eee en ie As a ie haat el mee ad ta eth hh PAARL sl Ms bee, Wh RL Adee Bd! AD Thiet sleek. Md by roid Poe me ce ee ry ne 4.5 CRC A Poe es Ikra | Cle Wr ir, eet er Ce ec cer Cl CC er a ten aren oe ted eb ya FE ee knw ae eA RAL AU ce ee SP a HM eg eee be wen BU tee RR Hr bit Fok mn Ash om Hop ob ei ie ie CC) OCC CO WI ete dey WM qe wel bay ea Parr ain Cc en a ee ee OT Sc Oe WT a Ao ET a oe ce ae ea ee oe ee Henn DEL AUMM Atm LL Bia Tl ae Sy WR Ne Ly eb delle ERI AEE Me ete arb BAT WAPATO tart ee he ate age ea Sener tee tyes Cb ew peer peewee CL Le oe, ek I oe ee er) Ee PS a rr oe rll ce tC ST AT Bnet 1 Ration be Ve Cen Wy mebat POC eev co naanpe ay er Oy Se eT CAs Cs CC Fb ian dre, eat Meena ah undhe ates yeu DE CL ALA ae ee Ne WSC SCAG CL TA Pe ir Ye Cube dan eee eerie baanta SL Oe A ee Sa ST NT Re TI TEE LE ee Mint ee tne ae fap brite Ue et mt Oia hed # ab. e ok a AON AW a Peo eee tas ee tM CER EM MUR TNR AEG MAD Ah kt Gey ark yt He gg Dusegsan sent abe bel APT ed thm ALE HW TAK GBR na ahedCood Des Be Qs tts bor nerfs OHI. AB ALAN eo eli ita boe a ee A A OC a aC a a a eran MT ert fo (Day OOTP ee he ae He be Voie Daten as HSN Er ke Bak oN A ADA te AA at AP HL op eee oa ks Raebeds Wik WE AAPA MeL ibelba Rie dinate Ramee a ibe libs eeu ane CC Ca Vb ARGU ALR CAC a CTT OST OT iE eee big COMMA DALA WA DDD gd ek 4 ig dee Cee Co OE DUC OC UC Te Arte er er ee ey Cn ee LR a OT AT Ve NT i BN POO Pa eRe Ne Me ny Chena ae Ce ee v4 Pare ia cr Vari Vi te eae See ea a a ACL ST CeO eT TN eS A Kab t PPE OO ee bea dete PA eta eb Pee Bat Cn Cee a ee AA ey NR HE oe ee a Bee MEE RE TA Ter (aot tbs et UME tke fet ety Oy Hoa eRe rar er arc’ We Py Wer at We Ma Pie ce Ce Pr QD ee CO Un Ve Wr Me VY Sa aT LA ee ce ittc won ts Mer ek Be are AO Ot Pa te te OPE OD he ee ee be Pe bee PO eer eg OE Re PD Wy eee ee bb re pg ag wy WOE dele Fe EE BH OR ree as egal “ tt We ea th te ree eet Deep hew ee hee ea ee hee ep ee ‘ Ce A ee oo) ere ee en ae Oy mele arate at Aa er PP ee eee ORE ee by bee ket “ PE UTAH Oey eg een CO eR CT Cote wa owe Pear ran Ser arr ay YT PT Ys SY Te ' oe a HE ee be ag pe ey Ce We a ea Benen: oye b ive ane m4 toned COT rer ar fir OS PPS eae SOT CT a Te ee CE ed he ee a rence ' CERO Re DCE ep Fa Sep bani eee aL Ae SEA CCN Oe OW PW SCRA ee a vbw een peat PU ea PUP A ee eV a wg Ce SP ee WT TCC re eC OTe er Pee her Le ee eh ete hee te eb et Ce Ce a et Wor ary ee GD EG Ror kop elk ini wea 7 ed Ped apa yeh ta ee ha yee gaia PA Oe ae he A ee RE CO ee oe CR TA IR ONY Uk CC es Or yee ee eT ae A eee ae en ree NT en ae 4 shan Pv ee Pe eee nye FOS Oe NH Ce y me ey Mom a kat Gute Ce feb bg ee tb pk eed PI pe pe Be Tes at POL Se ae pyopen SOS UL A OO CC ee Te att PhS vet Oey bet ky OY Sa Cae | ver ee PL CR OC OT PS TT WS va PV VEOV eb rt apap ea Cie ete ee a oh Wha een PN eke ei pe fee Ce ek Ea Ae Coke MRCP Ha’ Peavey bee ae vee Craw er cc Ce ce Wa (ec nC er Pe I ka Gee ey ee ASR Dey gee Yeni) Wilh Mise ae 9 ons HAGA UTP § ETN NaC tbs te tbeie to ihe ENGI RON ean ALVA. divers Eales BUR abs ABLU Ca Dk 0 0 hy pAlb Wire aL suth cP hick ap A yet hey EEA