to which the least dense materials are thrown, there is a cap of globules which presumably are oil or fat. Next to this is a dark brown zone in which the chloroplasts are concentrated, and within which the nucleus 1s concealed. The remainder of the ege, including * Most of the antheridial capsules dissolve first so that antherazoids, or sperm, are swimming about the ege capsules when they dissolve and set the eggs free in the sea water. SE) D. M. WHITAKER the centrifugal pole, is essentially transparent, although a few plastids may remain scattered especially in the cortical region, and the general texture appears slightly granular. Figure 1, B shows the same egg 15 minutes later, and material of the dark zone is already seen to be moving back toward the center of the egg. Soon afterward the nucleus migrates out of the dark band, moving toward the center of the egg. It is usually Fic. 1. Photomicrographs of typical developing centrifuged and normal Fucus eggs. (A) shows an egg 23 minutes after being centrifuged for 20 minutes at 3,000 X g. The centrifuging began shortly after fertilization, and the centrifugal end is below. Chloroplasts are concentrated in the dark centripetal band. (B) shows the same egg 15 minutes later, and already inclusions in the dark band are seen to be diffusing back to some extent. (C) shows another egg from the same sample 3 hours and 10 minutes later. The nucleus has emerged from the dark stratum within which it was earlier concealed. (2D) shows another egg from the same sample 13 hours and 25 minutes later. It may be seen, especially by re- ferring to the remnant of the centripetal cap, that the rhizoid protuberance has formed at the centrifugal end of the egg. (A’) and (D’) are normal non-cen- trifuged eggs in stages corresponding to (A) and (D) respectively. All eggs were reared in the dark at 15° C. at pH 7.9-8.0 (see text). closely followed, and sometimes it is flanked along the sides or sur- rounded, by redistributing dark material. Figures 1 C and 1 D show other eggs from the same dish at later times. In Fig. 1, D it can be seen by referring to the reduced remnant of the oil cap that the rhizoid has formed quite precisely at the centrifugal pole. Some of the eggs DETERMINATION OF POLARITY IN EGGS OF FUCUS Zs) do not remain stratified but redistribute completely. These will be referred to later. In six similar experiments of the first series, large numbers of eggs were reared in Petri dishes after being removed from the centrifuge, and the final records were taken 20 to 24 hours later. When centrifuged eggs are placed in a dish of sea water, most of them rotate and lie so that the heavy side is down. When the rhizoid extends downward it lifts the cell body which usually rolls over to one side. The strata can then be seen. Each egg was further rolled over by hand with a glass needle if necessary to observe the stratification when the results were recorded. Sample counts of over 1,000 eggs indicated that the vast majority of the eggs which remained stratified, so that the axis of centrifugation could be identified, formed rhizoids at or near the ends to which heavier materials had been thrown. In three experiments more precise esti- mates were made of the positions of the rhizoids. A random count of 490 eggs showed that 96 per cent formed rhizoids within 10° of the centrifugal pole, and the remaining 4 per cent formed them between 10° and 22° from the pole. In addition to the eggs which remained stratified (Fig. 1, D), and could be analyzed as above, some eggs re- distributed the stratified materials so completely that the axis of cen- trifugation could not be identified. The proportion of these was usually small, about 4 to 10 per cent, although in one set of eggs, for reasons which are not clear, approximately 50 per cent of the eggs had redis- tributed at 24 hours. There was usually a sharp distinction between eggs that remained clearly stratified and those that had completely redistributed, with relatively few transitional cases. From the experiments cited above, it was concluded that the rhizoid forms at the end of the egg to which heavier materials are thrown, at least in the eggs which remain stratified after 24 hours. The Fucus egg is spherical and has no early identifying mark of polarity (Fig. 1, A’). It is conceivable, however, that the egg actually has a pre- determined rhizoid pole which is also the heaviest part of the egg. It so, in centrifuging, the egg would be thrown to the bottom of the tube in an oriented position. This does not appear very probable, but if it were true the stratification would conform to the oriented position and might itself be irrelevant in the determination of polarity. A second series of experiments was therefore designed to test this and other points. THE SECOND SERIES OF EXPERIMENTS In the second series, eggs were embedded in 1.5 per cent agar-sea water before centrifuging, after centrifuging, and without being centri- 254 D. M. WHITAKER fuged. Eggs develop normally in this medium, and observations on stratified eggs in various positions indicate that they are firmly held and that few 1f any move or rotate within the agar-sea water after it has cooled and jelled, if it is undisturbed. If the agar is cut with a razor, eggs lying close to the cut may rotate as a consequence of the mechanical disturbance. One and a half per cent agar-sea water is a firm jelly at 15° C., and the surface of a fertilized Fucus egg attaches to it firmly. Eggs were embedded by pipetting a concentrated mass of eggs in a minimum of sea water into agar-sea water at 38° C. The agar-sea water gelates below this temperature. The mixture was taken in and out of the agar pipette once or twice to assure mixing and the dish was put on ice to cool rapidly and shorten the exposure of the eggs to high temperature. When a population of fertilized eggs which are not centrifuged is embedded in agar-sea water and reared in the dark, the rhizoids form at random in all directions as seen from above. The upward and down- ward components are not equal, however, when the population is con- sidered statistically. Rhizoids are found in every possible position, from straight down to straight up, but more are found with a downward com- ponent than with an upward component. The average position is some- what downward from the horizontal plane. For example, in a popula- tion of 413 eggs, 14 per cent of the rhizoids formed upward (+: 45°), 41 per cent to the side (+ 45°), and 45 per cent downward (+ 45°). In other populations more rhizoids formed to the side than downward. It has been shown elsewhere (Whitaker, 1937) that the downward component increases with acidity of the sea water and with increased concentration of eggs. It cannot be said at the present time whether the downward component is due to gravity or to concentration gradients of substances (e.g. CO,) which diffuse through the upper boundary of the medium but cannot diffuse through the bottom of the dish. When eggs were first centrifuged, and then mixed at random into agar-sea water, it was found by analyzing 600 eggs statistically after the agar had solidified that the eggs tended to be oriented with the heavy side downward. This orientation, while appreciable, was on the whole not pronounced. It presumably took place just before the agar solidi- fied, as a result of the strong density gradients which had been created in the eggs by stratification in the centrifuge. The results from these eges which bear on the determination of polarity will be considered after the principal results of the second series have been taken up. These are derived from eggs first embedded in agar-sea water and then cen- {rifuged. In six similar experiments of the second series, eggs were embedded DETERMINATION OF POLARITY IN EGGS OF FUCUS 250 at random soon after fertilization in 1.5 per cent agar-sea water in moulds which formed castings which fitted the bottoms of the centrifuge tubes. The castings were centrifuged, and then small blocks or strips of agar, bearing stratified eggs, were cut with a razor and placed in Petri dishes of sea water. All eggs in a block were oriented the same way with respect to their stratification, except for some eggs, especially those near the edge, which were mechanically disturbed and rolled when the block was cut and handled. These blocks were oriented in three different positions: with the centrifugal side (1) straight down, (2) to the side, and (3) straight up. In each experiment, blocks of agar cui from the same centrifuged piece were placed in either two or three of the positions. After the eggs in these blocks of agar-sea water had developed in the dark for 20 to 24 hours, they were inspected with the microscope and sample counts were made of the positions of rhizoid origin. When- ever necessary, eggs were rolled over within the agar with a glass needle. The eggs were classified as having formed rhizoids downward within 45°, or to the side within 45°, or upward within 45°. The positions of the rhizoids with respect to stratification were noted, and whether or not redistribution had taken place so that stratification was no longer dis- cernible. The results of the separate experiments were essentially sim- ilar, and therefore the averaged results will be considered. Of 423 eggs in agar-sea water blocks placed so that the centrifugal sides of the blocks were downward, 46 or less than 11 per cent had re- distributed, while the remaining 89 per cent remained stratified. Ninety- nine per cent of the eggs which remained stratified formed rhizoids within 10° of the centrifugal pole (see Fig. 1, D), while 1 per cent formed them farther away from the centrifugal pole, but still on the centrifugal hemisphere. Ninety-five per cent of the eggs which remained stratified had retained their position in the agar, without rotating after being cen- -trifuged, so that their centrifugal poles were downward, while 5 per cent had been rotated so that their centrifugal poles were to the side within 45°. Of the 46 eggs which had redistributed, 3 formed rhizoids downward, 25 to the side, and 18 upward. The distribution of these rhizoids is strikingly different from those on the eggs which remained stratified. Relatively many more are to the side and upward, which suggests either that the determination of rhizoid formation at the cen- trifugal poles is largely lost when eggs redistribute, or else that a large proportion of the redistributed eggs had been oriented with their cen- trifugal poles to the side or upward (as a result of rotating when the block was cut), and that such orientations favor redistribution. Evi- 256 D. M. WHITAKER dence that both of these factors operate is provided by eggs reared with their centrifugal poles upward. Eggs reared in agar-sea water blocks placed so that the centrifugal sides of the blocks were upward were held with the heaviest materials at the top and the lightest at the bottom. This is the exact opposite of the position which the eggs tend to assume when free in a dish, and it should tend to favor redistribution since the effect of gravity is added to that of diffusion. Of 488 eggs which developed in blocks in this position, 166, or 34 per cent had redistributed (compared with 11 per cent in blocks placed centrifugal side downward), and of these redis- tributed eggs 130 formed rhizoids downward, 33 to the side, and 3 up- ward. In other words, the percentage of redistribution was trebled, and the redistributed eggs formed rhizoids more nearly in the general pattern found in a population of eggs which have never been centrifuged. Even more rhizoids formed downward than in the average population of non- centrifuged eggs. Three hundred and twenty-two eggs remained strati- fied, and 93 per cent of these formed rhizoids within 10° of the cen- trifugal pole. The remaining 7 per cent formed them farther away from the centrifugal pole, but not more than 90°. Two hundred and seventy-two eggs formed rhizoids upward, and 99 per cent of these had remained stratified and formed rhizoids within 10° of the centrifugal pole. ‘Twenty-six of the 488 eggs remained stratified but had been ro- tated so that the centrifugal pole was to the side (+ 45°), and in 27 it was downward (+ 45°). Eggs reared in agar-sea water blocks placed so that the centrifugal sides of the blocks were to the side gave results which are on the whole intermediate between those from eggs in blocks in up and in down posi- tions. Four hundred and twenty-five eggs were analyzed. Sixteen per cent had redistributed, compared with 34 and 11 per cent, respectively, in the other two positions. These redistributed eggs formed 1 rhizoid upward, 39 to the side, and 27 downward (all + 45°). Three hundred and fifty-eight eggs remained stratified and 333 of these (93 per cent) formed rhizoids within 10° of the centrifugal pole. The remaining 7 per cent formed them between 10° and 90° from the centrifugal pole. Of the 333 eggs which formed rhizoids within 10° of the centrifugal pole, 314 (94 per cent) developed rhizoids laterally (i.e., they retained their original orientation in the agar blocks). As earlier mentioned, eggs were also centrifuged first and then em- bedded in agar-sea water. The axes of stratification lay at various angles with respect to the horizontal plane. The results obtained by analyzing 600 eggs selected at random were entirely in harmony with those just cited. DETERMINATION OF POLARITY IN EGGS OF FUCUS Dash Discussion Most of the centrifuged eggs remain stratified long after the rhizoid protuberances have formed, although the stratified materials begin to diffuse back to some extent quite early (see Fig. 1). The results show that, regardless of the position in which they are held and reared, the centrifuged eggs which remain stratified form rhizoids on their cen- trifugal hemispheres. With very high incidence, they do so quite pre- cisely at their centrifugal poles to which heavier materials have been thrown. Thus, of 1,057 eggs which remained stratified in blocks of agar-sea water placed so that the centrifugal sides were downward, to the side, or upward (comparable numbers of eggs being in each of these 3 positions), 1,007 or more than 95 per cent formed rhizoids within 10° of their centrifugal poles as marked by the strata. The remainder formed them between 10° and 90° from the centrifugal poles. These eggs were embedded in random positions in firm agar-sea water before being centrifuged, and presumably could not orient in the centrifuge in accordance with any pre-determined polarity. The developmental po- larity is therefore determined by the axis of centrifugation. If there is an earlier polarity (as found by Knapp (1931) in Cystosira; see intro- duction), it is completely altered. Some of the eggs do not remain stratified, but instead the visible inclusions have redistributed so completely when they are observed after the rhizoid protuberances have formed that they are indistinguishable from eggs which have never been centrifuged. When eggs are held in- verted in agar-sea water so that the centrifugal poles are uppermost, the proportion of the eggs which redistribute is considerably increased, al- though the redistributed eggs still remain in the minority. When in- verted eggs redistribute, the tendency to form rhizoids at their centrifu- gal poles disappears. They form rhizoids more nearly in the positions in which they would have formed them if they had never been centri- fuged. In fact, both in the blocks placed centrifugal side upward and in those placed centrifugal side downward, the redistributed eggs formed even more rhizoids on the sides of the eggs away from the centrifugal sides of the agar blocks than non-centrifuged eggs would have been expected to do in their place. _The fact that only some of the eggs redistribute, while most others held in the same position do not, must be attributed to variations in the properties of different eggs in regard to factors such as viscosity, etc., and perhaps to the fact that some of the eggs had been fertilized as much as 20 minutes longer than others at the time of centrifugation. Knapp (1931) found that rhizoids form at the centrifugal poles of 258 D. M. WHITAKER Cystosiva eggs, in high percentages if the eggs have been visibly strati- fied. The principal results on Fucus eggs are thus similar. Schechter’s (1934, 1935) results on pieces of the red alga Griffithsia are similar in that the location of organs was determined by centrifuging, but in this case shoots, not rhizoids, formed at the centrifugal pole. The place of origin of rhizoids was unaffected. It is possible that a special rhizoid-forming substance accumulates at the centrifugal pole of the Fucus egg as a result of centrifugation. However, since gradients of a number of factors which affect the rate of activities such as respiration, CO, production, etc. (e.g. temperature, pH, etc.; see Introduction) determine the point of rhizoid origin in this egg, it is also possible that polarity is determined in the stratified egg by gradients of such activities caused by the asymmetrical distribu- tion of cell inclusions. Moreover, these factors may influence the mi- totic figure which in turn may affect the polarity. The present experi- ments do not provide a basis for choosing between these possibilities. SUMMARY 1. When eggs of Fucus furcatus are centrifuged at 3,000 X g. for 15 or 20 minutes, beginning between 12 and 37 minutes after fertiliza- tion, the visible cell inclusions are thrown centripetally (see Fig. 1). 2: Most of the eggs remain visibly stratified until long after the rhi- zoid protuberances have formed. Ninety-three to ninety-nine per cent of the eggs which remain stratified form rhizoids quite precisely at the centrifugal pole (within 10°; see Fig. 1), even when they have been em- bedded in random positions in firm agar-sea water before being centri- fuged so that they could not orient in the centrifuge in accordance with any earlier polarity. The remainder also form rhizoids on the cen- trifugal hemisphere of the egg, but farther removed from the centrifugal pole. 3. In eggs which remain stratified, the place of rhizoid origin and the developmental polarity are therefore determined by the axis of cen- trifugation. ‘This is true regardless of the position in which the strati- fied eggs are held in agar-sea water during development. If there is an earlier polarity in the egg it is completely supplanted by the effects of centrifugation. 4. In a smaller proportion of the eggs, the cell inclusions have re- distributed so completely when the eggs are inspected after the rhizoid protuberances have formed that, in high contrast to the eggs which remain stratified, they are visibly indistinguishable from eggs which have never been centrifuged. Such redistribution is more prevalent when eggs are reared heavy side up in agar than when they are reared heavy DETERMINATION OF POLARITY IN EGGS OF FUCUS LSS) side down. The determination of rhizoid formation at the centrifugal pole is lost in inverted eggs which have redistributed. Under the con- ditions of the experiments, the determination therefore appears to cor- relate with the distribution of visible inclusions. The author is indebted to Mr. C. W. Clancy for assistance in carry- ing out the experiments. BIBLIOGRAPHY CoNKLIN, E. G., 1931. The development of centrifuged eggs of ascidians. Jour. Exper. Zool., 60: 1. Harvey, E. B., 1933. Effects of centrifugal force on fertilized eggs of Arbacia punctulata as observed with the centrifuge-microscope. Biol. Bull., 65: 389. Harvey, E. B., 1935. Some surface phenomena in the fertilized sea urchin egg as influenced by centrifugal force. Bzol. Bull., 69: 298. Harvey, E. B., 1936. Parthenogenetic merogony or cleavage without nuclei in Arbacia punctulata. Biol. Bull.,71: 101. — Hurp, A. M., 1920. Effect of unilateral monochromatic light and group orienta- tion on the polarity of germinating Fucus spores. Bot. Gaz., 70: 25. Knapp, E., 1931. Entwicklungsphysiologische Untersuchungen an Fucaceen-Fiern. 1. Zur Kenntnis der Polaritat der Eier von Cystosira barbata. Planta (Abb. E), 14: 731. Kniep, H., 1907. Beitrage zur Keimungs-Physiologie und Biologie von Fucus Jahrb. wissensch. Bot., 44: 635. Lowrance, E. W., 1937. Determination of polarity in Fucus eggs by temperature gradients. Proc. Soc. Exper. Biol. and Med. (in press). Lunp, E. J., 1923. Electrical control of organic polarity in the egg of Fucus. Bot. Gaz., 76: 288. Morcan, T. H., 1927. Experimental Embryology. Columbia University Press, New York. Morcan, T. H., ann A. Tyrer, 1935. Effects of centrifuging eggs of Urechis before and after fertilization. Jour. Exper. Zool., 70: 301. Navez, A. E., ann E. B. Harvey, 1935. Indophenol oxidase activity in intact and fragmented Arbacia eggs (abstract). Biol. Bull., 69: 342. | Rosenvince, M. K. L., 1889. Influence des agents extérieurs sur l’organisation polaire et dorsi-ventrale des plantes. Rev. Gen. Bot., 1: 53. Runnstrom, J., 1927. Experimentelle Bestimmung der Dorso-ventral-achse bei dem Seeigelkeim. Arkiv. for Zoologi., 18: (No. 4), 1. Scuecuter, V., 1934. Electrical control of rhizoid formation in the red alga, Griffithsia bornetiana. Jour. Gen. Physiol., 18: 1. ScuHecuTer, V., 1935. The effect of centrifuging on the polarity of an alga, Griffithsia bornetiana. Biol. Bull., 68: 172. Sapiro, H., 1935. The respiration of fragments obtained by centrifuging the egg of the sea urchin, Arbacia punctulata. Jour. Cell. and Comp. Physiol., 6: 101. Taytor, C. V., 1931. Polarity in normal and centrifuged eggs of Urechis caupo Fisher and MacGinitie. Physiol. Zo6l., 4: 423. Tver, A., 1930. Experimental production of double embryos in annelids and mol- lusks. Jour. Exper. Zool., 57: 347. Wuiraxer, D. M., 1931. Some observations on the eggs of Fucus and upon their mutual influence in the determination of the developmental axis. Bvol. Bull., 61: 294. 260 IDS IE NaVles Adie OSC DIRS Wuitaker, D. M., 1935. Induction of polarity in Fucus furcatus by a localized concentration of hydrogen ions. Proc. Soc. Exper. Biol. and Med., 33: 472. Wuiraker, D. M., 1936. The effect of white light upon the rate of development of the rhizoid protuberance and the first cell division in Fucus furcatus. Biol. Bull., 70: 100. Wuirtaker, D. M., 1937. The effect of hydrogen ion concentration upon the in- duction of polarity in Fucus eggs. 1. Increased hydrogen ion concen- tration and the intensity of mutual inductions by neighboring eggs of Fucus furcatus. Jour. Gen. Physiol., 20: 491. Witson, E. B., 1925. The Cell in Development and Heredity. The MacMillan Co., New York. ON THE ENERGETICS OF DIFFERENTIATION, VI CoMPARISON OF THE TEMPERATURE COEFFICIENTS OF THE RESPIRATORY RatTES OF UNFERTILIZED AND OF FERTILIZD Eccs ? ALBERT TYLER AND W. D. HUMASON (From the William G. Kerckhoff Laboratories of the Biological Sciences, California Institute of Technology, Pasadena, California) The results of these experiments show principally that the tempera- ture coefficients of the rates of respiration are the same for unfertilized eggs as for fertilized eggs over most of the temperature range investi- gated. They diverge somewhat at the lower temperatures, the ferti- lized eggs giving higher values. ‘THEORETICAL PART An unfertilized egg is generally considered as a resting cell, pre- sumably being concerned merely with keeping itself alive. It should thus exhibit simply maintenance (basal) metabolism. The maintenance metabolism of an organism is measured by the heat production or the respiration under “resting” conditions. The respiration of an unferti- lized egg may then be taken as a measure of its maintenance require- ments. A fertilized egg is also considered as having a maintenance require- ment, but in addition there are requirements for processes termed growth and differentiation. It is conceivable that by such an agent as change in temperature, these three processes might be affected differ- ently. But from experiments on the effect of temperature on the rate of development and the rate of respiration of fertilized eggs (Tyler, 1936), this does not appear likely, unless these changes are compensa- tory. The cited experiments had shown that there is the same total oxygen consumption in reaching a given stage of development at one temperature as at another. Thus the temperature coefficients of rate of respiration and development are the same and there is no temperature within the normal range at which development is accomplished with a minimum of respiration. It was to be expected, then, that the rate of respiration of unferti- lized eggs should give the same temperature coefficients as that of de- 1 This investigation was supported in part by a grant from the Penrose fund of the American Philosophical Society. 261 262 TYLER AND HUMASON veloping eggs. An investigation of this sort was made by Rubenstein and Gerard (1934) on the sea-urchin egg. They found much higher coefficients for the unfertilized than for the fertilized eggs, the average values for Q,, being 4.1 for the former and 1.8 for the latter. An ex- amination of their data shows a number of errors in the calculations which, while not greatly affecting the average values, make their con- clusions somewhat less convincing. Considering also the difficulties in- volved in measuring the respiration of unfertilized eggs, it seemed ad- visable to repeat these experiments on the sea-urchin and in addition to investigate other forms. EXPERIMENTAL PART Material and Methods The eggs used were those of the sea-urchin, Strongylocentrotus pur- puratus; the sand-dollar, Dendraster excentricus; the echiuroid worm, Urechis caupo,; and the ascidian, Ciona intestinalis. The manometric method of Warburg was employed for measuring the respiration. Since one of the principal difficulties with unfertilized eggs is their relatively low rate of respiration per unit mass, special ves- sels were constructed which would tend to make the measured pressure changes as large as possible, at the same time insuring adequate gas ex- change. Considering the various factors involved, the cylindrical type of vessel previously described (Tyler, 1936), but with calibration vol- umes of 18 to 20 cc. and capable of taking 8 cc. of egg suspension, was _ employed. ‘The vessel constants are of the order of 1.0. The eggs are pipetted into the vessels with special automatic pipettes. Errors in cal- ibration of the vessels and of delivery from the pipettes amount to less than 0.1 and 0.3 per cent and are therefore negligible. The amount of material employed was determined from the nitrogen content of the eggs, obtained by means of a modified Kjeldahl method. The error here depends upon the amount of material employed, but even for the most dilute egg suspension it does not exceed one per cent. In some instances, noted as “no Kjeldahl” in the tables, no nitrogen determinations were made, but the eggs were simply pipetted from a uniform suspension. The manometers are read to the nearest 0.5 mm., so the reading error | will be determined by the magnitude of the pressure change. In gen- eral no coefficients were calculated for pressure changes of less than 10 mm. and in most cases the readings used were between 30 and 80 mm. In cases in which only dilute suspensions of eggs are available, the runs must be continued over longer periods of time to get sufficient pressure change to reduce the reading error. This involves another difficulty. RESPIRATORY RATE OF UNFERTILIZED EGGS 263 One of the chief difficulties that was encountered is the variation in rate of respiration during the progress of arun. ‘The rate of respira- tion of the unfertilized egg does not remain constant, as is usually as- sumed, but rises after a shorter or longer period of time. This rise has been previously noted by Warburg (1915) and by Runnstrom (1930). We find that the unfertilized eggs of the different animals we have stud- ied vary in regard to the rate of rise. ‘This is illustrated in Fig. 1. It /' hour//m Go oS / a ee) abe as eee | Hours (after removal or intentitized eggs ue after fasemmigation for fertilized eggs) Rate of Oxygen Consumption (cmm. 05 1 EXPLANATION OF FIGURES Fic. 1. Rates of oxygen consumption of unfertilized and fertilized eggs of Ciona, Urechis, Dendraster and Strongylocentrotus. The unfertilized rates for the two latter rise much more rapidly with time than for the other two. may be seen that the eggs of Dendraster and Strongylocentrotus show a much more rapid rise than do the eggs of Ciona and Urechis. In fact, in Dendraster and in Strongylocentrotus the unfertilized egg respiration rate rises almost as rapidly as does the fertilized egg respiration rate. This rise in respiratory rate appears to be correlated with the loss of fertilizability of the eggs. Without presenting quantitative data at this time, it may be pointed out that Ciona and Urechis eggs remain fertil- izable after standing 24 to 48 hours in sea water whereas Dendraster and Strongylocentrotus eggs are no longer fertilizable after 5 to 8 hours. Different batches of eggs differ somewhat in the rate of rise, but the @ 264 TYLER AND HUMASON curves of Fig. 1 are typical. The unfertilized eggs of all the forms in- vestigated show sooner or later a rise in respiratory rate.” Occasionally in some runs (e.g. Experiment XI of Table 1) a de- crease in rate is manifested in the early part of a run. This appears to be due to the sticking of the unfertilized eggs to the walls of the vessels. In some cases a considerable amount (perhaps 5 per cent) of the eggs stick to the walls. The stuck eggs are presumably unaffected while under the sea water, but when the shaking is stopped for a reading some of the stuck eggs are left well above the water level in the vessels. Such eggs, if partially or completely cytolyzed, would give an abnormal res- piratory rate. In our experiments on cytolyzing eggs with dry ice or with distilled water we find an initially high respiration followed after about an hour by a very low rate. If this holds for the stuck eggs we would expect to get slight decreases in rate as well as slight initial in- creases, since most of the sticking occurs at the start of a run. Due to the variations in rate of respiration with time the temperature coefficients cannot be determined simply by placing a vessel first at one temperature then at another. It is necessary in addition to run eggs from the same batch simultaneously at the two temperatures. The ex- periments were therefore performed in the following manner. Usually, four vessels were prepared from the same batch of eggs as soon as pos- sible after removal from the animal and two of them were placed in each of two different temperature baths. After a certain number of readings were made the vessels in the high temperature bath were ex- changed with those in the low temperature bath. After another set of readings, the vessels were replaced in the original baths and the readings continued. At least fifteen minutes was allowed for temperature equi- librium to be attained. The temperature coefficients were then calculated in two ways; first, from the oxygen consumption in the vessels run simultaneously at the two temperatures; secondly, from the oxygen consumption in one vessel run alternately at the different temperatures. or the first type of coefficient it is important to know the quantities of eggs in the different vessels or to have identical samples of a uniform suspension in each. For the coefficients calculated in the second manner this is, of 2 The rise is not due to conditions in the respiration vessels, as an experiment with Urechis illustrates. Two samples of eggs, run continuously for 20 hours at 22°, gave Xo,’s of 3.05 and 3.00 cu. mm. in the first hour, and 5.12 and 5.35 cu. mm, in the twentieth hour. Two more samples of eggs from the same batch, that had stood for twenty hours and washed before using, gave Xo,’s of 4.81 and 5.06 cu. mm. in the first hour. Subsequent fertilization was 85 to 100 per cent. The capacity for fertilization is not lost until after a considerable rise in respira- tion is manifest. In some experiments 100 per cent fertilization was obtained after an almost two-fold rise in the unfertilized rate. RESPIRATORY RATE OF UNFERTILIZED EGGS 265 course, unnecessary. Errors in the determinations of the quantity of eggs therefore do not enter into the coefficients calculated in the second manner. However, variations in rate of respiration with time will sig- nificantly affect the coefficients calculated in the second manner. To compensate for this the average of two values is taken. For a vessei starting in the high temperature bath, one value is obtained by dividing the initial rate of oxygen consumption by the rate during the ensuing period in the low temperature bath. The other value is gotten by di- viding this same low temperature rate into the rate during the following period in the high bath. Where the rate rises, as it generally does, the first value will be lower than the second but the average of the two will be nearer the true value. -For a vessel starting in the low temperature bath the two values are obtained in a similar manner. In this case an increasing rate will make the first value too high and the second too low, but again the average will be nearer the true value. The principle in- volved is the same as in determining the rest point of a balance. In the tables, the coefficients calculated in the second manner are not given for each individual vessel, but the average for each pair of temperatures is listed in the line “ average for individual vessels.” The respiration values from which the coefficients are readily determined are given under the headings Xo,. The probable errors are given in cases in which fifteen or more values are averaged. For this purpose, the average for each vessel, determined as mentioned above, is considered as a single value. The last two columns of the tables give the coefficients determined in the first manner; that is, for oxygen consumption during the same period of time at the two temperatures. Where duplicates are run, the coefficient is the ratio of the average oxygen consumption in the two vessels at each temperature. The mean of these values for each pair of temperatures is given in the line marked “average ” in the tables. In determining this average an experiment with only two ves- sels is weighted one-half (rather than one-quarter); that is, two ex- periments with two vessels are considered the same as one experiment with four vessels. The probable errors are again given for cases in which fifteen or more values are averaged. In most of the experiments only unfertilized eggs were run, a con- siderable number of coefficients for fertilized eggs having been obtained in previous work (Tyler, 1936). These values for fertilized eggs are listed in the lines “average from previously published data” in the tables. In some experiments (e.g. VI, X and XI of Table I) two ves- sels with fertilized eggs were run along with two vessels with unfertilized eggs from the same batch. The figures in parentheses in the tables are for fertilized eggs. In some experiments (I, V and VI of Table V) 266 TYLER AND HUMASON after an unfertilized run, the eggs were inseminated in the vessels and the measurements continued. The fertilized eggs give in general much more consistent values, so relatively fewer experiments, in addition to those previously published, were required. ‘The coefficients for the ferti- lized eggs were calculated here in the same two ways as for the unferti- lized eggs. The values taken from previously published data were ob- » tained in a somewhat different manner (loc. cit.) which automatically took into account the rising rate of respiration of the fertilized eggs. However, since the rate rises very slowly at the start and since the methods of calculation also allow for the rising rate, comparable values are obtained for short runs. In prolonged runs, values for fertilized eggs obtained in this manner would tend to deviate because the rate of rise increases somewhat with time. TABLE I Eggs of Urechis. Xo, = mm.? O2 consumed per mg. egg nitrogen; figures in parentheses are for fertilized eggs, the rest are for unfertilized eggs. Quo = ratio of Xo, values (average Xo, where duplicates are run) for the same time intervals at the two temperatures. Vessels A B C D Experi- Tem- Tem- ment Time] pera- | XO» XO, | pera- | XOe XO» Quo unfert. Quo fert. No. ture ture hours| °C. 2G I 1 DD, 1.99} 1.90} 12 0.82] 0.83 2.37 1 12 0.84] 0.91 D2, 2.09] 2.17 2.43 1 2, DPA Sh |e Dee sl eel 0.97 | 0.96 2.41 22 6.10} 6.00) 12 2.30} 2.04 2.79 II 2 2 WZ DES 4 | 249 D2 6.43 | 6.28 Das) 2 22 GAT OO 1 2.19 | 2.04 2.98 III 22 4.35| 4.24 | 12 1.79 | 1.80 2739 IV 2 Di 4.77| 4.60 | 12 2AO 2 ern Os 2.30 VI 1 DD 1.38 | (3.38)} 12 0.67 | (1.43) 2.06 PSII 1 12 Ore @32) 22 1.64 | (3.38) 2.65 2.56 1 22 1.64 | (3.14)| 12 0.87 | (1.52) 1.88 2.07 RESPIRATORY RATE OF UNFERTILIZED EGGS 267 TaBLE I (Continued) Vessels A B G D Experi- Tem- Tem- ment Time| pera- | X02 KO, | pera- | XO» XO» Quo unfert. Qo fert. No. ture ture hours| °C. HG VII 1 D2 S50) o-35) |) ke ORS TAS, 2.74 3 12 3.87 | 3.93 | 22 | 10.91 }11.28 2.85 3 22 |10.14/10.28 | 12 3.99 | 4.17 2.50 VIII 1 DD 4,49 12 Load 2.54 2 12 3.00 22 8.23 2.74 3 22 A AOLST a 12 3.35 3.24 IX } 22 7.38 12 2.90 255 2 12 2.58 22 6.27 2.43 2 22 6.44 12 2.30 2.80 x 1 22 4.42 | (4.50)| 12 1.82 | (1.65) 2.43 DS 1 12 1207) (270) 22 4.46 | (4.75) SATA 2.79 1 2D, 3.69 | (4.69)| 12 1.82 | (1.68) 2.03 2.79 XI 1 22 5.06 | (4.75)| 12 1.96 | (1.62) 2.69 2.93 1 12 155) Clas) 922 5.12 | (4.80) 3.30 2.70 1 22 4.33 | (4.88)| 12 1.82 | (1.72) 2.39 2.84 EMV ETAC CMA TN CT tae EN vase eu rts Oe a Lore 2.57 + 0.03 | 2.64 Average for Experiments II, V and VII............... 2.70 Average for individual-vessels;. .: J..2046 0 oes be aes eee 2.57 + 0.04 | 2.65 Average for individual vessels, Exp. II, V and VII...... 2.77 Average for fertilized eggs from previously published LeU trata everett Poe MRR va Dm Meurn CAE Me LAR CHa Je oh 2 2.79 + 0.02 Samples of the eggs from the vessels were inseminated at the end of each run and gave in most cases 90 to 100 per cent fertilization. The experiments in which it was less are as follows: Table I, I, 85, 75, 90, OOF INES) D5, oo) oo); Lable UME My 707S5-0be 80/070; 90,85; Vable INGO CVn 7): lable Vile. 20. 3550. ie; 10,50; 207 (Phere is no evident relation between the low fertilization in these cases and values of the corresponding temperature coefficients, as an examination of the tables shows. It is clear from temperature experiments on biological material that coefficients determined at different parts of the temperature scale are not alike. They generally increase as the temperature range is lowered. In other words, Q,, (and even p of the Arrhenius equation) is not a constant. We must therefore make our comparisons for the same tem- perature intervals and not use the average of values obtained from all 268 TYLER AND HUMASON parts of the temperature scale. We have concentrated on one pair of temperatures in attempting to obtain consistent values and supplemented with fewer experiments at other temperatures. Urechis caupo Seventeen sets of experiments were run with the eggs of Urechis; eleven at the temperatures 22° and 12°, three at 20° and 10°, two at 18° ‘and 8° and one at 15° and 5°. The experiments at 22° and 12° are listed in Table I. The average of the coefficients for the unfertilized eggs is 2.57 by both methods of calculation. For the fertilized eggs (Experiments VI, X and XI) the average Q,,’s are 2.64 and 2.65 re- spectively by the first and by the second methods of calculation. From previous data the value for fertilized eggs is 2.79. Considering the variation in the individual values, we can only conclude that there is no significant difference between the coefficients for the unfertilized and the fertilized eggs in this temperature range. Comparison of the first and last values of Xo, in each experiment gives the change in rate of respiration. In some instances (e.g. Ex- periment IV) the unfertilized rate is greater during the last period than at the start. In other cases (e.g. Experiment II) it is fairly constant and in others (e.g. Experiment IX) it drops somewhat. But since the direction of change and relative miagnitude is the same at both tem- peratures, the O,,’s calculated in the first manner are not very greatly affected. Thus for Experiment IV we have 2.30, 2.22 and 2.35. The coefficients calculated in the second manner also are not greatly affected where the rate changes roughly uniformly. Thus in Experiment IV we have the O,, values 2.36, 2.40, 2.19 and 2.24 for the individual ves- sels. It would, of course, be better to consider only cases in which there is very little change in rate. Taking Experiments II, V and VII as such, we get average Q,,’s of 2.70 and 2.77 which are closer to the values for fertilized eggs. We could not, however, find any criterion, such as the extent of agreement between duplicate vessels, the behaviour of the eggs upon fertilization, etc. that would justify the exclusion of any of the experiments listed. The agreement between the coefficients calculated in the two ways described simply means that, where the rate of respiration changes, the change is fairly uniform. It does not, of course, measure the accuracy of the values. The experiments at 20° and 10°, 18° and 8°, and 15° and 5° are listed in Table II. At these temperatures, the unfertilized eggs give RESPIRATORY RATE OF UNFERTILIZED EGGS TABLE IT Eggs of Urechis. Same description as Table I. 269 Vessels A B Gs D Experi- Tem- Tem- ment Time} pera- XO, XO pera- XO XOo Qio unfert. | Qio fert. No. ture ture hours| °C. 2 a q XII 1 20 1.74) 1.84 10 0.73) 0.68 2.55 1 10 0.93} 0.88 20 1.99} 2.10 2.26 1 20 DIX) | Baile 10 1.00; 1.02 2.16 XIII 1 20 1.95 1.87 10 0.81} 0.84 2.30 1 10 0.77| 0.74 20 2.03} 2.07 2.71 1 20 1.89] 1.87 10 0.78| 0.79 2.40 XIV 1 2 Oa S223 (223) 10 | 12.3 | (18.2) 2.63 2.87 (no Kjeldahl) | 1 10 | 14.9 | (20.4) 20 | 37.0 | (52.0) 2.48 2.55 1 20 | 31.8 | (54.6) 10 | 11.9 | (16.4) 2.67 3.33 AN GESTENEOG a ov: © SS CHET Re die aPC SETI aR CEE Ee ree eye 2.46 2.92 Averacetorindivicduall vessels ap saan cas eae see oe 2.47 2.82 PMVCIAPeMOmpclea VAG el Gata es. sed es Meng wine ye ole 2 ciel 3.30 2 18 2.01 1.97 8 ss Os i es 0) 1.81 XV 2 18 DIES DPX) 8 1.05} 0.90 2.32 2 18 2.92} 2.80 8 0.83} 1.00 3.13 XVI il 18 1.42} 1.42 8 0.60) 0.67 2.24 1 8 0.83 | 0.79 18 1.52) 1.48 1.85 1 18 1.34} 1.28 8 0.73] 0.78 1.73 INS TETTEIIS DPE repair Rade eh aa ha ECR era naar pe ie ey 2.18 Averageton imdividuall vesselsijaans. ©. 242s. e ae aceon ae 1.94 XVII 2 15 | 46.4 | (37.9) 5 | 16.9 | (10.9) Pret) 3.48 (no Kjeldahl)| 2 5 | 19.5 | (11.2) 15 | 49.7 | (36.0) 2.55 3.23 2 15 | 55.5 | (37.9) By eSgs) Guiles) 3.58 3.29 JASH TRIS, & seer ash REG ebro eI at AOR Bt Poon ee aN Me Rta ava 2.96 938) 2.84 3.30 Average for individual vessels consistently lower Q,,’s than do the fertilized eggs. the experiments were fertilized eggs run. tion experiments were available from previous work, but the coefficient for cleavage (3.30) which should be the same as for respiration (Joc. cit.) is listed. For 18° and 8° and for 15° and 5° the cleavage value is In only two of For 20° and 10° no respira- 270 TYLER AND HUMASON Asie MUL Eggs of Strongylocentrotus. Same description as Table I. Vessels A B i C D Experi- . Tem- 2 Tem- ne | Be | Pee) Aon | nee | Moe | Oe rer ae hours) °C %G (aa I 2 DY 3.39) 3.60) 12 130) Sess 2.63 2 12 1.23] 1.48] 22 BAS) || SoA 2.39 2 22 Soi) 4220) “12 1.46] 1.58 2.61 Il D 22 4.74|(16.4) | 12 1.65 |( 6.8) Neel 2.42 Pualnton ol03 \( 7-2)-) 22) 5.03 \GneD) 2.57 2.97 2 2D 5.46 |(22-1))| 12 2.44 | (10.6) 2.24 2.08 Ill 1 22 |30.0 |(64.7)| 12 | 11.81 |(29.5) 2.54 2.19 (no Kjeldahl) | 1 12 ASS NGO 2 SAS) (oS) 2.37 2.08 Henao 13400 (66.4) |) 2) sh isetie \GoM) 2.60 DOA LNA ESTENERE a: 8.05 0! ois e528 AG a! 9 SLO CAE CRA Ce ROE ne iatee A iciahe 2.54 2.33 AV eErAgCeMOnmmciviclialyviesselsis ci, cus Nek sb ek byt oe seme 2.53 2.39 IV 2 20 Dre 552-65) 10 0.67} 0.70 3.84 2 10 EQ 2 e224 20 Papen Wie A) O)5) 2.24 2 20 POG S225, TESSa eal DS2. V 1 AQ | 2263 WEBS WO) 8.9 |(10.9) 2.49 2.08 (no Kjeldahl) | 1 LO LO GOL PAO PSs (PS?) 2.46 2.14 1 20 125-4 23.2) 10> se 12 saa Gisas) 1.99 1.69 VI 1 20 | 24.4 |(62.9)] 10 (24.3) 2.59 (ro Kjeldahl) | 1 10 | 10.6 |(29.0)| 20 (64.4) DD, 1 20 | 26.7 |(77.3) | 10 (Qi) 2.81 VII 1 20 1.48} 1.38] 10 0.49] 0.51 2.86 1 20 1F6O) 4-537 © 0.48} 0.51 3.16 1 20 1GO 1262) tO 0.54} 0.51 S407 VIII 1 20 1.03} 1.00} 10 0:33 |) 0:30 3.22 10 0.40} 0.42} 20 1.20] 0.92 2.59 1 20 1.09} 0.96; 10 0.55] 0.39 2.18 IX 1 20 eT 10) 0.42} 0.52 2.67 2D 10 PSO eA PAD 2.89 | 2.88 2-35 1 20 Led2 |) ew) Al@ 0.65} 0.67 2.67 x 1 20 PP eS al) 0.59} 0.33 2.56 1 20 tes Any ealle2S) esd 0 0.78| 0.41 2.20 1 20 1.41] 1.34] 10 0.95 | 0.39 2.05 PANNA Oa Net Shes BAN Hf a ese oA Lr RY oa ROM tA 2.63 + 0.06| 2.26 Averavenotandivaciall viesselsij.cis. 19s toe 64 gies eee 2-57 30105) 2-28 Average from previously published data.................... 2.30 RESPIRATORY RATE OF UNFERTILIZED EGGS Dia not given because Urechis eggs do not divide at 5° and only occasion- ally do so at 8°. By themselves, the experiments at these temperatures cannot be taken to demonstrate a difference between the coefficients for the unfertilized and fertilized eggs. They are, however, consistent with the results on the other forms investigated in giving at the lower tem- peratures somewhat lower values for the unfertilized eggs. TABLE IV Eggs of Strongylocentrotus. Same description as Table I. Vessels A B G D Experi- Tem- Tem- ment Time] pera- | XO, XOg pera- | XO, XO_ Qio unfert.| Qio fert. Number ture ture hours} °C. ew SG: Rear XI 1 18 |38.5] (48.2)| 8 16.0} (16.7) 2.41 2.88 (no Kjeldahl) | 1 | 8 |15.5| (19.5)| 18 |38.5| (48.6)| 2.48 | 2.50 (eS 40) 7055-3) 18) tG Sil CSE 246) «| 9.39 XII LS 2a ee (ONES) NS OFS)" (239) 2.61 2.56 (no Kjeldahl) 1 See OOS eS) el Semmi |Z | (Os-0) Desi 2-31 1 ; PAW CAG Chaney MGs ear Wen Maral el ep Mae ls tecneyssge wiencaahab anes aiohaue & 2.58 2.69 PAVerAae Lom ANGiviGd Wal Vessels. os Oi ieli ss | Sanh eae oe oe ce Shoes 2.59 Dei XIII ES AO COLO ere I ae (no Kjeldahl) | 2 | 7.5 |18.3| (38.4)! 17.5 147.7 | (106.3) | 2.60 De DN tes \SA oil d224)\" saeshl2ont ESUSEDIECO RE 0 ae hel Rea Ne ire ee NTL en ean eg et RG Re 2.67 2.79 PV erase tor individual vesselsiac wise andes Melani eyed anee oe 2.69 2.80 QO1s O1s XIV 2 320) 137-5 N@2to) |) SialitO-81| 128-0) |) G47 4.21 (no Kjeldahl) |-2 | 5 |12.0| (31.3)| 20 [44.6] (129.0)} 3.72 4.12 AN ES ee) SS) Aa OD) TE VGk eZee PNVSRNEDA oiblesd 6s 8 yea Win he LRM Ces Mane tye Sees. Bin gk Ane eke 3.61 4.22 Averare ror indivacualiyessels. sas yaee ese ane een oe ale ae 3.74 4 24 It is of some interest to compare the absolute values for the rate of oxygen consumption in different experiments. Considering only the first period in each experiment, we see in Table I that the rate varies from as low as 1.38 cu. mm. O, per hour per mg. N, as in Experiment VI, to as high as 5.06 cu. mm. as in Experiment XI. The differences 272 TYLER AND HUMASON are roughly correlated with the length of time that the animals were kept in the aquaria before the eggs were used, eggs from freshly col- lected animals giving higher absolute rates of respiration. Animals from different localities vary, but considering, for example, Experiments VI to XI in which the animals were collected at one time in one locality, the times after collection for numbers XI, VIII, X, IX, VII and VI are 14, 3, 4, 9, 10 and 40 days respectively, and the absolute rates of respiration decrease in just about that order. Eggs from the same animal were used in Experiments VII and XI. We shall not attempt an explanation at this time, but we may point out that Urechis does not store its eggs in the ovary (if there is a definitive ovary) but in “ neph- ridial”’ sacs. Also, it is evident that simply aging the eggs in sea water produces the reverse effect, namely a rise in the absolute rate. With the fertilized eggs we find no marked differences in the ab- solute rate of respiration. In addition to the figures of Table I we have ten more values for the absolute rate of oxygen consumption during the first hour after fertilization, all of which lie between 4.2 and 4.5 cu. mm. We find thus in Urechis cases in which the rate of respiration rises two or three-fold after fertilization, cases of no change in rate and cases (e.g. Experiment X1) of a decrease in rate. Whitaker (1933) showed that in different species the rate of respiration may rise, fall or remain unchanged after fertilization, the absolute rates for the fertilized eggs tending toward the same level'in all. Here we have one animal exhibit- ing all three types of behavior. Strongylocentrotus purpuratus Tables III and IV give the respiration data and temperature coef- ficients for fourteen experiments with eggs of Strongylocentrotus. There are three at 22° and 12°, seven at 20° and 10°, two at 18° and Sosone at l/c and 7.5°, and) one at) 20° andia 2a) tn) all ontnenisie rate of oxygen consumption of the unfertilized eggs shows a rise during the experiment. The shorter runs (e.g. Experiment III of Table IIT) show a smaller rise than do the longer runs (e.g. Experiment IT). In all of the experiments the temperature coefficients for the unferti- lized eggs are fairly close to those for the fertilized eggs. For the temperatures 22° and 12°, the difference is small. For the temperatures 20° and 10°, it is somewhat greater. But if we omit the high values 3.84 and 3.22 of Experiments IV and VIII, the average Q,,, by the first method, is 2.27, which is much closer to the value for the fertilized eggs. An examination of the corresponding respiration figures shows that there is some justification for omitting these values, since, in both these experiments, vessels C and D give oxygen consumption figures that are evidently too low during the first period. RESPIRATORY RATE OF UNFERTILIZED EGGS Daf) TABLE V Eggs of Ciona. Same description as Table I. Vessels A B Cc D Experi- Tem- Tem- ment | Time] pera- XO_ XO» pera- XO, XO, Qio unfert. Oro fert. No. ture ture hours| °C. bay Gos I 2 25 16.02 | 14.84 15 9.32 1.66 1 15 4.64 4.04 25 9.77 Dey) 1 25 8.71 8.28 15 4.76 1.79 I 2 DOA (2ONA2) 1 29236) US (S257) | (ALOT) 2.14 1 15 | ( 7.37)| ( 7.63)| 25 | (14.03) | (14.24) 1.88 1 25 ALO) 19:09) Sasi C esl) NG 7-07) 2.65 NG SARS. S S58 0 SHO MOOS CeO BER EERON GILEAD eo net eae ae 1.90 Dela Averace ton individual vessels.);. 222... 2c won uh cee Oe ewan 1.92 2.14 Average from previously published data................... 2.27 -| hours II lz || 22 il72 | Wileul 12 7.35 6.04 1.74 1 12 3.25 4.07 22 7.97 E22. 2.08 ily) 22 9.78 | 11.28 12 4.19 5.47 2.18 III 2 D2) 19.88 | 21.93 12 10.72 | 11.03 1.92 2 12 8.39 8.34 22 16.88 | 15.28 1.92 D DD 17.64 | 18.28 12 7.04 7.54 2.46 IV 1 22 7.43 6.30 12 4.06 3.40 1.84 $| 12 3.16 4.55 YD 8.89 8.28 2.23 1 22 6.29 5.06 12 4.68 4.86 1.19 San 200 et 2D AASO3" a tO 75) 10) (65.85 2.03 2 12 7.61 5.92 22 12.85 1.90 2 DY 13.48 | 12.30 12 5.86 2.20 V 2 22 | (22.50) | (22.39)| 12 | ( 9.64)| ( 9.89) DD) 2) 12 | ( 9.93)| ( 9.56)}| 22 | (20.94) | (20.41) DAD 2 22 | (25.00) | (25.47)| 12 | ( 8.84) ¢ 8.47) 2.92 VI 1 22 7.26 6.42 12 3.62 3.85 1.83 1 12 3.77 3.49 22 6.55 6.57 1.81 1 22 6.22 6.72 12 3.17 3.15 2.05 VI 1 22 |( 9.55)|( 9.30)| 12 | ( 4.16)| ( 3.67) 2.41 1 12 | ( 3.58)| ( 3.49)| 22 | ( 9.82)| ( 8.97) 2.66 1 22 | (10.90) | (10.04)} 12 | ( 4.08) | ( 3.85) 2.64 LESH ESREN SON etd SAMS 1 Oe Oty oc StL RO AS TA Me ee cea I 2 AG 1.96 + 0.05] 2.50 Average for individual vessels..................--+0+0-0-- 1.98 +0.06} 2.47 Average from previously published data................... 2.84 274 TYLER AND HUMASON For the temperatures 18° and 8° and 17.5° and 7.5° there are also no significant differences between the coefficients for the unfertilized and the fertilized eggs, as the figures in Table IV show. One experi- ment at 20° and 5° shows some difference which may be significant. The unfertilized eggs give lower coefficients (Q,, in this case) than do the fertilized, which is consistent with the results on Urechis at the lower temperatures. The coefficients for cleavage which could be ob- tained more accurately than those for respiratory rate, are not given here because the eggs do not divide at 5° C. TABLE VI Eggs of Dendraster. Same description as Table I. Vessels A B C D Experi- Tem- Tem- ment | Time| pera- XO. XO, pera- XO, XOo Qio (unfert.) | Quo (fert.) No. ture ture hours| °C. Es I 2 22 6.04 5.98 7 2.19 2.30 2.68 2 12 3.46 3.28 DY 7.73 8.00 2.34 2 22 11.08 | 11.02 1 3.96 4.08 Dele II 2 22 4.40 4.05 12 1.96 2.08 2.09 2 iD DD 2.31 D2, 6.10 6.19 2.44 2 2D 10.80 9.51 12 3.39 4.17 2.69 Ill 2 22 7.46 U5) 12 3.11 3.07 2.40 2 22 10.59 | 10.13 12 3.55 3.98 Dells) 2 DD, 12.56 | 11.82 12 4.26 4.66 2.74 IV 2 DY 7.83 12 3.01 2.60 2 22 12.58 12 4,29 2.93 PENNIES 5 1'e ti ests ERD Bah Ro ERR RU era Skeets te Mare on, 2.57 + 0.04 Aver ce nonindividialevessels. j\0u. 2.00... 22) oe eee 2.71 Average from previously published data................... 2.80 In the experiments of Tables III and IV marked “no Kjeldahl” the quantities of eggs employed were not determined, so the figures in these experiments cannot be compared with the absolute rates in the others. Also we cannot compare unfertilized and fertilized rates in those experiments, since the suspensions of unfertilized and of fertilized eggs were not of the same concentration. In the other experiments, if we compare the absolute rates of respiration for the unfertilized eggs during the first period of a run, we see no such differences as were ob- tained with Urechis. Here, the eggs used came from freshly collected RESPIRATORY RATE OF UNFERTILIZED EGGS 27s animals. Whether keeping Strongylocentrotus in an aquarium tank would affect the rate of respiration of the unfertilized eggs was not determined. Ciona intestinalis Certain difficulties in handling Ciona eggs were previously (1936) pointed out. In addition it may be noted that to secure large quantities of unfertilized eggs it is necessary to use many individuals and to wash the eggs thoroughly before mixing them, in order to remove the sperm that almost unavoidably comes out with the eggs. Since Ciona eggs tend to float rather than sink to the bottom of the dish, it generally takes about ten or more washings to remove the sperm. Six experiments were run with eggs of Ciona, one at 25° and 15° and five at 22° and 12°. The results are given in Table V. At the temperatures 25° and 15° there are no significant differences between the O,,'s for the fertilized and the unfertilized eggs. In this experi- ment the vessels 4, B and C were removed at the end of the unferti- lized run, the eggs inseminated, and the measurements continued on the fertilized eggs. In addition a fourth vessel, D, of freshly insem- inated eggs was added. Fertilization was 100 per cent in all. At the temperatures 22° and 12° the average Q,,’s are somewhat less for the unfertilized than for the fertilized eggs. Considering the magnitude of the difference in relation to the probable error and the fact that it agrees with Ureciis and Strongylocentrotus in giving at the lower tem- peratures * lower coefficients for the unfertilized eggs, we are inclined to regard it as significant. Comparison of the oxygen consumption figures for the unfertilized and fertilized eggs shows at most a less than two-fold rise upon ferti- lization. This is much lower than in the case of the sea-urchin egg. The rise is more of the order of that found with Nereis eggs (Whitaker, LOS: Barrom, 1932). Dendraster excentricus Four sets of experiments were done on Dendraster eggs, all of them dtiezZ and I22nC, Nhe =results are presented ine Vable Wie “nenone were large enough quantities of eggs obtained to get usable oxygen con- sumption values in less than two hours at each temperature. Dendraster shows a fairly rapid rise with time in the rate of respiration of the un- fertilized egg. The rate rises even more rapidly than in the case of Strongylocentrotus as Fig. 1 illustrates. In the last two hours of a six- hour run we may have more than twice the oxygen consumption ob- 3 Ciona eggs are adapted to a higher temperature range than the others; they fail to divide below 12° and above 26° C. 276 TYLER AND HUMASON tained in the first two hours, as in Experiment II. The temperature coefficients, however, are not particularly affected by the rise. This means simply that the relative rise is about the same at different tem- peratures. The average Q,,’s for the unfertilized eggs by the two methods of calculation are 2.57 and 2.71 respectively. These are slightly lower than the previously determined value of 2.80 for fertilized eggs. The Acie. Wall Average temperature coefficients (Qio) for unfertilized and fertilized eggs. Probable errors given where sample consists of 15 or more values. Coefficients under A are determined from the respiration of eggs in different vessels run simultaneously at different temperatures. Coefficients under B are determined from the respiration of eggs in one vessel run consecutively at different temperatures. Coefficients under C taken from previously published data. Unfertilized eggs Fertilized eggs Tempera- tures No. of No. of 4 ves- A B ves- A B C sels sels Strongylocen- | 22° and 12° 8 2.54 2.53 4 | 2.33 | 2.39 trotus 20° and 10° 23 |2.63 + 0.06/2.57 + 0.05) 4 2.26 | 2.23 | 2.30 18° and 8° 4 2.58 2.59 4 2.69 | 2.57 17.5° and 7.5° 2 2.67 2.69 2 2.79 | 2.80 20° and 5° y) 3.61F 3.747 D 4.224) 4.24+ Urechis 22° and 12° 34 |2.57 + 0.03/2.57 + 0.04) 6 2.64 | 2.65 | 2.79 22° and 12° 12* 2.70 DT 20° and 10° 10 2.46 2.47 2 2.92 | 2.82 | 3.30 18° and 8° 8 2.18 1.94 15° gyavel 52 2 2.96 2.84 2 3.33 | 3.30 Ciona 25° and 15° 3 1.90 1.92 4 DDN BAlah |) DLA 22° and 12° | 19 |1.96 +0.05/1.98 + 0.06] 8 | 2.50 | 2.47 | 2.84 Dendraster 22° and 12° 14 |2.58 + 0.04 Dial 2.80 * Selected experiments, included in line above. t Qis values. difference, however, cannot be taken as significant. The two average values for the unfertilized eggs differ more here than in the previous cases. That is because in two of the experiments (III and IV) coef- ficients could not be determined by the second method (from individual vessels) since the vessels remained at one temperature throughout the run. DISCUSSION The average values of the temperature coefficients for the eggs of the four animals investigated are listed in Table VII. As was pointed out RESPIRATORY RATE OF UNFERTILIZED EGGS DT in considering the individual cases, there are no large differences be- tween the unfertilized and fertilized eggs. At the higher temperatures in each case, there are certainly no significant differences. At the lower temperatures, the consistently lower values for the unfertilized eggs of the different animals incline us to regard the difference as significant. With fertilized eggs, or for that matter most biological material (see Belehradek’s review), Q,, increases as the temperature interval is low- ered. Here, it appears that for the respiration of the unfertilized eggs, Q,,) remains a constant or increases only slightly at lower temperatures. Thus, with Strongylocentrotus we get the values 2.54 at 22° and 12° aude aroonatliZe sande 7.50. Nath Omeehis) the @ 5s ane: 2.57 at 22° and 12° and 2.90 at 15° and 5°. We are not, however, particularly concerned here with the constancy of Q,,. Any other convenient meas- ure of variation in rate with temperature would serve for comparing unfertilized and fertilized egg respiration. We had expected to find the same values for unfertilized as for ferti- lized eggs. At the higher temperatures that appears to be the case. But if we accept the divergence at the lower temperatures as significant, then it would seem that one of the assumptions, upon which this expecta- tion was based, must be wrong. This might well be the assumption that an unfertilized egg is a resting cell exhibiting only maintenance. Other processes besides what we term maintenance may be involved. It would seem important then to determine with certainty whether real differences exist at the lower temperatures. We do not, however, consider it likely that with the present material and methods simply expanding the ex- periments will improve the data very much. Besides, there now appear to be other ways of getting at the questions involved. The unfertilized eggs of all the forms investigated exhibit a rising rate of respiration. In Strongylocentrotus and Dendraster the rate rises much more rapidly than in Urechis and Ciona (see Fig. 1), the differ- ence being correlated with the time of loss of fertilizability on the part of the eggs. Runnstrom (1928, footnote p. 4) has likewise noted that the sea-urchin eggs lose their capacity for fertilization after exhibiting a spontaneous rise in respiration. There are several agents that have been reported to prolong the fertilizable life of the egg; namely, cyanide (Loeb, 1912) thyroxin (Carter, 1931), alcohol and dextrose (Whitaker, 1937). It would be of interest to know whether these agents would prevent the rise in respiratory rate of the aging unfertilized egg. Under such conditions it is quite possible that different values for the tempera- ture coefficients would be obtained. 4 Cyanide and CO suppress this rise according to Runnstrom (1930), but the concomitant fertilization test is not given. 278 TYLER AND HUMASON Rubenstein and Gerard (1934) reported in Arbacia average values of 4.1 and 1.8 for the QO,,’s of unfertilized and fertilized eggs respec- tively. They point out, then, that as the temperature is increased the rise in respiration that occurs upon fertilization in the sea-urchin egg diminishes. By extrapolation they show that at 32° C. there would be no rise. In the sea-urchin, Strongylocentrotus, that we have used, as well as in the three other forms, no such differences in the coefficients are evident. At the lower temperatures, there are possibly significant differences, but in the reverse direction from what the above investigators find. However, if it is assumed that Q,, for the fertilized eggs de- creases as the temperature rises, while O,, for the unfertilized eggs re- mains constant, the unfertilized eggs would presently give the higher values. By extrapolation, then, if there is any point to it, we would find that the temperature, at which there would be no rise in respiration upon fertilization in Strongylocentrotus, approaches that of boiling sea water. SUMMARY 1. The effect of temperature on the rate of oxygen consumption of unfertilized and fertilized eggs of Urechis, Strongylocentrotus, Ciona and Dendraster was investigated. 2. The unfertilized eggs exhibit a rising rate of respiration with time in all four species. The rise is much more rapid in Strongylocentrotus and in Dendraster than in Urechis and in Ciona. ‘This rise appears to be correlated with the loss of fertilizability on the part of the eggs. 3. Methods of determining the temperature coefficients in such a way as to take into account the general rise (which is a significant factor in prolonged runs) and other variations are described. 4. Only the temperature coefficients for the same temperature in- tervals are compared, the respiration being determined at two tempera- tures in each experiment. With Urechis and with Strongylocentrotus eggs, experiments were run at temperatures between 22° and 5°; with Ciona, between 25° and 12°; and with Dendraster, 22° and 12°. The experiments with Urechis and Strongylocentrotus thus included low temperatures at which the fertilized eggs fail to cleave. 5. No significant differences between the temperature coefficients of the respiratory rate of unfertilized and of fertilized eggs of the four animals investigated are found over most of the temperature range in which development is possible. At the lower temperatures, there are differences that are possibly significant, the unfertilized eggs giving con- sistently lower values. 6. Comparison of the absolute rates of respiration of the unfertilized and fertilized eggs shows in Strongylocentrotus and Dendraster the rise RESPIRATORY RATE OF UNFERTILIZED EGGS 279 in respiration upon fertilization typical of the echinoids; in Ciona a less than two-fold rise is manifest; in Urechis, the rate may rise consider- ably, remain constant or decrease slightly, depending upon the particular batch of eggs employed. Eggs from animals kept some time in cap- tivity give lower unfertilized rates and manifest a rise upon fertilization ; eggs from freshly collected animals give higher unfertilized rates and no rise or even a slight decrease upon fertilization. IIMA IN UNees, (CMU SB) Barron, E. S. G., 1932. Studies on cell metabolism. I. The oxygen consumption of Nereis eggs before and after fertilization. Biol. Bull., 62: 42. BELEHRADEK, J., 1935. Temperature and living matter. Protoplasma Mono- graphien, vol. 8. ; Carter, G. S., 1931. Iodine compounds and fertilization. III. The fertilizable life of the eggs of Echinus esculentus and Echinus miliaris. Jour. Exper. Biol., 8: 194. Logs, J., 1912. The Mechanistic Conception of Life. University of Chicago Press. RuBENSTEIN, B. B., anp R. W. Gerarp, 1934. Fertilization and the temperature coefficients of oxygen consumption in eggs of Arbacia punctulata. Jour. Gen. Physiol., 17: 677. RunwNstroM, J., 1928. Die Oxydationserhohung bei der Entwicklungserregung des Seeigeleies. Arkiv. Zool., 20 (Art. 3): 1. Runnstrom, J., 1930. Atmungsmechanismus und Entwicklungserregung bei dem Seeigelei. Protoplasma, 10: 106. Tyzer, A., 1936. On the energetics of differentiation. IV. Comparison of the rates of oxygen consumption and of development at different temperatures of eggs of some marine animals. Bziol. Bull., 71: 82. Warsure, O., 1915. Notizen zur Entwicklungsphysiologie des Seeigeleies. P/flii- gers Arch., 160: 324. Wuitaker, D. M., 1931. On the rate of oxygen consumption by fertilized and unfertilized eggs. III. Nereis limbata. Jour. Gen. Physiol., 15: 191. Wuirtaker, D. M., 1933. On the rate of oxygen consumption by fertilized and unfertilized eggs. V. Comparison and interpretation. Jour. Gen. Physiol., 16: 497. Wuitaxker, D. M., 1937. Extension of the fertilizable life of unfertilized Urechis eggs by alcohol and by dextrose. Jour. Exper. Zod6l., 75: 155. SEXUAL AND ASEXUAL REPRODUCTION IN EUPLANARIA TIGRINA (GIRARD) ROMAN KENK (From the Miller School of Biology, University of Virginia, and the Department of Zoélogy, University of Ljubljana, Yugoslavia) Several authors have reported on the reproduction of Euplanaria tigrina (Synon. Planaria maculata Leidy), one of the commonest fresh- water planarians of North America. It has been known for a long time that this species often reproduces by fission. Curtis (1902), in his valuable paper on the life cycle of this form, also states that, at least in several localities, the animals grow sexually mature and deposit egg capsules in the early summer. Much material on the processes of fission has been gathered by experimental workers, particularly by Child and his co-workers. From these we know that the rate of fission may be controlled by external factors, such as temper- ature, the amount of food, and chemical properties of the water. Nevertheless, the relations between the two manners of reproduc- tion, the sexual and the asexual one, are as yet little known. The only extensive observations with regard to this question were under- taken by Curtis on material from the vicinity of Woods Hole, Massa- chusetts. This author investigated the life history of the species in four different localities through the course of three years and sum- marized his results as follows: “‘In some localities the species seems to have reproduced exclusively, so far as the observations go, by fission, in others only by the sexual process, while there are still others where both processes occur at different seasons”’ (1902, p. 556). In a later communication (Curtis and Schulze, 1924, p. 105) he writes: “It may also be noted that the differing habits of reproduction, originally reported . .. for the P. maculata in four different localities near Woods Hole, Massachusetts, have been confirmed by all our subsequent collecting in these localities.”’ These differences in the manner of reproduction in certain places suggested two possible interpretations: (1) there might be some differ- ences in the physical and/or chemical characteristics of these localities, which influence the life cycles of the animals living there; or (2) there might be more than one physiological race of Euplanaria tigrina, showing different habits of reproduction. If the latter be the case, 280 SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 281 the manner of reproduction would be determined, chiefly, by internal factors. It is the purpose of this paper to report on a series of experiments performed to analyze several factors that might control these processes. Most of the work has been carried out at the Miller School of Biology, University of Virginia. I wish to express my indebtedness to the Rockefeller Foundation which made this investigation possible, as well as my sincere gratitude to Professor William A. Kepner and Professor Ivey F. Lewis for the privileges extended to me during my stay at the University of Virginia. MATERIAL AND METHODS The animals used in the experiments were collected in four locali- ties: (1) Sinclair’s Pond, Park Street, Charlottesville, Virginia; (2) Big Spring, near Kerr’s Creek, Rockbridge County, Virginia; (3) Mary’s Lake, Naushon Island, Massachusetts; and (4) Pond south of Main Street, behind the Episcopal Church, Falmouth, Massachusetts. In the experiments on asexual reproduction, the influence of temperature, hydrogen-ion concentration, amount of food, and ir- radiation with ultra-violet rays upon the rate of fission were in- vestigated. The cultures were run in tap water. Temperature experiments were performed in three series of cul- tures: at indoor temperature, low and high temperatures. The low temperature cultures were kept in electric, thermostatically controlled refrigerators, while high temperatures were obtained in simple, electri- cally heated ovens. The experiments on the influence of the hydrogen-ion concentration in the culture water were carried out at indoor temperature. In order to keep the pH constant, small quantities (1:20) of buffer mixtures were added to the water. The buffers used were mixtures of KH,PO, and NaOH having a pH of 6.4, 7.0, and 7.6, prepared after Clark’s formule. Their addition to the tap water kept the media at a fairly constant acidity of respectively 6.5, 7.0, and 7.5. The animals were fed with beef liver at regular intervals; it has been proved by various workers that cultures of fresh-water planarians may be run for years on this food. The liver was freely taken. Ultra-violet irradiation was carried out by means of a Hanovia mercury-arc lamp kindly placed at the writer’s disposal by the Com- mittee on Effects of Radiation, National Research Council. These experiments did not yield any results bearing upon the process of re- production; a more detailed description of the technique employed may, therefore, be omitted. 282 ROMAN KENK EXPERIMENTS ON ASEXUAL REPRODUCTION For these experiments animals from Sinclair’s Pond and from Big Spring were used. Each experimental series comprised at least three cultures containing, at the beginning, 5 to 10 specimens each. When fissions occurred, the posterior pieces were eliminated from the cul- tures. It is well known that the two pieces derived from fission, the head and the tail piece, behave differently in the regeneration of the missing parts. The head piece does not undergo any considerable change, its prepharyngeal part remains almost the same size, while the posterior end appears to grow out of it; the head piece, therefore, regenerates chiefly by epimorphosis. The tail piece, however, shows decided morphallaxis, i.e. it rearranges its proportions and a large part of the old tissues is directly transformed into parts of the missing prepharyngeal region (cf. Curtis, 1902, p. 529). Moreover, in a short time, before the regeneration is completed, the tail piece may redivide spontaneously. In head pieces, the intervals between two fissions are more regular. We may, with the necessary caution, take these inter- vals as indicative of the rate of reproduction. We must, however, realize that an absolute regularity in fission cannot be assumed. Although we select specimens of the same origin, the same age, the same size, and keep them in the same aquarium, yet a great variety of fission intervals will be observed. This is obvious in view of the fact that we cannot control all the factors concerned in the induction of fission. Among these, the nutrition of the individual specimens and their locomotory activity play important réles. To eliminate the fluctuations of the fission intervals as much as possible, a sufficient number of specimens should be used for each experimental series and the cultures maintained for a sufficiently long time. In order to express the frequency of fission in a convenient way, two data were calculated for each experimental series: the average interval between two consecutive fissions, and the average fission rate. The latter is the number of fissions per day, calculated for a lot of 100 specimens. It is a disagreeable fact that, in cultures running for a long period of time, the numbers of specimens can hardly be main- tained constant, despite careful handling of the animals. The animals are apt to crawl up the wet walls of the culture dishes and then dry up if not noticed in time. Occasionally, individual specimens are lost while one is feeding and changing water. Therefore, the fission inter- vals and the fission rates were determined for shorter periods (usually 10 days) and from these data the average intervals and rates cal- culated for the entire length of observation. SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 283 Temperature Experiments The indoor temperature varied considerably during the course of the experiments. In general, it fluctuated between 20 and 30° C., occasionally, for a short time, dropping below these limits (minimum 13.8° C., March 7, 1932) or rising above them (maximum 32.6° C., August 8, 1932). The results obtained in parallel cultures at indoor temperature and in the refrigerator are shown in Table I. At low temperatures (10—12° C.) the processes of fission were greatly inhibited or even entirely suppressed for several months. In a culture of 20 specimens (the number decreased, during the experiment, to 18) running 330 days, only five fissions occurred, the first of them on the 267thday. This shows that at least 13 specimens for almost 11 months TABLE I Low-temperature experiments. Average fission rate and fission interval in cultures kept in the refrigerator (10—-12° C.) compared with those of control cultures at indoor temperature. Number of specimens Duration Average Average Origin of material Temperature | of experi- fission fission ce ee ment rate interval Initial | Final a. days days Sinclair’s Pond......... 20 18 | Refrigerator] 330 .084 1189 Sinclair’s Pond......... 15 4 | Indoor 341 5.9 16.9 IBIOP SPRING a) Wis a oike es oi 18 18 | Refrigerator] 329 all5 658 PISO PLUG ys ciate erie ss 18 2 | Indoor 329 UP 13.9 Biker Syoyinlers 4.4 sede oe ae 15 14 | Refrigerator} 174 041 2436 Bree Spring x22... < -ais'a.e\s 15 11 | Indoor 174 6.0 16.7 did not display fission at all. The animals were well fed, so the de- crease of the fission rate could not be due to starvation. On the con- trary, in the culture referred to, the length of the animals was from 5 to 6 mm. at the beginning of the experiment, while at the end lengths of 16 to 20 mm. were measured. The control animals, raised at indoor temperature, generally remained smaller, since they divided before attaining the size of the refrigerator specimens. Another factor that might be suspected of preventing fission is the decreased locomotory activity in the refrigerator culture. The stimulating effect of light was eliminated in both series of cultures by keeping the aquaria in the dark. There is, however, a decrease of the activity caused directly by the low temperature. Nevertheless, since the animals in the re- frigerator were often seen to move about in the aquaria, this difference 284. ROMAN KENK can hardly be responsible for the extreme rarity of divisions at low temperature. Several cultures were run at high temperature which varied be- tween 29 and 34.5° C., in the main amounting to about 32°C. They showed a significant change in fission frequency during the course of the experiment (Table II). At the beginning, the fission rates were TABLE II High-temperature experiments. Fission rate in cultures kept in the thermostat (about 32° C.), compared with those in control cultures at indoor temperature. Number of Fission rate specimens Origin of material Temperature Initial | Final 1st-—18th day | 19th—36th day} 37th—55th day Big Spring..... 14 8 | Thermost. 4.8 o2.8 81 Big Spring..... 2 11 Indoor 4.2 2.3 4.3 similar to those of the control cultures at indoor temperature. Soon the divisions became rarer and finally, after 43 days, no fissions occurred any more. Parailel with this decline in reproduction was a reduction of the size of the animals. The lengths of the specimens decreased from 7-13 mm. at the beginning to 3-5 mm. when the experiment was discontinued (after 55 days). Unfortunately, I had no opportunity to determine exactly the optimum temperature for asexual reproduction. Nevertheless, ob- servations on cultures at indoor temperature indicated that fissions were most frequent when the water temperature was about 25° C. or a little over. Hydrogen-ion concentration The observation of fissions in media of different acidity comprised only a small range of pH, viz. 6.5, 7.0, and 7.5. The addition of buffer mixtures to the culture water had no injurious effects on the animals. The reproduction continued in a normal way. No significant differ- ences in the fission rates were observed in cultures of different con- trolled acidity, nor between them and the cultures in tap water (Table III). It may be noted, however, that the range of pH used was relatively narrow. It is known that planarians tolerate a considerable variation of acidity, towards the acid as well as towards the alka- line side. SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 285 TABLE III pH experiments. Average fission rate and fission interval in buffered culture media and in tap water. Number of specimens Duration Average Average Origin of material pH of experi- fission fission Pegnren Pee Tok ment rate interval Initial | Final days days Bigg Spree taste eces-.- 15 11 6.5 168 8.2 12.3 [Biter Syormtnes ce oocaceaaas 15 13 7.0 168 6.3 15.9 BICC OPRING. | fe esse eee 15 10 US 174 7.4 13.5 IBICES PRINS acon his ses ss tS 11 tap w. 174 6.0 16.7 Irradiation with Ultra- Violet Light A number of specimens were exposed to the light of a quartz- mercury lamp on several consecutive days. During irradiation, the animals were kept in shallow dishes with little water. The animals so treated showed a slight decrease of the fission rate (Table IV). On TABLE IV Fission rates and fission intervals in cultures exposed to different doses of ultra- violet radiation and in control cultures not irradiated. Number of specimens Duration Average Average Origin of material Dose of experi- fission fission Eno ment rate interval Initial | Final days days Big Spring..... 15 7 Single 174 4.4 22.8 Big Spring..... 15 2. Double 174 4.6 Pill Big Spring..... 15 11 —— 174 6.0 16.7 the other hand, the irradiation apparently exerted a general injurious effect on the animals, which resulted in a high mortality in these cultures. I am, therefore, inclined to attribute the decrease of the fission frequency to the weakening of the animals rather than to a specific effect of the ultra-violet irradiation. Starvation It is well known that planarians can stand long periods of starvation very well. In prolonged starvation they grow smaller and, to a certain extent, simplify their anatomical structure. Their physiological condition becomes that of young animals. It has been proved re- 286 ROMAN KENK peatedly that starvation causes asexual (as well as sexual) reproduction to cease. Thisisshown very clearlyin Table V. For our experiments, well-fed specimens of a length of 12 to 15 mm. were taken. During the first week of starvation, fissions took place at an almost normal rate; later on, the frequency of division decreased rapidly. The last fission occurred on the twenty-second day after the starting of the experiment. On the sixty-fifth day the culture was discontinued. The size of the animals had decreased to 344-6 mm. I wish to point out that no traces of sexual reproduction or maturity were ever observed in any of the cultures of Euplanaria tigrina from TABLE V Fission rates in cultures starving and in control cultures well fed. Number of ele specimens Fission rate Origin of oe astavial Nutrition - te . 1st—10th 10th—20th 20th—30th 30th-65th Initial | Final day day dace aay Big Spring | 30 21 | Starving Os 2.4 a3 .0 Big Spring| 30 | 26 | Fed 12.0* 4.0 4.3 5.1 * The high fission rates during the first days were due to the fact that fully grown specimens had been selected for the experiment. Sinclair’s Pond and Big Spring, either in the stock aquaria or among the specimens subjected to various external conditions. EXPERIMENTS ON SEXUAL REPRODUCTION Since sexuality could not be induced in material from two localities in Virginia, animals were procured from those places where Curtis (1902) had made his observations on the reproduction of the species. In July, 1932, asexual animals were collected in Mary’s Lake; at the same time specimens brought in from the Pond in Falmouth proved to be sexually mature and laying egg capsules. Animals from these localities were kept in separate culture dishes, but under external condi- tions as identical as possible. They were raised at indoor temperature and fed on beef liver. As a result, the two lots retained, in the main, their characteristic manners of reproduction also in the laboratory. The animals from the asexual locality continued to undergo fission. There were no decided seasonal cycles of reproduction observed, apparently because of the SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 287 relatively favorable temperature all the year round. No sex organs developed in the course of 5 years.! The animals from the sexual locality continued to propagate sexu- ally in more or less regular periodical cycles. Though a definite coin- cidence of the periods of reproductive activity in the laboratory cul- tures with those observed in the field need not be expected, neverthe- less the periodicity conformed, in the main, with Curtis’ records of the life history in the natural habitat. The duration of the egg-laying periods in indoor cultures varied within wide limits. Usually the animals started depositing the first egg capsules in the late fall or in winter (middle of November to February). The breeding-reached its height in March and April, then declined, and generally ceased in June. Only once did I observe, in one culture, a breeding season extending from July till the beginning of September. Each specimen deposited several cocoons during one season. The size of the body gradually decreased from about 18-24 mm. to 13-15 mm. After the breeding season was over, the animals were rather sensitive, as if exhausted. The rate of mortality seemed to be higher at that time in the cultures. Nevertheless, most of the animals survived and recovered completely, if well attended to, and in the next egg-laying season again proceeded to produce cocoons. The individual lifetime of this species may be estimated at several years. Curtis’ data on the degeneration of the sex organs between two seasons of sexual activity were confirmed. Temperature Experiments Like the fission of the asexual form, the rate of breeding in the sexual form is influenced by the temperature of the medium. This is clearly shown in Table VI. At low temperature the laying of cocoons proceeded very slowly; it was almost inhibited; at high temperature the breeding rate increased. On the other hand, the breeding season lasted much longer at low temperatures than at high ones. In the refrigerator culture referred to in Table VI, maintained at 10° C., single egg capsules were deposited at long intervals and the animals remained in the sexual state for at least one year. 1One single seeming exception was noticed when, in March, 1933, one sexual specimen appeared in the asexual culture and laid two cocoons. This specimen was subsequently eliminated from the aquarium. Since then and up to now (May, 1937) no second case of sexuality has occurred. I am inclined to assume that this one and only mature animal had, by careless handling, been transferred to this vessel from the sexual culture while the dishes were being cleaned. Even though this assumption were not correct, the occurrence of one sexual individual among many hundreds of asexual animals would not be significant, and could not essentially affect the result of the experiment. 288 ROMAN KENK Besides reproduction by cocoons, the animals from the sexual locality also showed asexual reproduction by fission. Fission did not occur in all animals of the cultures and was confined to a short season of the year, from June to August. As this season followed the season of egg-laying, fissions were usually observed in animals which had previously deposited egg capsules. Young animals, hatched from cocoons in the preceding spring, likewise occasionally divided asexually, provided they had already reached a sufficient size and had not yet de- veloped sex organs. From the fact that the fissioning season coincides with the warmest season of the year, we may conclude that fission requires a high temper- ature. Temperature, however, is not the sole decisive factor: the animals must, at the same time, be in the state of sexual inactivity. If they are subjected to high temperature during the breeding season, they continue to deposit cocoons and do not divide. TABLE VI Temperature experiments on material from the pond at Falmouth. Daily breeding rate (calculated for 100 specimens) and breeding interval in cultures kept at indoor temperature, compared with those kept at lower and higher temperatures. Number Duration Number Average Average Temperature of of of breeding breeding specimens | observation cocoons rate interval days days Indoor (about 20° C.).. 20 31 29 4.68 21.4 Refrigerator (about 10° C. yes 19 147 10 36 279.3 Thermostat (about 28.6° C.) 9 35 98 31.1 3.21 Two tail-pieces of specimens which had fissioned after they had stopped laying egg capsules were studied anatomically. In both of them remainders of the copulatory organs were found, but in a state of apparent degeneration. The genital pore, parts of the atrial cavity, and the penis could still be identified, though their structure differed from that of the organs in function. Bursa and bursa stalk had entirely disappeared. In various places in the parenchyma of the genital region there were patches of brown substance, apparently the product of disintegration of tissues. In short, the pieces exhibited that typical picture of degeneration of sex organs which had been described by Curtis (1902, pp. 546-550) as occurring after the egg- laying. I should like to emphasize again that fission, in the laboratory cultures, did not occur in the case of every individual specimen. The animals that did undergo fission usually divided only once. After SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 289 the tail-piece had grown to a certain size, it occasionally divided a second time if the conditions were still favorable. In any case, fission plays only an insignificant réle in the life history of the planarian from the pond at Falmouth. The tendency to divide can be easily controlled by external factors, particularly those that influence the fission rate in the asexual form from Big Spring and Sinclair’s Pond, i.e., temperature and nutrition. The following experimental series may illustrate this statement: April 27, 1936, three cultures of 20 specimens each were prepared. The animals measured 18 to 24 mm. in length and were at the height of the breeding period. Culture 1. Kept at indoor temperature, fed twice a week with beef liver. Egg-laying ceased June 12, fissioning started June 19 and lasted until August 3. Thirteen animals had divided, 7 remained undivided. These, measured September 21, were from 13 to 15 mm. in length. They entered a new breeding season on March 8, 1937. Culture 2. Kept at indoor temperature, not fed. The last egg capsule was deposited on May 11; fissions occurred between June 30 and July 22. Only 3 specimens had fissioned, 15 were undivided, 2 had died. September 21, the animals measured 5 to9 mm. From that time on they were regularly fed. The new breeding period started February 18. Culture 3. Kept in a refrigerator at 10° C. Cocoons were laid at long intervals,—only 10 cocoons up to September 21, when the animals measured 15 to 21 mm. The egg-laying continued. It is interesting to note that sexually active animals which had been kept at low temperature for a long time showed an extraordinary increase of the breeding rate when transferred to indoor temperature. This appeared in the continuation of the refrigerator experiment re- ferred to in Table VI. Ten specimens which had been exposed to a temperature of 10° C. for 147 days had shown an average breeding rate of .36 (breeding interval of 279.3 days). Transferred to indoor temper- ature, they laid during the course of two months no less than 65 cocoons. This corresponds to a daily rate of 10.5 cocoons per hundred specimens or to a breeding interval of 9.5 days. Animals constantly kept at indoor temperature propagate at a much slower rate (see Table VI). From this experiment it would appear that low tempera- tures, while retarding the deposition of egg capsules at the time, do not inhibit the maturing of the germ cells to the same extent. Brought into normal temperature conditions the animals react to the accumula- tion of egg cells (and yolk material) in the body by an increased breeding activity. 290 ROMAN KENK DISCUSSION AND CONCLUSIONS From the foregoing paragraphs it is evident that there must be at least two physiologically different races of Euplanaria tigrina. They differ, mainly, in their respective habits of reproduction. We may call them, shortly, the ‘‘sexual” and the ‘‘asexual’”’ race. The differ- ent life cycles of this species in different localities near Woods Hole, Massachusetts, reported first by Curtis (1902 and 1924), are due chiefly to differences inherent in the animals themselves, not to physico- chemical properties of the environment. The asexual race lives in Sinclair’s Pond, Big Spring, and Mary’s Lake; the sexual one in the pond at Falmouth (see p. 281). Asexual reproduction has been observed in a relatively large number of fresh-water triclads. An excellent review of the material gathered has been given by Vandel (1921, pp. 370-374), who also assumed the existence of different races of Euplanaria tigrina, in which the manners of reproduction were hereditarily different. Few observations, how- ever, have so far been made on the entire life cycles of fissioning planarians and little has been done to investigate the processes of reproduction in an experimental way. Nevertheless, it is a striking fact that in almost all species subjected to closer examination, differ- ences in the incidence of sexual and asexual reproduction have been revealed in different localities (cf. Benazzi, 1936, p. 364). We know such physiological races in all European species of fissioning planarians: Polycelis felina (Dalyell) (Vandel, 1921, pp. 478-479; Thienemann, 1926, pp. 298-300), Crenobia alpina (Dana) (Vandel, 1921, pp. 478- 479), Euplanaria gonocephala (Dugés) (Vandel, 1925; Benazzi, 1936), and Fonticola vitta (Dugés) (Beauchamp, 1932, pp. 285-294). Among the American species, sexual and asexual forms are known in Eu- planaria dorotocephala (Woodworth) (Kenk, 19356, p. 451). The same phenomenon is now confirmed in Euplanaria tigrina. There is no reason to distinguish these races as separate taxonomic units, e.g. species or subspecies. Small morphological differences between them may occur, particularly in the shade of coloration, the size, and the proportions of the body; they are, however, of little tax- onomic value. Some of these characteristics, such as the size and shape of the animals, may be directly correlated with the mode of reproduction. The reproduction of the asexual form of Euplanaria tigrina is exclusively asexual. In this the field investigations of Curtis and the experiments described in this paper agree. External factors may accelerate the processes of propagation, retard, or even inhibit them, but they cannot change the process to a sexual one. The main factors SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 291 influencing the rate of fission are temperature and nutrition. At low temperatures division is suppressed, it takes place only at about 10° C. or more, provided the animals have attained a certain minimum size. The interval between two fissions is shorter, the higher the temperature. During this interval, the anterior piece at least has to regenerate the posterior portion and therefore the nutrition of the animal plays an important réle. The optimum temperature, though not exactly established, appears to be between 25 and 28° C. Above this temper- ature, the fission rate decreases. This decline probably is not the result of a reduced tendency to fission, but a sign of defective nutrition: at high temperatures the metabolism of the body is increased and the food taken does not suffice to cover the energy needs of both mainte- nance of body size and regenerative growth necessary to prepare the animal for the next fission. The animal is therefore in a state of inanition. This is seen also in the rapid decrease of the size. The maximum temperature tolerated continuously is a little above 30° C. (Lillie and Knowlton, 1898, determined this temperature for ‘‘ Planaria torva”’ to be 32° C.; the species used in their experiments was probably identical with Euplanaria tigrina.) There appears to be a correlation between the temperature and the size of the body at which fission takes place: at low temperatures the animals do not divide until reaching a considerable length, while at higher temperatures fission occurs already in smaller animals. This seems to apply to all fissioning planarians so far investigated. This phenomenon has not yet been subjected to a comparative study on a larger scale. Castle (1928, p. 420) e.g., observed it in another Ameri- can planarian, Fonticola velata. In their natural habitats, the life history of the asexual race of Euplanaria tigrina is simple: no reproduction in winter, fission during the warmer seasons. This cycle has been observed by Curtis (1902, p. 517) in ‘‘locality 4." Moreover, Taliaferro’s (1920, p. 63) planarian from an abandoned ice-pond near the University of Virginia, in which reproductive organs were not observed for nine years, apparently belongs to the same race. This applies also to Hyman’s (1920, p. 405) material from ‘‘the lagoon in Jackson Park in the city of Chicago”’ (p. 404). The sexual race of Euplanaria tigrina, on the other hand, is capable of both sexual and asexual reproduction. Generally speaking, it develops reproductive organs and deposits egg capsules. Even if conditions be favorable throughout the year, sexual activity is not continuous, but occurs in certain physiological cycles. These cycles, in the field as well as in the laboratory (at indoor temperature), 292 ROMAN KENK correspond to the seasons of the year. Apparently the yearly fluctua- tions of temperature are the decisive factor. The egg-laying season lasts several weeks or months. It goes on until the animals, especially their sex organs, are exhausted. Then follows a period of sexual inactivity during which the complicated copulatory apparatus first degenerates, to be newly reconstructed before the next breeding season. — A detailed account of these processes has been given by Curtis (1902). During the period of sexual inactivity the animals may undergo fission. This occurs only during the warm season of the year and is not at alla regular phenomenon. Many specimens remain undivided. The regular life cycle can be influenced, particularly by the tem- perature. Low temperature retards the laying of cocoons, high temperature accelerates it. The formation of egg capsules appears to demand a certain minimum temperature, which lies not far below LOC i These data derived from laboratory cultures agree, in the main, with Curtis’ observations in the field. Curtis (1902, p. 517) reports that Euplanaria tigrina in Locality 1 near Woods Hole, Massachusetts, lays egg capsules during May and June. After that, reproduction ceases and the reproductive organs degenerate. In August and Sep- tember, i.e. in the season when the water is warmest, the animals are found reproducing by fission. Then reproduction is again suspended and a regeneration of the reproductive organs takes place during the winter. In Locality 2, according to this author, the life cycle is similar to that in Locality 1, except that no fissions were observed there. There are several possible explanations of this fact: (1) the temperature may not rise so high as in Locality 1, i.e. it may not reach the minimum temperature necessary for fission in the asexual state; (2) the popula- tion in this place may belong to a physiologically different race with a fission temperature higher than that of the race from Locality 1 or even with the tendency to fission suppressed; and (3) there may be less food available in Locality 2, wherefore the animals may not reach the necessary size for fission. It would be desirable to investigate this question on the spot. It is interesting to compare the data concerning Euplanaria tigrina with those observed in regard to other fissioning planarians. In cases where both manners of reproduction, the sexual and asexual, alternate according to the seasons of the year, the asexual phase always coincides with the warmer season. The majority of the species concerned develop sex organs preferably in the winter and early spring: Polycelis felina (Dalyell) and Crenobia alpina (Dana) (according to various SEXUAL AND ASEXUAL REPRODUCTION, EUPLANARIA 293 authors); Euplanaria dorotocephala (Woodworth) (cf. Hyman, 1925, p. 65); E. gonocephala f. subtentaculata (Draparnaud) (see Draparnaud, 1801, p. 101, and Vandel, 1925, p. 502); Fonticola morgani (Stevens and Boring) (see Kenk, 1935a, p. 102); F. velata (Stringer) (see Castle, 1928, p. 419). In the natural habitats this rule is often obscured by two facts: First, certain localities have an almost constant temperature all the year round, e.g. springs, deep lakes, and subterranean waters; in these habitats often no alternation of reproduction takes place, though the animals in other surroundings would be capable of both sexual and asexual propagation. Secondly, there often occur physio- logical races, in which the tendency to sexual or asexual reproduction differs to a considerable extent (cf. Vandel, 1921, p. 478) and either one of them may be entirely suppressed; these latter forms, of course, are not considered here, since they have no alternation of reproduction. The sexual race of Euplanaria tigrina appears to be different from the other fissioning forms in so far as it becomes sexually active in the warm season of the year. This difference is, however, only apparent. Here, as well as in the others, fission takes place at a higher temperature (August, September) than sexual activity requires (May, June). Only, the minimum temperatures necessary for either kind of repro- duction are comparatively high. It is a matter of further investigation to decide whether this analysis of the processes of reproduction, carried out for two races of Euplanaria tigrina, is valid for other forms of this species as well. It may be as- sumed that additional material will furnish a still greater variety of physiological characteristics. SUMMARY 1. Euplanaria tigrina occurs in at least two physiological races which differ in the manner of reproduction: a sexual and an asexual race. 2. The asexual race, according to observations covering several years, reproduces exclusively by fission. Temperature and nutrition control the rate of fission but do not induce sexuality. 3. The sexual race periodically develops reproductive organs and lays cocoons. After the breeding period has ceased, the sex organs degenerate and fission may occur at high temperature. The in- dividual animal can outlive several periods of sexual activity. 4. In nature, all planarians that have alternating (sexual and asexual) reproduction, appear to propagate sexually during the colder season and asexually during the warmer season of the year. 294 ROMAN KENK LITERATURE CITED BEAUCHAMP, P. DE, 1932. Biospeologica. LVI. Turbellariés, Hirudinées, Bran- chiobdellidés. Deuxiéme série. Arch. Zool. expér. gén., 73: 113. BenaAzzi, M., 1936. Razze fisiologiche di Euplanaria gonocephala differnziate dalla diversa attitudine alla scissiparita. Rend. R. Accad. Lincet (Roma), Cl. Sc. fis., 23: 361. CasTLE, W. A., 1928. An experimental and histological study of the life-cycle of Planaria velata. Jour. exper. Zool., 51: 417. Curtis, W. C., 1902. The life history, the normal fission, and the reproductive organs of Planaria maculata. Proc. Boston Soc. Nat. Hist., 30: 515. Curtis, W. C., anp L. M. Scuuuzz, 1924. Formative cells of planarians. Anat. Rec., 29: 105. DRAPARNAUD, J., 1801. Tableau des Mollusques terrestres et fluviatiles de la France. Montpellier et Paris, (1801). Hyman, L. H., 1920. Physiological studies on Planaria. IV. A further study of oxygen consumption during starvation. Am. Jour. Physiol., 53: 399. ~ Hyman, L. H., 1925. The reproductive system and other characters of Planaria dorotocephala Woodworth. Trans. Am. micros. Soc., 44: 51. KENK, R., 1935a. Studies on Virginian triclads. Jour. of the Elisha Mitchell Scient. Soc., 51: 79. Kenk, R., 1935d. A morphological proof of the existence of zooids in Euplanaria dorotocephala. Physiol. Zool., 8: 442. LiLLiz, Fr. R., AND F. P. KNowiTon, 1898. On the effect of temperature on the development of animals. Zool. Bull., 1: 179. TALIAFERRO, W. H., 1920. Reactions to light in Planaria maculata, with special reference to the function and structure of the eyes. Jour. exper. Zool., 31: 59. THIENEMANN, A., 1926. Hydrobiologische Untersuchungen an den kalten Quellen und Bachen der Halbinsel Jasmund auf Riigen. Arch. Hydrobiol., 17: 221. VANDEL, A., 1921. Recherches expérimentales sur les modes de reproduction des Planaires triclades paludicoles. Bull. biol. France Belgique, 55: 343. VANDEL, A., 1925. Planaria subtentaculata Drap. n’est qu’une race asexuée de Planaria gonocephala Dugés. Bull. biol. France Belgique, 59: 498. INVESTIGATIONS AS TO THE LOCALIZATION OF THE MICROMERE-, THE SKELETON-, AND THE ENTODERM-FORMING MATERIAL IN THE UNFERTILIZED EGG OF ARBACIA PUNGRULA: SVEN HORSTADIUS (From the Zoétomical Institute, University of Stockholm, and the Marine Biological Laboratory, Woods Hole, Massachusetts) I. INTRODUCTION The available information about the localization of micromere-, entoderm- and skeleton-forming material, and about the stability of the egg-axis in the uncleaved egg of the sea urchin is not consistent. Driesch (1896, 1898, 1899, 1900) studied the cleavage of fragments (ob- tained by shaking) of unfertilized and fertilized eggs of Echinus. He came to the conclusion that the formation of micromeres is due to the cytoplasm, not to the nucleus. By way of explanation he suggested ‘ein polar-bilaterales Gerichtetsein der Teilchen ”’ of the whole egg, but he also spoke of a certain local specific structure of the cytoplasm. Morgan (1894) found in part a migration of pigment, in part a forma- tion of micromeres at the vegetative pole of Arbacia, even when the furrows were formed in an atypical sequence. Boveri (19010) called attention to the vegetative polar cap and its specific qualities, and Horstadius (1928, p. 14) confirmed the results of Boveri: vegeta- tive and meridional halves of Paracentrotus may show a typical cleav- age, whereas the animal half does not form any micromeres. By removing vegetative fragments of different sizes from the unfertilized Paracentrotus egg, Horstadius (loc. cit., p. 18) proved that the po- tentiality of forming micromeres gradually decreases from the vegeta- tive pole towards the animal and ceases about halfway between the normally micromere-forming area (unpigmented) and the equator. The micromere-forming region is thus restricted to the unpigmented pole-cap and the lower half of the pigment ring material in Para- centrotus. (Boveri (19010, p. 155) also saw a pigmented micromere.) Furthermore, Hoérstadius (loc. cit., p. 15) found, by aid of the pigment band, that the cleavage axis in fragments remained unchanged, the micromeres both in meridional and vegetative fragments being formed from the unpigmented pole-cap. Whereas Driesch (1900 etc.) in his early papers, spoke of all parts of 295 296 SVEN HORSTADIUS the sea urchin or starfish egg as equipotent, Zoja (1895), Terni (1914), v. Ubisch (1925) and Hoérstadius (1928) found that animal halves of cleavage stages could not gastrulate, nor form any skeleton. Boveri (19010, p. 158, 1902) used the pigment ring of Paracentrotus to study whether the polarity remains unchanged in fragments of unfertilized eggs. Animal fragments (without pigment) did not gastrulate. One vegetative and one meridional fragment gave larve in which the pig- mented and unpigmented regions were differentiated in such a way as to show that the polarity of the fragments was not altered. Haér- stadius (1928, p. 33) confirmed these results. Animal halves of both fertilized and unfertilized eggs of Paracentrotus developed in the same ways as animal halves of 8- or 16-cell stages. They do not gastrulate, nor do they form skeleton. Meridional and vegetative fragments, on the other hand, invaginate archentera and produce spicules. Thus not only the micromere-forming, but also the archenteron- and skele- ton-forming material is restricted, in the unfertilized Paracentrotus egg, to the vegetative half. Isolation of animal and vegetative fragments of the 64-cell stage in a plane corresponding to the middle of the pig- ment ring region (between veg: and vege, Hérstadius, 1935, p. 319) demonstrated that the upper level of the skeleton-forming area cor- responds roughly to that of the micromere-forming material in the uncleaved egg. This upper limit of the skeleton material was not determined in detail in the unfertilized egg (cf. above as to the micro- mere-forming material). By aid of the pigment ring Ho6rstadius (19362) also showed in agreement with Boveri, that the polarity of _ fragments of the unfertilized eggs does not change. These investigations on Paracentrotus have proved that both the micromere-, entoderm-, and the skeleton-forming material is localized in the vegetative part, roughly speaking, the vegetative quarter, of the unfertilized egg, and that the polarity of the egg remains unchanged in fragments, both as regards the cleavage axis and the differentiation axis. Some results obtained by other investigators on different material conflict considerably with this view. Harnly (1926) found the cleavage pattern of fragments of unfertilized eggs of Arbacia dependent, not upon the orientation of the plane of fragmentation in relation to the egg-axis, but in relation to the nucleus. Boveri (1901a) stated that the pronucleus of the mature egg of Paracentrotus may have any posi- tion in relation to the egg-axis. Harnly (1926) and Hoadley (1934) found the same thing in Arbacia. Thus neither the animal, nor the vegetative half is noticeably preferred. From the different cleavage patterns resulting on dividing eggs in different planes with relation to iti a LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 297 the nucleus, Harnly comes to the conclusion that the micromere- forming material prior to fertilization lies in the region between the pronucleus and the center of the egg. A nucleated fragment as large as half an egg, or larger, cleaved as does the whole egg. When an egg was divided into two equal halves in the plane shown in Fig. 2, the non- nucleated fragment did not form any micromeres. But when the egg was cut in a plane close to the nucleus, it was the non-nucleated frag- ment that cleaved as a whole egg. Shortly after fertilization Harnly found a different arrangement. Now only the half containing the egg nucleus will divide. Both this and the uncleaved half remained inside the intact fertilization membrane. Fragments that must have been meridional and vegetative cleaved as whole eggs, and other fragments segmented as animal halves. These results from fertilized eggs agree with those of Boveri and Ho6rstadius, although Harnly’s interpretation is somewhat different (see p. 309),—Harnly studied only the cleavage pattern; he did not follow the further development. Taylor and Tennent (1924), Taylor, Tennent and Whitaker (1926), and Tennent, Taylor and Whitaker (1929) obtained essentially differ- ent results for Lytechinus (Toxopneustes) variegatus. The polar bodies and the funnel (micropyle, made visible by octopus ink) served as landmarks for orientation. These authors state that a new polar axis is established in fragments. With a few exceptions, the first two planes of cleavage were at right-angles to the surface of section, regard- less of the orientation of the cut, and the micromeres were formed on the cut side at the end of the intersections of these two planes. Nor is there a complete segregation of the presumptive primary mesen- chyme or entoderm in the vegetative part of the mature, unfertilized egg. Their experiments indicated that, prior to fertilization, a mesenchyme-entoderm-forming substance has a uniform distribution through about 19/20 of the egg; the animal polar cap alone seemed to contain only ectoderm material. In the larger part of the egg, how- ever, the presumptive entoderm and mesenchyme was supposed to be restricted to the interior, surrounded by a superficial layer of ectoderm- forming material, as small superficial fragments never gastrulated. Thus both animal, vegetative, meridional and oblique halves of Lytechinus could form not only micromeres, but also mesenchyme, and could develop to plutei. In most cases a new polar axis was established, perpendicular to the cut side. In August, 1936, when spending the summer as a fellow of the Rockefeller Foundation at the Marine Biological Laboratory, Woods Hole, I had the opportunity of testing the localization of the micro- mere-, the entoderm-, and the skeleton-forming material in the egg of 298 SVEN HORSTADIUS Arbacia punctulata. Because of the short time available for this in- vestigation the material examined was scanty, but as the results of the observations are very definite, I think they are worth publishing. They indicate that localization in the Arbacia egg does not differ from that in Paracentrotus. The eggs were cut with a fine glass needle manipulated by hand. In order that the eggs should not slip away during the sectioning they were placed in a little scratch in a piece of celluloid (H6rstadius, 1928). The cut was orientated either in relation to the pronucleus (Fig. 2), or to the micropyle, which was made visible by ink from the squid (Loligo) (Fig. 1). Fic. 1. Cleavage and differentiation of animal and vegetative halves of the unfertilized egg of Arbacia punctulata. Orientation by means of the micropyle (a, 6). The egg nucleus may lie in the animal or the vegetative half (a, 0). c. Equal cleavage of the animal half. d. The vegetative half has formed macro- and micromeres, the latter at the original vegetative pole, antipolar to the cut side, where a cytoplasmic lobe has protruded through a slit in the fertilization membrane. e. Differentiation of the animal halves into blastulz with enlarged apical tuft. f, g. The vegetative halves gastrulate and form skeleton, developing to ovoid larve (g) or plutei (f). The results indicate a segregation of the micromere-, the entoderm-, and the skeleton-forming material in the vegetative part of the egg. II. IsoLateED ANIMAL AND VEGETATIVE HALVES, AND MERIDIONAL HALVES Twelves eggs were divided equatorially. At the operation the pronucleus in some cases lay close to the plane of section, in other cases far away from it. In 7 pairs the egg nucleus lay in the animal half, in 5 in the vegetative (Figs. 1a, 6). The animal and the vegeta- LOCALIZATION IN UNFERTILIZED ARBACIA EGGS DY tive halves of each egg were reared separately and kept in pairs. Of the twelve pairs only one vegetative fragment died early. A. Differentiation Zoja (1895), Boveri (19016), Terni (1914), and others have stated that animal fragments do not gastrulate and possess a very enlarged apical tuft. This phenomenon was studied in greater detail by Horstadius (1935, pp. 283-314). None of the twelve animal halves gastrulated. They all developed considerably enlarged apical tufts, thus differentiating completely in conformity with animal fragments in Paracentrotus (Fig. 1e). On the other hand, all the vegetative fragments gastrulated and formed skeletons. Some of them developed into good plutei with mouth and arms (Fig. 1f), the others showed a more vegetative type with ovoid body shape, no mouth, and a poor skeleton (Horstadius, 1928, 1935) (Fig. 1g). As controls for the animal and vegetative halves four pairs of meridional halves were also isolated. In two pairs only one partner developed. The other six larve (two pairs and two single larvz) all gastrulated and formed skeletons, developing into more or less typical plutei, or only ovoid larve. B. Cleavage After fertilization of egg fragments a fertilization membrane forms in the usual way. In Arbacia the membrane is not raised from, but lies close to the surface of the egg. When, on cleavage, the volume of the egg increases because of the division into blastomeres, the mem- brane often bursts on the cut side, and a lobe of cytoplasm protrudes through the opening (Harnly, 1926; Tennent, Taylor and Whitaker, 1929, p. 18; Hérstadius, 19360, p. 820). Sometimes this cytoplasmic bud is nearly cut off by the edges of the membrane, but very often it is connected with the egg by a rather broad base. Nuclei may migrate into the lobe, which may thus be divided into cells. In both cases the bud indicates the cut side. Of the 12 vegetative halves 11 formed micromeres. Morgan (1894) found that the red pigment in the Arbacia egg in the 4-cell, seldom in the 2-cell stage, migrates away from the vegetative pole. As a consequence the micromeres are lighter and whiter than the other blastomeres. The micromeres in our vegetative halves were of this whitish type. But there were not always 4 micromeres (only in seven cases). In one case there was only 1, in two cases 2, and in one case 3 micromeres. One egg showed an irregular cleavage,—it was the vege- 300 SVEN HORSTADIUS tative half that died. It is a well-known fact that the formation of micromeres may be suppressed, even in an entire egg, without any influence on the later development. Thus the primary mesenchyme is not dependent on the presence of micromeres during the cleavage. The suppression of micromeres is often caused by a slight mechanical disturbance (Boveri, 19016; Hérstadius, 1928, p. 124). The absence of one, two, or three micromeres is probably due to such secondary influences. . The essential point is the relation of the cytoplasmic bud to the micromeres, i.e. the position of the micromeres in relation to the egg- axis. In three eggs no lobe was formed, in all the other eight cases the cytoplasmic bud was antipolar to the micromeres (Fig. 1d). This shows, without any exception in the cases where the landmark was visible, that the micromeres were formed at the original vegetative pole. ' The cleavage pattern of the meridional halves was not so regular. In one pair both halves had cells of somewhat varying sizes, but no real micromeres were observed. In another pair one fragment had all cells of equal size, those of the other being slightly irregular. In the remaining two pairs, one half had no small cells, the other two either light micromeres close to the bud or only one small cell, the character of which is uncertain. (In the absence of the cytoplasmic bud its position could not be determined.) Eleven of the 12 animal halves showed cells of equal, or slightly varying sizes, but no micromeres (Fig. 1c). The twelfth fragment possessed two pairs of small cells, but they cannot be regarded as real micromeres, as they were not lying close together. C. CONCLUSIONS In dealing with the often irregular cleavage patterns of fragments, we have to define what is meant by a micromere. Not every small cell is a micromere. Not even every small, whitish cell. As we shall find below, small whitish cells can be formed, owing to particular factors, so that we cannot speak of real micromeres. We only define as micromeres those small, whitish cells which have been formed by a process of fundamentally the same character as in the normal egg. The isolation of animal and vegetative halves of the unfertilized egg of Arbacia gave exactly the same results as with Paracentrotus. Animal halves show partial cleavage (no macro- and micromeres are formed, Fig. 1c), and differentiate as isolated animal halves typically do, with enlarged apical tuft, and without gastrulation (Fig. 1e). The vegetative halves segment as whole eggs, and without any rotation of the egg-axis (Fig. 1d). The micromeres are formed at the original LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 301 vegetative pole, opposite the cut (equatorial) side, thus not on the cut side as in Lytechinus. The vegetative halves of Arbacia also differ- entiate as isolated vegetative halves of other species do, giving ovoid larve or plutei (Fig. 1f, g). This characteristic cleavage and differ- entiation of animal and vegetative halves takes place irrespective of whether the larve are haploid or diploid, and irrespective of the posi- tion of the pronucleus of the egg in relation to the plane of section. The results thus indicate a segregation of the micromere-, the ento- derm- and the skeleton-forming material in the vegetative part of the egg. The meridional halves also support this view. All surviving frag- ments gastrulated and formed spicules. It is important to note that micromeres were not formed very regularly. In the only positive case did they lie close to the cytoplasmic bud, thus close to the cut side, as one would expect. This absence of micromeres is probably due to the fact that the mechanism of micromere formation—especially in the unfertilized egg—is very sensitive to mechanical injury (Ho6rstadius, 1928, pp. 19, 124), and that the cut passes through the supposed micromere-forming region. In Paracentrotus I obtained micromeres also in meridional halves in most cases, but there the membrane is less tight, or perhaps the vegetative cytoplasm less sensitive. It is to be regretted, that more cases were not available for study. III. FRAGMENTS OF EQUAL SIZE ISOLATED BY A CUT AS FAR AWAY FROM THE NUCLEUS AS POSSIBLE To test Harnly’s results, unfertilized eggs were divided into two approximately equal halves by a cut as far from the nucleus as possible. Eggs were chosen in which the nucleus lay fairly close to the periphery. The plane of section was perpendicular to a line through the nucleus and the center of the egg and passing through the latter (Fig. 2a). All the micromere-forming material which, according to Harnly, lies between the nucleus and the center of the egg, should be found in the fragment containing the egg nucleus. The plane of section thus is at random in relation to the egg-axis. Fifty-one pairs were reared. A. Differentiation Not all the fragments started to develop after fertilization; some remained undivided. A few happened to get lost. But in 35 pairs both partners reached full differentiation (Fig. 2). In 13 only one of the two fragments developed. In 18 pairs both fragments gastrulated and formed skeletons, developing into plutei or ovoid larve. In 5 the diploid partner differentiated as an animal half (enlarged apical tuft, no invagination), the haploid as a vegetative (archenteron, 302 SVEN HORSTADIUS skeleton, ovoid larva or pluteus), and in 6 pairs I found the reverse, the haploid fragment behaving as an animal half. In addition there were 3 cases in which the diploid had only a small archenteron, with or with- out skeleton, but the haploid a typical or large archenteron, and skele- ton, whereas in another 3 pairs the reverse was the case (the haploid larve with a small invagination—Fig. 2). The 13 single larve devel- oped as follows: 7 diploid, 2 haploid larve with more or less typical archentera and skeletons; 1 diploid gastrula with too small invagina- tion; 1 diploid and 2 haploid blastule with enlarged apical tuft. eG Og ® OLYO@ Fic. 2. Diagram of the differentiation of halves of the unfertilized egg of Arbacia isolated by a cut perpendicular to the line nucleus-center of the egg. The cut thus at random in relation to the egg-axis. In 18 pairs both partners gastrulated and formed skeleton, developing like approximately meridional halves (6). In 11 ‘pairs the one larva differentiated as an animal half, which in 5 of those cases was diploid (c), in 6 haploid (d). In 3 pairs the diploid, in 3 the haploid fragment devel- oped into a gastrula with too small archenteron (e, f). The results demonstrate a localization of entoderm- and skeleton-forming material in a part of the unfertilized egg that must be smaller than half the egg, and independent of the position of the egg nucleus. B. Cleavage We have already seen that the cleavages of the meridional halves were somewhat atypical, as the micromere formation was often in- hibited. Moreover, the blastomeres may often vary more or less in size. In the halves isolated by a cut at random in relation to the egg- axis, we find all sorts of cleavage patterns. It has been emphasized above that probably not all small cells observed are micromeres. On the contrary, small cells evidently may be formed as a result of factors fundamentally different from those leading to the formation of the typical micromeres. One has to be very careful regarding the interpretation of the nature of the small cells. Sometimes the cytoplasmic bud, protruding through the open- ing of the fertilization membrane—although not containing any nuclei LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 303 or real cell membranes—may at the first glance look like a micromere, or a group of micromeres. In some cases the cell membranes at one or two divisions may also pass through the cytoplasmic lobe, but the small ‘‘cells’’ formed in that way will still lack nuclei. Very often, however, a nucleus migrates out into the bud and subsequently divides there. After two divisions, the lobe is divided into four small (if the lobe was small) cells, which are similar to micromeres, in that they are lighter than the other cells (the cytoplasmic bud is often not so markedly pigmented as the other part of the egg). When the lobe does not protrude much outside the membrane, these cells have a striking similarity to micromeres. But we can also find small cells, not micromeres, inside the membrane, but formed in connection with the bud. If the cytoplasmic bud is pressed out in the 2- or 4-cell stage, one of the quarter-blastomeres may be much smaller than the other ones. After two further divisions its descendants will constitute four small cells inside the fertilization membrane (or two or three, if one nucleus has migrated into the lobe, Fig. 3a). The bud may remain a b c Fic. 3. a. Formation of small cells at the cut surface, because one of the earlier blastomeres has lost the greater part of its cytoplasm by a cytoplasmic protrusion. At the following divisions the descendants of the part left inside the membrane naturally were much smaller than the other blastomeres. 0, c. Formation of small cells under the edge of the membrane, single (c) or in groups (8) at a distance from each other. or fall off. This group of apparent micromeres on the cut side are thus formed by a process essentially different from that of the normal formation of micromeres. Occasionally the first cleavages in fragments may produce blastomeres of more or less unequal size with or without the formation of a bud. In that way, after several divisions, the smallest blastomere may give rise to a quartet of small cells, without any particular relation to the egg-axis, or to the cut side. The small cells of probably non-micromere origin may be of any number from one to five or six, or even more. It seems to me, moreover, that the pres- sure of the edges of the membrane on the base of large buds produces small cells, not only as descendants from a small half- or quarter- blastomere, but as a result of the pressure at the cleavage leading to 304 SVEN HORSTADIUS the 8- or 16-cell stage. This appears more probable in that several times small cells were found just under the edge of the membrane, and not always as a single group, but separated from each other (Hicense DMG): Thus there are several possibilities of obtaining small cells, often of a lighter color than the other blastomeres, and these small cells are not real micromeres. One would expect from this that small cells, micromeres and atypical small cells, would occur much more fre- quently in fragments than normally. But the contrary is the case. As mentioned above (pp. 300 and 301), micromere formation is very sensitive to mechanical injury; e.g. after shaking, an egg may show an equal cleavage, but the differentiation will not be influenced. In Arbacia fragments the micromere-formation seems to be inhibited very often. As a result of these experiences, we have to deal with the following possibilities when dividing the unfertilized egg at random, if we assume the same arrangement of micromere material as in Paracentrotus. Vegetative and meridional halves ought to give whole cleavage, animal halves equal. But any fragment may cleave equally, i.e. if the micromere formation has become inhibited. Animal halves, and the other types of halves with inhibited micromere formation may show a more or less irregular pattern, the cells varying slightly in size. Any kind of fragment may form small cells which are not micromeres, either from the bud itself, close to the bud, or at any other place. It is naturally not easy to draw any conclusions from experiments where there are so many possibilities! According to Taylor, Tennent and Whitaker in Lytechinus all micromeres should appear on the cut side. According to Harnly, only the diploid halves form micromeres. One diploid and 3 haploid fragments showed whole cleavage, with a bud opposite the micromeres. They thus behaved like the vegeta- tive halves (p. 300 and Fig. 1d). Two diploid and 2 haploid fragments cleaved as whole eggs, having a lobe more or less on the side: they cor- respond to more or less meridional halves. Six diploid and 4 haploid fragments formed typical macro- and micromeres, as far as I could judge, but without any bud that made possible an orientation. Nine diploid and 5 haploid fragments cleaved equally, with or without bud. In 6 diploid and 4 haploid fragments the size of the blastomeres varied more or less, but no cells were as small as micromeres. In 7 diploid and 10 haploid fragments I found one or several small cells, the nature of which is open to question. They were in most cases formed either from or near the lobe, but sometimes not close to the lobe, or at a distance from each other. These observations are at variance with those of Taylor, Tennent LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 305 and Whitaker on Lytechinus, as micromeres were sometimes formed at a distance from or even opposite the cut side, and they are not in agreement with the results of Harnly on Arbacia, as of the cases re- garded as typical cleavage 9 were diploid while the same number, 9, were haploid. Let us now examine the relations between cleavage patterns and differentiation. We speak below of three kinds of larve (Fig. 2). One corresponds to the animal halves of Paracentrotus and of Arbacia as we have seen in the preceding chapter, a blastula with enlarged apical tuft. One constitutes a gastrula with a large archenteron and with skeleton, derived from vegetative, meridional, and oblique vegetative-meridional halves. It is not essential to mention whether the larva later becomes a typical pluteus or only an ovoid larva. The important point is whether it gastrulates or not. The third type is a gastrula with a too small archenteron, such as is derived from oblique animal-meridional halves (Fig. 2, e, f). Of the 18 fragments above characterized as showing whole cleavage, 17 developed to gastrule with large archentera and skeletons; one gave a blastula with enlarged apical tuft. The 24 fragments with equal (14) or slightly unequal (10) cleavages differentiated into 6 blastule, 2 gastrule with small, and 16 with large archentera. This result is what would be expected, as I consider that these two cleavage types correspond to animal halves and other types of half with sup- pressed micromeres. The 17 dubious cases must be described more in detail. In one pair one fragment showed two, the other one small cell inside the membrane, near the bud, but not quite in contact with it. Both developed into gastrule. I am inclined to interpret those as real micromeres, the halves as meridional fragments. In another case, some small light cells were formed at some distance from the bud, about 45°. The fragment differentiated into a gastrula. The nature of the cells is unknown. In one case the lobe itself divided into two small cells and one large cell. The former are certainly not micro- meres, as the latter lay between them. The development led to a blastula with enlarged apical tuft. Three fragments had three or four small cells lying inside the membrane, just under the bud (Fig. 3a). I assume that these cells were so small because the expulsion of the bud deprived one of the quarter-blastomeres of a large part of its cytoplasm. Another reason for believing that these small cells are not micromeres is that the other blastomeres cannot be classified as meso- and macromeres. All three fragments gastrulated, but one of them had a slightly enlarged apical tuft and only a small archenteron. In four cases a large part of the fragment had protruded through the 306 SVEN HORSTADIUS opening in the fertilization membrane and divided, just as well as the part within the membrane. In all four fragments small cells were formed, and in all they were situated under the edge of the membrane (Fig. 36, c). I do not believe that these small cells are comparable to micromeres, as in three of the four fragments they were found single or in groups at a distance from each other (Fig. 30, c). One developed into a gastrula, two into blastule of the animal type, and one died. The remaining six fragments had one or several small cells, but the sizes of the other cells were so varied, and the position of the ‘“‘micro- meres’? sometimes so scattered, that we cannot speak of a regular cleavage pattern with micromeres. In one case a close examination showed that the two apparent micromeres were not cells at all, only two small buds without nuclei. Three of these six fragments gastru- lated, two developed into blastule with enlarged apical tufts, and one died. C. Conclusions In this experiment we divided the egg with a random section in relation to the egg-axis, the plane of section being oriented in relation to the egg-nucleus (Fig. 2). The entoderm- and the skeleton-forming material cannot be limited to the region between the egg nucleus and the center of the egg as both the diploid and the haploid larve form archentera and skeletons with about equal frequency (in 27 and 26 cases respectively, from the pairs) and as the blastule with enlarged apical tufts were derived not only from haploid (6 cases) but also from diploid fragments (5 cases). Nor can the micromere-forming material be localized between the egg-nucleus and the center of the egg, as Harnly stated. Of the 18 cleavage stages that were considered to possess typical micromeres, 9 were diploid and just as many—9— haploid. Fifteen diploid and 9 haploid fragments formed no micro- meres, showing equal or slightly irregular cleavage. In 7 diploid and 10 haploid fragments, small cells were found, the nature of which has been discussed in detail. As 5 diploid and 6 haploid fragments of the pairs and 1 diploid and 1 haploid of the single halves failed to gastrulate, the entoderm- and skeleton-forming material cannot have the same widespread distribu- tion in Arbacia as in Lytechinus, in which, according to Tennent, Taylor and Whitaker, it occupies 19/20 of the volume of the egg. In that case every healthy fragment of the size of a half-egg would have gastrulated. Furthermore, the micromeres were not always formed on the cut side. In 4 cases they were found antipolar to the cyto- plasmic bud, in 4 this bud was situated laterally in relation to the LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 307 cleavage axis. In 24 fragments no micromeres were formed. In some of the 17 dubious cases we could show that the small cells on the cut side were not real micromeres. On the contrary, our results with one exception are in conformity with the view that both the micromere-forming and the entoderm- and skeleton-forming material is localized in a part of the Arbacia egg that is smaller than half the volume and comparable to the vegetative quarter of the Paracentrotus egg. Both diploid and haploid halves could gastrulate or differentiate into blastule of the animal type. Both diploid and haploid halves could form micromeres. These could appear on the cut side, opposite the cut side, or on any other side. With one exception, all fragments that formed “real’’ micromeres gastrulated. Fragments with equal or slightly irregular cleavage gave all kinds of larve, in accordance with the view that they com- prise the animal halves and any other kind of halves with suppressed micromeres. Of the dubious cases, some with small cells developed into blastule with enlarged apical tuft, i.e. as animal halves; but it was shown above that in all probability those small cells were not real micromeres, as either the whole cleavage stage was very irregular, or the small cells were located at some distance from each other. The only case that does not fit in with the assumption of a localization of both the micromere-forming and the entoderm- and skeleton-forming material in the vegetative part of the egg was one fragment which was supposed to have a typical 16-cell stage but differentiated into a blas- tula with large apical tuft (p. 305). Was that a case of micromere formation of an animal half, or was the interpretation of the cleavage stage wrong? IV. Discussion Although the micromere- and the skeleton-forming material occupy the same region in the Paracentrotus egg,—roughly speaking, the most vegetative quarter—the factors causing the formation of the micro- meres and those causing the formation of the primary mesenchyme of the skeleton-forming cells are not identical. As many authors have pointed out, normal primary mesenchyme is formed even if the micromeres are suppressed. i Harnly (1926) studied only the localization of the micromere material, but we have also paid attention to the differentiation of the fragments which were isolated in order to test his results. The out- come of his experiments, that the micromere-forming material in Arbacia prior to fertilization lies between the nucleus and the center of the egg, is contradicted by the two series of experiments described above. Animal halves did not form any micromeres, whereas the 308 SVEN HORSTADIUS vegetative fragments did, and in both cases irrespective of the position of the egg nucleus: the animal and the vegetative fragments could be diploid or haploid; the egg nucleus could at the operation lie close to the plane of section or at some distance from it. When the egg was divided into two equal halves by a section as far from the nucleus as possible, Harnly (see his Table 3 and Fig. 1a) obtained normal cleavage only in the nucleated fragments (23), whereas 28 haploid fragments showed ‘‘two tiers of eight equal cells.” The pairs described above formed (in the fairly clear cases) micromeres in 9 diploid and 9 haploid, and no micromeres in 15 diploid and 9 haploid fragments. In the light of my experiences, I have difficulty in understanding Harnly’s results. He found in his experiment (Table 1) that “in no case did a nucleated half that divided normally through the first three cleavages give other than a normal fourth cleavage.” Of 132 diploid fragments 30 showed normal fourth cleavage, 38 irregular and partial first cleavage, 50 endoplasmic buds (not studied), and 9 were undeter- mined. In Tables 3-5 no buds and no irregular and partial first cleavage are recorded. When the eggs were fragmented, as in our Fig. 2a, practically all diploid fragments segmented normally, but the haploid formed ‘“‘two tiers of eight equal cells.’ When the eggs were divided into two equal halves by a section close to the nucleus, all the fragments that were determined, diploid or haploid, segmented in a normal way, except a few haploid ones, which had only two or three macro- and micromeres. As we have seen, I have not found that fundamental difference between the diploid and haploid fragments cut as in Fig. 2a. Both kinds could form micromeres, and both kinds could show equal or slightly irregular cleavage. When equal, the blastomeres were hardly ever arranged as regularly as in two tiers of eight cells, as Harnly states for his 28 enucleated halves. Harnly also isolated non-nucleated fragments containing two-thirds or more of the material of the egg. One fragment was undetermined, 18 segmented normally. Harnly’s explanation is that these large haploid fragments contained all the micromere-forming material lying between the nucleus and the center of the egg. A more or less typical cleavage would in this case be expected also when assuming a localization of the micro- mere material in the vegetative quarter of the egg, since only very few of the fragments would be completely devoid of such material. Harnly (1926) also divided fertilized eggs into two equal halves. According to Harnly, the fusion nucleus is to be found in the egg-axis, just above the equator; Harnly tried to cut equatorially. The fertil- ization membrane could not be divided, the two halves remained inside the membrane, flattened against each other. Only the half LOCALIZATION IN UNFERTILIZED ARBACIA EGGS 309 containing the egg nucleus segmented. Three types of cleavage were found: whole cleavage with the micromeres opposite the cut side, whole cleavage with the micromeres close to the cut side, and equal cleavage. Harnly concludes from his experiments that, within five minutes after fertilization, the micromere-forming material changes its position. Both the nucleus and the micromere material are now oriented with regard to the primary axis. Harnly does not expressly state where along the axis the micromere material is now located, but he seems to indicate a central position. As far as I can see, his results accord just as well with the view that the micromere material is restricted to the vegetative part of the egg. The equal cleavage (the largest number, as the nucleus in most cases was lying above the equator) belongs to the animal halves; the whole cleavage with the micromeres opposite the cut side corresponds to the cases in which the nucleus came to lie in the vegetative half; and the two fragments with the micromeres close to the plane of division are probably, as Harnly himself presumes, meridional halves. Tennent, Taylor and Whitaker (1929, p. 66) state that their results “have compelled the conclusion that in the unfertilized egg of Lytechi- nus there is an animal polar cap and a superficial layer of ectoderm- forming material surrounding a core of undifferentiated material that is potential endoderm, primary and secondary mesenchyme, and mesoderm.’ The presence of an animal cap of ectoderm-forming material, occupying 1/20 of the volume of the egg (loc. cit., p. 66) was concluded from the fact that gastrule were obtained from vegetative fragments of 1/11 the volume of the egg (but there is only one of that size, No. 1926 253), while none of the larve from fragments of equal size from the region of the animal pole developed beyond the stage of blastule with mesenchyme (No. 1926 276, 279; see also pp. 47, 63, 64, 66). The smallest fragment that developed into a blastula was 1/21 of the volume of the egg. These results point, as it seems to me, rather to the size 1/11 than 1/20 for the animal polar cap. The reason for the conclusion that there is a superficial layer of presumptive ectoderm also in the other parts of the egg is not given in the paper in question, asfarasI can find. But in the preliminary report by Taylor, Tennent and Whitaker (1926), the same conclusion has been drawn from the fact that small superficial fragments—with a diameter of one-fifth of that of the egg—only gave blastule with or without mesenchyme, while larger, but still small fragments gastrulated. A fragment with a diameter of one-fifth of that of the egg has a volume that is only 1/125 of that of the egg. Nobody has so far been able to rear gastrule from such small fragments. The smallest blastomere that has been 310 SVEN HORSTADIUS observed to gastrulate is a half-macromere (Horstadius, 1936a). Its volume is larger than 1/32 but smaller than 1/16 of the egg. Morgan (1895) and Driesch (1900, 1902) estimated the smallest egg fragment that could gastrulate at 1/40-1/60 and 1/32 respectively of the volume of the egg. But these results are very uncertain, as the size of the fragment was concluded from the size of the larve. In the full report, Tennent, Taylor and Whitaker (1929, p. 62) state, contrary to the preliminary note, that the smallest fragment that gave a gastrula had 1/11 the volume of the egg, the smallest that gave a blastula 1/21 (cf. above). Under such circumstances I cannot see how the idea of asuper- ficial layer of ectoderm-forming material round the whole egg can be supported. Very small fragments probably do not gastrulate because they are too small, not because of lack of presumptive archenteron (if derived from a vegetative part of the egg). As the smallest blas- tule obtained were derived from fragments measuring 1/21 of the volume of the egg, no blastule seem to exist that could prove the presence of a superficial layer of ectoderm-forming material outside the entoderm-mesenchyme-forming substance that, in Lytechinus, is supposed to occupy 19/20 (10/11?) of the egg. We now turn to a comparison of the facts leading to the idea of an animal cap of ectoderm material, occupying only 1/20 (1/11?) of the egg in Lytechinus, with the facts demonstrating that in Paracentrotus and Arbacia the presumptive ectoderm is distributed over more than half the egg. Animal halves of the unfertilized egg of Paracentrotus, oriented at the operation by means of the pigment ring, never gastru- lated—cf. p. 296. None of our 12 animal halves of Arbacia (orientation by means of the micropyle) gastrulated, and all showed an enlarged apical tuft. Of the halves cut at random, many differentiated in the same way (Fig. 2). Tennent, Taylor and Whitaker (1929, p. 57) report the history of 27 pairs of fragments isolated by horizontal sections. But of these pairs, both fragments reached gastrulation age only in 9 cases. If the presumptive entoderm- and mesenchyme-forming ma- terial occupies 19/20 (10/11?) of the egg, every horizontal fragment larger than 1/20 (1/11?) should gastrulate. This was the case in 6 pairs, the relative size of the fragments varying from 1:1 to 1 : 3.4 (No. 1924 8, 42, 44, 54, 61, 1926 230). In one case both gastrulated with the size difference 10:1 (No. 1926 253), the animal fragment being the larger. In two pairs, however, the animal partner did not gastrulate. In the one pair its volume was only 1/17 of that of the egg (No. 1926 229), but in the other the fragments were of approximately the same size (1 : 13, No. 1926 231). The animal fragment formed a blas- tula without mesenchyme. In addition, 3 of the single larve, all of half LOCALIZATION IN UNFERTILIZED ARBACIA EGGS Sill size (No. 1924, 5, 45, 79) remained as blastule, while 11 gastrulated. These results, with 4 fragments of approximately half size differentiat- ing only to blastule, are at variance with the assumption that the pre- sumptive entoderm and mesenchyme occupies 19/20 or even 10/11 of the egg. Of 21 vertical sections in Lytechinus both partners developed to gastrulation age only in 5 pairs. Here we should expect gastrulation in every fragment of sufficient age (10 hours), and it occurred in 4 pairs (No. 1924 40, 53, 56, 80). But in the fifth pair (No. 1924 60) one frag- ment only formed a “blastula with mesenchyme,” while the other developed to a pluteus. These two fragments were of equal size. Furthermore, not less than 3 (No. 1924 23, 25, 1926 267) of the other 10 vertical halves (only one partner living) that lived longer than 10 hours failed to gastrulate. Seven pairs of fragments obtained by diagonal sections (Tennent, Taylor and Whitaker, 1929, p. 58) were reared (No. 1926 275, 279- 283, 285). In four cases the small, oblique animal fragment, of size 1/6, 1/11, 1/11, 1/21, formed a blastula with mesenchyme, while the larger vegetative fragment gastrulated. In two cases the animal frag- ment was the larger (5:1, 4:1); both partners gastrulated. In the seventh pair also both developed to gastrule, the size of the oblique animal fragment being 1/7 of that of the vegetative. The blastule obtained from the animal fragments are said to have a thickened posterior wall and often a number of mesenchyme cells. No mention is made of the size of the apical tuft in these blastule. Nevertheless, I get the impression from the drawings of such blastule (Figs. 44, 47) that they are typical animal fragments. Thus the thickened wall would correspond, not to the posterior, but to the animal and ventral side (cf. Lindahl, 1933; Hérstadius, 1935, p. 286, 1936a, p. 56). This interpretation is supported by the fact that the ““mesenchyme’”’ cells in the figures are located further towards the thin wall than towards the thickened wall. Furthermore, I very much doubt whether all the cells reported as mesenchyme cells have that character. In fragments, very often some cells pathologically migrate into the blastoccel. Many isolated animal halves of Para- centrotus may have a few, or a large number of cells free in the interior of the blastula. They have nothing to do with real mesenchyme cells. They may, of course, also appear in meridional and vegetative frag- ments, together with primary mesenchyme cells. It requires an inti- mate knowledge of the material to recognize the real mesenchyme cells from these pathological cells, and also to interpret correctly the animal blastulz. The cells in the blastule (Figs. 44 and 47) are rather 312 SVEN HORSTADIUS varied in size. This indicates, it seems to me, that here we have not to deal with real primary mesenchyme. Furthermore, the assumption that the mesenchyme material occupies 19/20 of the egg is contra- dicted by the fact that Tennent, Taylor and Whitaker report several blastule of half size without mesenchyme (No. 1924 5, 23, 25, 1926 231). The fact that 4 of the horizontal and 4 of the vertical fragments of half-size did not gastrulate is incompatible with the conception of the presumptive entoderm occupying 19/20 (10/11?) of the egg. It rather shows that it is restricted to less than the half of the egg. Also the small diagonal fragments illustrate the same thing, as in some cases they contain material from the equator region without gastrulating. The fact that both partners did gastrulate in 7 of the 9 animal- vegetative pairs, and that 4 of 20 vertical fragments did not gastrulate, seems to indicate that the orientation was not always that desired. When discussing the localization of the entoderm and mesenchyme material in Arbacia with some colleagues at Woods Hole in 1936, I said that one ought to obtain blastule of animal type from, roughly speaking, 15 per cent of the halves isolated at random, if the same conditions as in Paracentrotus prevailed. My material of random sections on Arbacia numbers 14 blastule with enlarged apical tuft and 65 larve that gastrulated. Thus 18 per cent did not gastrulate. This indicates that the presumptive entoderm occupies less than half of the egg, otherwise the animal blastulz would have been much fewer. Hight of the Lytechinus fragments of approximately half-size remained in the blastula stage, while 40 developed into gastrule or plutei. Thus 17 per cent did not gastrulate. These figures of all the horizontal and vertical sections together raise some doubts as to the accuracy of the orientation of the cut, particularly as 4 of 20 meridional fragments only formed blastule. Has the plane of section often been different from that desired? One possibility is that the egg slipped or turned during the operation. The three authors state (1929, p. 6) that the operator could control this. The micropyle and the polar bodies served as landmarks. Are they reliable? Tennent, Taylor and Whitaker (1929, p. 12) regarded it as possible that the egg might rotate in its jelly and in that manner make the micropyle of uncertain value as a landmark. ‘‘No evidence of rotation was found. Careful examination showed the polar bodies lying at the base of the micropyle. In one instance only were they out of line as much as five degrees.” But in the preliminary report (1926) it is stated that ‘‘our preliminary observations having convinced us that the egg might be rotated in its jelly, we felt that our orientation of the egg was dependable only when the polar bodies could be found at the base of the micropyle.” But LOCALIZATION IN UNFERTILIZED ARBACIA EGGS ol is it not possible that the egg can rotate inside the jelly, the polar bodies being held by the micropyle? (Hérstadius, 1928, p. 91.) If so, the egg-axis may lie at any angle to the axis indicated by the micropyle and the polar bodies. Or has the whole egg rotated during the slow sectioning with the micromanipulator? From the fact that, in certain fragments obtained by vertical sec- tion, the first cleavage amphiaster lies temporarily in the short axis of the cell, and that in those cases the size of the cleavage cells is atypical, Tennent, Taylor and Whitaker (1929, pp. 41-48) conclude that there is a prelocalization of an entoderm-forming substance. I can but support the reservation of Dr. Taylor (loc. cit., p. 47), who doubts the significance of cell-size as indicative of a segregation of entoderm-forming material in the unfertilized egg. In Paracentrotus, under certain circumstances, the four animal cells of the 8-cell stage may be much larger than the four vegetative (vorzeitige Mikromeren), or the spindles of the second and third division may stand obliquely, or the spindle of the first division in an animal-vegetative direction instead of equatorial, or a vegetative half may show a typical whole cleavage, all without any change of the position or amount of not only the micromere-, but also the entoderm-forming material (see Hoér- stadius, 1928, pp. 10, 14). As far as I can see, the results in question only permit conclusions as to the cleavage factors, not as to the localization of presumptive entoderm. Both Harnly (1926) and Tennent, Taylor and Whitaker (1929) state that the first two furrows in fragments stand at right angles to the surface of section, irrespective of the orientation of the egg. The last-mentioned authors maintain that a new polar axis has been es- tablished (loc. cit., p. 66). But, as we have seen above, the position of the first two furrows may be entirely independent of the polar axis of the egg and larva. In order to determine whether a new polar axis has been established, one has to ascertain by local vital staining whether the cleavage axis coincides with the axis apical organ—blastopore in the gastrula. In Paracentrotus the micromeres are always formed at the vegetative pole, irrespective of the position of the first two furrows, and differentiation takes place in accordance with the original egg- axis (HGrstadius, 1928, p. 10; 1936a, p. 53). Also, if micromeres are formed on the cut surface at the intersection of the first two cleavage planes, as stated for Lytechinus, this does not necessarily mean a change of polarity, as the factors for micromere formation and skeleton and entoderm formation are not identical (cf. the paragraph above and my S07). The statement that the micromeres are always formed on the cut 314 SVEN HORSTADIUS side in Lytechinus is surprising. Tennent, Taylor and Whitaker have studied the cleavage of fragments in a great many cases, and the drawings seem convincing, although they look rather schematized. But the statements are perhaps not entirely conclusive, as the micro- meres in our vegetative Arbacia fragments were formed opposite the cut side and as the three authors (1929, pp. 36-39) also made some experiments on Arbacia at Woods Hole,'which do not agree with those presented here. Only 14 of 45 operations were either fully or partially successful. If we count all the fragments that developed, single and pairs, 1-4 micromeres were found in 6 vegetative and 6 animal frag- ments, while 2 vegetative and 1 animal fragment cleaved irregularly, and 5 animal and 2 vegetative divided equally. Nothing is said as to the position of the micromeres in relation to the cut surface. These results contrast markedly with mine: 11 of 12 vegetative fragments showed micromeres, all opposite the cut side, while the animal halves did not form any real micromeres. Tennent, Taylor and Whitaker do not discuss any of the possible sources of error regarding the micromere formation. Moreover, perhaps the free hand is a better instrument for cutting the eggs than the micromanipulator, which may work so slowly that a rotation of the egg after the final orientation is possible. A renewed investigation on Lytechinus, as well as on Ar- bacia, with attention payed particularly to the orientation of the cut and the possible formation of false micromeres, with a large number of larvee that reach full differentiation, and with local vital staining to study the relation of the axis of the egg to that of the fragment and the relation of the micromeres to the cut surface, would be very desirable. The critical discussion may be summarized as follows. Harnly’s assumption (1926) that the micromere-forming material in the un- fertilized egg of Arbacia is situated between the nucleus and the center of the egg does not hold, as vegetative halves formed micromeres, and animal halves showed no micromeres, in both cases irrespective of the position of the nucleus, and as nucleated and non-nucleated halves isolated by a cut perpendicular to the line nucleus-center of the egg (Fig. 2a) formed micromeres with equal frequency. The conclusions of Tennent, Taylor and Whitaker (1929) that the egg of Lytechinus has an animal polar cap occupying only 1/20 (1/11?) of the volume of the egg, and a superficial layer of ectoderm-forming material are contradicted by their own results. The conclusion is drawn from the differentiation of one animal and two vegetative fragments of the size 1/11. But there are diagonal fragments of the same size, or larger, reaching down to the equator, that do not gastrulate. Moreover, several animal and meridional fragments of half size and sufficient LOCALIZATION IN UNFERTILIZED ARBACIA EGGS oS age did not gastrulate. These facts indicate that the presumptive ectoderm occupies at least half the egg. The fact that several merid- ional halves also differentiated as an animal fragment arouses the suspicion that the plane of section has not always been that desired, which would explain many of the discrepancies between Lytechinus and Paracentrotus. It is questionable whether some of the cells de- scribed as mesenchyme cells are not pathological. The position of the first two cleavage planes does not necessarily indicate the position of the axis of the larva. In fragments of Lytechinus eggs the micro- meres are reported by Tennent, Taylor and Whitaker to be formed on the cut side. In Arbacia I found the micromeres formed at the vegeta- tive pole, irrespective of the position of the section. The fact that the results of Tennent, Taylor and Whitaker as to the localization of the micromere-forming material in Arbacia do not agree with mine, which in turn confirm those on Paracentrotus, makes a reinvestigation of Lytechinus also desirable. V. SUMMARY Animal halves of unfertilized eggs of Arbacia punctulata segmented equally and developed into larve of animal type: blastule with en- larged apical tuft. Vegetative halves formed micromeres antipolar to the cut side, and gastrulated, differentiating into ovoid larve or plutei. Meridional halves all gastrulated. The type of development was independent of the presence or absence of the egg nucleus (Fig. 1). Unfertilized eggs were divided into approximately equal halves by a section perpendicular to the line nucleus-center of the egg (Fig. 2a), the plane of section thus being laid as far away from the nucleus as possible, and at random in relation to the egg-axis. Nucleated and non-nucleated fragments formed micromeres and gastrulated with equal frequency. Some of the fragments differentiated as animal halves (Fig. 2). Atypical small cells, particularly on the cut side, which may be mistaken for micromeres are often formed. The results indicate that the micromere-forming and the entoderm- and the skeleton-forming material in Arbacia is located in the most vegetative part of the unfertilized egg, occupying less than half the volume of the egg. The results of Harnly (1926) on Arbacia and of Tennent, Taylor and Whitaker (1929) on Lytechinus are critically discussed. LITERATURE CITED Boveri, THEODOR, 190ia. Die Polaritat von Ovocyte, Ei und Larve des Strongy- locentrotus lividus. Zool. Jahrb., Abt. Anat. u. Ont., 14: 630. Boveri, THEODOR, 19016. Uber die Polaritat des Seeigel-Eies. Verhandl. d. phys.- med. Gesellschaft. Wurzburg, 34; 145. 316 SVEN HORSTADIUS Boveri, THEODOR, 1902. Uber mehrpolige Mitosen als Mittel zur Analyse des Zellkerns. Verhand. d. phys.-med. Gesellschaft Wiirzburg, 35: 67. DriescH, Hans, 1896. Betrachtungen iiber die Organisation des Eies und ihre Genese. Arch. Entw.-mech., 4: 75. DriescH, Hans, 1898. Uber rein miitterliche Charaktere an Bastardlarven von Echiniden. Arch. Entw.-mech., 7: 65. DriescH, Hans, 1899. Die Lokalisation morphogenetischer Vorgange. Arch Entw.-mech., 8: 35. DriescH, Hans, 1900. Die isolirten Blastomeren des Echinidenkeimes. Arch. Entw.-mech., 10: 361. DriescH, Hans, 1902. Neue Ergainzungen zur Entwickelungsphysiologie des Echinidenkeimes. Arch. Entw.-mech., 14: 500. Harnty, Morris H., 1926. Localization of the micromere material in the cytoplasm of the egg of Arbacia. Jour. Exper. Zodl., 45: 319. Hoab ey, Lreicu, 1934. The relation between the position of the female pronucleus and the polar bodies in the unfertilized egg of Arbacia punctulata. Bzol. Bull., 67: 220. HOrstTapius, SVEN, 1928. Uber die Determination des Keimes bei Echinodermen. Acta Zoologica, Stockholm, 9: 1. HOrstTapius, SVEN, 1935. Uber die Determination im Verlaufe der Eiachse bei Seeigeln. Pubbl. Staz. Zool., Naples, 14: 251. HORSTADIUS, SVEN, 1936a. Weitere Studien tiber die Determination im Verlaufe der Eiachse bei Seeigeln. W. Roux. Arch. Entw.-mech., 135: 1. HoOrstapius, SVEN, 1936b. Studien iiber heterosperme Seeigelmerogone nebst Bemerkungen iiber einige Keimblattchimaéren. Mém. Mus. d’ Hist. Nat. de Belgique, Ser. 2, Fasc. 3: 801. LinpaH1, P. E., 1933. Uber ‘‘animalisierte” und “vegetativisierte” Seeigellarven. W. Roux. Arch. Entw.-mech., 128: 661. MoreGan, THomas Hunt, 1894. Experimental studies on Echinoderm eggs. Anat. Anzeig., 9: 141. Morean, THomas Hunt, 1895. Studies of the “partial” larve of Sphzrechinus. Arch. Entw.-mech., 2: 81. Taytor, C. V., AND D. H. TENNENT, 1924. Preliminary report on the development of egg fragments. Carnegie Inst. Wash. Yearbook No. 23. Taytor, C. V., D. H. TENNENT AND D. M. WHITAKER, 1926. Investigation on organization of Echinoderm egg. Carnegie Inst. Yearbook 25. . TENNENT, D. H., C. V. TAyLor anp D. M. WuiTaKeEr, 1929. An investigation on organization in a sea-urchin egg. Carnegie Inst. Wash. Publ. No. 391. TERNI, T., 1914. Studio sulle larve atipiche (blastule permanenti) degli Echinoidi. Mitt. Zool. Stat. Neapel, 22: 59. von Usiscu, L., 1925. Entwicklungsphysiologische Studien an Seeigelkeimen. Zettschr. wiss. Zool., 124: 361. Zoya, R., 1895. Sullo sviluppo dei blastomeri isolati dalle uova di alcune meduse (e di altri organismi). Arch. Entw.-mech., 2: 1. ” EXPERIMENTS ON DETERMINATION IN THE EARLY DEVELOPMENT OF CEREBRATULUS LACTEUS SVEN HORSTADIUS (From the Zodtomical Institute, University of Stockholm, and the Marine Biological Laboratory, Woods Hole, Massachusetts) I. INTRODUCTION The eggs of nemerteans, annelids, gastropods, lamellibranchs and scaphopods show the spiral type of cleavage. Most of these eggs,— those of the annelids and mollusks,—are assigned to the mosaic type of determination. Their isolated blastomeres have been found to develop in accordance with their prospective significance. None of these eggs has, however, been investigated in such detail as the sea urchin egg, of which we know that some parts may develop into a complete larva (Driesch, 1891, etc.) and that other parts may differen- tiate less typically than in the case of normal development, owing to the fact that interactions have to take place between different parts of the egg, bringing the formation of some organs (HGrstadius, 1928, 1935, 1936). [von Ubisch (1936) has denied the réle of interactions in the sea urchin egg. However, new isolation and implantation experi- ments, as well as constriction experiments (unpublished), completely confirm my earlier results as to the interactions.| The scope of this investigation was to see whether parts of an egg of the mosaic type might not show traces of interaction when transplanted atypically upon each other. As I could not find any annelid or mollusk egg so suitable for isolations and transplantations as the Cerebratulus egg, I had re- course to the latter. This egg shows a spiral cleavage but is not wholly a mosaic egg. It belongs to the regulation type insofar as fragments of the uncleaved egg and isolated half- and perhaps quarter-blastomeres may give rise to harmonic dwarf larve. But, on the other hand, the investigations of E. B. Wilson (1903), Zeleny (1904) and Yatsu (1910) indicate that the animal and vegetative parts of cleavage stages differentiate as fragments. Thus in that respect we seem to have to deal with a mosaic. The cell-lineage of the nemertean egg has never been worked out. For that reason, in order to be able to interpret rightly the results of the isolation and transplantation experiments, we had first to in- vestigate the prospective significance of the cell layers. As the experi- ments of the authors mentioned above are rather incomplete, we also have had to repeat and extend the study of isolated fragments. The results of previous investigators will be mentioned below in connection with each corresponding experiment. sili 318 SVEN HORSTADIUS II. MATERIAL AND METHODS The investigation was undertaken in July and the first part of August, 1936, at Woods Hole. The worms (Cerebratulus lacteus Verrill) were sent in several deliveries in ice-cooled jars from Lynn, north of Boston, and kept in an ice-box. In order to obtain the eggs, a piece of a female was put into a piece of cheesecloth in sea water. The cheesecloth retains the slime but allows the eggs to fall through. Coe (1899) has shown that the germinal vesicle gradually fades away, when the eggs come out in the sea water and the first polar spindle is formed, but the spindle remains in the metaphase until fertilization occurs. In order that uniform conditions as regards maturation and fertilization might be secured, the sperms were added half an hour after the shedding of the eggs (Zeleny, 1904). The egg forms no fertilization membrane but is surrounded by a jelly and a soft mem- brane, which are easily removed by gentle shaking. The blastomeres of the 2-, 4-, and 8-cell stages are easily separated with glass needles. The separation of the four layers of the 16-cell stage is more difficult, as the blastomeres from two adjacent layers lie partly between each other (Fig. 1, B). But it can be done by fa-Id 1A -1D A B Fic. 1. Eight-cell stage (A) and 16-cell stage (B) of Cerebratulus. cutting before the fourth cell division is quite completed, because at that moment the cells have not yet sunk in between each other. For transplantation the fragments were placed one on top of the other in a cavity of appropriate size, made in a celluloid plate, and a small glass sphere placed on top of the upper fragment to produce pressure (H6rstadius, 1928). In some cases the fragments fused after being made to adhere by a brief, gentle pressure of the needle. Local staining (of a single blastomere or of a layer of blastomeres) was made by leaning the cell or cells in question against a piece of agar (Vogt, 1923, 1925; v. Ubisch, 1925) or by staining an isolated cell layer and then transplanting it (Hérstadius, 1928, 1935). The number of experiments is rather limited. This is due partly to the fact that the termination of the breeding season already in early August cut short my experiments before the completion of an adequate study, partly to the fact that in many series nearly all larve died. DETERMINATION IN CEREBRATULUS EMBRYO 319 The reason for this was not a pathological condition of the larve. They looked very healthy until they suddenly disappeared by getting caught and expanded, just as if exploded, by the surface tension. The death ratio was considerably reduced when the larve were reared in a large drop of water between a slide and a coverslip on plasticine feet and protected from evaporation by vaseline. III. NorMAL DEVELOPMENT. NOMENCLATURE The following facts regarding the normal development have to be borne in mind. The Cerebratulus egg is very dark brown, opaque. But the polarity of the just-laid egg may be accurately determined partly by a conical protuberance at the vegetative pole, partly by a clear area (the fading germinal vesicle) at the animal pole. Later the polar bodies at the animal pole give a good landmark. They are also present in the cleavage stages. The first two furrows are meridional, the blastomeres of the 4-cell stage all being of the same size. Thus we have no predominance of the D-quadrant in the nemertean egg. The third cleavage is dexio- tropic, forming four micromeres (1a—id) and four macromeres (1A— 1D), but the micromeres are larger than the macromeres, as Coe (1899) first described (Fig. 1, A). At the next (leiotropic) cleavage, the micromeres (the first quartet of micromeres, 1a—1d) are divided into two layers each of four cells (1a'-1d! and 1a?-1d?), and the macromeres bud off a second quartet of micromeres (2a—2d). The macromeres are now called 2A—2D (Fig. 1, B). The use of the words micro- and macromeres at the 8-cell stage is rather confusing, as the micro- are larger than the macromeres. In the following account we prefer to call them the animal and the vegeta- tive cells of the 8-cell stage. Instead of speaking of the two layers derived from the first quartet of micromeres (1a‘1d!, 1a?-1d?), of the second quartet of micromeres (2a—2d), and of the macromeres (2A—2D) of the 16-cell stage, we may simply designate the four layers as ani, an, veg, and vege (Fig. 1, B), as has been done with the four layers in the sea urchin egg (Horstadius, 1931, 1935). The composition of a larva may then be expressed by a formula, e.g., an; + any + veg; means a larva from which vege has been removed, etc. The gastrula is still rather opaque, but the fully differentiated larva, the pilidium, is quite transparent. The blastopore forms the mouth, leading into an cesophagus and stomach; there is no intestine or anus (Fig. 2, A). One or two days after gastrulation two lappets grow out on the left and right sides of the mouth. The whole surface of the pilidium is ciliated, but the lappets are bordered by a special 320 SVEN HORSTADIUS ciliated band where the cilia are longer, more concentrated, and beat in a characteristic way as compared with the cilia of the rest of the ectoderm. The cells of the ciliated band are more yellowish than the other ectoderm cells. At the animal pole we find an apical organ which is a sense organ, a long flagellum in a thickened pit of the ectoderm. The flagellum is not a single structure but is composed of a bundle of fine threads. The right and left sides of the pilidium evidently are those which bear the lappets where the lappets are formed. As there is no anus it is difficult to say what corresponds to the dorsal, what to the ventral side. In the gastrula we call that axis dorso-ventral which is per- pendicular both to the animal-vegetative and the right-left axes, but in the pilidium the dorso-ventral axis probably, as the ventral side presumably is so short, only forms an acute angle with the egg (animal- vegetative) axis. After metamorphosis the egg axis corresponds to the dorso-ventral axis of the worm. Fic. 2. A. Normal pilidium. 8B, C. Pilidia from the isolated two half- blastomeres of one egg. IV. THE PROSPECTIVE SIGNIFICANCE OF ani, ame, vegi, AND vege The four animal cells of the 8-cell stage were stained by leaning them against a piece of agar. The whole ectoderm down to the ciliated band was blue, and there was also an abundance of blue granules in the band. When the four vegetative cells of the 8-cell stage were stained, the stomach, the cesophagus, the inner sides of the lappets, and the ciliated band were blue. Thus also in this case, we found the ciliated band markedly stained. When the animal cells DETERMINATION IN CEREBRATULUS EMBRYO oa were stained, the upper, most animal part of the whole band (with one probable exception, see below) contained blue granules, and when the vegetative cells had been stained, the color was restricted more or less to the lower, vegetative part of the band. It is very difficult to trace the limit of the stained area with certainty in Cerebratulus, but I think it is evident that the ciliated band is composed of material both from the animal and the vegetative cells. It was observed in several cases that on one day the young pilidium showed the ectoderm, including the ciliated band, bluish (staining of the animal cells); the next day the stomach too had turned blue. It was noted, however, that the cesophagus and the inner sides of the lappets were still unaffected by the Nile blue. Thus it would appear that we are not dealing with a general diffusion of the stain. It seems probable that at this stage the stomach acts as an excretory organ, as is the case with the digestive tract of turbellarians (Westblad, 1923). Fic. 3. In the 16-cell stage the four most animal cells, an:, were isolated, vitally stained and transplanted back on the an2 cells. The animal fragment was probably rotated 180° at the transplantation: note the little patch of ciliated band too high up on the posterior side. Normally it probably constitutes the most anterior part of the band. Stained area stippled. Staining of the four vegs-cells of the 16-cell stage resulted in a blue stomach. Whether a small part of the vego-material is or can be used to form a part of the cesophagus I cannot tell,—the limit of the stained area was not sufficiently sharp. Also in staining the four an;-cells I had great difficulties in observing the colored line. It seems to me that these four cells form the greater part of the pretrochal ectoderm down to about the equator of the pilidium. Thus ang would form the ectoderm on the outside of the lappets and a part of the ciliated band. I have the impression that _the an;-material in the anterior part of the larva goes down to the ciliated band and contributes to the formation of its foremost part, whereas posteriorly it does not reach so far down. This view is sup- ported by the following result. For the purpose of obtaining a sharper limit of the stain, the four an; cells were cut off, stained, and replaced. In one case they probably happened to be rotated 180°, because the Se SVEN HORSTADIUS larva showed a short piece of ciliated band way up from the ciliated band at the posterior end of the pilidium, and the stain border line ran from this isolated piece of band to the band at the anterior end of the larva (Fig. 3). This would indicate that the part of the ciliated band that is normally derived from the an;-material at the front end of the larva has undergone self-differentiation too high up on the posterior side. The prospective significance of the four layers thus seems to be as follows: an; gives the ectoderm of the upper half of the pilidium, including a very small piece of the ciliated band. ane forms the rest of the outside ectoderm (the lappets) and a great part of the ciliated band. veg; also contributes to the ciliated band and forms, moreover, the insides of the lappets, and the cesophagus. vege constitutes the material for the stomach and perhaps a small part of the cesophagus as well. V. THe POosITION OF THE FIRST FURROW In order to interpret the results of isolating the first two blasto- meres, it is necessary to know the position of the first furrow, whether it has a fixed position in relation to the median plane of the larva or may form any angle to this plane. In 34 2-cell stages one blasto- mere was stained by leaning it against a piece of agar. Owing to the large size of the blastomeres, the stain seldom penetrated the whole blastomere. Thus we did not get a very exact staining, with a sharp border line along a meridian of the egg. The stain rather indicated that side of the blastomere which was opposite the furrow. In 9 of the larvee the color was too weak or too diffuse. In the other 25 the following parts of the larvee were found to be blue: dorsal 5, ventral 3, left 6, right 3, dorsal-left 3, dorsal-right 4, ventral-right 1. The number of cases is small and the method not very exact. Nevertheless, I think these results show that the first furrow is not confined to a certain plane, but probably may have any position in relation to the median plane. Isolating half-blastomeres, we shall thus obtain dorsal and ventral or right and left or oblique meridional halves. VI. ISOLATION EXPERIMENTS A. Isolation of Animal and Vegetative Halves of Unfertilized Eggs Wilson (1903), Zeleny (1904), and Yatsu (1904, 1910a) found that fragments from any part of the unfertilized Cerebratulus egg may be fertilized and may segment as whole eggs. Zeleny and Yatsu found that fragments obtained during the stages between the fertilization and the completion of the first cleavage show a progressive specification DETERMINATION IN CEREBRATULUS EMBRYO 323 of cleavage factors: just before initiation of the first cleavage the fragments segment as halves. Wilson followed the further develop- ment of fragments from unfertilized eggs and found that, if of sufficient size (not smaller than one-quarter), they may produce normally formed dwarf pilidia. According to Yatsu (19106) even an animal half of an egg that has already given off the polar bodies is able to gastrulate and develop into a pilidium. As the cleavage of fragments has been studied in such detail, I only wanted to repeat the experiments on the localization of gastrula- tion potencies in fragments of the unfertilized egg. Animal and vege- tative fragments were isolated from the beginning of the fading of the germinal vesicle up to the formation of the first polar spindle (as far as the maturation proceeds before fertilization). The eggs used were not very satisfactory for they were the last of the season and the worms had lived for a long time in the ice-box. Eighteen animal and 20 vegetative fragments were fertilized. Five of the animal and 10 of the vegetative fragments died. The remaining 13 animal and 10 vegeta- tive fragments all gastrulated. The best of the vegetative fragments developed into pilidia, whereas the animal ones died before that stage. Our finding that not only the vegetative, but also all animal fragments gastrulate and that vegetative fragments may form apical organs supports Wilson’s results (1903) that fragments from any part of the unfertilized egg are able to produce dwarf pilidia. B. Isolation of Half-Blastomeres Charles B. Wilson (1900) was the first to obtain two pilidia from one egg by isolating the half-blastomeres. E. B. Wilson (1903), Zeleny (1904) and Yatsu (1910a@) found that the isolated half-blasto- meres segment, not as wholes, but typically as if the missing blasto- mere were present, thus a strictly partial cleavage. In the following development, sometimes closed and spherical, sometimes cup-shaped or flat, plate-like forms were observed. All these types of young blas- tule could produce pilidia, but those from the plate-like forms were usually asymmetrical. The best-developed dwarf larve seemed to be typical in every respect. But some abnormalities were frequent among the other pilidia. The apical organ was sometimes displaced towards the anterior end (Wilson). However, Wilson never found this organ duplicated in half-larve. Zeleny observed no constant defect except possibly in the case of the lappets. But he generally killed his larvz before the lappets could have developed fully. Yatsu (19105) obtained some perfect pilidia, some with slight abnormalities, among them not infrequently double apical organs. He concludes 324 SVEN HORSTADIUS that bilaterality of egg substances cannot be detected at the 2- or 4- cell stage. As the segmentation of half-blastomeres has been studied in detail by E. B. Wilson, Zeleny, and Yatsu, I restricted my observations to the differentiated larve. The two half-larve derived from one egg were reared together in order that the development of the bilateral sym- metry in the pairs might be studied. The lappets were formed in four days. Some larve developed into perfect dwarf pilidia. Others were more cylindrical in shape, with a short ciliated band, not forming any lappets. Some looked spherical from the side, namely when the main part of their bodies had the typical form, but the ciliated band was very short, forming only a narrow ring round the mouth. In some cases a lappet was formed on one side only. In many larve the apical organ was displaced towards the anterior end, as Wilson (1903) found, but a displacement to the posterior side was also observed. Contrary to Wilson’s statement, in several cases I saw the apical organ doubled . in the half-larve. Not only was the flagellum split up in two, but larve with two separated ectodermal pits were found. If we now turn to the pairs, there were 6 pairs with both larve typical, symmetrical, thus each larva with two lappets (Fig. 2, B, C). The largest group, 13 pairs, is characterized by one typical and one cylindrical larva. In the latter the shape of the cesophagus-stomach generally indicates a bilateral symmetry, which is, however, hardly expressed in the ectoderm. In 6 cases both partners were cylindrical. In 3 pairs one larva was typical, the other irregular and difficult to interpret. One larva had a lappet only on the right side, the partner being cylindrical. In one pair the two pilidia showed supplementary deficiencies: one had a left, the other a right lappet. This is what would be expected if we had a rather definite bilateral organization of the egg in the 2-cell stage. But against this one case were two where both partners had the lappets better developed on the same side, the right. . In the sea urchin, Hérstadius and Wolsky (1936) have demon- strated that the larve derived from isolated half-blastomeres, from halves of the 16-cell stage, and even halves of the unfertilized egg, show characteristic differences, according to their origin from the left and right, or dorsal and ventral parts of theegg. In Cerebratulus I was not able to refer the half-larve to certain parts of the egg. We know that the larve obtained may correspond to right and left, or dorsal and ventral, or oblique meridional halves of the egg (cf. above). But only in one case of 29 did we find a bilateral asymmetry that would indicate right and left halves. At the same time there were two pairs DETERMINATION IN CEREBRATULUS EMBRYO S25) with the right side better developed in both partners, thus detracting from the value of the supplementary deficiencies in the first-mentioned case. Furthermore, I was not able to detect any features that would incline me to ascribe one partner to the dorsal, the other to the ventral part of the egg. There is no sign of the cylindrical larve arising from a dorsal, or ventral fragment. Thus the abnormalities in the cylin- drical, rounded, or irregular larve do not, as far as I can judge, tell us anything about the origin of these larve. Our conclusion, then, would be like that of Yatsu, that the Cerebratulus egg in the 2-cell stage is not yet so markedly bilaterally organized that the half-larve show deficiencies on the side where material has been removed. C. Isolation of Quarter-Blastomeres The quarter-blastomere also undergoes a fractional cleavage analogous to that characteristic of the half-blastomere (Wilson, 1903; Fic. 4. Four larve from the isolated quarter-blastomeres of one egg. Zeleny, 1904). It gives rise to a much larger proportion of abnormal forms, and even at best the larvee were never normal in every respect (Wilson, 1903; Yatsu, 19106). Zeleny (1904) killed his quarter- larvee before they were fully differentiated. The apical organ was generally markedly displaced towards the anterior end (Wilson, Yatsu), or it might be lacking (Yatsu). The archenteron, also, could be abnormally formed, displaced towards the posterior end (Wilson). Seven quartets of quarter-blastomeres were isolated, the blasto- meres being reared in separate dishes and kept in such order of sequence as in the 4-cell stage, so that we know, for instance, that No. 4 had-had Nos. 3 and 1 as neighbors. The larve were all more or less irregular (Fig. 4). The apical organ was generally displaced, or might be absent; the archenteron was not always well differentiated. The position of the ciliated band was in most cases more or less oblique. But, as in the half-blastomeres, it was not possible to find any definite 326 SVEN HORSTADIUS relation between these asymmetries and a supposed bilateral organiza- tion of the egg. D. Isolation of the Four Animal and the Four Vegetative Cells of the Eight-Cell Stage Zeleny (1904) found that larve developing from the upper quartet have an apical organ, but no archenteron, those from the lower quartet have an archenteron, but no apical organ, while those from lateral four-cell groups have both apical organ and archenteron. Zeleny drew the conclusion that certain organ-forming materials are definitely separated by the third furrow and that the larve from the upper and the lower quartet have not the power of replacing the material lacking. The meridional halves, however,’ possess both kinds Fic. 5. Larve from the isolated four animal cells of the 8-cell stage. of material. Zeleny preserved his larve before they were fully differ- entiated. Yatsu (19100) isolated 7 animal quartets, six of which gave a blastula with apical organ, with 1 to 3 flagella, and, to judge from the drawings, a ciliated field or band at the opposite pole. Thus no gastrulation occurred. Strange to say, the seventh larva developed into an almost perfect pilidium (Yatsu, 19100, Fig. 18, D). Of the 3 vegetative quartets Yatsu mentions two as ‘‘defective.’’ The third is an almost normal pilidium (loc. cit., Fig. 18, J). My material numbers 10 pairs of animal and vegetative quartets, and moreover 9 animal and 4 vegetative fragments of the same kind, but of which the corresponding vegetative and animal quartet respec- tively are missing. None of the 19 animal fragments gastrulated. They gave a blastula with a pretrochal pavement epithelium, a thickened vegetative ciliated band or field, and one, or several, apical DETERMINATION IN CEREBRATULUS EMBRYO Sat organs (Fig. 5). Ten of these blastule had but one normal apical organ, 4 had two, and 3 had three flagella. In most cases, perhaps in all, every flagellum had its own ectodermal pit. In one case the apical organ was entirely lacking. One blastula was very peculiar, with seven flagella, three of them on the pretrochal part, the other four growing out from the ciliated band. The shape of the animal larve is shown in Fig. 5. It is rather rounded or ovoid, the ciliated band occupying the broader or the narrower end. The “ciliated band” has more the character of a ciliated field of high cells, with here and there some irregular protrusions. This ciliated area has the yellowish color of the ciliated band, and the cilia wave in the same typical way. Thus, on the whole, the animal quartet develops just as it would have done in the normal larva, giving rise to the same kind of tissues: pretrochal ectoderm with apical organ, and ciliated band tissue. A difference lies in the frequent duplication of the apical organ. Fic. 6. Larve from the isolated four vegetative cells of the 8-cell stage. The vegetative quartets also (Fig. 6) differentiate in accordance with their prospective significance. An archenteron invaginates, but it evidently corresponds only to the stomach of the pilidium, as it differentiates in the same way, containing the small crystals character- istic of the stomach. The “ectoderm” of the gastrula corresponds to the oesophagus and the insides of the lappets, and at the animal pole of this vegetative fragment we find a ciliated field, presumably derived from the material that normally contributes to the formation of the ciliated band. This field is, however, often divided into two or more ciliated patches of the same kind of tissue as the ciliated band. All the animal and vegetative fragments thus developed in accord- ance with their prospective significance. When Yatsu (cf. above) found that one animal and one vegetative fragment developed into almost perfect pilidia, he must have isolated two meridional halves instead of one animal and one vegetative. The orientation of the 8-cell stage is not always easy, unless the polar bodies are clearly seen. 328 SVEN HORSTADIUS E. Isolation of any In two cases Zeleny (1904) successfully separated the four animal cells of the 16-cell stage (ani) from the twelve lower (an + veg; + veg»). The two anj-larve possessed an apical plate but were not reared long enough to show the differentiation in detail. Twenty-five an; were isolated. One did not develop very well. Fourteen disappeared altogether (cf. p. 319), and the remaining 10 developed into blastule with apical organs (Fig. 7). Five of them had one flagellum, 3 had two flagella, and 2 none. I had the impression that the pit in some cases was missing or not so pronounced as in normal pilidia, but this observation may be erroneous. It may be mentioned that Yatsu (1904), in the uncleaved egg, localized the basis Fic. 7. A-—D. Larve from the isolated four most animal cells of the 16-cell stage:ani. . Isolated veg: (the four most vegetative cells of the 16-cell stage): of the apical organ, not in the animal pole, but in a broad zone a little above the equator. Wilson (1903) also observed flagella without ectodermal pit, namely in larve from vegetative egg fragments. There is need of further elucidation of the question whether the absence of the ectodermal pit in an;-larve, if this takes place, has any relation to the localization suggested by Yatsu. If so, it would indicate a trace of dependent differentiation. The ectoderm of these an;-blastule was ciliated in the usual way. At the pole opposite the apical organ they had small fields of ciliated band tissue (high, yellowish cells). In some cases this field had the form of a protrusion, with the ciliated band like a girdle round it (Fig. 7, D). Jinterpret this to mean that the patch of ciliated band tissue in an; corresponds to that part of the normal band which, on the basis of our staining and transplantation experiments, we ascribed to the an,-cells (See Fig. 3). DETERMINATION IN CEREBRATULUS EMBRYO 329 F. Isolation of ang + vegi + vege The best one of two fragments of this kind that Zeleny obtained (cf. above) formed a ciliated rotating embryo, with a large solid arch- enteron filling up the cavity of the blastoccel. Neither apical organ nor lappets were present. Of the five larve only one developed really well. In the early gastrula stage the archenteron seemed to fill up the entire blastoccel, as Zeleny found. Also in the early pilidium the archenteron looked very large in comparison to the ectoderm. A day later this was no longer the case to such an extent as one would have expected after removal of such a large part of the presumptive ectoderm (Fig. 8, A). A B C Fic. 8. A. any + vegi + veg. B, C._am + an, + vegi. The larva now appears as a small pilidium, save that the apical organ is missing. The stomach is of normal size. The cesophagus is smaller than normal but probably contains as much material as in the normal pilidium, as its wall is thicker. The ciliated band and the pretrochal ectoderm are too small, which is easily understood in that the larger part of the ectoderm has been removed. That, in spite of this, the ectoderm does not look still narrower is probably due to an extreme stretching and thinning of this material. G. Isolation of vege This region constitutes the material for the stomach (perhaps also a small part of the cesophagus). When isolated it does not live long. On the first day after fertilization these fragments were gener- ally all dead. Those which lived the longest formed a mass of cells, as seen in Fig. 7, E. Only in one case were some cilia developed, indicating an initial differentiation. 330 SVEN HORSTADIUS H. Isolation of any + ang + veg Only three fragments of this type were studied, but they all devel- oped in a similar way. They formed, much as the isolated four animal cells of the 8-cell stage, a blastula with an apical organ and a ciliated band (Fig. 8, B, C). Thus no real deep gastrulation took place, although the vegetative end of the larva at one stage may be a little curved inwards, as in a normal larva just at the beginning of the gastrulation. This part later differentiated into an epithelium which mostly looked like ectoderm, but perhaps was a little more densely ciliated. In the larva from the upper quartet of the 8-cell stage the “ciliated band”’ rather was a ciliated field of high cells, occupying the lower part of the larva (Fig. 5). Our present larve differ in the respect that we here have a real band, surrounding the ectoderm-like area just mentioned, which in turn must correspond to the inside of the lappets and the cesophagus. In one of the larve this area first curved inwards and was later turned out again (Fig. 8, B), whereas in the other two it remained as a very shallow invagination; the ectoderm formed like a lappet on one side (Fig. 8, C). Thus also a larva from which veg: has been removed seems to develop in accordance with the prospective significance of the material. I. Isolation of anz + veg, The same holds for the fragments ang + veg:. A great many larvee disappeared. The 9 surviving ones differentiated, although varying in shape, into blastule with a broad, ciliated band around the equator (Fig. 9). As the ectoderm does not expand, the ciliated band Fic. 9. Larvee from the two middle layers of the 16-cell stage: any + vegi. is very thick and broad. The pretrochal, animal epithelium which corresponds to the thin ectoderm formed by ane is generally thinner than the vegetative epithelium of the band (corresponding to the inside of the lappets and the oesophagus). Some larve are more or less massive, and irregular in shape (Fig. 9, C, D). DETERMINATION IN CEREBRATULUS EMBRYO Sol VII. TRANSPLANTATION EXPERIMENTS A. any, + veL2 The most animal and the most vegetative of the cell layers of the 16-cell stage were fused for the purpose of studying whether organs that normally arise from the middle part of the egg could be formed by regulation. The larve gastrulated, the vege-material being invagi- nated. The ectoderm appears identical with that of the isolated an, having an apical organ and a small patch of ciliated band tissue (cf. Figs. 10 and 7). In some cases the archenteron differentiated into just a stomach, very characteristic with its crystals in the wall. Butin other cases I also obtained what looked like a small part of an cesoph- agus (Fig. 10, C). This might be a structure formed by regulation, Fic. 10. A-—C. Larve composed of the most animal and the most vegetative layers of the 16-cell stage: an; + vega. D. an; + er -+Vveg. but, as pointed out on p. 321, it is possible that veg, normally also forms a part of the csophagus. Iam not sure that the third furrow always lies at the same level. I have the impression that the differences in size between the animal and the vegetative cells of the 8-cell stage may vary. As a consequence probably the cleavage plane vegi—vegs also may vary in relation to the limit between the presumptive cesoph- agus and presumptive stomach. This would explain the slightly different results (Fig. 10, A—C). B. any + + vege In three cases vege was added to an; + 2 ane-cells. This increase in ectoderm and ciliated band material gave a larva of essentially the same type as an; + veg, but with a larger ectoderm and a larger amount of ciliated band tissue, which now forms a complete ring (Bie OD): 332 SVEN HORSTADIUS C. Fusion of an Animal and a Meridional Fragment of the Eight-Cell Stage The problem is whether the entoderm material of a meridional fragment of the 8-cell stage can bring about any entodermization of adjacent presumptive ectoderm from an animal fragment, and whether a regulation will take place also regarding other organs (apical organ, ciliated band), so that a harmonic individual arises. The animal fragment of the 8-cell stage was vitally stained before transplantation. At the fusion the polarity of this fragment stands at right angles to that of the meridional one (Fig. 11, A). Fic. 11. A. Fusion of a meridional half and a vitally stained (stippled) animal half of the 8-cell stage. B, D. The larve have differentiated conformably to their prospective significance. C, E. Commencing regulation by redifferentiation. The 5 larve all gave the same result (Fig. 11, B, D). Each com- ponent differentiated in the same way as it would have done normally; there seemed to be no interaction between them whatsoever. At gastrulation none of the adjacent blue animal material was invaginated. The meridional partner developed an apical organ at its animal pole and half of a ciliated band, embracing the mouth like a horseshoe. The animal component clearly has a polarity at right angles to that of the meridional one. The apical organ of the animal fragment is situated DETERMINATION IN CEREBRATULUS EMBRYO 333 at its animal pole; its ciliated band forms a girdle round its base, at right angles to the horseshoe of the meridional component. In the larva, Fig. 11, D, the apical organ of the meridional fragment is not situated close to the stained ciliated band. The position of the two ciliated bands gives the explanation. They are not at right angles to each other. The fusion of the two components has occurred obliquely. Thus both halves differentiated according to their prospective significance. It is now highly interesting to note that, in spite of the lack of interaction in the period of embryological differentiation, we find a regulative interaction in the following development. Several days after the differentiation of the organs of the pilidium, the ciliated band of the animal component gradually disappeared. Only a short part of itremained to form, together with the horseshoe of the merid- ional part, a complete ciliated ring (Fig. 11, C; cf. also E). Thus we here witness a step towards the individualization of this hetero- geneous larva by means of redifferentiation. The larve did not live long enough to show any further changes. VIII. CoNcLUSIONS AND DISCUSSION By means of vital staining the prospective significance of the animal and the vegetative cells of the 8-cell stage, and of the four cell layers ani, ane, vegi, and vege of the 16-cell stage, was studied. The isolation experiments showed that, on the whole, the fragments isolated by cuts vertical to the egg-axis differentiated in accordance with the prospective significance of such parts interpreted through the presence of different organs, e.g., apical organ, ectoderm pavement epithelium, ciliated band, epithelium of the inside of the lappets, cesophagus, and stomach. There is some doubt, however, as to whether all fragments differentiate in every detail according to their prospective significance, as the latter could not be determined with accuracy in all cases. The staining experiment was not quite conclusive as to the fate of am, but the transplantation experiment (Fig. 3), too, indicates that ani normally forms a piece of the ciliated band, as the isolated fragment always does (Fig. 7). It is more uncertain whether veg: normally contributes to the cesophagus. We must thus leave open the question whether the small cesophagus in some of the larvae an: + vege (Fig. 10) was formed from presumptive cesophagus or by regulation. The rather extensive ectoderm in the larva ane + veg: + veg may result from an unusual stretching of the material. The mosaic development of the larve in Fig. 11 supports the view that the apparent deviations from the prospective significance in the differentiation of the cases just mentioned are more apparent than real. 334 SVEN HORSTADIUS It was suggested as possible that the apical organ is not definitely determined in an;. An interesting feature is the appearance of several apical organs in fragments (not only is the flagellum subdivided but there may also be several ectodermal pits). Is that a sign that the organ was not determined at the time of operation? Not necessarily, as it may, on the contrary, indicate a mosaic differentiation. If the apical organ is normally formed from all four quadrants, a slight rela- tive change of position of the animal parts of the blastomeres at or after the operation may cause the parts of the presumptive apical organ to become separated by other ectoderm. In my experiments I never got more than two apical organs in half-larve, only one in the quarter-larve, and, with one exception, not more than two or three in animal fragments, thus never more than the number of quadrants present. On the other hand, the exception, an animal fragment with seven separate flagella (p. 327) seems to show, as well as the possible lack of ectodermal pit in an;-larve (see p. 328), that the determination of the apical organs in fragments is more complicated. The problem must be left unsolved. The scope of this investigation was to compare an egg with mosaic development with that of the sea urchin. An animal fragment of the sea urchin egg (am, or an; + ane of so-called equatorial eggs (HG6r- stadius, 1935, p. 309)) forms less organs (no ciliated band, no stomo- deeum) than it would have done in normal development, whereas a vegetative half, by regulation, often forms more organs than the same material would normally have yielded (the vegetative halves may have apical tuft, mouth, oral arms). An isolated vegs-layer develops into a larva with both more animal (ectoderm) and more vegetative differ- entiations (skeleton) (loc. cit., p. 423) than it would have given rise to in normal development (archenteron). Another way of demonstrating the changes of differentiation that may occur in the sea urchin is to add a vegetative fragment to an animal fragment, e.g., to add the four micromeres to an; or to an animal half (an; + ane). As a result of an interaction between the animal and the vegetative qualities a complete dwarf pluteus develops (Joc. cit., p. 330). Our experiments with Cerebratulus have shown that there is proba- bly no change in differentiation in animal and vegetative fragments of cleavage stages. If any changes take place they are so slight that we have not been able to detect them with certainty. At all events, there is no regulation to compare with that of a vegetative half, or a vege, in the sea urchin, and, in the Cerebratulus fragments, there is no failure of some organs to appear (in comparison to the prospective significance) as in the isolated an; or an; + ane of the sea urchin. We have several DETERMINATION IN CEREBRATULUS EMBRYO 339 directly comparable experiments which show this very clearly. In the sea urchin it is possible to obtain two plutei from one egg, after cutting at right angles to the egg-axis, i.e., if we cut twice and put the two polar parts together (Hérstadius, 1936a). The middle part will thus give more animal and more vegetative differentiations than it would have done normally, the animal and vegetative fragments (an; + the micromeres, or an; + ang -+ the micromeres) will, by regulation, form the organs characteristic of the excised middle part of the egg (archenteron, etc.). The equivalents to these experiments are our isolations of ane + veg; (Fig. 9) and fusion of an; + vege (Fig. 10) in Cerebratulus. As far as we can judge, these larvee show mosaic development. The second experiment implies the adding of an animal half to a meridional half, at right angles to each other as regards polarity. This larva differentiates in the sea urchin into a perfect pluteus. The animal half becomes completely incorporated: a part of it is entodermized, now taking part in the formation of the archen- teron, and the ectoderm, too, acts in all details (ciliated band, stomo- dzeum) as a part of the new individual (Horstadius, 1928, 1935). In Cerebratulus this is not true (Fig. 11). We find a strict mosaic develop- ment, without any trace of entodermization or accommodation of the ectoderm of the animal fragment to that of the meridional half. I recall the fact that the ciliated band of the animal component of these larvae may disappear. We witness an initial change of the heterogeneous larva into a more harmonic individual. It has already been pointed out (p. 333) that this change is brought about by a re- differentiation after the embryological differentiation is already com- plete. In this connection we must mention a similar observation by Yatsu (19106). Whereas an apical organ is not formed in larve which in the cleavage stages were deprived of their most animal part, Yatsu found that the apical organ regenerated, when it was removed from late gastrule or young pilidia. We have to remember that the mosaic development of animal and vegetative parts mentioned above concerns the layers of the 8- and 16-cell stages. Our isolation of animal and vegetative halves prior to fertilization confirms the results of Wilson (1903) and Yatsu (19100), —that any fragment of the unfertilized egg may develop into a pilid- ium. Thus the transition from a regulation to a mosaic type takes place between the beginning of maturation (before fertilization, cf. p. 322) and the 8-cell stage. Yatsu (19100) observed that an animal half of even a mature egg could form a pilidium. Removal of the most animal part of the egg in the first cleavage of the two blastomeres of the 2-cell stage had no effect, but when the most vegetative material 336 SVEN HORSTADIUS was cut off, the archenteron was missing or too small (Yatsu). Thus at that time the localization of the presumptive entoderm seems to have advanced a great deal. When 2-cell stages were compressed so that the second furrow came to lie equatorially, instead of meridionally, and the two animal cells were separated from the vegetative, the former did not gastrulate, whereas the vegetative gastrulated but formed no apical organ. Thus already at the 4-cell stage we seem to have exactly the same animal-vegetative localization as in the 8-cell stage. Wilson (1903) and Zeleny (1904) have assigned the location of the entoderm material to the vegetative part of the egg and the deter- mination of the apical organ to the rearrangement of material at the breaking down of the germinal vesicle. Yatsu (1904) observed, in sections, that a segregation of egg material does actually take place at that period, the yolk accumulating in the lower hemisphere while the clear and more finely granulated protoplasm collects at the animal pole of the egg. Some of the results presented by Yatsu as to the time at which determination has taken place are contradictory. Further in- vestigations are desirable as to how much the determination becomes fixed along with the visible rearrangement of materials, and how much is a progressive process going on after the completion of this rear- rangement, and hence a process of a different character. The rearing of fragments of blastule has given startling results. It would seem as if these fragments were richer in potencies than fragments of cleavage stages. Wilson (1903), Zeleny (1904) and Yatsu (19106) found that animal halves of blastulae could gastrulate, although the archenteron was generally smaller than that of the corresponding vegetative fragment. This result is in conflict with everything we know of the progressive embryological determination which leads to gradual restriction of potencies in every zone. (We recall the fact that the regeneration of an apical organ in the young pilidium and the redifferentiation in our larve meridional half + animal half took place only after the completion of the embryological differentiation—cf. p. 335.) Wilson is inclined to explain at least some of his results on the grounds of oblique sections. In view of this, Zeleny (1904) took special care in determining the orientation, but nevertheless he found in his two cases that the blastula fragments had greater regulative power than those of the 8- and 16-cell stages. One of Yatsu’s animal fragments had a small gut, and the vegetative was provided with an apical organ. Until these results have been con- firmed by means of very careful experiments, I am, like Wilson, in- clined to explain them on the grounds of oblique sections. It is much more difficult to cut a blastula in a desired plane, than a cleavage SS. DETERMINATION IN CEREBRATULUS EMBRYO 33H stage. In the latter you follow a furrow all through the egg. More- over, if you start cutting a blastula equatorially with the correct orientation, the result may be oblique halves, for the knife may pass obliquely in animal or vegetative direction through the egg. This can only be detected by a careful examination of the fragments from all sides. The variability of the results on blastule, and the uni- formity of those on 8- and 16-cell stages speak in favor of this assump- tion. As the fragments of the uncleaved egg are more or less equi- potent, oblique cutting is most readily detected in the blastula stage. If we turn to the fragments in which all the layers are present in the same proportion as in the egg (isolated half and quarter-blasto- meres), we find that they are able to develop into more or less typical pilidia. All the organs are present, but the larvee may be more or less asymmetrical, especially the quarter-larve. On the other hand, several pairs of half-larvee showed a perfect bilateral symmetry (Fig. 2, B, C). These pairs probably correspond to both right-left, dorsal- ventral, and obliquely separated meridional halves (cf. p. 322). Furthermore, the pairs with asymmetrical larve did not show from which parts of the egg the halves originated. This seems to indicate that the bilateral symmetry is not fixed at an early cleavage stage when the animal-vegetative layers are already determined. In this respect a comparison with the sea urchin is of great interest. It has not been possible to study experimentally the unripe egg of the sea urchin. In the mature egg, before fertilization, we find not only an animal-vegetative segregation of presumptive ecto- and entoderm, but also slight traces of a bilateral organization. These traces are more marked in the cleavage stages. Right and left halves show comple- mentary deficiencies of the skeleton on the cut side. Ventral halves develop their ventral side faster and better than dorsal halves. It is very interesting that the dorso-ventral axis in the dorsal halves is inverted (Horstadius and Wolsky, 1936). I regret that time did not allow me to study in a similar way (by staining the cut side of the two isolated half-blastomeres) the relation of the median planes of the half-larve to that of the egg of Cerebratulus. Thus the animal-vegetative determination in Cerebratulus seems to take place between fertilization,—at the beginning of maturation,— and an early cleavage stage. In the sea urchin we find a localization of presumptive ecto- and entoderm already in the mature, unfertilized egg but the same degree of determination is not reached as in the 8-cell stage of Cerebratulus until much later; in the case of some organs not until the beginning of gastrulation (Horstadius, 1936b). On the other hand, the unfertilized, but mature sea urchin egg already seems to have a bilateral organization, which can be traced in meridional frag- 338 SVEN HORSTADIUS ments, whereas a bilateral organization of the Cerebratulus egg is not possible to detect, even in the early cleavage stage. It has many times been pointed out in the literature that there is no fundamental difference between the mosaic and the regulation eggs. In some eggs the determination sets in at an earlier stage than in others. This relative displacement in time is strikingly illustrated in these cases. In the Cerebratulus egg the animal-vegetative determination is ac- complished much earlier; the determination of the bilateral symmetry later than in the sea urchin egg. The formation of mesenchyme has not yet been mentioned. It would have been of great interest to determine whether, in fragments, mesenchyme can be budded off from other parts of the egg than those under normal conditions. In the figures, mesenchyme cells have been drawn in those larve in which they were clearly seen. In many iso- lated fragments we find cells in the blastoccel which are not real mesen- chyme cells, but pathological. Those are more rounded, and often larger than the true mesenchyme cells. We find typical mesenchyme cells in the half- and quarter-larve (Figs. 2 and 4), in the isolated vegetative cells of the 8-cell stage (Fig. 6), in an; + vege (Fig. 10), and in the meridional half + the animal fragment (Fig. 11). Many of the an; (Fig. 7, B, C) and the an; + ano-larve (Fig. 5, A, B) were empty, others had some cells of the pathological type in the blastoccel (Fig. 7, A, D, 5, C, D). Only in one case was just one single cell observed which looked like a true mesenchyme cel] (Fig. 5, C). The any + veg)-larve had some cells in their interior, but I could not determine their nature. To summarize, the larve with vege present show typical mesenchyme cells (Figs. 2, 4, 6, 84 cells not drawn, 10, 11). The animal fragments (Figs. 5 and 7) have with one exception no typical mesenchyme cells. As regards the réle of veg:, we cannot say anything with certainty. The character of the cells in Fig. 9 (ane + vegi) is unknown, and as to the larve an; + ane + veg: (Fig. 8, B, C), I have no records regarding the mesenchyme cells. Even with more detailed observations on the occurrence of the mesenchyme in fragments, it would be difficult to state whether the prospective potency as regards mesenchyme formation exceeds the prospective significance, as observations on normal mesenchyme and mesoderm formation are varied and contradictory. Coe (1899) traces the mesenchyme to the divisions of a large posterior pole cell, as in annelids, and to some of the entoderm cells. Charles B. Wilson (1900) speaks of micro- and macro- mesencytes, both derived from large entoderm cells close to the ecto- derm. E. B. Wilson (1903) describes two symmetrically placed mesoblast cells which, just before invagination, pass into the cleavage cavity near one end of the embryo, and from them smaller mesenchyme DETERMINATION IN CEREBRATULUS EMBRYO 339 cells are budded forth, without, however, giving rise to definite meso- blast bands, as in the annelid embryo. FE. B. Wilson finds it probable that the first two mesoblast cells do not arise from a single cell. All these statements refer to the genus Cerebratulus. Nusbaum and Oxner (1913) compare the observations of previous investigators with their own and come to the conclusion that the mesenchyme in the genus Lineus has a double origin: ‘‘(1) Aus Mesoblastzellen, die sich sehr friih aus den Mikromeren des vierten Quartetts differentieren und (2) aus nachtraglich schon im Gastrulastadium aus dem Entoderm sich abtrennenden Zellen.”” According to Nusbaum and Oxner (1913) the micromere 4d in Lineus divides into two teloblasts (Urmesoderm- zellen), whereas Lebedinsky (1897), in Tetrastemma and Drepanophorus, speaks of four mother cells for four mesoblast bands. In Malacobdella Hammarsten (1918) also traces the mesoderm to four mother cells, 2a—2d. Thus the observations diverge widely, even within one genus, and even within one species. A renewed study of the mesenchyme formation in Cerebratulus, combined with isolation experiments, would be of great value. One problem is whether veg; normally and in fragments buds forth any mesenchyme. As half-larve (which may be right-left as well as dorsal-ventral halves) and quarter-larve always have mesenchyme cells, a second problem is whether the mesenchyme normally arises in all four quadrants of the egg, or the mesenchyme in some of the meridional fragments is formed by regulation. It may be added that the mesenchyme of the pilidium corresponds to both the larval mesenchyme (ectomesoderm) and the entomesoderm of the trochophore, as a part of the mesenchyme cells in the pilidium preserve an indifferent character until the formation of the worm. It would lead us too far to consider here all experiments on other eggs with spiral cleavage. A full review of the literature on annelids and mollusks has been given by Schleip (1929); see also Huxley and de Beer (1934, Chapter V). The main difference between the nemertean egg and those of the other groups is the equality of the four quadrants of the egg in the former. In the uncleaved eggs of annelids and mol- lusks we are dealing with specific substances, more or less visible, which are of an ‘‘organ-forming”’ nature, the pole-plasms. They are precociously chemo-differentiated (Huxley and de Beer) to varying degrees in different species, and, moreover, the time at which they become specifically localized varies. Very often a polar lobe is formed temporarily during the first cleavage divisions. Removal of this lobe leads to an absence of the apical organ and to a defective post-trochal region; no somatoblasts will be formed. Because of this the isolated quarter-blastomeres develop in a very different way. Only the D- quadrant—the one that has the polar lobe—can produce somatoblasts 340 SVEN HORSTADIUS (which give rise to the ectodermal and mesodermal (entomesoderm) germ-bands. The A-, B-, and C-blastomeres cannot develop into complete larve. They may have larval mesenchyme (ectomesoderm), but are devoid of entomesoderm. As regards the apical organ, the conditions are, however, not so uniform. In Patella, for example, (Wilson, 19040) all four quadrants acquire an apical organ, in Dentalium (Wilson, 1904a) only the D-quadrant, whereas in Sabellaria (Hatt, 1932) the factor determining the apical organ goes to the C-quadrant. In mosaic development we thus find (at varying times) precociously differentiated substances distributed to particular regions of the egg, and in varying degrees by different forms, and subsequently isolated by the cell divisions. If the vegetative pole-plasm be equally distrib- uted to the first two cells (Tubifex, by heat or deprivation of oxygen, Penners, 1924; Chetopterus, Nereis, and Cumingia by compression, high or low temperature, centrifuging, or anaerobiosis, Titlebaum, 1928, Tyler, 1930) double monsters are formed, or the blastomeres may, if isolated, give rise to more or less complete larve. In isolated normal half-blastomeres we thus find a self-differentiation in AB, on the one hand, differing from that of CD on the other, but when the vegetative pole-plasm is evenly distributed, both are capable of pro- ducing all organs. An interesting problem still remains unsolved. The isolated normal blastomeres differentiate as fragments, but would the same blastomeres behave in the same way if placed in contact with each other atypically? Although the ‘‘organ-forming’’ sub- stances are now separated by cell-walls, could any kind of interaction be detected, any kind of change of the prospective significance of the material be brought about? Penners (1926, 1934) obtained, after killing teloblasts of Tubifex at different stages, a development in many respects of mosaic character, but also detected a certain dependence of the ectodermal and mesodermal components of the germ-bands on each other, and on the entoderm of the germ-bands, etc. In Cerebratulus we also have an early rearrangement of substances, namely at the time of the breakdown of the germinal vesicle. But these substances are not, as far as we know, unevenly distributed to the quadrants. As has been pointed out above, we do not yet know with certainty or in detail at what time the progressive animal- vegetative determination is accomplished. The potencies of the blastula are still particularly obscure. Nor do we know how much of that determination is due to a rearrangement of substances, or to another metabolic process. A close study of mesenchyme formation in Cerebratulus, both in normal development and in animal-vegetative and meridional fragments, would be of special interest in rendering possible a comparison with the conditions in annelids and mollusks. DETERMINATION IN CEREBRATULUS EMBRYO 341 IX. SUMMARY 1. The prospective significance of the animal and the vegetative cells of the 8-cell stage, and of the layers an, ane, vegi, and vege of the 16-cell stage (Fig. 1) was studied. an; forms the greater part of the pretrochal ectoderm, including (probably) the most anterior part of the ciliated band (Fig. 3). ame gives rise to the rest of the pretrochal ectoderm, and a great part of the ciliated band. veg) also partakes in the formation of the ciliated band, and, moreover, differentiates into cesophagus and the insides of the lappets. veg, corresponds to the stomach. The boundary between the layers could not, however, always be determined with complete accuracy. 2. The first furrow may form any angle to the median plane of the larva. 3. The two isolated half-blastomeres from one egg may develop into pilidia with perfect bilateral symmetry (Fig. 2B, C). When the isolated half- and quarter-blastomeres are less typical, their abnormal- ities do not show from which part of the egg the dwarf larve come. The bilateral symmetry is not determined in the early cleavage stage. 4. Animal and vegetative fragments differentiate, as far as we can judge from comparison with our results concerning the prospective significance, in the same way as they would have done in normal development: the animal (Fig. 5), and the vegetative (Fig. 6) cells of the 8-cell stage; an; (Fig. 7, A—D) ; vege (Fig. 7, £); ane + veg: + vege (Fig. 8, A); an; + ane + veg, (Fig. 8, B, C); ane + veg: (Fig. 9). 5. Larve composed of an; + vege did not form any of the organs normally derived from the excised middle part of the egg (Fig. 10). Fusion of an animal with a meridional half of the 8-cell stage also showed a complete mosaic development (Fig. 11). 6. The results are discussed and compared with those with sea urchins, annelids and mollusks. The desirability of further investiga- tions is emphasized. LITERATURE CITED Cor, W. R., 1899. The maturation and fertilization of the egg of Cerebratulus. Zool. Jahrb., Anat., 12: 425. DrigEscH, Hans, 1891. Entwickelungsmechanische Studien. I. Zezischr. f. wiss. Zool., 53: 160. HAMMaARSTEN, OLoF D., 1918. Beitrag zur Embryonalentwicklung der Malacobdella grossa (Miill.). Inaugural-Dissertation Stockholm. Uppsala 1918: 1. Also in Arbeten fr. Zootomiska Institutet, Stockholm, 1. Hatt, PIERRE, 1932. Essais expérimentaux sur les localisations germinales dans l’oeuf d’un Annélide (Sabellaria alveolata L.). Arch. d’Anat., Micr., 28: 81. HOrsTApius, SVEN, 1928. Uber die Determination des Keimes bei Echinodermen. Acta Zoologica, Stockholm, 9: 1. Horstapius, SvEN, 1931. Uber die Potenzverteilung im Verlaufe der Eiachse bei Paracentrotus lividus Lk. Arkiv. Zool., Stockholm, 23: No. 1. 