SO Gps Py Pid bone cl tr DE St eben PG Cb ea i Eke Oe Woke iek & er ea PATV Et Ves pe eo aha kd by ep ae ey POO EN ie bby wetnee cael ok eon eid mals Mra Meare Bs tA A ca Wa teat Bets He ALNW ie esl¥s obs th Wyiy nrg sh COA a Te ra ' paneer Vey eee Morin et Perea ere eC PC ek Pad eon ey aa Shien CV EM HEE bbe a ka be el Eby hp eb CO ie CW OO abel © AID Ahh abe BND Fess ga CU Nk ge AED, SL ELAR ALE AR Pik eke VIE OLD Aa GA Capen We epee. app SU eer Ce TEC YOON OAC GE TTP WaT Mesh em lay il obs sha ae be wabee ’ ' ‘ PERO OE Ce bee PE a CR A PL PD Pw WU ALT Wet Bie Rely oe te Bi kegee igh Muh oe rem A Teknik ob Shahn rarer ‘ ra CCA ee se Re PA A EO RTE foie hdc bl LP OA Ohne hear hey a Ribs Were eres ones ite attr verre CeO TOE Ee OT UT pee EO RE Ey UA TAL Opa op em bee DACA! UAE ig WA phereaheten gy fury SE 66 Qipeap api eccheap Ee eWD ete Wualbs Gade tevpe ie We fhesocw at MUb oho Lea es Welt he Cao Caren vera ar aa | CT Ee Lv me gee SOS eT a Me ne Ya % py Veuve enya Ce et Se ee We eS EY a yr Io LR NRO he We mt Re RT ete Me tG Me te Re Lith kay tee A i be : . AP A be Re Le AR VUE ee ED Ea ee Ry EN PO ON A ay ye es we Cag eet PME Clg pre Th, RAE ETE May genie aie han fh Sebi tek yee ge SE Ee ee EE EAD ge ne OR OP HEE iets Riv ab Boyt Men Tp a edema UAW web tae COV ee beh be eee ee td ee ROR AD Teta oe ae Ok eT) PU LIC AA Ae OU eS hit Wi teae Vea te gO OS meri ee aa ere On Bee wae Tree Ws Wd ee PE Mara COR Dike baw En are een ee So ie Wot be ih Pete eee eee Ee ea EE ew cas POL RR eB OE Tate Ble eR RA Neb Poneto shane CS ras ger We OC a OW a a OO , bow Pe wd wh Pee Ee Vp aibes NE oH ea Bea eo LR yet heap eH Ty Ce Monee MNS IB INEM he bak Be Pe MAE ohh AsVoy tthe my ve EAP LO ee RNS Rigen a ee ee i ee er Re STOOPS CW RC Bea eT SR Hs Is Wh fits ty a Phad te pa foes eet vrs tee ee : Ct ee eS Ora Ae, ee Wt . VOLT HBR he aes a PPL AAT PA ea es LOATH DP ee ae vipw ge Pee yg HR a ee ms er) ON Bede G yg tbe ahi PHO RR Vt ke, whe tae eats Pee Mn ts Pwr ete : es hen VPP Mbt ge COS er rere eS WC CTC Se pare omran POC War re et Crave eT pad hep fe pty ee pty ena yran fee Ue sate FE re EO KI emp AE eH lett 18 BOP CM RAEN Ae OA p scat Vole ene Verne ee yey pee rene ne ta Seep ree we Se UR Ce A Te RP vues Pete rey eva Wade eee tegen ge os . ‘ Sees STi Ae eS a rae at ee Wei e ane CLOUD eRe h ead VP ele ge be Boag e oe hens a A i Ce Wy a DR Ye Th CO) Yee ee ty CeCe a oe ee Veen wen ‘ CUP RU pete ee ke Hee Mee Lg Pt Py EN hd Le CEU me Vy iy dea Phobia ma eee! be au eee rs CRT CeCe ie ore Merny ; Aig éea td Mine bt EMA © Dee eb vA PP yt WD RT A ete EMC Hey PW ged Me ah MSRTE TOD Gey AA yale yey in Ub Sb sTaibelbe hice tac ted Nee ioe ese UMoagle Toe bonn te ait fury % Wy Qrk, Arey RN Se rarest vy Re ee eg We re WT BSC Cart WE Se er a CC De Aer MARL a SS CO Soe N Tee Sc nate te ee a an ere Oe a seas ‘ ve ‘ Pk em EE ke eR Ee PR pn PP Ee Mg GE EE HE TRG Vt ere tae whee La Ae PE ee A een eran nT ape eR oe ae Pam eee eh kr Ra Ys I ace rye VEVVEH EVE ECE CH Lee kde te hy wae ENN Re TAN te a ete dete Qe, Menu bes pele gibete fy @ ' ere aiy oy ean