342 SVEN HORSTADIUS Horstapius, SvEN, 1935. Uber:die Determination im Verlaufe der Eiachse bei Seeigeln. Pubbl. Stazione Zoologica, Naples, 14: 251. HORSTADIUS, SVEN, 1936a. Weitere Studien iiber die Determination im Verlaufe der Eiachse bei Seeigeln. W. Roux’ Arch. Entw.-mech., 135: 1. Horstapius, SVEN, 19366. Uber die zeitliche Determination im Keim von Para- centrotus lividus Lk. W. Roux’ Arch. Entw.-mech., 135: 1. HORsSTADIUS, SVEN, AND ALEXANDER Wo tsky, 1936. Studien tiber die Determina- tion der Bilateralsymmetrie des jungen Seeigelkeimes. W. Roux’ Arch. Entw.-mech., 135: 69. HUXLEY, JULIAN S., AND G. R. DE BEER, 1934. The Elements of Experimental Embryology. Cambridge University Press. LEBEDINSKY, J., 1897. Beobachtungen iiber die Entwicklungsgeschichte der Nemertinen. Arch. mikr. Anat., 49: 503. Nussavum, J., AND M. Oxner, 1913. Die Embryonalentwicklung des Lineus ruber Miill. Zettschr. wiss. Zool., 107: 78. PENNERS, A., 1924. Experimentelle Untersuchungen zum Determinationsproblem am Keim von Tubifex rivulorum Lam. I. Arch. mikr. Anat. u. Entw.- mech., 102: 51. PENNERS, A., 1926. Experimentelle Untersuchungen zum Determinationsproblem am Keim von Tubifex rivulorum Lam. II. Zeitschr. f. wiss. Zool., 127: 1. PENNERS, A., 1934. Experimentelle Untersuchungen zum Determinationsproblem am Keim von Tubifex rivulorum Lam. III. Zeztschr. f. wiss. Zool., 145: 220. SCHLEIP, WALDEMAR, 1929. Die Determination der Primitiventwicklung. Leipzig, Akad. Verlags. TITLEBAUM, A., 1928. Artificial production of Janus embryos of Chaetopterus. Proc. Nat. Acad. Sct., 14: 245. ' TyLer, A., 1930. Experimental production of double embryos in annelids and mollusks. Jour. Exper. Zodl., 57: 347. von Upsiscu, L., 1925. Entwicklungsphysiologische Studien an Seeigelkeimen, I. Zeitschr. wiss. Zool., 124: 361. von Usiscu, L., 1936. Uber die Organisation des Seeigelkeims. W. Roux’ Arch. Entw.-mech., 134: 599. Voet, WALTHER, 1923. Morphologische und physiologische Fragen der Primitivent- wicklung, Versuche zu ihrer Lésung mittels vitaler Farbmarkierung. Siéz.- ber. d. Ges. f. Morph. u. Physiol. Miinchen, 35: 22. Voct, WALTHER, 1925. Gestaltungsanalyse am Amphibienkeim mit Ortlicher Vitalfarbung. I. Arch. Entw.-mech., 106: 542. WESTBLAD, EINAR, 1923. Zur Physiologie der Turbellarien. Kungl. Fysiograf. Sdllsk. 1. Lund Handlingar, N. F., 33: No. 6. Witson, CHARLES B., 1900. The habits and early development of Cerebratulus lacteus (Verrill). Quart. Jour. Micros. Sct., 43: 97. Witson, Epmunp B., 1903. Experiments on cleavage and localization in the nemertine egg. Arch. Entw.-mech., 16: 411. Witson, Epmunp B., 1904a. Experimental studies on germinal localization. I. The germ-regions in the egg of Dentalium. Jour. Exper. Zodl., 1: 1. Witson, Epmunp B., 1904+. Experimental studies on germinal localization. II. Experiments on the cleavage-mosaic in Patella and Dentalium. Jour. Exper. Zoél., 1: 197. Yartsu, N., 1904. Experiments on the development of egg fragments in Cerebratulus. Biol. Bull., 6: 123. Yatsu, N., 1910a. Experiments on cleavage in the egg of Cerebratulus. Jour. Coll. Sct. Tokyo, 27: No. 10. Yartsu, N., 19106. Experiments on germinal localization in the egg of Cerebratulus. Jour. Coll. Sci. Tokyo, 27: No. 17. ZELENY, CHARLES, 1904. Experiments on the localization of developmental factors in the nemertine egg. Jour. Exper. Zoél., 1: 293. PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS PRE- SENTED AT THE MARINE BIOLOGICAL LABORATORY Jumy, 6.1937 Parthenogenetic merogony in the Naples sea-urchins. Ethel Browne Harvey. Parthenogenetic merogony, or development of non-nucleate parts of eggs which have been artificially activated, has been studied in the four species of sea- urchin commonly occurring at Naples. The non-nucleate fractions were obtained, as previously, by breaking the eggs apart with centrifugal force. In Arbacia pustulosa, which is similar to the Arbacia at Woods Hole, asters and cleavages occur in the non-nucleate half; when activated, similar to what has been described for Arbacia punctulata. In Parechinus (Echinus) microtuberculatus and Paracentrotus (Strongylo- centrotus) lividus, the fertilization membrane in activated non-nucleate halves is well separated from the cell surface, just as in normally fertilized whole eggs of these species. A large monaster is formed, an amphiaster and there are one or two cleavages. The cleavage planes frequently disappear and some time afterward the egg breaks up into a number of pieces. These become progressively smaller and more numerous until what resembles a blastula is formed. This breaks through the fertilization membrane, but no further development has been observed. The non-nucleate halves of Sphaerechinus granularis can also be activated and they break up into small pieces in a similar way. It might appear that this breaking up of the egg is a form of degeneration or cytolysis. But the same spontaneous breaking up of the cell has been observed also in normal whole nucleate eggs artificially activated. These sometimes gave rise to typical blastulae which became free swimming and looked normal in every respect. It has also been found that by treating the immature egg with parthenogenetic agents, there is formed a very definite layer on the periphery of the cell similar to the ectoplasmic layer in the fertilized or activated mature egg. This is formed both on the half containing the germinal vesicle and the heavier half separated from it by centrifugal force. Another evidence of activation is the failure to respond to sperm by the formation of blisters on the surface. The treated, im- mature egg, after some time, pinches off a small piece, always at the part most distant from the germinal vesicle in elongated eggs. Later there are two, three or more pieces and then a great number. A recent critical examination of prepared slides of the parthenogenetic mero- gones of Arbacia punctulata shows the presence of asters, usually in pairs, but no chromosomes. The Feulgen reaction, in the procedure of which I was assisted by Dr. Jean Brachet, is entirely negative. The parthenogenetic merogones show no red-staining nuclear material, whereas the fertilized merogones which were used as control material, show it very clearly. Some oxidative properties of isolated amphibian germinal vesicles. Jean Brachet. Amphibian germinal vesicles isolated from full-grown odcytes in a saline solution absorb oxygen and eliminate carbon dioxide during several hours; the 343 344 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS oxygen consumption was measured by means of a modified Garard-Hartline micro- respirometer using Triturus pyrrhogaster as material while the COz excretion of Rana fusca nuclei was estimated by a titrimetric micro-method. The QOs2 (cu. mm. O2) —_____________—_ of the isolated nuclei (the approximate wet weight being grams wet weight X hour calculated from the volume and the density) is about 13 while the full-grown odcytes have a QO: of 37. The respiration of one single isolated nucleus is only 1-1.5 per cent of the amount of gas exchanged by one full-grown odcyte deprived of its follicular epithelium. Removal of the germinal vesicle does not affect markedly the CO: elimination of full-grown odcytes during several hours. Ad- dition to the nuclei of glucose or of cytoplasm removed from the odcytes with a micropipette does not increase their respiration. Presence or absence of Ca ions (which, according to W. R. Duryee, greatly affect the physical properties of the germinal vesicle) does not affect appreciably the metabolic rate of the isolated nuclei. Influence of temperature and other agents on the respiration and devel- opment of marine eggs. Albert Tyler. (Most of the data given in this report appear in an article on p. 261 of this issue of the Biolog- ical Bulletin under the title, “On the energetics of differentiation, VI.” The report has also been abstracted in the Collecting Net of July 10.) aoe wants Effects of fatigue due to muscular exercise on the Purkinje cells of the cerebellum of mice at various ages. Warren Andrew. The problem of morphological changes in nerve cells as the result of func- tional activity has been studied by a number of investigators. Nevertheless, the question still remains open not only as to what changes occur due to activity but as to whether any such changes do occur. Among the factors to be taken into account in such work is that of age. The present experimental work is based on earlier work in which the Purkinje cells of mice and rats were studied from the time of their differentiation up to and including extreme senility of the individual. Sixteen black mice were used in the experimental work on fatigue. “ Fa- tigue”’ means complete exhaustion brought about by running in a motor-driven rotary cage. For each fatigued animal, a control animal of the same brood was killed at the same time and the tissues from the two animals carried through the technical processes together. The animals represent a range of ages including pairs of 23 days, 25 days, 43 days, 46 days, 98 days, 101 days, 746 days, and two animals with marked signs of senility—of 744 and 746 days, killed without fatiguing. In each case 100 cells were examined. The major conclusions to be drawn from the present work are: 1. There are morphological changes in nerve cells as the result. of fatigue carried to exhaustion, consisting primarily in a loss of Nissl material, an increased basophilic property of the nucleus, and an increased average cell size. 2. In senile animals there is also a loss of Nissl substance and an increase in the basophilic properties of the nucelus. The binucleate condition of the Purkinje cell is a phenomenon of senility. The differences between the Purkinje cells of senile and of young animals are far - more marked than are those between exhausted and fresh animals of the same age. PRESENTED AT MARINE BIOLOGICAL LABORATORY 345 Localization in the oculomotor nuclei of the goldfish. Zareh Hadidian, Milton S. Dunn and Roland Walker. The position of cells innervating the individual eye muscles was studied to see whether there is any pattern in the oculomotor nuclei which might be correlated with the type of eye coordination, which differs from that in animals with bi- nocular vision. The method was a study of chromatolysis in the oculomotor cell bodies after cutting a peripheral nerve or its muscle, since these operations gave similar re- sults. In normal animals there was a negligible proportion of cells showing ad- vanced degeneration, while in animals with both eyes removed practically all the cells showed chromatolysis. There were no changes in the nuclei studied after re- moval of the contents of an eyeball. When the contents of one orbit were removed, thus insuring complete damage to all fibers of III and IV on that side, about half of all the oculomotor and trochlear cells were altered.. Chromatolysis in the trochlear nucleus was con- tralateral, while the degenerate oculomotor cells were about 70 per cent homo- lateral, the contralateral ones being mostly in the ventromedial portion. Chromatolysis after damage to the inferior oblique muscle or its nerve was about 80 per cent homolateral, with a slight tendency toward ventromedial local- ization. For the anterior rectus degeneration was about 60 per cent homolateral with no localization. The inferior rectus showed about 85 per cent homolateral degeneration with definite localization in the dorsolateral nucleus anteriorly, changing gradually toward the ventromedial nucleus posteriorly. The superior rectus had only about 20 per cent homolateral cells, and the contralateral group was localized in the ventromedial nucleus. A consideration of these results leads to the conclusion that since there is poor definition of cell groups, and scattering of cells for any one muscle through- out the nucleus, any merely anatomical studies of oculomotor localization are in- adequate for the understanding of the type of eye codrdination in the goldfish. Some new observations on the secretory activity of neurones. E. Scharrer. (Followed by Demonstration. ) Nerve cells having more or less the appearance of secretory cells have a wide distribution. Among the invertebrates they have been found in annelids, molluscs, crustaceans and insects; among the vertebrates, in the diencephalon of selachians, teleosts, amphibia, reptiles and mammals, including man. In the case of the bony fishes, gland-nerve cells have also been found in the nucleus of the nervus ter- minalis, in the midbrain and in the caudal region of the spinal cord; the latter being especially well developed in selachians (see Speidel, 1919, 1922). All stages can be found, from typical nerve cells containing only a few granules in the cyto- plasma to cells with a spectacular formation and storage of droplets of a col- loid-like substance. There are even cases, such as that of the Mediterranean fish Cristiceps, where the nerve cells in the so-called “diencephalic gland” are trans- formed into gland cells lacking any nervous character. A marked nuclear poly- morphism is also typical for many gland-nerve cells and pericellular as well as endocellular blood capillaries are often observed in the secretory diencephalic nuclei of vertebrates. Observations of this kind, even when based on such extensive material, would not suffice to prove the glandular function of the cells in question. It must be demonstrated that there is a functional cycle in the production of the colloid ma- terial by the nerve cells in the neuro-secretory regions of the nervous system. This has been done with the diencephalic gland (nucleus praeopticus) of the American toad (Bufo americanus). Sections through this gland show all stages, 346 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS from the first appearance of fine granules in the cytoplasma of the cells, which stain brilliant orange in Azan preparations, to larger droplets, which finally are extruded and lie as colloid masses among the cells. This cycle can be shown clearly and the identity of those processes in nerve cells with the different stages of secretory activity in gland cells seems sufficiently demonstrated. The physio- logical meaning of the gland-nerve cells is still unknown and a wide field is opened for future investigation. Synaptic transmission in the sixth abdominal ganglion of the crayfish. C. Ladd Prosser. Transmission through the sixth abdominal ganglion of the crayfish was studied by simultaneous recording of impulses entering and leaving the ganglion in response to stimulation of caudal sensory hairs. Flexion of one hair gives rise to one sensory impulse. To set off one efferent unit, however, summation of two to four afferent impulses in different fibers is necessary. No efferent neurone is excited by one incoming impulse. When several efferent units are excited more afferent impulses are required to activate the first efferent unit in a response than to activate later ones. Thus both convergence and overlap play a part in conduc- tion through this ganglion. Ganglionic delays range from 3 to 30 milliseconds as measured from the time the first sensory impulse enters the ganglion. Most of the fastest units show delays of 5-6 milliseconds, and later ones fall in groups which are multiples of the first. This multimodal distribution of delays is interpreted as indicating the ex- istence of internuncial neurones. A given unit may show fluctuation of 2-3 milli- seconds in synaptic delay. The afferent neurones respond to stimuli separated by intervals as short as .01 second, whereas the efferents show recovery times of .05 to .1 second. This synaptic recovery time is longer than that of the fibers, hence no relatively refractory period can be detected in the responses of the indi- vidual efferent units. There are no connections from tactile receptors across the ganglion to contra- lateral efferent nerves of the sixth segment. Acetylcholine and eserine have no facilitating action upon the synapse. Eser- ine is toxic in high concentrations. Nicotine blocks conduction through the syn- apse. Excess potassium reduces action potentials in the afferent fibers and may block them before affecting the synapse. Adrenaline acts similarly to excess potassium. It is concluded that those humoral agents which mediate transmission in some mammalian ganglia do not have a similar action in this crayfish ganglion. Jwiex, 20 Chemical stimulation of the amphibian ectoderm. L. G. Barth. Further work on the chemical nature of the amphibian organizer indicates that the formation of the neural plate is due to some general stimulus imparted to the cells by a variety of substances. Earlier experiments using cephalin as an in- ductor showed that cytolysis occurred in the region of the implant. Following this other cytolytic agents such as digitonin, acids and bases were used and plates of neural cells were induced in the presumptive epidermis. Digitonin in concentra- tions of .5 to .05 °/o in powdered egg albumen and buffers at pH 3 and 10 have given positive results by implantation into the blastocoel of Amblystoma mexi- canum. With regard to the naturally occurring organizer, it was found impossible to extract it completely with fat solvents. In comparing the action of ether-alcohol extracts with protein residues, the protein residue gave better neural tubes. The PRESENTED AT MARINE BIOLOGICAL LABORATORY 347 ether-alcohol extracts were relatively weak in inducing power. The experiments suggest that the naturally occurring organizer is in the protein residue. Limb bud transplantation in chick embryos. Viktor Hamburger. Wing and hind-limb primordia of chick embryos, incubated 2-3 days (25-30 somites) were transplanted to the lateral trunk region or into the coelomic cavity of host embryos of the same stages. In a considerable number of experiments, the anterior-posterior and the dorso-ventral axes were inverted with respect to the axes of the host embryo. The transplants showed complete self-differentiation with respect to form, size and axial pattern. Transplants located near the spinal cord of the host were supplied by trunk nerves or by nerves branching from the brachial or the lumbo-sacral plexus of the host. The spinal ganglia contributing to the innervation of the transplant reacted always to even small increase in their peripheral fields by hyperplastic growth. In several cases, the number of the motor neurones in the level of the spinal cord which supplied the transplant was likewise found to exceed the number of motor neurones of the normal side, the hyperplasia ranging from 14 per cent to 30 per cent. Transplants located far ventrally were not innervated by spinal nerves. They showed, nevertheless, normal development and differentiation. The embryonic development of the limbs of the chick is therefore independent of innervation. Adult organizers and their action in adult tissues. Oscar E. Schotté. The development of a salamander, Amblystoma punctatum. L.S. Stone (motion picture. ) The development of the common black, yellow spotted salamander, Ambly- stoma punctatum, has been recorded in detail on one motion picture reel showing all the stages from the one-cell egg to the time the larva begins feeding. Rate in de- velopment was recorded at various speeds from a few hundred to several hundred times in order to analyse various stages more carefully. The detailed study in development extends over a period of about four weeks during which time a constant temperature of about 70° Fahrenheit was maintained. During segmentation one can see clearly the movement of blastomeres, the regional waves of cell division and movement of egg mass due to shifts in the center of gravity. The formation of the blastopore and the development of the neural plate and closure of the neural folds are shown in detail. All the changes in surface development are shown throughout the tail-bud stages from the time at which cilia on the surface of the embryo first come into action to the period when motor activity begins. Occasional local quivering movements in various parts of the embryo are also recorded. Following these periods in development, dorsal, ventral and lateral views of embryos are shown which carry the growth up to the time the larva begins feeding. The formation of many external features as well as the beginning of peristaltic movements in the intestines are clearly seen. In appropriate parts of the film are shown clusters of eggs as they are nor- mally laid, circulation in the tail fin, and a view of the larva taking its first meal. The picture begins with a view of a typical habitat and ends with a view of several stages in development from the egg to the adult animal in order to give a compari- son of sizes during growth. 348 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS Jury 27 The effect of standing on the carbon dioxide content of alveolar air and total ventilation volume. J. K. W. Ferguson and F. A. Hitchcock. A study of the respiratory exchange during consecutive ten-minute periods of reclining and standing, has revealed in every experiment a decrease of 7-15 per cent in the CO: content of alveolar air (confirming Higgins, 1914, Turner, 1927 and Main, 1937). This fall, however, cannot be attributed to overventilation in the usual sense of the word, because: (1) the R.Q. fell during standing in 21 out of 24 experiments, (2) the CO. output fell in 12 out of 24 experiments, (3) the total ventilation fell in 9 out of 24 experiments. When four consecutive periods, with the subject alternately reclining and standing, were studied, it was found that the Oz consumption and the CO. output were greater during the second reclining period than during the preceding stand- ing period. This is interpreted as evidence of oxygen debt and COs: retention during standing. It seems probable that this disequilibrium, during standing, be- tween metabolic requirements and the respiratory exchange is due not to inade- quate pulmonary ventilation but to inadequate circulation in the dependent parts of the body resulting from the erect posture. The mechanism of the loss of heat from the human body. Eugene F. DuBois and James D. Hardy. The three main channels of heat loss are radiation, convection and vaporiza- tion. The respiration calorimeter of the Russell Sage Institute of Pathology in New York Hospital measures the vaporization by weighing the moisture that comes from the skin and lungs. It measures radiation plus convection in a stream of cool water that flows through coils in the top of the calorimeter. Radiation is then determined independently by means of a Hardy radiometer pointed in rapid succession at 20 different places on the surface of the body. The total surface is measured and the effective radiating surface calculated as 80 per cent of the total. Convection is then estimated by difference. It has been possible for the first time to separate the factors of radiation and convection in persons exposed to varying atmospheric conditions such as are found indoors. Two normal men were studied naked at temperatures between 22 and 35°C. Their basal metabolism was uniform throughout this range not rising until a few minutes before the onset of shivering which occurred after exposures of two hours at 22°C. At the higher temperatures there was profuse sweating and practically all of the heat was lost through vaporization. The percentage of heat lost in radiation decreased steadily with rising temperatures. In the neutral zone of 28 to 32°C. convection accounted for 10-15 per cent of the heat loss and radia- tion 50-60 per cent. Convection was markedly increased by slight movements of the body. With moderate exercise or shivering it accounted for 25-30 per cent of the total loss. An electric fan raised the percentage to 33. Two athletes playing violent squash for 36 minutes showed a 2°C. rise in rectal temperature and about an equal fall in average surface temperature. The total heat lost through radiation increased but little, the percentage fell to 15. Convection dur- ing the exercise and recovery period accounted for 5-15 per cent. Vaporization dissipated 70-80 per cent of the heat. Peripheral inhibition of smooth muscle. Emil Bozler. The antagonism of vasoconstrictor and vasodilator nerves was studied using perfused frog legs and rabbit’s ears and recording the vascular responses by a sensitive flow-meter. Stimulation of the vasodilators of the dorsal roots blocks the action of single volleys of impulses of the vasoconstrictor fibers, whereas re- PRESENTED AT MARINE BIOLOGICAL LABORATORY 349 petitive stimulation produces'a response. Likewise acetyl choline blocks entirely the action of the first few impulses produced by repetitive stimulation. In an attempt to explain these results a difficulty was encountered in the observation that neither vasodilator stimulation nor acetyl choline antagonise the action of adrenaline. The simplest explanation seems to be the assumption that during in- hibition the passage of impulses from vasoconstrictor fibers to the muscle is partially blocked, thereby preventing the formation of the adrenaline-like mediator. It is suggested that the increase of polarisation, which has been observed in other cases of inhibition, is the immediate cause of the partial block. The mediator produced by the vasodilator fibers may be the cause of the change of polarisation of the muscle fibers and, therefore, indirectly also of the partial block produced by the activity of these nerve fibers. The relationship of tissue chloride to blood chloride. William R. Am- berson, Thomas P. Nash, Arthur G. Mulder and Dorothy Binns. A number of previous investigators have found that amphibian muscles, placed in, or perfused with, isotonic sucrose solutions, rapidly lose their chloride almost completely, while retaining almost all of their potassium and phosphate. When such muscles are soaked in solutions of varying chloride content, the tissue chlo- ride varies directly with that of the external fluid. Such observations have led to the conclusion that muscle chloride is extra-cellular, a view reinforced by its. low concentration in this tissue. Other students have recently attempted to extend this concept to other tissues, and to the mammalian body. In studying this literature it occurred to us that it would be useful to know whether the chloride of the mammalian tissues can be diminished when the plasma chloride is lowered. We have found it possible to produce very radical diminutions in the plasma chloride by a modification of the method of total plasmapheresis described by Stanbury, Warweg, and Amberson (Am. Jour. Physiol., 1936, 117, 230). We make up our artificial plasma with sulfates instead of chlorides, adding chloride-free gum acacia, and suspending in this plasma ox red blood corpuscles which have been rendered chloride-free by many washings through sulfate-Ringer-Locke solution. By long perfusion of the mammalian body we are able to remove most of the chloride before death ensues. In some tissues, such as skeletal muscle, liver and kidney, the tissue chloride falls off in direct proportion to the plasma chloride, as perfusion proceeds. In other tissues the straight line through all the points does not pass through the origin, but shows a y-intercept, suggesting that a portion of the chloride is indiffusible, and presumably intra-cellular. The stomach and spleen have particularly large intercepts of this character. The central nervous system resists removal of chloride, whereas most of the chloride of peripheral nerve may be washed away. ‘The brain and cord chlo- ride is held either (1) because it is largely intra-cellular, or (2) because the sulfate ion is unable to penetrate into the brain tissue. Certain tissues, such as tendon and lung, have chloride concentrations so high that it is impossible to explain it all by allotting it to extracellular fluid. AuGust 2 The use of diatoms from geological excavations at Clovis, New Mexico, as indicators of water conditions. Ruth Patrick. Mammoth Pit lies between Clovis and Portales in the Staked Plains region of New Mexico. The stratigraphy is as follows: The lowest stratum is coarse gravel devoid of diatoms. Above this is a stratum of speckled sand, in which is a diatom flora which seems to indicate a fresh to brackish water condition by the 350 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS dominance of Anomoconeis sphaerophora (Kiitz.) Pfitzner, Amphora ovalis Kiitz., and Amphora ovalis var. pediculus Kiitz. Horse skeletons are found in this level. The next stratum is a “bluish clay” of about the same constituents as the speckled sand plus considerable carbonized vegetable matter. In the lower part of this stratum, a very rich diatom flora consisting mainly of fresh water species such as Eunotia arcus Ehr., Cymbella affinis Kitz. Fragilaria brevistriata Grun. and Synedia ulna (Nitzsch.) Ehr. was laid down. The change in abundance and also in the kind of species from the previous level points to a freshening of the water. This agrees with other geological evidence that this was a period of much higher precipitation than now exists in New Mexico. Mammoth skeletons are most abundant in this level of the stratum. Passing from bottom to top of this stratum the typical fresh water species disappear. The dominant species near the top are the brackish or alkaline water types such as Epithemia argus Kiitz., Rhopalodia gibba (Ehr.) Mill., and R. gibberula (Ehr.) Mill. Coincident with the change in diatom species, the mammoth bones disappear and bison skeletons be- come much more numerous. The top stratum consists of brown dune sand devoid of diatoms. It is in the bottom of this layer that the bison bones disappear. Thus the change in water conditions as indicated by the diatoms seems to coincide with a change in the fauna as shown in this stratigraphy. Gemmuipary in Kalanchoe rotundifolia and other Crassulaceae. Harry N. Stoudt. The phenomenon of vegetative propagation in Crassulaceae has received much attention from students or morphology and physiology. A comparison of the morphological development of plantlets of this species with other members of the family should aid in understanding more adequately the phenomenon so character- istic of the group. Yarbrough (1936) reports that apparently mature parenchyma cells in the base of the petiole of Sedum Stahlii resume mitotic activity to form plantlets when the parent leaf is removed from the plant. Stoudt (1934) published an account of vegetative propagation in Byrnesia Weinbergu. In this species a plantlet forms at the base of the sessile leaf from a residual meristem that is undifferentiated into plantlet rudiments. Freeland (1933) discusses this phenomenon in Bryophyllum crenatum in which plantlets develop from residual meristems in the notches of the parent leaf. He finds that the amount of differentiation of plantlet rudiments varies. A stem primordium only may be formed by the time the parent leaf is mature. A nine-millimeter leaf of Kalanchoe rotundifolia reveals a meristematic region on the adaxial surface of the petiole. This region is undifferentiated into plantlet rudiments but when the leaf attains its maximum size, three to four centimeters in length, leaf and stem primordia have formed. The bud then remains dormant until the leaf is severed from the parent plant. Root, stem and leaf primordia are usually visible in the notches of the mature leaves of Bryophyllum calycinum according to Naylor (1932) etc. while in Kalan- choe daigremontiana and K. tubiflora plantlets consisting of root, stem, and leaf primordia are visible macroscopically even before the parent leaf has attained maximum size. Thus there is a definite sequence into which the various species arrange them- selves in respect to the degree to which the meristematic cushion becomes dif- ferentiated by the time the parent leaf is mature. Their greatest differences are in the stage of development attained by the meristem, or organ rudiments derived from it, at this time. PRESENTED AT MARINE BIOLOGICAL LABORATORY Sol Pollen analysis of the air in relation to hay-fever. A. O. Dahl. It is essential for the successful diagnosis and treatment of hay fever (“ pol- lenosis”’) that detailed data concerning the concentration of pollen in the air of any species at a given time or locality be made available. Records of the pollen content of the air have been obtained by exposing each day an oil-coated slide out of doors for 24 hours. The pollen grams observed in 25 systematically distri- buted low-power fields are identified and counted. The approximate number of pollen grains per cubic yard of air can be calculated by use of physical formulae. The pollen grains involved in hay fever in Minnesota vary from approximately 15 to 80 micra in diameter. Hay fever is a regional problem and atmospheric data from one locality will not apply in a detailed manner to another place. In Minnesota, pollen concen- trations of clinical interest occur between late March and early November. For purposes of clinical convenience, the pollens found in the air during the entire season have been placed into 16 groups. Thus, in diagnosis, one scratch-test for each group will test the patient’s sensitivity to all pollens present during the entire season. Prefaced by such procedure, successful therapeutic measures can be insti- tuted. (The complete report is to be published in joint authorship with Dr. C. O. Rosendahl and Dr. R. V. Ellis, under whose direction the study has been carried on for the last 5 years at the University of Minnesota). Avueust 3 A convenient test of physical agents as producers of dominant lethals. P. W. Whiting. Dominant lethals occurring in the spermatozoa cause failure of development of zygotes. Since in bisexual reproduction eggs also fail to develop unless fer- tilized, the two types of male sterility, due to (1) lack of sperm and (2) dominant lethals, cannot be statistically distinguished. In Habrobracon, reproducing males by haploid parthenogenesis, matings with males lacking sperm result in as many progeny per day (gd) as cultures from unmated females (4.73 gd) or as matings with untreated males (1.21 fd, 3.30 99). If fathers are X-rayed, males per day are not increased while females per day are decreased (to 2.22 with 2,500 R, 0.62 with 5,000 R, 0.16 with 7,500 R, 0.07 with 10,000 R, and 0 with 20,000, 40,000 or 75,000 R). Since neutrons induce dominant lethals and ultra-short (one-meter) radio waves do not, only the former may be expected to induce recessive lethals and visibles. Cytological observations on colchicine. Bernard R. Nebel. The action of the alkaloid colchicine on mitosis was studied in the following material: . Stamen hairs of Tradescantia, roots and shoots of Zea, Vicia, Tomato, Tagetes, Antirrhinum, Trifolium, Papaver, Dianthus, Solanum, and Lilium,— testes of Podisma, eggs of Asterias and Arbacia. In Tradescantia all stages of mitosis are easily seen in life. The cells will continue to divide in salt sugar solutions to which drugs may be added. Colchicine in concentrations of 5 X 10* to 5 X 10° molar will stop mitosis during metaphase. In concentrations of 5 X 10° to 10° molar it tends to produce binucleate cells. When blocked in metaphase the plate in Tradescantia is char- acteristically tilted, since a true spindle is not developed. A comparative study of phenyl-, amyl-, propyl-, ethyl urethane and chloral-hydrate showed that these anesthetics within their respective reversible concentration ranges will only oc- casionally produce binucleate cells. There is no particular evidence of a metaphase 352 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS block with the urethanes. In studying the action of colchicine on the developing egg of Arbacia punctuiata it was necessary to use fixed material to determine the nuclear stage accurately. Colchicine applied 10 minutes after insemination will block the first cleavage metaphase in concentrations above 10* molar. Between 6.5 X 10° and 3.5 X10° molar nuclear divisions of abnormal type proceed while cleavage is impeded. The rhythm of nuclear division may persist so that when controls are in third cleavage metaphase the eggs to which colchicine was added 10 minutes after fertilization will also show approximately 4 groups of chromatin, but the plates are hypoploid, often containing only one to six separate chromosomes, which may partly represent fused units. The micronuclei which form are not far apart from one another and their resting stage is relatively short; with lower con- centrations the rhythm of division is not affected in the early cleavages, but the subsequent development of the larvae is markedly stunted even at 10° molar. Colchicine in increasing concentrations thus first interferes with the normal course of cleavage, then astral rays are prevented from forming, next the spindle is re- duced in size and finally obliterated. Meanwhile chromosomes become pycnotic during a prolonged metaphase; they fail to divide orderly; the abnormal division may separate entire chromosomes rather than split halves. In the plants, in which root and shoot growth was studied (Ruttle) cuttings and seedlings respectively were subjected to colchicine treatments by immersion in aqueous solutions primarily. The active concentration ranges were found to be the same as in Arbacia and Tradescantia. All genera showed marked reactions, the tomatoes being the least sensitive. Where plant meristems were treated, the resulting tissues showed markedly irregular growth—incised and crumpled leaves as well as chlorophyll defects. Cytological investigation showed necrotic cell lineages and multinucleate cells in varying degrees. The drug is being studied further as an agent which may induce mutations and polyploidy. In cooperation with other investigators, some physiological effects of col- chicine were investigated. Respiration was tested on Arbacia eggs with col- chicine 1.8 X 10° to 7.5 X 10°. Respiration did not vary from the control in any of these concentrations (Tyler). Pectinmethoxylase (pectase) from tomato juice showed inhibition which is of doubtful significance since with higher concentrations of colchicine a precipita- tion was observed in the reaction mixture (Kertesz). The nitroprussiate reaction for S— H=S—S groups gives a color reaction in stamen hair cells of Tradescantia, the color being located in the chromonemata and in certain plasmatic granules. Cells under colchicine gave the same reaction (Medes). No significant influence was observed on the action of carbonic anhydrase from blood. No inhibition occurred of the reduction of methylene-blue by yeast or blood with and without admission of air. No sensitization of Arbacia eggs to X-rays was obtained by the addition of colchicine, when nuclear dbnormalities were used as a criterion. Demonstration of vital staining preserved in paraffin sections of lamprey embryos (Bismark Brown method). R. Weissenberg. In 1929 I recommended Bismark-brown for localized vital staining of the egg of lamprey because it is very easy to preserve these stained areas for paraffin sections. The quick-working method which I am using now in my studies of localiza- tion on lamprey egg is a very simple one. It is based on the surprising fact that Bismark-brown as a vital dye is alcohol-proof without further treatment in con- trast to Nile-blue sulfate. a ee PRESENTED AT MARINE BIOLOGICAL LABORATORY Soh) I fix the embryos in a mixture of 1.5 parts of absolute alcohol and 0.5 part of acetic acid for ten minutes, wash in absolute alcohol for a few minutes, and trans- fer them directly into cedar oil. The stained areas are very well preserved by this simple method and the eggs can remain in the cedar oil for years without any loss of the dye. They can be studied in the cleared condition in the cedar oil as total preparations, or they can be imbedded in paraffin at any time and cut with excellent preservation of the vital stain within the sections. The slides demonstrated are balsam preparations seven years old and still give a true representation of the stained areas of the living organism. The preservation obtained by this method is complete also in the finer localization of the staining. Carriers of the vital staining in the egg of lamprey are chiefly the yolk granules because here, in contrast to most amphibian eggs, pigment granules are missing in the earlier stages of the embryo. Microfilm on some experiments on isolated amphibian germinal vesicles. William R. Duryee. The film shows colloidal changes in the frog ovocyte nucleoplasm, nucleoli, and chromosomes brought about by relatively slight changes in the Na, K, Ca chloride concentrations of the medium. H* ions reverse the normal negative charge on the nuclear components to positive. When this change is gradual enough, as with 0.003 N HCl, a dark converging “ring” forms from flocculating particles in the approximate pH region of 4.0 to 5.0. Ca, Mg, Cu, Hg and basic dyes behave similarly to H* ions in causing a phase separation and an appearance of chromosomes from a previously transparent nu- cleus. On the other hand, K and Na and especially OH tend to disperse the nuclear colloids, thus stretching and separating the chromosome pairs, and at the same time making the nucleus transparent. Within narrow limits these changes are reversible. In Triturus pyrrhogaster, the “ Binnenkorper,’ or first maturation spindle anlagen situated at the center of the germinal vesicle, can be made to swell and separate the chromosomes radially, but not in typical bipolar directions. Jana fusca appears unique in having a differentiated coagulable capsule around the chromosomes, which may be important in forming the denser portions of the spindle. In R. pipiens the contraction of this material under the influence of calcium is less striking. Similar changes, including phase separation and violent contraction of the chromosomes, occur when acid fixatives are added or when the nuclei are exposed to their disintegrating cytoplasm. Hence this latter effect may be termed auto- fixation. Prominent differentiated areas or sac-like projections of the nuclear membrane reversibly swell and shrink in bases and acids respectively. Such struc- tures are obliterated by fixatives. It is concluded that in the forms studied Dar- lington’s assertion as to the absence of a nuclear membrane must be modified. With merely fine forceps and pipette it is easily possible to isolate various com- ponents of these giant nuclei (0.8 mm. diameter),— e.g. nuclear membrane, spindle anlagen, nucleoli, and chromosome pairs (100-200“)—during the fall, winter and early spring months of the year. Cortical cytolysis of the echinoderm egg. Robert Chambers. (Motion picture. ) Motion pictures were taken of those experiments designed to demonstrate the physical properties of the cellular cortex particularly in marine ova. Nearly all experimentally induced effects on living cells are extremely transitory. A later study of the motion picture record permits of a more careful and complete analysis 354 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS of these phenomena. For example, the film may be slowed down, stopped at any point or reversed at will. The film presented at the seminar is a compilation of many of these experiments. Included in the film is a scene showing the spontaneous coalescence of a de- nuded Arbacia egg with an oil drop. From such a record it is possible to measure with considerable precision the diameter of the oil drop at the instant of coalescence. These data are essential if the coalescence phenomena are to be treated quantita- tively, yet it is practically impossible to obtain the oil drop sizes in any other way. Another scene shows the natural elevation of the cellular surface and the re- action of this surface to mechanical injury. Over-insemination of an immature Arbacia egg results in the formation of many large insemination cones. Gentle manipulation with a microneedle will cause these cones to run together, thus bring- ing about an elevation of a continuous film which is separated from the granular cytoplasm by a liquid space. If this elevated film is ruptured at one spot with a microneedle, the entire film will rapidly disintegrate. The underlying cytoplasm becomes exposed and it quickly becomes converted into an irreversible coagulum. The following phenomena, inter alia, are also demonstrated: cytolysis of star- fish eggs in hypotonic sea water, effects of tearing eggs in pure salt solutions isotonic with sea water and the shrinking of the cortex of one blastomere following the puncture of the other. The photography was done by Mr. C. G. Grand. Aucust 10 Some aspects of normal and regulative development in the colomal ciliate, Zoothamnium alternans. F. M. Summers. A remarkable number of studies on metazoan “ organizers” have already demon- strated the importance of extrinsic factors for determination in specific parts. It was felt that additional information about these factors could be gained by ap- plying operative techniques to an animal type in which, presumably, the relation- ships of parts have not attained so great a degree of complexity. Zodthamnium alternans is a protozoan colony whose cells collectively possess in some degree many of the attributes of an individual organism. It is admirably adapted to this type of work for many reasons, particularly by virtue of the precision with which the characteristic colonial pattern develops. One of the most important consequences of this study of more than 200 normal and operated colonies is the demonstration of qualitatively different physiological relations between cells at different locations on the colonial framework. Under normal conditions a specific pattern unfolds. When the cell at the apex of the frond-like colony is cut away some cell of a lower order, one whose complete developmental potentialities are never otherwise expressed, assumes the dominant generative functions and the normal colonial pattern perseveres in the parts re- generated by it. These results are intelligible in terms of what Child (1929) calls physiological correlation: the relations of dominance and subordination between parts. Apical control appears to be continuous and quantitative. In this organism the transformation of the apical cell into an ex-conjugant initiates a developmental phase which furnishes another clue to the general nature of apical control. About four days after the union of gamonts, even before ex- conjugant generations are produced, the normal developmental relations are upset in an unusual way. The first three or four branches below the conjugant level begin to develop out of all proportion to the normal expectations. Each branch develops almost as an individual colony. Its cells divide precociously, forming secondary and even tertiary branch strains. The greatest effect obtains on the branch nearest the conjugant and diminishes basally as a gradient ; the basal branches are apparently unaffected. PRESENTED AT MARINE BIOLOGICAL LABORATORY 355 Under varying physiological conditions in the apical cell the coordinating in- fluences exerted upon the mitotic activity of neighboring cells may be inhibitory (as shown by the regulative response after the apical cell is removed) or excita- tory (when the apical cell is transformed into an ex-conjugant). The precocious development does not occur when the apical cell is present or when it is dissected away; it appears to be effected by some new quality in the coordinating mechanism arising in consequence of conjugation activities in one particular cell—the apical cell. These results invite the conclusion that the integrative factors in a colony of Zoothamnium are qualitative and discontinuous. Morphology, behavior and reproduction in Type A and Type B of Chaos chaos Linnaeus, the giant multi-nucleate amoeba of Roesel. A. A. Schaeffer. Chaos chaos Linnaeus 1767,.the first amoeba to be discovered, was described by Roesel von Rosenhof, a painter of miniatures, in 1755, in Germany. It has been seen 5 times since then: in 1900 by H. V. Wilson, North Carolina; in 1902 by E. Penard, Switzerland; in 1916 by W. A. Kepner in Virginia and by A. A. Schaeffer in Tennessee; in 1936 by A. A. Schaeffer in New Jersey. The general morphology and behavior of this amoeba are so much like those of the common laboratory amoeba, Chaos diffluens Miller, 1786, that there is strong probability that both amoebas belong to one and the same species. Conclusive evidence of such relationship is, however, still lacking and therefore, until such evidence is found, the two taxonomically specific names, C. chaos and C. diffluens, will be used to avoid confusion. C. chaos falls into at least two types which are distinct in some morphological details and also in antigenic reactions. Type A has discoid nuclei and divides usually while freely rolling around on the substrate. Type B has broadly ellipsoid nuclei, and divides while fastened to the substrate. The nuclei of type A are larger than those of type B. With the collaboration of Dr. J. A. Harrison, preliminary antigenic tests were made in which it was found that there is a marked difference between the two types, type B standing closer to diffiluens than to type A. Three of the most striking differences between chaos and diffluens are: in size, chaos being from 50 to 500 times as large as diffluens; in number of nuclei, chaos being multinucleate (up to 1,000 or more) ; in reproduction, chaos dividing at any single division, into 2, 3, 4, 5 or 6 daughters, with a strikingly marked mode at 3 daughters. Pieces cut from chaos grow up to full size, whether the piece contains one or more nuclei. Both types give off, when crushed, a strong cucumber-like odor which can readily be detected when only one amoeba is crushed on a slide. In diffluens this odor is also present but to a much smaller degree. aH ea j Observations upon the chemical composition and the metabolism of a larval parasitic nematode. Theodor von Brand. The experiments were performed with an immature Eustrongylides from Fundulus heteroclitus. The red color of the worms is due to the presence of hemo- globin in the body fluid. With its low fat and high glycogen content the general chemical composition resembles that of the adult Ascaris. The worms consume per unit weight much less glycogen than Ascaris, if kept in saline at 37° C. under aerobic conditions. They are able to keep their glycogen level high, even if their hosts starve for a long time and lose during this starvation period more than half of their polysaccharide stores. 356 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS Observations and experiments on sex change in the adult American oyster, Ostrea virginica. Paul S. Galtsoff. Sex of oysters was determined by inducing ovulation or ejaculation by in- creased temperature and chemical stimulation. During the summer of 1936 indi- vidual spawning records of 203 adult oysters were obtained and the discharged products microscopically examined. Oysters were measured, carefully marked by numbers engraved on the shell of each, and transferred to Milford, Connecticut, to be kept in tidal tanks through the winter. The same procedure was repeated dur- ing the summer of 1937. Results showed that 9.7 per cent of the oysters reversed their sex; the per- centage of reversals being higher among females (13.1 per cent) than among males (8.0 per cent). Mortality during the year was only 7.04 per cent and failures to react, due to lack of gonad development, was only 1.07 per cent, indicating healthy conditions under which oysters were kept. In the sex-reversed males the physiological set-up of the organism changes with the sex; typical female reaction, characterized by the rhythmicity of con- tractions of adductor and maintenance of a constant tonus level, develops in the former male. In several instances development of this reaction lagged and the newly-formed female retained physiological characteristics of the male, i.e, eggs were discharged through the cloaca instead of through the gills and rhythmical contractions of adductor were absent. All sex-reversed females reacted as true males. It is concluded that female reaction developed as a secondary adaptation providing mechanism for dispersing eggs throughout the water. From the simultaneous occurrence of sex reversal in the oyster population the conclusion is made that Orton’s theory of metabolism change (protein to carbohydrate) as the sex-determining factor in O. edulis is not applicable to O. virginica. Avucust 17 A sea water buffer for marine eggs. Albert Tyler and Norman H. Hor- owitz. (This paper has already appeared in full in Science, July 23, vol. 86, pp. 85-86. ) The effect of CO, upon the oxygen capacity of the blood of some fresh- water fish. Edgar C. Black and Laurence Irving. Conditions of respiration for fish differ from the respiratory conditions for mammals... The respiration of fresh-water fish must proceed in a medium in which the pressure of oxygen is always less and the pressure of carbon dioxide usually greater than in atmospheric air. At different levels in the water, pres- sures of gases are altered by the temperature changes. Types of blood which are suitable for the transport of CO, and oxygen under one set of conditions might be quite unsuitable under another set. The characteristics of the blood of the carp (Cyprinus carpio L.) and the common sucker (Catastomus commersonit) show examples of two types of blood, each suited for a different and limited range of pressures of oxygen and CO.. Oxygen dissociation curves obtained from. those two species are not as sigmoid as are those for mammalian blood. The presence of 5 or more mm. CO, (for the carp 10 or more) prevents the complete saturation of whole blood, even at high partial pressures of oxygen. This effect of CO, is quite different from the familiar effect of CO. upon mammalian blood. In the presence of relatively high pressures of CO, the blood of the carp is suitable for the transport of small quantities of oxygen, while the blood of the sucker would be quite useless. PRESENTED AT MARINE BIOLOGICAL LABORATORY Si At very low pressures of CO, and high pressures of oxygen in the water the blood of the sucker can serve to transport much more oxygen than that of the carp. Hemolysis by the addition of saponin to the blood of the sucker, carp and bowfin (Amia calva) abolished the effect of CO, upon the oxygen capacity at high pressures of oxygen (150 mm.). The CO, effect is in part at least de- pendent upon the integrity of the corpuscles. Oxidative mechanisms in the resting and fertilized sea-urchin egg. Irvin M. Korr. Since cyanide inhibits certain iron-containing systems, and since pyocyanine, a bacterial respiratory pigment, functions as a “hydrogen-carrier,” it was possible to vary the relative proportions of respiration going through simple, non-ferrous carrier and that going through the cytochrome-indophenol oxidase system; by adding KCN and pyocyanine, separately and in various combinations, to sea urchin eggs. The respiratory rates of the untreated fertilized and unfertilized sea urchin eggs, and those in which the mechanisms had been altered as above, were measured at different temperatures. These experiments were designd (1) to give the temperature relations of the two types of respiration, (2) to throw fur- ther light upon the factors determining temperature coefficients of cellular respiration and (3) upon the change in oxidative rate and mechanism that occurs upon fertilization of the sea urchin egg. It was found, first (in partial confirmation of Rubenstein and Gerard, 1934), that the temperature coefficient of unfertilized eggs was much higher than that of the fertilized egg. The effectiveness of KCN was found to increase with temperature. Increasing the respiration of the fertilized egg with added hydrogen-carrier did not appreciably change its temperature coefficient. Fertilized eggs, in which the iron system had been maximally inhibited and the respiration restored to or above normal with pyocyanine, also had the same temperature coefficients as the untreated fertilized eggs. In the unfertilized egg, whose respiration is cyanide-stable, the addition of carrier not only increases the rate of respiration, but also, above a certain con- centration, lowers the temperature coefficient. A concentration of pyocyanine which increases the respiratory rate to that of the fertilized egg also lowers the coefficient to that of the fertilized egg. The results, in conjunction with older work, lead to the conclusions that (1) respiration through a simple non-ferrous carrier and that through the cyto- chrome system do not, per se, have different temperature coefficients, that (2) these H-transfer mechanisms are the rate-controlling link in the respiration of the fertilized and unfertilized egg. (3) Temperature coefficients are largely de- termined by the ratio of the rate at which H-atoms are transferred, from sub- strate to oxygen, to the maximum rate at which they can be produced by the substrate-dehydrogenase systems. The more nearly the H-transfer rate ap- proaches the maximum H-production rate, the lower the temperature coefficient, and vice versa—within the limits set by the fertilized and unfertilized egg. Methods for the study of rapid chemical reactions and thew application to the kinetics of enzyme-substrate and enzyme-tnhibitor compound formation. Kurt G. Stern and Delafield DuBois. The observation of spectroscopically defined enzyme-substrate and enzyme- inhibitor compounds, made in the course of recent studies on catalase and peroxi- dase, offers an experimental approach to the detailed analysis of the mechanism of 358 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS action of these enzymes. A photoelectric method for the recording of such proc- esses and some preliminary results were reported last year (K. G. Stern and D. DuBois, J. Biol. Chem., 116, 575 (1936)). This method has since then been im- proved by replacing the single photoelectric cell by a differential photometer con- taining two photocells. The technique of mixing by injecting one reactant into the solution of the other with a spring gun has been retained. The new apparatus is quite insensitive to any but color changes. In addition, a simple spectro- graphic method has been developed which permits the recording of fast reactions without the use of photocells or electric instruments. This is achieved by replacing the plate holder of a spectrograph by a falling plate camera. While the plate is falling, the trigger of the spring gun is released and mixing is complete within 8 to 27 milliseconds. The time is recorded by a rotating time marker or a Neon tube flash circuit. With this method the reaction of catalase and methemoglobin with hydrogen peroxide, ethyl hydrogen peroxide, hydrocyanic acid, and hydrofluoric acid has been studied. The reaction rate is greatly de- pendent on the concentration of the reactants and on their ratio, as would be expected from a bimolecular process. The reactions studied appear to be slower than the reactions of hemoglobin or hemocyanin with oxygen or carbon monoxide which have been measured by Hartridge, Roughton, and G. A. Millikan with the flow method. Only when a considerable excess of substrate or inhibitor over the catalysts is employed does the rate of the reactions studied by the present authors approach the length of the mixing time which is, of course, the limiting factor in such experiments. Inasmuch as the rate of fall of the photographic plate may be varied within wide limits (0.3 to 29 cm. per second), a wide range of reaction rates may be studied. The use of supersensitive panchromatic plates permits the recording of changes of light absorption in the red and green region of the spectrum at rates of fall of the plate corresponding to an exposure time of less than 0.002 second for an individual spectrum. The continuous strip of spectra recorded on the plates corresponds to 350 individual spectra. Aucust 24 Mechanism of cellular death by freezing. B. Luyet. (The essentials of the paper were published in the August issue of Biodynamuca.) Binding and penetration of bivalent cations in Elodea cells. Daniel Mazia. The fact that Elodea cells contain soluble oxalates in their vacuoles makes it possible to study by a direct method the binding of Ca, Sr, and Ba ions. By subjecting the cells to strong electric currents or to ultraviolet radiation or to plasmolysis and deplasmolysis, one can set free the bound Ca in their protoplasm, which can then be observed as a precipitate of distinct calcium oxalate crystals in the vacuole. The Ca must come from the protoplasm, for the cells are kept immersed in distilled water or non-electrolyte solution. That the Ca actually is bound in living cells is indicated by the fact that it cannot be washed out by prolonged immersion (up to 14 days) in distilled water, although it can easily be removed by a few minutes of washing in a citrate solution and replaced then by a few seconds of immersion in a 0.01 M CaCl, solution. Further studies on leaves from which the Ca has been removed by citrate show that it can be rebound to a maximum level from Ca solutions as dilute as 5 X 10° M, the time required increasing with the dilution. This binding is influenced by Na and also by K ions, so that when the Na/Ca ratio is greater than about 100, the binding of Ca is largely prevented. It is possible to substitute Sr and Ba in the place of Ca in the protoplasm. They are bound in the same way as Ca. Cells in which they have been substituted PRESENTED AT MARINE BIOLOGICAL LABORATORY 359 for Ca seem to function normally, but when subjected to agents regularly causing the release of Ca show characteristic crystals of strontium oxalate or barium oxalate in their vacuoles. The Sr ion penetrates the protoplasmic layer easily, even when there is no concentration difference between the Sr inside and outside or even a higher con- centration inside. It can be demonstrated that this transport requires first the binding of the Sr in the protoplasm. I{ Na or K is added to an Sr solution in sufficient concentration to prevent the binding of Sr in the protoplasm, the pene- tration of Sr does not occur, whereas, in the control, a pure SrCl, solution of the same concentration, the transport does occur, and crystals of strontium oxalate can be seen in the vacuole. Factors governing cellular responses to nitro and halo phenols. G. H. A. Clowes, A. K. Keltch and M. E. Krahl. A recent comparison of respiratory stimulating and cell division blocking effects of three closely related compounds having pK values ranging from 4.1 to 4.5, mononitrocarvacrol, dinitrocarvacrol and dinitrothymol, shows that the first and third substances exert little or no effect on respiration but block division at the extraordinary dilution of 2 X 10° M. The second increases respiration almost four-fold to a peak at 10° M, at which concentration division is blocked. A fourth structurally related substance, 2,4-dinitro-o-isopropyl phenol, having a pK of 3.0, produces a moderate effect, giving a respiratory peak and cell division block at 10° M. These results lend further support to the conclusion previously reached that, while the substituted phenol anion undoubtedly exerts a significant effect inside the cell, it is quite impossible, from a knowledge of only the dissocia- tion constants of such substituted phenols, to predict with certainty the range of concentrations, if any, in which a given compound will affect either cell respira- tion or division or both. From experiments conducted in 1935 and confirmed in 1936, in which varying numbers of eggs were used in a constant volume of sea water medium, certain substituted phenols were observed, contrary to general experience with anes- thetics, to block division at greater dilution when larger numbers of eggs were employed and vice versa. This was believed to be attributable to a rise in intra- cellular acidity due to CO,. At last year’s meeting it was demonstrated that with varying CO, tensions, incapable in themselves of blocking division, the division-blocking effect of substituted phenols was greatly enhanced. In an attempt to evaluate the relative role of undissociated substituted phenol molecule and anion, the dissociation constants of some thirty substituted phenols have been determined during the past winter and used in analyzing the respiratory stimulating and cell division blocking effects obtained with the compounds in question at fixed exterior and varying interior acidities. These will be reported in the following paper. On the assumption that the substituted phenols penetrate the living cell only as undissociated molecules, for a given total concentration of the phenol, the intracellular concentration of phenol molecule and phenol anion in an intracellu- lar aqueous phase may be calculated for any levels of extracellular and intra- cellular acidity. The possible role of acidic dissociation in the physiological effects pro- duced by nitro and halo phenols. M.E. Krahl, G. H. A. Clowes and A. K. Keltch. From experiments performed during the summer of 1935, using a constant extracellular pH of 7.5 and a constant intracellular pH approximating the normal 360 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS 6.8, the intracellular concentrations of phenol molecule and phenol anion necessary to give 50 per cent reversible inhibition of division and approximately optimum respiratory effect (where this effect was present) were calculated for 30 substi- tuted phenols. The necessary concentration of phenol molecule was found to vary from 3X10" M for 2,4-dinitro-o-cyclohexyl phenol to 3 X 10* M for m-nitrophenol. The necessary concentration of phenol anions likewise varied over a wide range, from 1.0 X 10° M for 2,4-dinitro-o-cyclohexyl phenol to 2X 10% M for 2,4-dinitro-o-isopropyl phenol. In confirmation of our previous work and contrary to the results of Tyler and Horowitz (Proc. Nat. Acad. Sci., 23: 369, 1937), picric acid and o-nitrophenol, when free of impurities, produce no reversible stimulation of respiration or reversible cell division block. It is clear that no final conclusion regarding the precise roles played by the phenol anion and undissociated phenol molecule can be reached from such experiments, and that there is some as yet undetermined factor involved. The conclusions of Tyler and Horowitz, who used a limited number of compounds which happened to fall, for the most part, in the middle of the above series, are therefore not justified by the evidence at present available. With a constant extracellular pH and a decreasing intracellular pH, the ratio of intracellular anion to intracellular substituted phenol molecule decreases, while the concentration of phenol molecule is independent of intracellular pH and dependent only on the total extracellular concentration of phenol and the extracellular pH. In experiments on fertilized eggs of Arbacia punctulata, it was found, as anticipated, that at constant extracellular pH of 6.7, the concentra- tions of 2,4-dinitrophenol, 4,6-dinitro-o-cresol, 2,4,5-trichlorophenol, 2,4-dichloro- phenol, and m-nitrophenol necessary for 50 per cent division block were not af- fected by decrease in the intracellular pH, while the optimum levels of respiratory stimulation were decreased with decreasing intracellular pH, this decrease in respiratory optimum being largest with 2,4-dinitrophenol and 4,6-dinitro-o- cresol which have pK values of 4.1 and 4.4 respectively, and small or negligible with 2,4-dichlorophenol and m-nitrophenol which have pK values of 7.7 and 8.3 respectively. Depolarization of muscle and nerve membranes by organic substances. Rudolf Hober and Bernard R. Nebel. It is fairly generally accepted that some surface film of muscle and nerve fibers is the seat of a polarized state resulting from the high content of the in- terior of the fibers in free K ions and from the selective permeability of this film to cations. Furthermore, it is believed that the negative electric wave sweeping along the fibers after excitation is indicative of a local and reversible propagated depolarization due to an increase of ion permeability. This increase would be the result of an electro-chemical reaction, which involves a structural alteration of the surface film. Since the excitation process has been shown to be con- nected with an increased metabolic activity of the fibers, it seemed worthwhile to study whether organic compounds either identical or more or less related to normal constituents of the fibers would bring about depolarization. Experiments were performed on sartorius muscles and sciatic nerves of the frog, complemented in coOperation with Dr. M. Andersh, by studying nerves of the spider crab. Injury potentials were measured, following the usual pro- cedure. The experimental result is this, that not only certain organic cations, com- parable to the normally penetrating K ions, but also certain organic anions and non-electrolytes are enabled to depolarize the surface membrane, as disclosed by the arising electronegativity. The active cations concerned are those of higher dialkylamines and of alkaloids, the anions those of higher fatty acids and bile acids, the non-electrolytes anesthetics and saponin-like compounds. All these PRESENTED AT MARINE BIOLOGICAL LABORATORY 361 substances are likewise significant by their cytolytic power, which is associated with a polar structure of their molecule, with surface activity and lipoid-solubility. Lytic effects frequently appear to be irreversible. But under certain conditions, e.g., with the fatty acids by shifting the pH from a more acid to a more alkaline reaction, the depolarized state can be returned to the normal polarization. These statements tempt one to raise the question whether reversible cytol- ysis may play a rdéle in producing the traveling negativity, the propagated re- versible disturbance of the surface film of the excited fibers. Support may be lent to such an assumption by the facts that the phosphatides, characteristic con- stituents of the plasma membranes, are containing surface-active higher fatty acids and that electric currents, in passing artificial membranes, comparable to the action currents accompanying excitation, have been demonstrated to alter the ion concentrations, particularly the H ion concentrations in the electrolyte solutions bordering the membranes. Under such conditions, in the membranes various physico-chemical or chemical events could be released. GENERAL SCIENTIFIC MEETING Aucust 26 On some conditions determining sub-cooling in plant tissues. B. J. Luyet and E. L. Hodapp. Organisms found in frozen water or on the frozen ground are said to be sometimes killed by a slight concussion incapable otherwise of harming them. This would be attributable to the sudden freezing of the sub-cooled tissues under the action of the concussion. In the present experiments we studied the condi- tions determining sub-cooling and freezing after sub-cooling, in the potato tuber. Sub-cooling was found to occur with about the same frequency in living and in dead tissue. The temperature to which the material was heated before being cooled had an evident influence on preventing sub-cooling. Freezing of the sub- cooled tissues can definitely be induced by concussion, although the relatively high percentage of inefficient shocks indicates that some unknown factors, un- influenced by concussion, hold the system in the sub-cooled condition. On the double freezing point of some living tissues. B. J. Luyet and Sister P. M. Gehenio. Some plant tissues have been found to present sometimes two freezing points, one a few tenths of a degree below zero and the other about a degree lower. In the present work we investigated the conditions in which one obtains the double freezing-point in the potato tuber. From a large number of determinations it results that it is a character of living tissue and that a congelation of the material at the first freezing-point does not kill it, while at the second it does. The presence of the double plateau in the freezing curve, and its shape, have been studied in terms of the cooling velocity, of the occurrence of sub-cooling, of the size of the piece of tissue, of its degree of desiccation or of imbibition, and of the type of thermometric device used. The possible factors responsible for the double freezing point are discussed. Transverse electric impedance of the squid giant axon. H. J. Curtis and Ke SGCole. The transverse electric impedance of the giant axon of the lateral mantle nerve of the squid has been measured by means of a Wheatstone bridge over a frequency range from 200 to 2,500,000 cycles per second. The bridge current was 362 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS at all times kept well below that necessary to stimulate, and excitability was tested at the end of each run. The phase angles measured for this axon ranged from 70° to 85° which indicates that the membrane impedance is of the polarization type. These phase angles are considerably higher than those found for nerve bundles from the same animal, so it seems very likely that part of the low phase angles found for nerve bundles may be due to a statistical distribution of fiber diameters and membrane capacities. In several cases, impedance runs were taken both before and immediately after the fiber lost excitability, and none of the impedance char- acteristics of the axon changed when this occurred. Some time later, however, the membrane impedance dropped to zero which indicated the death of the cell. Membrane capacities found for this axon average 0.42 ui/cm.” at 100,000 cycles, which is in good agreement with the values previously found for nerve bundles. Electric impedance of suspensions of unfertilized and fertilized Arbacia COGS ieee olerands)ia vies Spencer, The alternating current resistance and capacity of suspensions of Arbacia eggs in sea water, measured at frequencies from one thousand to ten million cycles per second, give average membrane capacities, with 90° phase angles, of 0.84 wi/em.? for the unfertilized, and 3.5 ui/cm.’ for the fertilized eggs. Some slightly lower phase angles were probably indications of abnormalities. The previously reported additional capacity element in the fertilized eggs disappeared as the season progressed. It was found that the plasma membrane enclosed volume averaged 1.7 per cent less than the non-conducting volume for the un- _ fertilized egg and 2.5 per cent less for the fertilized egg, while the fertilization membrane enclosed volume averaged 32 per cent greater than the non-conducting volume. Thus the plasma membrane lies very close to the non-conducting mem- brane in both the unfertilized and fertilized egg, and the fertilization membrane is practically perfectly conducting. It is then probable that, on fertilization, the plasma membrane capacity increases to some four times its unfertilized value. Electric impedance of single Arbacia eggs. K.S. Cole and H. J. Curtis. The end of a two or three-mm. thin-walled glass tube is heated until it closes down to a short capillary about 50 in diameter. The tube is partially im- mersed in sea water, and when an egg, dropped in the upper open end, settles to the top of the capillary, the water level in the tube is raised until the egg is pushed into the middle of the capillary. Impedance measurements are then made between electrodes placed in the tube and in the outside solution. The low fre- quency resistance of a 48m tube rose from 24,000 ohms, when filled with sea water, to 840,000 ohms with an unfertilized egg in place. This increase might be due to a membrane resistance of 20 ohm cm.’ but this value is no more than a lower limit since a layer of sea water 0.25 thick between the egg and the glass would produce the same result. The low frequency resistance for a fertilized egg was equivalent to a 4.2 space, which is larger than the membrane elevation. The observed low frequency capacities and the higher frequency data give average membrane capacities of 0.8 uf/cm.’ for the unfertilized, and 2.8 ufi/cm.’ for the fertilized eggs. These ‘results are in agreement with these obtained from sus- pensions and the technique may be used for several problems which are not other- wise possible. The effect of NaCl on potentials in Nitella. Samuel E. Hill. The normal action current in Nitella requires about 15 seconds for com- pletion, including recovery. After the cells have soaked for 30 minutes or more PRESENTED AT MARINE BIOLOGICAL LABORATORY 363 in 0.01 M NaCl the action currents may become very brief, lasting not more than 1 or 2 seconds. The form of the action curve changes, showing 1 peak instead of 2, and the amplitude is usually less. After 24 hours in 0.01 M NaCl the action curve tends to become normal again. , The cells show no signs of injury after 24 hours or more in 0.01 M NaCl. The coalescence of a plant cell with oil drops. M. J. Kopac. The young aplanospores of Valonia ventricosa are essentially naked proto- plasts and coalescence with oil drops readily occurs. As the aplanospores become older, the tendency to coalesce with oil drops becomes decreased. Coalescence with oils of a high interfacial tension against sea water (paraffin oil, tension ca. 40 dynes* cm.*) is inhibited in aplanospores over 3% hours old. This decreasing tendency to coalesce with oil drops is believed to signify the building up of ex- traneous coats by the protoplast. Ultimately the aplanospores become coated with a cellulose wall. In aplanospores about 1%4 hours old, coalescence with low tension oils (oleic acid in paraffin oil, tension ca. 3 dynes* cm.*) occurs rarely. A small drop of this oil may be placed in contact with an aplanospore without coalescence occurring. If a few seconds later a small drop of a higher tension oil (oleic acid, tension ca. 10 dynes* cm.*) is applied to the opposite side, coalescence between it and the aplanospore immediately takes place. From % second to several minutes later, the first drop snaps into the aplanospore. In these young aplanospores no cellu- lose cell wall has been formed. The inhibition of coalescence with a low tension oil may be due to the preliminary solidification at the cell surface prior to the formation of a cellulose wall. Coalescence with oleic acid apparently induces a peripheral disorganization at the cell surface which then permits the protoplast to coalesce with a low tension oil. This disorganization in the case of the aplanospore may actually be a disintegrative action at the cell surface. Addi- tional evidence for this point is shown by the release of chloroplasts from the protoplast following coalescence with the two drops. These investigations were started at the Tortugas Laboratory this summer and are being continued at the Marine Biological Laboratory. The influence of length, tension, and tone upon the birefringence of smooth muscles (Phascolosoma and Thyone). Ernst Fischer. The retractor muscles of Phascolosoma and Thyone respond to direct stimu- lation with a twitch-like contraction. After indirect stimulation involving the ganglion the quick contraction of both muscles is followed by a sustained tonic contraction. Besides this “contractile tone” both muscles show marked “ vis- cosoidal tone.” A muscle stretched by a load is constantly lengthening, and when the muscle is released later on, it shortens only to a small extent. By alternatingly loading and releasing a muscle with increasing weights, the bire- fringence of the muscle can be measured for the same muscle length under no tension and under a well-determined tension. When no tension is exerted, the birefringence of the muscle is about proportional to the square root of the length, as found by Bozler for a smooth snail muscle under comparable condi- tions. For the same muscle length under tension the birefringence is higher, the increase being proportional to the tension present,—a true “ photoelastic effect.” In consequence, when a muscle is stretched quickly, the birefringence increases at a steadily growing rate until suddenly the birefringence diminishes and the muscle tears through. Under isometric conditions the tonic sustained contraction increases the double refraction under all conditions, while for the twitch-like contraction, as shown 364 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS previously, the direction of the birefringence change is dependent on length and tension. This indicates that “normal contraction” and “tonic contraction” are fundamentally different processes. For both muscles investigated the same re- sults were obtained with the only difference that in Thyone the birefringence is merely 60 per cent of that in Phascolosoma. The mechanism of salt penetration in Amoeba—some micromanipulative data. Samuel A. Corson. Addition of inorganic acids (to a pH of 4.5 or lower) to K or Na salt solutions in which the amoebae were immersed prevented the decrease in proto- plasmic viscosity and the cessation of movement which occurred when Ameba proteus (Chaos diffluens) was immersed in neutral or alkaline solutions of these salts. The same acids also prevented the marked increase in protoplasmic vis- cosity produced by immersion in neutral or alkaline Ca salt solutions. Utilizing the Chambers micromanipulator and a new method which permits quantitative injections (the method consists essentially of enclosing the solution between a layer of heavy and one of light oil and measuring the quantity of solution drawn into the pipette by means of a calibrated ocular micrometer) it was shown that while KCl injections produced the same effects as in the immersion experiments (a fact observed previously by R. Chambers), K,SO, injections failed to liquefy the protoplasm though they did inhibit locomotion. These effects, just as the effects of CaCl, injections, were not influenced by acidification of the injected solutions (to a pH of 4.2-3.0). Since in the immersion experiments the K effect was the same irrespective of the anion used, these results support the previously suggested hypothesis that the plasma membrane of this amoeba is selectively permeable to cations and relatively impermeable to anions. The efficiency of monochromatic ultraviolet radiation in the activation of Arbacia eggs. Alexander Hollaender. During an investigation of the effects of ultra-violet radiation on the eggs of Arbacia punctulata, it was observed that when the eggs are exposed for as short a time as 1/10 of a second to the entire radiation of a water-cooled high pressure quartz capillary mercury vapor lamp, a large percentage of the eggs went through one or more cleavages without fertilization. Exclusion of the infra-red did not inhibit while exclusion of the radiation below 3,000A inhibited the effect. The eggs were then irradiated with measured quantities of monochromatic radiation of nine different wave-lengths from 2,260 to 3,650A. Special care was taken to develop a method which would make certain that not only each egg within the dish but each part of each egg received an equivalent amount of energy. This was done by rotating a small dish in which the eggs were suspended in 3 mm. of sea water and blowing an air current against the water surface. The eggs were removed after irradiation to a larger volume of sea water and kept at 24° and 10° C. Three types of controls were handled in each series of experiments: (1) unirradiated, unstirred eggs, (2) eggs stirred in the usual manner but protected against the radiation, (3) sea water irradiated with the wave-lengths most effective in producing activation to which unirradiated eggs were afterwards added. In none of these controls could any activation be recognized if the original eggs were in good condition. The energy was measured with a standardized vacuum thermopile-gal- vanometer set up and the incident energy per egg calculated taking into account the total energy entering the dish, the time of exposure, the diameter of the egg and the fact that the eggs were exposed interruptedly ignoring for the present the energy reflected and scattered by the eggs. PRESENTED AT MARINE BIOLOGICAL LABORATORY 365 Three to five hours after irradiation a high percentage of the eggs (up to 98 per cent) were found activated (one or more cleavages), if certain wave- lengths and energies per egg were used. The wave-length found most effective was 2,260A, and the efficiency decreased with the increase in wave-length, be- coming negligible around 2,500A; 0.13 to 0.25 erg of incident energy per organism at 2,260A produce the highest percentage of activation. A plot of the percentage cleaving against the energy applied for each effective wave-length shows a typical S-shaped curve, a definite plateau, and finally a decreasing rate of cleaving with further increase of energy. A plot of the efficiency of radiation against wave-lengths with its most pro- nounced maximum at 2,260A and its low efficiency at 2,600A suggests interesting comparisons with the ultra-violet inactivation spectra of urease and the virus of typical tobacco mosaic, and the absorption spectra of other protein-like sub- stances. Activation of centrifuged whole eggs of Arbacia and their fractions by monochromatic ultra-violet radiation. Ethel Browne Harvey and Alexander Hollaender. Whole Arbacia eggs which have been stratified and elongated by centrifugal force (10,000 x g for 3 minutes) are activated by the same ultra-violet radiation as the whole uncentrifuged eggs; full arc 1/10-5 seconds, monochromatic 2,350- 2,480A for 2-8 minutes. They behave in exactly the same way as when activated by other parthenogenetic agents such as hypertonic sea water but pass through only a few cleavages. White half-eggs (nucleate) are activated by the same radiation but are much more sensitive to slight variations from an optimum treatment, and fewer cleave. Red half-eggs (non-nucleate) are activated by the same radiation and also by a band of longer wave-length, 2,650-3,050A for 4-12 minutes, which does not affect the whole eggs and the white halves. Fertilization membranes are formed, some large asters occur, the egg often becomes aspherical and somewhat amoeboid. A notch frequently appears at the equator of the more spherical red halves, indi- cating the beginning of a cleavage plane, and this usually completes itself. Stages with 8-12 cells have been observed usually unequal in size. There occur later on some eggs filled with many small asters, a possible precursor to a blastula. Yolk quarters (non-nucleate) are activated in just the same way as the red halves and by the same wave-lengths and dosage, and some 8-12-celled stages have been observed. The pigment quarters (non-nucleate) show some evidence also of being activated, since they form an ectoplasmic layer and become some- what amoeboid. Ultra-violet radiation, therefore, acts upon the whole eggs and their frac- tions obtained by centrifugal force exactly like other parthenogenetic agents such as hypertonic sea water. Since the radiation affects the non-nucleate fractions as well as the nucleate, the action must be on the cytoplasm, but since many abnormal and irregular mitoses are observed in stained sections of later cleavages, the action must be also on the nucleus. The cytology of Arbacia punctulata actwated by monochromatic ultra- violet radiation. B.R. Nebel, Ethel Browne Harvey and Alexander Hollaender. Unfertilized eggs of Arbacia treated in the summer of 1936 with the full output of a high pressure mercury vapor lamp for a few seconds, showed activa- tion in all the eggs and fairly normal first cleavages following the normal nuclear changes in 80 per cent of the eggs, but the cleavage was delayed by two or more 366 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS hours. A very characteristic feature of late prophase and metaphase stages in later cleavages as shown in prepared sections is the presence of small rings or spheres among the chromatic material. Occasionally the spindles of more than two plates appear united at various angles. Large cytasters are observed. Regu- lar dicentric spindles, typically non-astral, occur but the most frequent type is a monaster surrounding the chromosomes, which as the division proceeds appears to form two polar half-asters. The eggs developed usually only to about an 4 to 8-cell stage. Unfertilized eggs treated in the summer of 1937 with measured intensities of monochromatic ultraviolet of 2,350 and 2,260A for 4 to 6 minutes show up to 98 per cent activation indicated by a fertilization membrane and ectoplasmic layer and after 3-4 hours cleavage. Nuclear divisions are monastral and irregular, normal dicentric asters not being formed. The first division is frequently of the restitu- tion type leading to diploidy without cell or nuclear division. The length of the spindle is diminished. During subsequent stages the products of successive chromatic multiplication may be separated. Thus irregular blastomeres may be formed containing no chromatin, approximately haploid, diploid or polyploid nuclei. The average activation does not produce more than 4 to 6 nuclei con- taining approximately variable multiples of the haploid chromosome number, distributed irregularly in 2 to 8 asymmetrical blastomeres. Unfertilized eggs treated with the same and larger total energy given as ultra- violet \ = 2,480, 2,650 and 3,050A showed no marked activation. Fertilized eggs treated with ultraviolet of the short wave-length (2,260 and 2,350A) soon after insemination showed slight inhibition. Fertilized eggs treated with ultraviolet of longer wave-length (2,650 and 3,050A) showed marked in- hibition. In both treatments occasional nuclear irregularities were observed. The relationship of sperm extracts to the fertilization reaction in Arbacia. John A. Frank. , A specific egg-agglutinin, previously reported, is present in the filtrate from boiled Arbacia sperm suspensions. This substance is present in the fat-free resi- due on extracting sperm suspensions with alcohol and ether. It is not found in the lipid extract. Eggs fertilized in sperm extracts show a marked drop in fertilizability when compared with eggs fertilized in sea water. Experiments were performed to de- termine whether this inhibition of fertilization is due to the action of sperm ex: tracts on the egg alone, spermatozoon alone, or on both gametes. Sperm suspensions exposed for varying lengths of time to sperm extracts ex- hibit a marked loss in fertilizing power. Sperm extracts thus block fertilization by a rapid direct action on sperm. The fertilizability of jellyless eggs exposed to sperm extracts and subse- quently fertilized with fresh sperm decreases markedly. Sperm extracts there- fore exert an inhibitory effect on the cortex. When sperm extract is added to egg water containing fertilizin, the mixture will not agglutinate sperm. Some substance in the sperm extract has inactivated fertilizin. Sperm extracts inactivate fertilizin in definite quantitative proportions. The capacity for fertilizin inactivation varies directly with the concentration of sperm extract. Thus sperm extracts contain a substance which resembles Lillie’s anti-fertilizin. On ageing, the fertilizing power of sperm suspensions is lost concurrently with the capacity of extracts of these suspensions to inactivate fertilizin and to ageglutinate eggs. Evidence at present indicates that sperm extracts contain substances related in some way to the fertilization reaction. PRESENTED AT MARINE BIOLOGICAL LABORATORY 367 Stimulation and nuclear breakdown in the Nereis egg. L. V. Heilbrunn and Karl M. Wilbur. Recent studies of stimulation in diverse types of protoplasmic systems have indicated that one of the primary effects is a breakdown of a calcium proteinate gel in the cell cortex and a release of free calcium into the cell interior. As yet these studies of stimulation have thrown no light as to the rdle the nucleus may play when a cell is stimulated to divide. In the egg of the worm Nereis, various types of stimulating agents cause a breakdown of the germinal vesicle. Thus, such an effect is produced by heat, ultraviolet radiation, and Roentgen rays, as well as by various chemicals. If our theory is correct, one of the initial steps in the series of processes that result eventually in nuclear breakdown is a calcium release from the cortex of the cell. Our experiments lend support to this theory. If Nereis eggs are exposed for 6 or more minutes to an isotonic citrate solution, subsequent treatment with ultraviolet radiation causes no nuclear breakdown; al- though on return to sea water-the eggs again show a typical response following irradiation. Similarly previous treatment with citrate prevents the nuclear break- down which otherwise occurs very beautifully when eggs are placed in isotonic solutions of sodium or potassium chloride. The sodium and potassium ions appear to be capable of provoking a release of calcium ions from the cell cortex, and these calcium ions induce changes which eventually lead to a breakdown of the nuclear membrane. The movement of the egg nucleus in relation to the sperm aster in Ly- techinus and Echinarachnius. Edward L. Chambers. The observations indicate first that the egg nucleus is moved to the center of the sperm aster by centripetal currents of cytoplasm. This is in conformity with the early observations of Conklin. He held that the approach and union of the two nuclei were determined by protoplasmic currents in the odplasm. The granules in the cytoplasm move along with the egg nucleus. The curvature of the path of the egg nucleus is caused by the continual change in direction of the cytoplasmic currents due to the progressive movement of the sperm aster towards the center. The increasing acceleration of the egg nucleus indicates the existence of more and more intense centripetal currents of cytoplasm as the pronucleus migrates into the aster. These observations indicate, second, that the aster is a jellied mass. This con- firms R. Chambers’ conclusion. The diminishing acceleration of the egg nucleus as it moves down the ever- narrowing cytoplasmic path extending from the margin to the center of the aster demonstrates a resistance to movement due to the presence of a jellied material around the path through the aster. The deformation (ellipsoidal) of the egg nucleus as it moves along this path demonstrates that the path is gradually tapering cone of fluid cytoplasm in the jellied mass of the aster. The physical state of the wall of the furrow in a dividing cell. Robert Chambers. Evidence is accumulating that the wall of the furrow is solid and constitutes the most solid part of the cortex of the cell. The advance of the furrow displaces the interior toward the poles which bulge because of the relative weakness of the polar cortex. No symmetrical arrangement of the surface seems to be essential since strands of cortical protoplasm may be dragged out either at the equatorial or polar regions 368 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS without affecting the cleavage. The strands at the equatorial region are much stiffer than those at the poles. In tissue cultures fibroblasts often retain extended strands especially at the poles while the furrow forms at the equator. Conditions within the cell interior need not affect the division once cleavage is under way. A dividing epithelial cell sometimes shows a rhythmic back-and-fortk flow of the internal contents through the narrowing stalk connecting the twe daughter cells. This ceases only when the constriction of the stalk has completed the division. In the sea-urchin egg undergoing division in calcium-free sea water, the contents of one blastomere, when torn, pour out while the furrow continues to deepen. The continued pinching-down of the furrow on the connecting stalk frequently rescues the other blastomere from disintegration. The thickness of the cortex of the advancing furrow and the force with which it advances has been determined by injecting an oil drop into the equatorial region. The oil drop tends to come to lie in the central region at the equator so that the advancing furrow closes down on it and constricts it in two. This occurs when the surface of the floor. of the furrow is some distance from the surface of the oil-drop. The advance of the wall of the furrow must be considered as a growth phenomenon—material being added progressively to the gelated cortex of the fur- row analogous to the apposition of material along the plane of division of a plant cell. AvucGust 27 Chromosome studies in sundew (Drosera). A. Orville Dahl. For purposes of comparison with chromosomes in certain members of the Saxifragaceae, cytological material of Drosera filiformis Raf., D. longifolia L., and D. rotundifolia L. has been locally collected. A preliminary examination of root-tip meristem in aceto-carmine demonstrates 20 chromosomes in D. filiformis which is consistent with Levine’s (Mem. N. Y. Bot. Gard., 6: 125-147, 1916) re- port of 10 chromosomes in pollen mother-cell material from Lakehurst, N. J. Rosenberg (Ber. der Deutschen Bot. Ges.. 21: 110-119, 1903) found 20 chromo- somes in cells of root-tips, stems, leaves, and flowers of D. rotundifolia collected in Germany, Norway, and Sweden. A comparative study of the early somatic meta- phase in D. filiformis from Mashpee, Mass. and D. rotundifolia from North Fal- mouth, Mass. shows that the chromosomes of the former are approximately 1.49 times longer and 1.38 times wider than those of the latter. The metaphase chromo- somes are of comparatively small size, those of D. rotundifolia being about 1.90 u in length while those of D. filiformis are 2.82 in length. A visibly four-parted structure, with the distance between the half-chromatids 0.3 » to 0.4 4, could be de- tected at mid-prophase, late prophase (at which time the nucleus has a diakinetic appearance), early metaphase, and late telophase. Mitosis in the giant amoeba, Chaos chaos Linnaeus. M. Catherine Hinchey. Chaos chaos is from 50 to 500 times as large as C. diffluens and has from 50 to over 1,000 discoid nuclei. The chromatin granules are arranged in a layer underneath the nuclear membrane in the living amoeba. When mitosis begins, the nuclei become spherical and the chromatin granules congregate in a thick equatorial plate. Then most of the granules seem to dis- appear, leaving a spherical nucleus, with chromatin granules distributed in a thin plate. What appear to be spindle fibers become visible at this stage. These fibers can be seen in the living amoeba under the micro-compressor. PRESENTED AT MARINE BIOLOGICAL LABORATORY 369 Following this stage, the nucleus shortens along the polar axis and becomes wider at the equator, so that a ladder of fibers is seen, in optical section, with the chromatin distributed along the mid-points of the ladder-rungs. The chromatin then separates into two plates which move along the spindle fibers toward the poles, until the plates are three times their diameter apart. At this stage fibers can be plainly seen in both stained and living material, extending be- tween the plates and from the plates to the poles. Protoplasmic streaming next moves the plates apart. The fibers between the plates become twisted and disappear, but the fibers going to the poles persist for some time longer. The chromatin plate becomes thinner, and so homogeneous that it is extremely difficult to follow in living material, but in fixed material, a con- centration of chromatin occurs along the inner edge of the plate. The daughter nuclei next become wider and-more granular. This process continues by gradual steps until the interphase stage is reached. Cytoplasmic division—usually into 3 daughters—occurs during the reorganiza- tion of the daughter nuclei into interphase nuclei. The striking features of the nuclear division are: 1. All the nuclei divide at the same time. 2. Practically every stage of mitosis can be seen in the living amoeba under the micro-compressor. 3. Although the total number of nuclei doubles during mitosis, cytoplasmic division usually results in three daughters. Some effects of oxygen on polarity in Tubularia crocea. James A. Miller. | A chamber was constriicted by means of which the two ends of Tubularia stems could be exposed to different agents or to different concentrations of the same agent. This consisted of a double chamber with a partition which separated the solutions but which had perforations through which the stems could be passed. Using this apparatus preliminary studies were made upon the effects of high and low oxygen tensions on polarity. By placing the stems in alternating orientations each experiment served as its own control. Oxygen determinations by the Winkler method were made in all but preliminary experiments. When oxygen was bubbled on one side of the partition and boiled sea water was placed on the other, hydranths developed only on the side with high oxygen. One half of these were distal and the other were proximal hydranths. Similar results were obtained when oxygen was bubbled on one side of the partition and standing sea water (with 4.1 to 5.0 cc. Oz per liter) was on the other. That these results were caused primarily by the oxygen differential and not by a possible accumulation of carbon-dioxide was demonstrated when nitrogen was bubbled on one side of the partition and oxygen on the other. Here again there was a re- versal of polarity in all stems with their proximal ends exposed to the oxygen. The importance of circulation of the medium to sessile forms such as Tubu- laria was illustrated by experiments in which 95 per cent of the stems developed proximal hydranths in running aerated water when the distal ends were exposed to standing water, while only 6.3 per cent developed proximal hydranths when the conditions were reversed. Oxygen determinations in two of these experiments - showed a difference of only 0.1 cc. per liter in each case. Some effects of strychnine on reconstitution of hydranth primordia in Tubularia crocea. Faith Stone Miller and James A. Miller. Miller (1937) found that pieces of planarians regenerating in strychnine showed no evidence of stimulation. Since in Tubularia the size and time of de- velopment of hydranth primordia can be measured, uals form was used to con- tinue the study of the effects of strychnine. 370 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The behavior of unoperated individuals placed in solutions of strychnine sulphate in sea water was undistinguishable from that of controls anesthetized by magnesium sulphate. In reconstitution experiments stems selected for uniform appearance and diameter were used. Ten-millimeter pieces were cut with the distal end five milli- meters from the base of the hydranth. Continuous exposure resulted in a de- crease in size of hydranth primordia and increase in time of development. Con- centrations used ranged from M/20,000 which showed very little effect to M/5,000 in which few stems survived. Temporary exposures to M/1,000 for periods of % to 8 hours showed similar but less consistent results. The frequency of the oc- currence of bipolar forms was decreased in strychnine and none developed in the higher concentrations. The results obtained with strychnine on Tubularia are similar to those with inhibitory agents and indicate that it produces a definite depression. By bubbling oxygen through the solutions it was possible to antagonize the strychnine effect. The life cycle of Moniezia expansa. Horace W. Stunkard. Anoplocephaline cestodes are common parasites of herbivorous animals and one species is recorded from man. They are worldwide in distribution, and the group has been intensively studied for more than fifty years. The final hosts harbor sexually mature worms in their intestines, eggs of the parasites are voided with the feces of the hosts, but what occurs in the interval before the cestode re- appears in the intestine of the primary host has remained quite unknown. Stunkard (1934) published results of experiments which demonstrated that the final hosts could not be infected with eggs of the parasite and that an inter- mediate host is necessary for the completion of the life cycle. Experiments have been continued, using species of Momiezia from sheep and Cittotaenia from rabbits. Various minute, terrestrial invertebrates, chiefly in- sects, have been used in attempts to discover the intermediate hosts of these ces- todes. In the spring of 1936, tyroglyphid mites were fed eggs of Momiezia and onchospheres were recovered from the body cavity three days later. The struc- ture, habits, and life history of these mites indicate that they would not be suit- able intermediate hosts of Momiezia. The oribatid mites, however, appeared to be admirably suited, and representatives of this family were employed. Galumna sp. are abundant in regions where Moniezia occurs and specimens of this mite were collected from areas in which there were no sheep. These mites were fed eggs of both Moniezia and Cittotaenia and onchospheres of both species were recovered from the body cavity. During the past year thousands of Galumna sp. have been fed eggs of Moniezia expansa and a series of developmental stages, from the onchosphere to the mature cysticercoid, have been recovered from them. A new method for studying the pH of the intercellular substance in the living mammal. Richard G. Abell and Eliot R. Clark. This method involves the installation of phenol red within a transparent moat chamber in the ear of a rabbit. Such a chamber contains a thin space, called the ‘bay, into which living tissue, continuous with the subcutaneous tissue of the ear, grows through two small entrance holes at the proximal end. The bay has a glass bottom and a mica top, and is only 504 to 100 deep. Consequently the arterioles, capillaries, and venules, and other constituents of the tissue within it can be seen clearly with the microscope. At its distal end the bay opens directly into a reser- voir, called the ‘moat.’ Following the introduction into the moat of a 0.4 per cent solution of phenol red, made isotonic with rabbit’s blood by the addition of NaCl, the indicator diffuses into the bay, and there colors diffusely the intercellular substance of the PRESENTED AT MARINE BIOLOGICAL LABORATORY 371 tissue for a distance of approximately 1 mm. proximal to the most advanced capil- laries. It is not concentrated by the cells, and it is not toxic. The color of the indicator in the intercellular substance of tissue with an active circulation is pink, the shade of pink matching that of a phosphate buffer at pH 7.2, to which phenol red has been added, seen with a microscope in the counting chamber of a hemocytometer under the same conditions of illumination as used for the tissue. When the circulation is cut off, by compressing the main artery of the ear, the color of the intercellular substance changes to the orange-yellow of a buffer at pH 6.8 within 10 to 15 minutes, indicating accumulation of acid metabolites. Within 1 to 2 minutes after the artery is released and the circulation once more becomes active, the color of the intercellular substance changes from orange-yellow back to pink. By means of the present method, the pH of the intercellular substance can be studied under a variety of experimental conditions. The behavior of living mammalian arterioles, capillaries, and venules when exposed to CO,. Richard G. Abell and Eliot R. Clark. The experiments to be described were performed in a transparent moat cham- ber in the ear of a rabbit. The behavior of the vessels was studied with the micro- scope, and changes in the pH of the intercellular substance detected by means of the indicator method presented above. When CO: is passed through the moat, the color of the phenol red in the inter- cellular substance of the tissue in the bay turns from pink (pH 7.2) to yellow (pH 6.8-6.6). No increase in stickiness of the endothelium occurs when the amount of CO2 employed is small. If the endothelium is sticky toward leukocytes before CO: perfusion is started, it reverts to a state in which the leukocytes roll freely along the vessel walls. No increase in the diameter of the arterioles, capillaries, or venules occurs as long as the circulation continues to pass through them. The arterioles in these experiments were not supplied by nerves. The color of the indicator in the intercellular substance of the proximal tissue turns from pink to yellow more rapidly when the circulation is sluggish than when it is rapid. Such color change is followed shortly by thickening and vacuolization of the endothelium of the arterioles. If CO: perfusion is stopped at this stage, the vacuoles disappear, and the endo- thelium resumes its normal thickness. If perfusion is continued, a marked in- crease in permeability of the arterioles, and also of the capillaries and venules, occurs. The plasma passes through the walls of the vessels, leaving within their lumina only the formed elements of the blood. During this process extensive crenation of erythrocytes occurs. In vessels containing concentrated cells, the flow of blood is blocked. No increase in diameter of the arterioles, capillaries, or venules occurs at the time of onset of plasma hemorrhage, but may take place shortly thereafter, indi- cating a softening of the endothelium. These changes are reversible if CO: per- fusion is stopped when plasma hemorrhage first occurs. The control of peripheral circulation in the mammal. Eliot R. Clark and Eleanor Linton Clark. It has been possible, in transparent chambers introduced in the rabbit’s ear, to watch the mode of formation and behavior of extra-endothelial cells, with the following results: Fibroblast-like cells from outside the endothelium become flattened out on the wall of the capillary at a very early stage in capillary formation—often during 372 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS sprout formation. If the vessel remains a capillary, there may be no increase in their number and they may remain permanently as sparsely distributed inert, oval cells. If the capillary becomes a portion of an artery the number of out- side cells increases rapidly—in part at least by mitotic division—the axis of the cells changes quickly from a longitudinal to a transverse position and muscle cells develop which show typical contractility in case they receive a nerve supply. If the capillary becomes a part of a venule or small vein, the number of adventitial cells increases slightly, their long axis remains longitudinal and they do not develop contractility. Persistent observations of the living vessels made under a great variety of conditions has corroborated earlier findings that, in the mammal, neither the endothelium nor the adventitial cells, as found on the capillary or venule, mani- fests active contractility. The control of the peripheral flow resides in the muscle cells of arteries, arterioles and certain of the veins—providing the muscle cells are under nerve control. While there may be changes in the caliber of capillaries and bulgings into the lumen of endothelial nuclear thickenings or of adventitial cells, all such changes are apparently passive, secondary to a variety of factors, chief of which are changes in internal pressure and rate of flow, produced by contraction or dilatation of supplying arteries or arterioles, and changes in outside pressure occasioned either by variations in the amount of fluid accumulation in the inter- vascular spaces or by the elasticity of the enclosing wall. The structure of the liver lobule. Louis Loeffler. The liver of the pig is the only one which shows clearly defined lobules. By that is meant that the lobules have a connective tissue membrane separating one lobule from another. The capillaries, also, of an individual lobule are separated from the capillaries of the adjacent lobules. All the other livers of mammals, reptiles and fishes, so far as has been investigated, show no separating membranes and also show anastomosing capillaries throughout the whole organ. Nevertheless, one is justified to speak of liver lobules, because the vessels, sit- uated in regular distance from one another, form figures similar to the pig liver lobules. It is shown, however, that there are no so-called sub-lobular veins, because such veins, next in size to the central veins within each lobule, function as central veins quite the same. Hepatic veins come to lie outside the lobules, not before. The hepatic veins reach a diameter half or much more than that of a liver lobule. A collecting lobe of about 6 or 8 lobules around a sub-lobular vein as usually found in the diagrams of anatomical textbooks does not exist or is quite arbitrary. The pig’s liver should be explained on the basis of a physio- logical liver cirrhosis. A preliminary note on the innervation of the swim-bladder of the sea- robin. John B. Gaylor and Ernst Scharrer. This communication deals with the peripheral innervation of the sea-robin as investigated by Dr. Gaylor at Woods Hole; a combined paper will be published later when Dr. Scharrer has worked out the central connections. The swim-bladder of the sea-robin consists of a two-lobed sac in the ab- dominal cavity. It possesses intrinsic skeletal muscle which subserves the func- tion of noise production and, in the interior of the cavity, a gas gland which is in the form of a “rete mirabile” covered with secretory epithelium. A branch of the vagus on either side affords a motor supply to the striated musculature. Free endings, knob endings and ring terminations have been observed. There is no apparent difference in the mode of termination in the swim-bladder muscle from PRESENTED AT MARINE BIOLOGICAL LABORATORY 373 that of the usual somatic musculature. Ganglia in the striated muscle are sparse; the disparity between the wealth of fiber and the number of ganglia argues a double innervation to the skeletal muscle—one direct from the vagal nucleus and one relayed through peripheral ganglion cells. The presence of muscle-spindle organs in the region where the muscle takes origin from the fibrous sac is suspected but not yet definitely established. The “rete mirabile” contains fine non-medullated fibers which enter along with the vessels and which appear to be sympathetic. Between the vessels there is a large number of multipolar ganglion cells which are presumably parasym- pathetic relays. The origin and development of the thyroid in Eleutherodactylus, an anuran with no tadpole stage. W. Gardner Lynn. The Jamaican tree-toad Eleutherodactylus nubicola lays its eggs on land and the young hatch after about twenty-four days with a definitive body form. During the embryonic development some of the larval characters which are usually found in frogs appear very transiently but others, such as the formation of external and internal gills, are entirely lacking. A study has been made of the thyroid in embryos preserved at twenty-four-hour intervals throughout the period of development. The thyroid takes its origin from the pharynx at the sixth or seventh day and the cells exhibit signs of secretory activity almost as soon as the thyroid anlage is definitely recognizable. Throughout the succeeeding stages there is a steady increase in the amount of stored colloid. Intracellular vacuoles are abundant in the follicle cells at all stages. Vacuolation of the colloid mass is most striking during a period of about four days extending from the tenth to sixth day before hatching. There is no evidence of a sudden release of any large amount of colloid into the blood stream at any time during the embryonic history. The indication is, rather, that a regular release occurs even from the early stages. This would agree with the regular course of bodily differentiation and the absence of any striking metamorphic pattern. However, certain of the unusual features of development in this frog, such as the absence of external gills, cannot be attributed to precocious thyroid functioning. Thus it appears that while the evolutionary changes which have brought about the atypical life history of the species are some of them changes in the development and functioning of the gland complex, still others are changes which cannot be attributed to hor- monal influence but are direct alterations in the developmental potentialities of the tissues themselves. Some effects of mammalian follicle-stimulating and luteimizing hormones in adult female urodeles* Virginia Mayo. During October adult salamanders (Triturus viridescens) were subjected to preliminary tests on their ovulatory response to a single intraperitoneal injection daily of the following mammalian pituitary extracts made according to H. L. Fevold’s method: (a) physiologically pure follicle-stimulating hormone (F.S.H.) ; (b) physiologically pure luteinizing hormone (L.H.); (c) F.S.H. with a trace of L.H.; (d) an unfractionated extract containing both F.S.H. and L.H. Each injection contained 1/20 gram-equivalent acetone-dried sheep pituitary powder. The results indicated that fractions containing L.H. induced egg-laying whereas F.S.H. alone was almost entirely ineffective. Injections given both normal and hypophysectomized newts in November—December, May-June, and June—July- August corroborated the October findings. *This work was begun at the Biological Laboratories, Harvard University, and continued at the Marine Biological Laboratory. 374 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS Counts of eggs released by groups of 10 animals treated during the breeding season for 40 days gave the following: hypophysectomized L.H.-injected animals ovulated 801 eggs; normal L.H.-injected, 582; hypophysectomized F.S.H.-in- jected, 18; and normal F.S.H.-injected, 25. At the onset of the June-July—August series the newts’ ovaries were almost completely emptied of large eggs. By early August the ovaries of F.S.H. and F.S.H. + L.H.-treated animals were filled with yolked eggs, while those of L.H.- treated individuals were only slightly stimulated. Average ovarian weights of 7-14 animals were: untreated controls, 28 mg.; hypophysectomized + L.H., 69 mg.; normals + L.H., 88 mg.; hypophysectomized + F.S.H., 119 mg.; normals + F.S.H., 134 mg.; hypophysectomized + F.S.H. and L.H., 150 mg.; and normals + F.S.H. and L.H., 173 mg. On the basis of these results it seems that F.S.H. brings about a striking enlargement of the ovary while the L.H. is primarily responsible for egg release The relation of melanophore responses to vascular disturbances. G. H. Parker. It is difficult to cut nerves in experiments on melanophore control without at the same time cutting blood-vessels or at least introducing vasomotor disturb- ances. Does the stimulation of melanophore nerves thus brought about excite vasomotor changes which in turn excite responses in melanophores or do the nerves act directly on the melanophores? In the killifish FPundulus and in the catfish Ameiurus the melanophores have a double innervation, concentrating nerve-fibers inducing a blanching of the fish through a concentration of pigment in its melanophores and dispersing fibers darkening the fish through a dispersion of this pigment. In the dogfish Mustelus there are only concentrating fibers, the dispersion of the pigment being accomplished through a pituitary neurohumor in the blood. In Fundulus and Ameiurus when melanophore nerves are cut the dispersing nerve-fibers are stimulated and dark areas or bands result. In these two fishes and in Mustelus when the melanophore nerves are stimulated electrically the concentrating fibers are excited and the fishes blanch locally. One way of finding an answer to: the question under consideration is to ascertain whether these responses will occur or not in the absence of an active circulation of blood. To this end the ventral aorta of a given fish was ligated just anterior to the heart and the circulation of blood thus brought to a complete standstill. The melanophore nerves of such a fish were then subjected to electric stimulation or were cut. In all such instances there was either local blanching or darkening according to the stimulus. These responses, though less marked than in normal fishes, showed unquestionably that the complete loss of circulation was not accompanied by a loss of the power of color response and that therefore vasomotor or other vascular changes could not form any essential part in the chain of events between nerve and melanophore. Some effects of chloroform on the respiratory systems of yeast. E. P. Hiatt and J. K. W. Ferguson. The rate of reduction of methylene blue in suspensions of yeast is increased by small amounts of chloroform (0.05 per cent by weight). With larger amounts of chloroform, acceleration up to 40 times was obtained. No stage of inhibition was reached with fresh yeast. Dried yeast and yeast extracts, which have a much faster rate of reduction than equivalent amounts of fresh yeast, were retarded by chloroform. The same effects can be demonstrated with other oxidation-re- duction indicators, e.g., pyocyanin, thionin, and anthraquinone. PRESENTED AT MARINE BIOLOGICAL LABORATORY 375 When chloroform was exerting its maximal acceleration, the cytochrome bands showed a characteristic change. Only the C band was visible and it could not be made to disappear on shaking with air. Subsequently, different amounts of chloroform were found to affect the cytochrome spectrum differently. With concentrations up to 0.1 per cent by weight, the time for the appearance of all three bands (rate of reduction), was shortened. At about 0.3 per cent the A band disappeared, the B and C bands appearing slowly and remaining fixed. At about 0.5 per cent the B band disappeared. Similar but less marked effects were obtained with ether and 95 per cent ethyl alcohol. It is interesting to note that carbon tetrachloride has little effect. The oxygen consumption of the chloroform-treated yeast, as determined by the Warburg technique, was also greatly increased;—up to 20 times. Maximal acceleration was obtained in a medium of phosphate buffer at pH 6.6. Less effect was noted at pH 7.0 and little or none at 8.5. A similar acceleration was observed with fertilized and unfertilized Arbacia eggs. In view of the prevalent idea that narcotics act by depressing cellular respira- tion, it seems significant that these accelerating effects on respiration were obtained with concentrations of the same order of magnitude as are effective in producing general anesthesia. The oxygen consumption of activated and fertilized eggs of Chaetopterus. Jean Brachet. Unfertilized Chaetopterus eggs undergo activation when they are treated with 5 per cent isotonic KCl in sea water; maturation is followed by a series of monasterian cycles leading to the formation of unicellular larvae resembling gastrulae and trochophores (F. R. Lillie’s differentiation without cleavage). The oxygen consumption of these activated eggs has been compared with the respiration of unfertilized and of fertilized eggs during 7 hours (Warburg’s method). Activation is followed by a considerable drop in the oxygen consump- tion (49 per cent); and fertilization has exactly the same effect, as was shown first by Whitaker. The O, uptake increases then, but at a much slower rate in the activated eggs than in the fertilized ones: while these resume their initial respiratory rate after 3144 hours, the activated eggs need 6 hours to reach that level. The respiration of the unfertilized eggs remains constant for 7 hours. Control experiments showed that isotonic KCl in the concentration of 5 per cent used has no significant effect on the metabolism of Chaetopterus eggs: the slope of the curve is not changed if the KCl treated eggs have been re- peatedly washed or when KCl is added to the fertilized eggs. The reduced metabolic activity of the activated eggs must thus be linked to either their slower development or to the fact that they remain unicellular. Influence of respiratory inhibitors on stimulation of metabolism by nitro and halo phenols. M. EK. Krahl, Anna K. Keltch and G. H. A. Clowes. At-the 1934 meeting, experiments reported from this laboratory showed that the respiratory stimulation produced by 4,6-dinitro-o-cresol in fertilized eggs of Arbacia punctulata was progressively inhibited and could be completely abolished by increasing concentrations of potassium cyanide and that the division- blocking effects of the two reagents were additive. During the past three seasons these experiments with cyanide and 4,6-dinitro-o-cresol have been extended and similar experiments made with other respiratory inhibitors. The following concentrations of inhibitors have been found, with eggs in sea water at pH 8, give a suppression of normal respiration which is just measur- able (ie. 5 to 20 per cent): CO, 94CO:60,; ‘Amytal’ (Iso-amyl Ethyl Bar- 376 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS bituric Acid, Lilly), 2 X 10° M; malonic acid, 10° M; and, at pH 6, iodoacetic acid, 10* M. At these same respective concentrations of the inhibitors there is a progressive decrease in the extent to which the various inhibitors suppress the respiration stimulus by 4,6-dinitro-o-cresol, the first two members of the series giving almost complete, and the last two members almost negligible suppression of the stimulated respiration. With the exception of phenyl urethane and malonic acid, with which such experiments were not made, all inhibitors used gave complete suppression of stimulation by 4,6-dinitro-o-cresol when tried in sufficiently high concentrations. With partial inhibition by CO or KCN, the optimum stimulation of the residual respiration was produced by concentrations of 4,6-dinitro-o-cresol larger than those required in the absence of CO or KCN. With partial inhibition by the other inhibitors, optimum stimulation of the re- sidual respiration was produced by concentrations of 4,6-dinitro-o-cresol equal to or less than those required in the absence of inhibitor. In this series of experiments, the concentration of 4,6-dinitro-o-cresol re- quired, in the absence of inhibitors, to give 90 to 100 per cent block to division was 8 X 10° M. The concentrations of 4,6-dinitro-o-cresol required in the pres- ence of the concentrations of inhibitors mentioned above, which alone gave little or no block to division, were 4 X 10° M in CO; 8 X 10° M in low oxygen ten- sion; less than 10° M in KCN; 8 X 10° M in malonic acid. Substituted phenols as inhibitants of the fertilization of Arbacia and of ciliary movement of Arenicola larvae. G. H. A. Clowes, M. E. Krahl and Anna K. Keltch. It has already been demonstrated for a considerable series of nitro and halo phenols that the point of concentration required for maximum stimulation of respiration corresponds approximately with the point at which cell division is blocked in fertilized sea urchin eggs. In an attempt to throw further light on the nature of the mechanism involved, the concentrations were determined at which certain representative nitro and halo phenols blocked the fertilization of Arbacia eggs by sperm and anesthetized Arenicola larvae. The block to fertilization and anesthesia of larvae occurred at about the same concentration for each individual compound, but these concentrations differed in certain cases very greatly from the concentrations at which the respiration peak and cell division block occurred. In the case of 2,4-dinitrophenol and 4,6-dinitro-o-cresol, having pK values of 4.1 and 4.4 respectively, the ratios of concentration required for anesthesia to that required for internal cell division block were found to be 137:1 and 228: 1. That for 2,4,5-trichlorophenol, having a pK of 6.9, was found to be 19:1, whilst ‘Concentrations (moles per liter X 10°) of substituted phenols required to inhibit various physiological functions. pH 8.0. eal meciieeten IV : Compound pK Moreen agar Breton I a TV Arenicola Treated Arbacia Sperm 2,4-Dinitrophenol............... 4.1 205 410 3.0 137:1 4,6-Dinitro-o-cresol............. 4.4 205 205 0.9 228:1 2,4,5-Trichlorophenol...........| 6.9 13 26 1.4 191 o-Nitrophenolwnrn 42.4 7.2. | Noeffect | No effect | No effect — 2,4-Dichlorophenol.............. dedi 51 26 26.0 esi m-Nitrophenol.......:.....:..: 8.3 205 205 51.0 4:1 PRESENTED AT MARINE BIOLOGICAL LABORATORY 377 those for 2,4-dichlorophenol, having a pK of 7.7 and m-nitrophenol, having a pK of 8.3, were found to be 1:1 and 4:1. It is particularly interesting to note that orthonitrophenol, which although proved to enter the cell, had no effect on cell division or oxidation, had also no effect on the fertilization process or on ciliary movement of Arenicola. It appears advisable to refrain from speculation re- garding these results until further data are available. Stimulation of the rate of cell division of Arbacia eggs by carcinogenic hydrocarbons. Anna K. Keltch, M. E. Krahl and G. H. A. Clowes. As a part of an investigation into the mechanism by which certain polycyclic hydrocarbons produce cancer, a study has been made, during the seasons of 1935 and 1936, of the effects produced by three carcinogenic and two closely related non-carcinogenic hydrocarbons on cell division of fertilized eggs of Arbacia punctulata, using each hydrocarbon in the form of its water-soluble choleic acid, these being addition compounds of the hydrocarbon with desoxycholic acid. Unfertilized eggs were exposed to varying concentrations of each choleic acid in sea water solution for varying periods of time. They were then fertilized, and left in the same respective solutions and at the same respective temperatures during the periods of pretreatment, fertilization and division, with the single ex- ception that at 5° C., the eggs were raised to 15° for approximately five minutes in order to allow fertilization to take place and then returned immediately to 5° C., the controls in every case being subjected to treatment identical with that given the experimental material. Typical representative results obtained with a five-hour pretreatment are presented in the accompanying table. In these data it is desired to emphasize— (a) that the choleic acids of the three hydrocarbons which produce cancer in mice also produce a shortening of division time and that the choleic acids of the two closely related hydrocarbons which produce no cancer in mice do not pro- duce a shortening of division time; (b) that, with optimum concentration of 6-methyl cholanthrene choleic acid, there is a progressively smaller relative de- crease in the division time as the temperature is raised from 5° to 15° C. Minutes to 50 Per Cent First Cleavage Tem- Carcinogenic Choleic Acid pera- eae tap ah See Activity ture Minimum in Mice Control with Hydrocarbon © Ce 6-methyl cholanthrene........... 5 1717 1482 Positive 6-methyl cholanthrene........... 10 235 221 6-methyl cholanthrene........... 15 109 100 10-methyl-1,2-benzanthracene..... 15 122 116 Positive 1,2,5,6-dibenzanthracene......... 15 126 123 Positive phenanthrene ec a. 2. see oot sees 15 LUST 118 Negative Fluoranthene................... 15 115 115 Negative The molecular species concerned in the action of substituted phenols on marine eggs. Albert Tyler and N. H. Horowitz. In a recently published article (Tyler and Horowitz, 1937) the view was expressed that the substituted phenols penetrate as the undissociated molecule, 378 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS but exert their respiratory stimulating and reversible block to cleavage effects as the anion. The evidence for this rests on the fact that with any one of these compounds the concentration required for maximum effect varies with the pH of the solution, but the calculated concentration of the undissociated form present is the same at all pH’s. This might mean that the undissociated form is the active species. However, when the various compounds are compared, the con- centration of undissociated form at maximum effect shows enormous differences. On the other hand, when the comparison is made on the basis of calculated con- centrations of the anion present inside the cell the various substances give values of the same order of magnitude. The compounds investigated include the three mononitrophenols, 2,4- and 2,6-dinitrophenol, 2,4,6-trinitrophenol, the three mono- chlorophenols, 2,4-, 2,5-, 2,6- and 3,5-dichlorophenol, 2,4,6-trichlorophenol and 2,6-dichloro-4-nitrophenol. Four of these; namely, o-nitrophenol, trinitrophenol, o-chlorophenol, and 2,6-dichlorophenol, show large deviations, but these substances are actually inactive or only slightly active in stimulating respiration. Extension of the length of fertilizable life to more than twice the control is also obtained with dinitrophenol, confirming the findings reported by Clowes and Krahl. This prolongation occurs at high concentrations giving no respira- tory stimulation. Determinations of the temperature coefficient of the respiration of unfertilized eggs shows it to be the same in dinitrophenol as in sea water. Ovoverdin, a pigment chemically related to visual purple. Kurt G. Stern and Kurt Salomon. The eggs of the lobster (Homarus americanus) owe their green color to a pigment belonging to the class of carotenoid-proteins. According to G. Wald? visual purple is another member of this widely distributed group of chromo- proteids. The carotenoid contained in the lobster egg pigment is astacin which is esterified with an as yet unidentified organic acid.” * This “ovoester” is in turn linked up with a protein of albuminoid character. The name ovoverdin is pro- posed for the native pigment complex. Ovoverdin may be obtained in solution by grinding the eggs with sand and extracting them with distilled water. Treatment with an equal volume of sat- urated ammonium sulfate removes oil globules containing carotine and small amounts of globulins. The solutions may be further purified by repeated pre- cipitation of ovoverdin in saturated ammonium sulfate solution or by dialysis at low temperature. Ovoverdin has two absorption bands in the visible, at 6,400 and 4,700 A.; in addition it shows the typical protein absorption in the ultraviolet. The molecular weight, according to the rate of sedimentation in the ultracentrifuge as measured by Dr. R. W. G. Wyckoff, is of the order of 300,000. The isoelectric point is at pH 6.7. Organic solvents and weak acids liberate the orange red carotenoid by virtue of denaturation of the protein carrier. When solutions of ovoverdin or the whole lobster eggs are rapidly brought to 65 to 70° the color turns from grass green to bright orange red. The red form shows increased light absorption at 4,700 A and greatly diminished absorption at 6,400 A as compared with the green form. When the heated material is rapidly cooled, the green color returns. This reversible thermal dissociation is different from an irreversible dissociation which takes place upon longer exposure to these temperatures or upon raising the temperature to the vicinity of the boiling point. At the latter point coagulation *Wald, G., 1935-36. Jour. Gen. Physiol., 19: 351. * Kuhn, R., and E. Lederer, 1933. Ber. Deutsch. Chem. Ges., 66: 488. *Karrer, P., L. Loewe, and H. Huebner, 1935. Helv. Chim. Acta, 18: 96. PRESENTED AT MARINE BIOLOGICAL LABORATORY 379 of the protein occurs. The sequence of events and a photoelectric study of the color-temperature-time curves indicates a lower energy requirement of the first, reversible stage of dissociation as compared with the later, irreversible stages. It has been suggested** that the bleaching of visual purple by light is a disruption of the purple carotenoid-protein complex; the orange carotenoid, retinene, is liberated and the protein is denatured. This assumption, however, is open to the objection that the energy content of the effective wave-lengths of light is smaller than the energy required for inactivation of the visual purple complex. The present observations would suggest that the reversible bleaching of the retinal pigment does not involve a denaturation of the protein component but is rather of the type of the thermal phenomenon here observed. This hypothesis appears to receive support from the fact that the energy levels at which the latter occurs are lower than those at which protein denaturation takes place and secondly that the process observed here is rapidly reversible whereas protein renaturation in general is a time-consuming process and therefore not well suited for the regeneration requirements of visual purple during vision. The increase of CO, and decalcification in certain pelecypods. Louis- Paul Dugal and Laurence Irving. In three forms of pelecypods, Venus mercenaria, Ostrea sp. and Elleptio complanatus, the mantle cavity fluid gains CO, when they are kept out of water. The change is from 5 or 8 (in fresh ones) to 90 ml. per 100 ml. of fluid (for individuals kept out of water about 5 or 6 days). When the mollusks begin to die, the total CO, decreases. Return to water before death restores the CO2 to normal. The accumulation of CO, probably results from a disturbance of respira- tion. In Venus, which was most carefully studied, the total CO, of the M.C.F. increases rapidly. The pH decreases only from 7.4 to 7.2 and the Poo, increases only from 3 to a maximum of 25 mm. Hg, so that it is evident that the buffering capacity increases. The shell is eroded during these changes and a few analyses showed that the mantle cavity fluid gained Ca, so that it is easy to guess that the buffering is provided by the solution of CaCO; from the shell. The shell erosion is localized in the central inner part; the mantle tissue and no others, gain CO.. This indicates that the buffering is effected by the activity of a special tissue. The effect of pH and ionic strength on the activity of carbonic anhydrase. J. K. W. Ferguson and E. C. Black. The manometric method of Meldrum and Roughton for determining the ac- tivity of carbonic anhydrase by following the rate of evolution of CO2 from a mixture of phosphate and bicarbonate solutions, has been favored, because of its simplicity, for use in physiological and pathological studies. The reaction used in this method takes place in a medium of changing ionic strength and pH. As yet no adequate analysis of the effect of these variables on the activity of the enzyme has appeared. In this preliminary study the COz output method has been used and con- sequently the variation of pH was limited to the range from pH 5.7 to pH 8.0. The pH and ionic strength taken as corresponding to a certain activity was the mean pH and ionic strength of the range traversed by the stage of the reaction used in calculating the activity (usually from the beginning to two-thirds com- pletion). * Mirsky, A. E., 1936. Proc. Nat. Acad. Sci. (Washington), 22: 147. 380 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS Between pH 6.0 and 8.0 an eight-fold increase in enzymic activity was found, the activity increasing with pH. Between ionic strengths of 0.5 and 0.1 at constant pH a 2.5-fold increase in enzymic activity was found, the activity increasing as the ionic strength decreased. The distribution of carbon dioxide in dogfish blood. J. K. W. Fergu- son, S. M. Horvath, and J. R. Pappenheimer. After unsuccessful attempts to apply to fish bloods the chemical and kinetic methods of estimating carbhemoglobin, evidence concerning the state of CO, in erythrocytes of fish was sought by studying the distribution of COs between red cells and plasma. In this study smooth dogfish (Mustelus canis) and a few speci- mens of spiny dogfish (Squalus acanthius) were used. Defibrinated or heparinized blood was used exclusively after it was found that both oxalate and fluoride grossly affected the distribution of electrolytes between cells and plasma. The bloods were found to fall into two groups as regards total COz2 capacity. During June and early July the CO: capacities at a CO2 pressure of 40 mm. Hg. were on the average twice as great as in the bloods obtained during late July and August. In the early part of the season the curves were steeper, indicating greater buffer power. This difference could not be attributed to differences in hemoglobin content. The degree of oxygenation has no effect on the CO, capacity of dogfish blood. Usually the concentration of CO2 in the red cells (per units H2O) is greater than in the corresponding plasma. This is different from the situation in mam- malian blood and inconsistent with the view that the bulk of the CO: in the cells is in the form of active bicarbonate ions. When chloride distributions were studied the contrast was striking. The ratio of intracellular to extracellular con- centration of chloride, (rci), is always much smaller than the similar ratio for COs, (rco,). In mammalian blood rco, = € 1.25 Xrci. In these bloods rco, = 1.7 to 3 Xrce. If the intracellular CO. which is in excess of the amount to be ex- pected from the chloride ratios is assumed to be “ non-bicarbonate”’ COs, this “non-bicarbonate”” CO2 is found to comprise about two-thirds of the CO, in the red cell or about one-sixth of the CO, in the whole blood. The influence of certain alcohols on the permeability of the erythrocyte. M. H. Jacobs and A. K. Parpart. Low and moderate concentrations of the so-called indifferent narcotics have frequently been found to decrease cell permeability. Anselmino and Hoenig (Pfliigers. Arch., 225: 56, 1930) have reported such an effect in the case of the penetration of human erythrocytes by several non-electrolytes including glycerol. The present study, involving the erythrocytes of a number of species of mammals and of some other vertebrates shows somewhat more complicated conditions. Thus, while n-butyl alcohol in concentrations from 0.0156 M to 0.25 M may greatly de- crease the permeability of the erythrocytes, not only of man but also of the rat, rabbit, guinea pig, groundhog and several birds, the opposite effect is obtained with the erythrocytes of the ox, sheep, pig, horse, dog, cat and several reptiles and fishes. In general, these two groups of species are the same as those already distinguished by other properties of their erythrocytes (Jacobs, Glassman and Parpart, Jour. Cell. Comp. Physiol., 7: 197, 1935). In several cases involving de- creased permeability, the order of effectiveness of a series of alcohols is: methyl< ethyl guinea pig>rabbit >beef > pig, cat. In view of the slowing effect that butyl alcohol, ammonium salicylate and ammonium benzoate have on hemolysis in ammonium chloride it appeared of in- terest to study the effect of these substances on the exchange of Cl— for SO. In all of the species studied the above substances were found to slow the rate of this ionic exchange. Control experiments indicated that these substances, in the con- centrations used, had little or no effect on the volume of erythrocytes suspended in isotonic NaCl. PAPERS READ BY TITLE Oxygen as a controlling factor in the regeneration of Tubularia. L. G. Barth. Preliminary experiments in which the distal end of the stem of Tubularia was inserted into a glass tube gave complete inhibition of regeneration of this end with a marked increase in rate at the opposite end. This indicated that low oxygen inhibited regeneration. Following this the oxygen tension of sea water was varied and rate of regeneration was calculated by dividing the length of the primordia in micra by the number of hours necessary for its formation. Results of nine complete ex- periments agree in showing that from 4.0 to 21.6 cc. of oxygen per liter the rate of regeneration increased linearly with the logarithm of the oxygen tension. Below 3.0 cc. of oxygen per liter the rate of regeneration drops off to 0 at .35 cc. and the shape of the curve depends somewhat on the way in which the experiment is carried out. The lower limit for regeneration is between .35 cc. and 1 cc. of oxygen per liter. At .35 cc. complete inhibition results, which is reversible when stems are returned to high oxygen. The results are interpreted as showing that regeneration of JTubularia is closely dependent on the amount of oxygen which the tissues receive. It is suggested that the perisarc of Tubularia is relatively impermable to oxygen and that the stimulus for regeneration is the admission of oxygen due to cutting of the perisarc. The effects of different drugs on the melanophores of Fundulus hetero- clitus. Sinisha B. Bogdanovitch. In the following experiments the effects on melanophores of different drugs alone and in combination were studied. Isolated scales of Fundulus heteroclitus 382 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS were used and the drugs studied were atropine, pilocarpine, physostigmine, epine- phrine, acetylcholine, mecholyl and deuterium oxide. The technical procedures were the same as those described in a previous paper. A summary of the results obtained follows: Atropine: Expands melanophores and also does so after they had been pre- viously contracted by epinephrine, deuterium oxide, acetylcholine or mecholyl. Ergotized melanophores were also expanded by atropine. Epinephrine: Contracts melanophores and also does so after they had been previously expanded by pilocarpine, physostigmine or atropine. Ergotized melanophores are, as is well known, expanded by epinephrine. Physostigmine: Expands melanophores and also does so after they had been previously contracted by epinephrine, acetylcholine, or mecholyl. Well ergotized melanophores are not expanded by physostigmine. Pilocarpine behaves as physostigmine, but appears to be slightly more ef- fective. Acetylcholine contracts melanophores and also does so with melanophores previously expanded by physostigmine or pilocarpine. However, with melanophores previously expanded by atropine acetylcholine produces either very slight con- tractions or none at all. Ergotized melanophores are contracted by acetylcholine. Ergotized melanophores are, of course, already contracted, so that additional con- traction is sometimes quite slight. Mecholyl behaves as acetylcholine but the effects appear to be more lasting. Deuterium oxide contracts melanophores after they had been previously ex- panded by atropine, pilocarpine or physostigmine. Other effects of deuterium oxide on melanophores were described in a previous paper. Scales removed from fish and kept in balanced solutions for two hours or more showed an initial expansion in epinephrine followed by contraction. Such a result, while possibly due to a pathological change, may explain contradictory results of previous investigators and offer a clue as to the nature of the inner- vation of melanophores. Further investigations on the effect of tissue on different drugs. Sinisha B. Bogdanovitch. In a former paper it was established that epinephrine and acetylcholine are destroyed by tissue (in this case scales of Fundulus heteroclitus), but that deu- terium oxide protects these substances from such destruction. Using the same technique as in former experiments atropine, pilocarpine, physostigmine and mecholyl (acetyl 6-methylcholine) were similarly investigated with the following results. Solutions of atropine (sulphate) which had previously expanded melano- phores, no longer did so or at most only slightly, after treatment with fish scales for 24 to 48 hours. However, atropine solutions to which scales and deuterium oxide were added for the same length of time, showed no change in expanding melanophores. Physostigmine (sulphate) and pilocarpine (chloride) solutions were not affected by fish scales even after 48 hours. Solutions of mecholyl (chloride) were also practically unchanged by similar treatment. It thus appears that atropine is destroyed by fish scales as was epinephrine and acetylcholine, but that physostigmine, pilocarpine and mecholyl were not af- fected. In the former paper it was suggested that the destruction of epinephrine and acetylcholine by fish scales may be due to the action of enzymes such as acetylcholinase and some oxidase. In support of this view I observed that scales which had been heated at 100° C. for 10 minutes did not destroy atropine even after 48 hours. Similar experiments on epinephrine and acetylcholine gave the same results. This thermolability of the “destructive principle” in fish scales for these three substances is consistent with the enzyme hypothesis. PRESENTED AT MARINE BIOLOGICAL LABORATORY 383 A quantitative study of the staiming of marine eggs by neutral red. Barry Commoner. Unfertilized eggs of Chaetopterus and Nereis limbata were stained in sea water containing various concentrations of neutral red at 14.0°, 15.5°, 17.0°, 18.4°, and 19.5° . The pH was maintained at 7.6. Eggs were suspended in sea water acidified to pH 7.6. The density of eggs per ml. of suspension was determined and a known quantity of neutral red added. The same certified stain of 61 per cent dye content was used throughout. Con- centrations are expressed as mgm. of this preparation. Samples were removed periodically from the suspension and centrifuged to concentrate the eggs. The eggs were washed in sea water to remove unbound stain, again concentrated and shaken in 10 ml. of a neutral red extractant. (One volume of N/1 HCl and 9 volumes of 95 per cent ethyl alcohol.) The neutral red content of the egg sample was then determined by comparing (in a Duboscq colorimeter) the intensity of the extract with a standard neutral red solution. The neutral red content per unit volume of eggs was then calculated. (Maximum error; 3 per cent.) Curves were plotted indicating the neutral red content of the eggs after various periods of time in the stain. Curves of similar shape were obtained for Chaetopterus and Nereis eggs. The maximum staining rate occurs during the first 20 minutes. Thereafter, the velocity of the process decreases until after 60-80 minutes an equilibrium is reached and no further change in neutral red con- tent occurs. At concentrations below .02 mgm./ml. the staining rate is proportional to the concentration of neutral red in the suspension. The stain content of the eggs at equilibrium is also proportional to the concentration of neutral red. The initial staining velocity and the quantity of neutral red bound at equilibrium increase with temperature. Within the temperatures noted (con- centration: .01 mgm./ml.) the Q,;, obtained from the initial velocities of the staining curves of Chaetopterus eggs was 4.4. The adaptation of Paramecium to sea water. John A. Frisch, S.J. In cultures of hay and wheat infusions, in which the concentration of sea water was gradually increased by evaporation, many individuals survived and divided until a concentration between 35 and 40 per cent was reached; all died before a concentration of 45 per cent was reached. Daily observations of a culture in which the concentration of sea water increased to 40 per cent in 20 days showed that the average rate of pulsation and the average rate of feeding were lower day for day than in fresh water cultures; that both rates varied from day to day as in fresh water cultures, and that the rate of pulsation varied with the rate of feeding, increasing or decreasing from day to day, as the rate of feeding increased or decreased, just as in fresh water cultures. As the salt con- centration increased the animals decreased in length and volume and became emaciated. Addition of nutrient medium or of bacteria to ‘the cultures always resulted in an increase in the number and in the volume of the individuals, and in the rate of feeding and the rate of pulsation, except when the concentration of sea water had reached 40 per cent. The data indicate that the decrease in the rate of pulsation is not due to the increase in osmotic pressure of the medium, but to the decrease in the rate of feeding; that the decrease in the rate of feeding is due to a shortage of bacteria in the higher salt concentrations; that death is due to an increase in the viscosity of the protoplasm and to other toxic effects of the salts taken in by the cytostome, which result in the vacuolization of the protoplasm. 384 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The water and fat content of skeletal muscle in marine fishes. Charlotte Haywood and Abby Turner. In the course of a study of water balance in fishes of different life habits and phylogenetic relationships a series of determinations was made of both the water and fat content of skeletal muscle. This work was done in part at Bergens Museum Biologiske Stasjon, Herdla, Norway, and in part at the Marine Biologi- cal Laboratory, Woods Hole. Percentage of Water in No. of Percentage Indi- of viduals Fats Whole Fat-free Muscle Muscle Elasmobranchs: Mustelus canis.............. 4 Trace 78.4 78.4 Pristiurus catulus*...........- 4 — 80.0 — Galeus glaucus..............- 1 Trace 81.0 81.0 Carcharias Tauruse 2-2... 2 0 77.4 77.4 Squalus acanthias*........... 1 — 78.4 — Squalus acanthias*........... 7 5.4 75.0 79.4 Spinax niger*............--- 4 — 80.6 — ARO OMAUOPWOITES arias ee eis 2 eh 1 0 Sail oe IRI G SHOLODUOS = ds go Oo aoa 2 0 76.5 76.5 IROGO WOMORUG” 65 eo cba haahoo€ 1 — 79.0. — Raja oxyrhynchus*........... 3 — 81.1 — Narcacion nobilianus......... 3 0.4 83.9 84.2 Dasybatus marinus........... 2 Trace 717.9 717.9 Chimaera monstrosa*......... 3 —_ 81.3 — Teleosts: Anguilla vulgaris*........... 3 — 66.4 —_ Anguilla rostrata............. 1 4.7 73.3 ee Scomber scombrus*........... 6 — 66.5 — Scomber scombrus*........... 8 5.4 70.5 74.5 Centropristes striatus......... 11 i.) Useth 79.5 Cynoscion regalis.......5.5-- 2 Soll 76.7 79.2 Tautoga onitis............... 4 Trace 79.8 79.8 Cyclopterus lumpus*.......... 1 — (BL — Prionotus strigatus........... 2 0 79.9 79.9 Echenets naucrates........... 1 3.7 73.1 76.2 OPSUTIS UD si c0c0 en cco cone os 2 Trace 80.2 80.2 Anarhichas lupus*........... 6 — 84.0 — Gadus callarias*............. 8 — 80.3 Gadus callartas*............. 6 0 81.7 81.7 Gadus pollachius*............ 7 — 79.4 Brosmius brosme*............ 6 — 80.0 = Molva molva*................ Z — 80.3 = Lophius piscatorius*.......... 5 — 84.3 — Lophius piscatorius*.......... 2 0 87.1 87.1 * Determinations made in Norway. Reactions to light of different intensities in Dolichoglossus kowalevsky. Walter N. Hess. For studying the sensitivity of Dolichoglossus to different light intensities a 100-watt Tungsten bulb, supplied with neutral tint Wratten filters, was used and PRESENTED AT MARINE BIOLOGICAL LABORATORY 385 light of 115, 11.5, 1.15, and .115 m.c was thus obtained. For studying the rela- tive photosensitivity of different regions of the body a fused quartz rod was sus- pended so that its base was illuminated by an arc lamp. The distal end, which was used in testing photosensitivity of small areas, was drawn out into a very small blunt point. The animals reacted negatively to ordinary intensities of light but at .115 m.c. a rather large majority of positive responses occurred. By means of the pin- point light it was shown that the animal is photosensitive over its entire body though certain regions are more light-sensitive than others. The tip of the proboscis is the most sensitive to light. In general, the dorsal and lateral body surfaces are more photosensitive than the ventral surface. Mucous cells occur on the abdomen in patches interspersed with non-mucous cells. Very little response was obtained by illuminating the areas of mucous cells especially on the ventral surface, yet if the areas of non-mucous cells were illuminated the animals responded quickly. These results show that the photoreceptor cells must be widely distributed and that they must be more numerous or more specialized in certain regions than in others. Cells of a certain type which correspond in their distribution to the relative photosensitivity of the animal have been identified. In keeping with the early chordate characteristics of this animal these cells resemble in their general morphology the retinal cells of vertebrates. The hatching of the squid. Hope Hibbard. The hatching gland of Loligo pealit is the so-called Hoyle’s organ, a Y- or T-shaped gland which lies on the dorsal posterior surface of the larva, the arms of the gland extending out on the fins. It appears very early (about stage 12 of Naef in L. vulgaris), gradually matures, and disappears entirely shortly after the larva hatches. At the height of its development the gland protrudes slightly. A similar hatching gland has been described by Wintrebert and by Jung in several other cephalopods found in Europe. Staining the eggs or young embryos, even very heavily, with neutral red does not impede their normal de- velopment, and the organ in question can be very readily observed since it does not stain and appears as a white streak against a reddened mantle. In early development the larva rotates inside its shell by ciliary action, push- ing the posterior end up until gravity causes it to lose balance, and fall to the lower side of the shell whence rotation recommences. But as hatching approaches the animal adheres very firmly to the shell as if stuck by a secretion, in a disc- shaped area around the hatching gland. The head is directed downward, and rhythmic contractions of the free part of the mantle are constant. The gland itself appears to wriggle and squirm from time to time, due to underlying muscle cells, pushing ever closer to the shell. In some cases the weak place bulges con- siderably before giving way. Finally the shell appears to dissolve away and the animal swims out backward through a neat round hole. There is no tearing, and after the larva has emerged the round hole remains in the empty shell. . The cells of the gland are very long, slender ones with the nucleus at the proximal end deeply embedded in the mantle. The secretion is conspicuously granular and resembles well-fixed zymogen granules in a pancreas cell. Pressure on the cells forces the granules out, and they maintain their individuality out- side. After hatching, the cells remain for a few hours, but the distinctness of the unsecreted granules disappears, the material becomes more fluid and runs together. Further cytological examination of the evolution of these cells is under way. 386 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The vacuole system of the marine Amoeba, Flabellula mira. D. L. Hop- kins. In this study the vacuole system was observed when the amoebae were under normal conditions; slightly abnormal conditions; unstained; stained with the vital dyes, neutral red, Nile blue sulfate and Janus green B; and after impregnation with osmium tetroxide. Mitochondria which stain with Janus green B are not present in normal fully active amoebae, but appear and may be stained with Janus green B when the amoebae are placed under slightly abnormal conditions. The mitochondrial sub- stance in fully active amoebae appears to be contained in very small vacuoles. These vacuoles arise de novo from the protoplasm, grow, fuse with each other and engulf food. The food vacuoles and the vacuoles containing no food fuse with each other, forming larger and larger vacuoles, until finally a large fusion vacuole, the cloacal vacuole, is formed. The fluid and solid contents of this cloacal vacuole after a period of digestion are voided to the outside. Nile blue sulfate stains all these vacuoles blue from the time of their origin until after their entrance into the cloacal vacuole where the blue fades out, in some cases becoming clear of color directly, while in others it becomes pink or red and then clear. This indicates that fatty acids are present in all vacuoles, and perhaps some neutral fat. Both of these types of substances disappear from the vacuoles before the residue is voided. The substance stained by neutral red is present in all the smaller vacuoles but disappears entirely before they enter into the cloacal vacuole. Neutral red staining, and osmium tetroxide impregnation of amoebae pre- viously treated with neutral red show that the smaller vacuoles and the cloacal vacuoles are in a more highly oxidized, or in other words, in a less reduced con- dition than are the vacuoles of intermediate size. Effect of electrical shocks upon the division rate of Stylonychia pustulata as measured by the interdivisional period. Lois Hutchings. The normal interdivisional period of Stylonychia pustulata in summer was found to be six to ten hours. Most frequently the control animals underwent division once every eight hours. Every animal used in these experiments came from one protozoon. Isolation dishes were used. The culture medium was filtered hay-tea in which mass cultures had lived previously for five to twenty days, average pH 7.4. A piece of dry outmeal, one-millimeter square, was placed in each dish when the medium was changed each day. Two new dry-cell batteries were connected in series to the primary circuit of an inductorium, Harvard type. The secondary coil was kept at the eight-cen- timeter mark. The current which flowed through the secondary circuit had a non-smooth alternation of 106 times a second, known as tetanizing current. A foot switch closed the circuit at will. From the secondary circuit were two insulated copper wires to whose tips were soldered one-inch lengths of No. 30 platinum wire. Each point electrode was so manipulated by hand that a con- stant distance of twice the body-length between it and the animal was maintained. Even when the protozoon moved it was usually possible to keep the interelec- trodal distance constant. Each Stylonychia received treatment hourly. For several reasons physiologi- cal indices of sufficient treatment were adopted. In order of importance these were: degree of swelling, loss of lateral orientation to the current, and rapid spinning. The length of treatment varied from 10-75 seconds. Disregarding the five cases in which the treated Stylonychia had the same length of interdivisional period as the control animals, the results may be simply stated. Fifteen treated animals underwent an average shortening of the inter- PRESENTED AT MARINE BIOLOGICAL LABORATORY 387 divisional period by 2 hours, but 102 animals had an average lengthening of the interdivisional period by 634 hours. The range of shortening was %-3%4 hours and the range of lengthening was 14-23 hours. In other words, although in 13 per cent of the cases application of electrical shocks caused an apparent acceleration of the division rate, in far the greater number of cases, 87 per cent, such shocks caused a slowing down of the division rate. Further comparative studies on the permeability of the erythrocyte. M. H. Jacobs and H. N. Glassman. In earlier papers (Proc. Am. Philos. Soc. 70, 363, 1931; Jour. Cell. Compar. Physiol. 7, 197, 1935) certain characteristic differences in the permeability of the erythrocytes of different species of vertebrates to dissolved substances were reported. The results previously obtained by means of the hemolysis method were not very satisfactory in the case of many species having nucleated erythrocytes be- cause of the failure of suspensions of the latter to become completely transparent on hemolysis. By taking advantage of the fact that this difficulty is largely removed by the addition to the hemolytic solutions of 0.001 M NaHCOs, without greatly altering the time of hemolysis in the case of most of the species studied, further data have been obtained on the permeability to ethylene glycol, glycerol, urea, and thiourea, of the erythrocytes of 9 species of elosmobranchs, 14 of teleosts, 2 of frogs, 6 of turtles, 4 of snakes and 4 of birds. In addition to various highly characteristic specific differences the following general peculiarities of the larger groups—subject to some exceptions which cannot here be men- tioned—are of interest. Fishes: permeability to ethylene glycol greatest; that to urea and glycerol highly variable from species to species; permeability to thiourea usually greater than that to urea. Amphibia: (data as yet too scanty to permit generalizations). Birds (gull, tern, English sparrow, pigeon): permeability to ethylene glycol and glycerol both very great and nearly equal; that to thiourea much less and to urea least of all. Earlier experiments showed similar conditions in the starling but not in the duck and the chicken. Reptiles: permeability to urea relatively great, followed by that to ethylene glycol and at a much greater dis- tance by that to thiourea. Permeability to glycerol slight as compared with that to the other substances. Mammals: (for comparison) permeability to urea ex- tremely great; that to ethylene glycol much less and to thiourea still less; per- meability to glycerol in some species greater and in some species less than that to thiourea. The attenuation of toxins by interfacial adsorption. J. M. Johlin. In view of the fact that some toxins, upon standing, are changed into toxoids, the writer thought it likely that similar attenuations might be brought about more speedily by methods which induce interfacial adsorption, and has applied such a method to the attenuation of ricin, tetanus toxin and snake venom. This speedy attenuation may be regarded as the result of the catalytic influence of the interface in bringing about an increased surface concentration of properly oriented molecules of the toxin at the interface. Adsorption was brought about by emulsification of the toxin with a volatile inert liquid such as ether or chloroform which could be easily removed by evaporation under reduced pressure afterwards. Ricin, thus attenuated, was found to be ten thousand times less active when injected intracutaneously into rabbits. One thousand M.L.D.’s of tetanus toxin, when attenuated, could be injected subcutaneously into mice repeatedly at two- day intervals without killing them or causing any apparent prolonged discomfort. Multiple lethal doses of mocassin and rattlesnake venom could also be injected into mice without causing death or producing the usual signs of damage caused by the untreated venoms. In such immunization experiments with rabbits as 388 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS have been carried out it was found that the attenuated toxin had retained its antigenic property and could be injected daily in large amounts without producing visible symptoms of any kxind. The action of acetylcholine on the skeletal muscle fibers of the frog. _Elsa M. Keil and F. J. M. Sichel. In a previous report by the present authors (Biol. Bull. 71: 402, 1936) it was shown that the application or injection of very small amounts (50004°) of acetylcholine in concentrations ranging from 1 in 10* to 1 in 10* had no effect on the isolated single fiber (adductor muscle of frog’s leg, Rana pipiens), beyond the effects attributable to the medium in which the acetylcholine was dissolved. Recent experiments with acetylcholine on single muscle fibers isolated from the sartorius of the frog confirm this. However, if the sartorius be dissected from the frog, and even if both the tibial and pelvic ends be cut across, then acetylcho- line will cause a propagated twitch when applied as a small droplet with a micro- pipette to the dorsal surface of the muscle in the region of the nerve twigs. This twitch is noticed in only those muscle fiber bundles which are in the vicinity of the point of application of the acetylcholine. If either the pelvic or tibial half of the sartorius is split lengthwise into two parts, one containing about twenty fibers, the other the rest of the fibers, then acetylcholine applied to the surface of either of these bundles also causes a similar propagated twitch response. If the smaller bundle of fibers is progressively split into still smaller groups, a degree of subdivision will eventually be reached such that even dilute acetylcholine will no longer evoke twitches in fibers still irritable to electrical stimulation. This apparently occurs when the subdivision has interfered with the nerve supply, the fine nerve twigs being injured by the process of dissection. We conclude that these experiments offer further evidence in support of our previous statement that the action of acetylcholine is not upon the contractile mechanism of the muscle directly. We further conclude that acetylcholine evokes twitches in the sartorius muscle of the frog only when the terminal nerve supply or possibly some junctional tissue or receptor is intact. These conclusions are in accord with Garrey’s experiments on turtle and Limulus heart, and with Arm- strong’s experiments on embryonic Fundulus heart. The effect of pituitary on nuclear changes in the egg of the frog. John A. Moore. In an effort to separate the processes of maturation and ovulation in the frog, Rana pipiens, mature females were given homoplastic anterior lobe injec- tions during the middle of the summer, at which time it has not been possible to stimulate the release of eggs. Parts of the ovary were removed before and after injection so the same animal serves as a control and as an experimental. It was found that by thirty to forty hours after injection (at 25-27° C.) the germinal vesicle had moved from its position in the center of the egg to the surface at the animal pole. This movement is found only in the large eggs with evenly distributed yolk. By one hundred and twenty hours the eggs are degen- erating and have been heavily invaded by phagocytes. This latter condition is similar to that found by Miss King in the few mature eggs that remained in the ovary of the toad after the breeding season. The effect of urea upon the surface of unfertilized Arbacia punctulata eggs. Floyd Moser. Following the procedure of Moore (Protoplasma, 1930) Arbacia punctulata eggs were treated with molar urea solutions. After a total exposure of no PRESENTED AT MARINE BIOLOGICAL LABORATORY 389 more than two minutes, the eggs were transferred to sea water. Some of them were then inseminated and others served as uninseminated controls. Within three or four hours irregularly cleaving eggs (two to eight cells) may be found among both the treated inseminated eggs and the treated lunin- seminated eggs. At the same time apparently normal blastulae with membranes perhaps a little thinner than normal may be found among the treated inseminated eggs, but not among the uninseminated eggs. In some cases it is possible to demonstrate very thin and but slightly elevated membranes in these irregularly cleaving eggs. If a small drop of sea water containing centrifuged or uncentrifuged eggs be placed on a slide and a larger drop of molar urea solution be caused to flow into the drop of eggs while the latter are being observed under the high power of the microscope, a breakdown of cortical granules occurs. Immediately following this cortical response membrane elevation occurs with the formation of a very © wide perivitelline space. Within a few seconds the elevated membrane begins to recede toward the egg surface, becoming thinner as it moves, until finally, in some cases, no vestige of the membrane remains. The cortical response secured in these urea-treated eggs is essentially like that obtained upon stimulation with sperm cells or with agents which induce parthenogenesis such as saponin. Similar cortical responses are secured in molar thiourea and molar glycerine solutions. It is evident, therefore, that these non-electrolyte solutions do not prevent membrane elevation in Arbacia punctulata eggs (either irreversibly or otherwise), but may actually stimulate the egg to cortical response and subsequent membrane elevation. Indeed, delayed and irregular cleavage often follows the cortical re- sponse after treatment with these agents. The cortical response of Arbacia punctulata eggs to direct current. Floyd Moser. In a previous abstract (Biological Bulletin, 1935) it was noted that a layer of granules located in the cortex of the unfertilized Arbacia egg breaks down when the egg is subjected to a number of different stimulating agents, as well as to normal insemination. The present experiments with direct current extend the list of stimulating agents reported earlier. A non-polarizable system of Cu-CuSO, electrodes with agar bridges was used. The eggs were placed in small glass tubes both ends of which were plugged with 2 per cent agar made up in sea water. The current intensity varied from 1.5 to 20.0 milliamperes. As soon as the circuit is closed the eggs begin to move toward the anode. Within a few seconds cortical layer granules on the anodal side of the egg break down, releasing the fertilization membrane from that portion of the egg over which the response has taken place. When the current is reversed the same re- sponse takes place on the opposite side of the egg. Generally some of the cyto- plasmic granules beneath the cortex also exhibit this breakdown phenomenon. Indeed, when the current flows for too long a time the cortical response is followed or accompanied by complete cytolysis. In the uncentrifuged eggs the cortical response stops when the circuit is broken, but generally completes itself in wave-like fashion in the centrifuged egg. The so-called fifth layer usually breaks down in the centrifuged eggs. If the eggs are treated with M/4 ammonium oxalate, 0.3 M potassium oxalate or 0.35 M potassium citrate there is no granule breakdown or membrane elevation upon exposure to direct current. The observations here recorded are in general agreement with the point of view developed by Heilbrunn and his students in their studies of stimulation. 390 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The histology of the retractor muscles of Thyone briareus Lesueur. Magnus Olson. A decided gap exists in our knowledge of the comparative histology of in- vertebrate muscle. This fact has made desirable a more comprehensive study of invertebrate muscles particularly since a considerable number of these have been employed in recent physiological research. The only extensive paper on the histology of holothurian muscle (Hall, 1927 on Cucumaria) misinterprets their essential structure and has been responsible for some erroneous conclusions by workers in muscle physiology. The present study has shown that the retractor muscles of Thyone consist of enormously long narrow unstriated fibers imbedded in a connective tissue network. The fibers occur in bundles of 2-15 and in each bundle are arranged circularly in a single layer. Each bundle is surrounded by a comparatively dense layer of connective tissue which penetrates only sparingly into the interior of the bundle. Hall interprets the bundles of fibers as being single cells each of which contains many enormously large fibrils. That these so-called fibrils of Hall are actually the muscle fibers can be determined readily not only by the presence of connective tissue within the fiber bundles but also by a study of the arrangement of fiber nuclei. Critical examination of this arrangement reveals a peculiar pic- ture. In the contracted fiber the nucleus lies on the outside of the fiber and may be connected to it only by a narrow protoplasmic strand. In the extended con- dition of the fiber the nucleus is elongated in the axis of the fiber and lies closely pressed against it. Occasionally nuclei are found directly within the fibers. As far as it has been possible to determine, the fibers appear to be uninucleate. Con- nective tissue nuclei occur not infrequently within the bundles of fibers. The fibers are extremely extensible. The muscles may vary in length from about 5 mm. in the contracted condition to 5 or 6 cm. in normal extension or after anaesthetization with magnesium chloride. The fiber diameter varies from 5-10 mu in a contracted condition to 2-4 mu in an extended condition. From teased preparations it may be seen that although the fibers are ex- tremely long, they do not extend the full length of the muscle. The terminal portions of the fibers are long and tapering. The fibers are spherical or hemi- spherical in cross-section in a contracted condition, laterally compressed in an extended condition. The connective tissue fibers appear to run at right angles to the muscle fibers and form a dense network at the periphery of the muscle. Strength-duration curves of nerve fibers in the squid. C. Ladd Prosser and A. H. Chambers, Jr. Giant nerve fibers and fin nerve fibers of the squid were stimulated by con- denser discharges through calomel electrodes and strength-duration curves were obtained. Excitation constants (Hill’s characteristic time, k) range from .33 to .60 milliseconds and average .47 milliseconds for the isolated giant fibers. The time constants are essentially the same (characteristic time averages .42 milli- seconds) for the giant fiber intact in the stellar nerve as for the isolated fiber, but the whole strength-duration curve is shifted upward with the intact nerve. Hence the effect of other fibers in a nerve trunk upon the excitation of one fiber is to serve as a shunt for the stimulating current. Decreasing the electrode separation from 12 to 3 mm. shifts the strength- duration curve diagonally upward and to the left, with the result that the time constants are shorter at the smaller separation. As the preparation deteriorates the threshold rises and the characteristic times become shorter, i.e. the change in the strength-duration curve is similar to the change resulting from shortening in- PRESENTED AT MARINE BIOLOGICAL LABORATORY 391 terelectrode distance. With electrodes 7 mm. in diameter the time constants are approximately 1.5 times as long as with electrodes less than 0.6 mm. The diameters of the giant nerve fibers are approximately 100 times those of the fin nerve fibers. The fin nerve fibers have characteristic times approximately twice as long as the giant fibers (.88 millisecond with electrode separation of 12- 14 mm.). New structures induced by implants of adult nerve cord in the polychaete, Clymenella torquata. Leonard P. Sayles. For several summers a study has been made in which a piece of adult nerve cord, removed from one Clymenella, was transplanted into another. All implants were of one to two segments in length. They were placed in the body wall in a dorso-lateral region to minimize the chance of injuring the nerve cord of the host. No bud has been formed in any case in which the nerve cord slipped into the coelom free from the body wall. To date 63 buds have been induced by this type of implant. These include 12 heads, 29 tails, 2 double buds each consisting of a head and a tail, and 20 irregular or weakly developed buds. No head buds have appeared posterior to the tenth segment of the host. One of the double structures was formed at the fourth, the other at the eighth. Both the source of the implant and the region of implantation in the host seem to influence the type of bud to be formed. Many more cases must be secured before any conclusions can be drawn concerning this point. The possibility that orientation of the implant may also be a factor has not been fully eliminated. Cytoplasmic division in type B of the giant amoeba Chaos chaos Lin- naeus. A. A. Schaeffer. Although type A and type B of Chaos chaos are very closely similar, morpho- logically, to the common laboratory amoeba, Chaos diffluens, there are certain characteristics which set them apart very definitely. One of these characteristics is the striking phenomenon of frequent division into three daughters instead of two, which occurs in both types of chaos. During division the amoebas of type A round up into a mulberry-like shape, while those of type B, which are also berry- shaped at first, soon flatten out somewhat, so that one cen predict to a certain extent the number of daughters about to be produced, by counting the well-defined lobes of the flattened-out amoeba. The number of daughters may be 0, 2, 3, 4, 5, or 6, and the number of lobes formed before separation is well correlated with the number of daughters produced. The amount of undigested food in the amoeba influences the number of daughters to be formed, to a markedly greater extent than it influences the number of lobes formed during division. In both cases the number is reduced by large quantities of undigested food. In type A, more divisions result in 3 daughters than in all the other classes put together, while in type B, in a culture with food in excess of needs, as many amoebas divide into 2 as into 3. But the number of amoebas showing 3 lobes prior to division, under these cultural conditions, is much greater than those with 0, 2 and 4 lobes combined. The average variation in size between twins from a 3- lobed parent is also greater than that between twins or triplets whose parents showed 2 and 3 lobes respectively before separation of daughters. More sex- tuplets occur than would be expected on pure chance, although not enough data are at hand to warrant the conclusion that the curve of frequency rises again to some degree at 6 daughters. Something akin to a trigonal field of force in an apparently turbulent medium exists here which stands in striking contrast to the common halving type of cyto- plasmic division. 392 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The sense of taste in the free fin rays of the sea robin (Prionotus). Ernst Scharrer. Experiments were carried on to show that the free fin rays of the sea robin are chemoreceptors as well as organs of locomotion. Blinded sea robins react after 3-4 weeks of acclimatisation in the aquarium with snapping reactions when the juice of a spider crab or of a clam is gently poured on the free fin rays by means of a pipette. A water current caused by a pipette never is followed by a reaction. The behavior of the sea robin is therefore identical with that of the Mediterranean genus Trigla (Scharrer, 1935) and it might be concluded that all Triglidae use the three free rays of the pectoral fins as chemoreceptors in search for food. Calcium and magnesium in relation to longevity of Mactra, Nereis and Hydroides egg cells. Victor Schechter. Experiments on the relation of calcium to longevity of unfertilized Arbacia egg cells (Biol. Bull., 72: 366) were extended to the cells of the mollusk, Mactra, and to those of Nereis and Hydroides. The Mactra egg, which in change in rate of hypotonic cytolysis with age (Biol. Bull., 71: 410) follows the Arbacia egg’ closely, also exhibits prolongation of life in low calcium most clearly. This ac- tion, as in Arbacia, seems to be one rather specific to calcium as a decrease in magnesium alone does not produce it nor does lowering of both magnesium and calcium have an added effect. Increase of magnesium above the normal concen- tration in sea water did result in an additive effect on longevity. The two phenomena need not, however, be related. The first sign of deteriora- tion in the Mactra egg is usually the breakdown of the germinal vesicle. Insemina- tion also causes germinal vesicle breakdown. In eggs which cannot be activated by sperm, cytoplasmic cytolysis may occur with age while the germinal vesicle is still intact. With such eggs, increasing the magnesium concentration retarded cytoplasmic disintegration; and the eggs then lived longest if calcium was low. Therefore, in so far as the phenomena of aging may be partitioned between the various cellular components, the effect of low calcium is tentatively regarded as one which retards deteriorative changes in the nucleus, whereas the action of high magnesium may be primarily in the nature of an anaesthesia. Magnesium may act chiefly on the cell membrane since with increased magnesium the egg cells retain their spherical shape beyond the time when controls appear plasmolyzed. Preliminary experiments with Hydroides eggs show prolongation with low calcium. In one-sixth the normal calcium concentration the reaction of Nereis eggs was variable. This is possibly due to egg condition and it may be that with different calcium and magnesium concentrations prolongation of life in these cells will also be obtained consistently. A convement method for the measurement of nerve respiration. Francis O. Schmitt and Otto H. Schmitt. For the determination of the effect of electrical stimulation on the oxygen con- sumption of single nerves it is desirable that readings be made over short in- tervals of time (every 2-3 minutes) and that the variations between individual readings be small (ca 2-5 per cent). Even with differential volumeters of small volume this requires that temperature fluctuations between the two vessels be re- duced to a minimum. This may be accomplished without the use of a precision thermostat by immersing the vessel in a mercury bath, the latter in turn being en- closed in a well insulated box. The mercury is contained in a copper box lined inside and outside with insulating material and provided with a glass window which PRESENTED AT MARINE BIOLOGICAL LABORATORY os) permits following the movement of the index droplet with a traveling microscope, the capillary being only partially immersed in mercury and bent so that the vessels are submerged well below the surface of the mercury. Wires from the electrodes are brought out through rubber tubes. The capillary is illuminated by filtered light brought in by a glass rod fastened on the carriage of the travelling micro- scope. The copper box is surrounded by kapok or eiderdown contained in a well- insulated box which also carries the traveling microscope. The entire unit is easily portable and, at least for work at room temperature, the method has been found quite as satisfactory as the conventional method which requires a precision thermostat of special design. Temporal relations in the excitation of the isolated muscle fiber. F. J. M. Sichel and C. Ladd Prosser. The possibility of spatial summation in the excitation of the isolated muscle fiber has been reported previously (Sichel and Prosser, Biol. Bull., 69: 343, 1935). In the present case the excitatory effects of two stimuli separated by a varying interval have been studied. The stimuli were obtained from two condensers, each controlled by a gas triode, discharging into a common resistance. The interval between the two shocks was varied by a Lucas’ spring rheotome in the grid cir- cuits of the gas triodes. The time-constant for the discharge of each condenser was 1 millisecond. The fiber (adductor longus, Rana catesbiana) was isolated and mounted in a manner previously described (Sichel, Jour. Cell. Compar. Physiol. 5: 21, 1934) and the isometric tension recorded by a micro-lever (Brown and Sichel, Jour. Cell. Compar. Physiol. 8: 315, 1936). The electrodes were silver-silver chloride, and were sufficiently large to make the spreading of the field at the ends of the fiber unimportant. The tension developed falls continuously as the interval between the shocks increases, rapidly at first, then more and more slowly toward a steady value reached when the interval between shocks is about 20 milliseconds. The tension is then one-third to three-quarters of the tension developed when the stimuli are simul- taneous. If a similar experiment is done on the whole sartorius using maximal stimuli, as the interval between stimuli increases the tension at first remains constant, this being related to the absolute refractory period, then increases rather abruptly to a new level. With submaximal stimuli there is an initial fall to the constant level followed by a subsequent rise. The absence in the case of the isolated fibers of the low constant level of tension for short intervals, and the smooth nature of the tension-interval curve we attribute to the absence of an absolute refractory period in the excitation of these fibers. Electrolytes in Phascolosoma muscle. H. Burr Steinbach. The retractor muscles of the marine annelid Phascolosoma gouldi consist of closely packed long smooth muscle fibers. Muscles freshly excised from the animal contain the following average amounts of inorganic elements, expressed as milliequivalents per hundred grams wet weight: Na 10.4, Cl 9.1, K 10.6, Ca 0.85. The body fluid surrounding the muscles contains approximately: Na 38, Cl 43, K 3.5, Ca 2.0. On immersion for two hours or more in sea water, the tissues gain weight by 11 per cent, chloride increases by 16 per cent, Na by 14 per cent, Ca by ' 13 per cent while 11 per cent of the K is lost. These changes can be partly ac- counted for by assuming an extra-cellular chloride space which alone is involved in the tissue swelling and the substitution of sea water for body fluid. Both Cl and Ca increase somewhat more than would be expected on this simple hypothesis. 394. PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS The Na increase is less than would be expected while a large part of the K loss must involve the fibers themselves. Muscles were soaked for two hours or more in artificial sea water containing known concentrations of K or Ca, or in sea water diluted with isotonic sucrose solution. Na, Cl and Ca contents of the muscle bear a simple linear relationship to the concentrations of these ions in the immersion fluid. All of the Na and Cl ap- pears perfectly free and diffusible, the slope of the curves Na-tissue/Na-fluid and Cl-tissue/Cl-fluid indicating about 30 per cent chloride (extra cellular) space at all concentrations. About half of the initial Ca content does not diffuse into Ca- free sea water or isotonic sucrose, but using that figure as zero concentration of Ca in the muscle, the curve for Ca-tissue/Ca-fluid is linear and has a slope indi- cating about 40 per cent diffusion space. Eighty per cent of the K does not diffuse into a K-free medium. As the K content of the fluid is increased the K of the muscle increases, rapidly at first then more slowly until finally a linear relation- ship is shown between K-tissue and K-fluid at concentrations between 0.05 M and 0.25 M K in the fluid. Leaf development and vegetative propagation in Polystichum plasch- nickianum. Harry N. Stoudt. Material for this work was collected by the writer in the rain forests along the Vinegar Hill Trail, Jamaica, B. W. I. in the summer of 1936 and the investi- gation was undertaken at the suggestion of the late Professor Duncan S. Johnson. Leaves of this plant are produced from segments of the apical cell of the rhizome. The leaf grows, apparently, by means of a bifacial initial which persists in the mature leaf. This cell continues to cut off segments to form the first em- bryonic leaf of the plantlet at the apex of the leaf. An apical cell is cut out from a marginal cell of one of the last segments of the parent leaf to form the shoot of the plantlet. Later leaves are formed from segments cut off from this cell. A meristematic cushion forms at the apex of the parent leaf and adventitious roots arise endogenously in this region and grow by means of a tetrahedral initial. By this time the roots penetrate the moist soil and the plantlet matures. Simultaneous with plantlet development the parent leaf develops wings that encircle the plantlet. Further investigation is needed to understand the origin and development of these wings. When several leaves and roots have formed the plantlet becomes inde- pendent and this region of the parent leaf decays. The remaining portion of the mother leaf persists to carry on spore formation. Morphological and experimental cytology of lobster spermatozoa. (Pre- liminary note.) T. Terni. In Woods Hole I have begun to microdissect the spermatozoa of decapods, most appropriate material because of its immobility and large size. I refer here only to my preliminary results. 1. The three long rays of the head have a permanent shape which is slightly curved. They are elastic, and if deformed by the needles, they immediately resume their original form. The rays are, on the other hand, neither plastic nor extens- ible if pulled with the needle. The mechanical stimulation does not induce visible contraction of the ray-like filament. Movements of these, if really existing, are in all cases so slow that only the microcinematographical technique will reveal them. 2. If with the tip of the needle one exerts considerable pressure, although not sufficient to injure, on the anterior part of the spermatozoa, there always appears the phenomenon of the so-called “explosion” (Koltzoff). The process seems to be irreversible; I have followed for 20 hours the fate of the exploded spermatozoa without changes. In order that the explosion should take place after pricking, it is necessary that the spermatozoa be freshly removed from the spermaduct. PRESENTED AT MARINE BIOLOGICAL LABORATORY 395 3. In the posterior part of the small tube contained in the chitin capsule of the tail there exists before the explosion material strongly acidophilous. After the explosion the capsule and its content become suddenly basophilous while the nu- clear chromatin loses its strong basophily, perhaps because it diffuses in the capsule. 4. With microdissection it is easy to follow the very rapid process of invagina- tion of the anterior part of the spermatozodn into the capsule, which simul- taneously becomes swollen. The examination of good preparations (ac. osmic-iron hemat.) shows the fact that the basis of the cephalic rays are pulled into the tail portion like an umbrella that closes itself into a sheath. Simultaneously from the posterior pore of the tube there flows out a basophilous body, perhaps the anterior centrosome. Observations on arterial pressure in marine fishes. Abby H. Turner and Charlotte Haywood. In an effort to determine representative values for the arterial pressure sup- plying the organs of fishes the. anterior mesenteric artery was employed for can- nulation. Pressures were measured in centimeters of isotonic saline, using a manom- eter. Respiration was well maintained by a stream of water over the gills, but obvious disadvantages in the method are that the fish is not entirely submerged in water and that the body cavity must be opened. A number of the determinations were made in the summer of 1936 at Bergens Museum Biologiske Stasjon, Herdla, Norway. No. of Range of Values Average Values Individuals in cm. Saline in cm. Saline Elasmobranchs: WVustelus CONUS). 6.3 ce ens ss 4 11. —25.5 16.7 Carcharius taurus. 75-5. 05..0- 2 16. —19. Wes Squalus acanthias............ 7 8.5-18.8 foal Raja dia phranes a.) 25... -- 1 Me RFC SIQUULLT ONES: o-oo ecient 2 12°5—2 5 Iie OD OXNTIN WENUSE on ee 1 8.3 Narcacion nobtlianus......... 3 Extremely low Dasybatus marinus........... 2 9. -10.1 9.5 Teleosts: Centropristes striatus......... 1 37. VAOTTAGHOSILPUS.. fa. 4 9.7-15. HES Gadus callartas. 4:0) e458 oe 12 10.9-43. ISTE - Gadus pollachius............. 2 ID D2 WHY 12.2 Lophius piscatorius.......... 6 9. -23.7 16.7 Respiratory rate and length of fertilizable life of unfertilized Arbacia eggs under sterile and non-sterile conditions. Albert Tyler, Nelda Ricci and N. H. Horowitz. The respiratory rate of unfertilized eggs of marine animals does not remain constant with time, but sooner or later shows a marked rise. This rise occurs at the time when the eggs lose their fertilizability (Tyler and Humason, 1937). Measurements of the respiration of Arbacia eggs in 2 per cent alcohol or other agents that are known to prolong the fertilizable life (Whitaker, 1936) show that the rise is correspondingly delayed, while the initial absolute rate is unaffected. 396 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS By plating out for marine bacteria from the respiration vessels, it is found that the bacteria increase as the rate of respiration rises both in sea water and in alcohol. When sterile or almost sterile cultures of eggs are run it is seen that the respiratory rate remains low and shows even a slight drop. Such eggs have been run for four days without showing a rise in respiratory rate, whereas under ordinary conditions the rise begins at about 10 hours. Culturing eggs under sterile conditions in sealed dishes prolongs their fer- tilizable life as Gorham and Tower (1902) found, some of the eggs lasting as long as 20 days. Under such conditions alcohol appears to shorten the fertilizable life. Addition of bacteria accelerates somewhat the disintegration of the eggs. Alcohol in ordinary sea water or in the presence of small amounts of bacteria considerably prolongs the life of the egg. In the presence of large quantities of bacteria, 2 per cent alcohol is toxic within a few hours, such cultures showing a drop in pH to as low as 4.60. The results evidently mean that bacteria contribute to the destruction of the eggs only after certain disintegrative changes have occurred. The agents that prolong the life do not act bactericidally, but by retarding these initial spontaneous changes on the part of the egg. The effect upon gastrulation and differentiation in Arbacia of NiCl,, CuCl,, and Na,SiO, in modified artificial sea water. A. J. Water- man. A formula for artificial sea water consisting of 390 cc. Im NaCl, 9 cc. lm KG 23:35) ce. im MeCL” 25.5 ce) lm MesO,, 9:35)ce) In (Ca@l, about On vearand NaHCoOs, additional water to make one liter (Chambers), and corrected for pH supports development of Arbacia from the blastula stage in a typical manner. Modifications and omissions of salt proportions have been studied in relation to effects on gastrulation and embryonic differentiation (Runnstrom, Lindahl, Dalcq). The addition of NiCl, and CuCl, to the above has been studied and comparison made with previous observations with these heavy metals in sea water. Increase, decrease or omission of Mg, SO,, K, and Ca influence type and rate of effect of the metals. Exogastrulation occurred in some experiments and gas- trulation could be temporarily inhibited. The blastula stage was used in all experiments. Na.SiOs was tested in relation to respiratory and cell division rates of the embryos in sea water and in the above modifications. In relatively large concen- trations silicon is non-toxic and does not affect gastrulation. Normal cleavage rate is increased, top-swimming blastulae appear earlier, plutei mature sooner, and gastrulation may be accelerated. There is no noticeable differential accelera- tion of germ-layer differentiation. This has also been studied in artificial media where the salt balance has been disturbed or metallic salts added. Toxicity may be increased. Addition of silicon (Mast and Pace) to cultures lacking sulphur may or may not influence developmental rate. Compounds of silicon are very abundant in nature and present in all waters (Reynolds, 1909). Both animals and plants use silicon and it is found in animal tissues. According to Mast and Pace, 1937, it increases rates of respiration and division in Chilomonas and has a catalytic action on synthesis of complex organic compounds. Studies on living conjugants of Paramecium caudatum. Ralph Wich- terman. A new method of approach to a better understanding of the problem of sex- uality in Paramecium is presented. By means of a precision micro-compressor, it PRESENTED AT MARINE BIOLOGICAL LABORATORY 397 is possible to observe in the living condition, the behavor of nuclear phenomena and establish time relationships during conjugation. With a micro-pipette, a single pair of recently joined conjugants is placed on the glass slide of the compressor in a small drop of the culture medium and studied at 26° C. The metal top holding the glass cover-slip is screwed down while observations continue with the microscope. The animals are prevented from spiraling between the two pieces of glass and allowed to move around slowly. Adjustments within a few microns make this possible. The divisions of the micronuclei and their behavior have been seen and photographed in the living condition. The micronucleus of each conjugant is seen to leave its place near the macronucleus and increase in size. While this enlargement takes approximately four hours, the end of the division (including the anaphase and telophase stages) is more rapid requiring only about 18 minutes. The “crescent” prophase stage and long anaphasic-telophasic separation spindle so characteristic of this division are clearly shown. The swelling at the center of the separation spindle becomes separated and is passed into the cytoplasm. Cyclosis moves it about until it ultimately degenerates. Each product of the first division enters into the second where again two long spindles in each con- jugant are visible. The second division requires 50 minutes for completion from the time the first division products are formed. The anaphase and telophase stages are still more rapid, requiring only nine minutes. Degeneration of three of the four products of the second division is observed. Behavior of the pro- nuclei after the third division is being more carefully studied in an effort to record their movements accurately. The pronuclei of a given conjugant have been seen to fuse and form a synkaryon in the same individual. In a number of observations, definite evidence has been obtained to show that crossing-over of pronuclei does not occur. Too, no evidence has been obtained to show that crossing-over of pronuclei takes place. In view of these observations, any conclusions, genetical or otherwise, based on the assumption that there is crossing-over of pronuclei, are open to grave question. Photographs are being made of the living conjugants and will be included in the full paper. It is hoped also to make a motion picture of the entire process. Conjugation in Paramecium trichium Stokes (Protozoa, Ciliata) with special reference to the nuclear phenomena. Ralph Wichterman. A cytological study was made of conjugation in a race of Paramecium trichium Stokes. The preconjugants, which are smaller than the vegetative individuals, fuse along their oral grooves. The centrally located oval macro- nucleus undergoes complete fragmentation involving the formation of a twisted ribbon which becomes thinner and longer resulting in small irregular rod-like elements. These result in spherical bodies which disappear in the cytoplasm after exconjugant reorganization. The deeply staining oval-shaped micronucleus divides three times. The first pregamic division results in two micronuclear products; the second division, four products, three of which degenerate. The remaining micronuclear product enters into the third division to produce the pronuclei. The developing third pregamic spindle is commonly seen to press against the opposite conjugant where a cone is clearly visible. There is strong evidence to believe that the pronuclei cross approximately in the mid-region of the conjugants. This point is being investigated further by observing living conjugants under the micro- compressor. Pronuclei fuse to form a synkaryon which divides three times to produce eight products. Two of these divisions occur while the conjugants are together; the third division occurs in the exconjugant. Exconjugants with four 398 PROGRAM AND ABSTRACTS OF SCIENTIFIC PAPERS macronuclear anlagen and a single micronucleus are commonly seen. This re- duction in number from eight nuclear products of the synkaryon to five, is being studied further. Variations occur in regard to the distribution and fate of the eight nuclear products. Individuals with four macronuclear anlagen and one micronucleus divide with two anlagen going to each daughter while the micro- nucleus divides mitotically. In exconjugants with two anlagen and one micro- nucleus, a final division distributes a single macronuclear anlage to each daughter as the micronucleus again divides mitotically thus restoring the original nuclear condition to each animal. This work confirms Diller’s observations as reported in his SEES: in Anatomical Record, 1934. Mitotic activity of stimulated rat adrenals measured by colchicine tech- mique. Opal M. Wolf. In connection with a recent investigation it was noted that daily injections of an alkaline extract” of the anterior lobe of the pituitary into rats at 1 cc. and 2 cc. levels gave evidence of activity in the adrenals as measured by the colchicine technique. Control and injected animals were etherized the morning following the 2nd, 4th, 6th, and 10th days of injection. Nine to fifteen hours before the animals were killed, 0.3 mg. (per 100 grams of body weight) of colchicine ° dissolved in distilled water was injected sub-cutaneously.. At autopsy tissues were fixed in Bouin’s, dehydrated by the aniline oil-dioxan method, cut at 10 micra and stained in Delafield’s haematoxylin. Mitoses were counted with an oil im- mersion lens. Preliminary observations were made of every 20th section in order to esti- mate the effect of colchicine on the organs. A 171 gram control male showed the following total counts: sDhiy Olden areca eek cere 157 Parathyroid. c)ih scarce cere 213 TANGENT Meme ieee oat wish ww 625 (highest count for controls) An average count for the three largest sections of the adrenal, 45 in the con- trols, was used as a basis for comparison. Counts of both adrenals varied by 11-17 mitoses. Injected animals showed slight stimulation after 2 days, and greater stimulation after 4 days, but the differences were not marked for the two levels. The greatest stimulation occurred after 6 days of injection. One series of males weighing 170 grams gave: Control. oases 21 1h, CS HRB eRe ie che 90 CCN c/a se kh CR 232 (2149—total for every 20 sections) With one exception (126 mitoses on 2 cc. level), after 10 days injection the animals showed a count of 44. Nine to eleven hours after colchicine many of the cells were in early prophase, at 15 hours the greatest number were counted in the later stages. Most of the mitoses occurred in the outer part of the zona fasciculata, but a few cells in the zona glomerulosa and reticularis were dividing. No late stages of mitosis were observed in the medulla. In several instances very large nuclei appeared to be in early prophase. * The experimental work was performed at Goucher College, the sections were cut and studied at the Marine Biological Laboratory. * The growth extract was purchased from E. R. Squibb and Sons. * Colchicine U.S.P. was purchased from Merck and Co. in the form of a powder. PRESENTED AT MARINE BIOLOGICAL LABORATORY Oey The effects of the drug on the adrenal cells were similar to the descriptions by Ludford, Allen et al, and Nebel. Apparently normal stages of mitosis as well as stages leading to multi-nuclear cells were observed. The latter were probably caused by the varying concentration of the drug due to the rate of absorption and elimination. It was noted that the level of dosage had a toxic effect on some rats, others apparently showed no ill effects. ) Induced breeding reactions in isolated male frogs, Rana pipiens Schreber.1 Opal M. Wolf. Adult male frogs give a characteristic call, show ridged epidermal thumb- pads with large, active mucous glands on the first finger of the fore limbs, de- velop amplexus, normally in the presence of the female, and shed sperms at, the time of reproduction. In earlier work it was shown that the male as well as the female could be stimulated. by the injection of frog anterior lobe to repro- duce normally as early as the latter part of October and throughout the winter months. It seemed of interest to report experiments on individually isolated mature male Rana pipiens. In the first experiments injections of frog anterior lobes from Rana pipiens, an extract of whole pig pituitary, a purified extract of the gonad-stimulating fraction of horse anterior lobe* and an extract of pregnant mare’s serum, at times other than the breeding season and the summer, caused the shedding of sperm. Inspection of the cross-section of the testes showed empty seminiferous tubules and were comparable to the picture of the testes of frogs which were known to have fertilized eggs following pituitary stimulation. Following in- jection of the extract of pregnant mare’s serum sperm were recovered as they passed from the red and swollen cloacal opening and stained preparations were made. In some cases the characteristic call or trill was heard 8 to 9 hours following the injection of as small an amount as one frog anterior lobe. The call was not heard following the injection of the extracts. The thumb-pads showed active mucous glands but the effect on the epidermal ridges was not so clearly shown because the thumb-pad of the adult male is ridged during the winter season. Examination of the thumb-pads of adult males in the middle of July showed a smooth surface with slight indication of ridging and small mucous glands. A small piece from the thumb-pad of the experimental animals was removed as a control and the individually isolated animals were injected daily for 9 days with the anterior lobe from male Rana catesbiana. The injected animals gave the characteristic call of the breeding season, sperm were shed, active spermatogenesis was going on, the thumb-pads had developed ridges and the mucous glands were active. The thumb-pads of the experimental animals resembled those prepared from male Rana clamitans caught in the field and prepared immediately for histological examination. This animal breeds during the summer from June to the middle of August while Rana pipiens breeds from the first of April to the middle of May depending on the season and latitude. It appears from an examination of the testes of the isolated experimental animals killed at varying times after prolonged pituitary treatment that the mature sperm are shed and spermatogenesis induced by frog anterior lobe and the extracts used in these experiments. 1Part of the experimental work was performed at the University of Wis- consin Department of Zoology and part at the Marine Biological Laboratory, 1936. * The extract of horse anterior lobe, Prephysin A, prepared by Chappel Bros., and the other extracts prepared in Dr. Hisaw’s laboratory were obtained through the courtesy of Dr. Frederick L. Hisaw. Experimental Biology Monographs eA unique cumulative reference library for biologists, providing full accounts of both the experimental and the theoretical aspects of all significant current research work in the various biological sciences. NEW VOLUMES Autonomic Neuro-Effector Systems By Walter B. Cannon and Arturo Rosenblueth. The first complete summary of the nature of the chemi- cal mediation of nerve impulses. $4.00 Phytohormones By F. W. Went and Kenneth V. Thimann. An inclusive, critical survey of experi- mental work, covering experimental techniques and a- large number of new and unpublished experiments. $4.00 PUBLISHED AND FORTHCOMING VOLUMES PACEMAKERS IN RELATION TO ASPECTS OF BEHAVIOR By Hupson Hoaguanp. $3.00 NEUROEMBRYOLOGY, AN EXPERIMENTAL STUDY By Samuget R. DetwiterR. $3.75 THE EGGS OF MAMMALS By Grecory Pincus. $3.75 THE RECEPTOR PROCESS IN VISION By Srenig Hecut GEOTROPISM: A STUDY OF DETERMINISM IN BEHAVIOR By W. J. Crozier anp GREGORY PINCUS THE BIOLOGY AND CHEMISTRY OF OVARIAN HORMONES By Grorce W. CorNER AND Wituarp M. ALLEN BIOLOGICAL OXIDATIONS By E. 8. Guzman Barron THE HYPOPHYSIS By J. B. Couuie, D. L. THomson, anp H. SELYE NITROGEN METABOLISM IN ANIMALS By Henry Borsooxk TEMPERATURE CHARACTERISTICS By W. J. Crozizr CYTO-GENETICS AND PLANT PHYLOGENY By Ernest B. Bapcock BIOELECTRIC PHENOMENA IN PLANTS By L. R. Buinks PERIODICITY IN ANIMAL BEHAVIOR By T. J. B. Sripr THE GENETIC ANALYSIS OF SPECIES DIFFERENCES By T. H. DopzHansky DYNAMICS OF INFLAMMATION: AN INQUIRY INTO THE MECH- ANISM OF INFECTIOUS PROCESSES By Vary MENKIN The Macmillan Company 60 Fifth Avenue New York CONTENTS | | Page WELSH, J. H., F. A. CHACE, JR., and R. F. NUNNEMACHER | 185 The Diurnal Migration of Deep Water Animals........... COONFIELD, B. R. and A. GOLDIN The Problem of a Physiological Gradient in Mnemiopsis During Regeneration: 109. oat hs Wh eel ke he oa al GLASER, OTTO and GEORGE P. CHILD The Hexoctahedron and Growth. ..............000..-5.5. CARVER, GAIL L. Studies on Productivity and Fertility of Drosophila Mutants. BALL, ERIC G. and C. CHESTER STOCK The pH of Sea Water as Measured with the Glass Electrode GOLDSMITH, E. D. The Relation of Endocrine Feeding to Regeneration, Growth, and Egg Capsule Production in Planaria maculata......... PROSSER, C. LADD and JOHN Z. YOUNG ) Responses of Muscles of the Squid to Repeutive Stimulation of the Giant Nerve Fibers................. ieee ARNE SPARROW, F. K., JR- ! The Occurrence of Saprophytic Fungi in Marine Muds WHITAKER, D. M. 227 237 242 Determination of Polarity by Centrifuging Eggs of Fucus 7 PITCATIS ce Oe Re ee Tea a uM ep eT ne ee TYLER, ALBERT and W. D. HUMASON On the Energetics of Differentiation, VI. Comparison of the temperature coefficients of the respiratory rates of unfertilized and of fertilized eggs SL Tae PONTE Mette | age Sac otal, deeta fee KENK, ROMAN Sexual and Asexual Reproduction in Euplanaria tigris (Garay i) RAR TE LOC ISTHE yu Aah or a Lt da a HORSTADIUS, SVEN | Investigations as to the Localization of the Micramere=, Skeleton and Entoderm-forming Material in Unfertilized Egg Of ArbAatidiy eR) ee See i pal a he atten een ae HORSTA DIUS, SVEN i Experiments on Determination in the Early Peverpiaeu of GETEDFA tS HACTELLSS ee ee Oe ON Ti A eth Fk ac PROGRAM AND ABSTRACTS OF SCIENTIFIC MEETINGS, SUMMER OB LOR ee ee ORM e 21g 1c Pa eare MAS be al Pecalerk Soa 249 261 280 205, 317 Eee Oe Rae en ee ee ee a - Volume LXXIII 3 : Number 3 THE BIOLOGICAL BULLETIN | PUBLISHED BY THE. MARINE BIOLOGICAL LABORATORY Editorial Board GARY N. CALKINS, Columbia University E. E. JUST, Howard University E. G. CONKLIN, Princeton University FRANK R. LILLIE, University of Chicago E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden LEIGH HOADLEY, Harvard University T. H. MORGAN, California Institute of Technology M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University H. 8S. JENNINGS, Johns Hopkins University EDMUND B. WILSON, Columbia University ALFRED C. REDFIELD, Harvard University Managing Editor DECEMBER, 1937 | Printed and Issued by LANCASTER PRESS, Inc. PRINCE & LEMON STS. LANCASTER, PA. NEW BLAKISTON BOOKS The Microtomist’s Vade Mecum. 10th Edition Includes Both Animal and Plant Microscopic Anatomy and Technique 11 Illustrations. Appendices. 784 Pages. Washable Sturdite Covers. $9.00 In this new edition an important change has been made by the inclusion of plant technique. This section has been written by Dr. D. G. Caicheside of London University. There are new chapters on Frozen Section Staining; Vital Staining. Dr. Conn has recast the whole chapter on Staining. Dr. Duthie has done the same for the article on Blood and Glands. Other sec- tions which have been thoroughly revised are those on Celloidin Imbedding by Mr. Richardson; Fats by Drs. Kay and Whitehead; Protozoological Technique by Dr. Helen Pixell-Goodrich; Nervous System by Drs. Green- field and Stern. Dr. Theophilus Painter has contributed a masterly article on Animal Chro- mosome Technique. The authors of this edition are: Editors: J. Bronte Gatenby, M.A., Ph.D., Sc.D. GfTrinity College, Dublin), and Theophilus S. Painter, M.A., Ph.D. (Texas University). Contributors: D. G. Catcheside, Sc.D. (London Univ.), Harold J. Conn, Ph.B. (N. Y. Agric. Station), E. S. Duthie, M.Sc., M.B., Ph.D. (Sheffield Royal Hosp.), Helen Pixell-Goodrich, M.A., Sc.D. (Oxford Univ.), J. G. Greenfield, M.D., F.R.C.P. (Nat. Hosp. Lond.), W. W. Kay, M.Sc., M.B. (Manchester Univ.), Reginald: Ludferd, Ph.D., Se.D. (Cancer Research Lab. Lond.), K. C. Richardson, M.Sc. (Univ. College, Lond.), Ruby O. Stern, M.D. (St. Andrew’s Hosp. Northampton), Raymond Whitehead, M.Sc., M.D. (Manchester Univ.). Barnes— Textbook of General Physiology 166 Illustrations. 554 Pages. Washable Cloth Covers. $4.50 By T. Cunliffe Barnes, Sc.D. (Asst. Prof. of Biology, Yale Univ.) P. BLAKISTON’S SON & CO.,INC. ‘“Svindcinnia Pan Vol. LXXIII, No. 3 December, 1937 THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY EFFECT OF SALTS OF HEAVY METALS ON DEVELOPMENT OF THE SEA. URCHIN, ARBACIA PUNCTULATA A. J. WATERMAN (From the Marine Biological Laboratory, Woods Hole, and the Thompson Biology Laboratory, Williams College) I In previous experiments an examination has been made of the effects on gastrulation and embryonic differentiation in Arbacia of ether, changes in temperature, alcohols, X-rays, hypertonic and hypotonic sea water, and of the disturbance of the salt balance in sea water. Disturbance of the salt balance of sea water and the action of certain alkaline earth metals etc., have been tested on the sea urchin (Paracenirotus lividus) by Runnstrém (1935), Lindahl (1936), Lindahl and Stordal (1937) and others in relation to gastrulation, ectodermization and endodermization of the embryo and specifically to differentiation. References to other workers have been given in earlier papers. These studies have been concerned with the un- fertilized and fertilized egg rather than with the blastula stage. At gastrulation physical movements take place that result in an invaginated endodermal tube from which arises much of the meso- dermal material. These movements may be delayed temporarily, retarded or inhibited, and they also may be influenced independently of differentiation. In the latter case an exogastrula may or may not form. What initiates the movements is another problem, but at least it has been shown that a host of factors, producing quite similar effects, can interfere with gastrulation; disturb the proportions of potential ectoderm and endoderm; and influence the differentiation of skeleton, body form, ectoderm and endoderm. As regards embryonic differentiation, the effect has been in general very much the same for these various environmental modifications. Some types of differentiation, taking place as development proceeds, can be slowed down to a greater extent than others. Ectoderm, and to a lesser extent mesoderm, may grow and differentiate when endo- derm formation is entirely inhibited. Furthermore, the ectoderm is 401 402 A. J. WATERMAN the most resistant to environmental modifications. Different cultures may vary under apparently identical conditions; cultures from one female may vary from those of another; while individuals in the same culture always vary in their susceptibility. The exact condition of early development, i.e. fertilized egg to early gastrulation, seems of minor importance, since the results are somewhat similar under the conditions of the experiments. Some metallic salts, which are a natural constituent of sea water, are more toxic to animal life than are others. On the whole, it is probable that the concentration of a particular metal changes very slightly throughout the ocean. The possible exception may be where large bodies of water empty into the ocean causing dilutions of certain constituents of sea water and concentrations of others, especially metals. What concentrations can the developmental stage of a littoral animal such as the sea urchin, Arbacia punctulata, with- stand and what is the effect of increased concentration of different metals on such a developmental process as gastrulation? Finally, are the effects on development comparable to those secured by other experimental methods? II In this study ZnCl, ZnSO., Zn(C2H3O02)2, FeCl, PbCle, CuCl, HegCle, AlCls3, NiCle and CdCle have been used in various concentra- tions. As in previous studies, the blastula stage was selected for study and exposure was made for long periods of time. Reference to Tables I-IV will show the approximate concentra- tions which produced certain results. The concentrations are ex- pressed as proportions and represent the amount of the metallic salt which was added to the sea water. No estimate was made of the normal amount of the metal in sea water. As nearly equal numbers of embryos as possible, without actually counting them, were used in each culture. By mixing the embryos secured from several females more or less similar cultures were secured. All results were based on random sampling from three places in the culture. The exposure time was varied, and in certain cases where developmental arrest had occurred without death, the culture was washed and returned to fresh sea water for recovery. In some instances samples were trans- ferred at intervals from a lethal solution to fresh sea water to study the rate of the toxic action and the recovery ability of the embryos. It is said that many of the heavy metal salts are precipitated by sea water. The concentrations specified in this study may represent initial rather than final concentrations and some of them may have changed in the course of the experiment. At this point it may be EFFECT OF HEAVY METALS ON DEVELOPMENT 403 said that those experiments were discarded in which visible precipita- tion had occurred. The whole subject of heavy metal action is obscure. This study was made at room temperature during the months of July and early August. Hence there was some variation as some of the experiments ran for several days. To avoid repetition only the results secured from the exposure to ZnCle will be described, but comparisons with the other metals will be made. The data for the others are included in the tables. Ill A survey of the more recent literature has revealed numerous references to the effects of heavy metal salts on animal development and embryonic differentiation. The effect on gastrulation has ap- parently not been tested. Hammett and Wallace (1928) found that the lead ion retarded growth, and differential development of the head and optic regions was markedly inhibited in chick embryos. Child (1929) has used CuSO, in the study of physiological gradients in the chick embryo. Féré (1893) obtained monsters following the injection of lead nitrate into developing chick embryos (see also Franke, 1936, for selenium salts). Galtsoff (1932) has shown that marine invertebrate animals can concentrate different metallic elements in their bodies. Certain groups concentrate zinc, others copper, etc. Copper salts affect oyster larve by inducing attachment (see also Prytherch, 1931), and by initiating metamorphosis. Metallic silver causes sperm of the sea urchin to lose their fertilizing power, and paralyses plutei (Drzewina and Bohn, 1926). Hoadley (1923) has studied the effects of the salts of the heavy metals on the fertilization reaction in the sea urchin, Arbacia. The inhibiting concentration varied for the different metals tested. Gold chloride was most toxic for membrane elevation and cleavage and cadmium or cobalt chloride least. Other salts tested included CuCl, ZnClz, LaCls3, AICl3, PtCls, PbCl, NiCl, in the order of the toxicity. HgCl, differed from the others in that membranes elevated at a con- centration of 1 part in 600,000, which was toxic for cleavage. Con- centration of these metals varied slightly for different batches of eggs and thus showed the influence of a time factor. HgCl, has a harmful effect on cleavage at different concentrations following very short exposures (Hoadley, 1930). It affects the cortical region resulting in membrane elevation. After longer exposures it affects the pigment which has mercury-avid properties. The pigment 404 A. J. WATERMAN may be extruded and such an egg may develop if not injured. A con- ceivable mechanism is thus available by which the mercuric ion, which has entered the egg, may be bound and removed. Copper salts are known to have an injurious effect on many | organisms. In a very low concentration which inhibits fertilization, sperm may be still active. Inhibition is marked in a concentration as low as 1 part in 2,500,000 parts of sea water and is reversible provided the eggs are not injured too much. Copper appears to injure the vitality of eggs and acts as a slow poison (Lillie, 1921). A concentra- tion of 1/62,500 is necessary to suppress cleavage. The effect of HgCl, is different from that of copper: initial stages of fertilization are little affected, susceptibility increases as fertilization progresses, fertilized eggs show the effect more rapidly than unfertilized eggs, and the movement of sperm is suppressed at great dilutions. It produces membrane elevation alone and favors it in fertilization. The effects are reversible if exposure is not continued too long. Lillie concludes from these studies that the effect of mercury and copper on fertilization following membrane formation may be due to enzyme poisoning. The effect on the initial stages of the fertilization reaction does not correspond so well to the enzyme analogy. _ Glaser (1923) shows that in the egg of Arbacia copper becomes concentrated in the chorion, vitelline membrane and cortex. It is diffused throughout the cytoplasm. Copper occurs normally in egg pigment, membrane chorion and cortex associated with lipolysin. Normal eggs secrete copper compounds as well as removing copper sulphate from sea water. Parker (1924) has shown that marine animals will grow upon any heavy metal plate provided the metal does not liberate ions or soluble compounds. ‘The effect of CuCl, MnCh, and FeCl; upon cardiac explants cultured in vitro has been studied by Hetherington and Shipp (1935). The survival time was tested. Other interesting studies could be given but the above are sufficient to indicate the type of work which has been done. IV NiCl,.—In an experiment to test the effect of this metallic chloride upon the development and differentiation of the germ layers after their formation had been initiated, oval blastule to early gastrulz were exposed to various dilutions of a 1 per cent stock solution made up in sea water. The cultures were examined after 21-42 hours of exposure and in some instances the embryos were returned to fresh sea water. NiCl, gave the best and most numerous examples of exogastrulation of all the metallic salts which were tried. Further- EFFECT OF HEAVY METALS ON DEVELOPMENT 405 more, the exogastrulz underwent further differentiation. The lowest concentrations employed gave also marked inhibition of development. After 21 hours exposure in a concentration of 1 part in 60,000, development had only progressed to a large triangular stage with skeletal spicules or short rods and initial flap formation. A few simple exogastrulz with tri-radiate spicules were seen in the culture. During the same length of time the control embryos had advanced to a medium pluteus stage. A 42-hour exposure killed most of the experimental embryos although a few large blastulae and triangular stages showed slow movement. In concentrations of 1/40,000, 1/20,000, 1/14,285, development was progressively inhibited but the number of exogastrule increased markedly. The embryos were all on the bottom of the culture dish. Both the evaginated and invaginated endodermal tubes failed to differentiate although some growth occurred, while the mesenchyme either formed small spicules or failed to show any skeleton-forming activity. The ectoderm was the least affected and continued growth. It acquired its characteristic appearance and the apical plate appeared. In such cases where the formation of the endoderm was entirely inhibited, large circular or oval ectoblastule were seen. A longer exposure killed the embryos, while those which had been transferred to fresh sea water showed increased activity but only slightly ad- vanced development, especially in the case of the ectoderm and skeleton. The endoderm of these embryos showed no further change. The ectoderm in the region of the vegetal pole tended to be more or less irregular and the shape of the body was often lumpy. When transferred to fresh sea water after 21 hours exposure, the embryos either died or failed for the most part to continue develop- ment. The activity of the survivors increased. Blastule and exo- gastrule with spicules, stomodeum and apical plate were seen. Some of the survivors showed no formation of endoderm while others contained a simple endodermal tube but no skeleton. Thus the embryos were unable to overcome in sea water the poisonous effects of the metal. It acts in very low concentrations (Table I) and within a relatively short time. The effects are not reversible and embryos in a culture react differently. The various abnormal types seen in the cultures are the result of differential susceptibility among the embryos and also in the developmental processes involved. This has been a common observation in previous studies of gastrulation in the embryos of Arbacia. The number of exogastrule decreased in concentrations from 1/10,000 to 1/6,000. Differentiation was further inhibited while A. J. WATERMAN 406 “asueyo ON "DAT IV *asueyo oN “OAT? IV LS8‘Z/T *SOxO *saynoids MaJ ‘sayndids| ON ‘soxa YUM Jepnsue |Jomoaq “wn -11} 0} wpnyserg | -sejq [[euIS “SOX9 *sOxe | 1YSI[S Aue MJ ii (o) a ‘x[nayses -Eynajses ON | = [eIUL May 000'S/T 999'9/T *sOxa *x[N14se3 [e1} | "sOxO MIJ Auey ‘sojn | -1ul Maj ‘soxa| *uWI07}30q “aA0qe oY} Ul sayn | -oIds YIM May | a10][ ‘Sie[NOIs | UO Aso] -o1dg ‘soxe Auey] |‘e[n1jse3 a]duris -0A xINIselg *Jo][eus *x[NJse]q [[eUIS 0} e[NAse[g | *WO}30q UO ][V |JeYMoWIOS w]NI4seyy [EIU] 0} wjnj4se[q 93e7 *UOTJEULIOJ UWepopus Iepnse1 *SOXO -1I Io ‘soxe aid a]duiis a31e] -WIS Josi1e] Aue Md ‘“JaMOysS ‘uO}IIOyS | wnzWses ‘Uu0l}e[N.AYses 3e JUSUIZAOPY | ON ‘xNsWse3 94] 02 sjdu19}7e Jamey | ‘x[nqselq a10y{ | 97e] 0} w[nNAsepgq | wmIpeyy x[Nqselg [eotteayds 07 [eAC 000'01/T S8'PT/T 000'02/T 000°0#/T *SOx9 MOT “SaTpITeUr -I0UGE BUIOS “U0J9]94S YA Je[NSsuels as.1e7 sgjnoids YIM siejnsue -I1} 0} w[n41}se5 000‘09/T Toqnjd wUNIpeTl 1aynjd UINIpsUl 07 wznijsey Jorqu0g ‘UOI}VEIJUSJEHIp MuUOAIqUIa PUL UOT}E[NAYses UO Z{DIN JO SUOT}eIQUaDUOD JUSIAYIP Jo poy” [| A14vL EFFECT OF HEAVY METALS ON DEVELOPMENT 407 growth and movement became slower. The ectoderm was the most resistant. In a concentration of 1 part in 5,000, there occurred some increase in size but no development. After 21 hours in a 1/2,000 concentration there were few intact individuals left which had not progressed beyond the state at which they were placed in the solutions. In an effort to see how quickly the embryos would be affected, samples from a series of different concentrations were transferred at certain intervals to fresh sea water. These were examined about 24 hours later. The results are summarized in Tables II and III. They show that a relatively strong concentration. of NiCl, which inhibits any further development in the experimental solution acts very quickly upon the embryo. For example, 6-7 hours exposure to a 1/2,000 concentration was fatal to further differentiation but the embryos lived and increased in size (growth) in fresh sea water. In the tables, comparative data are given for concentrations ranging from 1/10,000 to 1/1,666. In a 1/2,000 concentration endo- derm formation was entirely inhibited following an exposure of about 6 hours. In a 1/10,000 concentration the gastrulation processes first showed inhibitory effects after 7 hours in some individuals of a culture. Up to this time, the effect had been upon differentiation and growth. In several repetitions of this experiment, a peculiar result was seen. After 3-4 hours exposure followed by transfer to fresh sea water for 18-19 hours, quite normal appearing embryos were found. Longer exposure times likewise gave more active and better differentiated embryos than did the shorter exposures. This result is surprising since it was expected that continued exposure would eventually result in death for all individuals. Is it possible that over a long exposure period the embryos lost their sensitivity to the toxic effect of the salt and were able to differentiate normally or that the salt was taken out of solution in some way? This result was observed in three out of the five repetitions. In a 1/5,000 concentration development was progressively inhibited up to about 43 hours while after 63 hours there was a marked advance in differentiation. In a 1/2,857 concentration this ‘‘pick up” was ob- served after 44 hours exposure. In a 1/2,000 concentration the improvement was noticed in cultures exposed to the salt for 3¢ hours or longer. In other experiments, Tables II and III, the above phenomenon was not observed. These show a progressive depression of develop- ment in the higher concentrations. In Table III it can be seen that the capacity to gastrulate was destroyed in a 1/5,000 concentration after about 8 hours exposure, and no recovery was made in fresh sea A. J. WATERMAN *3n3 OU SIe[NoIsa/A, "WI0}}0q UO [|e JsOW]y “Apoq 3404S *sajnords uojajeys ‘3n3 YUM Sie[NSuelI} VIO[\ *‘soAIqUIO Ie[NdISaA ][euls pue o9.1e] ore Iso *saseo awios ul Adwin] wisp -0104 ‘“Sajnoids pue 3n3 YIM wnaysey ‘iegnyid suno< 0} dn sad4} [|v *4ns sAey MO ‘SIe[NIISO/A *4n3 ou YM IP[NISIA 31 QWIOS ‘soynords yIIM Jen suels 1e[ndoIt1] yns YIM SUIIOJ JE[NIISa/ *]eulIOU -Ge Ud}JO U0}2]94S *“sajnoids pue ns YIM wnIYWse3 pue siejnsuers ][eWs Moy ‘uoTs -a1 ajod ]e1930A Je[Nsei11 10 pause Shey MENUS ae ‘soAIquis 1e[nN -JISaA ‘adie] AUeT “suse peseds Ajeprm pue Apog 3104s YIM t1agnjd sunoA 07 dn sedAQ [TV 9 ‘aseis Iejnsuelsy dy} payoval savy J9}}2] VY} JO Jama ‘SNISes JOM] pue x[nyseI[q s10;\[ uO}I[a4s [eulIouqe YIM iegnjd sunoA 0} SUIIO} pozUNIS ‘ent -Ses0x9 o]duIIs Ma, ‘seynords Yum x[N1} -ses Jejnsuelsy 0} e[NIAse[q jeoreyds osie7T *Jasu0] SUIe pue 1931e] Apog ‘uo1 -epiejai sso] v OPeUl SEA OSULYD VOY S[eAIOIUI INOF{ “Ie[nsat -4I U9}JO U0}IJB4S ‘ans yyIM soAiquia Je[NSsuel.y od1e7 *sdey wie [elit yim ouos 0} U0}JI]OYs pue yns YIM Ie]NISaA Os[y “UOT}epse}a1 BUIOG € “repn3uers ‘asieqT | 00s‘z/T “sdey wie JO suIIe JOYS YIM awio0g ‘1eynjd Bunox | e¢e'e/T JO1}U09 O¥X1T | 000‘OT/T wornjos *ase1S 408 ' *709]J9 DIXO} JO vaIZap pue 9301 ay} MOUS e}ep IT, snajnjd WNIpsul & }e SAA [O1]UOD 9Y} 9UIT] SITY JY JOP] SINOY G[-Q] Poulurexy “319}eM Las YSoIJ 0} S[eAIOJUI }e paAoutel sajdures pue uoljnjos ur posed xjnysejq ‘eutt) Jo sported polseA JO} 2{DIN Jo suoI}e1]UBDUOD JUSIEIpP 0 e[Nyse][q Jo simsodxa jo WeyyA Il S1avL 409 EFFECT OF HEAVY METALS ON DEVELOPMENT wo1}e13 -9}UISIP paMmoys IUIOG *e[Nyses -oxo podeys Ajiepnsei1 pue pazejUusIoyIpun Mj VY ‘ouljjno ul rejnso1 -II WO1de1 ojod [eIad0/ ‘4nd [eUtoJUI ON “YSsIp jO W10}30q UO [Je jsO;, e]ngseiq [IV "sary -IJe[Nseiit [eusto} -X9 1O ‘S][99 Jo sseur 10 ‘aqn} Wiepop -u9d OU MOYS ISO *Ie[NIIsaA Ae] -nseii soAIqUIy “UOT}e[N14SeS [eIzIUI pemoys Moy Aro/ ‘zxnyseiq AjIsoy “xn I4Se3 JaMa,J “UOJIPIYS ON *QUI]INO UT IejNse1 -1I pue Iejnsisa/, AS UES ACES ING 9 S sain -o1ds ou ‘e[n13 -ses UINIpeul 0} zjnyseIg *sdey UII [eIQIUI ON "sojniods |[eus ‘xen I4Sse3 Te[NSIsvA 910Ul pue Jayyeus v SPEUI SBM OsUeYD VOYM S[eAIOIUT IMOF{ ‘w2][N.1QSes0xe o[durIs Moy ‘Sopnords 4AM X[NIses repnsuersy 0} KINASE] [[eUIS sojnords yAIM X[NI4ses 99] [[eUIS ‘uoT}eWI10; dey-urse [eIUI pamoys VUIOG ‘sojnoids |[ewis pue asiey “Je[NoISaA IOJ, “JOT[CUIS Beis | *paueq -yey Apog ‘Auvur UI pessel UIIEpo} “uor}eUlIoy dey -WIe [eI}IUI pamoys auI0S ‘saynords Je[nse111 10 [euliou pue yn3 YIM soA1q “Wd Ie[NSuUeII} as1e7] 999'T/T 000°2/T wornjos PenuyuUoy)—_j{] ATaV LL A. J. WATERMAN 410 wz[nqseiq [eAo 03 [eortayds Je[NIISaA IIOJ\[ [01}U09 ayxTT ct “Z] —nijses eI} “SOX9 Md, “soznords do} uo /y}1M Jepnsuery JOMaT “aules 0} JeNdIsa/\, yee) -njd sunoé eee wI0}}0q uO [l@ 2[nYses Jepnsuersy 0} xN\selq eee‘e/T x[nyseyq | 2[nysejq aie YSOJ | 912 JSOT[ “sain -o1ds YAM Ie] —NSUPII} MJ ‘xen IYWse3 pue 1e[noIsoA 0} x[NAselq auou Jo sojnaids UOT} epsejOI 2IOW IUIOG u0j2]94S YIM 1e[ns -UPII} [[eWIsS 0} e[Nz4se3 Jen se11] quoul -9AOU 97} “HWY = ~eynay Ses [CIIUI 07} eNjse[g aures UO0J9[OyS YIM Je[Nsuel} 0} z[NI4ser) -IUT 0} xn SIE]NISAA -sejq ‘do OBIE] 0} X] DUES uo 9UON | -Ny4se]q s107\, "u049] XS ONG *x[N4] | -nyse[q s107q “ses JoMoy “QUITINO UT “Iejnoisaa |Aduin] pue 1e] Jeorseyds | [eorteyds jaroy, “WOR |-nSerzy =" W077 -10q UO |[V | -30q uO JsoTJ sojnords Taynjd | epnsqses 07 YIM 1eins poor | sze[noisa, | -uerI} [[eUIS 6 EL 9 {01 mie v JO}EM LOS YSolj 0} OPEUI SEAA OSULYO USYM S[eAIOzUT INOFT Toynjd [[ews 09 Jen suersy, $]0.1]U09 SPN en fa UOI}NIOS Ut UT 3J9] eSOyL, 000'S/T 000‘OT/T wolnjos *Ja}e] SINOY FZ JNoGe pauturexs soinj4jnd vsey J, ul ynd xjnjsejq [eAo 03 jeol1ayds *{99]J9 OIXO} 9Y} JO d1Sep puv 9301 ay} MOYsS e}ep ay, “1a3N,[d pamoys jo1}U0>7 II] adv LL "Ja Jed1IY} S[PAJOIUI 72 JoJeM IS YSIIJ 0} PSAOUIII UsYy} PUe OUT UTe}I00 & Ye SUOTINIOS ‘aUII} JO spotted palizeA JOJ *{DIN JO suor}eI1}UBNUOD JUdIBYIp 0} wjnyse[q Jo sinsodxe jo Wey” 411 EFFECT OF HEAVY METALS ON DEVELOPMENT aules qynoqy QUIS QUIES x[nqselgq xz[n1} “ses YSIS aules | 0} e[Nyse[q “UOIJEUISEAUI [Reg es aese al ‘x[NIselg ‘gyn -IdsON “&] -NIjSes 99] 0} w[NIse[g uO} eUl -SeAUL JY SIIS JO x[N}se|q dIOJ, “WOR OC LLOn TIN, eco) al -€139}UISIp dUIOG “WIO} -J0q uO ]TV *u0}}0q UO TV “9utes auIesS "u0}9]94S ON ‘9ules *sajnoids YUM xn} -Se3 9}P] 0} &] -nysejq “uwloy -jJoq uo Aueyy “u0y Jays YUM Jejnsuelsy 0} 2[Nyselq *W10930q uO }SOJ u0}29]94S YIM Jens -UPII} 99°] 0} 2[Nyselq sojn -o1ds yyIM re[nsue -11} 0} e[N1 -SeS 9307] *u01}e.139) -UISIP 9UIOS ‘xNAselgq xz[nqselq wzny -SE]q [PAO 03 jeotrayds 999'T/T 000‘2/T 00S‘2/T sodeys SNOLIGA JO aues | xniselg *W01}2.139]UISIP DUIOG ‘aes aUIeS aues | xniselg rat #01 JO}EM aS YSolj 0} OPEUI SEM OSULYD USA S[eAIOIUT INOF, panuyuoj—{] ATAv uOlyNjos ur eT SsOuL WOINjoS A. J. WATERMAN 412 *S][20 Jeoleyds jo} ‘worje1s9qUISIpP E2jbal -ses sunoA wx[niqses WINIpour Livohls) E>: (5) {NOYIIM siTe[nsueiy Sossep, “peesq|yonyL ‘“x[nIselg 0} enysel[g| 0} x2[nyselq 0} E[nayses Apieq 000‘000'S/T 000'000'8/T 000‘000‘9T/T |000°000‘ZE/T 000‘000‘84/T “sul -JsIsiod soueiq *ai10d XI “Ul 2] : -O}SE[G os1e7T| -n1ses [eI -nijses sunoAé lojnjd sunoA uonepie, ‘oures Jnoqy 0} wlnyselg aUlOS 0} wnyselg ejnaqse5) 0} x[N14se5) -o1 WSIS [O1}U09 OyIT “peep BUOS “XI -Nijses SunoA uory aules ynoqy 0} xinIselg aUleS w]Niqse45 SIe[NSUeI | -eprejor JUSS] [01}U0D oxTT [O1}U0D oyIT “41S -1od soueiquioul *uo1}e[NI1} E2jabal *uI0}30q UO “JUS sie[ns Aqdwuq ‘xT -Ses [CIUT -Ses sUNOA!| JSsO;, ‘“x[nI -9AOU I9MOT[S -UCII} Vd1P] -nysejq Asopy |’ 2[nyselq s1077 0} X[Nysejq| -Ses UMIpeyy| “ewes ay} YON] x[nzWses 97e7]] 0} L[n1se5 [01}U09 oy] *110}}0q uO = sO. ‘“Soueiq *U10}0q *WI0}}0q UO} -WoUl UL [[Ss\ul Aue; *3UTqSIS “u01je13 IlV ‘Seueiq) Aueyy ‘x[n13) -19d souviquiour *soynoIdsS “HONCIZIIUISIP] -d}UISIP BWIOS|-UoU ul AUST, -SPS8 [III] BWIOG *xX[NIWsed w[nijses| ‘srepnsueisy eo, ‘“aues| ‘A[UO x[NIse[g)| “A[UO ~z[nyse[g 0} L[nyse[q] SuUNoA 0} [eI}IUT] 923e] 0} BuNOA] 07 K[NIWSe5 JO13U09 oy] 000°‘002/T 000‘0SZ/T 000‘00¢/T 000°00F/T 000'00S/T 000‘008/T 000°009'T/T 000'002'‘¢/T [O19U09 94} WOI} sdUSTOIp ou AyjeonoeIg e[niy wnt} “Sed IoMoj pue] -ses WINIpew wz[ni4se3 siepnsueny yo19 x[Nqse[q sop, 0} L[NIse[g] 93e] 0} sunoy, 0} L[N14Sse4)|UOTJeprejoeI YSIS JO1}U09 ayI7T -U09 o¥IT 000‘0S/T 000‘09/T 999'99/T EES‘ Es/T 000‘00T/T 0S8'CPT/T 000'00Z/t | 000‘00#/T *Sopnords sojn ON ‘siey] -o1ds 471M “WO1}e139}UL “4sIsiod wnt} -nsueldy sie[ns -SIq ‘UWI03}0q SoueIquIsUI “SIOUIIMS w[ngse[q UOe[N14Ses| -Ses UINIPsU|/x[n414ses 9}e]}] Ajies 03) -UeIIZ 07 UI [JV “eules) suIOG ‘“aUIeS| May ‘“eWIeS} 9}e] 07 SUNOA| [eWIUI [eUOISeDDO 0} w[nIselg 0} Ssuno,| w[N1WseyH} wN4IWses *s][29 *e[niqse3 “SIOUIWIIMS doz JO SeSSeUI [eDI] [eIIUI [euOIs MOT “xL[NI17 x[nijse3 Teyny{d uoneis -IaydS ‘uone]] -e20Q ‘xn we[n14Sse3 -Sses [CI}IUT Ooze] AIOA aures| SuNOoA 07 -9}UISIP BUIOS eules] -e[Nniyses ON| -selq ATIsOyT [equ Tome 0} X[Nyse[q| 0} x[nIse[g qnoqy| e[naqses 000°S/T 000'9/T 999'9/T eee's/T 000'0T/T C8e PT/T 000'02/T 000'0F/T | 999‘99/T “sIq ct “SI EET “Siu FET ainsod -xo Jo yjsueT To3njd sunoj, Toynj{d suno, jeynyjd suno x, Teqnjd suno jx, leqynjd sunoj, Toyn[d suno jx Toyn{d suno x Teqnyd sunoj reqnyid sunoj, yor -u09 %10°0 73H 2(@O&H2D)UZ %1'0 %1'0 (‘OSUZ) 10 OLYA %10 Ono %V0 JOe4 VO 104d %O't 40) 30) %O't TOV Te} "Ia}eM aS UI JUasaId Apeai[e VSOY} 0} S}][eS OI|[eJOUI JUSJEYIp Jo sJUNOWe snoliea Jo uOIIIPpe ay} Jo JusuIdOjeAep uodn yayq Al 314V.L EFFECT OF HEAVY METALS ON DEVELOPMENT 413 water. The length of time necessary to destroy the ability to gastru- late varies according to the concentration employed; Tables II and III. The results summarized in the tables represent single experiments, but they are indicative of what was secured by several repetitions of each. Al,Clg.—In dilutions of a 1 per cent stock solution of Al,Cle, gastrulation in all individuals was practically inhibited at a 1/8,333 concentration during an exposure of 133 hours. Table IV shows that the first indication of the inhibition of gastrulation in some of the individuals was observed at a concentration of 1/20,000. Further inhibition of more individuals occurred in higher concentrations until at a 1/6,666 concentration no gastrulation took place during the experimental period. Samples transferred to fresh sea water from concentrations between 1/14,285 to 1/8,333 after this interval re- covered. Gastrulation and development to the pluteus stage took place. Abnormalities of arm and skeleton formation appeared in the sample from the 1/8,333 culture. Samples transferred to sea water from the 1/6,666 culture died. To test the rapidity of the toxic effect, samples were transferred from a 1/6,666 concentration, at certain intervals, to sea water. Those transferred after $ to 23 hours of exposure gave plutei. The longer exposures caused progressive retardation, and more gastrule and fewer plutei were found for the same length of time. After 5 hours exposure, a few gastrulated but most died in the sea water; after 63 hours exposure, all died without any development. When the toxic effect of a 1/5,000 concentration was tested, an exposure of $ hour markedly retarded development, 1 hour inhibited almost all gastrulation and 24 hours killed the embryos. CdCl.—Table IV includes the effect of this metallic chloride in different concentrations. Over a 133 hour period, gastrulation was inhibited in a 1/8,333 concentration. In sea water, these blastule grew in size, became irregular in outline, but did not gastrulate or form skeleton. Samples from a 1/10,000 concentration formed large, vesicular, irregular blastule and gastrule in sea water. PbCl,.—Lead chloride is slightly more toxic than the three metallic chlorides just mentioned, and hence a 0.1 per cent stock solution was used. Table IV includes part of the data secured in one experiment which may be considered typical. Slight retardation became evident in a concentration of 1 part in 100,000. In aconcentration of 1/60,000 gastrulation was inhibited in some individuals. The latter increased in number in a concentration of 1/50,000 after which little change was noted in the cultures until a concentration of 1/35,714 was reached. 414 A. J. WATERMAN One result was the persistence of the fertilization membranes even after the embryos had escaped. Thus this salt causes the membranes to harden and prevents their dissolution during the experimental period. In the 1/35,714 solution late gastrula to triangular stages were found, while in a 1/33,333 concentration even young plutei occurred. This condition was found in all cultures up to a concentration of 1/22,222. No higher concentrations were tried. This peculiar change may perhaps be accounted for by the precipitation of lead seen in the bottom of the container, during the experimental period. The precipitation did not appear when the experimental solutions were made up, but only after standing. Enough was left in solution to retard development slightly. FeCl,.—This metallic chloride is apparently not toxic to any ex- tent. Very little if any difference was observed in graded cultures up to a concentration of 1 part in 14,285. CuClh.—This metallic salt is far more toxic for development than PbCle. A 0.1 per cent stock solution was used. Table IV summarizes the results of a typical experiment. Gastrulation was inhibited in most individuals in a concentration of one part in 400,000 and entirely in a 1/300,000 concentration. The fertilization membranes showed a hardening in these concentrations and failed to dissolve. This must occur rather quickly because in these and higher concentrations more of the embryos failed to escape from the membranes. The same persistence of the membranes was also seen in the PbCl, solutions but to a lesser extent. In the latter case the embryos had escaped so the hardening occurred more slowly. This lack of digestion of the membranes in both the PbCl. and CuCl, solutions may be accounted for by enzyme-poisoning which is characteristic of some of the heavy metal salts. As will be seen below, the higher concentrations of the Zn salts likewise produced persistent membranes. After transfer to fresh sea water, the original 1/500,000 culture continued development to the pluteus stage. Also large vesicular blastulz without skeleton were found, indicating growth of the ecto- derm but inhibition of endoderm and mesoderm formation. Most of the individuals in the 1/400,000 concentration gastrulated in fresh sea water and in 24 hours formed triangular embryos with spicules. More large blastulee were present as well as a few exogastrule showing tri-partite gut, anal opening, and apical plate. Thus in some indi- viduals this metallic salt inhibits the gastrulation processes without inhibiting endoderm formation. As seen in Table IV, no gastrulation occurred in a 1/300,000 EFFECT OF HEAVY METALS ON DEVELOPMENT 415 concentration during the experimental period. However, in sea water some did gastrulate and a very few formed simple exogastrule. No skeleton was seen in any individual, indicating the inhibition of mesoderm formation or the formation of skeleton during the period of observation. The endoderm did not differentiate. Only a few individuals from a 1/250,000 concentration recovered in sea water. Large, spherical and oval blastulz were found as well as a few initial gastrule. The individuals from a 1/200,000 culture failed to gastrulate even after 48 hours; many died; and many failed to get out of the membranes. The time required for recovery varies of course with the concentra- tion. Embryos which did not gastrulate or in which no skeleton formed during a 24-hour period in sea water, did so after 48 hours. Since it is known that certain marine organisms may store copper in their tissues (oysters) this inhibition may be due to the concentration of the toxic metal, and the slowness of recovery and differentiation be due to the slowness with which the metal left the cells and again reéntered the sea water in such a dilute concentration as not to inhibit further development. The metal could not have irreversibly injured the protoplasm, but rather inhibited its function. Angerer (1935) has found that after Arbacia eggs were exposed to CuCl, an interval of time elapsed during which the metal produced no visible effect in the protoplasm as regards its viscosity. This time interval is a function of the concentration of copper in solution. At the conclusion of this latent period, there ensues a sharp rise in viscosity values resulting in an irreversible gelation (coagulation) of the protoplasm. In the case of Arbacia egg protoplasm, there is no concentration at which gelation is reversible. In the light of Angerer’s results the temporary inhibition of differentiation cannot have been due to gelation, otherwise recovery would not have occurred. An irreversible effect was seen first in the concentration of 1/200,000 where no recovery occurred even during 48 hours in sea water. To test the rate of lethal action of CuCl, a concentration of 1 part of stock solution in 150,000 parts sea water was used, which inhibits all development and causes complete disintegration in an 18-hour period of exposure. Embryos were removed from this solution to fresh sea water at various intervals. An exposure of 3} hours resulted in some inhibition and many persistent membranes from which numerous embryos had failed to escape. After 94 hours in the solution only a few embryos gastrulated during the following 28 hours in sea water. Those which did not escape from the membranes, or were only 416 A. J. WATERMAN partially successful in it, did not differentiate. A concentration of 1/250,000 produced no inhibitory effects during 8 hours exposure. Membranes hardened while the embryos which did not escape differ- entiated slightly. HgCl,.—This metallic chloride is very toxic. A concentration of 1 part in 48,000,000 parts of sea water retarded development while a concentration of 1/8,000,000 prevented gastrulation during a 15-hour exposure (Table IV). After return to sea water, only very few of the more hardy individuals were able to gastrulate, but they developed no further. In a concentration of 1/5,000,000, the swollen cells showed a clumping of the pigment granules. To test the rate of action of HgCl:, spherical to oval blastulez were placed in a concentration of 1/2,500,000. The embryos transferred to sea water after 5 minutes exposure were able to recover with only slight evidences of retardation in 24 hours. A 10-minute ex- posure visibly retarded development. More gastrula and triangular stages were found, while the control showed medium plutei. After 15-minute exposure, fewer individuals were able to gastrulate. These decreased in numbers following exposures of 20 to 45 minutes. During this interval more of the embryos were killed and movement became progressively slower. An exposure of 55 minutes killed most of the embryos, though a few survivors showed attempts to gastrulate. This metallic salt does not appear to affect the gastrulation process independent of other developmental processes. As long as any sur- vived they still attempted to gastrulate. Development went no further, however, during the period of observation. The most pro- nounced toxic effect on the majority of the embryos occurred within the first 15 minutes exposure. ZnClo.—This metallic salt is slightly less toxic than CuCl. Gastru- lation was inhibited in most individuals when the concentration was about 1/200,000. The most pronounced inhibitory effects appeared in concentrations up to 1/400,000. Persistent membranes appeared in the higher concentrations, and embryos failed to escape. Also an increasing number of embryos were killed in concentrations of 1/120,000 and higher. In these concentrations only an occasional attempt at gastrulation was seen; while in a 1/100,000 concentration, no attempt at gastrulation was seen and many were dead. On return to fresh sea water, the surviving individuals formed large globular structures without gut or skeleton. The ectoderm was often irregular and lumpy in appearance and movement was lethargic. To test the rate of action of ZnCl, samples were transferred to fresh sea water from a 1/10,000 dilution every five minutes following EFFECT OF HEAVY METALS ON DEVELOPMENT A17 their introduction at 6:50 P.M. and were examined the following morning at 8-9 A.M. The control at this time showed young plutei. A 5-minute exposure gave blastule to late gastrule. Membranes persisted and some of the embryos failed to escape completely. The latter had not gastrulated. A 15-minute exposure gave blastule to young gastrule. Many had died and disintegrated or failed to escape from their membranes. The toxic effect of this metallic chloride is manifested very quickly after exposure. The most pronounced effects occurred within the first 10-15 minutes. After this the noticeable changes occurred very gradually. Thus even after one-half hour of exposure, gastrulation of many of the surviving embryos did occur during the experimental period. Movement, however, was very lethargic. After 1-hour exposure no normal gastrule were found, although some had attempted it. ZnSO,.—A stock solution of 0.1 per cent was used. As seen in Table IV, a concentration of 1/500,000 caused conspicuous retardation. In a 1/250,000 concentration, some individuals failed to gastrulate and those which did gastrulate were still in the young gastrula stage after 15 hours exposure. In a concentration of 1/175,000 most failed to gastrulate and some died. The hardened membranes failed to dissolve. The few individuals which attempted to gastrulate showed very broad invaginating regions. This effect was also observed in the ZnCl, study. In a 1/100,000 concentration only an occasional individual showed an attempt to gastrulate. Zn Acetate.—This salt gave results quite comparable to the other zinc salts used (see Table IV). It caused the membranes to persist and the initial gastrule in the higher concentrations showed broad blastopores. The blastoccel was packed with cells. Gastrulation seemed to be inhibited at concentrations of 1/100,000 and higher. V It is well known that cyclic changes in the distribution of marine invertebrates are often accompanied by changes in the chemical composition of the sea water. Reproduction, embryonic development and growth are dependent on the presence of various constituents but the relative proportions of the various necessary elements may vary without any essential detrimental results. The absence cf a necessary element or its presence in a non-utilizable form naturally disturbs development while the presence of too much of the element (experi- mental study) will also bring about developmental changes. The latter probably does not play as important a part in marine life as does the lack of the necessary amount of the element, but nevertheless 418 A. J. WATERMAN the study furnishes information on the specific effect of the element upon development. The chlorides of some heavy metals, especially NiCle, whose effects on gastrulation and subsequent differentiation in the sea urchin, Arbacia punctulata, have been described in this study, may be added to the growing list of physical and chemical agents which may provoke exogastrulation. As is well known from studies by others, some metals are more toxic than others (Table IV) and the more toxic ones exert their toxic effect on gastrulation very quickly. The relative toxicity of the metallic salts seems to be as follows: HgCle > CuCl, > ZnClo, ZnSOu, Zn(CeH302)e > PbCl, > AleCle, CdCh, NiCl. > FeCle. In the case of most of the salts employed, the initial effect is upon growth and differentiation. Retardation and inhibition become increasingly more conspicuous in progressively higher concentrations, while the gastrulation process is the last to be effected. Finally it has been found that certain of the metals, especially NiCle, give a larger number of exogastrule at certain con- centrations than in some of the previous studies where other types of environmental modifications were made, but they are not as effective in provoking exogastrulation as LiCl, (Runnstré6m). It would appear, therefore, that such different physiological processes as gastrulation, differentiation and growth have different thresholds of inhibition for the same toxic substance. Information is given on the concentrations which inhibited gastrula- tion. Providing the exposure has not been long enough to injure the embryo fatally, gastrulation and even further development will take place in fresh sea water. The rate at which some recovery takes place depends upon the concentration employed and the length of exposure. No cases of fused embryos were found. As seen in previous studies (cf. Runnstrém, Lindahl), differentia- tion of the ectoderm and even of mesoderm, within limits, may take place independently of gastrulation or of the formation of the gut tube. Of the three germ layers, ectoderm is the most resistant to injury. In general the types of inhibitory or retardational effects are similar to those produced by other environmental modifications and by other workers. These include the behavior of the skeleton forming mesenchyme, the development of body form and size, the relative proportions of the potential ectoderm and endoderm, the inhibition of the gastrulation process, etc. The various metals give quite similar results but at different concentrations. Information is also given on the rate of action of a lethal solution of each metallic chloride tested. The more toxic the metal, the EFFECT OF HEAVY METALS ON DEVELOPMENT 419 quicker are inhibitory effects shown even in very dilute concentra- tions. In higher concentrations of certain heavy metals like Zn, Cu and Pb, the fertilization membrane does not disappear. In such cases, and even where partial escape has occurred, the embryos do not differentiate. This lack of digestion of the membrane may be due to enzyme-poisoning which is so characteristic of some of the heavy metal salts (Ishida, 1936). Although the differentiation of organs and tissues in both gastru- lated embryos and in those where exogastrulation has occurred has been of interest to numerous workers, the effect of environmental modifications on the relative proportions in the amount of ectoderm and endoderm has attracted much study (Runnstrém, Lindahl, Lindahl and Stordal, and others). By appropriate stimulation of the egg the embryo can be animalized or vegetalized and the differentiation followed. Furthermore, this phenomenon can be inhibited by various means. For example, Li salts will vegetalize the embryo while SO, deficiency will animalize (ectodermize) the embryo. Chlorides of certain heavy metals, acting on the blastula stage, likewise produce ectodermal embryos which lack gut or any endo- dermal material as far as can be seen. All transitions between this and typical gastrulation occur. The ectodermal embryos differentiate skeletal spicules or rods, apical plate and sometimes stomodzum. If treatment has not been too severe, the exogastrulz likewise differ- entiate these structures. Arms never develop but oral and anal flaps may appear and the body tends towards typical shape which, however, is modified by the abnormal distribution of skeletal material. In extreme cases only large globular, ectodermal embryos are found in which the skeleton does not go beyond the spicule stage and neither stomodzum or apical plate differentiates. Interpreted in the light of Runnstrém’s hypothesis, the heavy metals employed may effect in a differential manner the ectodermal and endodermal gradients which he believes to exist in the fertilized ege (ectodermization and endodermization of the embryo, Lindahl, 1936). It has been shown previously that these same gradients may be present at the blastula stage and hence the limits of the endoderm and ectoderm are not definitely established even at the oval blastula stage which immediately precedes gastrulation. It is doubtful if the limits are established even at gastrulation. Lindahl (1936) considers that the two gradients have different metabolic rates and also that hydrocarbon metabolism dominates the animal pole while protein metabolism dominates the vegetal pole. The heavy metals used in this study may therefore exert their toxic action upon these metabolic 420 A. J. WATERMAN processes or produce substances which give the same effect which would account for some of the various developmental abnormalities and inhibitory effects described above. LITERATURE CITED ANGERER, C. A., 1935. The action of cupric chloride on the viscosity of Arbacia egg protoplasm. Anat. Rec., Supplement No. 1, 64: 81. CHILD, C. M., 1929. The physiological gradients. Protoplasma, 5: 447. Drzewina, A., AND G. Bonn, 1926. Action de l’argent métallique sur le sperme et les larves d’Oursin. Compt. Rend. Acad. Sct., 182: 1651. Féreé, M. Cu., 1893. Note sur l’influence des injections de liquides dans |’albumen sur l’incubation de l’oeuf de poule. Compt. Rend. Soc. Biol., 45: 787. FRANKE, K. W., ET AL., 1936. Monstrosities produced by the injection of selenium salts into hens’ eggs. Anat. Rec., 65: 15. GatsorfF, P. S., 1932. The life in the ocean from a biochemical point of view. Jour. Wash. Acad. Sci., 22: 246. GLASER, O., 1923. Copper, enzymes, and fertilization. Bzol. Bull., 44: 79. Hammett, F. S., AND V. L. WALLACE, 1928. Biology of metals. VII. Influence of lead on the development of the chick embryo. Jour. Exper. Med., 48: 659. HETHERINGTON, D. C., AND Mary E. Suipp, 1935. The effect of cupric, manganous, and ferric chlorides upon cardiac explants in tissue culture. Bvzol. Bull., 68: 215% Hoab.ey, L., 1923. Certain effects of the salts of the heavy metals on the fertiliza- tion reaction in Arbacia punctulata. Biol. Bull., 44: 255. Hoan.ey, L., 1930. Some effects of HgCle on fertilized and unfertilized eggs of Arbacia punctulata. Biol. Bull., 58: 123. IsHipA, J., 1936. An enzyme dissolving the fertilization membrane of sea-urchin eggs. Annot. Zodl. Japon., 15: 453. Lire, F. R., 1921. The effects of copper salts on the fertilization reaction in Arbacia and a comparison of mercury effects. Biol, Bull., 41: 125. LinDAHL, P. E., 1936. Zur Kenntnis der physiologischen Grundlagen der Determi- nation im Seeigelkeim. Acta Zoologica, 17: 179. LINDAHL, P. E., UND A. StorDAL, 1937. Zur Kenntnis des vegetativen Stoffwechsels im Seeigelei. Arch. f. Entw.-Mech., 136: 44. PaRKER, G. H., 1924. The growth of marine animals on submerged metals. Bzol. Bull., 47: 127. PRYTHERCH, H.F., 1931. The réle of copper in the setting and metamorphosis of the oyster. Science, 73: 429. RUNNSTROM, J., 1935. An analysis of the action of lithium on sea-urchin develop- ment. Biol. Bull., 68: 378. THE CYCLE OF ORGANIC PHOSPHORUS IN THE GULF OF MAINE ALFRED C. REDFIELD, HOMER P. SMITH AND BOSTWICK KETCHUM (From the Woods Hole Oceanographic Institution’ and the Biological Laboratories Harvard University) It is generally recognized that the fertility of the sea depends upon a cycle in which carbon, nitrogen, phosphorus and other substances are assimilated under the influence of photosynthetic processes in surface waters and are set free again by processes of digestion or decay. The total organic productivity of a region is limited to the rate at which this cycle is completed. While it is apparent that in many localities the principal limiting factor is the rate of restoration of the inorganic products of decay (NO3, PO.) to the surface or photosynthetic zone, little is known concerning the exact locus within the sea at which decomposition actually sets these substances free, or of the rate at which the cycle as a whole or in part is completed. The concentration of the ultimate products of decomposition, such as NO; and POu,, tell us little about these points since they are stable substances capable of accumulating over a long period of time to high concentrations, and of being transported far from their place of origin. The actual site of decomposition is better indicated by the presence of intermediate products of decay such as ammonia, nitrite, and organic compounds of nitrogen and phosphorus. The present paper is an account of the distribution of phosphorus throughout the year at a standard station in the Gulf of Maine. At all depths the amount of phosphorus present has been measured in three forms: (1) inorganic phosphate (PO,), (2) dissolved organic phosphorus, and (3) particulate organic phosphorus. The analysis of these data gives some indication of the magnitude of the synthetic and dis- integrative processes undergone by phosphorus compounds in different parts of the water column at different times of year, and of the extent to which phosphorus is transported from one depth to another. The station chosen for study was located in the deeper portion of the western basin of the Gulf of Maine, 30 miles northeast of Highland Light. The surface water in this region is apparently relatively free from strong non-tidal currents, being sufficiently far offshore to avoid the coastal drift which accompanies the freshening of the water along 1 Contribution No. 127. 421 422 REDFIELD, SMITH, AND KETCHUM the margin of the Gulf. In this part of the Gulf, as the result of freedom from strong currents, there develops each summer maxima! surface temperatures and maximal stability of the water column. In this location one also finds relatively deep water (230-270 meters) free from strong non-tidal currents. Preliminary measurements made by Dr. E. E. Watson with current-meter indicate maximal tidal velocities of 11 cm. per second at 40 cm. above the bottom, and 14 cm. per second at 177 cm. above the bottom. The oxygen content in the deep water below 200 meters is low, varying from 4 to 4.5 cc. per liter. The point chosen for study thus presents conditions in which there exists above the bottom a considerable column of water too poorly illuminated to permit of photosynthesis, which terminates at a depth of 40 to 50 meters (Clarke and Oster, 1934) and in which decomposition occurs in sufficient magnitude to maintain a low oxygen concentration, and one TABLE [| Station positions and dates. Atlantis Depth Station Date Location of Number water meters 2440 May 18, 1935 42° 22’ N. 69° 35’ W 249 2468 August 20, 1935 42° 20’ N. 69° 32’ W 232 2493 November 8-9, 1935 42° 21.5’ N. 69° 32’ W 256 2495 February 26, 1936 42° 22’ N. 69° 33’ W 270 2558 May 14, 1936 42° 27’ N. 69° 31.5’ W 254 as well chosen as may be to avoid disturbances due to the non-tidal drift of the water. Stations were made on five occasions separated by intervals of three months, thus completing one yearly cycle. Their positions and dates are given in Table I. The approximate position is indicated by a circle in Fig. 1. ANALYTICAL PROCEDURE Inorganic Phosphate Phosphate samples were collected in black bottles and analyzed for inorganic phosphate at room temperature on shipboard by the Deniges- Atkins method, except that the solution of stannous chloride used contained 0.1 gram SnCle.2H.2O in 10 ml. of 1 to 10 hydrochloric acid. Salt effect correction factor as measured was 1.35. Corrections for salt error and reagent blank were both applied. CYCLE OF PHOSPHORUS IN GULF OF MAINE 423 Particulate Organic Phosphorus Samples of water of about 300 ml. volume were filtered on shipboard as soon after collection as possible. The procedure by which particulate organic phosphorus was determined is as follows. The particulate matter is filtered out by suction on a precipitate of barium sulfate on a 3G4 Jena sintered glass funnel with polished surface. The barium sulfate precipitate is Fic. 1. The distribution of inorganic phosphate, PO., throughout the Gulf of Maine in May, 1934, at the surface and at the depths of 60, 120, 180 meters. The circle indicates the position of the stations at which the present investigations were made. prepared by stirring 0.6 ml. of normal barium chloride into about 10 ml. of hot water containing excess sulfuric acid. It is poured over the funnel, the liquid sucked through, and filter and flask washed thor- oughly with distilled water. The sea water sample is run through the 424 REDFIELD, SMITH, AND KETCHUM filter, and the volume of the filtrate measured. Plankton and barium sulfate are washed off the filter into a 125-ml. Erlenmeyer flask with a stream of distilled water. The sides and bottom of the filter are cleaned with a rubber policeman and all the precipitate transferred to the flask to insure removal of the plankton. The funnel is cleaned by reverse suction, treatment with sulfuric and chromic acids and thorough washing. After the addition of 2 ml. of 38 per cent (by volume) sulfuric acid to the plankton, the flask is evaporated on the steam-bath to charring of the organic matter. The flask is ignited to fumes of sulfuric anhydride, and a drop of phosphate-free hydrogen peroxide (Cooper, 1934) (prepared by vacuum distillation of concentrated hydrogen peroxide) is added. A few seconds heating without loss of sulfuric anhydride fumes suffices to destroy the organic matter. The excess hydrogen peroxide is decomposed by heating the sample at 120° C. for one hour on an oil-bath. The cooled samples are diluted with ca. 50 ml. of distilled water and warmed on a steam-bath before filtering through a sintered glass funnel to remove the barium sulfate. The filtrate is diluted to 100 ml. in a glass-stoppered bottle. After the addition of 2 ml. of 23 per cent ammonium molybdate each sample is shaken. Standards of similar phosphate content as potassium dihydrogen phosphate are made up with 2 ml. of 38 per cent sulfuric acid, 2 ml. ammonium molybdate and distilled water to 102 ml. and shaken. Two drops of stannous chloride solution containing 2.5 grams SnCl».2H.2O in 100 ml. of 1 to 10 hydrochloric acid are added to each sample and standard with immediate shaking. After five minutes the samples are compared with the standards in a colorimeter with about 30 cm. depth of solution. Phosphate in the reagents for the development of the color (designated as Blank A) is determined by intercomparison of dilute standards. Phosphate introduced in the treatment of the unknowns (designated as Blank B) is determined by carrying through the procedure on a barium sulfate precipitate not treated with sea water. Blank A is added to the standards to give their true phosphate content. Both Blanks A and B are subtracted from the phosphate found in the unknowns. The net phosphorus in 1,000 ml. sample the plankton phosphorus in grams X 10~* per liter. Errors in the analysis due to loss of phosphorus by volatilization or change in acidity by loss of sulfuric anhydride are shown to be negligible by carrying through blank analyses with known amounts of added phosphorus. A procedure employing nitric acid as the oxidizing agent (Robinson grams X 10-* found in the unknowns multiplied by gives CYCLE OF PHOSPHORUS IN GULF OF MAINE 425 and Kemmerer, 1930), in which the nitric acid was evaporated off and the last traces removed by an evaporation with hydrochloric acid gave check results with the hydrogen peroxide procedure but a poor color match with the distilled water standards. It might be possible to use a procedure similar to that of von Brand (1935) for separation of the diatoms from sea water. No effort was made to differentiate between phosphorus and arsenic in the plankton phosphorus determinations. It was found, however, that the procedure of Zinzidze (1935) using bisulfate and sulfuric acid will bring about the complete reduction of at least 9 X 10~® grams of arsenic to the limit detectable by the cerulio- molybdate method (2 X 10-’ grams) and make possible a distinction between phosphate and arsenic in plankton analyses. Dissolved Organic Phosphorus The phosphorus in solution in organic combination was obtained by subtracting the inorganic phosphorus from the total phosphorus in the filtrate from which the particulate matter had been separated. The following procedure for the determination of total phosphorus in a sea water sample was devised with a view to avoiding the interference of pentavalent arsenic with the ceruleo-molybdate phosphorus determi- nation and to insure the complete destruction of organic matter. In the more rapid methods of analysis which have been used, in which sea water is treated directly with oxidizing agents, arsenic in the sample is oxidized and not subsequently reduced, while organic matter is not completely destroyed in solutions containing large amounts of chloride. Fifty-milliliter samples of sea water are treated with 3 ml. of concentrated sulfuric acid (arsenic-free) in a 125-ml. Erlenmeyer flask. After evaporation on a steam-bath under a hood to carbonization of the organic matter, each sample is evaporated to the formation of fumes of sulfuric anhydride with swirling to avoid bumping. A drop of phosphorus-free 5 per cent hydrogen peroxide (prepared by vacuum distillation of 30 per cent hydrogen peroxide) is added. The flask is heated in the same manner without loss of sulfuric anhydride for half a minute more to make the solution colorless. If necessary, more hydrogen peroxide may be added. About 60 ml. of distilled water is added to the cooled sample, which is set on the steam-bath for complete solution of the precipitated salts. After transfer to a 500-ml. Erlen- meyer and addition of 8.5 ml. of concentrated ammonia water, the excess of ammonia is boiled off and the volume of solution reduced to about 30 ml. The sample is transferred to a small ground-glass- stoppered bottle and warmed on the steam-bath, open, with 2.47 ml. of 38 per cent (by volume) sulfuric acid. Four milliliters of 10 per cent 426 REDFIELD, SMITH, AND KETCHUM sodium sulfite (anhydrous, or hydrated salt in double the concen- tration) is added, and the stopper held in with a clip to prevent the escape of sulfur dioxide. Excess hydrogen peroxide is immediately destroyed, but eight hours heating on the steam-bath is necessary to reduce pentavalent arsenic to the trivalent form. The sample is transferred to a 500-ml. Erlenmeyer, boiled for five minutes to remove sulfur dioxide, cooled, and diluted to 100 ml. in a ground-glass- stoppered bottle. The residual acid in the sample is 2 ml. of 38 per cent sulfuric. After the addition of 2 ml. of 2.5 per cent ammonium molybdate the sample is shaken. Standards which will develop about the same color intensity as the unknowns are made up from a potassium dihydrogen phosphate solution with the addition of 2 ml. 38 per cent sulfuric acid, 2 ml. of 2.5 per cent ammonium molybdate solution, and distilled water to 102 ml. These are shaken. Each standard and sample is again immediately shaken after the addition of two drops of a solution containing 2.5 grams of SnCle 0.2H2O in 100 ml. of 1 to 10 hydrochloric acid. After four or five minutes for the development of the color the samples are compared with the standards in a colorimeter with about 30 cm. depth of solution. Intercomparison of dilute standards in distilled water is made to estimate the phosphate in the reagents for producing the color (desig- nated Blank A). A blank determination carried through with reagents alone gives the phosphate introduced by reagents in the treatment of the sample (designated Blank B). Blank A is added to the standards to give their true strength in calculating the phosphate in the unknowns. Both Blanks A and B are subtracted from the phosphate found in the unknown. The salt effect correction has been determined as 0.885 for unknowns by analysis of the same sea water with varying amounts of phosphate added. The slope of the graph of phosphate added against phosphate recovered is the salt effect cor- rection. No variation in salt effect correction was found with salinity from 31.5 to 38.5. A similar procedure gave a salt effect correction of 0.81 in determining Blank B. Change of salt effect correction with temperature was not studied but extremes of temperature were avoided in making the analyses. Since several samples of sea water gave identical analyses with or without the addition of as much as 165 mg. per cubic meter of arsenic (about eight times the amount found in sea water), arsenic was assumed to be reduced completely by the procedure used. Reduction of arsenic in solutions for determining Blank B was found to be much slower than in sea water samples and quantitative only for amounts of added arsenic equivalent to 50 mg. per cubic meter or less. The CYCLE OF PHOSPHORUS IN GULF OF MAINE 427 difficulty was avoided by the use of reagents containing negligible amounts of arsenic. Schering-Kahlbaum “pro analysi’’ sulfuric acid was found suitable. The method can easily be adapted to use with a photometer although this was not attempted. Table II, containing representative data from the analysis of samples of sea water, illustrates the procedure used in computing the dissolved organic phosphate, and gives some indication of the de- pendability of the methods. The total phosphorus in the samples obtained by combining the phosphorus in the filtrate with the par- ticulate phosphorus retained by the filter (column IV) is compared with the total phosphorus in the water obtained by direct analysis without filtration (column V). It is evident that some small losses result from filtration, but these do not greatly exceed the normal variation of the Deniges-Atkins method. The values for the particulate phosphorus are small, but are consistent within themselves. The smaller values do not greatly exceed that of the blanks employed and in one set of data (for Station 2468, August 20, 1935) the values for the particulate phosphorus are in doubt owing to uncertainty in the value of the blank. The values for dissolved organic phosphorus in these samples are consistently positive and larger than the apparent errors of the method. During a large part of the year, as will be shown, much less soluble organic phosphorus is present in the water. Since this fraction is obtained by taking the difference of two large measures, each subject to considerable errors, it sometimes eventuated at such times that negative values were obtained for this fraction. The methods em- ployed evidently do not give an exact measure of the distribution of the phosphorus fractions, but they do serve to indicate the general magni- tudes of the quantities in which each occurs. The meaning of the particulate and dissolved fractions of organic phosphorus depends upon the properties of the barium sulfate filter. When a suspension of diatoms, Nitzschia closterium, containing a known amount of phosphorus is filtered, the phosphorus may be recovered quantitatively from the filter. The refiltration of a filtrate leaves no detectable amount of phosphorus upon the filter. Evidently diatoms and microorganisms of similar size are completely retained in the particulate fraction. Dr. Charles E. Renn has kindly tested the filter for the retention of bacteria. After filtering sea water containing some 400 bacteria per milliliter the filtrate contained about one-sixth that amount. The particulate organic phosphorus fraction probably con- tains the phosphorus of all the larger phytoplankton and smaller zooplankton and the greater portion of the bacteria and detritus as well. The dissolved organic phosphorus fraction may include a small 1)$) 428 REDFIELD, SMITH, AND KETCHUM TABLE II Specimen of data on phosphorus fractions in samples of sea water. Concentrations expressed as milligrams PO, per cubic meter. I II III IV Vv VI VII Depth Total Total Organic in Inorganic | Particulate | Phosphorus phos- phosphorus. phos- meters phos- phosphorus in filtrate phorus Direct Difference phorus phorus Il + Ill analysis Ill —I 1 45 10 64 74 62 70 80 86 68 78 67 77 71 + 6 22 10 27 13 64 Hil 68 60 73 86 64 77 63 76 Hdl — 1 36 20 39 14 66 80 86 70 84 86 45 59 60 74. 86 —12 21 60 110 6 116 130 116 — i 6 122 123 100 110 4 126 142 142 130 142 —12 — 16 125 115 5 132 138 146 137 142 — § 17 160 156 150 111 5 142 147 158 —11 31 200 123 5 152 154 162 157 158 — 1 29 225 108 8 148 162 158 156 160 — 4 40 CYCLE OF PHOSPHORUS IN GULF OF MAINE 429 portion of the bacterial flora, perhaps some minute nannoplankton, and detritus which has been reduced to the smallest dimensions in addition to organic compounds of phosphorus in colloidal form or in true solution. Renn (1937) has estimated that a bacterial population of 100,000 cells per milliliter would represent only 2.9 mg. PO, per cubic meter. It is evident that the much larger values of dissolved organic phosphorus obtained in our analyses can be due in only negligible part to the presence of bacteria. We believe, consequently, that our measurements represent chiefly the presence of phosphorus compounds in solution. TABLE III Inorganic phosphorus. Concentrations expressed in milligrams PO, per cubic meter. Depths not corrected for wire angle, which in no case would reduce by more than 4 per cent. Depth in May 18 August 21 November 8-9 February 26 May 14 meters 1935 1935 1935 1936 1936 1 35 5 45 105 14 10 28 Dall 27 102 10 20 34 53 39 103 15 30 59 96 78 99 82 40 70 92 73 100 95 50 — — 77 96 97 60 97 93 110 88 116 80 105 110 89 99 117 100 108 109 110 95 120 125 114 107 115 99 129 150 122 122 111 96 147 175 138 132 123 131 152 200 146 137 123 141 146 225 161 136 106 143 157 250 170 139 139 The distribution of phosphorus in the three forms into which it has been separated is recorded in Tables III, IV, V and VI. Inorganic phosphate represents by far the greater quantity of phosphorus in the water amounting to 72 to 92 per cent of the total at different times. It is rather uniformly distributed at depths greater than 80-100 meters throughout the year, the concentrations increasing somewhat with depth. In the superficial layers, as has been frequently observed elsewhere, the inorganic phosphate becomes greatly reduced in quantity in the spring and is restored to concentrations characteristic of greater depths during the winter. Particulate organic phosphorus, representing organisms and detritus, is in general the smallest of the three fractions, amounting to about 5 per cent of the total. It occurs in greatest quantity in the 430 REDFIELD, SMITH, AND KETCHUM upper layers—above 40 meters—corresponding to the observed distribution of phytoplankton (Gran and Braarud, 1935). The quanti- ties in these layers are highest in spring; in midwinter the quantity in surface water is scarcely to be distinguished from that in deep water. The values obtained in August are subject to doubt. Below the photosynthetic zone the distribution of filterable organic phosphorus is on the whole very uniform, amounting to about 5 y PO. per liter. The distribution and magnitude of the concentrations of filterable phos- phorus agree well with that of particulate nitrogen observed by von Brand (1937) in these waters in the summer of 1936. TABLE [IV Particulate organic phosphorus. Concentrations expressed in milligrams PO, per cubic meter. Depths not corrected for wire angle, which in no case would reduce by more than 4 per cent. Depth in May 18 August 21 November 8-9 February 26 May 14 meters 1935 1935 1935 1936 1936 1 20 13 10 6 17 10 18 14 13 5 13 20 15 15 14 5 20 30 20 7 9 4 12 40 12 8 7 4 0 50 — = 5 4 8 60 5 § 6 6 9 80 8 6 4 4 12 100 5 5 4 4 3 125 4 3 5 3 2 150 3 = 5 3 3 175 5 3 13 3 wa | 200 4 3 5 3 7 225 4 4 8 4 7 250 6 5 In May dissolved organic phosphorus occurred in only minimal quantities in the water of all depths except near the surface. During the summer the concentration increases markedly until November, when over 20 per cent of the total phosphorus is in this form. In the early winter there is a rapid disappearance of this form of phosphorus, associated in time with the increase in concentration of inorganic phosphate in the upper waters. The appearance of organic phosphorus commences in May at the surface, and the concentrations appear to grow from the surface downward. Not until November are high concentrations observed near the bottom. These observations suggest that considerable decomposition is taking place throughout the water column and in particular in those depths where phyto- and zoéplankton CYCLE OF PHOSPHORUS IN GULF OF MAINE 431 are known to exist in greatest numbers; and that at this station decomposition at the bottom may be relatively unimportant. The findings concerning organic phosphorus are somewhat similar to those of Kreps and Osadchik (1933), who made studies in Barents Sea. They found organic phosphorus to show a gradual increase from August to January reaching concentrations of some 40 mg. P.O; per cubic meter and actually exceeding the inorganic phosphorus during the latter month. Their observations, which did not cover the earlier part of the year, showed the greatest concentrations in the deeper waters. As will be pointed out in a subsequent publication, the seasonal distribution of soluble organic phosphorus has some resemblance to TABLE V Dissolved organic phosphorus. Concentrations expressed in milligrams POs, per cubic meter. Depths not corrected for wire angle, which in no case would reduce by more than 4 per cent. Depth in May 18 August 21 November 8-9 February 26 May 14 meters 1935 1935 1935 1936 1936 1 30 49 21 — 2 12 10 15 36 35 0 0 20 —1 50 29 De, 35 30 —4 14 7 14 18 40 3 24 36 — 3 3 50 —_— = 43 5 3 60 8 36 7 1 26 80 0 24 50 9 55 100 2 30 24 Zyl 22 125 —3 29 20 21 22 150 y) 6 38 WM 9 175 —6 20 20 4 —11 200 —7 22 30 6 8 BOS) —2 17 43 — 4 10 250 8 —10 that of ammonia, which may be considered to be a somewhat analogous stage in the nitrogen cycle. The data recorded in Tables III, IV and V have been submitted to further analysis with a view to determining in so far as possible, just what alterations take place in the phosphorus cycle at various depths and at different times of year. The column of water is considered to be virtually a closed system in which every exchange with the surroundings is exactly balanced by an equal and opposite exchange. By dividing the column into a number of segments lying at different depths, in which the quantity of phos- phorus in the different forms is recorded from time to time, and by 432 REDFIELD, SMITH, AND KETCHUM observing certain general biological and hydrographic principles, it becomes possible to estimate to what extent changes in the concen- trations of each form of phosphorus may be derived from processes taking place in situ, and to what extent vertical movements of phos- phorus from one segment or layer to another must be postulated. To obtain a workable body of data, Table VI has been drawn up recording the quantity of phosphorus present in each of the three TABLE VI Summary of distribution of phosphorus fractions. Depth in | May 18 | Aug. 20 | Nov. 8 | Feb. 26 | May 14 meters 1935 1935 1935 1936 1936 Total Pas grams PO: per sq. m. 0-240 | 28.7 31.0 32.9 29.9 34.4 Percentage of total phosphorus 0-60 16.4 20.6 18.0 22.0 14.9 in each 60-meter layer 60-120 | 23.7 23.2 25.0 23.0 28.0 120-180 |. 27.0 26.3 28.0 25.0 27.9 180-240 | 32.9 29.9 29.0 30.0 29.2 Total 100.0 | 100.0 | 100.0 | 100.0 | 100.0 Inorganic phosphorus as per- 0-60 11.9 12.6 11.3 19.7 10.6 centage of total 60-120 | 22.1 16.8 18.7 19.4 20.8 120-180 | 26.0 7 ess) 21.6 21.1 25.0 180-240 | 32.0 25.6 Dee, RST 26.5 Total 92.0 ies 72.8 88.9 82.9 Dissolved organic phosphorus 0-60 1.5 6.2. 5.0 1:4. \> 23 as percentage of total 60-120 0.4 5.4 5.4 Dea 6.0 120-180 0.2 3.2 5.0 3.3 2.5 180-240 0.0 3.7 6.4 0.6 1.7 Total 2.1 18.5 21.8 8.0 1225 Particulateorganicphosphorus| 0-60 3.0 1.8 ier 0.9 2.0 (organisms and detritus) as| 60-120 1.2 1.0 0.9 0.9 1.2 percentage of total 120-180 0.8 0.6 1.4 0.6 0.4 180-240 0.9 0.6 1.4 0.7 1.0 Total 5) 4.0 5.4 Sell 4.6 forms for each of four layers each of 60 meters depth at each time of observation. ‘The values are obtained by graphical integration and are expressed as percentages of the total phosphorus in a water column of 240 meters depth at each time of observation. In analyzing these data the following premises are held: 1. The horizontal exchange due to the drift of water past the station may be neglected. ‘This premise is not justified on the ground that the CYCLE OF PHOSPHORUS IN GULF OF MAINE 433 station is located in a region of minimal drift. There can be little doubt that water is constantly drifting past the station. The observed changes in salinity demonstrate this. The total phosphorus recorded varies + 7 per cent from the mean value throughout the year. Since this variation shows no seasonal sequence, the total phosphorus being lowest in May, 1935 and maximal in May, 1936, there evidently is some variation in the character of the water occupying the station at different times. These differences are eliminated by expressing the phosphorus fractions as percentages of the total, a procedure which imposes artificially the character of a closed system upon the set of data. The justification of this procedure lies in the relatively small differences in total phosphorus observed from time to time, and in the fact that on the whole the horizontal distribution of phosphorus throughout the Gulf at any time, at least to judge by inorganic POs, is much more uniform than is the vertical distribution. The horizontal distribution of PO, at various depths throughout the Gulf as observed in May, 1934 are shown in Fig. 4 and illustrate this fact. The general character of the phosphorus cycle may be supposed to be similar in all parts of the basin. 2. Phosphate present as zodplankton and nekton and not sampled by the water bottle may be ignored. A large number of vertical zodplankton hauls made throughout the years 1933-34 in all parts of the Gulf yielded an average catch of 40 cc. dry plankton per square meter of surface. It may be estimated from analyses made on such material that this would contain about 0.4 per cent of the total phosphorus in the water from which it was strained. Since the particulate organic phosphorus amounts to about ten times this quantity, it may be seen that the neglect of this fraction does not introduce a significant error. 3. All synthesis of particulate or soluble organic phosphorus com- pounds from inorganic phosphate takes place in the upper layer. This is justified by Clarke’s measurements on the penetration of light into the Gulf of Maine and on determinations of the compensation point in photosynthesis by diatoms in bottle experiments at different depths (Clarke and Oster, 1934). 4. All downward movement of phosphorus is due to the sinking of organisms (particulate organic phosphorus). ‘This is the only fraction affected by gravity. It is also the only fraction displaying a well- marked concentration gradient decreasing downward—a condition essential for downward dispersal by eddy conductivity. 5. All upward movement of phosphorus is due to the transport of inorganic PO. by eddy conductivity. The gradient of concentration of inorganic PO, increases downward and is well marked except in mid- 434 REDFIELD, SMITH, AND KETCHUM winter. The soluble organic phosphorus never develops a strong gradient in this direction. Since it is present in much smaller concen- trations than is the inorganic phosphate, it may safely be ignored in considering vertical transport by eddy conductivity. 6. All observed transformations in any layer are attributed to processes occurring in that layer, except so far as vertical transport must be postulated to account for the transformation. ‘This premise is introduced since without it a unique solution cannot be obtained. It implies that all values arrived at for vertical exchange are minimal. 7. The portion of the exchange in which the cycle runs to completion 1s necessarily ignored. All values for the magnitude of the exchange are ea minimal. Fic. 2. See text. A consideration of this limitation may serve to make clear the general basis of the analysis. If we start with the system in a steady state and if, during the period between two sets of observations the cycle has proceeded without change in the relative velocity of the processes in any part, the distribution of the fractions of phosphate in all parts of the system will be the same at the end as it was in the beginning. This does not mean that no exchanges of phosphate between different parts of the system have taken place, rather that all exchanges are exactly compensated. While the system is in a steady state an unobservable quantity of phosphate is undergoing transfor- mation from each stage in the cycle to the next stage. If the system is disturbed, as through seasonal changes in the physical conditions, then CYCLE OF PHOSPHORUS IN GULF OF MAINE 435 transformations of one sort may proceed more rapidly than those of another with the result that differences in the distribution of phos- phorus are observed, and from these differences the magnitude and nature of the processes which have caused the differences may be deduced. The observations tell us nothing, however, of the basal level of activity on which the differences are superposed.” Figure 2 illustrates the principle of the method. At any time there will be a basal level of activity represented by the transformation of an unobservable quantity of phosphorus, x, through each stage in the cycle. This quantity will have been transported upward from layer 2 into layer 1 to be synthesized into particulate organic form. If the system is to remain unchanged, the equivalent of this material must have sunk back into the deeper layers and been decomposed, passing through the soluble organic form, to exactly replace that which was transported upward. Portions of x, designated as x, ¥2, etc. may sink to deeper layers before undergoing transformations from organic to inorganic form. ‘The general conditions are that the quantity of x entering and leaving any part of the system shall be equal, that « move upward as inorganic phosphate (Postulate 5), and downward as filterable organic phosphorus (Postulate 4), that it represent synthesis of filterable organic phosphorus only in the upper layer, and that it represent a transformation of organic into inorganic phosphorus in any layer. If the system is disturbed between observations, then changes in the quantity of phosphorus, AY, in any form and part of the system may be observed. These changes may be accounted for only by additional exchanges between the various parts of the system. The problem is to determine the minimal additional exchanges of this sort, Y1, Y2, Ysetc., which will account for the change in each part of the system in ac- cordance with the postulates laid down above. The general conditions are that AY, the change in any fraction in any layer shall equal the difference in Y;, Yo, Y3 etc., the amounts of phosphorus entering or leaving that fraction and layer during the period between observations. Furthermore, AY must be accounted for so far as possible by exchanges taking place within the layer in question (Postulate 6). An attempted analysis of the changes in phosphorus distribution is presented in Tables VII, VIII and IX. February to May (Table VII) represents the period in which the 2The method is analogous to the integration of a differential equation, the unobserved basal activity corresponding to the constant of integration. It is only in proportion as the system undergoes great seasonal fluctuation that the partial effects observed approach the total exchanges taking place. The method is applicable consequently particularly to studies made in high latitudes. 436 REDFIELD, SMITH, AND KETCHUM great spring flowering of phytoplankton occurs. During this period an amount of phosphorus equivalent to 9.1 per cent of the total disappears from the inorganic phosphate of the upper layer. Of this only 2.0 per cent can be accounted for as an increase in particulate and soluble organic phosphorus remaining in that layer. Seven and one-tenth per cent must have sunk to the deeper layers following its synthesis into organic matter. Only one-eighth of the phosphorus absorbed in TABLE VII Balance sheet of phosphorus exchanges February 26 to May 14, 1936. Numbers represent the change in the phosphorus fractions as percentages of total phosphorus in water column. : Soluble Particulate I 5 : Depths phosphorus ahgchare Shosphorae Photosynthesis —9.1 —> +9.1 Decomposition +0.9<— —0.9 0-60 0 =< Qo meters Exchange with layer below 0 0 —7.1 Net change —9.1 +0.9 +1.1 Exchange with layer above 0 0 +7.1 60-120 Decomposition +4.7 <— —-4.7 meters +1.4 <— —1.4 Exchange with layer below 0 en) — eal Net change +1.4 +3.3 +0.3 Exchange with layer above 0 0 +2.1 120-180 Decomposition +0.9 <— —0.9 meters +1.7 <— —-1.7 Exchange with layer below +2.2 0 —1.4 Net change +3.9 —0.8 —(0.2 180-240 Exchange with layer above —2.2 0 +1.4 meters Decomposition 0 +1.14<— —-1.1 Net change aed +1.1 +0.3 photosynthesis has remained as particulate matter in the upper layer. This observation accords with the conclusion of Harvey (1934) that several times more vegetation is produced during the spring flowering of diatoms in the English Channel than is found there at the time of its maximum. To account for the increasing concentrations of inorganic phos- phorus in the deeper layers after allowing for the greatest possible decomposition in situ at least 2.1 per cent must sink past the 120-meter CYCLE OF PHOSPHORUS IN GULF OF MAINE 437 level and 1.4 per cent past the 180-meter level. The phosphorus removed from inorganic form in the upper layer by photosynthesis is redistributed during the spring through considerable depths by the sinking of particulate matter. It is unnecessary to assume that any of the particulate matter sinks beyond the lower level before undergoing decomposition, though it is possible that this may be the case. May to November (Table VIII) includes the greater part of the TABLE VIII Balance sheet of phosphorus exchanges May 18, 1935 to November 8, 1935. Numbers represent the change in the phosphorus fractions as percentages of total phosphorus in water column. : Soluble Particulate I c : a phosphorus | ,,o°85RiC, | organs 0-60 Photosynthesis —19.2 — > +19.2 meters Decomposition 4+3.5<— —3.5 0 =< oO Exchange with layer below +18.6 0 —17.0 Net change —0.6 +3.5 —1.3 60-120 Exchange with layer above —18.6 0 +17.0 meters Decomposition +5.0 <— —5.0 0 < 0 Exchange with layer below +15.2 0 —12.3 Net change —3.4 +5.0 —0.3 120-180 Exchange with layer above —15.2 0 +12.3 meters Decomposition +4.83<— —4.8 0 < 0 Exchange with layer below +10.8 0 —6.9 Net change —44 +4.8 +0.6 180-240 Exchange with layer above —10.8 0 +6.9 meters Decomposition +64<— -—64 Net change —10.8 +6.4 +0.5 growing season. The important feature of this period is the appear- ance of large quantities of dissolved organic phosphorus at all depths. One-fifth of all the phosphorus in the water is in this form in November. This material can have been produced only by photosynthetic processes taking place in the upper layer. It must have been set free for the most part by decomposition of the particulate fraction in the layer in which it is observed (Postulate 5). In order to account for the 438 REDFIELD, SMITH, AND KETCHUM quantities of dissolved organic phosphorus observed, a large vertical movement of inorganic phosphorus upward through all depths must be postulated as well as an equivalent sinking of organisms to the sites at which the soluble organic phosphate appears. Over 17 per cent of the total phosphorus in the water must pass through the zone of photo- synthesis in the course of the six summer months. Since the account is balanced without supposing any phosphorus to pass from the organic TABLE IX Balance sheet of phosphorus exchanges November 8, 1935 to February 26, 1936. Numbers represent percentages of total phosphorus in entire water column. : Soluble Particulate I : - ash phosphorus | oa. | pram Photosynthesis 0 ==> 0 Decomposition +0.6<— —0.6 0-60 +4.2 <— —4.2 meters Exchange with layer below +4.2 0 —0.2 Net change +8.4 —3.6 —0.8 Exchange with layer above —4,2 0 +0.2 Decomposition 0 <— 0 60-120 +2.7 <— —-2.7 meters Exchange with layer below +2.2 0 —0.2 Net change +0.7 —2.7 0 Exchange with layer above —2.2 0 +0.2 120-180 Decomposition 0 <— 0 meters +1.7 <— —-1.7 Exchange with layer below 0 0 —1.0 Net change —0.5 —1.7 —0.8 Exchange with layer above 0 0 +1.0 180-240 Decomposition +1.7<— -—-1.7 meters +7.55 <— —7.5 Net change +7.5 —5.8 —0.7 back to the inorganic form, a process which must certainly be taking place, this figure may be far below that actually obtaining. November to February (Table IX) is marked chiefly by the re- generation of inorganic phosphorus, which increases by 16 per cent of the total, and by the equalization of the concentration of this fraction throughout the water column. The table shows that this regeneration is made to a large extent at the expense of the soluble organic phos- CYCLE OF PHOSPHORUS IN GULF OF MAINE 439 phorus, that the transformation takes place throughout the entire _ range of depths, though greatest near the bottom. The decomposition of organic phosphorus compounds in situ and the vertical transport of inorganic phosphate are about equally important in effecting the equalization of the concentration of the latter throughout the water column. THE MECHANISM OF VERTICAL TRANSPORT The foregoing analysis indicates that very considerable exchanges of phosphorus take place between various depths of water. At the same time these exchanges appear to diminish in extent as the depth in- creases. Downward movement has been attributed to the sinking of particulate matter under the influence of gravity. There appears to be no difficulty in considering that in depths of a few hundred meters organized particles of the dimensions of diatoms would sink to the bottom before undergoing decomposition. If such were the case very large quantities of phosphorus would be withdrawn from the water during each growing season. This may be the case in shallow waters, but it does not appear to be happening in the western basin of the Gulf of Maine. If so the total phosphorus in the water should show a marked seasonal change. The situation is probably complicated by biological considerations. Harvey (1934) has presented evidence that the stock of phytoplankton is grazed down by zodplankton during the summer. This conclusion suggests that the zodplankton are important active agents in converting particulate organic phosphorus into its decomposition products. Since these animals, and particularly the copepods, make extensive diurnal vertical migrations, and since some time must elapse between the taking of food near the surface and its elimination as waste products, they provide an agency for a limited vertical transport of organic material. From this viewpoint the zooplankton become an important agency in maintaining the fertility of the water for phytoplankton, since they hasten the conversion of bound nutrients into inorganic form and prevent these nutrients from becoming unavailable by the sinking of particulate matter to great depths or to the bottom. The vertical transport of inorganic phosphate is simpler since it can be effected only by the mixing of the water. It is pertinent to inquire whether the conditions are such as to permit of the amounts of transport deduced during the various periods of observation. The amount of a constituent, Q, passing through unit horizontal surface in unit time depends upon the gradient of concentration of the constituent 440 REDFIELD, SMITH, AND KETCHUM dc/ds and the coefficient of eddy conductivity, A (Austausch coeffi- cient). QO = Adc/ds The coefficient of eddy conductivity, A, represents the volume of water exchanged through each horizontal unit surface in unit time. The gradient of phosphate concentration observed in February, May and November is shown in Fig. 3. Between May and November, when large vertical movements have been deduced, a well-marked gradient exists particularly in the upper layers, as is required for such movements. In February this gradient has disappeared, and the fo) 30 60 90 120 150 = 180 210 NOVEMBER FEBRUARY 240 60 80 100 120 140 160 80 100 120 140 Fic. 3. Distribution of inorganic phosphate concentration with depth in November, 1935 and February and May, 1936 at standard station in the western basin of the Gulf of Maine. Depths in meters measured downward along the ordi- nate; concentrations in milligrams PO, per cubic meter along the abscissa. concentration of phosphate is equal at all depths down to 150 meters. No amount of mixing can effect a change in its vertical distribution. Since this condition must exist during a considerable portion of the winter, it is not surprising that the vertical transports deduced from our data between November and May are smaller than those observed in the summer. A knowledge of the coefficient of eddy conductivity, A, is the key to understanding the nutritive conditions in deep bodies of water. CYCLE OF PHOSPHORUS IN GULF OF MAINE 441 Methods of estimating its value are so indirect that little is known of its magnitude under any circumstances. Although our data are admittedly very unprecise and yield only minimal values for the exchange, it is nevertheless of some interest to use it in estimating the coefficient required to account for the transport of phosphate in the Gulf of Maine. The gradient of phosphate concentration is sufficiently uniform throughout the period May—November to permit a single value to be taken at any level as representative of the entire period, This is not the case during the remainder of the year. We have (0) 30 MAY 6 (e) NOVEMBER 90 120 150 180 210 240 FEBRUARY 25.0 25.5 26.0 265 27.0 26.0 265 270 Fic. 4. Distribution of density, o:, with depth in November, 1935, and Febru- ary and May, 1936, at standard station in the western basin of the Gulf of Maine. Depths in meters measured downward along the ordinate; density, o:, measured along the abscissa. calculated the value of A for the boundary of each of the layers, using the data presented in Table VIII for estimating Q and the slopes of the curves in Fig. 3 for dc/ds. The result is shown in Table X. The values of A are minimal and are less reliable at the greater depths. The values obtained are not unreasonable. Seiwell considered A to equal 2 C.G.S. units in the thermocline of the tropical Atlantic, whereas values increasing to 50 C.G.S. units have been obtained in various waters (Seiwell, 1935). 442 REDFIELD, SMITH, AND KETCHUM The value of the coefficient of eddy conductivity depends upon the forces responsible for mixing and varies inversely with the stability, oo of the water. If K represent the mixing forces, | des ds Seiwell considers that K = 4.73 X 10-4 in the thermocline of the North Atlantic. Figure 7 shows the distribution of density, o:, with depth at the station in the Gulf of Maine in May, 1935 and in No- do t ds in Table X are taken. The values of K given by multiplying = K=AX vember. From these curves the representative values of entered by A are of the order obtained by Seiwell. It is concluded that the TABLE X Estimation of coefficient of eddy conductivity in Gulf of Maine, May to November, 1935. Depths Q Felis A dox|ds Haran et per cent in mg. mg. ml. grams meters 3 months cm.2 X sec. ml. cm. cm. X sec. | ml. X cm. 60 >18 >3.5X10- | 6.610%] > 5.2 | 1.01074) >5.2x10~4 120 >15 >2.4<107-8 | 2.0X10-9 | >12.0 | 0.51074 | >6.010% 180 >10 >2.010-8 | 3.0X1079 | > 6.6 | 0.4X104 | >2.6x104 vertical transport of inorganic phosphate deduced from the seasonal change in the distribution of the various fractions of phosphorus compounds does not require unreasonable assumptions concerning the magnitude of the eddy conductivity. SUMMARY 1. Methods are described for the determination of the phosphorus present in particulate form and of the total phosphorus in a sample of sea water. 2. The distribution of phosphorus present as inorganic phosphate, as dissolved organic compounds, and as particulate matter (detritus and microérganisms) has been determined at all depths throughout the year at a station in the western part of the Gulf of Maine. 3. In late winter over 90 per cent of the phosphorus is in inorganic form and three-quarters of the remainder is present as soluble organic compounds. CYCLE OF PHOSPHORUS IN GULF OF MAINE 443 4, In the spring—February to May, inorganic phosphorus is con- verted to organic form by photosynthesis in the upper layer of water. Most of this fraction sinks to considerable depths before undergoing decomposition. 5. During the summer—May to November, large quantities of dis- solved organic phosphorus appear at all depths, indicating a very con- siderable transport of inorganic phosphate from deep water to the surface and the sinking of an equivalent amount of phosphorus in particulate form to the depths in which organic compounds are liberated by decomposition. Decomposition appears to take place throughout the water column. 6. During the winter—November to February, the organic phos- phorus compounds are converted to inorganic phosphate. This and vertical mixing of preformed phosphate are about equally important in bringing about the equalization of phosphate concentrations through- out the depth of water. 7. A method is described for analyzing quantitatively the factors producing a seasonal change in the distribution of a compound such as phosphorus. It is shown that the vertical transport of material within the water mass demanded by such an analysis may be accounted for reasonably by the hydrographic conditions obtaining. 8. Values of the coefficient of eddy conductivity at several depths are obtained. BIBLIOGRAPHY VON BRAND, T., 1935. Bol. Bull., 69: 22. Von BRanp, T., 1937. Biol. Bull.,'72: 1. CLARKE, G. L., AND R. H. OsTER, 1934. Biol. Bull., 67: 59. Cooper, L. H. H., 1934. Jour. Mar. Biol. Ass’n, 19: 755. Gran, H. H., anp T. BRAARuD, 1935. Jour. Biol. Bd., Canada, 1: 279. Harvey, H. W., 1934. Jour. Mar. Biol. Ass’n, 19: 755. Kreps, E., AND M. Osapcuik, 1933. Intern. Rev. Hydrobiol. und Hydrograph., 29: Beile RENN, C. E., 1937. Biol. Bull., 72: 190. Roginson, R. J., AND G. KEMMERER, 1930. Trans. Wisconsin Acad. Sci., 25: 117. SEIWELL, H. R., 1935. Papers in Physical Oceanography and Meteorology. Vol. 3, No. 4, pp. 1-56. ZINzIDZE, Cu., 1935. Ind. and Eng. Chem., Anal. Ed., 1: 227. GROWTH AND VARIABILITY IN DAPHNIA PULEX BERTIL GOTTFRID ANDERSON, H. LUMER AND L. J. ZUPANCIC, JR. (From the Biological Laboratory, Western Reserve University) INTRODUCTION The aim of this study is to determine the number of pre-adult instars, growth, relative growth, and variability of individually reared female Daphnia pulex De Geer. Numerous studies on growth of Cladocera have been made. Most of them are based on size-frequency distributions in natural popula- tions. Some are based on experiments with individually reared animals. These studies have been carried on largely by Woltereck and his students, especially Rammner. Size-frequency distribution methods are inadequate for growth determinations as will be brought out more fully later. Previous studies carried out by means of indi- vidually reared animals deal with small numbers of organisms, a short part of the possible life span of an individual, or both. Recently Banta and his co-workers have carried out intensive studies on large numbers of individually reared Daphnia longispina which lived for long periods. In the present study a large number of animals has been observed for a relatively long time. PROCEDURE Essentially the same procedure was used as that employed by Anderson (1932) for Daphnia magna. Individual female Daphnia pulex of a single clone were isolated within eight hours after their release from their mothers and reared in manure-soil medium (Banta, 1921). Each individual was placed in a separate glass vial containing 20-25 cc. of the medium. Semi-weekly one-half of the total volume was replaced with fresh medium. The animals were kept at room temperature (15°—22° C.). At the time of isolation and daily thereafter, each individual animal was placed in a watch glass together with a few drops of culture medium. Just enough saturated chloretone solution was added to bring about cessation of movement. Measurements as shown in Fig. 1 were made by means of an ocular micrometer. The animals were never in the chloretone solution for more than five minutes at any one time. 444 GROWTH AND VARIABILITY IN DAPHNIA 445 Immediately before the daily mensurations note was taken of cast carapaces and the number of young released. These were removed at the time of their discovery. RESULTS AND DISCUSSION Longevity Some 82 animals were observed in these experiments. Figure 2 is a survival curve. Fifty-one animals lived for twenty instars and of these, 3 continued to the twenty-fifth instar. Bourguillaut de Ker- herve (1926) observed two Daphnia magna for nineteen instars. Each one released nineteen clutches of young and since not more than one clutch of young is released each instar and at least five instars precede the release of the first clutch of young, these two Daphnia magna lived Fic. 1. Diagram showing method of making measurements. T, total length, longest dimension of animal exclusive of spine. C, carapace length, longest dimension of the carapace exclusive of spine. H, height, the shortest distance between two lines tangent to the carapace and parallel to the line of T. This measure of height is affected very little by the number of young in the brood chamber. for twenty-four instars or more. He cites 4 other individuals which must have reached the twenty-second, the thirteenth, the tenth, and the sixth instar, respectively. Anderson (1932) observed some 30 Daphnia magna for fourteen instars, 32 for thirteen instars, and others for a smaller number of instars, all from the time of release from the mothers. Rammner (1928) observed one Scapholeberis mucronata for seventeen instars and another for nine. Rammner (1929) cites others that have been observed for shorter times. Ingle, Wood, and Banta (1937) have observed individually reared Daphnia longispina for over twenty-five instars. Of the 82 animals observed in the present study, 71 were primiparous during the fifth instar and 9 during the sixth, while the remaining 2 died during the second and third instars. The number of 446 ANDERSON, LUMER, AND ZUPANCIC pre-adult instars is therefore variable in this species as well as in others (Anderson, 1932). The minimum number of pre-adult instars for Daphma pulex is probably four. Of the 71 animals primiparous during the fifth instar, 47 lived through the twentieth instar. The data from these 47, summarized in Table I, are the basis for the growth studies which follow. Absolute Growth Figure 3 is a group of growth curves in terms of total length, carapace length, and height. The curves are similar in shape. The ap pe Too] pe To &0 NUMBER OF INDIVIOUAL S w S a 10 15 £0 £5 INSTAR Fic. 2. Survival curve for the eighty-two animals observed in these experiments. point of inflection in each curve comes immediately before the instar during which the animals are primiparous. Attempts have been made to fit the Robertson and Gompertz equations to the curves. In neither instance was the result judged satisfactory. The time unit used in this work is the instar. Adult instars have been considered as equivalent physiological time units (Anderson, 1933). At room temperature, each of the first three pre-adult instars lasts approxi- mately one day, the fourth or last pre-adult instar one and a half days, the first adult instar about two days, and each subsequent instar becomes increasingly longer. The use of the instar as a time unit GROWTH AND VARIABILITY IN DAPHNIA 447 may be the cause of unsatisfactory results in attempting to fit the equations to the present data. When the logarithms of total length are plotted against time in instars, a curve is obtained which may be broken up into three distinct segments each of which approximates a straight line. The first seg- ment takes in the first five instars, the second segment includes the next five, and the third segment embraces instars eleven to twenty. Figure 4 shows the growth increments in terms of total length, carapace length, and height. The increments increase up to the fourth instar then gradually decrease until the eleventh instar, after which they remain much the same. It is of interest to determine to what degree such characteristics of MM. Ae 7 20 C 1S Hf LO OS 5 /0 (5-29 WSTAR Fic. 3. Growth curves based on data from the 47 animals which were pri- miparous in the fifth instar and lived a twenty instars or more. JT, total length; C, carapace length; H, height. the growth process as initial size, final size, duration of growth, and initial velocity of growth are interdependent. As measures of these quantities, total length in the first instar (a), total length in the twentieth instar (A), the number of instars required to attain a length of approximately 0.8A (¢), and the increment between the first and second instars (7), respectively, were employed. For these the follow- ing coefficients of correlation were obtained: toa = 0.2309 = 0.0931 fai = — 0.4481 + 0.0792 ra; = — 0.1483 + 0.0963 fat = 0.1641 + 0.0957 — Wiss) se WH038S = fe I 448 ANDERSON, LUMER, AND ZUPANCIC It is apparent that only 7,; and ra; are significantly different from zero. ‘There is thus an indication of an inverse relationship between initial size and initial velocity of growth, and between duration of growth and final size. On the other hand, there is no evident relation- ship between initial and final size, between initial size and duration of growth, or between initial velocity and final size. Similar coefficients of correlation have been computed for Daphnia longispina by Wood and Banta (1936). These authors found that with unlimited food early growth tends to be inversely related to size at mm. rv par ed 025 020 QI5 0.10 O05 5 Ogee Ti 1.20 INSTAR Fic. 4. Growth increment curves for the same animals as in Fig. 3. T, total length; C, carapace length; H, height. — the time of release, but that when the quantity of food is limited these factors vary independently. They found also that initially larger animals tend to become larger adults, which is contrary to the results obtained from our data. A negative correlation between initial size and initial growth rate has been found to occur also in certain mammals. Thus it has been observed that in man (Hammett, 1918) and in the cat (Hall and Pierce, 1934) the percentage increase in body weight in the period immediately following birth is inversely proportional to birth weight. GROWTH AND VARIABILITY IN DAPHNIA 449 Hammett (1918) interprets these results as indicating that differences in birth weight correspond to differences in physiological age at the time of birth, the physiologically younger individuals having the greater percentage growth rate. In multiparous mammals the situation is often complicated by variation in litter size and attendant differences in food supply of the sucklings. Crozier and Enzmann (1935) showed that in inbred albino mice there is a hyperbolic relationship between birth weight and litter size, also that individuals in large litters grow more slowly at first than those in small litters, due to the decreased quantity of available milk per animal. After the suckling period, however, the growth of the smaller individuals is accelerated so that they eventually catch up with the others, and all attain the same adult weight regard- less of birth weight. In order to eliminate differences in nutrition, Kope¢ (1932), working with a non-inbred stock of mice, reduced all litters to the same size. He found that birth weight varied inversely with litter size, and that individuals born in large litters exhibited a higher initial percentage growth rate than those born in small litters. When the data were seriated on the basis of birth weight without regard to litter size, however, there was found to be no correlation. Kope¢, who regards the different size groups in the second case as representing different genetic types, concluded that genetic differences in birth weight have no effect on subsequent growth, as do environmental differences. The validity of this argument seems open to question. It would be interesting in this connection to determine whether or not the same results occur with an inbred stock, also to investigate similarly the effects of brood size in Cladocera. The occurrence of a significant negative correlation between dura- tion of growth and final size in the present case appears to be in line with the results of McCay, Crowell, and Maynard (1935). These authors observed that in white rats, individuals whose growth period has been prolonged by partial starvation attain on adequate feeding a final size lower than that of controls given sufficient food throughout. Ingle, Wood, and Banta (1937), however, in a similar experiment on Daphnia longispina, found practically no difference in final size between the experimental animals and the controls. Also Merrell (1931) found that in rabbits there is no significant correlation between the time required to reach one-half the adult weight and the adult weight attained. On the other hand, Merrell’s observation that initial growth rate is not associated with adult weight is in agreement with our results. 450 ANDERSON, LUMER, AND ZUPANCIC Reproduction Figure 5 gives the average number of living young released during each instar. The number increases until the tenth instar followed by a gradual decrease. The period of increase corresponds to the time when the growth is falling most rapidly. The period of decrease corresponds to the time when growth has practically ceased. The number of young released by a daphnid during any one instar has been used as an index of its condition (Anderson, 1933). The data available in the present study may serve as a test for this assump- tion. The young released in any instar develop from eggs produced during the instar before. Consequently, the coefficient of correlation must be computed from the growth increment with respect to total £9) NUMBER OF YOUNG S S SS G, o /0 15 £0 INSTAR Fic. 5. Curve of the average number of living young released during each instar by thesame animals as in Fig. 3. length for instar five and the number of young released in instar six, etc. The coefficients for the fifth, sixth, and tenth instars were determined. These are .690 + .025, .649 + .028, and .645 + .030, respectively. Hence one may conclude that the number of young produced may serve as a valid index of the condition of a daphnid. Relative Growth Relative growth studies show certain characteristics of this species. In Fig. 6 the logarithms of carapace length were plotted against those of total length, those of height against those of total length, and those of carapace length against those of height.. The relation of carapace length to total length can be expressed satisfactorily on this log log GROWTH AND VARIABILITY IN DAPHNIA 451 plot by three straight lines. The first of these can be drawn through the points for the pre-adult instars (1-4), the second through the points for the next eight instars (5-12) and the third through the last eight instars (13-20) represented. The relations between height and total length are similar to those between carapace length and total length. The relations between carapace length and height are dif- ferent. Two lines may be used, one for the first ten instars and a second for the last ten (11-20). These linear relations can be ex- pressed as Ve DI & ON Ste S J SD IO Nh ALLE. LS A 3 mm. Fic. 6. Double logarithmic plots of the relations between carapace length and total length C/T, height and total length H/T, and carapace length and height C/H during each instar for the same animals as in Fig. 3. where 6 is a constant—the initial growth index, a is the equilibrium constant, and x and y are values of two parts (Huxley and Teissier, 1936). The values of the constants are given in Table II. These values were calculated by the least squares method. The relationships shown above are much the same as those secured for Daphnia magna by Anderson (1932). No very marked change in proportions occurs as may be seen in Fig. 7. The marked change in relative growth that occurs at the twelfth and thirteenth instars as indicated in Fig. 6 can hardly be recognized by inspecting Fig. 7 alone. The lines representing the thirteenth to the twentieth instars in Fig. 6 are very short since growth is very much retarded during this period. 452 ANDERSON, LUMER, AND ZUPANCIC TABLE [ Mean values of total length, carapace length, and height and their probable errors in millimeters for each instar. Also the standard deviation (c) and the coefficient of variation (V) for total length. Standard Coefficient deviation of variation 1 | 0.5665 +0.0034 | 0.0347 +0.0024 | 6.1340.43} 0.4194 +-0.0027| 0.3056+0.0021 2 |0.7214+0.0038 | 0.0389 +0.0027 | 5.39-0.37| 0.5361 +0.0034| 0.4015 +0.0027 3 | 0.9609 =+0.0056 | 0.0568 0.0040 | 5.92 +0.41] 0.7244 +0.0048) 0.5433 +0.0039 4 | 1.2653+0.0092 | 0.0934+0.0065 | 7.38+0.51] 0.9848 +.0.0078} 0.7346+0.0054 5 | 1.6079=+0.0095 | 0.0964-+-0.0067 | 6.00+0.42) 1.2850+0.0082) 0.9597 +0.0059 6 7 8 9 Instar Total length Carapace length Height 1.8134+0.0106 | 0.1081 0.0075 | 5.96+0.41| 1.4572 +0.0097| 1.1015 0.0084 1.9332 0.0117 | 0.1196+0.0083 | 6.19+0.43| 1.5591 =+0.0110) 1.1778 0.0094 2.0571 +0.0130 | 0.1316+0.0092 | 6.40-+-0.45| 1.6615 +0.0111) 1.2445 +-0.0096 2.1673 +0.0126 | 0.1275 +0.0089 | 5.88--0.41| 1.7587 +0.0110} 1.3213+0.0090 10 | 2.2466+0.0116 | 0.1174+0.0082 | 5.23+0.36) 1.8200+0.0103] 1.3738--0.0088 11 | 2.3061 +0.0108 | 0.1095 +0.0076 | 4.75+0.33] 1.8653 0.0093) 1.3976+0.0083 12 | 2.3401+0.0099 | 0.1007 =-0.0070 | 4.31--0.30| 1.8891 +0.0088} 1.4131+0.0071 13 | 2.3705 0.0092 | 0.0939 +0.0065 | 3.96+0.28) 1.9034+0.0081| 1.4227 0.0066 14 | 2.3853+0.0095 | 0.0962 +0.0067 | 4.03+-0.28) 1.9135 0.0081} 1.4262 0.0071 15 | 2.4032+0.0092 | 0.0935 0.0065 | 3.89+0.27/ 1.9326+0.0075| 1.4274+0.0065 16 | 2.4223+0.0085 | 0.0866-+0.0060 | 3.58+0.25} 1.9433+0.0072| 1.4274+0.0053 17 | 2.4461+0.0072 | 0.0736=-0.0051 | 3.01+0.21] 1.9594 +0.0055} 1.4340 +0.0047 18 | 2.4604+0.0072 | 0.0730+0.0051 | 2.97+0.21] 1.9689 0.0060} 1.4340 0.0046 19 | 2.4835+0.0072 | 0.0730+0.0051 | 2.94--0.20| 1.9862 0.0061) 1.4435 +0.0049 20 | 2.4890+0.0072 | 0.0731-0.0051 | 2.94+0.20) 1.9928+-0.0061| 1.4429 +0.0047 Variability in Body Size Variability in body size and its fluctuations during growth have been studied in other forms by a number of investigators. The present data are particularly suitable for such a study, since it may safely be assumed that the animals are genetically homogeneous and that genetic differences are thus eliminated as a source of variation. TABLE II Values of the constants 6 and a@ for the data shown in Fig. 7. Relation Instars b a Carapace length y.............. 1-4 0.762 1.06 Total length COR ie ues ee 5-12 0.790 1.03 13-20 0.814 0.985 Height CURR Kati ee kode? 1-4 0.417 1.09 Total length BG aie Gye heEOtete re 5-12 0.530 1.03 13-20 12.49 0.315 Carapacellengthivacs. i . fs sede: 1-10 1.333 0.99 Height Bereich A ele tie «js 11-20 0.848 2.34 GROWTH AND VARIABILITY IN DAPHNIA 453 DELO GG nie eed ee | AM Fic. 7. Camera lucida outline drawings of a single animal for each of the first eighteen instars. Arabic numerals designate pre-adult instars; Roman numerals— adult instars. 454 ANDERSON, LUMER, AND ZUPANCIC Total length was employed as the best available measure of body size. Its use in this way is justified by the fact that there is apparently very little change in the shape of the body during growth (see the preceding section). The standard deviations and coefficients of variation for total length are given in Table I and are plotted against instar number in Fig. 8. It may be noted that both these quantities tend to increase at first, but that after the eighth instar they fall off rather steadily until the seventeenth instar, after which they remain approximately constant. Hence as the growth process reaches com- pletion, body size becomes less variable both absolutely and relatively. fort op er rp hy o) /0 1S 20 INSTAR Fic. 8. Standard deviation and coefficient of variation of total length in relation to instar number. A comparison of the curve for the coefficient of variation with the increment curve (Fig. 4) suggests that these two quantities are corre- lated. When the coefficient of variation is plotted against the loga- rithm of the increment (Fig. 9), the points tend to group themselves along a straight line, indicating that there is roughly a linear relation- ship between the relative variability and the logarithm of the growth rate. Whether or not any general significance can be attached to the precise character of the relationship as indicated here is open to ques- tion; however, there is no doubt that a distinct relationship exists. These observations are essentially in agreement with those of previous investigators. A trend in the coefficient of variation similar GROWTH AND VARIABILITY IN DAPHNIA 455 to that obtained here has been found to occur in the case of both stature and body weight in man by several workers, including Bowditch (1877), Thoma (1882), Porter (1894), Boas (1897), Boas and Wissler (1904), and others. A summary and discussion of the data of Bow- ditch, Boas, and Boas and Wissler is given by Thompson (1917, pp. 78-80). Essentially the same trend has been observed for body weight in the albino rat by Jackson (1913), King (1915), and Hanson and Heys (1927), in the Norway rat by King (1923), and in the cat by Hall and Pierce (1934). King (1918, 1919) has shown also that the trend persists in albino rats inbred for as many as twenty-five genera- mime Tee yo ype ay eS 400 200 100 .060 .040 020 0/0 006 J 4 rr) 6 Uf V Fic. 9. Graph of the coefficient of variation of total length plotted against the logarithm of the increment. The values of the coefficient employed are averages of successive pairs in Table I. tions, although the coefficient of variability in such rats is consistently lower than that in non-inbred animals of the same average body weight. Weymouth and McMillin (1930) have found that in the Pacific razor clam, Siliqua patula, relative variability in shell length, as measured by the ratio interdecile range: median, decreases steadily with increasing length. Several of these investigators have remarked also the existence of a correlation between the coefficient of variation and the growth rate (cf. Thoma, 1882; Porter, 1894; Boas, 1897; Boas and Wissler, 1904; King, 1915, 1923). Jackson (1913), on the other hand, states that his data show no evident correlation between these quantities. 456 ANDERSON, LUMER, AND ZUPANCIC Thompson (1917) computed from the data of Boas the coefficients of variation for annual increments in human stature between the ages of five and eighteen, and found that these increased steadily with age. He attributes this to increasing differences in phase of growth among individuals of the group as growth proceeds. Analogous coefficients of variation computed for the present data fluctuate somewhat irregularly, but show on the whole a marked upward trend. In fact, from the thirteenth instar on, the values of the coefficients are con- siderably higher than 100 per cent. These values, however, are of doubtful significance, since the mean increments for these instars are small. When the value of the mean is close to zero, as Philiptschenko (1927) has pointed out, the coefficient of variation becomes unreliable as a measure of variability. As a matter of fact, one would expect that as the growth process comes to a close, the differences in phase of growth would tend to decrease (i.e. the increments for all individuals would ultimately become zero), and that variability in growth rate would correspondingly decrease. The similarity of results obtained in the cases thus far investigated suggests that the relationships described above are of rather general occurrence, and that it may be worthwhile to seek a general explanation for them. In attempting to find such an explanation, it is necessary first of all to consider the sources of the observed variability, and the relative importance of each. There are in all four possible sources, namely errors in measurement, genetic differences, environmental differences, and the fundamental nature of the growth process itself. It does not seem likely that the observed trend in variability can be accounted for to any significant degree on the basis of error in measure- ment. Assuming that the distribution of errors is Gaussian, we should expect that in measurements of length or weight, the absolute error, as indicated by the standard deviation, would remain constant or would, in some cases, increase with increasing magnitude of the object being measured. In the former case, the relative variability would decrease with growth; in the latter its behavior would vary, depending on the particular conditions. There is, however, no evident feature of such observational error which would account for a trend in the standard deviation such as that shown in Fig. 8, nor for the existence of a relationship between the coefficient of variability and the growth rate. Some information concerning the rdéle of genetic differences can be obtained from the data of King (1919) on the growth of inbred and non- inbred albino rats raised under identical laboratory conditions. Although the trend in the coefficient of variation is essentially the same GROWTH AND VARIABILITY IN DAPHNIA 457 in both stocks, the degree of correlation between the coefficient and the growth rate, as indicated by the scatter of points on a graph, is much higher for the inbred rats than for the others. This may be interpreted to mean that variability due to genetic factors tends to be largely independent of the velocity of growth. On the other hand, variability due to environmental differences - would be expected to vary with the growth rate, for an individual in a state of rapid growth is relatively highly sensitive to the action of environmental agents, and a given fluctuation from the norm would produce a greater deviation in growth than it would in an individual with a low growth rate. This has been recognized by Plunkett (1932) in connection with developmental processes in general. He points out that as such a process asymptotically nears completion, the organisms tend to become more stable and less sensitive to the effects of environmental factors such as, for example, temperature. Hence the variability with respect to the particular character involved will decrease. However, he erroneously assumes that the converse is also true, namely that if one group of organisms is less variable in a certain respect than another group, it is therefore nearer to the completion of the underlying developmental process. This may or may not be generally true, but it clearly requires further demonstration. It has generally been assumed that variability in growing organisms, aside from that introduced through errors in measurement, must be due to either genetic or environmental differences. According to a recent theory proposed by Rahn (1932), however, such variability arises at least in part from the physico-chemical nature of the growth process itself. The theory is briefly as follows. The division of a cell must be preceded by the doubling of all its genes. Let us consider a particular gene in each of a number of unicellular organisms which are identical with respect to genetic and environmental factors. If we assume that the doubling of this gene in the various cells con- forms to the law of mass action, then it follows that it will not double simultaneously in all the cells, but will do so over a definite time interval. If the same is true of each of the remaining genes, there will be a variability in division time of the cells. Rahn has developed a simple mathematical formulation of the theory which he has applied to data on the division rate of bacteria. For multicellular forms, he has succeeded in analyzing only the case of a hypothetical organism in which all the cells are alike and in which the percentage rate of cell division is constant. He has shown that such organisms will vary with respect to the time required for the completion of a given number of cell generations. Moreover, as the 458 ANDERSON, LUMER, AND ZUPANCIC number of generations increases, the frequency curves become flatter while the relative spread of variation decreases. In other words, there will be an increasing standard deviation and a decreasing coefficient of variation. If the theory is valid, it would follow that variability in cell number (and in body size insofar as it depends on cell number) in successive time intervals should behave in a similar fashion. Obviously these conclusions are not directly applicable to growth in actual multicellular organisms for a number of reasons. First of all, such growth is generally characterized by a decreasing percentage growth rate; secondly, it involves increase in cell size, as well as in cell number; and thirdly, the average reproductive rates of different types of cells are not equal. Before the conformity of the theory with the observed results can be tested, the former must receive a much a 8 NUMBER OF INDIVIDUALS S 4 . Lana IEA \N Ss \\ fF H L \ 5 ‘ y \ * IN \ H : Nae PN BON SP ia ase 4 MIN 5 AA 6A MIAN IN /\ AN te Ne /\ yy) Ui NG HENNE / 4 INF x vA \\ 6 8 10 12 14 16 18 2.0 Ge, Fic. 10. Size frequency distributions during the first seven instars for all eighty-two animals observed. Broken lines designate the individual instars. The solid line is a composite curve for all instars. The vertical bars near the upper edge of the figure represent the mean total lengths for each of the first seven instars. more elaborate mathematical formulation, and a theoretical analysis must be made in which these factors are taken into account. It may therefore be tentatively concluded that the observed trend in variability in body size with growth and its relation to growth rate are explicable largely in terms of the action of environmental factors, and perhaps also in part in terms of Rahn’s theory. Size-frequency Distribution Most studies of growth in Cladocera have been made by analysis of size-frequency distributions in natural populations. These have been of two types (see Woltereck, 1929). One consists of plotting the number of individuals in a size class against size. In the graph that results a number of size modes appear which are taken as repre- GROWTH AND VARIABILITY IN DAPHNIA 459 sentative of the growth stages or instars. A second type consists of plotting the value for one dimension of an individual against another dimension of the same individual for each animal in a population. Usually the points fall into groups which are more or less distinct. Each group of points is taken as representative of a single instar. No one so far as we are aware has attempted a comparison between the results secured by these methods and those secured by observations on individually reared animals. To make such a comparison, Figs. 10 and 11 were constructed. First the values for total length fre- mm. 6 ro 10 hé 14, 16 /8 20 Qe TOTAL LENGTH Fic. 11. Size frequency distributions in two dimensions during the first seven instars for all 82 animals observed. Each instar is represented by a different type of circle. The area of each circle is directly proportional to the number of animals of the particular dimensions represented by the circle. The crosses are the mean values for each of the first seven instars. quency during the first instar for all 82 animals observed were plotted. Then those for the second instar and so on to the seventh instar. After that a composite curve for all the values was plotted. The separate instar curves are shown in broken lines and the composite curve in a solid line in Fig. 10. The vertical lines in the upper part of the figure represent the mean total lengths for each instar up to and including the seventh. It is obvious that each mode in the com- posite curve does not represent the mean total length for an instar. The first three distinct modes are near the mean values. No distinct mode is apparent for the fourth instar. The mean value for the fifth 460 ANDERSON, LUMER, AND ZUPANCIC instar falls near a mode but the means of the sixth and seventh instars fall midway between modes. Further, considerable overlapping occurs between the values for various instars. The greatest value for the first instar is larger than the smallest for the second. Similar are the cases for the second and third instars, the third and fourth, and the fourth and fifth. Some of the values for the fifth are greater than the lowest for the seventh instar. This size-frequency method may be of some use for the pre-adult instars, but certainly not for adult instars. Figure 11 was constructed by plotting the values of height against that of total length for all 82 animals for the first instar, followed by those of the second instar, and so on through the seventh. The area of each circle is directly proportional to the number of individuals represented. Where values for two or more instars fall on a point separate circles are used for each instar. The mean height and the mean total length for each instar are represented by crosses. This method differs from that described by Woltereck (1929) in that each individual was represented by a point on the graph while in the present case if more than one individual is of the same size the circle at point on the graph is larger and its area proportional to the number of indi- viduals represented. In analyzing Fig. 11, it will be noted that a band of circles is secured but no distinct groups. The circles of the greatest area may be con- sidered to represent the instars. For the first three instars, the circles of greatest area fall near the mean values for the instars. No con- spicuously large circle is found to represent the fourth instar. The cross for the fifth instar falls near a large circle but the crosses for the sixth and seventh instars do not. In most plots of natural populations (Woltereck, 1929) the points scatter more widely, i.e., they form a wider band than is the case in Fig. 11. Here again it is obvious that size-frequency distribution does not lend itself to a satisfactory study of growth. Studies on growth by means of individually reared animals are advantageous over the above methods in several respects. The individual life histories are fully known. Genetic constancy can be maintained since diploid parthenogenesis is the only means of repro- duction. Selection of young from one clone insures genetic constancy except for mutations. The environment can be maintained fairly constant. By natural population analysis methods the individual histories are unknown, the genetic constancy cannot be controlled, and the environment is subject to considerable fluctuation. GROWTH AND VARIABILITY IN DAPHNIA 461 SUMMARY Eighty-two individually reared female Daphnia pulex were ob- served from the time they were released from the brood chamber of their mothers until they died. Measurements of total length, carapace length, and height were made daily on each animal. The number of young released during each instar was recorded. Seventy-one animals were primiparous during the fifth instar and 9 during the sixth. Consequently the number of pre-adult instars is variable, the minimum being four. Data from the 47 animals which were primiparous during the fifth instar and which lived for twenty instars or more were used in con- structing growth curves.. Growth in the three dimensions studied is sigmoid. The point of inflection in all curves comes during the fourth instar, the last pre-adult. The Robertson and the Gompertz equations do not fit the data satisfactorily. This may be due to the time unit employed, which in this case is the instar. The growth increment is greatest during the fourth instar. The increments increase up to the fourth instar, then decrease gradually until the eleventh instar, after which they remain low and relatively constant. A significant negative correlation exists between initial body size and initial growth rate, also between duration of growth and final body size. Other characteristics of the growth process investigated appear to vary independently. The number of young released during the adult instars increases to a maximum at the tenth instar followed by a gradual decrease. The number of young released during any adult instar is sig- nificantly correlated with the growth increment for the instar preceding the one during which the young are released. Relative growth in the dimensions studied may be expressed satis- factorily by the equation : i OE No marked changes occur in the relations between total length and carapace length. Marked changes in the relations between carapace length and height and between total length and height occur at the thirteenth instar. The standard deviation and coefficient of variation of total length tend to increase somewhat during the early instars, but after the eighth instar both decrease rather steadily. Relative variability in body size, as measured by the coefficient of 462 ANDERSON, LUMER, AND ZUPANCIC variation of total length, is roughly directly proportional to the logarithm of the growth rate. The existence of this relationship is ascribed largely to the mode of action of environmental factors during growth. It may perhaps also be explicable in terms of Rahn’s theory of the physico-chemical origin of variability in growth rate. Size-frequency analyses of natural populations as methods of studying growth are shown to be inferior to the method using indi- vidually reared animals. LITERATURE CITED ANDERSON, B. G., 1932. The number of pre-adult instars, growth, relative growth, and variation in Daphnia magna. JBzol. Bull., 63: 81. ANDERSON, B. G., 1933. Regeneration in the carapace of Daphnia magna. I. The relation between the amount of regeneration and the area of the wound during single adult instars. Biol. Bull., 64: 70. Banta, A. M., 1921. A convenient culture medium for daphnids. Science, 53: 557. Boas, F., 1897. The growth of Toronto school children. Rept. U. S. Comm. of Ed. 2 (1897): 1541. Boas, F., anp O. O. WissLER, 1904. Statistics of growth. Rept. U. S. Comm. of Ed. 1 (1904): 25. Bownitcu, H. P., 1877. The growth of children. Rept. Mass. State Board of Health, 1877, p. 278. Crozier, W. J., AND E. V. ENzMAnn, 1935. On the relation between litter size, birth weight, and rate of growth, in mice. Jour. Gen. Physiol., 19: 249. Hatt, V. E., AND G. N. PiERcE, JR., 1934. Litter size, birth weight, and growth to weaning inthecat. Anat. Rec., 60: 111. Hammett, F. S., 1918. The relation between growth capacity and weight at birth. Am. Jour. Physiol., 45: 396. Hanson, F. B., AND F. Heys, 1927. Differences in the growth curves of albino rats born during the four seasons of the year under uniform laboratory conditions. Anat. Rec., 35: 83. Hux.ey, J. S., unp G. TetssiER, 1936. Zur Terminologie des relativen Gréssen- wachstums. Biol. Zentralbl., 56: 381. INGLE, L., T. R. Woop, AnD A. M. Banta, 1937. A study of longevity, growth, reproduction, and heart rate in Daphnia longispina as influenced by limi- tations in quantity of food. Jour. Exper. Zoél., 76: 325. Jackson, C. M., 1913. Postnatal growth and variability of the body and of the various organs in the albino rat. Am. Jour. Anat., 15: 1. KERHERVE, J. B. DE., 1926. La descendance d’une Daphnie (D. magna) ou ses millions de germes en une saison. Ann. de Biol. Lacustre, 15: 61. Kine, H. D., 1915. Growth and variability in the body weight of the albino rat. Anat. Rec., 9: 751. Kine, H. D., 1918. Studies on inbreeding. I. The effects in inbreeding on the growth and variability in the body weight of the albino rat. Jour. Exper. Zool., 26: 1. Kine, H. D., 1919. Studies on inbreeding. IV. A further study of the effects of inbreeding on the growth and variability in the body weight of the albino rat. Jour. Exper. Zoél., 29: 71. Kine, H. D., 1923. The growth and variability in the body weight of the Norway rat. Anat. Rec., 25: 79. Kopsé, S., 1932. Die prozentuelle Wachstumsgeschwindigkeit der Mduse in bezug auf das Gewicht der Neugeborenen. Arch. Ent.-mech., 126: 575. GROWTH AND VARIABILITY IN DAPHNIA 463 McCay, C. M., M. F. CRowELL, AND L. A. MAYNARD, 1935. The effect of retarded growth upon the length of life span and upon the ultimate body size. Jour. Nuirition, 10: 63. MERRELL, M., 1931. The relationship of individual growth to average growth. Human Biol., 3: 37. PHILIPTSCHENKO, J., 1927. Variabilitat und Variation. Gebriider Borntraeger, Berlin. PLUNKETT, C. R., 1932. Temperature as a tool of research in phenogenetics: methods and results. Proc. Sixth Int. Cong. Genet., 2: 158. PorTER, W. T., 1894. The growth of St. Louis children. Trans. St. Louis Acad. Sct., 6: 263. Raun, O., 1932. A chemical explanation of the variability of the growth rate. Jour. Gen. Phystol., 15: 257. RAMMNER, W., 1928. Uber die postembryonale Entwicklung der Cladocere Scaphole- beris mucronata. O. F. Miller (Ergebnisse aus Einzelzuchten). Arch. Ent.-mech., 113: 287. - RAMMNER, W., 1929. Uber periodische Erscheinungen am Cladoceren-Individuum. Int. Rev. Hydrob. Hydrog., 21: 402. THomMA, R., 1882. Untersuchungen iiber die Grésse und das Gewicht der anato- mischen Bestandtheile des menschlichen K6rpers im gesunden und im kranken Zustande. F.C. W. Vogel. Leipzig. TuHomeson, D. A. W., 1917. On Growth and Form. University Press, Cambridge. WEymouTH, F. W., AnD H. C. McMILLin, 1930. Relative growth and mortality of the Pacific razor clam (Siliqua patula, Dixon) and their bearing on the commercial fishery. Bull. Bur. Fish., 46: 543. WOLTERECK, R., 1929. Technik der Variations- und Erblichkeitsanalyse bei Crustaceen. Abderhalden Handb. biol. Arbettsm., Abt. 1X, Teil 3. Woop, T. R., anp A. M. Banta, 1936. Interpretations of data on growth and reproduction (Daphnia longispina) (abstr.). Anat. Rec., 67:55. SEASONAL PRODUCTION OF ZOOPLANKTON OFF WOODS HOLE With SPECIAL REFERENCE TO CALANUS FINMARCHICUS ! GEORGE L. CLARKE AND DONALD J. ZINN (From Harvard University, Cambridge, Mass., and Rhode Island State College, Kingston, Rhode Island) — The purpose of the work described in this paper was the investiga- tion of the distribution of the plankton in the waters around Woods Hole during the summer months and the study of the seasonal changes in the plankton at one locality throughout an entire year. Our primary object specifically was to ascertain whether the copepod, Calanus finmarchicus—a form used extensively for laboratory experi- ments—breeds in this region, and if so during which months and with what success. It was proposed also to include measurements of the physical and chemical factors in the environment, and the collection of nannoplankton and phytoplankton, in order that the sequence of biological events from season to season in this locality might be followed. Cause and effect relationships in the growth of the various plankton forms might thus be unraveled. Since copepods play a prominent réle in the economy of the sea, it is important to know what conditions promote their growth and what factors tend to reduce their numbers (Clarke, 1934). To this end a series of laboratory experiments was initiated in which the nutrition of copepods, particularly of Calanus finmarchicus, was investigated (Clarke and Gellis, 1935; Fuller and Clarke, 1936; and Fuller, 1937). The nearest point from which copepods of this species could be obtained in sufficient numbers was in the deeper water off Gay Head. The population of Calanus at this locality was found to persist throughout the summer, but we had no knowledge either of the subsequent fate of these copepods or of the time and conditions of their appearance here. In our studies of nutrition we had assumed that the Calanus found in this area were living in equilibrium with their environment and that the smaller organisms found in the same body of water were adequate both qualitatively and quantitatively for their food supply. If, however, these copepods had been produced in another locality and transported hither by currents, the possibility existed that the conditions at our point of observation were not suitable 1 Contribution No. 112, Woods Hole Oceanographic Institution. 464 SEASONAL PRODUCTION OF ZOOPLANKTON 465 for the growth or even the maintenance of Calanus and that the copepods found here were destined shortly to die off. The proposed field observations were therefore especially desired and when worked out in conjunction with further laboratory experiments should eventually yield information on the factors controlling copepod pro- duction which would have a general application. Let it be understood, however, that the work described in this paper was of the nature of a reconnaissance program. The locality chosen for the observations throughout the year was a point as far offshore as could be reached within a day’s sail from Woods Hole, but it probably is not entirely removed from fluctuating land in- fluences. However, evidence will be given below for believing that this station is typical of a considerable area. LOCATION OF STATIONS AND HYDROGRAPHY During the summer of 1935 the five stations in the waterways around Woods Hole shown in Fig. 1 were visited about twice a week. Buzzards Bay is a broad, shallow body of water with a temperature a few degrees higher and a salinity slightly lower than Vineyard Sound. A strong current flows in Vineyard Sound, and offshore water seems to be carried into the Sound much more readily than is the case with Buzzards Bay (Haight, 1936). Since access to the open sea from Woods Hole is most direct through the western entrance of Vineyard Sound, a line of stations was run in that direction to study the transi- tion between inshore and offshore conditions and populations. Station 3, which was selected for the continuation of the observa- tions throughout the year, was visited once a month during the autumn and winter and at least twice a month from April to Sep- tember. The authors are indebted to Dr. Renn, Mr. Iselin, Mr: Woodcock, and Mr. Butcher for assistance which made these trips possible. A rough sea is almost always encountered at the exposed location of Station 3, and during the extremely cold weather when spray froze on the deck and rigging, work was extremely difficult. Station 3 is about eight miles from the nearest points of land on the north and east, and twenty-five miles from Block Island on the west. On the south the continental shelf water extends without obstruction. Since at Station H2 (see Fig. 1) the flood tide flows NNE and the ebb tide flows WSW, the main body of water debouching from Vineyard Sound and Buzzards Bay probably passes largely to the west of Station 3. However, the currents in this region are extremely complex and unfortunately the measurements of the U. S. Coast and Geodetic Survey (Haight, 1936) do not extend as far out GEORGE L. CLARKE AND DONALD J. ZINN 466 “pueysy] jayonjueN JO sepis y30q uO vas ay} 0} Uddo SI puNog jJeyxONJUeN] SeoIOyM ‘deur ay} JO WU] ay2 Jo y}10u jsnf pus ue 0} sautOD Aeg spiezzng ‘peazy Aes) Jo yyNos ay} 0} deur ay} Yo qsnf sol] puey] suey] ON JO purysi jews eyy, “AeAing o1Jepoar) pue ysv0Z *S ‘(A 243 Aq pardnoo0 ZF] uol}eIS Jo puke ‘epeul orm s[neYy uozyZued yoIyM Ye G—] SUOT}eIS JO UOT}VIOT “]T “Oly @YVAINIA v4 JUNE SWE Y A\UG SEP JUNE JULY AUG SEPT Fic. 2. Percentage distribution of Calanus finmarchicus copepodid stages III, IV, and V and adult @ and o& taken with the scrim net at Station 3 during the summer of 1935. The open circles represent catches made on an incoming tide (E. 1 hour—-W. 1 hour) and the solid circles represent catches made on an outgoing tide (W. 1 hour-E. 1 hour). 468 GEORGE L. CLARKE AND DONALD J. ZINN as our station. Our plankton catches during the summer of 1935 at the stations nearer shore showed a considerable difference depending upon whether they were made on an incoming or an outgoing tide. There is some evidence that the effect of the tide was felt as far out as JUL AUG SEP OCT NOv DEC JAN FEB MAR APR MAY JUN JUL AUG SEP Yoo) SAL 33 dy Nn) inv} JUL AUG SEP OCT NOV DEC JAN FEB MAR APR Maéy. J UL Fic. 3. The salinity and temperature at depths of 0, 15, and 30 meters at Station 3 during 1935-6. The vertical markers along the abscissa indicate the dates on which observations were made. Station 2, but at Station 3 no consistent difference due to the phase of the tide was observed, as may be seen for the Calanus population from Fig. 2. Transparency measurements similarly indicated the presence SEASONAL PRODUCTION OF ZOOPLANKTON 469 of typical offshore water at this station (Clarke, 1938). However, for the sake of uniformity, our observations throughout the year were made as far as possible only during the periods when the “‘east,’’ or flood tide, had flowed for at least one hour, and before the ‘‘west”’ tide had flowed for more than one hour. These tidal oscillations are superimposed upon the much larger but slower coastal current which flows in a southwesterly direction over the continental shelf all along this coast. Water is moving continuously from Nantucket Shoals on toward New York. It is presumably from the inner edge of this current that the water at our station is derived. Yet there must be a certain admixture of inshore water, since the salinity value at Station 3 (see Fig. 3) was slightly lower at all seasons of the year than that characterizing the water farther offshore? and was slightly higher than that found within Buzzards Bay and Vineyard Sound, according to determinations which we carried out during the summer of 1935. The prevailing wind, which is from the southwest, would tend to blow offshore water in toward our station, but gales from this direction or from other points of the compass probably disturb the normal hydrographic situation rather profoundly. The fluctuations in the salinity at Station 3 from July, 1935 to September, 1936, were for the most part confined between 31.5 and 32.5 Joo and generally uniform at the depths of 0, 15, and 30 meters (Fig. 3). On several occasions, however, a noticeable freshening of the surface layer was observed, and on December 29, 1935, an unusually great difference in salinity existed between the surface and the bottom. Since a pronounced transparency anomaly occurred on the same day (Clarke, 1938), we may assume that an abnormal interchange of water masses took place on this occasion. The temperature change at Station 3 during the year was ex- tremely great for a marine environment. The maximum value, observed in August, was over 23° C. higher than the minimum value, in February (cf. also Allee, 1919). During the summer months the water was highly stratified, for a difference of 8° or 10° was usually found between the surface and the bottom. This would seem to indicate that no great stirring was produced by wind or tide in these months, but it is conceivable that the temperature difference could be brought about by some persistent differential water movement. In September the surface water began to cool off rapidly and to mix more effectively with the bottom water, thus raising the temperature of the latter. By the first of November the temperature had become 2 At Station ‘‘Martha’s Vineyard I.” See H. B. Bigelow and M. Sears (1935). 470 GEORGE L. CLARKE AND DONALD J. ZINN uniform from top to bottom. This condition persisted throughout the autumn and winter indicating strong stirring action during this period. In the spring months warming took place more rapidly in the upper layers than in the deeper with the result that a progressively greater difference in temperature was found between the surface and bottom until by June or July a pronounced stratification was re- established. METHODS During the summer of 1935 10-minute horizontal hauls for zooplankton were made using the scrim closing nets described by Clarke (1933). At Stations 1, 4, and 5 hauls were made at depths of 2 meters below the surface and 2 meters above the bottom; at Stations 2 and 3 an additional haul was made at a depth mid-way between the other two. Water samples were taken and temperatures determined at each of these depths. A fraction of each water sample was turned over to Miss Lois Lillick for qualitative and quantitative analysis of the phytoplankton. Other fractions of the water samples were bottled for determination of salinity, and of phosphate and nitrate content. ‘The authors are indebted to Mr. Alfred Woodcock, Mr. Bostwick Ketchum, and Mr. Homer Smith, for these chemical analyses. For the work at Station 3 from October, 1935 to September, 1936, the zodplankton was taken in ‘‘oblique”’ hauls in which the net was lowered to 30 meters (just over the bottom), towed for 1 minute, then raised 1 meter and towed for 1 minute, then raised another meter and so on, until the surface was reached. A double-action hand plankton pump was added to our equipment in order that the eggs, nauplii, and younger copepodid stages of the copepods, which were too small to be retained by the scrim net, might be caught whenever spawning took place. The free end of the hose was attached to the cable about 2 meters above the weight and then lowered to the bottom. The outflow from the pump was piped directly to a 1’’ Hersey Water Meter and then discharged into the top of a No. 20 silk phytoplankton net hung vertically in a metal barrel. Ten gallons (about 38 liters) were pumped at each meter from the bottom to the surface. At 30, 15, and 0 meters bottles were filled from the hose for the chemical analyses and for quantitative phytoplankton counts by Miss Lillick. At 20, 10, and 0 meters the pump was stopped and the contents of the silk net drawn off and bottled separately, thus dividing the catch into three parts representing three strata in depth. On each occasion temperatures were taken and light penetration was measured as is described elsewhere (Clarke, 1938). When time SEASONAL PRODUCTION OF ZOOPLANKTON 471 permitted, a second oblique zodplankton haul and pumping operation were carried out to compare with the first. The net hauls and the pump catches were analyzed by suitable dilution and subsampling (Clarke, 1933). Our thanks are-due Mr. David Bonnet for assistance in this task and in the preparation of the tables and diagrams. GENERAL FEATURES OF THE ZOOPLANKTON The number of species of zodplankton represented in our catches by one or more specimens is enormous (cf. Fish, 1925)—too numerous for detailed analysis in an investigation devoted primarily to the TABLE I ' Total number of Calanus finmarchicus tn hundreds calculated on a basis of a 30-minute haul. Date Station 1 Station 2 Station 3 Station 4 Station 5 1935 sf) 2A ik es 6 106 90 — ~- RLS Oe 17 65 142 5 37 Sle Saeeneeee 8 90 109 — 0 SS) caf Steam x 91 = BK 15 I rarer x 108 216 0 4 Se aca aks 11 152 123 — — ifile os Le 14 , 74 124 xi me 21) Ae ae 6 20 209 — x DS) aa ae 14 288 147 Xx x Deis 5 hh 9 529 267 5 3 JNU al as reais 84 385 232 3 0 Din esate 5 52 171 6 OF Lo) SN i aa vt 25 72 — — aera es 3 92 94 Bre x TOR crc: 8 137 50 Xx 0 2 OW adie: 0 28 113 0 0 Average....... 13.6 140.1 143.9 Sil tee — indicates station omitted. x indicates less than 100. study of the production of a few important forms. Even a cursory glance at the material, however, reveals that a considerable difference exists in the plankton at the stations near shore and at those farther out. At Stations 1, 4, and 5 large numbers of larval forms of both bottom-living and pelagic species were encountered mixed in with a varying number of mature individuals of truly planktonic types. The composition of this inshore population changed almost daily in contrast to the more gradual fluctuations which took place offshore. At Stations 2 and 3 fewer types were found in the plankton and, 472 GEORGE L. CLARKE AND DONALD J. ZINN although quantities of immature specimens appeared at certain seasons, these belonged mainly to the same species as the adult indi- viduals. The difference in the population offshore is well illustrated in Table I, in which the total numbers of Calanus finmarchicus at each station are set forth. Calanus was consistently scarce at Stations 1, 4, and 5 and, although the average number at Station 3 was only slightly greater than that at Station 2, a population of at least moderate dimensions was always to be found at the offshore station. TABLE II Total zoéplankton taken in scrim net at Station 3, 1935-36. Thirty-minute oblique hauls with 75 cm net. Approximate volumes after settling one month. Date Haul Vol. cc Av Date neu Vol. cc Av Septyesie. 63 181 <10 Whey 25.52 o00c 204 240 ; 182 <10 <10 205 160 | 200 Oct. 1 185 105 Junev ids sees: 206 270 186 70 88 207 215 | 246 Noa 2 (Ge, Dall ter 215 June 25....... 208 105 188 160 188 209 140 | 123 ID Yeroge (kis Saeco 189 140 140) uly OF 210 70 Deer ZO ee ee: 191 105S 211 90 80 192 1OO'S) 134 afuly e225 see 2D 45 [ta eA ena el etre 193 105 S 213 40 43 194 185S} 145 | July 30....... 214 240 | 240 Rely 2 Sine nose 195 700S| 700] Aug. 6....... 215 210 | 210 IMME Ye ATi tee ecdena 197 10 10 } Aug. 14....... 216 70 pie S)\ 32. {ae 198 40 217 70 70 199 65 58) || Abe ZO) os 65 5 218 210 IN pie Ore seca 200 115 219 160 | 185 201 115 115 | Aug. 31....... 221 70 Wiens) celia We eer o 202 65 BRE 40 55 203 90 Hou Septeslom rere 223 <10 224 10 10 Note: Hauls mostly copepods except those marked ‘S’” which were mostly sagittee. The net used catches effectively animals as large as, or larger than, cope- podid Stage IV of Calanus finmarchicus. Many smaller forms are not retained by the net. The seasonal variation in the total population may be traced from the approximate volumes of the zodplankton taken at Station 3 throughout the year (Table II). Plankton was scarce during Sep- tember and October, but increased in November and December. The mid-winter hauls were characterized by large quantities of sagitte. After this plankton again became scarce, but during the spring months volumes increased reaching a maximum early in June SEASONAL PRODUCTION OF ZOOPLANKTON 473 The plankton remained abundant through the summer with few exceptions until in September a sudden reduction was encountered. At all the stations Crustacea—usually copepods—formed the bulk of the catch with the exception of a few occasions on which large numbers of medusz or, at the offshore stations, of sagitte were encountered. The most common copepods, in addition to Calanus finmarchicus, which will be treated in a separate section, were the following (cf. also Fish, 1925): Centropages typicus......... July to Dec. Numerous in August Centropages hamatus........ May to July, and Numerous in June and in Sept. and De- July cember Pseudocalanus minutus... .. .Jan.to Oct. Numerous Mar. to Aug. Paracalanus parvus......... July to Sept. and Numerous in Aug. in Feb. Acariia tonsd.............. April to Dec. Otthona similis............ Aprilto Sept. and probably throughout the year Labidocera @stiva.......... Oct. to Dec. Many of these copepods appear to breed in this region because immature specimens of most of the list were found on one occasion or another. The copepodid stages of Centropages (species not de- termined) were particularly numerous during the summer months, and the locality is a veritable nursery for Pseudocalanus as judged by the large numbers of eggs, nauplii, and copepodites taken in the pump catches, particularly in the spring. Sagitta elegans was represented by at least a few specimens in almost every haul at Station 3 and the large catches of sagittz re- ported on several occasions consisted entirely of this species. Sagitta enflata occurred in small numbers in October and November, 1935, and in August, 1936. Sagitia serratodentata was taken from Sep- tember to December, 1935, and in August, 1936, and was abundant on only one occasion. It was noticed that the size of Sagitta elegans varied greatly from month to month and that the smallest individuals appeared in January, May, July, and September. Since these months agree almost exactly with the four periods of the year during which Russell (1932-33) believes the main breeding of Sagitta elegans to take place at Plymouth, we may conclude that the seasonal production of this species is approximately the same on both sides of the Atlantic. THE PRODUCTION OF CALANUS FINMARCHICUS Comparison of Net and Pump Catches The numerical analysis of our zo6plankton hauls was limited almost entirely to Calanus finmarchicus. The numbers of this species taken in the scrim net ran into the tens of thousands. If we assume 100 per cent 474 GEORGE L. CLARKE AND DONALD J. ZINN straining efficiency, the scrim net filtered 564,000 liters of water during the standard 30-minute haul. In the standard pump operation, on the other hand, the total volume of water delivered to the phytoplankton net in pumping from 30 meters to the surface amounted to 1134 liters, One copepod in the pump catch would theoretically correspond to about 500 copepods in the net haul provided that the two methods are equally effective in catching all stages of Calanus. This was not expected to be the case, but since a few of the older copepodites and adults were usually to be found in the pump catches, these older stages were counted as well as the younger stages in order that the sampling efficiency of the pump and the net might be compared. The total number of each stage of Calanus in the pump catches throughout the year is shown by the solid areas in Fig. 4. All the nauplius stages have been lumped together because the duration of each of these stages is short compared to the intervals between our observations. Very few of the earlier nauplius stages were taken. Certain eggs, which may have been those of Calanus, were found, but the number of these was so trifling as not to be worth plotting. The scarcity of eggs and early nauplii suggests that the actual spawning ground may be farther offshore than our station. The numbers of Calanus taken in the net hauls is indicated by the superimposed single line (Fig. 4), but using a different scale. The magnitude of the scale which has been used for the net hauls was determined by finding the ratio between the two sets of data for copepodid Stage IV. This stage was chosen for the comparison of the numbers taken with the pump and with the net because it seems to have been caught effectively by both methods. Since the number of Calanus in this stage was high both on July 30 and August 6, the average number taken by the net on these two days was divided by the average number taken with the pump. The resulting quotient was very nearly 100. Accordingly all the totals for the net hauls were divided by 100 before plotting. The scrim zodplankton net should theoretically remove animals from 500 times as much water as the pump. Since the volume of water passing through the pump was accurately measured, we reach the conclusion that the straining efficiency of the scrim net is 20 per cent. The two curves for Stage IV which have been brought together at one point in the manner described are seen to correspond very roughly for the rest of the year. For copepodid Stage V and for the adults, however, the curves for the pump catches consistently lie below those for the net hauls. This indicates that the latter method is relatively more effective than the former for these larger individuals either OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 200 NUMBER OF COPEPODS fe) 400 200 (eo) - 600 400 200 ‘0; 800 600 400 200 200 () O OCI NOV DEC AN GES em ViAh Arh MAY OUN i 00a AUG an SE Fic. 4. Thesolid black area indicates the total numbers of Calanus nauplii (N), copepodid stages I-V, and adult 9 and @ taken with the pump per standard haul (see text). The superimposed line shows the numbers of Calanus taken with the scrim net per standard haul, but for these the scale must be multiplied by 100. 476 GEORGE L. CLARKE AND DONALD J. ZINN because fewer slip through the mesh of the net or because more escape the mouth of the hose. The opposite situation is seen in the case of copepodid Stage III in which it appears that a disproportionate number of this group fails to be caught by the net. Seasonal Changes in Abundance Nauplii occurred in significant numbers on only three occasions, namely March 27, April 13, and May 11. Copepodid Stages I and II have peaks of abundance on April 13, May 11, and June 25. Stage III was numerous also on April 13 and May 11 and a third period of abundance extended from the end of June to the first part of August. The numbers of individuals in Stage IV were moderate in April and May, increased rapidly in June, and reached a maximum early in August. The net caught representatives of Stage V in every haul throughout the year, but with the exception of December 29, numbers were extremely low from September until the end of April. This stage was particularly abundant at the end of May and again in August, but never equalled the maximum of Stage IV. Adult specimens of Calanus were most numerous in May and June, but at all times were relatively scarce, particularly the males. Taking the species as a whole, we may conclude that the population is of very small dimensions during the autumn and winter. By the first of April reproduction has begun and from then on through the summer Calanus is abundant. At some time after the end of August the species suddenly becomes depleted and the population is not restored again until the following spring. Succession of Generations To determine the number of generations? which succeed one another during the year recourse is best made to the percentage dis- tribution of the various age stages within each of the hauls from season to season because this procedure removes the confusing effect of fluctu- ations in the size of the total catch and allows the progression from stage to stage to stand out in relief. The percentage diagram for the pump catches (Fig. 5) shows that on the first of October the total catch ’The term “generation” is used here instead of ‘‘brood,” which has been employed by others in this connection, because the former word expresses correctly the relationship of the two groups of copepods which appeared during the course of the year. The term ‘‘brood” should be limited to its strict sense in order to avoid confusion. Nicholls (1933) regards it possible for each female copepod to produce more than one brood of ova. Accordingly there may be circumstances in which the young of the second generation exist contemporaneously with a late second brood of the first generation. The necessity for using these terms with their exact meanings thus seems obvious. SEASONAL PRODUCTION OF ZOOPLANKTON 477 Uf OCT NOV DEC JAN FEB. MAR APR OG NOV DEG JAN BEB S MAR Fic. 5. Percentage distribution of Calanus nauplii, copepodid stages I-V and adult @ and & taken with the pump at Station 3 during 1935-6. consisted of 66 per cent Stage V and 34 per cent Stage IV. During November and December practically 100 per cent of the species occurred as Stage V. But in January some of these individuals matured into adult males and in February an even larger proportion of 478 GEORGE L. CLARKE AND DONALD J. ZINN females appeared. A month later the males and females had largely disappeared and Stage V was reduced to 1 per cent, while nauplii comprised 60 per cent of the catch and copepodid Stage I amounted to 30 per cent. During April other copepodid stages appeared in suc- cession, but these gave rise to only a very small percentage of adult females and practically no males. In May asecond series of peaks is to be found progressing up through the copepodid stages and culminating in a peak for the adult males on May 25 and a prominent peak for the females on June 11. The relationship between these two series of peaks in April and May is obscure. Although we know that in other areas two or more generations may follow one another in rapid succession during the spring and summer, the time interval between our two series appears to be too short for them to represent successive generations. In labora- tory experiments Nicholls (1933) found that 27 days was the minimum time possible from the shedding of the ova to the appearance of the adults at the temperature at which he worked (apparently 11—14° C.). The observations of Ruud (1929), Lebour (1916) and Fish (1936) indicate that a longer time than this is required for the entire develop- ment of Calanus. Marshall, Nicholls, and Orr (1934) state that “the eggs appear after two to four weeks, spawning may last for several weeks, and the adults then die out.’”’ Taking fourteen days as the minimum time for the maturation of the eggs and 27 days as the minimum for development, it appears that 41 days is the shortest interval which can exist between the spawning of one generation and the spawning of the next—or between the two generations at any corresponding stage of development. At certain stages in our April and May series the peaks are less than 30 days apart and the tempera- ture of the water at the time was only about 6-8° C. It seems very doubtful, therefore, that the copepodites appearing in May were produced by the individuals which were found as copepodites in April. It is possible that the May group is a late second brood from the same parents as the April group. Adult specimens of Calanus were present in the water throughout April although their numbers were low. Another possible explanation of the situation is that at the end of April an unusually extensive movement of water along the coast swept away the population which had reached Stage IV on April 25 and brought in an entirely distinct population of copepods whose develop- ment was two or three weeks behind that of the first group. The reduction in the abundance of the first group when it reaches Stage V and its practical disappearance thereafter support this idea (cf. Fig. 4). Another group of Stage I copepodites appeared late in June and SEASONAL PRODUCTION OF ZOOPLANKTON 479 continued into July, but in lesser numbers. In the succeeding weeks the older copepodid stages became numerous one after another until by August 20 a prominent peak had been built up in Stage V. This summer generation was undoubtedly produced by the individuals which we observed as mature adults in May and June, although the JUN , JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP % 20 100 " gal JUN JUL AUG SEP OCT NOv DEC JAN FEB MAR APR MAY JUN JUL AUG SEP Fic. 6. Percentage distribution of Calanus copepodid stages III, IV, and V and adult 2 and @ taken with the scim net at Station 3 during 1935-6. eggs and nauplii which should have appeared in June were not found. Evidently the spawning took place during the two-week’s interval between our visits to the station in June, or else at a point farther off- shore. There is some indication in the diagram that this summer 480 GEORGE L. CLARKE AND DONALD J. ZINN generation consisted of more than one brood, but the irregularities were so great that no definite conclusion can be reached. It seems clear, however, that Stage IV moulted only slowly into Stage V during August and that these individuals did not mature into adults im- mediately, but formed the population destined to carry over the autumn and winter as Stage V. The catches with the scrim zodplankton net have similarly been calculated on a percentage basis to serve as a check on the conclusions reached with the pump catches (Fig. 6). The numbers of the older stages of Calanus taken with the pump were low and the possibility existed that the larger individuals escaped the 1-inch opening of the hose, whereas loss from this cause would be much less in the case of the scrim net. Since the net hauls were begun four months earlier than the work with the pump, we have a record for two summers and the intervening winter. At the end of June, 1935, relatively large numbers of adult males and females were found. The spawning of these produced the summer generation which appeared as Stage III copepodites in July. This generation formed the winter stock which did not mature until January, February, or March. Individuals in Stage III appearing in the net hauls in April (with a subsidiary group in May) constituted the spring generation of 1936. This generation matured and spawned in June and we find the summer generation as Stage III in July, thus nicely confirming the net haul results of the previous year. As before, these individuals grew slowly during the summer, and when work was discontinued in September, the population was about equally divided between Stage IV and Stage V. From both the pump catches ° and the net hauls we conclude that at Station 3 a short-lived generation of Calanus occurs during the spring and is followed by a long-lived generation which has its origin in the early summer and lasts through the following winter. One or more subsidiary groups were found during the spring and summer which may represent second broods from the same adults, or immi- grants in a different stage of development, or possibly additional generations of very short duration. These two main breeding periods correspond closely with those observed by Fish (1936) in the Gulf of Maine for the ‘“‘western stock”’ 4 During the summer of 1937, however, incidental observations showed that the Calanus population was extremely small in this region and consisted mostly of Stage V. Evidently annual variations may be considerable. 5 The fact that the pump catches obtained through a 1-inch hose gave the same general picture of the production of Calanus as the net hauls suggests that a device with a relatively small opening is adequate for plankton studies and could be more widely used (cf. Hardy, 1936). Such a device would have the important advantage that the volume of water strained could be accurately measured. SEASONAL PRODUCTION OF ZOOPLANKTON 481 of Calanus. Fish believes that this stock forms the principal source of supply for the Gulf and he reports that these animals breed chiefly in March-April and in June-July, but possibly also in September. The developmental period of two and one-half months which Fish calcu- SEP OCT NOV DEC JAN FEB MAR APR MAY. JUN JUL AUG SEP’ CEPHALOTHORAX LENGTH SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP Fic. 7. Cephalothorax length of Calanus copepodid stages III, 1V, and V and adult 9 at Station 3 during 1935-6. The shaded area indicates the range and the heavy line the mean of 25 measurements. Since males were sufficiently numerous to measure on only five occasions, the limits of the range and the mean are indicated by the co’ symbols placed on the upper diagram. It will be noted that the scales of the upper and lower diagrams overlap to a certain extent. lated for his Calanus agrees well with the approximate interval between our spring and summer generations. Marshall, Nicholls, and Orr (1934) found that in the Clyde Sea three generations of Calanus occurred during the course of a year. The spawning periods for these 482 GEORGE L. CLARKE AND DONALD J. ZINN generations took place in March, May, and June. Our observations differ only in that the batches of nauplii appearing off Woods Hole from March to May seemed to belong to the same generation. Bogorov (1934) reports that three generations occur in the Plymouth area. He believes spawning to take place in February-March, May, and intermittently from June to August. Other investigators (Paulsen, 1906; Bigelow, 1926; Farran, 1927; Russell, 1928; Ruud, 1929; and Sdémme, 1934) have found a similar succession of two or more genera- tions in their respective areas, but since thorough reviews have been made by Fish (1936) and by Nicholls (1933), their results will not be discussed here.® Variation in Size The cephalothorax length of 25 specimens of copepodid Stage V was measured for each haul throughout the year and for Stages III and IV and for the adults on those occasions when these groups were suffi- ciently numerous (Fig. 7). The range of the measurements for each group in each haul was greatest for the adults and progressively less for the younger stages. The average length of Stage V did not fluctuate widely but was found to be somewhat greater from January to May than during the rest of the year. Stage IV similarly diminishes in size after the spring months have been passed.- The adult males and females averaged about 2.65 mm. in length from January through > April, but in May a sudden drop occurred to about 2.45 mm. and this smaller size persisted through the summer. The abrupt reduction in the size of the adults in May coincides with the appearance in the hauls of the mature males and females of the new spring generation (Figs. 4and 5). These individuals were spawned at the end of April or early in May when the temperature ranged between 5° and 9° C. We have already seen that the generation which was produced in June at temperatures of 11° to 15° C. apparently did not reach maturity until the following January or February. Conse- quently, the summer generation resulted in larger adults than the spring generation although the latter group was spawned at a tempera- ture about 6° C. lower. This relationship with temperature is just the reverse ’ of that reported by Marshall, Nicholls, and Orr (1934) and by Bogoroy (1934). These investigators found that the individuals of the spring generation were the largest of the year and Bogorov reported 6 Results of a similar nature have been reported by Wimpenny (1936) in a publication which appeared since this paper went to press. ’ The adult females which were taken in April and which were of large size may possibly have been spawned in March and, if so, present a case of correlation of large size and low temperature during the spawning period. ae - ee SEASONAL PRODUCTION OF ZOOPLANKTON 483 that the females of this generation were 1.2 times larger than those of the autumn-winter generation. Comparison with Phytoplankton and Nannoplankton The phytoplankton obtained in the water samples taken at 2, 15, and 30 meters on each occasion has been analyzed by Lillick (1937). She reports that during the summer of 1935 the phytoplankton was dominated first by Chetoceros and associated forms and later by Rhizosolenia. On October 1 the diatoms reached a maximum.® During the rest of the autumn and early winter the phytoplankton was extremely poor. Latein January numbers increased somewhat but by the end of February the phytoplankton was scarce again. On no occasion was a typical spring diatom maximum encountered but Miss Lillick cites evidence for believing that a great flowering of diatoms took place early in February at a time between two visits to the station. During the late spring and summer of 1936 the phytoplankton re- mained at a low ebb, with Guinardia appearing as the dominant form, until August 9 when for a period of four days a flowering of Rhizosolenia and Guinardia occurred. Diatoms were the most prominent element of the flora at all times in this region, but dinoflagellates appeared in smaller numbers through- out the year especially in summer and fall. In addition Miss Lillick lays stress upon a group of flagellated and pigmented forms which appeared in all the samples in very large numbers. Although this group was made up entirely of exceedingly small forms, she states that their frequency was so great as to make them significant in the phytoplankton population and she believes that “‘they must doubtless play an important part in the general food cycle of the region.”’ Special water samples for the study of the bacterial population were taken by Dr. C. E. Renn at 5-meter intervals from the surface to the bottom on each trip to Station 3 from October 1, 1935 to May 26, 1936. The numbers of cells per cc. as determined by plate counts ranged from 30 to 350 during the first three months but from February to May no counts greater than 33 were obtained (Renn, 1937). Although the actual numbers of bacteria in the sea may have been a thousand times greater than this, we have shown elsewhere that populations of these dimensions cannot be a significant source of food for Calanus (Fuller and Clarke, 1936). Besides the bacteria, and the pigmented forms 8 Our observations on the zooplankton were not made at sufficiently short inter- vals to determine conclusively whether or not this flowering coincided exactly with a marked reduction in the number of grazing animals (cf. Harvey, Cooper, Lebour, and Russell, 1935). In 1935 a sharp drop in the number of Calanus occurred on Nov. 2; in 1936 a similar diminution of numbers took place on August 31, 484 GEORGE L. CLARKE AND DONALD J. ZINN mentioned above, to which Miss Lillick confined her attention, other nannoplankton types undoubtedly existed—possibly in significant quantities. The presence of great numbers of these, during the summer months at least, has been demonstrated by Lackey (1936). In most routine investigations many of the smaller and more delicate forms are probably overlooked because the usual preservatives destroy them or render them unrecognizable. Since Marshall, Nicholls, and Orr (1934) found that in Loch Striven the periods of diatom increases coincided with the three main spawning periods of Calanus, it is interesting to examine our data for similar correlations. For this purpose observations should have been taken at much shorter intervals than was possible in the present TABLE III Phytoplankton at Station 3, March to June, 1936. The number of cells per liter at 2, 15, and 30 meters have been averaged. Mar. 27 Apr. 13* oP May 11* ey. Jute June 25 Diatomseeee ee ae 33 9,450 600} 4,700 {1,467 |2,300 700 Dinoflagellates.... 67 100 1,300 0 67 0 3,767 “Flagellates’..... 300 5,750 {1,433 950 {2,633 |7,533 6,767 Other forms...... 0 350 OF uae 67| 100 167 MOA o eee 400 iS}G50N 3,059 5,650 4,234 |9,933| 11,401 Calanus Nauplii Abundant |Abundant Abundant Copepodid ESN eve sacra Abundant Abundant Abundant * Average of two depths only. investigation. Conclusions are also made difficult in our case because of our lack of information in regard to the state of the fauna and flora in contiguous areas and the effect which horizontal movements of water masses would produce. It is clear, however, that whatever diatom flowering may have occurred in the early spring months between two visits to the station, this could not have served as a food supply for our Calanus since nauplii did not appear until March 27. On this date and throughout the ensuing four months diatoms were not abundant. Similarly no production of Calanus was observed at the times of the secondary diatom maximum in August or October. Thephytoplankton which did exist during the breeding periods of Calanus may be studied from Table III. Diatoms were relatively numerous on April 13 and May 11, occasions when nauplii and early copepodites were abundant, SEASONAL PRODUCTION OF ZOOPLANKTON 485 but they were scarce on March 27 when the first nauplii of the season appeared and on June 25 when the early copepodites of the second generation came into prominence. On this latter date dinoflagellates were especially numerous. The forms designated as ‘‘flagellates”’ were abundant throughout this period. Although our data are not sufficient to allow us to conclude which are the essential food organisms for Calanus nor in what quantity they are required, nevertheless we have obtained certain facts which limit the possibilities. We have seen that the spawning periods of Calanus occurred at times which did not coincide with diatom maxima. We know from Miss Lillick’s analysis that throughout the year the diatoms and dinoflagellates appeared suddenly at intervals of a week or two, flourished for a few days only, and then disappeared again. The “flagellates,” on the other hand, although small in size, were found in relatively large numbers on all occasions. If Calanus feeds chiefly on diatoms, we must conclude either that the small number of cells always present as a minimum is sufficient for their nutrition, or that the food obtained by the copepods at times of local flowerings can be converted into reserve tissue (e.g. oil) which will tide them over until the next period of diatom abundance. The number of diatoms present between flowerings seems too low to fulfill the nutritive requirements of Calanus (Fuller and Clarke, 1936; Fuller, 1937) but further investi- gations may show this not to be the case. As regards the second possibility, we know that Calanus, in the later copepodid and adult stages at least, can live for a week or two without food, but is unable to moult successfully under these conditions (loc. cit.). The copepods might thus be able to survive without moulting from flowering to flowering—and this may be the actual situation during the autumn and winter—but rapid growth such as took place from April to July would probably be impossible unless adequate nutriment was con- tinuously available. A third possibility which our present observations suggest is that the nannoplankton is important as food for copepods. _ The “‘flagellates’’ which were found at Station 3 at all times in large numbers would represent only a small amount of substance because of their minute size, but if, as seems likely, these pigmented forms are only a small fraction of the total nannoplankton, the latter may turn out to be significant as a food source. SUMMARY 1. The seasonal production of zodplankton, particularly of Calanus jinmarchicus, was investigated by means of scrim nets and a plankton pump at five stations in the vicinity of Woods Hole during the summer 486 GEORGE L. CLARKE AND DONALD J. ZINN of 1935 and at one offshore station throughout the ensuing year. Collections of phytoplankton and of nannoplankton and measurements of temperature, salinity, phosphates, nitrates, and illumination were carried out at the same time. 2. The zodplankton consisted largely of copepods, but medusz, sagitte, and, at the stations near shore, larve of both benthonic and planktonic forms occurred irregularly. Sagitta elegans exhibited four main breeding periods during the year. 3. The numerical analysis of the Calanus population revealed the presence of nauplii in significant numbers in March, April, and-May, and of early copepodites in April-May and June-July. Stage V was found at all seasons. Adult specimens were relatively scarce at all times, but most abundant in May and June. The species as a whole was reduced to small numbers during the autumn and winter. 4. The pump catches and the net hauls agree in indicating that a short-lived generation of Calanus occurs during the spring and that this is followed by a long-lived generation. The latter has its origin in the early summer, passes through the autumn and winter as Stage V, and matures early the following spring to give rise to the next short- lived generation. 5. The measurement of the cephalothorax length of 25 specimens in each stage older than copepodid Stage II showed that the average lengths of the various groups did not fluctuate widely during the year except in the case of the adults which exhibited a sudden drop in size in May. 6. The spawning periods of Calanus did not occur at times of diatom maxima, and therefore the two phenomena are not directly related in the present case. In regard to the nourishment of this copepod we must conclude (a) that the small number of diatoms always present as a minimum is sufficient, or (0) that the animals build up sufficient reserve on occasions of small local flowerings of diatoms, which occur at intervals of a week or two, to tide them over the intervening periods, or (c) that ‘‘flagellates,’’ which were found continually in large numbers, and other types of nannoplankton are important as a food source. BIBLIOGRAPHY ALLEE, W. C., 1919. Note on animal distribution following a hard winter. Biol. Bull., 36: 96. BiceLow, H. B., 1926. Plankton of the offshore waters of the Gulf of Maine. Bull. U. S. Bur. Fish., 40 (Part 2): 1924. (Doc. No. 968). BiGELow, H. B., AND M. Sears, 1935. Studies of the waters on the continental shelf, Cape Cod to Chesapeake Bay. II. Salinity. Papers in Physical Oceanography and Meteorology, 4 (1): 1. Bocoroy, B. G., 1934. Seasonal changes in Biomass of Calanus finmarchicus in the Plymouth Area in 1930. Jour. Mar. Biol. Ass’n., 19: 585. SEASONAL PRODUCTION OF ZOOPLANKTON 487 CLARKE; G. L., 1933. Diurnal migration of plankton in the Gulf of Maine and its correlation with changes in submarine irradiation. Biol. Bull., 65: 402. CLARKE, G. L., 1934. The rdle of copepods in the economy of thesea. Fifth Pacific Science Congress. Vancouver, B. C. AS, p. 2017. CLARKE, G. L., 1938. Seasonal changes in the intensity of submarine illumination off Woods Hole. cology (in press). CLARKE, G. L., AND S. S. GELLIS, 1935. The nutrition of copepods in relation to the food-cycle of the sea. Biol. Bull., 68: 231. FarrAn, G. P., 1927. The reproduction of Calanus finmarchicus off the south coast of Ireland. Jour. du Conseil, 2: 132. Fis, C. J., 1925. Seasonal distribution of the plankton of the Woods Hole region. Bull. Bur. Fish., 41: 91 (Doc. 975). Fisu, C. J., 1936. The biology of Calanus finmarchicus in the Gulf of Maine and Bay of Fundy. JBzol. Buil., 70: 118. FULLER, J. L., 1937. Feeding rate of Calanus finmarchicus in relation to environ- mental conditions. Biol. Bull., '72: 233. Futter, J. L., anp G. L. CrarxKe, 1936. Further experiments on the feeding of Calanus finmarchicus. Biol. Bull., '70: 308. Hatcat, F. J., 1936. Currents in Narragansett Bay, Buzzards Bay and Nantucket and Vineyard Sounds. U. S. Dept. Commerce. Coast and Geodetic Survey, Special Pub. No, 208. Harpy, A. C., 1936. Thecontinuous plankton recorder. Discovery Reports, 11: 457. Harvey, H. W., L. H. N. Cooper, M. V. Lesour, AND F.S. Russgex, 1935. Plank- ton production and its control. Jour. Mar. Biol. Ass’n., 20: 407. Lackey, J. B., 1936. Occurrence and distribution of the marine protozoan species in the Woods Hole area. Biol. Bull., '70: 264. LeBour, MARIE V., 1916. Stages in the life history of Calanus finmarchicus (Gun- nerus), experimentally reared by Mr. L. R. Crawshay in the Plymouth Laboratory. Jour. Mar. Biol. Ass’n., N.S., 11: 1. Liuicx, L., 1937. Quantitative and qualitative studies of the phytoplankton of Vineyard Sound, Mass. Biol. Bull., 73: 483. MarsHALL, S. M., A. G. NICHOLLS, AND A. P. Orr, 1934. On the biology of Calanus finmarchicus. V. Seasonal distribution, size, weight and chemical compo- sition in Loch Striven in 1933, and their relation to the phytoplankton. Jour. Mar. Biol. Ass’n., 19 (2): 793. NicHotts, A. G., 1933. On the biology of Calanus finmarchicus. I. Reproduction and seasonal distribution in the Clyde Sea area during 1932. Jour. Mar. Biol. Ass’n., 19; 83. PAULSEN, O., 1906. Studies on the biology of Calanus finmarchicus in the waters around Iceland. Meddel. fra Komm. for Havunders, geleser, 1 (4): 1. Renn, C. E., 1937. Conditions controlling the marine bacterial population and its activity inthesea. (Abstract.) Jour. Bact., 33: 86. RussELL, F. S., 1928. The vertical distribution of marine macroplankton. VII. Observations on the behavior of Calanus finmarchicus. Jour. Mar. Biol. Ass’n., 15 (2): 429. RussELL, F. S., 1932-33. On the biology of Sagitta. Pts. I-IV. Jour. Mar. Biol. Ass’n., 18: 131. Ruup, J. T., 1929. On the biology of copepods off Mére, 1925-27. Rapp. Proc. Verb., Cons. Internat. Explor. Mer., 56: 1-84. SgmmE, J. D., 1934. Animal plankton of the Norwegian coast waters and the open sea. I. Production of Calanus finmarchicus (Gunner) and Calanus hyperboreus (Froyer) in the Lofoten Area. Fisk. Skrtft., Ser. Havunderso- kelser, 4: 1. Wimpenny, R. S., 1936. The distribution, breeding and feeding of some important plankton organisms of the south-west North Sea in 1934. Part I. Fusher- tes Invest., Ser. II, 15: 1-34. SEASONAL STUDIES OF THE PHYTOPLANKTON OFF WOODS HOLE, MASSACHUSETTS}? LOIs C. LILLICK (From the Department of Botany, University of Michigan, and the Woods Hole Oceano- graphic Institution, Woods Hole, Mass.) During the summer of 1935 certain investigations of the plankton in the waters near Woods Hole, Massachusetts, were undertaken by workers at the Oceanographic Institution. The zodplankton studies have been described in the preceding paper by Clarke and Zinn (1937); the observations on phytoplankton form the subject of the present paper. At the outset five stations were established for regular obser- vations, four in Vineyard Sound, and one in Buzzard’s Bay. For the phytoplankton work water-bottle samples were taken at each station twice weekly, and at three depths, surface, mid-depth, and bottom. Along with these, determinations of the temperature, salinity, nitrate and phosphate content of the water were made. During the first two months the water at all five stations was found to be so homogeneous in nature with regard to the phytoplankton, that it seemed unprofitable to continue observations at all stations. Therefore all subsequent collections were made at Station 3 alone, which is located at the Whistle Buoy near the western entrance to Vineyard Sound, Latitude 41° 17’ 35’’, Longitude 71° 0’ 0’” (see Clarke and Zinn, 1937, for map). This station was chosen because conditions there, more nearly than at any of the others, simulate those of open waters, as will be seen later. From September, 1935, through the winter months samples were taken once a month; from April through July, fortnightly; during August, weekly. The final phytoplankton data were collected on August 20, 1936, giving a range of fourteen months for the entire survey. Samples were taken at the 2-meter, 15- meter, and 30-meter levels, the bottom being at 32 meters. In this paper only such forms have been included as fall naturally within the plant classes, the chlorophyll-bearing or food-producing groups, which include the Cyanophycex, Diatomacex, Dinoflagellate, Stlicoflagellate, Coccolithinex, and certain of the pigmented flagellated groups. All of the true protozoan classes have been omitted. In making the quantitative determinations of the phytoplankton 1 Papers from the Department of Botany of the University of Michigan, No. 629; Contribution No. 156 of the Woods Hole Oceanographic Institution. 488 PHYTOPLANKTON OFF WOODS HOLE 489 for the majority of the samples Gran’s centrifuging method was used, modified somewhat by having the organisms concentrated over a mem- brane filter, and then centrifuged. Since by this method others have found that there may be a 10 per cent loss of organisms, an original sample of 1,100 cc. was used instead of 1,000 cc. to allow for this error. For the winter and spring samples the precipitation method of Nielsen and von Brand (1934) was used chiefly because the amount of sample available was not sufficient to permit the use of the other method. Most of the collections were made by Mr. Donald Zinn and the crew of the Oceanographic Institution boat Asterias. Chemical determina- tions were made by Dr. Homer P. Smith and Mr. Bostwick Ketchum. The writer is indebted to Dr. H. B. Bigelow and Dr. F. K. Sparrow, Jr., for certain guidance at the Woods Hole Oceanographic Institution, and to Dr. William R. Taylor and Professor H. H. Bartlett under whose direction the work was completed at the University of Michigan. HYDROGRAPHIC OBSERVATIONS Certain physical features of the water, currents, tides, temperature, and salinity, play an undoubtedly important part in the distribution of plankton. The observations made on these factors during the present investigation have been presented by Dr. Clarke (Clarke, 1937; Clarke and Zinn, 1937). Although the control of phytoplankton species distribution, at least in part, must be attributed to the influence of temperature and salinity, no obvious correlations could be made between the variations in these factors and in the seasonal distribution of phytoplankton. Nitrate determinations ranged from 4 milligrams of N. as NOs per cubic meter of sea water at 15 meters on October 1, to 62 milligrams at 2 meters on December 29 (see Figs. 1, 2, and 3). Through the summer months of the first year, the nitrate content of the upper strata of water was low, exhibiting slight fluctuations. Somewhat higher values were obtained in the lower strata. During the winter nitrates were gradually built up throughout the entire water mass to their highest concentration in early January. It is noteworthy that there is a decided lag in this accumulation of nitrates behind that of phosphates, a feature which is consistent with Cooper’s theory (1933) based on observations in the English Channel, that the phosphates are returned directly to the water upon the decomposition of the plankton; whereas the nitrates are regenerated only after a series of intermediate steps. Terminating the winter accumulation, there was a sudden dropping off in the amount of nitrates in late January and early February, followed by a more gradual decrease throughout March and April. The summer was marked by small amounts of nitrates with 490 LOISJ@G. LILLIcCk PHOSPHATES PHYTOPLANKTON TEMPERATURE Fic. 1. Distribution of nitrates, expressed as milligrams of N as NO; per cubic meter of water; phosphates, as mg. of PO, per cubic meter; phytoplankton, in thousand cells per liter; and temperature, in degrees centigrade, at 2 meters depth, Station 3, from July, 1935, to Aug., 1936. PHYTOPLANKTON OFF WOODS HOLE 491 PHOSPHATES PHYTOPLANKTON TEMPERATURE Fic. 2. Distribution of nitrates, expressed as milligrams of N as NO; per cubic meter of water; phosphates, as mg. of PO, per cubic meter; phytoplankton, in thousand cells per liter; and temperature, in degrees centigrade, at 15 meters, Station 3, from July, 1935, to Aug., 1936. 492 BOISTE, VILLI CK PHOSPHATES PHY TOPLANKTON TEMPERATURE ° Fic. 3. Distribution of nitrates, expressed as milligrams of N as NO; per cubic meter of water; phosphates, as mg. of PO, per cubic meter; phytoplankton, in thousand cells per liter; and temperature, in degrees centigrade, at 30 meters, Station 3, from July, 1935, to Aug., 1936. PHYTOPLANKTON OFF WOODS HOLE 493 only slight fluctuations. In August there was an accumulation to © form a second lesser peak, which again lagged behind the phosphates, and which was followed immediately by a decrease in concentration. Variations with depth were never extremely pronounced. However, there was usually some degree of stratification. From May through August, the nitrates tended to become concentrated at the bottom. By September they were more or less uniformly distributed, and remained so until December, when the surface waters became very rich in nitrates, the bottom waters much lessso. This general relation- ship continued through April. The phosphate distribution differed somewhat from that of nitrates. The total amounts found in the water were always much higher, | though the total range was greater, from none in the surface waters through most of August and September, to 112 mg. PO. per cubic meter at the surface in December. The seasonal phosphate cycle was as follows: low through July and August of the first summer; increasing through September and October; reaching a maximum during the winter months of November, December and January; decreasing from the end of January through February and March, and dropping to the lowest point for the year in April, remaining low through May and June; in July and August gradually increasing, with another drop at the end of August. Stratification of phosphates was frequently much more pronounced than that of nitrates. From June through August the phosphates were found to be more concentrated at the bottom. In October the waters became more nearly uniform through- out, and remained so until spring, when the amounts at the surface increased over those at lower layers, this relationship remaining until June. The possible correlation of phytoplankton production with nitrates and phosphates will be discussed later. PHYTOPLANKTON (See Tables I and II) The larger part of the phytoplankton in our collections was made up of diatoms, dinoflagellates, and flagellates. During certain seasons of the year members of the Silicoflagellatee and Coccolithinez became rather numerous, and stray members of the Cyanophycez appeared in the collections on occasion. More than 100 different species appeared during the year, divided among the several groups as follows: Diatompacez tee. Vey Ae prone sprees. co sue 57 Dinoflagellatzes che yy See res oe ae 31 Silicoflagellates cers tan pews recone atin aes 3 Coccolithinesss7 8 ales Saeco ees 2 Cyanophy ceceintrteiectac seisenohe ae oii dens « 22 Blagellatesac cerita ae cte ee a adie ? EOS (Gr LILLick 494 000°6€ | 00E*4'|00Z | 00S | 000'8Z OE oos'e |o0e |ooe | ‘°'* | 008. Ce ee ea Oe eee Oe 000‘0¢ oe Sie ee 00 Gees ,005 ign teal ans So. Lapa ye eees ro 45 00E'T 008‘S | 009‘¢ | 009‘LT 00€ |009 | 099 weve [ores | gga OOL em OO each 00¢ | 009 | 00T 007‘2 006 00S 00E*Z 0L0°7 | 000'TZ 006 | 189s | see's 6IE | 996 cos‘z | ces'z PLL | P8T'e eee eee ialeca Uo cckal Were ices cop | 942 9 | coEz Se0'T | 784'97 89€ 90€'FT 82¢ | 801'¢ $96'CP PL3'F | | 968'T 8F0'7 | 8OL‘9¢ 967'T | SET oF | H8I 00S | 781 82Z | 86S L@-uefl 162 °29d | T99d | Z°AON | 1°390 «| ¢'3deS]9z‘3nVy| ct “sny 49} Ted sijap O00'TT | OFZ'T | OOF 000°9€T| 000'SE] 000‘9T 000‘681 Sadi: OxcteOmoms oN. Oubed cacr bec siajou 7 7e [eIO], cae e Pe g9e Ue oes ee 961'T O&¢ eo 90S Dia alecee “tes | og ele 081 oO Tsny | sc A[nf | cy Aine ¢ 4qnf SIZ ‘T SBOoooCKn HOO oe re roo o Doo oo oo 6 8 oie} oie) (| Oy OPO. ROR O Or OD Onde C86 OUDaeteto OF ad Og OG elteydsoor1AG Gro \[Skolo oNduo prOsted ssenen Okottdkekead’o aro IUTYI]0I90-) een i pinta CceO-Ceamehig a chnmose.5 unjnrads snunydaysiq sploWeiherye oll buset(emtensentertaigeiee fale al Taree Seeks eateries LL[aseyooris eieimommeltane d ltetomencsyretienohicNsircMteWm oyiais sh tcerep Licata ee WNIaINIS * OST 9 Oat) DeO Oo Deu"o-Osoeod Nooo WNnUAM WUNMUBIOLOAT Bats pam | DM ov A owalecichkctaice -felmriin eesoeome: Doering a9 3 eto xajqdulas "if GAO OOOO OOOO MOO OO Od ‘dds mnMpLday O6e en ee ee: SISUAADULD] XDINDKU0D Stn 0 one hook osnadernT eaten ny tiso cece oseke coos 2 e][eseyourq CC CRISES ES LUPIOLYSUIPAO NT DAISOUWSDIDY T, $96 1 os * SapLovyISajvU DUAUOISSDIDY T, G: beoNG. — Io Saonqstieg a O20 .neb oF NOC 40 UINIDISOD DULAU0IA1095 OL aes en eee oars purgsauas DIDIDGaY “W S000 |loo-Dobodopaoan oo oU Oo OU DUassyIsDAL AV) 60 00 0856 ono O-O6.p.0.8 4 00.00 0-0 SUA “T &%.d Ml'S-dun-piGa 0, a woes snovunp snapuyir0jdaT OOM OOO Gd God wOO-o-0 6.04 ppvony, DIPADUIN®) COU D-o Ose Oo cob Oo D-o5 ‘Dooe So XUAISKY UOAYJOL07Z) 020900 OSG D608. 000 8G OA. ON SOS sSnui pip ‘YD OLZ Cono oe on iano Goo ood O00 0 suardigap YD BOO modo gonoOn moon ooo oo a5 $110940Q YD Ce i i i ) dds SOLBI0JLY 7 a ey 1haq4Darg DIJAU01AAIS V Fe ee ew eee ee ee VIOCUIO}LLIG] soloed ee ee on ae, Se SS 00888 SS OOOO OOOO ‘OC6T “Isnsny ysno1yy ‘cesT ‘A[N[ ‘¢ uorzeIS ‘siaj0ur ST 32 sotsads uo0jyuejdozAyd jueqsodur aiour ay} Jo uoOlyNqI4ysIp Teuoseas [ HIV PHYTOPLANKTON OFF WOODS HOLE 008 |00€ | 000‘ZT| 00E'T | 006'S | O0E'E | 0026 | OOO'TT| 00Z'S 000'Z | 008°9 | 00¢ 000'Z | 006'2 | 000'TT| OO0'T | * 006‘'T | 000'6Z| O02 OST'L | 008‘Z | 009'Z | 00S‘z | 000‘9Z | 009‘ST | O0S‘Z | 002 008‘z | 00¢'z | 00S OOF'T | 007‘ 00Z‘T | 00Z‘T | 00Z‘OT | 000'T 008‘T | 00S‘6 000'T OO ae ae visite 009 00z2 | 002 OOT OOT | 008 002 | 002 Ae es MG | 007 00s 00S | 00¢ 000'T bo lee 00e 00r 004 | 004 004 | 008. 009 00z't | 008 | | 008. 008'81 Ee: FI sny | 9sny | zz Aint | of Anf | sz cunt JayWT Jed sjjad Wee, Tp eunf | sc Ae | tt APN | Sc adv | et Ady | 17 4eIN | £7 Gea sJ9joul Z 1 [BIO], S00 0CU bo GO CO OU DOO 0.00 K6 HO 2IETaSeLy Oto m.obdad inp o ooo Doorn uo tod on elieydsoorsAG O08 CO Cdio-o co Aum DMO BI OOO IUTY}I]OII07, Gath 020 OXOLD On G-Olo, G@-anc"o wunjnoads snunydasiq OD OOOO 10 520.5000 0,5 oo ho Ud BLIL][ISEYOOT|IS quid! Ds BAOsoScNOSONO. OG; Qudvbed-o-n O60 ago) O40 mnyyaInIS “I OTR Lood eerie cere pti. 0 WNUWAUIU WNAJUZIOLOAT Gg Gi 4 GO HeOuost yO. 10107010 dU OU Oxon OO xarduas oa Sonogo Gove co og a0 60000 0g dds UnwMepiagd Aro OL0-00-0 O10 do O60 SISUaADULD] XDINDKUOL) nel (eed sukents’ (ot offs igculsiiey loi celNe geen isiesiateiate x e[[eseyourq poOOoOoONO oOo GoD oOo od GS WN4DISOD DUAU0IA7IIIS Gaps semtnib. . ChoNo: Dy Geomo dno purdsuuas DIDIDGaY ‘ Domo OO Od 0-00 000 6 0-0-5010 -400 DUASSYUBDAL ‘YW MUCMOOUOOOoDOoODoOoOmMDoDoooOo OG SHUM "T GOOD CCoOr on Doo Ot 5 snaMnp SnApUYKIoIgaT GoOoenhooOoD O00 O00 GO Oo ppi9py DIPADUINE) Too 6 Oo 0 OR oe oo Oo Oo Oo Oo XIAISKY UOLYIALOD pOUoHoOoCoOoGDOOUDDOnBOoOo odo ow ao SHU pUp 79) Se suaidvap ‘YD OOO AHDDOO OOD oO moOo GO bobo oo oo $1409409 Yd DOoOe DOD OndoOo DODO oe ooo dds SO4990JXYID SOO O06, 6 0) 0.6.0 6-H oho 18314 DI1g DIJIUOLAAS epretieiieCiuute tamiel (aie siren iatisitcitpitemtel ie) {altatiotes ies) LIICUIOILIG, sotoedsS panuyuoj—] ATAV I, 496 EOS Ney ELeLICkK TABLE II Complete list of the species found at Station 3 during the survey of 1935-1936. DIATOMACEZ Achnanthes teniata Asterionella Bleakley1 japonica Bacteriastrum hyalinum Cerataulina Bergonii Chetoceros spp. affinis borealis borealis var. concavicornis compressus constrictus curvisetus debilis decipiens didymus gracilis laciniosus teres Cocconeis placentula Corethron hystrix Coscinodiscus centralis excentricus lineatus Grammatophora marina Guinardia flaccida Hemiaulus Hauckii Leptocylindrus danicus MInImus Lichmophora abbreviata Melosira moniliformis sulcata Navicula spp. Nitzschia Closterrum longissuma sertata Pleurosigma Normanti Rhabdonema adriaticum Rhizosolenia alata calcar-avis fragullisima hebatata var. semispina imbricata var. Shrubsolet setigera styliformis Sceletonema costatum Striatella unipunctata Surirrella Gemma ? Synedra Gallionti ? Thalasstonema nitzschioides Thalasstosira decipiens gravida Nordenskioldit Thalasstothrix Frauenfeldi longissima DINOFLAGELLAT Cerattum Fusus lineatum longipes longipes var. oceanicum macroceros Tripos Dinophysis acuminata norvegica Ovum Exuviella baliica Glenodinium trochoideum Gonyaulax tamarensis Mesoporos asymmetricus * Minusculus bipes Noctiluca miliaris Peridinium spp. breve Cerasus conicotdes conicum denticulatum depressum gracile Granti simplex Phalocroma irregulare ? Prorocentrum micans minimum Scutellum SILICOFLAGELLAT Dictyocha Fibula Distephanus speculum Fbria tripartiia COCCOLITHINE Rhabdosphera tubulosa Syracosphera spp. FLAGELLATE CYANOPHYCEE Gleocapsa Oscillatorta Ss PHYTOPLANKTON OFF WOODS HOLE 497 Both oceanic and neritic forms occurred with several important species belonging to either class. Neritic forms included both pelagic and bottom types; and occasional tychopelagic forms appeared. The summer flora (see Table I) of 1935 was broadly typical of the area for that season. In July various species of Chextoceros were important. This is the normal summer flora for regions lying to the north of Woods Hole, and to be expected here. This genus was accompanied by Corethron hystrix, Guinardia flaccida, Leptocylindrus danicus and WNitzschia seriata, all temperate or boreal forms. In August these forms were replaced by species of Rhizosolenia (R. calcar- avis, R. alata, and R. setigera) and Thalassionema nitzschioides, a temperate flora. The summer maximum was reached during August, and consisted of Rhizosolenia and Guinardia. These are such large forms that the numbers recorded give a false impression of the actual state of the water. Plankton nets drawn through the waters during one of these maxima became clogged within a few minutes with Rhizosolenia. ‘This summer maximum is very typical of the general area, and is similar to that found by Fish (1925). The entire flora 2 Since the name Porella given by Schiller (1928) to a genus of the family Prorocentracez, Dinoflagellatz, is identical with that of the well known liverwort genus Porella [Dill.] L., order Jungermanniales, and since this latter group antedates Schiller’s genus by a great many years, it seems necessary to change the name of the dinoflagellate group. For this genus the name Mesoporos is offered here in substi- tution for Porella Schiller. Since Schiller’s generic name was not established by a Latin diagnosis, one is given here for the nomenclatorial validation of the genus, in accordance with the International Rules. Mesoporos gen. nov. Porella Schiller non Porella [Dill.| L. Cellula ovalis plusminusve lateraliter applanata. Flagella per fenestram edentatam poriferem exserta. Hemitestae poro singulo mediano intus projecto praeditae, si lateraliter viso invaginationem conicam obtusam apice perforatam formanti. Chromatiphora dua vel tria flava vel brunneo-flava ad hemitestas applanata, absque pyrenoideis. Species 5, a cl. Jos. Schiller sub nomen Porellam descriptae, Arch. f. Protistenkunde 61:54. 1928. Mesoporos bisimpressus (Schiller) comb. nov. Exuviella bisimpressa Schiller, Arch. f. Protistenk. 38: 257. 1918. Porella bisimpressa ibid., 61: 54. 1928. Mesoporos asymmetricus (Schiller) comb. nov. Porella perforata (Gran) Schiller, Arch. f. Protistenk., 61: 55. 1928. Porella asymmeirica Schiller, Dinoflagellata, Bd. 10, 3 Abt., S. 29, in Rabenhorst’s Kryptogamen-flora. 1931. : Mesoporos globulus (Schiller) comb. nov. Porella globula Schiller, Arch. f. Protistenk., 61:56. 1928. Mesoporos adriaticus (Schiller) comb. nov. Porella adriatica Schiller, Arch. f. Protistenk., 61:56. 1928. Mesoporos perforatus (Gran) comb. nov. Exuviella perforata Gran, 1915, Bull. plankt. for 1912. Cons. Perm. int. p. l’explor. de la Mer, Copenhagen. 498 IQS), (C,, EINE IKONS for this season was much more characteristic than that found in the following summer. The winter flora, following the fall maximum, was extremely poor (see Table I). Collections made on October 1 showed a relatively large number of diatom species, occurring in considerable abundance. These collections were very similar to Fish’s (1925) typical winter flora and indicate that the season of 1935 resembled those of 1922 and 1923 at that time. In October 1935 the species which Fish had found to be dominant were again dominant, namely, Rhizosolenia alata, Sceletonema costatum, and Leptocylindrus danicus. Certain of the forms which he mentions as occurring in abundance were found in our collections to be important here in October and at no other time throughout the winter (Chetoceros decipiens and Rhizosolenia setigera). This similarity is surprising in the face of the total lack of similarity throughout the rest of the winter. For the rest of the winter months our collection showed that the plankton as a whole was extremely poor. The four important species, and at times nearly the only species found, were Thalassionema nitzschioides, Rhizosolenia hebatata var. semispina, Thalassiosira Nordenskiéld1i, and Nitzschia seriata. All are boreal forms, the first two oceanic, and the last two neritic. Nearly all of the supplementary species were likewise boreal. Along with these forms there was a relatively high percentage of flagellated forms, a group which appeared throughout the year and which will be discussed later. These results are in the main very different from those of Fish (1925), who found a definite winter ‘‘maximum”’ consisting of a greater variety of abundant species than the summer. It is extremely unfortunate that the great difficulty in making these collections necessitated the spacing of the trips rather far apart. During much of the year, because of the relative stability of the flora, this should have made little difference. Nevertheless it has resulted in our failing to observe any especial spring flowering of diatoms which may have occurred and which is so characteristic of the region north of Cape Cod. It is well known from the work of Bigelow, Gran and Braarud, and others that at some time during the early spring, and for a period of short duration, there is a flowering of diatoms dominated by Thalassiosira Nordenskidldii in the Gulf of Maine, Bay of Fundy, and Massachusetts Bay, which results in phytoplankton numbers far in excess of those of any other season of the year. Fish seems not to have found a similar flowering, though he mentions T. Nordenskiéldii as occurring throughout January, February and March. However, the conditions with which he dealt at the Bureau of Fisheries’ dock per PHYTOPLANKTON OFF WOODS HOLE 499 were far different from those at our more off-shore station. He does, however, record the occurrence of a winter maximum beginning in November or December and lasting into March, for Vineyard Sound and Long Island Sound. Whether or not a flowering of either type occurred at our station during the spring of 1936 must remain problematic; certainly no such conditions appeared in our collections. It is entirely possible that in the interval between any two of our spring collections such a flowering could have burst forth and dis- appeared again without showing up at all in our collections. The weight of evidence which we have at hand seems to indicate that if such a flowering occurred during the spring of 1936, it could not have happened in all probability later than early February. Collections taken at the end of January show that the phytoplankton had in- creased decidedly over the winter months, yet was not abundant enough to be considered as a “‘flowering.’’ Further, there was a decided increase in the number of cells of T. Nordenskiéldii. Other species which began to appear or to become prominent were Riizo- solenia fragillissima, Nitzschia seriata, Thalassiosira decipiens, Thalasst- onema nitzschioides, and certain of the peridinians, all boreal forms of the neritic or oceanic type. By the end of January the nitrates and phosphates in the water had decreased markedly from their winter maximum concentration in December. Similar sudden depletions of these salts have been found by others in the English Channel, Loch Striven, and elsewhere (Harvey, 1928; Atkins, 1927; Marshall and Orr, 1929) to be coincident with the outburst of spring phytoplankton. It seems logical to infer that such an explanation may account for the sudden decrease in these salts at our station. The collections of late February and the following months show a marked falling off of the phytoplankton, and of the critical species, a condition which would normally follow the spring maximum. All of these arguments combine to indicate that any spring flowering of diatoms which may have occurred at our station must have done so between the end of December and the earlier part of February. Following February the plankton became exceedingly scarce, and entered the summer period, which for 1936 was the poorest for the entire year. A few species of Chextoceros, which usually make up a large portion of the summer flora in this latitude, and which were prominent the year before, appeared early in the season but died out as the year progressed. The important species throughout the season was Guinardia flaccida, a temperate neritic form. The flagellates also became prominent for a time in June, and then decreased. ‘Temperate 500 ILIOUES) (Ce IEIMEI BINS Ie forms became more common; whereas boreal forms tended to die out. Certain of the smaller dinoflagellates appeared during the summer, though never in great numbers. The summer maximum, which is of very short duration, did not show up in our collections. However, we do have other information which gives the time and nature of this summer flowering. Collections made in another connection at the Oceanographic Institution showed that during a period of three or four days, from August 9 to August 11 or August 12 the waters in Vineyard Sound were teeming with a diatom flora which excluded nearly everything else, coloring the water a deep olive green. An examination showed this flowering to be made up almost entirely of the two large diatom species, Rhizosolenia calcar-avis, and Guinardia flaccida, both of which were prominent in the summer maximum of the year before. This growth appeared very suddenly and disappeared just as rapidly. As can be seen from our collections of August 6 and August 14, both species were present in the waters, but in very small amounts. Table I shows the general occurrence and abundance of the more important species found throughout the fourteen months at a depth of 15 meters. Diatoms were obviously the most important element of the flora throughout the year. Dinoflagellates appeared throughout the year, more frequently in summer and fall, but their numbers were never sufficient to outweigh the diatoms. A group of forms which were exceedingly small, flagellated and pigmented, which therefore must be considered in a treatment of phytoplankton, kept appearing in all of our collections throughout the year, and in very large numbers. The formaldehyde used to preserve the collections apparently so distorted these forms that identification was impossible. There is no question that many widely separated organisms appeared in this class, which have been lumped together indiscriminately here under the term “Flagellatz.’’ At times certain of the forms were obviously swarm spores of certain of the diatoms and peridinians. It is quite possible that others may have been spores of other algal groups. Some were evidently independent flagellate forms. Much difficult work will be necessary before this general group can be clearly under- stood. No mention of such a prominent group as we have found seems to have gotten into the literature of this general region, probably because much of the previous work has been done with a no. 20 plankton net, which would allow these small types to slip through the meshes. Poor though our general grouping of the forms may be, it seems desirable to include them in this account because they doubt- less play an important part in the general food cycle of the region. PHYTOPLANKTON OFF WOODS HOLE 501 In Table II is given a complete list of the phytoplankton found to occur at Station 3, during the survey. Broadly speaking, then, our results correspond well with those of Fish, whose work is the only one of a similar nature conducted in a nearby region. Making allowances for the great differences between the two localities, and for the limits placed upon our collections, the seasonal cycle of the two surveys corresponds remarkably well. Comparison of the species given in Table II with those listed by Fish shows that a certain number of littoral or bottom forms appeared in his collections and not in ours. A number of these species appeared in our collections made during the first summer at the more littoral stations. Other features characteristic of more offshore waters are shown by our station. Along with stratification of the water column is seen a certain degree of stratification of phytoplankton. One of the outstanding characteristics of Fish’s station is the homogeneity of the water from surface to bottom. The seasonal cycle at Station 3, if we consider that we have sufficient data from which to judge, more nearly resembles that of offshore waters, with its spring and fall maxima, and summer and winter lulls, than it does the neritic waters. Conditions here must be considered to be somewhat intermediate between the other two types. For a long time oceanographers have held to the theory that phytoplankton production is tied up with and dependent upon the amounts of nitrates and phosphates found in the water in any given latitude. Ina broad sense, although this does not seem to be the final factor which controls the cycle, there must be, certainly, some close correlation between the two. Some such broad relationship can be seen in the results of our survey. An examination of Figs. 1, 2, and 3 will show that the nitrates and phosphates were relatively plentiful during July and early August of 1935, and that they dropped suddenly during August. They gradually accumulated during the fall and rose to their highest point during the winter months. During January there was a sudden drop to a summer low which continued until July, when the salts began to accumulate again to a lower peak in August. Another decrease occurred here. Considering the curves for the plank- ton, it can be seen that there are certain points of correlation, assuming that the spring flowering occurred, as has been indicated earlier, before the beginning of February, and the summer one in the middle of August. When the phytoplankton was low, the salts began to accumulate. As the phytoplankton began to increase, the salts dropped off suddenly, implying that they were used up in the plant cells. The dying off of the phytoplankton was again followed by a 502 LOIS "@, LIELICK gradual accumulation of salts. These were no doubt supplied in part by the decomposition of dead organic matter, and in part by new water entering the Sound. There are minor differences between the curves throughout, but in the major phases this general relationship holds. Nitrates seem to be somewhat more critical for phytoplankton in these particular waters than phosphates. There was always more PO, in the water column than NO3, probably always enough for some plant growth. It is with the curve for NO; that the phyto- plankton shows the greatest correlation, and it may be that only when the PO, curve agrees with that of NO; in trend that it can be considered related to the plankton production. There is a much more pronounced relationship between the phytoplankton and the salts at 2 meters and 15 meters than at the bottom, the 15-meter level showing the clearest correlation. The general ratio of nitrates to phosphates which is shown by Redfield (1934) to exist in the open oceans, and which corresponds in general to the proportions in which these occur in plankton organisms, does not hold at our station. Our data reveal that throughout most of the year the phosphates exceeded the nitrates, a condition which is contrary to that normally found. Quantitatively, a greater number of organisms were found near the surface than at the bottom. The greatest development of phyto- plankton is usually found to be in the lower strata of water, but at our particular station the amounts of detritus found in the water at all levels was so noticeably greater than that found in more open waters, that it is quite possible that throughout much of the year light is not available at the lower levels for photosynthesis. It happened very frequently that the number of cells at the bottom greatly exceeded that at 15 meters. This is probably explainable by the fact that cells sink rapidly once they become inactive, hence the active cells are those at the surface; those at the bottom have become inactive or are dead. In general the important species found at all three depths at any one time were the same. No counts obtained during the entire survey were high, the highest number recorded being 214,000 cells per liter at the surface in October, a number which is normally much exceeded during the spring maximum at other localities. Doubtless the numbers during the spring flowerings which occur in this region would exceed 200,000 many times. The lowest number recorded was 200 cells per liter at 30 meters in December, and at 15 meters in February. Collections made in March for the entire water column were lower than at any other time of year. This survey, incomplete though it is, and lacking two important phases of the phytoplankton cycle, adds definitely to our knowledge PHYTOPLANKTON OFF WOODS HOLE 503 of the waters near Woods Hole. It is the first study of the sort in this region of a station which is far enough from land to show some of the hydrographic features of the open sea. The quantitative work gives a reasonably accurate idea of the numbers of phytoplankton cells which occur in these waters at the different seasons; and the possible relationship between the plankton and the concentration of nitrates and phosphates is indicated. LITERATURE CITED Atkins, W. R. G., 1927. The phosphate content of sea water in relation to the growth of the algal plankton. Part III. Jour. Mar. Biol. Ass’n., N. S., 14: 447. BicELow, H. B., 1914. Oceanography and plankton of Massachusetts Bay and adjacent waters, Nov., 1912—May, 1913. Bull. Mus. Comp. Zodl., 58: 385. BicGELow, H. B., 1926. Plankton of the off-shore waters of the Gulf of Maine. Bull. U. S. Bur. Fish. (Part II), 40: 1. Caxins, G. N., 1902. Marine Protozoa from Woods Hole. U. S. Fish. Comm. Bull., 1901: 413. CLARKE, G. L., anD D. J. ZINN, 1937. Seasonal production of zodplankton off Woods Hole with special reference to Calanus finmarchius. Biol. Bull., 73: 464. Cooper, L. H. N., 1933. Chemical constituents of biological importance in the English Channel, November, 1930 to January, 1932. I. Jour. Mar. Biol. Ass’n., N. S., 18: 677. Fisu, C. J., 1925. Seasonal distribution of the plankton of the Woods Hole region. Bult. U. S. Bur. Fish., 41: 91. Gran, H. H., 1908. Diatomeen, in K. Brandt u. C. Apstein, Nordisches Plankton.., 19: 1. Gran, H. H., AND T. BRAARUD, 1935. A quantitative study of the phytoplankton in the Bay of Fundy and the Gulf of Maine. Jour. Biol. Board, Canada, I: 279. Harvey, H. W., 1926. Nitrateinthesea. I. Jour. Mar. Biol. Ass’n., N. S., 14:71. Harvey, H. W., 1928. Nitrate in thesea. II. Jbid., 15: 183. Hustent, F., 1930. Die Kieselalgen Deutschlands, Oscseeing, und der Schweiz. iRalbannoes? s Kryptogamen-flora. Bd. 7, Leipzig. MARSHALL, S. M., AND A. P. Orr, 1929. A sends of the spring diatom increase in Loch Striven. Jour. Mar. Biol. Ass’n., N. S., 6: 853. NIELSEN, E. STEEMANN, AND T. H. v. BRAND, 1934. Quantitative Zentrifugen- methoden zur Planktonbestimmung. Rapp. Cons. Explor. Mer., 89 (Part III, App.) (1933-1934), p. 99. PAULSEN, O., 1908. Peridiniales, in K. Brandt u. C. Apstein, Nordisches Plankton., 18: 1. REDFIELD, A. C., 1934. On the proportions of organic derivatives in sea water and their relation to the composition of plankton. James Johnstone Memorial Volume, Univ. of Liverpool. SCHILLER, J., 1928. Die planktischen Vegetationen des adriatischen Meeres. Arch. f. Protist., 61: 45. i; SCHILLER, J., 1930. Coccolithineze Deutschlands, Osterreichs, und der Schweiz. eepenhorss s Kryptogamen-flora. Bd.10,2 Abt. Leipzig. SCHILLER, J., 1931. Dinoflagellate. Jbid., Bd. 10, 3 Abt. SERUM PROTEIN MEASUREMENTS IN THE LOWER VER- TEBRATES) 1h) THEVCOLLOID OSMOTIC PRESSURE, NITROGEN CONTENT, AND REFRACTIVE INDEX OF TURTLE SERUM AND BODY FLUID MILDRED L. CAMPBELL AND ABBY H. TURNER (From the Physiological Laboratory, Mount Holyoke College, South Hadley, Massachusetts) The significance of the blood serum proteins in the maintenance of the fluid balance between blood and tissues was first suggested by Starling (1), who postulated that the osmotic pressure they exert is the force which, acting in opposition to the hydrostatic pressure of the blood, prevents the loss of excessive amounts of fluid to the tissues. The numerous studies of the serum colloid osmotic pressure which have since been made in man and the lower mammals have indicated that for mammals Starling’s hypothesis holds true, if not completely, at least as a major factor. A comprehensive review of the subject, with bibliography, is that of Landis (2). A recent modification of method is that of Wies and Peters (3). The question of the applicability of this hypothesis to conditions which obtain in the lower vertebrates has been investigated very little. Data on the amphibia are given by Landis (2) and also by Drinker and Field (4). The most recent work has been that of Keys.and Hill (5), who studied six species of fish. It has seemed important to us to make a series of studies of the colloid osmotic pressure of the serum in several of the lower vertebrates, correlating the findings whenever possible with other determinations on the serum proteins. In this paper we present the results of a study of a reptile, the common “slider’’ turtle. Determinations were made of the colloid osmotic pressure, nitrogen content, and refractive index of the serum and of the body fluid of normal animals and of those subjected to long- continued starvation at two different temperatures. METHODS Fourteen adult, healthy, female turtles of the species Malacoclemmys geographica, Les. were obtained in October at about the end of the period of summer activity and feeding. ‘Two of these were studied at once as normal fall animals and the remainder divided into four experi- 504 SERUM PROTEIN MEASUREMENTS IN LOWER VERTEBRATES,I 505 mental groups, of which one was kept during the succeeding months under each of the following conditions: (1) room temperature, 19°—20° C., with feeding, (2) room temperature without feeding, (3) winter temperature, 7°-10° C., with feeding, (4) winter temperature without feeding. The food consisted of common garden worms ad libitum; all animals were kept constantly in large tanks of fresh water. During April at the end of the winter period of hibernation and fasting, two fresh specimens were obtained for use as spring controls. Blood was drawn by syringe from the aorte and sinus venosus and allowed to clot in tubes. After centrifuging, the supernatant serum was removed by pipette. A clear, colorless fluid which was frequently found in considerable amounts in the body cavity was removed by syringe without contamination by blood. It is this fluid which is referred to as body fluid. Colloid osmotic pressures were obtained by the second method of Krogh and Nakazawa (6), using the modifications suggested by Turner (7). The membrane was Du Pont cellophane no. 450, the outer liquid 0.6 per cent sodium chloride. For every determination three or more osmometers were set up. The figure taken for the colloid osmotic pressure of any sample was the average of values given by osmometers which conformed to the requirements of the Krogh technique. Determinations of the total and non-protein nitrogen concentrations were made by the micro-Kjeldahl method combined with direct Nesslerization as given by Peters and Van Slyke (8). The difference between total and non-protein nitrogen gave the con- centration of protein nitrogen. Total refractive indices were deter- mined by a Zeiss dipping refractometer. Whenever the quantity of serum available made it possible, a determination was made of the refractive index of an ultrafiltrate prepared by the use of a Thiessen ultrafiltration apparatus with cellophane 450 as membrane. The total refractive index minus that of the ultrafiltrate gives the refractive index of the colloid fraction. RESULTS The protein data on sera and body fluids are given in Table I. The serum colloid osmotic pressure in animals of the warm, fed group was within the range of the normal fall values for a period of three months. The progressive decrease in pressure observed in all the other groups may be explained as due to the effects of starvation modified by the metabolic rates of the animals. In the warm, starved group total starvation plus a relatively high metabolic rate resulted in marked and immediate changes in the serum proteins as shown by the lowered colloid osmotic pressure. Edema was expected in these animals, but 506 MILDRED L. CAMPBELL AND ABBY H. TURNER was not found either in this or in any other group. Since the turtles in the cold, fed group ate very little, both cold groups may be regarded as starved. Due apparently to the depression of the metabolism by the low temperature, the colloid osmotic pressure in these animals decreased more slowly than in the warm, starved group and the total fall was less. That the rate of metabolism in animals when subjected to different environmental conditions is so affected was shown by TABLE I Protein data from blood serum and body fluid, Malacoclemmys geographica. Body fluid was found only in the animals indicated. Protein nitrogen was obtained by subtracting the analytically determined non-protein nitrogen from the total nitrogen. The refractive index of the colloid fraction was obtained by subtracting the refractive index of the ultrafiltrate from the refractive index of the whole serum. An asterisk after the colloid refractive index indicates that in this case no ultra- filtrate was available. The figure was obtained as stated in the text. Colloid 5 e Length] osmotic Bees ie pene Refractive index tatus of pressure of Date | experi- animal killed | mental period | Se- | Body | Se- | Body | Se- | Body} Se- | Colloid | Body rum | fluid | rum | fluid | rum | fluid rum |fraction] fluid mm.| mm. |\mgm./| mgm./|\mgm./| mgm.| months H20 | H20 | cc. CC. CC. CC. Normal, fall. ..| Oct. 30} 0.0 81 81 — — —_ —_— 1.34350] .00768 | 1.33586 Be “',..| Nov. 7} 0.0 2s — — — — | 1.34453) .00886*| — Warm, fed....| Jan. 4] 2.0 98 | 46 | 6.54} 0.75 | 6.29 | 0.61 | 1.34364] .00799 | 1.33586 of (a5 co|| es Al S40) 99} — | 683} — | 660] — | 1.34399] .00809 — Warm, starved | Dec. 21] 1.75 58 | — | 6.24] — | 602} — | 1.34508 00907 —_— oe 2 Jan. 17] 2.5 58} 28 | 5.85 |} 0.83 | 5.45 | 0.51 | 1.34291] .00758 | 1.33632 oe as Mar. 1] 4.0 39 | 36 | 6.58] 0.79 | 6.34 | 0.58 | 1.34342 | .00779 | 1.33636 Cold, fed...... Dec. 14} 1.5 94) 50 | 5.64} 1.77 | 5.51] 1.65 | 1.34341] .00773 | 1.33659 Be erie era Jane Lon|2e5 77| — |695 | — | 6.73} — | 1.34474} .00907*| — SS ees ayer Feb. 12} 3.5 72! — | 6.14] — | 604] — | 1.34339] .00766 —_— Cold, starved..} Dec. 11] 1.3 102 | 65 | 5.37 | 0.66 | 5.19 | 0.54 | 1.34356] .00789*| 1.33604 45 oll eins 1) || 583 57] 41 4.70 | 1.23 | 4.55 | 1.07 | 1.34291] .00724*| 1.33519 My we soll Jer, ZA We7/ 58 34 6.51 | 1.23 | 6.27 | 1.00 | 1.34350}| .00760 | 1.33580 a ool ads 7] Ses 66} — | 5.74} — | 5.56} — | 1.34548} .00981*) — Normal, spring | Apr. 7] 0.0 68} — | 4.18} — | 4.02 | — | 1.34208] .00671 — He > Apr. 9] 0.0 48 | — |5.11} — | 490] — | 1.34235] .00699 — Hand (9) in his compilation of the results of inanition on the higher groups of animals. The occurrence of amounts of body fluid adequate for determina- tions was irregular and not limited to any group or groups nor to any particular time in the experimental period. Although the variations in the colloid osmotic pressure were wide, the value in each case was found to be lower than that of the corresponding serum except for one instance of equality. No proportional relation between the two pressures was discernible. SERUM PROTEIN MEASUREMENTS IN LOWER VERTEBRATES, I 507 Table I shows total nitrogen and protein nitrogen figures, the latter obtained by subtracting the non-protein nitrogen from the total nitrogen. The non-protein nitrogen content was low and varied irregularly in the sera from .10 to .40 mgm./cc., with eleven of the fourteen values between .15 and .25 mgm./cc. In the seven analyses of body fluid the variation of non-protein nitrogen was from .12 to .32 TABLE II A table showing percentage changes as the experiment progressed. The two normal fall animals are taken as 100 per cent. In the case of the nitrogen values, since no analyses were made on the fall animals, the 100 per cent is furnished by the two warm, fed animals whose other findings showed a negligible decrease as the months went by. The two spring animals were fresh from the collector, after a hibernation period of unknown duration probably much longer than the duration of the experiment. Length of Colloid Status of animal expenmental pempite ae Rehacuye months per cent per cent per cent INGOT IGS gina peo uosen ee 0.0. 100 — 100 \Wanmmesteds oa 2 bo tee see 2.0 101.5 100 96.5 a SC HUE SE Sa aA 3.0 102.5 100 98 Warinm Starved: ss... 62. sls fe 60 93.5 110 oe PU 4 bok 8S ARS ai ee ae DES 60 84.5 92 oe Cie) Ni aie ie Me eae 4. 40.5 98 94 Cold Miedier sau ease ea eae o. 1.5 97.5 85.5 93.5 ne a Ce ya Nu A te 2.5 80 104 LQ) ue SRS ey ciate Seth a8) 2 cael ae yet 3.5 75 94 93 Goldiictarved =) Aya sas 163 106 80.5 95.5 of Ce ae Ed aaah ied Pea A DES 59 70.5 87.5 i Ri reece SRE a ary Rae eh Dall 60 97 92 of RSM SEER LD ae. Be AUN 3.25 68.5 86 118 INormialiispring:. 94/511. fh fae —_— 70.5 62 81 a ibid nige Maelo pe ar — 50 75.5 84.5 mgm./cc. The concentration of protein varied widely yet it seemed clear that starvation in this species fails to cause a fall in the amount of protein parallel to the fall in colloid osmotic pressure. Table II gives the values in percentages, thus showing the lack of correspondence between the effect of starvation on colloid osmotic pressure and on protein content. The average of the two fall specimens is taken as 100 per cent for the colloid osmotic pressure series. It will be noted that the warm, fed animals show no fall. Nitrogen concentrations 508 MILDRED L. CAMPBELL AND ABBY H. TURNER were not determined for the two fall animals but in the absence of that standard it seemed permissible to use the level of the two warm, fed animals as 100 per cent for protein comparisons. Determinations of total protein, albumin, and globulin in the sera of mammals have been reported by several observers (10, 11, 12, 13, 14) who have shown that during starvation a reversal of the mammalian albumin-globulin ratio occurs, indicative of a relatively higher globulin concentration. Because of the larger size of the globulin molecules this shift results in a marked fall in colloid osmotic pressure without a corresponding decrease in total protein concentration. From our observations it seems possible that a similar shift in albumin-globulin relations may occur in the turtle. The two spring specimens which had doubtless spent the entire winter in hibernation showed a fall in total protein more nearly approaching that in colloid osmotic pressure. The con- centration of protein in the body fluid was much lower than in the serum and showed wide variations. No direct correlation with the colloid osmotic pressure was apparent. “The total refractive index of normal, fall turtles averaged 1.34377. With two exceptions the figures for animals with lowered colloid osmotic pressure were below this average, falling even to 1.34222 as an average for the two spring controls. The range of the refractive indices for the ultrafiltrates was narrow, with an average of 1.33569. Where no ultrafiltrate could be prepared, this average figure was used to calculate the probable value for the colloid fraction. (See Table I.) When the refractive index of the colloid fractions was plotted against the colloid osmotic pressure a good correspondence was evident, though there is a closer correspondence between protein concentration and refractive index of colloid fractions. The refractive index of the body fluid was consistently lower than that for the corresponding ‘serum and higher than that of the serum ultrafiltrate. Since the crystalloids of serum and body fluid are probably closely alike, this difference between body fluid and ultrafiltrate from serum would indicate a colloid fraction in the body fluid, as verified by analysis. DISCUSSION Our interest has centered upon the inferences which may be drawn from these determinations as to the significance of the serum proteins. Certain points stand out: first, the fall in colloid osmotic pressure with prolonged starvation, especially at the higher temperature; second, the much greater stability shown during starvation in the quantity of serum protein as compared with its colloid osmotic pressure; third, the presence in the body fluid of substances exerting a colloid osmotic SERUM PROTEIN MEASUREMENTS IN LOWER VERTEBRATES,I 509 pressure far from negligible though the protein percentage is low. While the relation of plasma and true tissue fluid—meaning by this the fluid which has just made its exit through capillary walls—may not be that shown by the comparison of the serum and body fluid, it is nevertheless probable that capillary permeability to proteins, at least those of the smaller molecular size, is indicated by the very perceptible colloid osmotic pressure of the body fluid. Such proteins as may be outside capillary walls will, of course, lessen the effective colloid osmotic pressure of the plasma. Further, the marked decline of the colloid osmotic pressure in starvation without evident edema speaks against it as a major controlling factor in water balance between blood and tissues in this species, while the persistence of the protein content of the serum at a relatively high level may indicate the importance of plasma proteins for some other function than the maintenance of colloid osmotic pressure, possibly as nutritional reserves. SUMMARY The average serum colloid osmotic pressure of two normal fall turtles of the species Malacoclemmys geographica, Les. was 96 mm. water pressure. This pressure was lowered by starvation, especially at laboratory as compared with winter temperature, but the serum protein concentration was much less affected. A body fluid obtainable from about half of the animals showed a colloid osmotic pressure of more than half the serum pressure in these individuals, though the protein content was low. These findings are interpreted as indicating that in this species the colloid osmotic pressure may not be the only controlling factor in the water balance between blood and tissues, and that other functions of plasma proteins may be of importance. BIBLIOGRAPHY 1. STARLING, E. A., 1896. On the absorption of fluids from the connective tissue spaces. Jour. Physiol., 19: 312. 2. Lanpis, E. M., 1934. Capillary pressure and capillary permeability. Physzol. Rev., 14: 404. 3. Wiss, C. H., AND J. P. PETERS, 1937. The osmotic pressure of proteins in whole serum. Jour. Clin. Invest., 16: 93. 4, DRINKER, C. K., AND M. E. FieELp, 1933. Lymphatics, Lymph, and Tissue Fluid. Baltimore. 5. Keys, A., AND R. M. Hitx, 1934. The osmotic pressure of the colloids in fish sera. Jour. Exper. Btol., 11: 28. 6. Krocu, A., AND F. Nakazawa, 1927. Beitrage zur Messung des kolloid- _ osmotischen Druckes in biologischen Fliissigkeiten. Bzochem. Zeitschr., 188: 241. 7. Turner, A. H., 1932. The validity of determinations of the colloid osmotic pressure of serum. Jour. Biol. Chem., 96: 487. 510 MILDRED L. CAMPBELL AND ABBY H. TURNER 8. Sh 10. 11. 12, IS 14, PETERS, J. P., AND D. S. VANSLYKE, 1932. Quantitative Clinical Chemistry, Vol. II. Methods. Baltimore. Hanp, H. M., 1934. Concentration of serum proteins in the different types of edema. Arch. Internal Med., 54: 215. Friscu, R. A., L. B. MENDEL, AND J. P. Peters, 1929. The production of edema and serum protein deficiency in white rats by low protein diets. Jour. Biol. Chem., 84: 167. BRuCKMAN, F.S., L. M. D’Esopo, ann J. P. PETERS, 1930. The plasma proteins in relation to blood hydration. IV. Malnutrition and the serum proteins. Jour. Clin. Invest., 8: 577. SHELBURNE, S. A., AND W. C. Eciorr, 1931. Experimental edema. Arch. Internal Med., 48: 51. WEEcH, A. A., AND S. M. Line, 1931. Nutritionaledema. Observations on the relation of the serum proteins to the occurrence of edema and to the effect of certain inorganic salts. Jour. Clin. Invest., 10: 869. WEEcH, A. A., E. GOETTSCH, AND E. B. REEVES, 1935. Nutritional edema in the dog. I. Development of hypoproteinemia on a diet deficient in protein. Jour. Exper. Med., 61: 299. SERUM PROTEIN MEASUREMENTS IN THE LOWER VER- TEBRATES. II. IN MARINE TELEOSTS AND ELASMOBRANCHS ABBY H. TURNER (From the Woods Hole Oceanographic Institution; 1 Bergens Museum Biologiske Stasjon, Bergen, Norway; and the Physiological Laboratory, Mount Holyoke College, South Hadley, Massachusetts) The reasons for studying the plasma proteins of the lower verte- brates as shown in the first paper of this series (1) have to do with the validity of Starling’s hypothesis for these forms and with the possible relation of the protein fraction of the plasma to nutritional needs. References to contributions in this field relating to the higher vertebrates will not be repeated. For the lower vertebrates, the list given by Drinker and Field (2) summarizes the information available up to their date of publication. A brief paper by Keys and Hill (3) gives the only findings for the colloid osmotic pressure of fish sera except those in two preliminary notes to this paper (4,5). Interésting data allied in various ways are to be found in references (6) to (15) inclusive. The last two papers (14, 15) have excellent bibliographies. The serum of various species of marine teleosts and elasmobranchs was studied at the Woods Hole Oceanographic Institution during the summers of 1933 and 1934, and during the summers of 1935 and 1936 at the Bergens Museum Biologiske Stasjon on the island of Herdla off the west coast of Norway. ‘To both of these institutions most sincere thanks are due for the assistance generously rendered by them. A grant in 1936 from the Tracy McGregor Fund helped to defray expenses in that year.” The determinations made were as follows: (1) colloid osmotic pres- sure, on all samples; (2) refractive index of the whole serum, on all samples; (3) refractive index on an ultrafiltrate, on many samples of the Norwegian series, whenever quantity of serum permitted; (4) total nitrogen and non-protein nitrogen in 1934, 1935, and 1936, a con- 1 Contribution No. 148. 2 The nitrogen determinations of the Woods Hole series were made by Catherine Goffin and those of the Norwegian series by Cand. mag. Francis Wolff, of Bergens Museum. Acknowledgment is made of the careful and helpful work done by both. At Woods Hole the Marine Biological Laboratory made its facilities available to Miss Goffin for her work; at Bergen the Norwegian Fiskeriforséksstasjon allowed the use of its equipment for cold storage and for chemical work. Without this help it would have been impossible to make the nitrogen analyses. 511 912 ABBY H. TURNER siderable number of determinations, whenever quantity permitted. In all, 121 individual samples representing 19 species of teleosts were studied and 48 individual samples from 11 species of elasmobranchs. A few of these samples were assembled from two or more small fish each. ‘There were in addition some scattered determinations on other species. METHODS Obtaining the Fish The fish whose blood is to be studied should be in a healthy con- dition and must be uninjured, to prevent access of the surrounding medium to the blood. To secure fish that meet these standards is difficult. The ordinary methods of fishing are extremely rough when viewed from the standpoint of obtaining true samples of the circulating blood. The effect of the severe struggling often incident to capture may not be negligible. Grafflin (8) has told in detail of his care in this matter of reliable sampling. The fish included in this study have been progressively better and none are included in the figures reported about which there is appreciable doubt. ‘Two alternatives present themselves. Blood may be drawn the moment the fish is taken and this is for some species, for example those from deep water which will not live in tanks, the only possible method; or the fish may be allowed to rest for some hours or longer in cars or tanks of water suitable in temperature and aeration. Feeding is possible in some cases where the fish live well in captivity. The effect of the initial struggling is thus eliminated and the state of the fish is steadier. At least four of the most consistent sets of determinations in this study have been made mainly on fish thus treated, Opsanus tau, Tautoga onitis, Anarhichas lupus, and Raia erinacea. In certain species the skin is so delicate that it is probably best not to use them in such studies as these unless unusual care can be taken. It is possible that the two pleuronectids in the Woods Hole series belong in this group. Enough is known about the relation existing between the blood and tissues of fishes and the demands of the breeding season to make avoidance of this period desirable in a preliminary survey. However, the length of this season is often far beyond the months mentioned in books so that it has not always been possible to avoid it without sacrificing the use of fish otherwise very suitable. The depression in the breeding season well-known to trout hatchery men, when the fish will eat nothing or very little, is of indeterminate length and not easily avoided. ‘The breeding fish used in this series are so described. When it is not mentioned that a species was apparently ripe it is to be under- stood that as far as known the samples were not from breeding fish. SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 513 Blood Sampling The blood was taken in 1933 by syringe from gill vessels or heart; in the other three summers, in most cases from tail vessels, occasionally from the heart. For the suggestion of taking blood from the tail, the author is indebted to Dr. Homer W. Smith. The fish is wrapped in a cloth, laid in a deeply grooved board, ventral side up, restrained by lacing tapes across the board, and held by an assistant. The syringe needle is inserted in the mid-ventral line, somewhat behind the anus, often along the anal fin, at an angle which will take it between the hzemal spines. It is pushed in until the vertebral column is felt, is slightly withdrawn and pushed in again quickly, with a good chance of puncturing a tail vessel from which blood can be withdrawn with a minimum of injury and a maximum of ease. If the tail vessels are too small for use or are covered by too firm a membrane between the spines, the heart is exposed, the ventral aorta clamped off by a hemo- stat, and blood withdrawn from the bulbus or venous sinuses. The blood must be handled gently when discharged from the syringe into the centrifuge tubes else hemolysis may occur. The same thing may happen if the vessel used is not entered quickly and neatly. Heparin has been used in both syringe and centrifuge tubes, but some clotting is usually evident after centrifuging, hence the term serum has been thought more appropriate than plasma. The blood was centrifuged three or four hours after being with- drawn, not earlier because in that case delayed and repeated clotting sometimes occurs and is very troublesome if in osmometer tubes. After centrifuging, the serum was set up at once in the osmometers or kept in a very cold refrigerator until next day. Because it was impossible in Norway to have the nitrogen determinations made on the first day, various preservatives were tried, but all resulted in the appearance of a precipitate. Serum kept at about 0° C., on the other hand, remains clear for a period up to at least three weeks, and re- fractometric measurements show little or no alteration. At rather low room temperatures, 16°-19° C., osmometers containing teleost serum have held steady as long as three days. In the case of elasmo- branch blood, the occasional rather warm intervals at Woods Hole, with room temperatures up to 25°, in a few instances caused the appearance of a thin whitish film on the membrane in the osmometer, but the colloid osmotic pressure in these cases seldom varied from the usual range. These films were not seen at the temperature of the Herdla laboratory, usually 15°-18° C. A sterile technique was not attempted. The place where the needle was to be pushed through the skin was carefully wiped with 514 ABBY H. TURNER cotton. Syringes, needles, centrifuge and transfer tubes, and all other parts of apparatus coming into contact with the serum were cleaned systematically as follows. They were washed thoroughly with water, with soap when needed; they were rinsed thoroughly first with tap water, then with distilled water, finally with 70 per cent alcohol; they were dried in an oven at 70°-80° C. Contamination was not apparent except in the elasmobranch samples of the Woods Hole series in the hottest weather. In general, then, while the securing, care, and preservation of the blood have presented difficulties, the samples included in this report TABLE [| Protein determinations on serum of Opsanus tau, toadfish. Fish all of one lot, kept in aquarium up to 10 days, fed. One or two months after the breeding season. The length of the specimens varied from ten to twelve inches. Nos. 1 and 6 were females, the others males. C.O.P. Colloid osmotic pressure in mm. water pressure, determined from one or two osmometers. N-P.N. Non-protein nitrogen. No. IN CiOnle= SRG ag Total N. N-P. N. Protein N. mm./water mg.|cc mg.|cc mg.|cc 1 104 1.34401 — — — 2 119 1.34498 8.13 .866 7.26 3 76 1.34498 9.28 .809 8.47 4 100 1.34470 6.73 494. 6.24 5 95 1.34375 6.77 497 6.27 6 127 1.34369 6.55 419 6.13 a 125 1.34457 5.63 423 5.21 8 111 1.34386 7.44 368 7.07 9 97 1.34509 — — — 10 84 1.34468 — —_ — Averages 104 1.34443 7.22 554 6.67 are considered reasonably good. However, it is to be said that the history of the individual fish is in many important points entirely unknown. Its age, except as indicated roughly by its size, its recent activity when the fish has been bled immediately on being taken, its nutritive condition except that it may or may not have food in the digestive tract, all these factors of importance are not known. A con- siderable range in findings is to be expected. The Identification of Species The identification of species was made at Woods Hole by the use of Bigelow and Welsh (16) and in Norway by Otterstrém (17). In both places much help was received from various persons who know the SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 515 fishes of the region. Especial mention should be made of the Director at the Herdla laboratory, Professor August Brinkmann, and his son, and of the Assistant-Director, Dr. Rustad. Methods of Testing Colloid osmotic pressures were determined by the method of Krogh and Nakazawa (18) somewhat modified (19). The membrane used was cellophane no. 450. Refractive indices were determined by the dipping refractometers of Bausch and Lomb and Zeiss on both whole serum and its ultrafiltrate. The latter was secured from samples of TABLE II Protein determinations on serum of Anarhichas lupus, wolf-fish, 1935. Blood taken a few hours or a day after fish was brought to laboratory. Not near the breeding season. The length of the specimens varied from 18 to 30 inches. C.O.P. Colloid osmotic pressure in mm. water pressure. Derived from two to four osmom- eters in all cases except where only one osmometer was used, Fish No. 7. N-P. N. Non-protein nitrogen. No. ae p. | %p Serum Oe AD, Sone Total N. | N-P.N. | Prot. N. mm.|water mg.|cc. mg.|cc. mg.|cc. 1 119 1.34530 | 1.33532 | 0.00998 6.15 5 5.90 2 178 1.35588 | 1.33609 | 0.01979 10.94 51 10.43 3 147 1.35051 | 1.33609 | 0.01442 7.96 oo 7.44 4 152 1.35128 | 1.33636 | 0.01492 8.55 .62 7.93 5 145 1.34702 | 1.33545 | 0.01157 8.12 ati 7.A1 6 153 1.34832 | 1.33551 | 0.01281 — — — 7 150 1.34415 | 1.33551 | 0.00864 5.82 slit 5.05 8 177 1.34984 | 1.33578 | 0.01406 8.30 ATT 7.53 Averages 153 1.34916 | 1.33576 | 0.01328 7.98 69 7.38 the Norwegian series by a Thiessen apparatus, with care as to uni- formity of technique. Two cubic centimeters of serum were placed in the apparatus with cellophane no. 450 as membrane and subjected to a pressure of some three atmospheres for a period of 12 hours. At the end of this time a clear, limpid filtrate was obtained, always protein- free. The residue on the membrane at this time was small in amount and slimy in character. Tests for chloride and reducing sugar showed them to be present in the ultrafiltrate. It was assumed that the mem- brane was permeable to practically all the serum crystalloids. Nitrogen determinations were made in 1934 at Woods Hole by the micro-Kjeldahl method with direct Nesslerization. For the non- protein determinations the precipitant was trichloracetic acid. In 516 ABBY H. TURNER 1935 and 1936 the micro-Kjeldahl method was by titration. The precipitant was metaphosphoric acid. All methods and reagents were suitably controlled and tested. TABLE III The colloid osmotic pressure of the sera of marine teleosts. Arranged in order of height of osmotic pressure, in two series. Number of C.O.P, C.O.P. Name of species individuals Range Average mm. water | mm. water Woods Hole Series Sarda sarda, Bloch. Bonito................. 1 233 233 Echenets naucrates, Lin. Remora*........... 1 216 216 Hantoga ons MinkwMautog*.. 5 ase ).s- 4. - 24 79-193 129 Paralichthys dentatus, Lin. Summer flounder. . . Pseudopleuronectes Americanus, Wal. Winter ko tnauakere rs dst eee-B eee alee eo ae rena ae lee ee 6 79-180 120 Prionotus Carolinus, Lin. Common robin*.... 7 86-122 106 Prionotus strigatus, Cuv. & Val. Red-winged LOD Ula ee ee RO Pere ney epee Mev oe.) Su evan 7 87-117 102 Opsanus tau, Lin. Toadfish................. 10 76-127 104 Norwegian Series Scomber scombrus, Lin. Mackerel............ 3 samples 164-212 198 fr. 11 fisht Anguilla vulgaris, Tur. Eel................. 4 samples 167-204 189 fr. 13 fisht Belone acus, Risso. Hornfish................ 2 162, 184 173 Anarhichas lupus, Lin. Wolf-fish............ 13 85-181 147 Brosmius brosme, Asc. Cusk}............... 8 105-210 139 Molva molva, Flem.}..............0.000005. 4 108-177 139 Gadus morrhua, Lin. Cod.................. 11 70-192 117 Gadus pollachius, Lin. Poellack.............. 8 samples 31-172. 94 3 fr. 6 fisht Cyclopterus lumpus, Lin. Lumpfish*......... 3 75-87 81 Lophius piscatorius, Lin. Goosefish.......... 8 25-60 42 Gadusiurrens Meter tebe eee vcys gers, see a ance ais 1 fr. 2 fisht 40 40 * Breeding season. {| From deep trawl, 300-400 meters. Fish with swim-bladders protruding or burst and with organs extruded in some cases on arrival at surface. Blood samples taken immediately on board the collecting boat. {Samples from more than one fish because of small size of samples available from single individuals. The refractive indices were found to be similar before samples were combined. RESULTS The results of the study are presented in a series of annotated tables. The determinations on teleosts are given first. To show the range of variation among individuals of a single species Opsanus tau, Lin., toadfish, was chosen from the Woods Hole series of 1934, Table I, SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 517 and Anarhichas lupus, Lin., wolf-fish, Table II, from the Herdla series of 1935. The toadfish was chosen as a sluggish fish, much studied at Woods Hole laboratories, and the wolf-fish as a more vigorous species much studied in Norway. TABLE IV Refractive indices of sera and ultrafiltrates from marine teleosts. Arranged in order of height of C.O.P. Figures in parentheses indicate the number of indi- viduals used for the following average, when this is less than the total number. Name of species Aner Oe ny Serum Ps mee Woods Hole Series ~ SOR SUS aobeian su0 co hws ean og bb. il 1.35220 = — WIGHENCUS NAUCKOLES Mersin eon 2 1.35158 = — TE OSXGD. OBES A Van Gio Cee eo Sor 24. 1.34629 = — Paralichthys dentatus............... Pseudopleuronectes Americanus...... 6 1.34313 — — Prionotus Carolinus.............-.- 7 1.34329 — — LEO OWS SH ALOUND org beaiule Wie tele alee if 1.34205 = — CO PSUTLUSELILIER ON Det rain it eke ers Sclaye 10 1.34443 = — Norwegian Series ISGOMIDCTASCOMORUS eae eiels ar saeco 3 samples | 1.35000 |(2)1.33690 | 0.01306 fr. 11 fish* Alon io, SOVGT IS yds = Bele as On cee 4samples | 1.35152 1.33552 | 0.01600 fr. 13 fish* JBOD EO ICSE ance ae ie ee 2 1.35084 1.337247] 0.01360 Anarhichas lupus... .......0.0000.. 13 1.34852 1.33577 | 0.01275 Brosmius brosme.... 1.6.2... eevee 5 1.34703 1.33574 | 0.01169 VIG UCIT OLUA ih een pe Acie eae caraee 4 1.34597 | (2)1.33601 | 0.00912 (GODS GUAT DE pega hence ORC 11 1.34421 1.33588 | 0.00834 Gadus pollachius.............00005. 8 1.34401 1.33594 | 0.00807 3 of these fr. 6 fish* CNG G PICTUS ANU DUS A oasis «sos eee oe 3 1.34388 1.33535 | 0.00853 Lophius piscatorvtus.............05. 8 1.34168 1.33545 | 0.00623 (ROIS, TEAZ ISS Rae ATOR ee 1 sample 1.34158 1.33559 | 0.00599 fr. 2 fish* * See note on Table III. {+ Samples used for obtaining ultrafiltrates in these cases were only one cubic centimeter instead of two as usual. This may explain the aberrant value. Table III summarizes the data on colloid osmotic pressure from the teleosts of Woods Hole and Herdla; Table IV, the data on refractive indices; Table V, the nitrogen determinations. The fish from Woods Hole and Herdla are listed separately, since the procedures were more complete in the later series. The order of names in all tables is that of the relative heights of the colloid osmotic pressure, even though the 518 ABBY H. TURNER order of refractive indices and of nitrogen content does not follow exactly the same course. In a similar manner Tables VI-IX present the data for elasmo- branchs, Table VI showing the variation among individuals of a single species, Raia erinacea, Lin., small skate, and Tables VII-IX the colloid osmotic pressure, refractive indices, and nitrogen findings re- spectively for the several species of elasmobranchs. An effort was made to study Myxine glutinosa, Lin., hagfish, but it TABLE V Nitrogen determinations on sera of marine teleosts. Number of Total Protein N. N-P. N. N. Name of species san Ferra Fle Woods Hole Series WECITEIVEUS LOM CTOLOS PD nine n tm tele re chetis of ekc Soa 1 6.68 PAID 6.41 MGS ULO ZA ON IS Ra ee Ney ten ieee he cs iesdere ee «63 6 6.62 498 | 6.12 Paralichthys dentatus......-0..0.20eeeeeee Pseudopleuronectes Americanus............ 3 5.48 452 5.03 IPetONOLUS COV OIINUS Mee een eecares see aa eh os 1 7.90 747 Ua ESOL TEPID we a.ceoasa 8 Ate Nicks eRe 7 7.22 554 | 6.67 Norwegian Series SS COMMUCHESCOMUOIAUS een a tedets Welel Ac\aiel fete +)6%- 2samples | 8.30 65 7.66 fr. 9 fish* Mpaontts SACS 33 ae 60 8 6 On 8 on Soe re Ree 3 samples | 8.47 | 1.19 7.28 fr. 11 fish* PAHOA RUS [TALIS Vee o 00 oaks SOR aoe 10 8.11 .70 7.48 > IEOSTTRIS HADES oi, SG 6 30 0k a DO oe ae 1 6.95 7) 6.38 Gadusnmornhugaereoee eon ae eee: 5 5.93 .60 5.33 GMb Usp POUAChiUse Ow eee tee See ei sarns 3 samples | 6.11 Ay {ae Ls fi Salishin t Ciclo Piers Pus) Vie wert) he a Nokes che a oie 3 Sasi 125 5.09 WO PIAS PUSGOLOTIUS |: due wiry Ae eco. oe ess: 8 4.85 44 4.41 (CHGS TARA ES 55 GAO Old Be ee Dieters ene 1 sample 3.99 39 3.60 fe.) 2 fish* * See Table III. was found very difficult to secure blood samples entirely uncon- taminated by slime. The individuals were not large enough to yield more than about one cubic centimeter of blood, often less, and there- fore of necessity the determinations were made on mixed samples, though the refractive indices were taken before mixing and only samples of similar indices combined. From one such mixed sample two osmometers gave an average of 24 mm. c.o.p., the total serum showed a 7p of 1.35073, the ultrafiltrate mp 1.33923, thus giving a colloid index of 0.01150. The total N of the serum was 5.88 mgm./cc. TABLE VI Protein determinations on serum of elasmobranch, Rata erinacea, small skate. Fish of different lots, kept in aquarium not more than three or four days. Height of breeding season, several eggs laid in aquarium. The length of the specimens varied from 17 to 19 inches. Nos. 2, 3 and 9 were males, the rest females. C.O.P. Colloid osmotic pressure, determined from one to three osmometers, usually two. N-P. N. Non-protein nitrogen. No. AvaGO Pam echves |! “Total N. N-P. N. Protein N. mm.|water mg.|cc. mg.|cc. mg.|cc. 1 63 1.34694 16.83 — == 2 73 1.34886 16.95 — = 3 14 1.34665 — — = 4 50 1.34581 — — — 5 47 1.34895 — — — 6 71 1.34765 — — = a 17 1.34661 15.91 = a 8 16 1.34661 14.78 13.55 1.23 9 14 1.34849 13.13 13.02 0.11 10 34 1.34683 16.39 13.19 3.20 lal 19 1.34618 16.39 13.71 2.68 12 29 1.34640 16.75 13.88 2.87 Averages Sif 1.34715 15.497 13.47} 2.02 { Five only used for average. TABLE VII The colloid osmotic pressure of the sera of elasmobranchs, arranged in the order of the height of the osmotic pressures. C.O.P. Colloid osmotic pressure. : Number of COLE: C.O.P. Name of species individuals Range Average mm.|water | mm./water Woods Hole Series Carchartas taurus, Raf. (sp.?) Shark.......... 1 46 46 Mustelus canis, Mitch.{ Smooth Dogfish....... 4 33-58 41 Raia erinacea, Mitch.} Small Skate........... 12 14-73 37 Dasybatus marinus, Klein. Stingaree.......... 1 36 36 Norwegian Series IROOE: HOUND PAIR, NODA ON Re Gale Bis ere a Ge 2 14, 49 32 Chimexra monstrosa, Lin.t.................... 5 * 10-40 27 Galeus vulgaris, Flem. Shark................. 1 Dil 27 Squalus acanthias, Lin.{ Spiny Dogfish........ 2 De, oil 27 Prastiurus catulus, Gun.t.s..1. ssc e 7 5-59 26 ROG ORM APSARA LIS, NUM ha ah oe bs aca ch eb asnee 6 11-45 22 HSE O PLer US SPEMas leitledy anise oie Aor eee ieee 7 samples fr. 14 fish* | -—8-34 17 * Individual fish small, about 12-14 inches. Samples were therefore combined, but only after refractive indices had shown them to be in the same range. Two negative values may indicate the error of the method when the pressure is extremely low. { Breeding season. t From deep trawl, 300-400 meters. No outward sign of injury on coming to the surface. LE. spinax and R. fullonica lived often as long as 24 hours in laboratory tanks. 519 520 ABBY H. TURNER TABLE VIII Refractive indices of sera and ultrafiltrates from elasmobranchs. The number in parentheses (4) indicates that the following ~p was derived from 4 individuals only. See notes on Table VII. Name of species muntabee Gt np Serum foe 1 Colloid Woods Hole Series Carcharias taurus.............. 1 1.34601 = — Mustelus canis}............... 4 1.34618 — — Rava erinaceay............+..- 12 1.34715 — — Dasybatus marinus............ 1 1.34682 — — Norwegian Series Ratenulanico eee eee 2 1.34852 1.34079 | 0.00774 Chimexra monstrosat........... 5 1.34596 | (4) 1.34063 | 0.00486 Galeusiaulearis sone. che 1 1.34422 1.34113 | 0.00309 Sgualuscacanthiasiten. = ck veureae 1 1.34671 1.34096 | 0.00575 Pristuurus catulust............. 7 1.34423 1.34076 | 0.00331 Raia oxyrhynchus} t........... 5 1.34412 1.34070 | 0.00342 Himopreris spinaxy i)... 50s o. a. 7 samples | 1.34353 1.34055 | 0.00288 fr. 14 fish* the N-P.N., 0.22 mgm./cc. These figures do not seem concordant, but are reported since data on this form are few. Body fluid was present in one instance only in quantity sufficient for collection and study. This was in a wolf-fish of medium size and TABLE IX Nitrogen determinations on sera of elasmobranchs. See notes on Table VII. Number in parentheses (1) indicates that in only one instance was there serum enough for the determination of the N-P. N. ‘Name of species Buber ier Total N N-P.N. Protein N. a mg.|cc. mg.|cc. mg.|cc. Woods Hole Series RO TONCLI ACEO ee ea er 5 15.49 13.47 2.02 Dasybatus marinus. ..........0.0.05. 1 16.85 11.81 5.04 Norwegian Series ROTOR ON TCU IIR ten peice oe tes 5 3 17.50 12.16 5.34 Chimera monstrosat................. 1 13.91 11.15 2.76 Galeuswuul carts era hie ers. 1 13.76 11.47 2.29 aS COUTTS Ss, oe 6 bein Geel aie 1 14.98 12.68 2.30 Pristvurus catulusi.................. 3 | 12.72 9.58 3.14 (ROTANOGN HIN ICIS LR ey re ee ce 6 15.07 12.31 2.76 Etim pleres SPrTann. mate lee ase = A - 4samples | 12.18 | (1) 10.07 1.60 SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 521 medium serum values. There was nothing to indicate that the fish was not in a healthy condition. The data are given in Table X. The body fluid showed a c.o.p. nearly four-fifths that of the serum, and while the refractive index did not show quite as high a colloid fraction, yet it is clear that this fluid contained a considerable amount of protein, an amount comparable to that found in the body fluid of the ‘‘slider’’ turtle (1). DISCUSSION OF RESULTS Regarding the results on teleosts, it is apparent that the findings for a single species show a range proportionately wider that we expect in mammals. Such a lack of constancy has been found by other workers on fish blood. - (See references in the first paragraph of this paper.) Whether this is really true or due, in spite of efforts to the contrary, to the use of unsuitable specimens is not known. Studies are in progress on fresh-water species living in the controlled environ- TABLE X Comparison of serum and body fluid from one specimen of Anarhichas lupus from which enough body fluid was obtainable for colloid osmotic pressure and re- fractometer tests though not for nitrogen analyses. A n, Serum n,, Ultra- . Hae ea or fluid filtrate tore es mm./water Blood serum......... 129 1.34707 1.33563 0.01144 Body fluid........... 98 1.34184 1.33532 0.00652 ment of fish hatchery ponds to see if the variation from individual to individual is as great as that seen in the toadfish, 76-127 mm. or in the wolf-fish, 119-178 mm. water pressure. Also the yearly rhythm is being followed to see whether colloid osmotic pressure shows the effect of elaboration of sex products and of spawning depression. There is obviously a wide difference between the colloid osmotic pressure of the goosefish, 42 mm. and that of the mackerel, 198 mm. This latter value is similar to that found by Keys and Hill (3) for Anguilla, the European eel, and verified in this series. The ranges for individuals of these species are seen to be mutually exclusive, but those for species nearer the middle of the total range overlap freely and some are indistinguishable, as for instance the two species of Prionotus. To what this characteristic range for a species is related is not known though various suggestions have been made. Arterial pressure determinations on fish are few and reliable determinations still fewer (20), while determinations of capillary pressure have not been 522 ABBY H. TURNER made at all. The degree of habitual activity has been suggested as a clue, for Lophius is proverbially sluggish and Scomber very active, while the extensive migrations of Anguilla are well-known. ‘The series is as SPECIES Fs a D TELEOSTS SCOMBER SCOM BRUS ANGUILLA VULGARIS BELONE ACUS ANARHICHAS LUPUS BROSMIUS BROS ME MOLVA MOLVA GADUS MORRHUA GADUS POLLACHIUS GYCLOPTERUS LUM PUS LOPHIUS PISGATORIUS GADUS VIRENS ELASMOBRANCHS RAIA FULLONICA CHIMAERA MONST ROSA GALEUS VULGARIS SQUALUS ACANTHIAS PRISTIURUS CATULUS RAIA OAY RHYNCHUS ETMOPTERIS SPINAX CuHarT 1. Refractive indices of colloid fraction in Norwegian series of teleosts and elasmobranchs. The left end of each line shows the refractive index of the ultrafiltrate, the right end that of the whole serum. The length of the line therefore shows the index of the colloid fraction. The ultrafiltrates in all of the teleosts except two are closely grouped, indicating relative constancy in the non-colloid components of thesera. The two exceptions are in fish which were represented by limited samples only, in the case of Belone by samples for ultrafiltration of only half the usual size. The high position of the ultrafiltrate index for all elasmobranchs is related to their high non-protein nitrogen fraction. yet too short to give an answer here. The possibility of other sub- stances which may add to the colloid osmotic pressure of the serum, e.g. fats (21, 22) is not to be excluded particularly when one occasion- SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 523 ally sees globules of fat at the top of a centrifuged tube of blood and infers the presence of associated substances in solution in appreciable amounts. Also the wide variation in the fat content of fish is a matter of common knowledge. No determinations on serum fats have been made. The albumin-globulin ratios in fish are said by Lepkovsky —— HA "sate 222% COLLOID ex @ PROTEIN N tb hes Ps SCOMBER SCOMBRUS ANGUILLA VULGARIS BELONE ACUS ANARHICHAS LUPUS BROSMIUS BROSME MOLVA MOLVA GADUS MORRHUA, GADUS POLLACHIUS CYCLOPTERUS LUMPUS LOPHIUS PISCATORIUS GADUS VIRENS Cuart 2. Comparison of colloid osmotic pressure, refractive index of colloid fraction, and protein nitrogen in Norwegian teleost series. The several ranges have been adjusted so that the maximum values are shown by lines of approximately equal length. The relative values of the three measurements thus appear, with a range much wider in c.o.p. than in np or protein N. This is probably to be explained by a variation in the albumin-globulin ratio in the direction of large molecules when the total quantity is small. (11) to vary widely. Such variations, with their attendant contrast in the size of the protein molecules, may help to explain not only differences between species but irregularities in the relationships be- tween colloid osmotic pressure and other protein data. A few pre- liminary studies on plasma protein regeneration after depletion by 524 ABBY H. TURNER . hemorrhage seem to indicate that protein may come back rather quickly, perhaps more quickly in total quantity than in colloid osmotic pressure, which may be the same as saying that the globulins re- generate—or are gotten into the blood—more rapidly than the albumins. This obviously suggests the experiments on mammalian protein regeneration from the laboratories of Whipple and Weech, to whose papers only partial references are given (23, 24, 25). As the refractive indices are scrutinized, it is apparent that the range for the ultrafiltrates is narrow, indicating constancy in the non- protein fraction of the plasma, while the refractive index for the colloid fraction varies widely from species to species and accompanies the c.o.p. though the correspondence is not too close. (See Chart 1.) The analyses for total and non-protein nitrogen similarly give values for the serum proteins which vary approximately with the c.o.p. but follow more closely the refractive indices of the colloid fractions. These correspondences and differences are brought out in Chart 2. It is to be noted that the range of c.o.p. from the species of low to those of high values is much greater than the range in protein content. This fact may be associated with the ready appearance of protein molecules of large size mentioned above. It is possible that in the life economy of teleosts it is of especial importance to hold the total protein content of the serum at a definite level while the assortment of proteins may vary. It would follow from this that a rigid maintenance of colloid osmotic pressure is impossible or unnecessary. For the elasmobranchs it is to be said that all averages for colloid osmotic pressure are low, though occasional individuals have been found as high as. 70 mm. water pressure. As expected, the total and non-protein nitrogen findings, the refractive indices for total sera and for ultrafiltrates are all high, reflecting the high urea or trimethylamine oxide content of elasmobranch blood. The protein nitrogen figures are low with two exceptions, one specimen of Dasybatus marinus, stingray, and three of Raia fullonica, a Norwegian skate. Differences in the physiological make-up of elasmobranchs and teleosts thus include the protein blood picture. SUMMARY AND CONCLUSIONS 1. In a study of 121 individuals of 19 species of marine teleosts it has been found that the range of colloid osmotic pressure shown by the blood serum is wide for a single species; that the c.o.p. shows a characteristic range in each species; and that for the species studied the c.o.p. varies from an average of 42 mm. water pressure in Lophius piscatorius, Lin., goosefish, to about 200 mm. in Scomber scombrus, SERUM PROTEIN MEASUREMENTS LOWER VERTEBRATES, II 525 Lin., mackerel, Anguilla vulgaris, Tur., eel, and probably others. Intermediate values for averages within species were found most commonly, from about 100 to 150 mm. 2. The refractive indices for the colloid fractions of the sera and the protein nitrogen values follow the same general distribution as the c.o.p. though correspondence is not perfect. The c.o.p. seems to vary through a much wider range than the protein N or the refractive index of the colloid fraction. 3. All protein values for the 48 individuals of 11 species of elasmo- branchs were found to be low, distinctly lower than in all save the very lowest of the teleosts. 4. The refractive indices of the ultrafiltrates from the sera in all teleosts were very constant as compared with the wide range of the colloid figures. ‘This constancy, reflecting the crystalloid status of the sera, is also characteristic of the elasmobranch ultrafiltrates though the absolute level is materially higher in the latter group, a fact associated obviously with the high non-protein nitrogen content of elasmobranch blood. BIBLIOGRAPHY 1. CAMPBELL, M. L., AnD A. H. Turner, 1937. Serum protein measurements in the lower vertebrates. 1. The colloid osmotic pressure, nitrogen content, and refractive index of turtle serum and body fluid. Bzol. Bull., 73: 504. Preliminary note in Am. Jour. Physiol., 116: 24 (Proc.). 2. DRINKER, C. K., ano M. E. FIeLp, 1933. Lymphatics, Lymph, and Tissue Fluid. Baltimore. 3. Keys, A., AND R. M. Hitt, 1934. The osmotic pressure of the colloids in fish sera. Jour. Exper. Biol., 11: 28. 4. Turner, A. H., 1935. The colloid osmotic pressure of the blood plasma in fishes. Am. Jour. Physiol., 113: 132 (Proc.). 5. TurNER, A. H., 1936. Serum protein of fishes: colloid osmotic pressure and other data. Am. Jour. Physiol., 116: 155 (Proc.). 6. Dents, W., 1913. Metabolism studies on cold-blooded animals. II. The blood aril nice of fish. Jour. Biol. Chem., 16: 389. 7. Denis, W., 1922. The non-protein eens constituents in the blood of marine fish. Jour. Biol. Chem., 54: 693. 8. GRAFFLIN, A. L., 1935. @ilecde and total osmotic pressure in the blood of marine teleosts. Biol. Bull., 69: 245. 9. GraFFLIN, A. L., 1936. Renal function in marine teleosts. III. The excretion of urea. Biol. Bull., 70: 228. 10. Keys, A., 1933. The mechanism of adaptation to varying salinity in the common eel and the general problem of osmotic regulation in fishes. Proc. Roy. Soc., Ser. B., 112: 184. 11. LeprKovsxy, S., 1930. The distribution of serum and plasma proteins in fish. Jour. Biol. Chem., 85: 667. 12. Smitu, H. W., 1929. The composition of the body fluids of elasmobranchs. Jour. gia Chem., 81: 407. 13. Smitu, H. W., 1929. The composition of the body fluids of the goosefish (eve aisea terre). Jour. Biol. Chem., 82: 71. 526 ABBY H. TURNER 14. 5 16. il7fe 18. iQ). 20. alle Dyup US). 24. 25% SmitTH, H. W., 1932. Water regulation and its evolution in the fishes. Quart. Rev. Biol., 7: 1. Situ, H. W., 1936. The retention and physiological réle of urea in the Elasmo- branchii. Biol. Rev., 11: 49. BIGELow, H. B., AnD W. W. WEtsu, 1924. Fishes of the Gulf of Maine. Bull. U.S. Bur. Fish., 40 (Part 1): 1. OTTERSTROM, C. V., Fisk. Danmarks Fauna. 3 vols. No. 11, 1912; No. 15, 1914; No. 20, 1917. Copenhagen. Krocu, A., AND F. Nakazawa, 1927. Beitrage zur Messung des kolloid- osmotischen Druckes in biologischen Fliissigkeiten. Biochem. Zeitschr., 188: 241. TurRNER, A. H., 1932. The validity of determinations of the colloid osmotic pressure of serum. Jour. Biol. Chem., 96: 487. VON SKRAMLIK, E., 1935. Uber den Kreislauf bei den Fischen. Ergebn. d. Biol., 11: 1. FisHBERG, E. H., 1929. The relations of serum proteins and lipids to the osmotic pressure. Jour. Biol. Chem., 81: 205. Man, E. B., AND J. P. PETERS, 1933. Permeability of capillaries to plasma lipoids. Jour. Clin. Invest., 12: 1031. MADDEN, S. C., P. M. WinsLow, J. W. HowLanp, AND G. H. WHIppPLe#, 1937. Blood plasma protein regeneration as influenced by infection, digestive disturbances, thyroid, and food proteins. A deficiency state related to protein depletion. Jour. Exper. Med., 65: 431. KnutTl, R. B., C. C. Erickson, S. C. MappEn, P. E. REKERs, AND G. H. WuHippLeE, 1937. Liver function and blood plasma protein formation. Normal and Eck fistula dogs. Jour. Exper. Med., 65: 455. WEECcH, A. A., E. GOETTSCH, AND E. B. REEVEs, 1935. Nutritional edema in the dog. I. Development of hypoproteinemia on a diet deficient in protein. Jour. Exper. Med., 61: 299. CENPMES SAND ei aISiOlOG i tOr Trl COLOR 2 Adan IN THE NORMAL AND ALBINO PARADISE FISH, MACROPODUS OPERCULARIS L.! H. B. GOODRICH AND MAURICE A. SMITH (From Wesleyan University and the Marine Biological Laboratory, Woods Hole, Mass.) The paradise fish is well known to fish fanciers. It is thought to have been the first ‘‘fancy”’ tropical fish to be bred in Europe, having been first introduced in Paris by Carbonier in 1868. Specimens of the albino variety utilized in this investigation were obtained from dealers in New York City during the fall of 1934. These were at that time a novelty and it is stated by Innes (1935) that they were first imported from German fanciers in 1933. The paradise fish is one of the labyrinth fishes and is referred to Order Labyrinthici, Family (368) Osphronemidz in the classification by Jordan (1923). It is also known in the literature as Macropodus viridi-auratus. A description is given by Regan (1909). This fish is found in the lowland streams of China, Formosa and Cochin China. It is well fitted to live in stagnant waters because its accessory re- spiratory apparatus, the labyrinth organ, allows it to use atmospheric oxygen as well as to breathe by gills. On this account almost no attention needs to be given to aeration of water in an aquarium and it may be kept in very small containers. The paradise fish is a “bubble nest’’ builder. The male, pre- sumably by aid of some oral secretion, makes a floating structure of bubbles into which he shoots the eggs as they are laid by the female. The mating and spawning activities, in our experience, take place at temperatures ranging from 26° C. to 29° C. Certain fish dealers, how- ever, state that they will breed at as low a temperature as 20° C. Studies made by Goodrich and Taylor (1934) on a species Betta splendens, a related genus, showed a remarkably precise limitation of the breeding activities to a temperature of about 26°C. It is neces- sary, under aquarium conditions, to remove the female after laying, as she may eat the eggs, and to remove the male after hatching as he may eat the young fish. The young fry are first fed from Paramecia cultures and later with Daphnia. The adults are fed with enchytraeids 1 This paper is published as a part of a research program at Wesleyan University supported by the Denison Foundation for Biological Research. The senior author wishes to acknowledge his indebtedness to his assistant, Miss Priscilla Anderson, for carrying certain parts of this work to completion. 527 528 H. B. GOODRICH AND MAURICE A. SMITH (the white worms of dealers), minced earthworms, liver and dry pre- pared fish foods. Further details of the care of the fish are given by Innes (1935). The albino variety is much more difficult to rear. It is less viable and takes food less readily. It seems to avoid light more than the dark variety and possibly for this reason it is less likely to discover food. GENETIC EXPERIMENTS The dark or normal colored paradise fish when crossed with the albino variety produced all dark type in the F; generation. The F2 in two separate crosses gave the results shown in Table I. The results TABLE I Fy from cross between normal and albino varieties Date of spawning ee of Number. of Ratio OES =3 Susie esate ees 204 62 3.29/1 IEDAES GAG ct weet ase see Sais 438 145 3.02/1 F, from cross between heterozygous dark and albino Date of spawning Supe of ee of Ratio OE ES Spe te ok rn. wire 226 221 1.02/1 SoIDeS On eee rn eare hoes ine OE 552 551 1.002/1 Counts in both cases made eight days after spawning. of the back cross between the F; and the albino are also shown in Table I. Similar results were independently obtained by Kosswig (1935) though with somewhat greater deviation from the expected ratios. Our counts were made eight days after spawning and before the onset of the high mortality which is unfortunately frequent in fish cultures. These results clearly indicate that the normal and albino types are a pair of mendelian allelomorphs in which the normal is dominant over the albino. THE COLOR PATTERN The normal or wild type paradise fish (Fig. 1) is strikingly marked on the sides with vertical stripes which are referred to in this paper as green and red stripes. The color, however, varies with environmental conditions. ‘The striping may largely disappear when the fish is placed against a white background. The green stripes may be more accurately described as being a metallic blue-green when viewed from an angle which strongly reflects light. From other angles these stripes may COLOR PATTERNS IN THE PARADISE FISH 529 appear much darker, giving a mottled blue and black appearance. The red stripes are of an orange-red hue. The markings in the male are more brilliant than in the female, especially at the breeding period. The albino (Fig. 2) is light colored and has pink eyes but shows faint orange and blue-green stripes. As will be shown later in detail, and as has been mentioned by Kosswig (1935), the black pigment is entirely absent. HISTOLOGY OF THE COLOR PATTERN The cells involved in the production of the color pattern are the black cells or melanophores, yellow cells or xanthophores, red cells or erythrophores and those containing reflecting crystals or iridocytes. There is some evidence for considering that the erythrophores develop from the xanthophores in this fish. The xanthophores contain yellow pigment granules soluble in 95 per cent alcohol in about two hours. F1¢. 1. Normal © The erythrophores ee contain orange-red granules less read- ily soluble in alco- Fic. 2. Albino ‘ paradise fish. The hol but sometimes pictures of the two xanthophores are fish were taken observedwhichcon- simultaneously in tain a few granules C7> 79° similar to those in the erythrophores. The erythrophores appear later in development and old fish show more red pigment than young fish. We recognize three zones or areas of disposition of chromatophores in relation to the individual scales. All of these are probably dermal as they lie beneath the stratified epithelial layer of the epidermis. The relationships of these areas are illustrated by Figs. 3 and 4 which show diagrammatic reconstructions of median longitudinal sections of a series of scales taken perpendicular to the body surface. ‘These diagrams are based upon microtome sections made by three freezing methods and upon dissections of the body wall made under a dissecting binocular microscope. ‘The first or superficial zone (a) (Figs. 3 and 4) lies on the upper outer surface of the scale and is connected by the second and upward sloping intermediate zone (6) with the third or deep zone (c) located on the underside of the next anterior overlying scale. This deep zone also lies above a stratum of fat cells (d) which separates it from the inner upper surface of the underlying scale. 4 aK ———— . . | | 1 | | | Tics aoe | S| we Sf nano \ETAUINAAT LT} es 9 aa ao ava Pissscee GEEZ H mal | a 0) EY) ED =e WIRIAAT Al ba ESAS VS OQWES FIG. 3 NORMAL AK GIRIZIEIN) SWRulee GIRIEZIEIN) SWRI aA AWA ZF eavnau' ALBINO oN mae CS 2p HEX a0 BY VYYT eT a (US RED STRIPE Rizo) Siyailrte* DAP eT = SAY BOD: ZS R LD AY IRIDOCYTE 2 ERYTHROPHORE © XANTHOPHORE Lat @ MELANOPHORE PAR bese; wy, zy LF Woe xz = anvanes200s8428 rs OSE a ue (0 8 8 Fic. 3. Normal dark fish. Diagram of median longitudinal section of a series of scales. Magnification < 80. (A) Superficial zone of chromatophores; (B) inter- mediate zone of chromatophores; (C) deep zone of chromatophores; (D) stratum of fat cells; (Z) connective tissue layer; (F) stratified epithelium; (G) scale; (H/) muscle. Chromatophores reduced to about 75 per cent of actual number for clarification of diagram. Fic. 4. Albino fish. Details as in Fig. 3. 530 COLOR PATTERNS IN THE PARADISE FISH 531 TABLE II Dark-colored fish. Counts of chromatophores in area of 0.054 sq. mm. M, melanophores, X, xanthophores, E, erythrophores, I, iridocytes. Superficial zone Green stripe Red stripe M x E I M x E I 2D, 20 0 6 15 26 0 3 23 18 0 3 12 20 0 D, 17 17 0 5 13 20 0 1 18 24 0 4 14 18 0 2 19 ‘23 0 3 18 21 0 5 19 17 0 3 19 20 0 0 25 19 0 4 20 19 0 6 15 22 0 2 21 26 0 3 18 22 0 4 21 23 0 6 19 18 0 0 15 25 0 7 Average first 10....] 19.5 | 20 0 3.4 | 16.8 | 20.6 0 3.5 | Total Average first 20....| 21.2 | 20.2 0 3.6 | 18.5 | 20.8 0 3.6 | 87.9 Deep zone Green stripe Red stripe M x E I M x E I 43 0 34 19 11 3 4 42 0 40 19 5 5 2 38 0 36 17 9 3 2 25 0 25 15 3 3 0 36 0 36 18 5 3) 1 30 @) 20 17 14 3 3 27 0 27 18 8 5 0) 28 0 28 15 15 3 1 38) 0 35 18 9 3 3 37 0 37 14 5 3 3 Average first 10....| 34.1 0 SiS Waly 8.5 | 3.4 1.9 | Total Average first 20... .| 34.1 0 32.5 | 18.5 That Nero) 2.8 | 99.2 Only the first ten of twenty counts are tabulated but the average of all twenty is given. Counts of iridocytes have a greater possibility of error than that of other chro- matophores because it is difficult to determine the boundaries of individual cells. Counts of the xanthophores from the deep zone of the dark fish (Table II) are not tabulated because they were so covered by melanophores and iridocytes that no accurate count could be made, but they are few in number or absent. 532 H. B. GOODRICH AND MAURICE A. SMITH In the dark or normal type of paradise fish (Fig: 3) the upper zone contains a fairly uniform distribution of melanophores, xanthophores and iridocytes. The iridocytes tend to be superficial to and often TABLE III Albino fish. Counts of chromatophores as in Table II. Superficial zone Green stripe Red stripe M xX E I M x E I 0 20 0 14 0 30 0 16 0 27 0 13 0 28 0 19 0 26 0 17 0 30 0 14 @) 29 0 16 0 31 0 17 0 25 @) 14 0 31 0 18 0 27 0 18 0 29 0 10 @) 23 0 10 0 26 0 14 0 24 0 11 0 30 0 15 0 28 0 14 0 23 0 10 0 31 0 17 0 24 0 19 Averace first 10s Oe 26.8 0 faa ON E2802. iO, 15.29) Total Average first 20....| 0 27.4 0 16.4 0 27.6 0 Tots eves Deep zone Green stripe Red stripe M x E I M xX E I 0 17 0 20 0 13 4 18 0 18 0 23 0 7 3 17 0 12 0 19 0 8 6 10 0 15 0 20 0 6 5 10 0 20 0 17 0 10 3 21 0 15 0 16 0 4 5 11 0 Bil 0 22 @) 13 3 22 0 34 0 23 0 6 4 20 0 31 0 26 0 11 6 15 0 33 0 34 0 10 4 18 8.8 | 4.3 | 16.2 | Total Average first 10.... : : 0 AMNeTh 0 Dae 0 8.2 | 4.4 | 15.4 | 72.8 Average first 20.... directly above melanophores in all zones. Table II shows the relative numbers of these cells in both the green and the red stripes. The counts in the table were made on scales removed from the body and treated with adrenalin to cause a concentration of pigment in the cells COLOR PATTERNS IN THE PARADISE FISH 533 and so to help to distinguish one cell from another. The counts were of cells seen within one quadrant of an ocular counting disc. It is calculated that the area so outlined is 0.054sq.mm. These counts will therefore when multiplied by the factor 18.4 give the approximate number of cells per square millimeter. It will be noticed that there is probably no significant difference between numbers of cells of the upper zone in the green and in the red stripes. It is the deep zone which provides the basis of the striping. Counts were made on the deep zone after removal of scales and after washing the side of the body with an adrenalin solution. It will be noted in Table II that in the deep zone the green stripes, in contrast with the red stripes, show a significant excess of melanophores and of iridocytes and but few or no xanthophores or erythrophores and that these latter are found in the red stripes. The intermediate zone very seldom contains any erythrophores and there are also fewer iridocytes than in the outer zone. It is much the same whether located in a green or a red stripe and therefore is, like the outer zone, neutral in regard to striping. No melanophores are found in either stripe in the albino fish (Fig. 4 and Table III). Tests with adrenalin to concentrate pigment (cf. Goodrich, 1927) or use of the ‘“‘dopa”’ reaction (cf. Goodrich, 1933) failed to indicate the presence of any ‘‘colorless’’ melanophores. Otherwise the disposition of cells is similar to that in the normal type except that there are more xanthophores and iridocytes. DEVELOPMENTAL HISTORY The melanophores first appear at about the 16-somite stage -(nineteen hours in our cultures). These are at first rather irregularly distributed and there is no indication of pattern formation until the fish is from nine to twelve days of age (about 3 mm). At this time from six to eight spots appear on both the mid-dorsal and mid-ventral aspects of the body and tail region. ‘The striping first becomes visible in the 10 to 14 mm. fish. The number of stripes is comparable with the number of earlier spots but no certain relationship has been demonstrated. DISCUSSION Histological studies on the light color phases of various fish have shown that different cellular complexes may produce these light types. In A plocheilus (Oryzias) latipes a double recessive (Aida, 1921) owes its absence of color to a reduction of amount of melanin produced in melanophores and of xanthine in xanthophores (Goodrich, 1927). This circumstance in other animals has given rise to the term ‘‘color- 534 H. B. GOODRICH AND MAURICE A. SMITH less’’ chromatophores. In this case the cells seem incapable of producing the normal amount of chromogen although sufficient oxydase is present (Goodrich, 1933). In the goldfish it is found that the light type, the transparent shubunkin, owes its condition to the early disintegration of melanophores and erythrophores (Goodrich and Hansen, 1931). In the paradise fish as described above no melano- phores of any type have been seen at any stage of development. This latter form is a true albino, as no pigment is present in the eyes in contrast with the other types named above. The senior author of this paper has unpublished observations on an albino trout which has no pigment in the eyes but does have melanophores with reduced pigment as in Oryzias. In each case among fish so far investigated in histolo- gical detail a different developmental mechanism is concerned in producing the light phase. SUMMARY 1. In the paradise fish, Macropodus opercularis, the dark-colored or normal type is a Mendelian dominant to the albino. 2. A description is given of the cell groupings which form the basis of the color pattern. 3. The melanophores are entirely absent from the albino but all other types of chromatophores are present. LITERATURE CITED Apa, TATuo, 1921. On the inheritance of color in a fresh-water fish, Aplocheilus latipes Temmick and Schlegel, with special reference to sex-linked in- heritance. Genetics, 6: 554. ; Goopricu, H. B., 1927. A study of the development of Mendelian characters in Oryzias latipes. Jour. Exper. Zo6l., 49: 261. Goopricu, H. B., 1933. One step in the development of hereditary pigmentation in the fish Oryzias latipes. Bzol. Bull., 65: 249. Goopricu, H. B., AND I. B. HANsEN, 1931. The postembryonic development of Mendelian characters in the goldfish, Carassius auratus. Jour. Exper. Zodl., 59: 337. GoopricHu, H. B., AND HoytC. Tayior, 1934. Breeding reactions in Betta splendens. Copeia, December 31, p. 165. INNES, WILLIAM T., 1935. Exotic Aquarium Fishes. Philadelphia. : Jorpan, D.S., 1923. A classification of fishes, including families and genera as far as known. Stanford Univ. Pudl., Univ. Ser., Biol. Sct., 3: 77. Kosswic, Curt, 1935. Uber Albinismus bei Fischen. Zool. Anzeig., 110: 41. REGAN, C. TATE, 1909. The Asiatic fishes of the family Anabantide. Proc. Zodl. Soc. London. Nov. 9: 767. CHROMATOPHORE REACTIONS IN THE NORMAL AND ALBINO PARADISE FISH, MACROPODUS OEE RGULARIS VE: H. CLARK DALTON AND H. B. GOODRICH 1! (From the Shanklin Laboratory of Biology, Wesleyan University) The paradise fish, Macropodus opercularis L., was chosen for this study because there are two varieties, the normal dark type and the albino. It was thought that the albino would afford special oppor- tunity for studying the erythrophores and xanthophores because these are not hidden by the melanophores, as they are in the dark type. These two varieties would then give a good basis for comparing the reactions of the three types of chromatophores. The paradise fish is well known to fish fanciers. Its native habitat is the shallow coastal streams of southeastern Asia. Details in regard to the characteristics and care of this fish may be found in the book on exotic aquarium fish by Innes (1935). The albino variety apparently was first imported into this country from Germany in 1933. This variety has been shown by Kosswig (1935) to be a Mendelian recessive to the normal dark form. The histological details of the color pattern of the two varieties are described by Goodrich and Smith (1937). CoLoR REACTIONS The adaptations of the paradise fish to various colored backgrounds have been studied by methods used by previous investigators (cf. Mast, 1916; Connolly, 1925, and bibliography by Parker, 1930). The fish were subjected to tests in small aquaria with white, black, red, yellow, and blue paper jackets and uniform illumination from above. Macroscopic observations on the normal dark type showed typical color adaptations to each environment. The color responses to the red and the yellow backgrounds were relatively similar but yet seemed to be definitely distinguishable. Thealbino fish, in which melanophores are entirely lacking (Kosswig, 1935; Goodrich and Smith, 1937), showed differing responses against black, white, red and yellow backgrounds. The greatest contrast, however, was between the very pale appearance, when placed against the white background, and the 1 This paper is published asa part of the research program of Wesleyan University supported by the Denison Foundation for Biological Research. 535 536 H. CLARK DALTON AND H. B. GOODRICH various slightly differing combinations of yellow, orange and red given in the reactions to the black, yellow and red backgrounds. The results of microscopic examinations of the changes in the chromatophores are summarized in Table I. Twenty normal dark fish were available but only four albinos. Every one, however, of these albinos was tested against each background. It will be noted that in the normal dark fish the black, yellow and red cells react essentially alike to environments of black or white or red by showing the dispersed pigment in the black and red environments and concen- trated pigment in the white environment. On the yellow background, however, the pigment in the black cells is concentrated, while that in the yellow and red tends to be dispersed. This gives the fish a light yellow color. The reverse situation occurs in blue surroundings, where the yellow and red cells show concentrated pigment and the black cells TABLE [| Chromatophore behavior in the paradise fish D = dispersed pigment; C = concentrated pigment; J = intermediate state. Dark variety Albino variety Mel. Xan. Ery. Mel. Xan. Ery. Bilan mata kas eee D D D and C — D D AVAL C= a sees a eos ins C C Ge — I Cand I Rede hemes D D DandC — D D Mellowey Struss eee ee: C D ITandC — D D Beng nse ce ee D C C — Cand J | Cand J dispersed pigment. The reactions of the melanophores are more rapid than those of the other two types. The erythrophores behaved in general similarly to the xanthophores, except that they were slightly slower and less extensive in their reactions. EXPERIMENTS WITH DENERVATED CHROMATOPHORES After the study of the color changes and their relation to the activities of the chromatophores, attention was next turned to the problem of the physiological control of the chromatophores. The methods used by G. H. Parker (19340) were adopted.? The nerves supplying a part of the caudal fin are severed by making a cut across one of the fin rays. This operation leaves a small area between the cut and the edge of the fin without central nervous control. The 2 Kamada (1937), whose paper has just come to hand, has also induced caudal bands in Macropodus opercularts. CHROMATOPHORE REACTIONS IN PARADISE FISH 537 normal dark fish lay quietly in the dish when the incision was made, but the albino required an anaesthetic, for which purpose ether was used. The experiments, except as otherwise noted, were carried out at room temperature, which remained relatively uniform, the water tempera- ture averaging 23.5° C. Immediately upon sectioning the nerve of a dark-colored, or normal fish the melanophores in the area between the cut and the outer edge of the caudal fin began to disperse their pigment. A clearly-defined dark band formed within one minute and it extended laterally to include half of the area between the cut ray and the two adjoining rays. When the pigment of the melanophores is dispersed, it is more difficult to observe other types of pigment cells. Nevertheless, it could be seen that the pigment in the xanthophores was also dispersed, but the reaction was somewhat slower than that of the melanophores. Three to five minutes was required for a degree of dispersion comparable to that shown in one minute by melanophores. The erythrophores lagged slightly behind the xanthophores in dispersing their pigment. These observations were confirmed later on albinos, where no melanophores obscure other types of cells. When fish with these dark caudal bands were placed in white porcelain dishes illuminated from above, the dark bands faded in about ten hours, as the denervated melanophores gradually assumed the punctate condition. The reverse reaction, of denervated melanophores assuming the dispersed phase after concentration, is much more rapid, taking place in two to three hours. As noted above, the xanthophores and erythrophores can be easily observed in the albino. The sectioning of a nerve was followed in one or two minutes by a detectable dispersion of pigment in these cells, which, however, take several minutes to complete the reaction. This band, although it contains no melanophores, is nevertheless striking because the rest of the fin with punctate xanthophores and erythro- phores is very light. Figure 1 shows the edge of such a caudal band in the albino fish. Although these albino fish remained under the same conditions as the normal ones, there was a marked difference in that the caudal bands faded in about four hours, or less than half the time required by those in the dark-colored fish. Certain other experiments devised by G. H. Parker (1934c), which have special bearing on the neurohumoral hypothesis, were tried with especial reference to the reactions of the erythrophores. The material was limited, but the results seem to be clear. Dark fish were operated on to produce a caudal band in each and then placed in white dishes until the bands were completely faded. Other cuts were then 538 H. CLARK DALTON AND H. B. GOODRICH made producing two fresh dark bands, one on either side of the distal half of the faded band. The chromatophores in the proximal half of the faded band remained concentrated, as were those in other parts of the tail (the new dark bands excepted). In the distal half, however, the chromatophores became dispersed, corresponding to the condition of their immediate neighbors in the fresh dark bands. The dispersal occurs first and to the greatest extent at the sides of the faded band. It will be recognized that these observations parallel those made by G. H. Parker on Fundulus and so support the theory of neurohumors as he has applied it to explain the reactions of chromatophores in fish. The humor is thought to spread from the new dark band, where it has . tae io Fic. 1. Photomicrograph of edge of caudal band in albino fish. The pigment cells are erythrophores. 95 X. been recently liberated, to the faded band and to stimulate there the dispersion of pigment in the melanophores. The same experiment was tried on albinos to test the occurrence of a similar dispersing neurohumor governing the erythrophores. A definite expansion was obtained in the distal half of the flanked bands in two albinos in which both lateral bands were successfully induced. It did not occur in two cases where only one flanking band was produced. The reactions of the denervated melanophores differ from those of the xanthophores and erythrophores. This was shown by observations made when normal dark paradise fish with faded caudal bands were transferred from a white dish to a yellow environment. The melano- CHROMATOPHORE REACTIONS IN PARADISE FISH 539 phores remained concentrated, but the xanthophores and erythrophores became gradually dispersed. Each type of chromatophore assumed the same condition of pigment distribution as the corresponding type of innervated pigment cells in the surrounding area. TEMPERATURE EXPERIMENTS As a further test of the neurohumoral hypothesis, an experiment was carried out to determine the effect of temperature on the rate of disappearance of bands produced by cutting nerves in the caudal fins of normal fish. If the fading of such bands is indeed due to the dispersion of an oil-soluble substance from cell to cell, as Dr. Parker’s theory indicates, it might be expected that the dispersion should occur more rapidly at a higher temperature than at a lower one, in accordance with laws of diffusion. Accordingly, fish were operated on as previ- ously described for the formation of dark caudal bands and isolated in white porcelain dishes illuminated from above. Two determinations were made with each fish, one at a temperature of 20° C. and the other at about ten degrees higher. The bands were considered faded when they could no longer be distinguished by the unaided eye when the fish were observed in a paraffin-lined Petri dish with a white surface in the background. The following table summarizes the data obtained: TABLE II Effect of temperature on rate of caudal band disappearance in normal fish Observations 1-4: Low = 20° C., high = 29° C. Observation 5: Low = 23° C., high = 27°C. Figures are elapsed hours after operation when bands became so faint as to be practically indistinguishable. Low High TN e a SaaS 18 hrs. 8 hrs. DR Aa RR 10 5 ES Oi ote 10 5 AR OS Charo), 10 5 Bio SAG eee US 3 The results of this experiment are what would be expected on the basis of the neurohumoral hypothesis; i.e., that the dark bands disappear more rapidly at a higher temperature than at a lower one. Furthermore, it appears that with a rise in temperature of approxi- mately ten degrees, the time required for disappearance is reduced very nearly one half. DISCUSSION The action and control of chromatophores in the paradise fish are found to be similar to that reported for certain other species. The -rate of reaction of these processes, however, is strikingly different. 540 H. CLARK DALTON AND H. B. GOODRICH This disparity shows particularly in the rate of disappearance of dark bands produced by denervating chromatophores in the caudal fin. Caudal bands in Fundulus (Parker, 1934a) fade in twenty-two to ninety-six hours, while those in the catfish (Parker, 19345) fade only after two or three days and may last as much as seven days. In contrast to these conditions, bands in the normal paradise fish fade in approximately ten hours, and in the albino, in about four hours. The activity of the concentrating neurohumor, then, is more rapid than in the fish previously reported. This greater speed of action applies also to the dispersing neuro- humor. Parker (1934a) reported that in Fundulus light bands, produced by transferring fish with faded caudal bands from white surroundings to black, fade in seventeen to twenty-seven hours. In the paradise fish, however, similar bands fade in only two and a half hours. That caudal bands in the albino should disappear in approximately half the time taken by fading bands in the normal fish is a puzzling problem. Conceivably the neurohumor (or humors) produced in regulating the xanthophores and erythrophores might be more soluble than that for controlling melanophores and, consequently, become dispersed more readily than the latter. At any rate, a difference in chemical or physical properties of the two substances seems to be indicated. The independence of action of melanophores and xanthophores is similar to that described by Abramowitz (1936) for Fundulus majalis. In the paradise fish, however, the erythrophores have also been studied and their reactions are found to be similar to but a trifle slower than those of the xanthophores. This gives additional evidence of the pos- sible close relationship between xanthophores and erythrophores as suggested by Goodrich and Smith (1937). The results obtained which show that an increase of temperature brings about an increase of rate of disappearance of the dark caudal bands may have bearing on the theory of neurohumoralism. It supports the concept that the effect is produced by a diffusing chemical substance. SUMMARY 1. The normal paradise fish adapts itself by appropriate color changes to environments of black, white, red, yellow, and blue. 2. Analogous but less adaptive changes occur in the albino. 3. The reactions of melanophores, xanthophores and erythrophores which produce these color changes are described. 4. Dark caudal bands, formed by cutting chromatophore nerves in the caudal fin of normal paradise fish fade in approximately ten hours. Similar bands in the albino fish fade in about four hours. CHROMATOPHORE REACTIONS IN PARADISE FISH 541 5. The evidence indicates the presence of independent dispersing neurohumors for melanophores and for erythrophores. 6. The rate of caudal band disappearance is directly proportional to the temperature, and a rise of about ten degrees in temperature very nearly doubles this rate. LITERATURE CITED ApraAmowitz, A. A., 1936. The non-identity of\the neurohumors for the melano- phores and the xanthophores of Fundulus. Am. Nat., 70: 372. ConnotLy, C. J., 1925. Adaptive changes in shades and color of Fundulus. Bzol. Bull., 48: 56. Goopricu, H. B., anp M. A. Smitu, 1937. Genetics and histology of the color pattern in the normal and albino paradise fish, Macropodus opercularis L. Biol. Bull., '73:527.. InnEs, W. T., 1935. Exotic Aquarium Fishes. Innes Publishing Co., Philadelphia. 463 pp. KamapA, TAKEO, 1937. Parker’s effect in melanophore reactions of Macropodus opercularis. Proc. Imp. Acad. Tokyo, 13: 217. Kosswic, Curt, 1935. Uber Albinismus bei Fischen.. Zool. Anz., 110: 41. Mast, S. O., 1916. Changes in shade, color, and pattern in fishes, and their bearing on the problems of adaptation and behavior, with especial reference to the flounders Paralichthys and Ancylopsetta. Bull. U. S. Bur. Fish., 34: 173. ParKER, G. H., 1930. Chromatophores. Bzol. Rev., 5: 59. Parker, G. H., 1934a. Cellular transfer of substances, especially neurohumors. Jour. Exper. Biol., 11: 81. ParkKER, G. H., 1934. The prolonged activity of momentarily stimulated nerves. Proc. Nat. Acad. Sct., 20: 306. Parker, G. H., 1934c. Neurohumors as activating agents for fish melanophores. Proc. Am. Phil. Soc.,'74: 177. AUTOSOMAL LETHALS IN WILD POPULATIONS OF DROSOPHILA PSEUDOOBSCURA A. H. STURTEVANT (From the William G. Kerckhoff Laboratories, California Institute of Technology, Pasadena, California) INTRODUCTION It has been recognized for some years, following the work of Muller, that lethals are especially convenient material for the study of muta- tion rates. This is because they occur with a frequency that is great enough to be measured, and because their occurrence can be detected by a technique that is independent of the personal equation of the observer. These same two advantages apply to the use of lethals in the study of the constitution of wild populations. Two other ad- vantages are also to be noted in this field. Owing to the extensive studies on the mutation rates of lethals, there is available a large body of data on the frequency of the occurrence of new lethals under a variety of conditions. On the other hand, the rate of elimination of lethals from a population furnishes the minimum possible difficulty of quantitative estimation. It is, therefore, not surprising that there is a rapidly increasing literature in the field—to which the present paper belongs. MATERIAL AND METHODS A large series of wild strains of Drosophila pseudodbscura have been studied in this laboratory (see, for example, Dobzhansky, 1935; Dobzhansky and Boche, 1933; Sturtevant and Dobzhansky, 1936). Tests have been carried out on a series of these, to determine the fre- quency with which autosomal lethals occur. Most of the work has been with the third chromosome, though a few tests on the second will be described below. Wild males, or a single son from each of a series of wild females (i.e., females already fertilized when trapped), are mated individually to females carrying the recessive gene for orange eyes. From each such mating a single son is mated to females of the multiple stock orange Scute (dominant, bristle reduction) purple (recessive, eye color). All the Sc, not-or, offspring of this mating carry one particular 542 AUTOSOMAL LETHALS IN DROSOPHILA 543 third chromosome from the wild specimen; when they are mated ale a OSG pre Or Sc pr for one of the two third chromosomes of the original wild male. If this chromosome carries a lethal, the not-Sc class is absent. The difficulty with the method is that, as just outlined, it neglects the possibility of crossing over. This possibility exists only in the females used in the final test generation, since everywhere else the tested chromosome is kept in males, in which crossing over is absent (or at most negligible in frequency). Even in the females in question, it happens that in the majority of cases crossing over is not a serious source of error. As reported by Sturtevant and Dobzhansky (1936), there is a wide variety of sequences in this chromosome. The or Sc pr stock has the Standard sequence; and females heterozygous for Standard and any other sequence give few crossovers. The presence of the three mutant genes in the experiment makes it possible to distinguish between Standard, on the one hand, and all the remaining sequences, on the other hand, in the tested third chromosome. Incase the tested chromosome does not have the Standard sequence (and the majority of them do not have it) the presence of a lethal is at once apparent from the absence or extreme rarity of wild-type flies in the test generation. Most of the sequences do give occasional crossovers in one region or another when tested against Standard; for this reason special tests are required to distinguish between semi-lethals and lethals that cross over with Sc. These have not, in general, been carried out in the present experiments; but it is clear that very few semi-lethals are present, since in most cases the wild-type class is either wholly absent or approximately half as numerous as the heterozygous Scute class. This results from the fact that few of the lethals happened to lie in the sections that give crossovers with Scute. If the test chromosome has the Standard sequence, lethals can still be detected, though it is necessary to make more extensive counts to be certain no lethal is present. This is due to the fact that Sc lies near the middle of the chromosome (about 30 units from the left end, 40 from the right), so that any lethal shows appreciable linkage with it. Here, however, the distinction between lethals remote from Sc, and low-viability genes close to it, requires special tests. The test, applied in most cases, has been to use a chromosome carrying the dominant Emarginate (eye shape) and the Arrowhead sequence. Em J Em | ots an outlined above are obtained. together ( ) the not-Sc offspring are homozygous From the mating , results comparable to the Sc test 544 A. H. STURTEVANT In the case of the second chromosome, the test procedure is similar. Wild males or single sons of wild females are mated to glass (recessive, eye structure) females, and a single male from each such mating is mated to Bare (dominant, bristle form) glass females. The resulting Bare offspring (= ) are mated together, and counts are made of ae their offspring. This test is less efficient than that used for the third chromosome, since there is more crossing over. No inversions are present in the material tested, and Bare is about 60 units from the left end of the chromosome, 40 from the right. Lethals even at this distance do cause a disturbance of the ratios; and all suspected lethals were verified by making a series of matings of Ba X Ba from sibs. In every case these tests agreed in showing a lethal to be present. TABLE I Frequency of lethals Lethals Chromosomes tested Number Percentage Chromosome II LOS OP ae cr nee ane een Ce Se 21 4 19.0 Oldkstocks ee Rees een sae 16 3 18.8 Chromosome III OSS ise Sit te RR Rance oho 25. 67 13 19.4 LOS ORE Se acne reach aks 120 23 19.2 Oldtstocks skeen ee 29 7 24.2 However, there is no test for distinguishing lethals and semi-lethals; and the technique is more laborious than that used for the third chromosome. For these reasons relatively few tests were made on the second chromosome. The mutant strains available did not permit reasonably efficient tests for lethals in chromosomes IV or V, and none was attempted. In the case of the X chromosome no detailed tests were made, but obser- vations on the sex ratio from wild females and their daughters indicated that no lethals were present in any of the wild strains studied. It should be pointed out that what is being studied here is lethal- bearing chromosomes, rather than lethals. That is to say, if a chromosome carries more than one lethal the technique used does not make the fact evident. With frequencies of lethals as high as those AUTOSOMAL LETHALS IN DROSOPHILA 545 found it seems likely that such duplications of lethals in a single chromosome do in fact occur. The net result will be that the fre- quencies recorded are too low. Further complications arise in con- nection with the tests for identity of lethals, but since few cases of identity were found this does not seem a serious source of error. The results obtained are shown in summary form in Table I. The material studied in 1935 came largely from the Rocky Moun- tains and from the Mexican plateau; that in 1936 largely from southern California. The ‘‘old stocks’’ were, many of them, from the Pacific Northwest. These latter had been kept in the laboratory for at least a year before the lethal tests were begun. The striking thing about the table is that, in spite of these differences, there was little variation in the frequencies found for a given chromosome or even as between the second and third chromosomes. ‘Table II, giving the data for the WABrE UT Lethals Chromosomes tested Number Percentage Chromosome III San Gabriel Canyon, Azusa, Calif........ Diff 8 29.6 airilicrnby @alliteewee eters et tus se e eieeee os 21 3 14.3 Sam Goreonio Nite Calif... ose. eel. oo 7 3 17.6 Samia, Cavs liscl, Caltissosceubeapoodsoue 16 3 18.8 most-studied individual localities (included in Table I) shows a similar relative constancy of the frequency of lethals. All these results are from Race A; 5 Race B third chromosomes (all from the Olympic Peninsula of Washington) were tested, and one lethal was found, giving a frequency of 20 per cent. As reported by Sturtevant and Dobzhansky (1936), the third chromosomes of this species exist in a wide variety of sequences. This phenomenon does not appear to be related to the occurrence of lethals. Lethals have been found in the Standard, Arrowhead, Pikes Peak, Santa Cruz, Chiricahua, Cuernavaca, and Klamath sequences. Their frequency seems to be essentially the same in regions such as southern California or southern Mexico, where several different sequences exist in fairly large numbers within each population, and in the area near the common corner of Utah, Colorado, Arizona, and New Mexico, where only the Arrowhead sequence is found. Fifteen third chromo- 546 ated > URE VAN somes tested from this pure Arrowhead area gave 3 lethals—exactly 20 per cent. The data on Chromosome II also indicate that diversity of sequences is not important, since all the strains from the regions con- cerned in these tests had the Standard Race A sequence in that chromosome. . TESTS FOR ALLELOMORPHISM If two lethals, different in origin, are kept in stocks of the form lethal or Sc pr If two distinct lethals are concerned, wild-type offspring will appear in numbers approximately equal to half the size of the heterozygous Scute class; if the two lethals are identical, no wild-type offspring will be produced except as a result of crossing over in the female used, and this will be infrequent if the lethal-bearing chromosome does not have the Standard sequence. Numerous tests of this kind have been carried out. Fifteen of the 1935 and ‘‘old stock” lethals were tested in all possible combinations; the result showed 14 lethals to be present. One from Metaline Falls, Washington (1934) was identical with one from Florence, Texas (1935). Thirteen of the 1936 lethals were tested against each other, and were found to include 12 different ones. One of three tested from Julian was found to be identical with one of eight from San Gabriel Canyon. The Metaline 1934 lethal was tested against the 12 different 1936 lethals, and was found to be allelomorphic to one from Barton Flats, on the slopes of Mt. San Gorgonio, Calif. The San Gabriel-Julian duplicate was different from all the seven surviving members of the 1935 series with which it was tested. These tests, with a few miscellaneous others, total 225 crosses between separately found lethals, with 3 cases of allelomorphism. That is, an , it is easy to test their identity by crossing the two strains. average of about = = 1.3 per cent of the lethals found may be expected to be allelomorphic to any given one selected at random. Or, since the total frequency of lethals in this chromosome in wild stocks is just under 20 per cent, the average frequency of a given lethal may be estimated at about one-fourth of 1 per cent. It may be observed that no cases of recurrence have been found within a locality. That is to say, the same lethal has not been re- covered from any two specimens collected in the same place. Asshown above, one lethal was found both at Julian and at San Gabriel Canyon, in southern California, in 1936. While the two localities are only about AUTOSOMAL LETHALS IN DROSOPHILA 547 100 miles apart, the populations of their respective mountain ranges (San Jacinto and Sierra Madre) appear to be distinct in that the former contains the Santa Cruz sequence while in the latter Tree Line occurs (along with Standard, Arrowhead, and Chiricahua, common to both regions—see Dobzhansky and Sturtevant MS in press—Genetics). The other recurrent lethal was found in stocks from Metaline Falls, Washington (1934), Florence, Texas (1935), and Mt. San Gorgonio, Calif. (1936). It seems clear in the latter case, and probable in the former, that we are dealing with recurrent mutations rather than with the persistence of a lethal gene in heterozygous form. RATE OF ORIGIN OF NEw LETHALS A preliminary lethal-accumulation experiment has been carried out. The technique used is not essentially new, and the results are not yet extensive. It seems sufficient at present, therefore, to record that 2 new lethals occurred in a total of 120 chromosome-generations, giving a frequency of 1.7 per 100—with a large probable error. PROPERTIES OF NON-LETHAL CHROMOSOMES The frequency of lethals found in the tested wild stocks is un- expectedly high. A possible explanation of this result seemed to be that the species is in a condition approximating that of balanced lethals. That is to say, that most or all its third chromosomes are so constituted that flies homozygous for any one of them cannot compete with heterozygotes. Such an assumption amounts to supposing that the phenomenon of heterosis is well developed within a single pair of chromosomes. Under these conditions it might result that the rate of elimination of a lethal would be greatly decreased; since lethals are eliminated only when homozygous, and the assumption is that homozygotes play little part in the perpetuation of the species, so that their properties are unimportant. The simplest way to test this hypothesis is to study the properties of individuals homozygous for a series of non-lethal chromosomes, from the same wild populations as those that contained lethals, in com- parison with individuals heterozygous for two such chromosomes. There is another question that can conveniently be investigated at the same time—are the lethals completely recessive? This latter question is of importance in any calculation of the probable rate of elimination of lethals from wild populations. Only one characteristic of the flies concerned has been investigated —namely their viability. It is clear that fertility, length of life, re- 548 AEE Sa Ewa Ve sistance to unfavorable conditions, and other properties, would all be of importance in determining the frequencies found in natural popula- tions; but the labor involved in studying a series of combinations for such a variety of characteristics has seemed prohibitive. It also seems probable, a priori, that there will be, in general, a rough correlation between all these properties, such that data on one of them will give a satisfactory picture of the general situation. The technique used was the same as that used in the detection and testing for identity of lethals—in fact many of the cultures served both purposes. For the study of lethals partial counts were adequate, and were used in many cases. However, if complete counts are made on a culture there results a measure of the viability of the wild-type anes or Sc pr’ the absence of crossing over, three types are expected in the ratio 1:2:1. The homozygous or Sc pr is usually present in rather small numbers, and has been neglected in making comparisons, since the numbers seem to be very much influenced by minor environmental differences and are often so low as to be inconveniently sensitive to the error of random sampling. The index of viability actually used is the number of wild-type + the number of heterozygous Scute X 100—i.e., the percentage that the wild type is.of the Scute. Evidently, with equal viability of all classes the index will be 50. Several objections may be raised to this method. ‘The most serious is that each of the classes with which we are concerned is compared, chromosome concerned. From the mating together of in not with a common standard type, but with a class that or Sc pr varies in constitution from one test to another. In the case of a test of a homozygote the comparison concerns a heterozygote of that same chromosome; in the test of a heterozygote the cee class is made up Se of heterozygotes for each of the two + chromosomes, in approximately equal numbers. In a few cases tests were repeated, and the result shows a definite, though far from complete, correlation between suc- cessive tests of a given combination. ‘The use of this technique can, therefore, not be considered as giving more than an indication of the situation, and accordingly it does not seem desirable to present the re- sults obtained in detail. In general, the homozygotes for non-lethal third chromosomes from wild stocks are viable and fertile in both sexes, with the result that homozygous stocks are easily established. The test outlined above AUTOSOMAL LETHALS IN DROSOPHILA 549 indicates that some homozygotes are definitely below par, others apparently at no disadvantage. The average viability index of 21 tested chromosomes was 45.91, the minimum 19.4 (weighted average of three tests—16.3, 21.6, 33.3). The tests of heterozygotes for two-third chromosomes from wild strains were made chiefly with lethal-bearing chromosomes. Twenty- five different combinations of lethal X lethal gave an average index of 53.0—.e., more viable than the separate lethals heterozygous for or Sc pr. The lowest index among the 25 was 38.3. The most probable conclusions are that (1) lethals are wholly recessive; (2) homozygotes for non-lethal chromosomes are slightly less viable, on the average, than heterozygotes; (3) the viability of homozygotes is quite variable, some being definitely low, others apparently as high as the heterozygotes. COMPARISON WITH RESULTS FROM D. MELANOGASTER The frequency of lethals in wild populations of D. melanogaster has been studied by several authors. The most extensive work is that of Dubinin and co-workers (1934, 1936) on material from the Caucasus. These studies were carried out on a large scale. Altogether 4819 second chromosomes are recorded, with 470 (= 9.8 per cent) lethals. The frequencies range from 0 (in 92 tested, from Delizhan) to 16.1 per cent (161 tested, from Ordzhonikidze). From one locality, Gelendzhik, tests were made in three successive years, giving 7.98 per cent (877 tested), 12.86 per cent (616 tested), and 8.78 per cent (797 tested). Tests were also carried out for identity, but only within regions. The average value here was 2.2 per cent—i.e., an average of 2.2 per cent of the lethals found in one year at one locality may be expected to be allelomorphic to any one lethal found in that year at that locality. The average frequency for a given single lethal is 0.22 per cent (.022 X .098). There are no tests for identity of lethals from different localities, but those found at Gelendzhik in successive years were so tested, with the result that ‘‘among the 33 lethals of 1933 and the 55 lethals of 1934, nine lethals were common to both years.”” The conclusion is drawn that this represents survival of these lethals. The results reported in the present paper throw some doubt on this conclusion, for a comparable frequency of identity was found in pseudoobscura for all years and all localities. One may surmise that, in melanogaster also, any two series of lethals would show a fairly large proportion of common members—i.e., that recurrent mutations are relatively frequent. 550 A. H: STURTEVANT The absolute frequency of lethals found in the Caucasian melano- gaster populations is only about half that in the American pseudo- obscura, though the chromosome concerned (II) is roughly twice as long as the III of pseudodbscura and includes the same material plus that of the pseudodbscura IV (Donald, 1936; Sturtevant and Tan, 1937). The preliminary experiment described above suggests a higher lethal mutation rate in pseudodbscura; but until more data are available on this point it is scarcely possible to decide on the relative rates of elimination of lethals in the two species. One difficulty in evaluating the data of Dubinin and his co-workers may be pointed out. The results are recorded by localities and years; but we are nowhere given more specific data. How large an area is included in a single “‘locality’’? Over how long a period were speci- mens taken in one year? What kinds of places were collected—woods, grocery stores, garbage dumps, fruit orchards? These questions are of importance in judging the size of the populations sampled and the probable degree of relationship of the tested individuals. SUMMARY 1. Approximately ‘20 per cent of the third chromosomes found in wild populations carry lethals. 2. Less extensive data indicate a similar frequency for the second chromosome. 3. On the average, about 1.3 per cent of the lethals found may be expected to be allelomorphic to any given one. 4. The average frequency for any one lethal is about one-fourth of one per cent of the chromosomes of wild strains. 5. The lethals, so far as studied, are completely recessive. 6. Study of flies homozygous for non-lethal third chromosomes shows a considerable variation in their viability. Some are at a definite disadvantage, others apparently not. On the average they - are not quite as viable as flies carrying two different third chromosomes. LITERATURE CITED DospzHAnsky, TH., 1935. The Y chromosome of Drosophila pseudodbscura. Genetics, 20: 366. _DoszHansky, TH., AND R. D. Bocue, 1933. Intersterile races of Drosophila pseudodbscura Frol. Bzol. Zentralbl., 53: 314. Dona.p, H. P., 1936. On the genetical constitution of Drosophila pseudodbscura, Race A. Jour. Genet., 33: 103. Dusinin, N. T., ET AL., 1934. Experimental study of the ecogenotypes of Drosophila melanogaster. Bzol. Zhurn., 3: 166. AUTOSOMAL LETHALS IN DROSOPHILA Sol Dusinin, N. T., et al., 1936. Genetic constitution and gene-dynamics of wild popu- lations of Drosophila melanogaster. Bzol. Zhurn., 5: 939. STURTEVANT, A. H., AND TH. DoOBZHANSKY, 1936. Inversions in the third chromo- some of wild races of Drosophila pseudodébscura, and their use in the study of the history of the species. Proc. Nat. Acad. Sct., 22: 448. StTuRTEVANT, A. H., anp C. C. Tan, 1937. The comparative genetics of Drosophila pseudodbscura and D. melanogaster. Jour. Genet., 34: 415. THE EFFECT OF SALINITY UPON THE GROWTH OF EGGS OPSBUCUS FURCATRUS?Z D. M. WHITAKER AND C. W. CLANCY (From the School of Biological Sciences, Stanford University) INTRODUCTION The effect of salinity upon fertilization and early development in 3 species of Fucus (F. serratus, F. vesiculosus, and F. spiralis) was studied by Kniep (1907) on the Norwegian coast. He investigated especially the tolerance for dilution of the sea water and found it to differ con- siderably in the different species. Kniep was able to correlate the tolerance with the distribution of the species (and, by inference, of additional species of Fucus and other related marine alge as well) especially in the Baltic Sea and in near-by regions in which brackish water communicates with the ocean, thus establishing gradients of salinity. The present investigation is an attempt to measure the effect of increased and decreased salinity upon the percentage germination and the growth rate of the fertilized eggs of Fucus furcatus f. luxurians throughout the range of salinity in which germination and growth take place, and to do so under strictly controlled conditions. The eggs were transferred abruptly into sea water of altered salinity, and therefore, any effect which a prolonged gradual adaptation might have on the tolerance is eliminated from the measurements. ‘The data are useful in conjunction with other experimental investigations of the development of this egg in which the salinity of the medium varies. MATERIAL AND METHOD Fruiting tips of this hermaphroditic Fucus were collected at Moss Beach, California in May, June and July, 1937. The material was cared for in the manner described in an earlier paper (Whitaker, 1936). Sea water was collected at the same place and was filtered twice upon arriving at the laboratory. In each experiment, eggs from the same batch were reared in this sea water and also in samples of the same sea water with artificially altered salinity. Salinity was decreased by adding triple glass-distilled water. It was increased by boiling under 1 This work has been supported in part by funds granted by the Rockefeller Foundation. 552 SALINITY AND GROWTH OF FUCUS EGGS 5S) vacuum so that the temperature did not rise enough to cause pre- cipitation of salt. By this method the volume of sea water was reduced to half without precipitation. The measure of salinity which has been used is the total salt content per unit volume expressed as a percentage of the total salt content per unit volume of normal sea water. In all cases hydrometer readings were made which confirmed the salinity as originally determined volumetrically. The specific gravity of the normal sea water used was 1.027. The pH of the samples of normal sea water, as determined with a glass electrode, ranged from 7.9 to 8.2. The pH tended to rise slightly when the salinity was artificially increased, and to fall when it was decreased. Such shifts were usually not great, being of the order of a few tenths of a pH unit or less in most experiments. In the most extreme cases the shift amounted to 0.5 units. The maximum pH range of the media in the experiments upon which Fig. 1 is based was 7.7 to 8.5. In some of the experiments, all of which gave essentially similar results, the range was considerably less. This pH range is much narrower than the limits of normal development for this species of Fucus (Whitaker, 1937), and it is small enough so that it appears safe to conclude that it is a minor factor in the present results. Since Fucus furcatus is hermaphroditic, and sheds male and female capsules together at the same time, it is not feasible to inseminate a population of eggs at any precise moment. The male capsules dissolve in the sea water first, however, so that fertilization takes place when the egg capsules dissolve and liberate the individual eggs into the sperm bearing sea water. By selecting eggs only from capsules which break down during a limited period, it is possible, in effect, to confine fertiliza- tion to this period. In these experiments eggs were used which were fertilized during a 10-minute period. In order, for the present purpose, to rule out the effects of altered salinity upon fertilization and entrance of the sperm, eggs were fertilized in normal sea water in all experiments. Twenty minutes after the end of the fertilization period samples were transferred with a negligible amount of normal sea water into the media of altered salinity. Eggs had thus been fertilized 20-30 minutes when they were removed from normal sea water. They were thinly seeded in covered Petri dishes which were stored in a moist chamber in a dark, humid, constant temperature room at 15°C. Fertilization and all subsequent development took place in this constant temperature room. The eggs were observed only with red light until the end of the experiment. 554 D. M. WHITAKER AND C. W. CLANCY RESULTS The fertilized Fucus eggs develop in a relatively great range of salinity. As the limits of tolerance are approached, all eggs in a population do not have the same end point, and the results are repre- sented graphically by the blocks in Fig. 1 to show the percent of the eggs in the populations which form rhizoids. Each block in the figure represents the average of the results of from 4 to 10 experiments, each involving counts on 300 or 400 eggs. The counts were made two days after fertilization. The blocks in Fig. 1 show that very high percentages of the eggs 100 70 ¢o 50 40 30 20 CURVE: LENGTH RATIOS IN PER CENT BLOCKS: PER CENT OF EGGS WITH RHIZOIDS [Sa 20 30 40 50 60 70 80 90 100 110 120 130 140 150 IGO 170 180 190 SALINITY OF SEA WATER IN PER CENT Fic. 1. Blocks: The percentages of eggs which formed rhizoids in normal sea water, in diluted sea water and in concentrated sea water, as observed after two days. Curve: The average over all length of embryos at 92-96 hours, reared in sea water of the salinity indicated, expressed as a percentage of the average length of control embryos reared in normal sea water. In all cases the eggs were fertilized and reared in the dark at 15°C. The specific gravity of the normal (100 per cent) sea water was 1.027 (see text). develop in sea water ranging in salinity from 50 or 60 to 150 per cent of that of normal sea water. Beyond these limits the percentage drops off rapidly, although some eggs develop rhizoids when the salinity is as extreme as 30 per cent or 180 per cent. Beyond these limits no rhizoids are formed. In the more dilute sea water (10 per cent and 20 per cent) eggs burst and cytolyze at once, and many of the eggs which fail to form rhizoids in 30 per cent or 40 per cent sea water also cytolyze. In the concentrated sea water, on the other hand, most of the eggs which fail to form rhizoids do not cytolyze, at least in the course of several days. Eggs which have remained for two days in sea water of twice the normal salinity form rhizoids after being returned SALINITY AND GROWTH OF FUCUS EGGS So) to normal sea water. The suppression at high salinity thus bears some resemblance to anesthesia. The Fucus egg is spherical until a bulge forms on one side, about 16 hours after fertilization (Whitaker, 1936). This bulge extends in filamentous fashion by elongation and cell divisions until its length is many times the diameter of an egg. Ultimately the rhizoid of the new plant forms from this filamentous structure, while the remainder of the embryo gives rise to the thallus. The rate at which early development takes place, as indicated by the rate of extension of the rhizoid filament, is a function of the salinity of the medium. In both extremes of the range of salinity, in which many or most of the eggs form no rhizoids at all, the rate of extension is greatly retarded in those eggs which do form rhizoids. Within the ranges of salinity (50 or 60 to 150 per cent, see blocks, Fig. 1) in which practically all of the eggs form essentially normal rhizoids, the salinity affects the rate of extension as shown by the curve in Fig. 1. The curve in Fig. 1 shows the average full length of the embryos reared in media of the salinities indicated, expressed as a percentage of the average full length of the control embryos from the same batches of eggs reared in normal sea water. The lengths were measured with an occular micrometer 92-96 hours after fertilization. Each point was obtained by averaging the results of 4 to 10 experiments, each involving counts of 50 or 100 eggs, except that the point at 90 per cent salinity is based on only 3 experiments. The results of the separate experiments are very similar. It is seen from the curve in Fig. 1 that the embryos grow as rapidly in 90 per cent as in 100 per cent sea water. The average absolute length of the embryos in these optimum salinities was 303 microns at 92-96 hours. Since the embryos were growing in the dark in the absence of photosynthesis, no nutrients were available to the eggs except those stored in the unfertilized eggs. It is therefore doubtful if much new protoplasm was synthesized, although internal conversion of stored foodstuffs may have supported some protoplasmic synthesis. The growth or extension observed is presumably largely a develop- - mental elaboration, i.e., an extreme change of form. SUMMARY 1. Fertilized eggs of Fucus furcatus f. luxurians have been reared in the dark at 15° C. in diluted and concentrated sea water. 2. When the salinity of the medium is between 60 per cent and 150 per cent that of normal sea water (sp. gr. 1.027), practically all of the eggs in a population form rhizoids and develop. The rate of elongation 556 D. M. WHITAKER AND C. W. CLANCY of the embryos, as measured at 4 days, is the same when the salinity is 90 per cent or 100 per cent. When the salinity is greater or less, the growth rate is retarded as shown in the curve in Fig. 1. 3. As the salinity is reduced below 60 per cent or is increased above 150 per cent, the percentage of eggs which form rhizoids declines rapidly as shown in the blocks in Fig. 1. In 10 per cent and 20 per cent sea water the eggs burst and cytolyze. In concentrated sea water which inhibits development the eggs do not cytolyze and the develop- mental inhibition may be reversible. BIBLIOGRAPHY KwniEp, Hans, 1907. Beitrage zur Keimungs-Physiologie und -Biologie von Fucus. Jahrb. f. wiss. Bot., 44: 635. WHITAKER, D. M., 1936. The effect of white light upon the rate of development of the rhizoid protuberance and the first cell division in Fucus furcatus. Biol. Bull., 70: 100. WHITAKER, D. M., 1937. The effect of hydrogen ion concentration upon the induc- tion of polarity in Fucus eggs. I. Jour. Gen. Physiol., 20: 491. STIMULATION AND NUCLEAR BREAKDOWN IN THE NER EIS PEGG L. V. HEILBRUNN AND KARL M. WILBUR (From the Zoological Laboratory, University of Pennsylvania) When a protoplasmic system is exposed to a so-called stimulating agent, a series of changes occurs which results eventually in activity of the protoplasm. Various opinions have been expressed as to the physico-chemical basis of stimulation. In our laboratory for several years we have been interested in proving that stimulating agents all cause a release of calcium from a calcium proteinate gel in the cell cortex. This calcium is then supposed to initiate a protoplasmic clotting similar in some ways to blood clotting. The evidence on which this theory is based has been summarized in a recent book (Heilbrunn, 1937, Chapter 37), and the details of the theory are also given there. Although in previous studies various types of animal and plant material have been used, and the theory has thus been applied to cells very different morphologically, as yet no attempt has been made to determine what effect stimulation may have on the cell nucleus. Certainly in a dividing cell, the nucleus plays an all-important réle. It is of interest, therefore, to inquire what effect stimulation may have in those cases in which the primary result of stimulation is cell division. As is well known, the egg cells of many marine invertebrates are sensitive to a wide variety of stimulating agents, and following stimulation, division is usually initiated. Typically, during mitosis the nuclear membrane breaks down. The mechanism of this breakdown has scarcely been investigated experimentally, although an understanding of the causes of such muclear breakdown would be an important link in any complete theory of cell division. In the case of many marine invertebrates, the egg is released into the sea water in an immature condition, and it contains a large nucleus or germinal vesicle. Sometimes contact with sea water is in itself sufficient to cause a breakdown of the germinal vesicle and an initiation of the maturation or polar body divisions. This is true of the starfish egg. In other instances, the egg remains immature and retains its large germinal vesicle until insemination. Then the 1 This work has been aided by a grant from the Radiation Committee of the National Research Council. 557 558 L. V. HEILBRUNN AND KARL M. WILBUR entrance of sperm into the egg, or its contact with the egg cortex, provides the stimulus for nuclear breakdown and maturation divisions. The egg of the annelid worm Nereis limbata has a large germinal vesicle when it is shed into sea water, and this is broken down after insemina- tion. The Nereis egg has been widely used both in studies of normal fertilization (Lillie, 1911), as well as in studies of artificial partheno- genesis (Just, 1915; Heilbrunn, 1925). It is readily obtainable at Woods Hole, and is an easy object to study. Following activation, the breakdown of the germinal vesicle can readily be detected in the living egg, especially if the egg is somewhat compressed between slide and coverglass. Some years ago, Just showed that the Nereis egg could be artificially activated by heat (Just, 1915; see also Heilbrunn, 1925). Later, Just (1933) found that ultra-violet radiation could also cause activation. In general, it is believed by radiologists that Roentgen rays can produce the same effect as ultra-violet radiation. Accordingly an attempt was made to discover if activation could be produced by Roentgen rays. It was found that in order to obtain results, extremely high dosage was necessary. In three experiments in which eggs were exposed to 7600 r units per minute for 75 minutes, the percentage of eggs activated was found to be 74, 52, and 5. Three other experiments gave negative results. The activation was not the result of heat, for the dishes containing the eggs were packed in ice to keep them from becoming overheated. The Nereis egg may be activated by chemical agents as well as by physical means. It was found that isotonic solutions of sodium chloride cause a high percentage of germinal vesicle breakdown. Similar effects may also be obtained with isotonic solutions of potassium chloride. Doubtless various other chemical agents are also effective, but no attempt was made to list all the effective reagents. However, it may be noted that a dilute ether solution (3 per cent in sea water) is somewhat effective in causing germinal vesicle breakdown. Other fat solvents doubtless act in similar fashion. The literature on artificial parthenogenesis contains many papers in which activation of marine eggs has been found to follow various types of treatment. Our concern has not been to prepare an ex- haustive list of the effective agents for the Nereis egg. We have attempted to discover the nature of the effect of several physical and chemical activating agents, and we have assumed that if we can throw light on the true mechanism of activation in two or three cases, such information may perhaps be applied to all types of activation. In our experiments we have studied the action of ultra-violet STIMULATION AND NUCLEAR BREAKDOWN 909 radiation and of isotonic salt solutions, and we have tried to discover whether the calcium release theory of stimulation will hold for these cases. Our experimental procedure was very simple. First we de- termined the percentage of germinal vesicle breakdown following exposure to the stimulating agent. Then we immersed eggs in'*citrate solutions, to discover if previous exposure to citrate would prevent activation. As is well known to students of blood clotting, citrate solutions tend to prevent the action of calcium. Opinions as to the mechanism of the effect differ. (Compare Hastings, McLean, Eichel- berger, Hall and DaCosta, 1934.) Some unpublished work of D. Mazia indicates that when sea-urchin eggs are immersed in citrate TABLE I Nuclear Breakdown following Irradiation Percentage breakdown Immersion Percentage breakdown after 60 seconds . time in after 60 seconds irradiation in 0.35 M Na irradiation in sea water citrate Na citrate minutes OOPS hits sextet. 4 0 OO eer iene ec 4 il 98... 4 7 MOORE O eee ice ei ane 4 2 NOOBS ees set hc aha y 4 6 AND Gaetan eye aaa epee 4 5 TOO) sed ea tea rt rel rts 1 4 D, NOOR eax u ake ccals, 3 4 37 Oe ear peak ai sie NS 4 11 OE Riper nee veritas ea: 4 5 HOO) eiamaanenearn ts in ee enh 5 0 NO Oper arta tot ctse 5 5 T1010) ea pees Hie een eee 5 0 OU ilapte eh ee ty ayn cch Se 5 19 CO [os EUR ut EE 29 0 solutions, a large percentage of the calcium normally present in the cell cortex is removed. If citrate solutions are able to remove calcium from the cell cortex, or if they prevent release of calcium to the cell interior by some other mechanism, then citrated eggs should be incapable of activation. This was found to be the case. Two types of experiments were tried. In the first place, eggs were exposed to ultra-violet radiation, both in sea water and in sodium citrate solutions. Ultra-violet radiation was obtained from a Uviarc Laboratory Outfit. This type of lamp is well known and its charac- teristics have been described in a paper by Buttolph (1930). The lamp was operated at 160 volts and 5 amperes. In our work we were not interested in the energy output of the lamp nor in separating 560 L. V. HEILBRUNN AND KARL M. WILBUR different wave-lengths of radiation. A few preliminary tests soon showed that when Nereis eggs were placed in a small quantity of sea water and exposed to the mercury arc at a distance of 26 cm. for 60 seconds, all of the eggs, or at least more than 90 per cent of them, showed activation. Shorter exposures (e.g. 30 seconds) were not quite so successful. On the other hand, following exposures of 3 or 4 minutes, slightly lower percentages of activation were obtained. It is thus clear that under the conditions of our experimentation, a 60-second exposure to the Uviarc lamp was an effective stimulus, and we have reason to believe that other Uviarc lamps act approximately as ours did, unless of course, the lamps have suffered very serious deteriora- TABLE II Nuclear Breakdown in Isotonic NaCl Immersion time Percentage breakdown Percentage breakdown in 0.35 M Na in NaCl following in NaCl citrate citrate treatment minutes OO pte een scribe era 4 4 D Oe AD aeebane hse ea! 4 4 OOF tac Pate cei nicat cac ane 4 7 O18) od alas UR a i a eles et 4 24 SIO Chala ac ne Me RL 5 0 SO agement: My aa cc tligesn e VaN 6 2 Coates cee ee aa Seen eee hee 6 0 GOREN ee ee crane wae, 6 0 SO RP eMac sh iy eta a 6 0 O10) wished ices Ahn een ee 6 0 SIO ONS eke ae es ae ais 6 1 OO ME Preteen cea ces cere e 6 0 OS ee eee cece pee rs 6 0 QAP Means ae amete sk stag Sh 6 1 OOO ett Feet Cy ac A ant a 6 0 WON O ea ies oats eek oka bus rere ae 8 2 tion. In any case, it is a simple matter to determine the exposure necessary to obtain activation, for there is a wide range over which this occurs. Table I shows the percentages of nuclear breakdown following irradiation in sea water, as compared with the percentages obtained following irradiation in citrate solutions. The difference is striking. In sea water typically, practically all the eggs respond by nuclear breakdown. On the other hand in the citrate solution, there is almost no response. Similarly, when eggs are exposed to isotonic sodium or potassium chloride solutions (0.53 molar), there is a high percentage of response, but if the eggs are first immersed in 0.35 molar sodium citrate and then STIMULATION AND NUCLEAR BREAKDOWN 561 exposed to the sodium or potassium chloride solutions, the eggs fail to respond. This is shown in Tables II and III. From these tables it will be noted that although a 4-minute immersion in the citrate solution is sufficient to prevent most of the eggs from showing nuclear breakdown, occasionally there is some response following such im- mersion. Thus in the fourth experiment of Table II, 24 per cent of the eggs were activated after 4 minutes of citrate treatment. If the - eggs are allowed to remain in the citrate solution for 6 minutes before ‘being transferred to the sodium chloride or the potassium chloride solution, there is almost no nuclear breakdown. Apparently, the TABLE III Nuclear Breakdown in Isotonic KCl Percentage Immersion time Percentage breakdown breakdown in 0.35 M Na in KCl following in KCl citrate citrate treatment minutes AWE yeas Severe eck 18 CETL IaneE cheat Geren me 4 10 II OKO eee tener vere cee 4 Some eggs broken up. Count impossible. OO er Papers srs moines 4 Some eggs broken up. Count impossible. Dis cee aed tahaea 5 0 DSi ey rede Mt nee i a 6 1 DISH IN egy nM ge nes 6 0 IAs eet ate eaves eile | Seat 6 0 UE Meet) Sa trer tes tagral ts 6 0-2 BAER Gen casieyaiet eae ier 6 0 ILS Same to cere are agrees 6 0 218) bela ls. a aaa MeN ue 6 0 OOM i ceo cen een 6 0 OO Ree rcamicume- chee 6 0 O10) hE aire nearer 6 0 MOORE Sree -ctein cate es 8 4 citrate solution must act 5 or 6 minutes if it is to be completely effective in preventing response to stimulation. In these studies of activation, our attention was focussed on the nuclear breakdown, and no attempt was made to study the cortical changes which occurred in the eggs. These changes are interesting, but they were not considered in the present study. It might well be urged that the effect of the citrate in preventing the stimulating effect of ultra-violet radiation and of sodium and potassium solutions was essentially due to injury. Thus, if eggs are fixed in formalin or some other poison, they presumably would not respond to any type of stimulation. This objection is very easily met. If citrated eggs are returned to sea water, they immediately become 562 L. V. HEILBRUNN AND KARL M. WILBUR sensitive to radiation again. This is clearly shown in Table IV. Eggs exposed to citrate for 6 minutes and then returned to sea water show nearly 100 per cent of germinal vesicle breakdown on irradiation. In this connection, it should be pointed out that even without irradia- tion, a certain percentage of eggs may show germinal vesicle breakdown after transfer from the citrate solution to sea water. This is an interesting fact and will be considered again later. As a whole, our results show that both physical and chemical stimulation of the Nereis egg, as evidenced by the breakdown of the germinal vesicle, are ineffective if the eggs are first treated with citrate. There is thus support for the theory that stimulation to be effective must involve a calcium release from the cell cortex. It is rather easy to understand why ultra-violet radiation might cause a breakdown of a calcium proteinate gel with consequent release of calcium, for this TABLE IV Nuclear Breakdown resulting from Irradiation in Sea Water following Citrate Treatment Immersion time Immersion time Percentage breakdown in 0.35 M Na in sea water following citrate before irradiation irradiation . minutes minutes GRU etre shensieeb ae are 4 92-95 (C203 6 is oir Coe teeoe 4 98 POE Eins ee 8 4 98 (Ore ot ner Me a aaa 10-1034 99 (Deeg Gee aan 10-103 100 (55 tail ne ARE Re aK 10-104 98 (CEN Sa 4 ee anne ROT 10-1034 100 GO omrnie eipseel con anes 5 97 is consistent with various types of earlier work both on protoplasm and on proteins (for a review of this literature, see Heilbrunn and Mazia, 1936). On the other hand, it is rather paradoxical to assume that immersion of a cell in pure sodium or potassium chloride solution would cause a release of free calcium to the cell interior. However, there is real evidence that this type of phenomenon does actually occur (see Heilbrunn, 1937, Chapter 33). If Nereis eggs are immersed in isotonic calcium chloride solutions (0.29 molar) there is no breakdown of the germinal vesicle. Mag- nesium solutions are also without effect. This apparently means that the calcium ions are unable to penetrate the cell with any rapidity, although this is not the only possible explanation. In the case of the clam Mactra, some unpublished experiments of Miss R. A. Young show that activation (which in this egg also involves germinal vesicle STIMULATION AND NUCLEAR BREAKDOWN 563 breakdown) follows immersion in calcium chloride solutions. (See also earlier work of Dalcq, 1925, 1928; Hérstadius, 1923; Hobson, 1928; Pasteels, 1935.) From these studies with marine eggs we can conclude that for some, calcium chloride solutions are stimulating agents, whereas for others calcium chloride solutions are without any very apparent effect. This is entirely comparable to the situation with respect to smooth muscle. Thus Tate and Clark (1922) showed that calcium caused a contraction of the uterine smooth muscle of the rabbit and cat, but had no effect on similar muscle in the guinea-pig and rat. In the case of Nereis, the eggs become sensitive to calcium chloride solutions and to a lesser extent to sea water, following immersion in sodium citrate solutions. Thus if eggs were immersed in 0.35 molar sodium citrate for 60—65 minutes and then placed in 0.29 molar calcium chloride, 80 to 100 per cent of them showed germinal vesicle breakdown (four experiments gave percentages of 80, 100, 92, 82). It almost appears as though the loss of calcium from the cell cortex makes it possible for calcium to enter the cell more freely. It is possible that there may be a relation between this phenomenon and the calcium paradox as observed in invertebrate hearts (see, for example, Chao, 1934). Our results lead us to conclude that the breakdown of the germinal vesicle in the Nereis egg is the result of a release of calcium from the cortex. Previously Dalcq had suggested that the calcium ion was responsible for the breakdown of the germinal vesicle in the starfish egg, but he offers no clear explanation as to why sodium or potassium should act like calcium, either in the starfish egg or in other eggs. (See Dalcq, 1925, 1928; also Pasteels, 1935.) Our results offer an interpretation of this difficulty. As to the details of the mechanism involved in the disappearance of the nuclear membrane, we are still at a loss. Presumably, calcium ion initiates a series of changes and these changes eventually cause nuclear breakdown. There is obviously more than a mere action of calcium on the nucleus, for when eggs are crushed in sea water, the germinal vesicle does not break down. SUMMARY When the Nereis egg is stimulated either physically by ultra-violet radiation, or chemically by immersion in sodium or potassium chloride solutions, the germinal vesicle breaks down. This response to stimula- tion is prevented if the eggs are previously immersed in sodium citrate solutions. The results are in agreement with the calcium release theory of stimulation. 564 L. V. HEILBRUNN AND KARL M. WILBUR LITERATURE CITED Buttovpy, L. I., 1930. The electrical characteristics of commercial mercury arcs. Rev. Sci. Instruments, 1: 487. Cuao, I., 1934. Paradox phenomena in the cardiac ganglion of Limulus polyphemus. Biol. Bull., 66: 102. Daca, A., 1925. Recherches expérimentales et cytologiques sur la maturation et l’activation de l’oeuf d’Asterias glacialis. Arch. de Biol., 34: 507. Datcg, A., 1928. Leréle du calcium et du potassium dans l’entrée en maturation de l’oeuf de pholade (Barnea candida). Protoplasma, 4: 18. Hastines, A. B., F. C. McLEAn, L. EICHELBERGER, J. L. HALL, AND E. DACosta, 1934. The ionization of calcium, magnesium, and strontium citrates. Jour. Biol. Chem., 107: 351. HEILBRUNN, L. V., 1925. Studies in artificial parthenogenesis. IV. Heat partheno- genesis. Jour. Exper. Zodl., 41: 243. HEILBRUNN, L. V., 1937. An Outline of General Physiology. Philadelphia. HEILBRUNN, L. V., AND D. MaziA, 1936. The action of radiation on living proto- plasm. Duggar’s Biological Effects of Radiation. Vol. 1, p. 625. New York. Hosson, A. D., 1928. The action of isotonic salt solutions on the unfertilised eggs of Thalassema neptuni. Brit. Jour. Exper. Biol., 6: 65. Horstapius, S., 1923. Physiologische Untersuchungen iiber die Eireifung bei Pomatoceros triqueter L. Arch. f. mikr. Anat. u. Entw.-mech., 98: 1. Just, E. E., 1915. Initiation of development in Nereis. Bzol. Bull., 28: 1. Just, E. E., 1933. A cytological study of effects of ultra-violet light on the egg of Nereis limbata. Zeitschr. f. Zellforsch. u. mikr. Anat., 17: 25. Lituig, F. R., 1911. Studies of fertilization in Nereis. I. J. Morph., 22: 361. PASTEELS, J. J., 1935. Recherches sur le déterminisme de |’entrée en maturation de l’oeuf chez divers Invertébrés marins. Arch. de Biol., 46: 229. TATE, G., AND A. J. CLARK, 1922. The action of potassium and calcium upon the isolated uterus. Arch. intern. de pharm. et de thérap., 26: 103. STUDIES OF THE MITOTIC FIGURE. VI. MID-BODIES AND LHEIRV SIGNIFICANCE FOR THE CENTRAL BODY PROBLEM HENRY J. FRY (From the Department of Anatomy, Cornell University Medical College, New York City, and the Marine Biological Laboratory, Woods Hole, Mass.) INTRODUCTION Purpose of the Study Mid-bodies have received less attention than any other component of the achromatic figure. In many animal cells they appear, during the time of division, as deeply staining thickenings on the fixed spindle fibers, in the equatorial plane. They are generally regarded as in some way comparable to the cell-plate of plants—possibly a vestigial homolog (Wilson, 1928, pp. 144 and 159; Sharp, 1934, p. 169). Flemming (1891) first suggested this concept. In a previous study of central bodies in brain cells of Squalus embryos, in which considerable attention was also given to the mid- bodies, I reached the conclusion that in this case they are nothing but accumulations of dye at places where the spindle materials have been pinched together by the division furrow, and have no existence as an individualized cell component. The question was raised: ‘‘ Whether or not mid-bodies generally will eventually be explained in terms of focalization phenomena remains to be seen”’ (1933a, pp. 177-178). This problem is the subject of the present paper. Discussion of the Term “ Focal Body”’ Before presenting the data it is advisable to clarify further the concept focal body or focalization phenomenon which I have discussed previously (1932, pp. 181-182; 1933a, p. 177; and 19330, pp. 233—235).1 These terms are used to describe a structure of the achromatic figure which appears to be produced as a result of the convergence to a common center of either spindle materials, aster materials, or both. If this assumption is correct, there are three possible classes of focal bodies. Living Focal Bodies.—These probably exist in the living cell as minute pools of spindle or aster materials at centers of focalization, and are preserved by fixation. The situation here may be somewhat like 1 Some of these studies, referred to here and later, were carried out with the as- sistance of other investigators, as indicated in the bibliography. 565 566 HENRY J. FRY that which occurs when thin lines of a substance of relatively low viscosity are placed upon a piece of glass in a converging configuration: a pool of material forms where the lines coalesce at the center, the size of the focal body thus produced depending upon the number and thickness of the lines, and the viscosity of the substance. Such a body is non-radial and more or less demarked, but although it looks very different from the surrounding radial area and appears to be an indi- vidualized structure, it is actually only the inner ends of the lines which have fused. Coagulation Artifact Focal Bodies.—According to the hypothesis presented here and in earlier papers, these do not exist in the living cell but are created by the act of coagulation. It is suggested that the process of fixation may effect a breakdown of the converging materials at the focal center and form a body, the structure of which is de- ‘termined largely by the coarseness of the fixed rays and fibers. As I have shown in previous studies, the comparative coarseness of such materials is modified by the technique—the fixative used or the environmental condition prior to fixation. For example, in eggs of Echinarachnius (1928, 19296, and 1929c) and Chetopterus (1932 and 19330) there is a definite relationship between the structure of bodies at centers of focalization and the coarseness of the fibers and rays, as determined by the technique. An analogy to this situation is found in a model similar to the one described above, but employing material of such a composition that the converging lines do not coalesce at the point where they meet. Upon subjecting this model to a given treatment (the equivalent of the fixation procedure), the character of the material is so changed that the lines coalesce at their meeting-point and a focal body is formed. Staining Artifact Focal Bodies.—These are only accumulations of dye at the focal centers of some fixed mitotic figures, such as I demon- strated in brain cells of Squalus embryos (1933a). An analogous situation occurs if a model is made in which a large number of delicate threads are attached to the edge of a circular frame and brought together at the center. When such a model is dyed, the focal area takes more stain than the peripheral region, owing to the aggregation of threads there. If this model is viewed at the proper distance (dupli- cating the effect produced when stained mitotic figures are studied with a microscope) the central region may in some cases look like a sharply demarked body, depending upon the number of threads, the chemical nature of both threads and dye, whether they are thick or thin, their surface rough or smooth, and their paths separate or interwoven. MID-BODIES AND CENTRAL BODIES 567 Assuming that the suggestion is correct that the focalization of spindle and aster materials may result in the production of focal bodies, we still know too little about the structure of living spindles and asters, the effects of coagulation, and the staining process, to explain adequately in physico-chemical terms exactly how the bodies are produced. For these reasons the models just described are not intended to give exact parallels of what occurs when focal bodies are formed in mitotic figures; they only illustrate how focalization phe- nomena may be involved in any configuration of converging substances. Depending upon the type of cell used and the technique employed, such bodies may be minute or large, demarked sharply or vaguely, granular or homogeneous or vesicular in structure, and they may differ in shape. Such variations can often be demonstrated in the same material when different techniques are used. Focal bodies, however, have certain characteristics by which they can be surely differentiated from typical cell components. They exist only at places where spindle and aster materials come to a focus. They arise only as such places of focalization are formed, and disappear when such areas disintegrate; they therefore do not maintain conti- nuity from cell to cell. They change shape as the focal area changes shape, and often become double when the focal area elongates. They may vary widely in structure from one cell cycle to another in the same organism, or they may vary in the same cell cycle in closely related organisms; sometimes, on the same slide, they show differences from cell to cell which are identical as to species, cell cycle, mitotic phase, and the method used. They are usually unstable in structure and easily modified by variations in the technique. In general their morphology is related to the size of the mitotic figure and the coarseness of the fixed focalized materials. In all of these respects focal bodies differ markedly from such cell components as the centriole-blepharoplasts of spermatocytes and Protozoa, the diplosomes of vertebrate cells, and other similar struc- tures, which have undoubted existence in the living cell as typical individualized cell components, and in some cases can be seen in the living condition. They exist regardless of whether areas of focalization are present or absent, and they exhibit great stability of structure despite wide variation in the technique. Hence the suggestion is made that at centers of focalization, structures are formed which are often minute, homogeneous, sharply demarked, and look like typical cell components, but are actually nothing more than the result of the convergence of spindle and aster materials to focal areas, i.e., focal bodies or focal phenomena. 568 HENRY SER Y METHODS The behavior of mid-bodies was studied in eggs of Cerebratulus, Cumingia, Nereis, Chetopterus, and Asterias, during the formation of both polar bodies, during first cleavage, and also in cells of blastule; in Arbacia eggs, during the first three cleavages as well as in blastule; in brain cells of Squalus embryos; and in primary spermatocytes of Romalea. Many fixatives were used and the cells were exposed to various environmental modifications prior to fixation. They were sectioned at a thickness of 5 uw and stained with Heidenhain’s hematoxylin. In the case of eggs mid-bodies were studied only in those which happened to be sectioned in a plane passing through both the long axis of the mitotic figure and the animal pole, since only in such sections are the relations between the structure of mid-body and spindle shown clearly. In each case an adequate number were examined to ascertain the extent of variation occurring in mid-body structure and associated spindle remnant. In order to check the conclusions reached by these original studies I also examined a considerable portion of the pertinent literature— almost two hundred papers—to be sure that no important types of mid-bodies described by other workers should be overlooked. In most of these invéstigations the author was not interested in the mid-bodies; in many cases they are not even mentioned in the paper, although they are shown in the drawings. In only a few studies do the figures delineate the detailed changes from their origin to their disappearance. The illustrations of the present paper, whether original or repro- duced from the literature, show all types of mid-bodies found, typical as well as atypical.” THE BEHAVIOR OF MID-BODIES Conditions Under Which Mid-bodies Are Present Mid-bodies arise only when a remnant of the spindle still exists when cell division occurs; this is necessarily pinched together by the advancing furrows. No mid-bodies are found if, as often happens, the spindle disappears prior to division. Such contrasting conditions may occur in the same cell type of closely related species, as illustrated in cleaving eggs of Echinoidea. In a series of experiments, done in collaboration with Dr. G. H. A. Clowes 2 Those illustrations taken from the literature (Figs. 13-69) were for the most part reproduced by photographing the drawings of the original papers. In four cases, however (Figs. 14, 62, 65, and 66), photographs were made of illustrations, redrawn from the originals, in Wilson’s The Cell in Development and Heredity. MID-BODIES AND CENTRAL BODIES 569 and Dr. M. E. Krahl, I studied Arbacia eggs under about 75 experi- mental conditions, in an attempt to analyze the relation between modifications of respiration and the behavior of the mitotic figure. Usually the eggs were fixed in Bouin’s fluid, but a number of other reagents were used. Mid-bodies are present (Fig. 5) with a spindle remnant (with rare exceptions when the material is very abnormal and both are absent). Boveri (1901) illustrates a similar condition in cleaving eggs of Echinus (Fig. 59). On the other hand, Echinarachnius eggs were also studied, using about 20 modifications of the technique, and the situation is in contrast to that in Arbacia eggs: the spindle disintegrates before division occurs and mid-bodies are not found. Toxopneustes eggs (Wilson and Leaming, 1895) also show neither body nor remnant. Here, therefore, are eggs of four species of Echinoidea, two with mid-bodies associated with a focalized spindle remnant, and two with neither. The same relation between the presence and absence of a spindle remnant and the presence and absence of a mid-body may exist even in the same material, under different experimental conditions. When Arbacia egg-sets are run at various temperatures, those which cleave at temperatures from 10° to 25° C. show both mid-body and spindle remnant, but at 30° neither is present. As just noted, a mid-body does not arise unless a remnant of the spindle is present when division occurs. Conversely, when a remnant is present, a mid-body is practically always formed, but a few exceptions to this are reported in the literature. For example, in the spermato- gonia of Enteroxenos (Fig. 55), the remnant is present without the body. Occasional instances like this are not surprising in view of the fact that the behavior of mid-bodies is so variable, depending upon species, cell cycle, fixation, and depth of stain. Scarcely a single generalization can be made about them to which some exceptions cannot be found. History of Mid-bodies There are a number of divergent structural types of mid-bodies, and their history differs. One type is illustrated in Arbacia eggs. In this case first cleavage begins with the appearance of a furrow at the animal pole (Fig. 1), but no mid-body arises until the furrow has advanced far enough into the egg to impinge upon the spindle; and the body which then forms at its tip is just a faint thickening (Fig. 2). Another furrow appears at the opposite pole about a minute later, and a similar body arises at its tip when it too presses upon the spindle. Meanwhile the first furrow has advanced more deeply and its body enlarges. By the time the two furrows are ready to touch each other, a mid-body is 570 HENRY J. FRY present at the end of each, and when contact is made these bodies join to form a dumbbell-shaped structure which undergoes no further change until it disappears when the spindle remnant fades (Figs. 3-5). Another type is illustrated in the brain cells of Squalus embryos, in which division occurs from one side only. A mid-body appears at the point where the spindle is first compressed by the approaching furrow. As it advances, other bodies arise, their size and number differing from cell to cell, but the end result in all is a single body at the opposite side of the cell, where the spindle is finally focalized. It too disappears when the spindle disintegrates (Figs. 6-11). Fics. 1-11. Successive stages in the formation of mid-bodies. Figs. 1-5: Arbacia eggs at first cleavage. Figs. 6-11: brain cells of Squalus embryos (Fry and Robertson, 1933). Several studies in the literature illustrate the history of other types. In erythrocytes of duck embryos (Heidenhain, 1907) a group of nondescript thickenings are aggregated into a centriole-like body (Figs. 16-20). In smooth muscle cells of Amblystoma (Pollister, 1932) an atypical elongate mid-body undergoes progressive condensation (Figs. 21-26). And in Arion eggs (Lams, 1910) slender elongate thickenings which arise at the middle of the spindle fibers are pinched together into a dense mass (Figs. 27-34). MID-BODIES AND CENTRAL BODIES 571 Although these types differ as indicated, they all have in common a gradual aggregation of mid-body material associated with an increasing focalization of the spindle brought about by the division process. The final position of the fully formed mid-body is determined by the point where the division furrows make contact. For example, in red blood cells of the duck, the furrows meet at the middle of the cell, and the mid-body lies there (Fig. 19). In Arbacia eggs during first cleavage one furrow arises about a minute before the other; their meeting point is therefore considerably off center, and the body lies there (Fig. 5). In Squalus brain cells there is only a single furrow arising at one side which presses the spindle against the opposite cell membrane, and the body lies there (Fig. 11). In most instances mid-body and spindle remnant disappear simultaneously. This is generally true of the materials examined in the present study. It is also shown in the illustrations in the literature, where telophase figures which have both body and remnant are usually succeeded by a drawing of the next mitotic phase without either. In some cases, however, the body persists for a short time after the remnant has disappeared. This point was studied in Arbacia eggs at first cleavage. After division is completed, all cells have both mid- bodies and spindle remnants, but about fifteen minutes later, just before the next prophase figure makes its appearance, the remnant is doubtful or absent in about 25 per cent (count of 50), although the bodies are still present in all cells. By the time the prophase figure arises mid-bodies have, with rare exceptions, disappeared. A few cases are also found in the literature where the mid-body is shown as persisting after the spindle remnant has faded, as for example, in red blood cells of the duck (Fig. 20). In other words, if one considers the converging configuration as a whole, the focal area is obviously the most condensed part, and hence this may explain why it occasionally persists longer than the surrounding region of the converging fibers, although it may have been originally created as a result of their becoming focalized. Configurations of M1id-bodies The configurations which mid-bodies show during their formation, when cell division is not yet completed and spindle fibers are only partially pressed together, must be distinguished from their final structure, when the aggregation of fibers is fully accomplished. The data here presented refer only to fully formed mid-bodies; the limited number of atypical ones will be described later. Many mid-bodies are round, smooth, and centriole-like (Figs. 11, Sy} REN Re ERY 19, 46, 58, and 62); or their surface may be rough (Figs. 56, 57, and 60). Another common type is elongate: it may be ovoid and smooth (Fig. 47), ovoid and rough (Figs. 45 and 59), dumbbell-shaped (Figs. 5 and 61), or irregular (Figs. 48 and 49); it may lie parallel to the major axis" of the spindle (Fig. 61) or at right angles to it (Figs. 5, 45, 47, 48, 49, and 54). In addition to these common forms, other types are found oc- casionally. The body may be ring-shaped (Figs. 29, 36, and 37), cone- like (Figs. 30 and 52), linear in form (Fig. 26), or it may be composed of thickenings which arise at the middle of the spindle and are then aggregated loosely (Fig. 51) or tightly (Fig. 33). A specialized typeisa large irregularly outlined area containing a small vesicle (Fig. 43). In general the above types are found only when the division process has focalized the spindle to a point or to a very limited area. But associated with a specialized kind of remnant (mitosome) which is not focalized to a point, and persists as a relatively broad band of fibers, is a very different type of mid-body, composed of a row of granules (Fig. 14). In the studies examined this type was found only in spermatogonia and spermatocytes, and not in all of them. Of 15 investigations of spermatogenesis taken at random from the literature, 13 show this unusual remnant and the other two have a fully focalized remnant with a single mid-body (e.g., Enteroxenos, Figs. 56 and 57). In cleaving eggs of Arion the mid-body is also a broad band of aggregated spindle fibers, but no granules are present (Fig. 33); and in the polar body cycles of this species there is a similar band associated with a ring-shaped mid-body (Fig. 29). ; It is significant that the mid-body which is composed of a row of granules is found only in the specialized broad spindle remnants of male germ cells, whereas the other types which are usually more aggregated, are found with remnants which are focalized to a greater degree. In some cases the very configuration of the mid-body indicates that it is focalized spindle substances. In cleaving Arion eggs, for example, thickenings which appear at the mid-region of the spindle are pressed together by the division furrow into a bundle, which is later bent into a U-shaped figure when the nuclei move from their former position toward the surface of the egg (Figs. 32-34). In cone-shaped bodies, the very form shows that they are spindle substances which have been pinched together (Figs. 30 and 52). And when the atypical, elongate mid-body of smooth muscle cells of Amblystoma makes its first appear- ance it is obviously a condensation of spindle materials (Fig. 22). The other types—dots, dumbbells, ovoids, rings, lines, and vesicles —which do not appear to the eye to be focalized spindle substances, MID-BODIES AND CENTRAL BODIES Ss nevertheless behave like those which do. In general, all types arise after focalization of the spindle has begun, become more aggregated as focalization is completed, and disappear as the spindle remnant disappears. A series of structural groups could be arranged, beginning with the centriole-like type, and passing by gradations to those which are clearly aggregations of spindle materials. Relation Between Structure of Mid-bodies and Coarseness of Spindle Fibers When mitotic figures are subjected to various experimental con- ditions and fixed with different reagents, mid-bodies often vary in size and shape, and the fibers vary in coarseness. The relation between the physical structure of the mid-body and the distinctness of the fixed fibers was studied in primary spermatocytes of Romalea and in cleaving eggs of Arbacia. TABLE [| The relation between the size of the granules composing the mid-bodies in primary spermatocytes of Romalea microptera and the coarseness of the spindle fibers. In general, the coarser the fibers, the larger are the granules, but there are various exceptions to this generalization. Forty-eight modifications of the technique were used, and the depth of stain was similar in all. The table shows the percentages of each type of association of mid-body and spindle remnant. Coarseness of Spindle Fibers Size of Granules Composing Mid-bodies Vague or Delicate Coarse Very Coarse per cent per cent per cent INGER eae thie waite teeta sae 8 — 3 Snell (OL ee eae ee 18 13 21 Medium (1.0u)............ — 18 15 anoen(lesis)i. se) e.cee sees A — — 4 Romalea spermatocytes were subjected to 48 experimental modifi- cations, differing both as to environmental factors prior to coagulation and the fixatives employed. They were all stained to the same depth with Heidenhain’s hematoxylin. Under usual conditions the mid- body is composed of a row of practically contiguous granules, each about 1 w in diameter, and the spindle fibers have a certain degree of distinctness. The same situation obtains in spermatogonia (Fig. 14). After modifications of the technique, however, the granules may be absent, small, medium, or large, and they may vary in shape; the fibers may be vague, delicate, coarse, or very coarse. The relation between variation in size of granules and coarseness of fibers is reported in Table I. In general, the more delicate the fibers 574 HENRY J. FRY are, the smaller the granules. But the presence of very coarse fibers does not guarantee large granules, and those of small size are associated with fibers of all degrees of coarseness. In Arbacia eggs, in the 75 experiments previously noted, the dumbbell-like mid-body is in most cases of usual length—about 1.5 y»— and the small spindle remnant is composed of delicate fibers. But in 6 of the experiments the fibers are unusually coarse and the bodies are larger. In this material too, however, there are a few exceptions: the remnant may have the usual delicate appearance and yet be associated with exceptionally large bodies, or the fibers of the remnant may be very coarse while the bodies are of usual size. In general, therefore, the size of the mid-body is directly pro- portional to the coarseness of the spindle fibers, but there are ex- ceptions. These are probably explained by the uncertainties of focalization phenomena—which is another way of saying that we do not know what the structure of the living spindle is, and we know even less about the effects of coagulation upon it, especially at points of focalization. Furthermore, when we are dealing with areas of focaliza- tion, we must also remember that slight differences in depth of stain may produce marked differences in appearance. Variations in Mid-body Structure from One Cell Cycle to Another in the Same Species Within the cell types of any one species—o6gonia, o6cytes, sperma- togonia, spermatocytes, large blastomeres, small blastomeres, and the innumerable kinds of somatic cells—there may be wide variation in mid-body structure. . In eggs of Cerebratulus, Cumingia, Nereis, Chetopterus, and Asterias, the mid-bodies of the first polar body figure are larger than those of the second. In Arbacia eggs, also, during the first three cleavages, the average length of the dumbbell-shaped bodies is successively 1.5 u, 1.4 uw, and 1.2 » (25 measurements in each case). And in Cumingia the diameter of the mid-body at first cleavage is 1.6 u, but at the second it is 1.0. In all these cases Bouin’s reagent was employed. It is probable that differences in the size of the mitotic figure may explain why mid-bodies are larger in the first polar body figures than in the second, and larger in first-cleavage figures than in those following. In Nereis eggs, during the first polar body cycle, the mid-body is round and relatively large, with a rough surface (Fig. 60); during the second cycle it is dumbbell-shaped and small, with a smooth surface (Fig. 61); at first cleavage, however, it is absent (Bouin’s fixation). This is probably explained by the fact that a spindle remnant is MID-BODIES AND CENTRAL BODIES SS) present when the polar bodies are pinched off, but not when the egg divides. Crepidula eggs (Conklin, 1902) show striking differences in mid- body structure from cycle to cycle. In first polar body figures it is a ring, which appears as two dots when seen in cross-section (Figs. 35 and 36) ; in the second, it is a smaller ring with a dot in it (Fig. 37), or just a dot (Fig. 38); in first-cleavage figures it is a large area containing a vesicle (Fig. 43); in third-cleavage figures it is a small irregular blob (Fig. 45), and in fourth-cleavage figures, a minute dot (Fig. 46). The difference in size of the mid-bodies is apparently related to the size of the mitotic figures concerned. The factors which produce rings in some cells and vesicles or dots in others are not as yet understood. Variation in the structure of mid-bodies is also exhibited in Enteroxenos cells in different cycles (Figs. 47-57). B00B80 868 080008 08u [00880889 Fic. 12. Variation in mid-body structure in Arbacia eggs at first cleavage. Twenty-five bodies selected at random from a single slide are shown in outline and without their accompanying spindle remnants. The unit of the scale is 0.5. All eggs are in late telophase just after cleavage is completed. Bouin’s fixation. Instability of Mid-body Structure Mid-bodies vary in size and shape from cell to cell on the same slide. Outline drawings of mid-bodies in Arbacia eggs at first cleavage (Fig. 12) illustrate the variation in size and contour which may occur in cells at the same stage treated identically and lying side by side on the same ‘slide. The extent of this variation is typical of the materials reported in this paper. Mid-bodies of odgonia (Figs. 47-49) and those of the second polar body figures (Figs. 52-54) in Enteroxenos cells also exhibit variation in shape. 576 HENRY J. FRY Mid-bodies vary in structure due to the use of different fixatives. In some instances a wide modification of the technique has but little effect upon the structure of the mid-body, as in Arbacia eggs at first cleavage (Fig. 5), but in others the effects are considerable, as in primary spermatocytes and spermatogonia of Romalea (Fig. 14 and Table I). As a general rule, if fixation demonstrates coarse spindle fibers, the mid- bodies are large; if the fibers are delicate, the bodies are smaller. Mid-body structure is varied by modifications of environmental factors. In Arbacia eggs the size of the mid-body at first cleavage is modified by temperature. At temperatures from 25° to 20° C. the length is 1.5 uw; at 15° C. it increases to 2.8 uw; at 10° and 7.5° it again has an average length of 1.5 4. There is, however, much greater variation at the lower temperatures than at the higher ones. (Bouin’s reagent used; counts of 25 in each case.) When Cerebratulus eggs cleave at 20° C., mid-bodies are present only in about 2 per cent of the cells (Boveri’s picro-acetic reagent), and are less than 1 » in diameter; but when these eggs are allowed to develop at 15° C., and are fixed in the same manner, mid-bodies over 2 wu in diameter are present in about 10 per cent of the eggs. (Counts of 50 eggs.) Mid-bodies may be present in some cells and absent 1n others prepared im an identical manner. A given egg-set may be fixed under usual laboratory conditions and run up in the regular manner, yet cells on the same slide and at the same mitotic phase may show mid-bodies and remnants in some cases and neither in others. For example, Cere- bratulus eggs, as just noted, have only 2 per cent with mid-bodies and spindle remnants at first cleavage. Cumingia eggs at first division show 18 per cent (Bouin’s fixation), and Arbacia eggs show 30 per cent when dividing at 7.5° C. (Bouin’s reagent). These percentages are based on counts of 50 eggs. Various similar examples could be cited. But no matter how the experimental technique is modified, a mid- body is present only when there is a spindle remnant, with the rare exceptions previously mentioned (pp. 569 and 571). How shall we explain, then, the presence of a spindle remnant in some eggs and its absence in others which are at the same mitotic phase, the two kinds lying side by side on the same slide? The probable cause of this variation is an uncontrollable factor in the fixation process: eggs are added to the reagent in a certain amount of sea water, necessarily causing some dilution of the fixative. Each egg is coagulated within the first second after exposure to the reagent (based on a study of Chetopterus eggs (Fry, 1932, pp. 173-176)), but several seconds are required for the egg suspension and the fixative to mix completely. MID-BODIES AND CENTRAL BODIES S00 Some of the eggs are thus coagulated by the reagent at full strength and others at various degrees of dilution. In Chetopterus eggs such uncontrollable dilutions of the fixative produce marked differences in the structure of astral rays and central bodies (Fry, 1932, pp. 161-167). This factor may also operate in the case of mid-bodies and their associated spindle remnants, and may explain their instability of structure, their presence in some cells and absence in others on the same slide and their differences in size and contour from cell to cell.’ Mid-bodies in Protozoa Mid-bodies are usually absent in Protozoa. In the majority of species mitosis is intranuclear, the nucleus separating into two parts while the membrane remains intact. The morphological details of the process of separation vary widely: in many cases the connecting strand, prior to the final separation, is long and thin, while in others it is relatively broad; in some the spindle fibers are visible up to the time the break occurs, and in others they disappear before then. But in most species there is no special point of focalization and no mid-body. There are a few exceptions, however, where the mode of nuclear constriction aggregates the spindle fibers to a point or a limited area, and these may have mid-bodies, as in Cerattum (Lauterborn, 1895) and Coccidium (Schaudinn, 1900). In the occasional instances where the nuclear membrane disappears and the division figure lies in the cytoplasm, a mid-body may arise as the spindle is focalized by the process of cell division, exactly as in eggs of Metazoa. This is illustrated in Acanthocystis (Schaudinn, 1896). But if cell division does not occur at the time when such a cytoplasmic figure is present, and the spindle is hence not focalized, there is no mid-body, as shown in Monocystis (Muslow, 1911). Atypical Mid-bodies The mid-bodies thus far described arise without exception only as spindle fibers are focalized. There are, however, several cases illus- trated in the literature where thickenings appear either just prior to the process of focalization or in its complete absence. 3 With the assistance of Dr. George Child, an attempt was made to produce mid-bodies artificially. Somewhat in advance of the time when focalization of the spindle occurs naturally, eggs of Arbacia and Cheiopiterus were individually con- stricted with a glass hair in a plane passing through the middle of the late anaphase spindle, in order to focalize it artificially. They were fixed instantaneously while the needle was still in position, and then run up individually. The experiment failed, because the protoplasm was rendered completely hyaline by the manipulation, in the plane through which the needle passed, and no structure of any kind could be seen. 578 HENRY J. FRY In Arion eggs (Lams, 1910), both in polar-body and first-cleavage figures, thickenings arise at the middle of the spindle during anaphase, before division has begun. Thereafter these are aggregated in an orthodox manner into a bundle-like body when division occurs (Figs. 27-34). A similar situation is found in spermatogonia of Blaps (Nonidez, 1920). In spermatocytes of Llaveia (Hughes-Schrader, 1931) the spindle is composed of tubes, each associated with a tetrad or a dyad. Soon after the chromosomes separate, during anaphase, the central part of each tube shrinks to a cord-like structure. At this time in some tubes, or a little in advance of the shrinking process in others, mid-bodies make their appearance—one in tubes associated with dyads, two in those with tetrads. A little later the tubes are pressed together by the division furrows and coalesce, the mid-bodies still maintaining their identity (Fig. 63). In developing Drosophila eggs (Huettner, 1933) mid-bodies arise at the middle of the disintegrating spindle during telophase, in the complete absence of cytoplasmic division (Fig. 64). This also occurs in maturating eggs of Aspidiotus (Schrader, 1929). These rare cases which, with the exception of Arion, I found only in cells of insects, do not invalidate the fact that the great majority of mid-bodies arise only in connection with the focalization of spindle fibers. No explanation can be given for these exceptions, but such behavior is not surprising in disintegrating gelatinous material. Certain other cases reported in the literature might be regarded as atypical, but they are only additional examples of the instability of mid-bodies. For example, Buchner (1915, p. 28) illustrates in Ascaris eggs a mid-body of irregular shape without any spindle remnant, whereas Carnoy and Lebrun (1897, Plate II), who studied the same material and used a similar, though not identical reagent, illustrate neither body nor remnant. Such different results may be due to differences in the reagents employed, or, if the mid-body here is a staining artifact, to variations in the depth of stain. Furthermore, where Buchner shows it without an associated spindle remnant, this may be one of those occasional cases where the body persists after the remnant has disappeared. DISCUSSION The Nature of Mid-bodtes The foregoing facts indicate that mid-bodies are phenomena of focalization. They do not arise in cells in which the spindle dis- integrates prior to the time of division; and in those in which the spindle MID-BODIES AND CENTRAL BODIES 579 is still present at that time, mid-bodies appear only as the fibers are focalized by the process of cell division; they disappear, with rare exceptions, when such areas of focalization disintegrate. Their wide variability as to mode of formation is associated with the manner in which the spindle fibers are aggregated. Their final position in the cell is determined by the point at which the spindle remnant is pinched together by the division furrows. They often show variability in appearance from species to species, from cell cycle to cell cycle, and even from cell to cell on the same slide. They are frequently modified by differences in the fixatives employed, or variations in environmental factors prior to fixation, and such differences are usually related, with exceptions, to the size of the mitotic figure concerned, and the coarse- ness of the spindle fibers. In many cases their structure is unques- tionably nothing but aggregated spindle substances; and it appears that such focalization phenomena may take many forms, sometimes even simulating centrioles. Three classes of focal bodies were mentioned earlier: living focal bodies, coagulation artifact focal bodies, and staining artifact focal bodies. A number of experiments were carried out to determine to which of these categories mid-bodies belong. Some of them un- doubtedly exist as a delimited structure in the coagulated cell. In these cases, when the materials are slowly destained, the result being closely observed at every stage by the use of a high power water- immersion objective, it is seen that the bodies retain the dye (Heidenhain’s haematoxylin) to a greater extent than any other cell component, but they too finally yieldit. During the destaining process they grow lighter in color but show no change in size. When the fully destained slides are run up and studied with an oil immersion objective, the bodies are still seen distinctly as refringent structures, although they are colorless. Examples of this class of focal bodies are found in the first polar-body cycle in Nereis (Fig. 60) and in cleaving Arbacia eggs (Fig. 5). Whether they existed in the living cells as minute pools of fiber material at the focal center (living focal bodies), or the living point of focalization had no such structure and the body was produced by fixation (coagulation artifact focal bodies) cannot be determined because no structure can be seen in the living spindle. Other mid-bodies, though sharply demarked and sometimes looking exactly like centrioles, are clearly nothing but accumulations of dye at focal points. If the materials are heavily stained with Heidenhain’s heematoxylin, the bodies are relatively large; if stained in the ordinary 580 IBIS WN IROSZ ifs IISSNY manner, they are smaller. If these bodies are gradually destained, while watching the process, they do not lose their color while main- taining their characteristic size, but throughout the process they remain intensely black and become progressively smaller until they disappear. When such completely destained preparations are run up and studied in the usual way they show fibers which come to a point without the presence of a body. This is the case for example, in brain cells of Squalus embryos which have a centriole-like mid-body (Fig. 11). The usual hypothesis that mid-bodies of animal cells are related in some way to the cell-plate of plant cells is not consistent with the behavior of mid-bodies as here described. The cell-plate is associated with a broad spindle which has distinct fibers in the mid-region (Fig. 13), whereas the mid-body is associated with a disintegrating spindle, and only after its fibers have been focalized. In most in- Fries. 13-15. Resemblances between the cell-plate of plant cells and mid-bodies of animal cells. Fig. 13: cell plate in pollen mother cells of Fritillaria (Strasburger, 1888). Fig. 14: fully formed mid-body in Romalea spermatogonia (Wilson, 1928). Fig. 15: early stage during formation of mid-body in epithelial cells of the salamander lung (Flemming, 1891). stances, regardless of their diversity in shape, mid-bodies are single structures which do not resemble cell-plates in any way. In certain cases, however, there is enough resemblance to have given rise to the hypothesis. The mid-body which is composed of a row of granules simulates the fixed cell-plate to some extent, but the differences are marked (Figs. 13 and 14). Also, there is occasionally a brief phase during the formation of a single mid-body when several granules are present (Fig. 15) that calls to mind the cell-plate. In general, however, it seems that the mid-body and the cell-plate have nothing in common. But regardless of the relation of the mid-body to the cell-plate, the question may be raised as to whether or not the mid-body is a true cell component playing some réle in the process of cell division and the consequent focalization of the spindle. May it be a causative factor MID-BODIES AND CENTRAL BODIES 581 and not just an effect of focalization? The findings of this study do not support such an interpretation. (1) In some cases the mid-body is only an accumulation of dye. (2) Arbacia eggs cleave whether mid- bodies are present (at temperatures from 10° to 25° C.) or absent (at 30° C.), indicating that they play no essential réle. (3) This con- clusion is also supported by the fact that mid-bodies are present in cleaving eggs of some species of Echinoidea but notin others. (4) And all the other data presented in this investigation make it highly improbable that we are here dealing with a cell component which plays a causative role. Cells exhibit other areas of focalization—aside from central bodies which will be discussed in a moment—that behave much like mid-bodies. In his classic study of Crepidula eggs Conklin (1902) illustrates an unusual case. During anaphase of the second maturation division the minute centriole becomes a vacuole, while the centro- some enlarges. Fora very brief period, during middle anaphase, a number of minute bodies appear, each associated with a bundle of spindle fibers (Figs. 67-69). No such bodies arise during the first polar-body cycle in this species. In Chetopterus eggs, however, such phenomena occur during the first maturation division but not in the second (Mead, 1898, Plate 17). In Ascaris eggs (Carnoy and Lebrun, 1897, Plate I) they occur in both cycles. Beélar (1928, p. 35) illustrates a somewhat similar situation in Monocystts. Another possible phenomenon of focalization is shown when protoplasm and certain artificial emulsions are fixed with reagents which demonstrate foam structure (Biitschli, 1894). Minute bodies are frequently found at the points where the lines of the foam structure meet (Figs. 65-66) and these may be focal bodies. A still different phenomenon of focalization is illustrated in telophase of the first polar body cycle of Arion eggs (Fig. 30). By the time the center of the old aster has disintegrated two new central bodies and asters have appeared within it. The rays of the old aster are most aggregated about this center, where they form a dark, diffuse, ring-like area of focalization. The Significance of Mid-bodies for the Central Body Problem If focalization of the middle of the spindle may result in the formation of structures which in some cases simulate individualized cell components, it is in order to examine the situation at the ends of the spindle, where similar focalization areas occur, of both rays and fibers, or fibers alone. In certain cases the similarity in appearance between mid-bodies and central bodies is so marked that it compels attention. For example, Conklin (1902, p. 43) 4 notes the similarity in first-cleavage 4 In using this quotation I have taken the liberty of substituting modern termi- nology (according to Wilson, 1928, p. 675) for that used in the original paper. The modern term centriole is used instead of the old term centrosome to indicate a minute, sharply demarked body; the modern term centrosome is used instead of the old term sphere to describe a larger less sharply demarked area which surrounds the centriole or exists alone; the term central body is a general one which includes either or both of the others. 582 Ta EIN[IRNE IJ LEM ENC figures of Crepidula eggs: ‘‘This mid-body is for all the world like a centriole with its surrounding centrosome and aster, and recalls Watasé’s (1893) comparison of the mid-body to an intercellular centriole. This apparent resemblance is still further supported by the fact that the mid-body in this case becomes a hollow sphere before it finally disappears, just as the centriole does. The mid-body is sur- rounded by a darkly staining substance which resembles the centrosome substance”’ (Figs. 39-44). In short, the structural changes which occur in cleaving Crepidula eggs at the ends of the spindle during the early history of the mitotic figure are later repeated at the middle of the spindle when it is focalized by the division process. Furthermore, mid-bodies and central bodies are again identical in appearance at fourth cleavage, though here both are minute dots (Fig. 46). Such a similarity of appearance occurs frequently: in Squalus brain cells (Fig. 6), in leucocytes (Fig. 58) and connective tissue cells (Fig. 62) of the salamander, in erythrocytes of the duck (Fig. 19), in cleaving Echinus eggs (Fig. 59), and in many other cases. There are numerous instances, however, where the two bodies differ completely in appearance. The sharply demarked, minute, dumbbell- like mid-body in cleaving Arbacia eggs (Fig. 5) bears no resemblance to the large non-demarked granular central body. The elongate mid- body in the smooth muscle cells of Amblystoma is unlike the typical centriole (Figs. 21 and 26). In the first maturation division of Arion eggs the mid-body is a band of fibers surrounded by a ring, in contrast to the minute dot-like centrioles (Fig. 29); during first cleavage, however, it is a similar band but without a ring, whereas the central body is a large area concentrically differentiated and without centrioles (Fig. 33). Crepidula eggs during maturation (Figs. 35-38) and Enteroxenos cells at various cycles (Figs. 47-57) also exhibit differences in appearance between mid-bodies and central bodies, and many other examples could be cited. The uncertainties of focal phenomena may explain why mid-bodies and central bodies look alike in some cases and different in others, the instances of dissimilarity being somewhat more numerous. But the significant point is not whether mid-bodies and central bodies look alike, but whether or not they behave alike. In my studies of central bodies in various cell cycles of several species, some of which were used in the present investigation of mid- bodies, I demonstrated that the structure of the central body in these cases is related to the structure of the rays or fibers. In cytasters of artificially activated Echinarachnius eggs (1928), where no spindle is MID-BODIES AND CENTRAL BODIES 583 present, a central body occurs only when rays reach the center, regardless of how distinct the rays are peripherally. When a spindle is present without asters, as in Squalus brain cells (1933a), the structure is again related to the coarseness of the fibers. When the history of the central body is followed from the beginning to the end of a mitotic cycle, as it was in cleaving Echinarachnius eggs (1929a) and Squalus brain cells (1933a), it is apparent that the central body undergoes changes related to the coarseness of the converging rays and fibers, and their general configuration. The continuity of the bodies from one cell cycle to another cannot be demonstrated; they arise as areas of focalization arise, and disappear as such areas disintegrate, even though the peripheral region still has distinct rays. When, at any given mitotic phase, rays and fibers are modified by the use of different fixatives or environmental factors, the structure of the central bodies is changed, as shown in Echinarachnius eggs at metaphase of the first-cleavage figure (19296 and 1929c). This relationship is also shown with unusual clearness in Chetopterus eggs (1932 and 19330). Here, furthermore, when supposedly typical centrioles are demon- strated, it is found that they differ in size and contour from cell to cell, at the same mitotic phase, on the same slide. Hence the conclusion was reached that in these cases the supposed central bodies are phenomena of focalization: staining artifacts in Squalus brain cells, and coagulation artifacts in the others. The technique used in studying the mid-bodies reported in the present paper was the same as that employed in the investigations of central bodies just noted. Many fixatives were used to modify the structure of the fibers, in order to determine the effects of such modifi- cations upon the structure of the bodies; for the same reason, cells were also subjected to abnormal environmental conditions prior to fixation. In each case the sample studied was large enough to determine the extent of structural variation, and all classes were reported and considered in arriving at the conclusion. If focalization phenomena are involved in the formation of both mid-bodies and central bodies, it is rather surprising that in mitotic figures which have asters, those bodies occurring at the middle of the spindle where only fibers are present, should ever look like those occurring at the spindle-ends where both fibers and rays are concerned. But, as previously mentioned, in some cases they may be identical in appearance, and in others completely dissimilar. At present it is as impossible to explain this fact as it is to explain why, for example, the mid-bodies of Crepidula eggs are ring-like during polar body formation, but vesicle-like at first cleavage, and centriole-like at fourth cleavage 584 HENRY Vann Fics. 16-34. Successive stages in the formation of mid-bodies. erythrocytes of duck embryos (Heidenhain, 1907). Figs. 21-26: non-striated muscle cells of Amblystoma larvae (Pollister, 1932). Figs. 27-34: first polar bodies and first cleavage in Arion eggs (Lams, 1910). MID-BODIES AND CENTRAL BODIES 585 Fics. 35-46. Mid-body structure and central body structure in various mitotic cycles of Crepidula eggs (Conklin, 1902). Figs. 35-36: first polar bodies. Figs. 37-38: second polar bodies. Figs. 39-44: first cleavage. Fig. 45: third cleavage. Fig. 46: fourth cleavage. Fics. 47-69. Mid-bodies in various mitotic cycles of Enteroxenos cells (Bon- nevie, 1906). Figs. 47-49: odgonia. Figs. 50-51: first polar bodies. Figs. 52—54: second polar bodies. Fig. 55: spermatogonia. Fig. 56: primary spermatocytes. Fig. 57: secondary spermatocytes. Mid-bodies of various cells. Fig. 58: salamander leucocytes (Béla¥, 1928). Fig. 59: cleaving Echinus eggs (Boveri, 1901). Figs. 60-61: first and second polar-body figures of Nereis eggs (original). Fig. 62: con- nective tissue cells of the salamander lung (Flemming, 1891). Fig. 63: primary spermatocytes of Liaveia (Hughes-Schrader, 1931). Fig. 64: developing Drosophila eggs (Huettner, 1933). Bodies occurring at the junction of the walls of alveoli. Fig. 65: fixed eggs of Sphaerechinus (Biitschli, 1894). Fig. 66: artificial emulsion of olive oil and NaCl (Biitschli, 1894). Multiple bodies, in addition to central bodies, occurring at spindle-ends during middle anaphase in the second maturation division of Crepidula eggs (Conklin, 1902). Fig. 67: early anaphase without bodies. Fig. 68: middle anaphase with bodies. Fig. 69: late anaphase without them. MID-BODIES AND CENTRAL BODIES 587 (Figs. 35-46). If, however, the focalization of fibers or rays may cause the production of focal bodies, it is not surprising that there is variation in their structure according to the size of the configuration as a whole and the coarseness of the converging fibers. But our knowledge of the physical chemistry of such systems is not sufficient for us to explain the structural diversity in focal bodies, from species to species, and cell cycle to cell cycle, in mitotic figures of the same size and distinctness. If the conclusion of the present study is valid, we cannot accept the presence of a sharply demarked body in an area of focalization—whether at the middle or the ends of the sbindle—as evidence that we have demonstrated a true cell component. This is, however, the generally accepted practice. Structures which appear to the eye to be as individualized as chromo- somes or plastids may nevertheless be merely ephemeral transient phenomena of focalization—sometimes nothing but focal accumulations of dye, as in both mid-bodies and centrioles in brain cells of Squalus (Fig. 6) (Fry and Robertson, 1933). The eye sees such a structure, and the mind accepts it on the basis of its appearance and attempts to imagine its function. The fact that bodies do not arise at the middle of the spindle until it is focalized by the division furrows should make us alert to possible pitfalls—errors in interpretation—when structures are found at points where fibers or rays or both converge. In this connection it is sig- nificant that bodies are not present at the ends of anastral spindles unless the tips are sharply focalized. The blunt, anastral spindles found in some odcytes and spermatocytes do not have them (with possibly a few exceptions in the latter when centriole-blepharoplasts are present). If the conclusion of this investigation is correct, the current central body hypothesis must be reevaluated. Focal bodies have probably been confused with such cell components as centriole-blepharoplasts of male germ cells, diplosomes of vertebrate cells, and similar struc- tures of Protozoa and other cell types, which for the most part are concerned with the formation of axial filaments and flagella, no matter what their rdle in mitosis. The great majority of cell types have no such components, and it is yet to be ascertained to what an extent their supposed central bodies are actually artifacts of focalization.® 5 The data of the present study suggest that the spindle, despite its homogeneous appearance in the living cell, has some kind of linear organization. The very fact that bodies appear in the plane where the division furrow exerts pressure on the spindle indicates that there are differentiated materials there. If the spindle were actually homogeneous, it is probable that its materials would flow to one side or the other when it is pinched together, in which case no focalization bodies would be formed. 588 HENRY J. FRY RESUME Mid-bodies were studied in various cell cycles in a number of species; many fixatives were used and cells were subjected to different environmental modifications prior to fixation, to modify the structure of both bodies and fibers in order that the relation between them might be analyzed. Mid-bodies are found only in cells in which the spindle is still present when division occurs; they are absent in the numerous cases where the spindle disappears before that time. They arise only as the spindle fibers are gradually brought to a focus by the advancing division furrows; they usually disappear simultaneously with the spindle remnant. Their final position is determined by the point where the division furrows meet. There are many structural types: centriole-like dots, blobs, ovoids, dumbbells, rings, cones, lines, rows of dots, and other configurations. Some are obviously nothing but aggregated spindle materials, whereas others look like individualized cell components. Different structural types may occur in the same species in successive cell cycles. They may vary in size and contour from cell to cell on the same slide. They are generally unstable and easily modified by the use of various environmental factors or different fixatives. Such structural modifications are usually related to the coarseness of the fibers. Mid-bodies are generally regarded as a vestigial homolog of the cell-plate of plants. The present study, however, interprets them as phenomena of focalization or ‘‘focal bodies.”” It is suggested that these are produced as the result of the concentration of spindle or aster materials at points where they are focalized. In some cases mid- bodies are only an accumulation of dye at the focal area; in others they are probably produced by the process of fixation. If bodies which look like typical cell components can be formed at the middle of the spindle, as the result of the focalization of its materials brought about by the advancing division furrows, similar phenomena may occur at the areas of focalization at the spindle ends where both fibers and rays may converge. Previous studies of central bodies, in some of the same cells used for the present study of mid- bodies, indicate that they too are produced as a result of the focalization of spindle and aster materials. It is, therefore, suggested that in the formulation and development of the current central body hypothesis, focal bodies may have been confused with true cell components such as centriole-blepharoplasts, diplosomes, and similar structures. MID-BODIES AND CENTRAL BODIES 589 BIBLIOGRAPHY Bear, Kari, 1928. Die cytologischen Grundlagen der Vererbung. Gebriider Borntraeger, Berlin. BoNNEVIE, KrisTINE, 1906. Untersuchungen iiber Keimzellen. I. Beobachtungen an den Keimzellen von Enteroxenos 6stergreni. Jena. Zeitschr. Naturwiss., 41: 229. BovERI, THEODOR, 1901. Zellen-Studien, Heft4. Ueber die Natur der Centrosomen. Gustav Fischer, Jena. BucHNER, PAUvL, 1915. Praktikum der Zellenlehre. Gebriider Borntraeger, Berlin. BitscuHii, O., 1894. Investigations on Microscopic Foams and on Protoplasm. (Translated from the German work of 1892 by E. A. Minchin.) Adam and Charles Black, London. Carnoy, J. B., AnD H. Lesprun, 1897. La fécondation chez !’Ascaris megalocephala. La Cellule, 13: 63. ConkKLIN, Epwin G., 1902. Karyokinesis and cytokinesis in the maturation, fertilization and cleavage of crepidula and other gastropoda. Jour. Acad. Nat. Sct. Phila., Ser. 2, 12: Part 1. FLEMMING, W., 1891. Neue Beitrage zur Kenntniss der Zelle. II. Arch. mikros. Anat., 37: 685. Fry, Henry J., 1928. Conditions determining the origin and behavior of central bodies in cytasters of Echinarachnius eggs. Bzol. Bull., 54: 363. Fry, Henry J., 1929a. The so-called central bodies in fertilized Echinarachnius eggs. I. The relationship between central bodies and astral structure as modified by various mitotic phases. Biol. Bull., 56: 101. Fry, Henry J., 19296. The so-called central bodies in fertilized Echinarachnius eggs. II. The relationship between central bodies and astral structure as modified by various fixatives. Bzol. Bull., 57: 131. Fry, Henry J., MATTHEW JACoBs, AND H. M. Lies, 1929c. The so-called central bodies in fertilized Echinarachnius eggs. III. The relationship between central bodies and astral structure as modified by temperature. Bzol. BU Sdicnlode Fry, Henry J., 1932. Studies of the mitotic figure. I. Chetopterus: central body structure at metaphase, first cleavage, after picro-acetic fixation. Bzol. Bull., 63: 149. Fry, Henry J., anD C. W. ROBERTSON, 1933a. Studies of the mitotic figure. II. Squalus: the behavior of central bodies in brain cells of embryos. Anat. Rec., 56: 159. Fry, Henry J., 19336. Studies of the mitotic figure. III. Chztopterus: central body structure at metaphase, first cleavage, after using diluted fixatives. Biol. Bull., 65: 207. HEIDENHAIN, Martin, 1907. Plasma und Zelle. Gustav Fischer, Jena. HvETTNER, ALFRED F., 1933. Continuity of the centrioles in Drosophila melano- gaster. Zeitschr. Zellfor. mikr. Anat., 19: 119. HUGHES-SCHRADER, SALLY, 1931. A study of the chromosome cycle and the meiotic division-figure in Llaveia bouvari—a primitive coccid. Zettschr. Zellfor. mtkr. Anat., 13: 742. Lams, H., 1910. Recherches sur l’oeuf d’Arion empiricorum. Mém. Acad. roy. Belg., Classe Sci., Deuxiéme Série, 2: 1. LAUTERBORN, R., 1895. Protozoenstudien. I. Kern- und Zelltheilung von Ceratium hirundinella. Zeztschr. wiss. Zool., 59: 167. Meap, A. D., 1898. The origin and behavior of the centrosomes in the annelid egg. Jour. Morph., 14: 181. Mustow, Kart, 1911. Uber Fortpflanzungserscheinungen bei Monocystis rostrata, n.sp. Arch. Protist., 22: 20. NOoNIDEZ, José F., 1920. The meiotic phenomena in the spermatogenesis of Blaps, etc. Jour. Morph., 34: 69. 590 HENRY) Jo ERY POLLISTER, ARTHUR W., 1932. Mitosis in non-striated muscle cells. Anat. Rec., 53: 11. ScHAUDINN, F., 1896. Uber das Centralkorn der Heliozoen, ein Beitrag zur Centro- somenfrage. Verhandl. Deutsch. Zool. Ges., 6: 113. ScHAUDINN, F., 1900. Untersuchungen iiber den Generationswechsel bei den Coccidien. Zool. Jahrb. Abt. Anat., 13: 197. SCHRADER, FRANZ, 1929. Notes on reproduction in Aspidiotus hederze (Coccide). Psyche, 36: 232. SHARP, LESTER W., 1934. Introduction to Cytology. Third Edition. McGraw- Hill Book Co., Inc., New York. STRASBURGER, ED., 1888. Histologische Beitrage. I. Ueber Kern- und Zelltheilung im Pflanzenreiche usw. Gustav Fischer, Jena. WATASE, S., 1893. Homology of the centrosome. Jour. Morph., 8: 433. ‘Witson, Epmunp B., AND Epwarp LEAmMING, 1895. An Atlas of the Fertilization and Karyokinesis of the Ovum. Macmillan Company, New York. WItson, EpmunpD B., 1928. The Cellin Development and Heredity. Third Edition, with corrections. The Macmillan Company, New York. A PHYSIOLOGICAL AND HISTOLOGICAL STUDY OF THE PRONDAL, CORTES Of VTHE, SEAL (PHOCA DY FOULINA) + DAVID McK. RIOCH (From the Department of Anatomy, Harvard Medical School) The so-called ‘“‘motor areas’”’ of the cortex of the cat, dog and bear have been investigated recently with the combined methods of electrical stimulation and histological examination (Langworthy, 1928; Smith, 1933, 1935, etc.). The general pattern, both functional and anatomical, of these three fissiped carnivores is quite similar, although minor variations are distinct. It was therefore of interest to make a comparative study of the harbor seal (Phoca vitulina), a pinniped carnivore which is functionally adapted to an aquatic existence and in which the trunk and extremities are very considerably modified (Howell, 1928). I should like to express my thanks to the Superintendent, Mr. Thomas H. Dorr, and his assistants, particularly Mr. E. C. Barter, at the Boothbay Harbor Station of the United States Bureau of Fisheries, for obtaining the animals and assisting in handling them. METHODS Experiments were successfully completed on 6 seals approximately 4 to 5 months old. The animals were in excellent condition and weighed between 18 and 25 kgm. Satisfactory anesthesia was ob- tained with Dial fluid (Ciba) 0.3 cc. per kgm. intraperitoneally, supplemented by a few whiffs of ether during the opening of the skull. It was found preferable to keep the body of the seal sub- merged in a tank of water during the early stages in order to assist respiration.2 The thin skull was readily removed with trephine and rongeurs. Since the electrically excitable cortex was located behind the eye, the latter was enucleated and the orbital roof removed in 1The seals were obtained and the physiological observations made at the Boothbay Harbor Station of the U. S. Bureau of Fisheries. 2 When on land the seal normally breathes rhythmically and regularly. Five to ten minutes after the injection of the dial, respirations became intermittent. Four to five deep breaths were taken, ending in inspiration, following which the nostrils were tightly closed and the breath held for 15 to 40 seconds, the cycle being then repeated. The administration of ether by a cone during three or four of these cycles resulted in regular, slow deep respiration. 591 592 DAVID McK. RIOCH the later experiments. The rectal temperature of the animals was kept between 37° and 40° C. by pouring cold sea water over them occasionally. Stimulation was bipolar by means of a pair of silver-silver chloride electrodes with an interpolar distance of 2mm. Current was supplied by the 60 cycle a.c. line through a potentiometer. The effective voltage varied in different animals from } to 5 volts. For single Fic. 1. Outline drawing of a lateral view of the brain of seal 5. X 1.5. The position of the electrically excitable cortex is indicated by the shaded area. an, S. ansatus. co, S. coronalis. cr, S. cruciatus. e lat, S. endolateralis. Jat, S. lateralis. Jat p, S. lateralis posterior. olf, Bulbus olfactorius. p cr, S. post- cruciatus. pr cr, S. precruciatus. pr s, S. presylvius. pro, S. proreus. ri, Fissura rhinalis. s,S. pseudosylvius. ssa, ssp, sss, S. suprasylvius anterior, posterior and superior. shocks a Harvard inductorium, with 3 volts in the primary circuit and the secondary coil set at 7 to 9 cm., was employed. Between observations the cortex was kept moist by application of cotton pledgets wet with warm Ringer’s solution. At the end of the experiment the brains were removed and pre- served in formalin. Serial sections in the sagittal plane were cut at 35 w of the frontal poles of the left hemispheres of seals 3 and 4, and STUDY OF FRONTAL CORTEX OF THE SEAL 593 in the horizontal plane of the right hemisphere of seal 4. Every twentieth section was stained with thionine and mounted. RESULTS In conformity with the rounded shape of the skull, the high position of the nostrils and the microsmatic habits of the seal, the brain tends to a more spherical form than in the terrestrial carnivores. fF Fic. 2. Outline drawing of a frontal view of the brain of seal 4. X 1.5. The sulci are labelled as in Fig. 1. F The electrically excitable cortex is indicated on the left hemisphere. The numbers are referred to in the text. The histological areas are indicated on the right hemisphere. fr. agr, Area frontalis agranularis. fr. gr, Area frontalis granularis. g pyr pc, Area giganto- pyramidalis postcentralis. g pyr prc, Area gigantopyramidalis precentralis. gr pc, Area granularis postcentralis (the caudal boundary of this area is not defined in the figure). The general pattern of the sulci, however, was found to be similar, as illustrated in Figs. 1 and 2. A number of additional shallow sulci were present, which resulted in a reduplication and subdivision of the gyri, a condition more pronounced in other aquatic mammals (cf. Langworthy, 1932, 1935). Certain features in which the frontal lobe differed from that of other carnivores may be noted. The sulcus cruciatus was relatively small and was quite shallow laterally, but 594 DAVID McK. RIOCH medially became deeper and, behind the olfactory bulb, ran into the well-developed sulcus postcruciatus. The sulcus ansatus was well marked and entered the S. suprasylvius anterior. In some instances it extended medially as a shallow groove into the S. postcruciatus, but in no case into the S. lateralis. The S. coronalis was very shallow, appearing merely as a slight depression in some of the hemispheres. In one case it extended ventrally as far as the S. precruciatus, which was also shallow. The large Gyrus proreus was divided into three parts by the S. proreus and the thin olfactory stalk. PHYSIOLOGICAL OBSERVATIONS The boundaries of the electrically excitable cortex under the present experimental conditions were as follows. The posterior limit was defined by the S. cruciatus. Lateral to this a small area giving movements of the tail extended onto the G. sigmoideus posterior, medial to the S. coronalis. The medial and ventromedial boundaries were sharply defined at the lateral margin of the olfactory bulb and by the S. presylvius. The lateral and ventro-lateral boundaries did not follow the superficial markings except along the posterior third of the S. coronalis, but extended well onto the G. suprasylvius anterior (see Fig. 2). In two of the animals movements of the hind limb were obtained by stimulation of high intensity (10 volts) from the posterior lip of the S. cruciatus. In no other case, however, were striped muscle responses evoked by superficial stimulation outside of the designated area. The motor cortex was readily subdivided, on the basis of the movements evoked, into the following areas: (1) tail area, (2) tail and hind-flipper area, (3) fore-flipper area, (4) neck area, and (5) face area. These are correspondingly numbered in Fig. 2. The tail area (1) was small and sharply defined. The only move- ment obtained consisted of elevation and deflection of the tail to the contralateral side. EXPLANATION OF FIGURES 3-6 Fies. 3, 4, 5 and 6. Microphotographs of portions of a sagittal section of the left hemisphere of seal 4 in the plane indicated by arrow A in Fig. 2. Thionine stain. 35. X 35. 3. Area gigantopyramidalis precentralis, caudal to the S. precruciatus (hind- flipper area). 4. Area gigantopyramidalis precentralis, rostral to the S. precruciatus (fore- flipper area). 5. Area frontalis agranularis, rostral to the S. presylvius. 6. Area granularis postcentralis, through the S. ansatus. SHPUONS Ola WINONA (COKINa.< Ols Islay Sis se, 595 ee Sent s i Set ow 6G sat NR Sa ae . - ae er oe “ee : ee SN re a OS ate FIGURES 3-6 596 DAVID McK. RIOCH The tail and hind-flipper area (2a, 2b, 2c) was the most extensive and merged into the fore-flipper area (3a, 3b) in a zone (2-3) from which various combinations of movements of the contralateral fore- flipper, both hind-flippers and tail were evoked. From 2a the tail and contralateral hind-flipper were elevated and deflected contra- laterally. From 26 and c similar movements resulted of the tail and both hind-flippers, together with ventral arching of the back so as to raise the hindquarters of the animal off the table. From 6 the flippers were flexed at the digits, whereas from c they were extended. The line of demarcation between the fore-flipper and neck areas was much sharper than that between the fore- and hind-flipper areas. Movements of the contralateral, but not of the ipsilateral, fore-flipper were obtained. From 3a there was extension of the digits, moderate extension at the elbow and adduction at the shoulder. From 30 there was flexion of the digits and elbow with abduction at the shoulder. The boundaries of the neck area (4) were moderately well defined. Movements of the shoulder girdle, drawing the shoulder forward, and of the neck, with deviation of the head to the contralateral side, were obtained either alone or in combination. The shoulder was localized posterior to the neck. The face area (5) was somewhat larger than the tail area, but much smaller than those for the flippers. Retraction of the contra- lateral angle of the mouth and closure of the contralateral eye, usually separately, occasionally together, were the only movements elicited. The boundaries between areas 3 and 4, and 4 and 5 could be defined within approximately 1 mm. in a single series of stimulations, but there was a shift of the boundary of as much as 2 mm. backwards or forwards in successive series of observations depending on whether the previous stimuli had been applied in front or behind respectively. Similarly, the predominating movements evoked from the zone 2-3 depended on whether the preceding stimulation had been of the hind- or of the fore-flipper areas. The most excitable area of the motor cortex was found in the central region of the hind-flipper area (the region of the largest giganto- pyramidal cells, see below). Here, in four of the seals, an intensity of 0.5 volts with a.c. stimulation evoked strong movements and in one case a single break-shock (Harvard inductorium, 7 cm.) resulted in a jerky contraction of the contralateral hind-flipper. Epileptiform after-discharge of varying duration was regularly elicited by a.c. stimulation of 3—5 volts for 10 to 20 seconds from any of the areas, but was most marked from areas 2 and 3. STUDY OF FRONTAL CORTEX OF THE SEAL 597 Rete ay Ot “se oa = tage * ot sles! pad os " Pais oe eS Moles. J — 2 . ; . ae oacas Fics. 7 and 8. Microphotographs of portions of sagittal sections of the left hemisphere of seal 4. Thionine stain. 35. 35. 7. Transition from the Area gigantopyramidalis precentralis (face area) to the Area frontalis agranularis, the lower border of the electrically excitable area in the plane of arrow B, Fig. 2. 8. Area gigantopyramidalis postcentralis and the transition to Area granularis postcentralis above, just rostral to the S. ansatus in the plane of arrow A, Fig. 2. 598 DAVID McK. RIOCH HISTOLOGICAL OBSERVATIONS The exposed cortex of the frontal lobes was well developed and the several cell layers clearly defined. The average thickness was approximately 2mm. In the depths of the sulci, however, it became much thinner, at the expense of the lower layers, and less well differen- tiated. This was particularly marked in the cruciate, precruciate and presylvian sulci. The histological areas conformed in general to the pattern and structure described by Smith (1935) in the dog. Their extent on the exposed surface of the frontal lobe is shown in Fig. 2. Certain charac- teristics of the classical six-cell layers of the several areas, as seen in sagittal and horizontal sections, may be summarized as below. Area gigantopyramidalis precentralis—(Figs. 3, 4, 7, 9 and 10.) Layer I was better developed than in the other areas, but Layer II was relatively thinner and less dense, with many small to medium- sized pyramidal cells. Layer III was very broad, with numerous medium-sized pyramidal cells. Layer IV, absent. Layer V _ con- tained typical giant pyramidal cells which tended to be arranged in groups between fiber bundles. This layer showed more variation than did the others in different parts of the area. The cells were largest in the upper, central region (Fig. 1) (the central portion of the hind-flipper area), becoming smaller toward the periphery (Figs. 7, 9, 10) Gncluding the tail, neck and face areas). There was a rather abrupt change from the posterior to the anterior lip of the precruciate sulcus (cf. Fig. 3 with Fig. 4), but elsewhere the transition was gradual. These cells were also considerably smaller around the sulci, particularly in the lower half of the cruciate sulcus, although a narrow band of gigantopyramidal cells again appeared in the posterior lip of this sulcus. Layer VI was broad, with numerous small, fusiform cells. The myeloarchitecture was prominent throughout the area and the cells of layers III, V and VI appeared to be grouped between radiating bundles of fibers. Around the bottoms of the sulci, however, the orientation of the cells seemed to be determined by U fibers. The boundary of the Area gigantopyramidalis precentralis and the EXPLANATION OF FIGURES 9 AND 10 Fics. 9 and 10. Microphotographs of portions of a horizontal section through the right hemisphere of the brain of seal 4 in the plane indicated by arrow C, Fig. 2. Thioninestain. 35yu. XX 35. '9. Transition from the Area gigantopyramidalis precentralis (right), to the Area frontalis agranularis (left), at the lateral margin of the Bulbus olfactorius (upper left). 10. Transition from the Area gigantopyramidalis precentralis (extreme left) through the Area gigantopyramidalis postcentralis (center) to the Area granularis postcentralis (extreme right) between the S. coronalis and the S. suprasylvius anterior. LOSS Nae ind CE as 8 SECURIT Oe : eae SE Ge nat Raat eS : WEES) ma eh : fo gece i Be Pipa ° Wd SL Cu sme A te a ~ : Pre eat FT Rg i Spies es eis iS aries 5 oe s . ae $F Se, fd ee a FIGURES 9 AND 10 599 600 DAVID McK. RIOCH Area frontalis agranularis was quite sharp (Figs. 7 and 9). The transition, however, to the Area gigantopyramidalis postcentralis (Fig. 8) and to the granular cortex of the G. suprasylvius anterior was very gradual (Fig. 10). Area frontalis agranularis.—(Figs. 5 and 7.) In comparison with the precentral gigantopyramidal cortex this area showed the following characteristics. Layer I was approximately half as thick. Layer II contained more cells, but with a smaller proportion of pyramidal forms. Layer III also showed an increase in number of cells. Layer IV was absent. Layer V varied in thickness at different regions and the cells varied in size from slightly smaller to slightly larger than those of layer III. Occasional cells contained heavily staining Nissl substance, but the majority were relatively pale. Layer VI was similar to that in the area gigantopyramidalis. The myeloarchitecture was less prominent, although in places the radiating arrangement of cells between fiber bundles was noticeable. This type of cortex extended over a wide area, including most of the G. proreus and the ventral portions of the G. genualis where a transition to a granular type of cortex occurred posteriorly. Anterior to this area, between it and the undifferentiated cortex of the basal ol- factory areas, was the Area frontalis granularis. The latter was poorly defined and small in extent. It was somewhat thicker, but the cell types were similar with the addition of the granular cells of layer IV. Area gigantopyramidalis postcentralis—(Figs. 8 and 10.) The following changes occurred through the transition zone between this and the Area gigantopyramidalis precentralis. Layer I was reduced to about one-third as thick. Layer II increased in density and breadth, most of the cells being rounded in form. Layer III was considerably reduced and the cells became smaller. Layer IV ap- peared as a narrow, ill-defined band of granular cells. The giant pyramidal cells of layer V persisted, but greatly reduced in numbers and size. Layer VI showed little change. The radiating bundles of fibers through the lower layers remained prominent. This area covered the greater part of the exposed surface of the G. sigmoideus posterior, the transition to the Area granularis post- centralis being sharply defined at the postcruciate and ansate sulci (Fig. 9). A similar, though narrower, zone extended round the lateral border of the Area gigantopyramidalis precentralis, between it and the granular cortex of the G. suprasylvius anterior (Fig. 10). Area granularis postcentralis.—(Figs. 6 and 8.) This area differed strikingly from those described above. Layer I was narrow. Layer II was broad and dense, the cells being mainly small and rounded. It appeared to fuse with layer IV, which contained similar cells. STUDY OF FRONTAL CORTEX OF THE SEAL 601 Layer III showed as a band of scattered, small to medium-sized, pyramidal cells amongst the granular cells at the junction of layers II and IV. Layer V was sharply defined, consisting chiefly of fibers, with a few cells of the small, pyramidal type. Layer VI was well developed and for the most part contained rounded cells with occa- sional fusiform and triangular elements. The decrease in thickness of this type of cortex round the base of the sulci was much less marked than in the other areas and was mostly due to diminution in layer VI. This area extended caudally over the G. lateralis and rostro- laterally over the G. suprasylvius anterior. In the latter location layer V was better developed than it was caudally. Laterally it was bounded by an area in the anterior wall of the S. suprasylvius anterior which showed a poorly developed layer III, a narrow, irregular layer IV, and sparsely scattered, but very large, pyramidal cells in layer V. Rostral to the Area gigantopyramidalis precentralis the transition to the Area frontalis agranularis was relatively sharp. DISCUSSION The present experiments were performed under a single form of anesthesia and on animals otherwise normal. Because of these limitations it is obvious that conclusions with regard to correlations between so-called electrical excitability and the cytoarchitecture of the different cortical areas are of little or no significance (cf. Tower, 1936; Rioch and Rosenblueth, 1935). It is of interest, however, that the lowest threshold was found in the region of the largest Betz cells; and the observation that a strong movement could be evoked by a single shock in this area is unusual in carnivores under dial anesthesia. The extent and differentiation of the Area gigantopyramidalis, on the basis of both the physiological and anatomical observations, compare favorably with these features in other carnivores as ob- served by the present author under similar conditions of anzesthesia (dog and cat) and as described in the literature (cf. Langworthy, 1928; Smith, 1933, 1935, etc.). In contrast with this high degree of develop- ment of the central representation stands the apparently simpler and less differentiated form of the extremities of the seal. It may be concluded that the extent of central representation is to be correlated with function and not with form. Further evidence for this hypothesis is furnished by the following considerations. In the terrestrial carnivores electrical stimulation of the cortex evokes movements of the extremities which resemble certain phases of the normal activity of the animal when initiating locomotion, feeding, seizing prey, etc. In the seal, however, the movements of the trunk and extremities 602 DAVID McK. RIOCH elicited from the cortex resembled parts of the normal swimming actions, which are finely and swiftly executed, and not the clumsy movements of locomotion of that animal on land. The rostral position of the gigantopyramidal area and the shallow- ness of the S. cruciatus may be due in part to the relatively small size of the olfactory brain, allowing more room for superficial develop- ment of the frontal lobes of the hemispheres. The cerebellum, particularly the lateral lobes, and the pons were found to be very large in the seal as compared with the dog and cat, resembling the development of these structures in other aquatic mammals (Langworthy, 1932, 1935). This is probably to be corre- lated with the wide extent of the Area frontalis agranularis. The teleological explanation which suggests itself is the necessity for par- ticularly fine codrdination of movements controlling the position of the animal in three-dimensional space. SUMMARY The exposed cortex of the frontal lobes of the left hemispheres of six young seals (Phoca vitulina) under dial anesthesia was stimulated electrically. Movements of the tail, both hind-flippers and the contralateral fore-flipper, neck and face were evoked. The cortical localization is charted in Fig. 2. Histological examination (thionine stain) of three frontal lobes revealed a well-developed cortex divisible into cytoarchitectural areas structurally similar to those in other carnivores, but differing in their extent and their relation to the sulci (Fig. 2). On the basis of a comparison of the seal with the terrestrial carni- vores it is concluded that the degree of differentiation of the cortical representation is to be correlated with function and not with form. BIBLIOGRAPHY Howe tt, A. B.,1928. Contribution tothe comparative anatomy of the eared and ear- less seals (genera Zalophusand Phoca). Proc.U.S. Nat. Museum, 73: Art. 15. LanGwortTHy, O. R., 1928. The area frontalis of the cerebral cortex of the cat, its minute structure and physiological evidence of its control of the postural reflex. Bull. Johns Hopkins Hospital, 42: 20. LanGwortHy, O. R., 1932. A description of the central nervous system of the porpoise (Tursiops truncatus). Jour. Compar. Neurol., 32: 437. LanGwortTsxy, O. R., 1935. Thebrain of the whalebone whale, Balznoptera physalus. Bull. Johns Hopkins Hospital, 57: 142. Riocu, D. McK., anp A. ROSENBLUETH, 1935. Inhibition from the cerebral cortex. Am. Jour. Physiol., 113: 663. SmiTH, W. K., 1933. Motor cortex of the bear (Ursus americanus). A physiologic and histologic study. Avch. Neurol. and Psychiat., 30: 14. SmitH, W. K., 1935. The extent and structure of the electrically excitable cerebral cortex in the frontal lobe of the dog. Jour. Compar. Newurol., 62: 421. TOWER, SARAH S., 1936. Extrapyramidal action from the cat’s cerebral cortex: motor and inhibitory. Braim, 59: 408. INDEX A BRAMOWITZ, A. A. The réle of the hypophyseal melanophore hor- mone in the chromatic physiology of Fundulus, 134. Amoeba, rate of locomotion, as affected by luminous intensity and adapta- tion to light, 126. ANDERSON, BERTIL GOTTFRID, H. LUMER and L. J. Zupancic, Jr. Growth and variability in Daphnia pulex, 444, Arbacia, localization of micromere-, skel- eton-and entoderm-forming material in unfertilized egg of, 295. , effect of salts of heavy metals on development, 401. Ascaris eggs, suppression of cleavage by ultracentrifuging, 99. Autosomal lethals in wild populations of Drosophila pseudoobscura, 542. Autotomy in brachyuran, Uca pugnax, 155. BACTERIAL content of clover nodule cells, 112. BALL, Eric G., AND C. CHESTER STOCK. The pH of sea water as measured with the glass electrode, 221. BAvER, Hans. See Tyler and Bauer, 164. Beams, H. W., AND R. L. Kinc. The suppression of cleavage in Ascaris eggs by ultracentrifuging, 99. Blood and body fluid, of turtle, colloid Osmotic pressure, nitrogen content and refractive index, 504. Boyp, WILLIAM C. Cross-reactivity of various hemocyanins with special reference to the blood proteins of the black widow spider, 181. BuTLER, MARGARET RUTH. The effect of its nitrogen content on the de- composition of the polysaccharide extract of Chondrus crispus, 143. ALANUS finmarchicus, seasonal pro- duction off Woods Hole, 464. CAMPBELL, MILDRED L., AND ABBy H. TURNER. Serum protein measure- 603 ments in the lower vertebrates. I. The colloid osmotic pressure, nitro- gen content, and refractive index of turtle serum’ and body fluid, 504. Carver, Gar L. Studies on productiv- ity and fertility of Drosophila mutants, 214. Centrifuging, determination of polarity by, in eggs of Fucus, 249. Cerebratulus lacteus, determination in early development of, 317. Cuace,F.A.,JR. See Welsh, Chace and Nunnemacher, 185. Cuitp, GeEorGE P. See Glaser and Child, 205. Chondrus crispus, effect of nitrogen con- tent on decomposition of polysac- charide extract, 143. Chromatic physiology of Fundulus, réle of hypophyseal melanophore hor- mone, 134. Chromatophore reactions in normal and albino paradise fish, 535. Ciancy, C. M. See Whitaker and Clancy, 552. CLARKE, GEORGE L., AND DONALD J. ZINN. Seasonal production of zo- oplankton off Woods Hole with spe- cial reference to Calanus fin- marchicus, 464. Cleavage and polar body extrusion in artificially activated eggs of Urechis caupo, 164. , suppression of, in Ascaris eggs by ultracentrifuging, 99. Clover nodule, bacterial and alleged mito- chondrial content of cells of, 112. Color pattern, genetics and histology, in normal and albino paradise fish, 527. CooNFIELD, B.R., AND A. GOLDIN. The problem of a _ physiological gradient in Mnemiopsis during re- generation, 197. Cortex, frontal, of seal, physiological and histological study of, 591. DALTON, H. CLarkK AND H. B. Goop- RicH. Chromatophore reactions in the normal and albino paradise fish, 535. 604 Daphnia pulex, growth and variability in, 444, Determination, in early development of Cerebratulus lacteus, 317. Differentiation, energetics of, VI, 261. Diurnal migration, deep water animals, 18520» Drosophila mutants, fertility and pro- ductivity, 214. pseudodbscura, autosomal lethals in wild populations of, 542. NDOCRINE feeding, relation to re- generation, growth and egg capsule production in Planaria, 227. Entoderm-forming material, localization of, in unfertilized egg of Arbacia, 295. Environmental factors, differential effect of, on Microbracon hebetor Say and its host, Ephestia kuhniella Zeller, 147, FERTILITY and productivity of Dro- sophila mutants, 214. Fry, Henry J. Studies of the mitotic figure. VI. Mid-bodies and their significance for the central body problem, 565. Fucus furcatus, effect of salinity on growth of eggs of, 552. Fundulus, chromatic physiology of, 134. , development of pituitary gland, 93. , histochemistry of ovary, 67. Fungi, saprophytic, occurrence in marine muds, 242. ENETICS and histology of color pattern in normal and albino para- dise fish, 527. GLASER, OTTO AND GEORGE P. CHILD. The hexoctahedron and growth, 205. Gotpin, A. See Coonfield and Goldin, 197. Go tpsmiTH, E.D. The relation of endo- crine feeding to regeneration, growth, and egg capsule production in Planaria maculata, 227. GoopricH, H. B., ann Maurice A. SmiTH. Genetics and histology of the color pattern in the normal and albino paradise fish, Macropodus opercularis L., 527. GoopricH, H.B. See Dalton and Good- rich, 535. Growth and the hexoctahedron, 205. INDEX and variability in Daphnia pulex, 444, , effect of salinity on, eggs of Fucus, ozs , relation of endocrine feeding in Planaria to, 227. GuTHRIE, Mary J. See Marza, Marza and Guthrie, 67. HEILBRUNN, L. V. AND Kart M. WILBUR. Stimulation and nuclear breakdown in the Nereis egg, 557. Hemocyanins, cross-reactivity of, with special reference to blood proteins of black widow spider, 181. HeERSHKOWITZ, S. G. See Sayles and Hershkowitz, 51. Hexoctahedron and growth, 205. Histochemistry of ovary of Fundu- lus heteroclitus and differentiating odécytes, 67. HoapLey, LeicH. Autotomy in the brachyuran, Uca pugnax, 155. Hormone, hypophyseal melanophore, role in chromatic physiology of Fundulus, 134. HO6rstTapiIus, SVEN. Experiments on determination in the early develop- ment of Cerebratulus lacteus, 317. —, . Investigations as to the localization of the micromere-skele- ton and entoderm-forming material in unfertilized egg of Arbacia, 295. Humason, W. D. See Tyler and Huma- son, 261. ENK, ROMAN. Sexual and asexual reproduction in Euplanaria tigris (Girard), 280. KetcHuM, Bostwick. See Smith and Ketchum, 421. Kine, R.L. See Beams and King, 99. Redfield, ETHALS, autosomal, in wild popula- tions of Drosophila pseudodbscura, 542. Light, adaptation to and rate of locomo- tion in Amoeba, 126. Litiick, LorisC. Seasonal studies of the phytoplankton off Woods Hole, Massachusetts, 488. Localization of micromere-, skeleton and entoderm-forming material in un- fertilized egg of Arbacia, 295. Locomotion and relation to luminous intensity and adaptation to light in Amoeba, 126. INDEX Lumer, H. See Anderson, Lumer and Zupancic, Jr., 444. ARINE Biological Laboratory, thirty-ninth report of, 1. Marza, V. D., EUGENIE V. MARZA AND Mary J.GuTHRIE. Histochemistry of the ovary of Fundulus heteroclitus with special reference to the differ- entiating odcytes, 67. Mast, S. O., AND NATHAN STAHLER. The relation between luminous intensity, adaptation to light, and rate of loco- motion in Amoeba proteus (Leidy), 126. MattTHEews, SAMUEL A. ‘The develop- ment of the pituitary gland in Fundulus, 93. Microbracon hebetor Say and _ host, Ephestia kithniella Zeller, differen- tial effect of environmental factors, 147. Micromere-forming material, localization of, in unfertilized egg of Arbacia, 295. Mid-bodies and their significance for the central body problem, 565. Migration, diurnal, of deep water ani- mals, 185. Mitter, E. Dewitt. A study of the bacterial and alleged mitochondrial content of the cells of the clover nodule, 112. Mnemiopsis, physiological gradient dur- ing regeneration, 197. Muscles, responses of, in squid to repeti- tive stimulation of giant nerve fibers, 23s Mutants, in Drosophila, fertility and pro- ductivity, 214. NEREIS egg, stimulation and nuclear breakdown in, 557. Nitrogen content, colloid osmotic pres- sure and refractive index of turtle serum and body fluid, 504. , effect on decomposition on polysaccharide extract of Chondrus crispus, 143. Nucleus, breakdown, and stimulation in Nereis egg, 557. NUNNEMACHER,R.F. See Welsh, Chace and Nunnemacher, 185. VARY, histochemistry of, Fundu- lus heteroclitus, and differentiat- ing odcytes, 67. 605 PARADISE fish, chromatophore reac- tions in normal and albino, 535. , genetics and histology of color pattern, in normal and albino, 527. Payne, NELLIE M. The differential effect of environmental factors upon Microbracon hebetor, Say (Hymen- - optera: Braconide) and its host, Ephestia kiihniella Zeller (Lepidop- tera: Pyralidz). III, 147. pH of sea water, measured with glass electrode, 221. Phosphorus, organic, cycle of, in Gulf of Maine, 421. Phytoplankton, seasonal studies of, off Woods Hole, Massachusetts, 488. Pituitary gland, development of, in Fundulus, 93. Placoid scale types and their distribution in Squalus acanthias, 51. Polar body extrusion and cleavage in artificially activated eggs of Urechis caupo, 164. Polarity, determination of, by centrifug- ing eggs of Fucus furcatus, 249. Productivity and fertility of Drosophila mutants, 214. Prosser, C. LADD, AND JOHN Z. YOUNG. Responses of muscles of the squid to repetitive stimulation of the giant nerve fibers, 237. REDFIELD, ALFRED C., Homer P. SMITH AND Bostwick KETCHUM. The cycle of organic phosphorus in the Gulf of Maine, 421. Regeneration, and physiological gradient during, in Mnemiopsis, 197. , relation of endocrine-feeding to, in Planaria, 227. Reproduction, sexual and asexual, in Euplanaria tigris, 280. Respiratory rates, comparative tempera- ture coefficients, unfertilized and fertilized eggs, 261. Rroco, Davin McK. A physiological and histological study of the frontal cortex of the seal, 591. SALINITY, effect on growth of eggs of Fucus furcatus, 552. Salts, of heavy metals, effect on develop- ment of sea urchin, 401. SAYLES, LEONARD P., AND S. G. HERSH- KowiTz. Placoid scale types and their distribution in Squalus acan- thias, 51. 606 Seal, physiological and histological study of frontal cortex of, 591. Serum protein measurements in marine teleosts and elasmobranchs, 511. Sea urchin, development, effect of salts of heavy metals on, 401. Sea water, pH, of measured with glass electrode, 221. Skeleton-forming material, localization of, in unfertilized egg of Arbacia, 295. SmitH, Homer P. See Redfield, Smith and Ketchum, 421. SmitH, Maurice A. See Goodrich and Smith, 527. SPARROW, F. K., Jk. The occurrence of saprophytic fungi in marine muds, 242. Squalus acanthias, placoid scale types and their distribution, 51. STAHLER, NATHAN. See Mast Stahler, 126. Stimulation and nuclear breakdown in Nereis egg, 557. , repetitive, of giant nerve fibers, responses of muscles in squid to, 237. Stock, C. CHESTER. See Ball and Stock, 221. STURTEVANT, A. H. Autosomal lethals in wild populations of Drosophila pseudoodbscura, 542. and ‘TEMPERATURE coefficients of res- piratory rates of unfertilized and fertilized eggs, 261. Thirty-ninth report of the Biological Laboratory, 1. TURNER, ABBY H. See Campbell and Turner, 5G4. ——— —. Serum protein measure- ments in the lower vertebrates. II. In marine teleosts and elasmo- branchs, 511. Turtle serum and body fluid, colloid os- motic pressure, nitrogen content and refractive index of, 504. Marine INDEX TYLER, ALBERT, AND HANS BAUER. Polar body extrusion and cleavage in artificially activated eggs of Urechis caupo, 164. TYLER, ALBERT, AND W. D. HuMAson. On the energetics of differentiation, VI. Comparison of the tempera- _ture coefficients of the respiratory rates of unfertilized and of fertilized eggs, 261. UA pugnax, autotomy in, 155. Ultracentrifuging, suppression of cleav- age in Ascaris eggs by, 99. Urechis caupo, artificially activated eggs of, polar body extrusion and cleav- age in, 164. V ARIAB ILITY and growth in Daphnia pulex, 444. WATERMAN, A.J. Effect of salts of heavy metals on development of the sea urchin, Arbacia punctulata, 401. WE sH, J. H., F. A. CHACE, JR., AND R. F. NUNNEMACHER. The diurnal migration of deep water animals, 185. Wauitaker, D. M. Determination of polarity by centrifuging eggs of Fucus furcatus, 249. » — —, AND C. M. Criancy. The effect of salinity upon the growth of eggs of Fucus furcatus, 552. Wivpur, Kart M. See Heilbrunn and Wilbur, 557. YOUNG, Joun Z. See Prosser and Young, 237. INN, Donatp J. See Clarke and Zinn, 464. Zooplankton, seasonal production of, off Woods Hole, 464. Zupancic, L. J., Jr. See Anderson, Lumer and Zupancic, 444. Experimental Biology Monographs eA unique cumulative reference library for biologists, providing full accounts of both the experimental and the theoretical aspects of all significant current research work in the various biological sciences. PUBLISHED: AND FORTHCOMING VOLUMES AUTONOMIC NEURO-EFFECTOR SYSTEMS By Watrer B. CANNON AND ArtuRO RosENBLUETH. $4.00 PHYTOHORMONES By F. W.. Went AND Kennets V. Tarmann. $4.00 PACEMAKERS IN RELATION TO ASPECTS OF BEHAVIOR By Hupson Hoacuanp. $3.00 NEUROEMBRYOLOGY, AN EXPERIMENTAL STUDY $3.75 ' THE EGGS OF MAMMALS By GreGcory Pincus. $3.75 THE RECEPTOR PROCESS IN - WISION By Seria Hecur GEOTROPISM: A STUDY OF DETERMINISM IN BEHAVIOR By W. J. CRozIER anp GREGORY PINCUS THE BIOLOGY AND CHEMISTRY OF OVARIAN HORMONES By Grorge W. CoRNER AND Wiuarp M. ALLEN BIOLOGICAL OXIDATIONS By EK. 8. Guzman Barron THE HYPOPHYSIS By J. B, Coup, D. L. THomson, AND H. SELYE By Samuet R. DETWILER. NITROGEN METABOLISM IN ANIMALS By Henry Borsoox TEMPERATURE CHARACTERISTICS By W. J. Crozier CYTO-GENETICS AND PLANT PHYLOGENY By Ernest B. Bascock BIOELECTRIC PHENOMENA IN PLANTS By L. R. Brinks PERIODICITY IN ANIMAL : BEHAVIOR By T. J. B. Srrer THE GENETIC ANALYSIS OF SPECIES DIFFERENCES By T. H. DospzHansky DYNAMICS OF INFLAMMATION: AN INQUIRY INTO THE MECH- ANISM OF INFECTIOUS ~ PROCESSES By Vaty MEmnxKIN The Macmillan Company 60 Fifth Avenue New York CONTENTS : Page WATERMAN, A. J. Effect of Salts of Heavy Metals on Development of the Sea Urchin; Arbacia:punctulata )y40 3) arse eek ee Se 401 REDFIELD, ALFRED C., HOMER P. SMITH, AND BOSTWICK KETCHUM The Cycle of Organic Phosphorus in the Gulf of Maine. .... 421 ANDERSON, BERTIL GOTTFRID, H. LUMER, AND L. J. ZUPANCIC, JR. Growth and Variability in Daphnia pulex................. 444 CLARKE, GEORGE L., AND DONALD J. ZINN Seasonal Production of Zooplankton off Woods Hole with special reference to Calanus finmarchicus................. 464 LILLICK, Lois C. Seasonal Studies of the Phytoplankton off Woods Hole, Massachusetts: a CenicS ty Wes SON rea 488 CAMPBELL, MILDRED L., AND ABBY H. TURNER Serum Protein Measurements in the Lower Vertebrates. I. The colloid osmotic pressure, nitrogen content, and re- fractive index of turtle serum and body fluid.............. 504 TURNER, ABBY H. Serum Protein Measurements in the Lower Vertebrates. II. In marine teleosts and elasmobranchs.....:.......... 511 GOODRICH, H. B., AND MAURICE A. SMITH Genetics and Histology of the Color Pattern in the Normal and Albino Paradise Fish, Macropodus opercularis L....... 527 DALTON, H. CLARK, AND H. B. GOODRICH Chromatophore Reactions in the Normal and Albino Paradise 1 5 Ce arg MARE GAS Sam ANAC ay aia are ete | o\ 535 STURTEVANT, A. H. Autosomal Lethals in Wild Populations of Drosophila pseudo- ray eS) by Ob of BS CON gee MN Op eR ee ce me Vaan ME Ocala Sic Gacwa on. 542 WHITAKER, D. M., AND C. M. CLANCY The Effect of Salinity upon the Growth of Eggs of Fucus PUT CAS 5. Si car Pie ba Shy aoe Gy ee SMa Peg eee 552 HEILBRUNN, L. V., AND KARL M. WILBUR Stimulation and Nuclear Breakdown in the Nereis Egg..... 557 FRY, HENRY J. Studies of the Mitotic Figure. VI. Mid-bodies and their significance for the central body problem................. 565 RIOcH, DAVID McK. A Physiological and Fiuctalosicnl Study of the Frontal Cotter of the’ Seale a res et a eee 591 ya) is wate en Van aN ~{7 = NAA >V VAN nN g AA Aa SaAWAN Anns agen ARR RAR ARE AAAA AA tess Aa AAAAAA AAR [ \y aRANARA AAA Anna sAAARAARIR DB 3 oN NN al aol WAN naa Mra AR AAA ms SA, AAAA, AAA @Aary NAAAAY AAAAA \- = 2 Aa pacer\- 2 aa ARARARERRARARA AABN i AAAAAAGIAS - cass aaceenoaaannnce y $3 2 PANAAAAAAAAAAAae Ai | OOO AARAAARCE CTY YA, -e Ana 2 py» | Na RAREARRARR AA ARRR RARERARR ARRAY AY ANA AR : A | AAAnaal|.|\eleae\\\\ AAAAas pas VAAAAAAAANAA AR AAA Bs, VAY AAA AN ARARAARAARAAAAAAmAn AAPAAAAARAARARAAAA AAAAA AA ARARA BARBARA AeA NAAN Aaa eases pa soeternacnnes AAA, AAA AARAAAAS Anas Awan penn alaalenN AA \" | As oauctet at atiaee AAPAPAARARAAMMAG STA aAAAAAAAAARWA\, y B/N! v la ye Is pit AARA ARAL ARARAAR wp ~ Aap panei Me BRIA ‘wiitiiiiin 088 01228 1564