Sey ee ee ay Vivesyevae DC a TC RT a CORT AE hee it Yorn gee rapt PH Co gg ti el ty eeyanen ever tee pert, Pay wie ai} tos cary Voce eee ae gy Ee 1 ie) Hie peter ee Oe ai ee he ree ee eae Bak pe enh Soar on vedes og iy we Cee ee AN ge ee Cn te Peet fot Tesh BB ay te Ve Wop pe Bere vita Held Mp ey |e PHP P OY Ath eb Wey bebe se te ue eon) ve CI Tie Pee wh hy eee Eee be a Beg Pens We ck Wrata se Srtk tein y A won EC OO TU YO Sk TRY fq Win asl bee ey Op thee Be Web on be Veen a ‘ ei ot Nut uigT4 ote tenn Sh pee yy EN ke rig yy le way ‘ WA op ge bie obe da re tot we Mop AU ea Vor: KO aR gM Th Myles brits by Ean, Wwe Au CAH KMS HG Ae ete dee pct ALA OWNe Deeb TD ey Ce eC Oe Care eM eee a Mra Ss SS OT a Cyne ee yg Ofgn y ater weyebae Nee or on ie gp ys reste ad SOM PM Pe ep Roheb bw hte ER ee ei aa eg TWEE te MODAN et agg ey Ly eden PON PE DEO EE EE a pe ey ht bee a pt be sip Merny p SP rea eet an 4 1 WW BL te ed kW, CR RAGE MCG TEAL Ty TALI eA eA ee rey Veo pete bya epee RM SVE MTL ed NN bees fie NONE Se We dled pe My chete) Ela AG Wa bate Rett: Wom ANA mL AAAS BAG AS BLN Iho tbs 7 s. ‘ Cae aR arte eT cae Sr ae RR no Ca ee Os MYO PS a Pear Ok CITT ORs a On MS SC OY OTTO TR Dare ee is roi we haa eee Oe OA a GN veypeenge DEM UP VEY Wey ed wy ERM APR gy Syme eh pip Od) Whe Ce ee ee Oe wes hone wey rare pe es Poe eee vay Cir era era OS Be SA ic ier See Pe ig ek ¥ REDE eee a eas oe aren OV a yee yee ee da Pee eae i te Ke ee Cp ew eee y eth Che lyse weed Pye ke Hee! PN SUG AN ayy ag m4 ’ ' . PEhee ve pe hp essa ee pa eye eae Vee Pee eh Oe ee Sab wae TR ek Se ey ten ty sy hanes fey Owe ka Uae Ce rene ee Sep bw en Cee ahha Hie WA ee ere sre vetghey Pee yuneta CPE ND ey Say eee pe POY ey ie wy eye WE eel ee Fe Bas ' ih ee ar ie a ’ viecbaay . veep pe eee vey Pe Re ee i i ry Wim ae bee pe a reas Cpe ee Pee ENE PE Ot yp pM bee hype bays PAVE oe A OED OR a 8 gy lh py Deau mB, y Mh eta Aiptibl yb Crean . ' , ‘ peo vote yee te Pe Pee ee Pee eh ey PR EV ES AM le ie Tide bbe holy AVATAR ROR git iw ' An os oN a8 Aap vise ee NW Ey OLN 8 wep Gee AA Le Mp Rep pie ye eae eh F ¥ treed ' ve vous Pe ee ye ey yh Pe LAMY A Sop ea ecy bee ne %, aren arets ) seb vr gen Ls re Ea ee} shy wet ony een preys aie ne Ving - TMP I AS Rae fev oe cbs ‘ eres ron PCr ph teva ya hy yy tpl ee PV VE A eae wine ay ‘ “ AT EM AYE wy : ' Nea yeh a ae ey Ce Oe ea a ee ‘ Oe Ny MRA Ek WY pre go eve y 4 6 Fe hen gee Vere pda wey beget ee de ee ey a Vebmeealy ‘ on WS Litegit ae ok F : el ‘ Vmeets "a pee oye VE Spe ee eta ae ele py Ae Var gy hig at ars a F ean TeV e a) pg aes Pe Pee et ye ee eee pee ES ' when's eden de setae ais s bg) vit v4 eye yea 4 say Vee Eee L ee Tae bee ke thee peel iy SAT it Seah ae ae oe eee 1 a Moat pre Vere wees GRU NE UE ps A CPL Ts Wh a8 a Sry iran Ree aA ’ ‘ 4 Ly . . pyetege PAR ge ' Youd Le diya ys ve hy A rev hag vy Liew he eat peden gfe a fig path ria a ja wh yts : rope pp ye Veg Poca ee tee ae 1 SCO ape ipa ig vehi 8 ’ eperges on ’ ' . se yee . ve . eyo hee eee ’ wha : tia ’ ” ed el oe ree Wag noe j ‘ , Lpeb ete y : eye yet ewe a eel ey tye seyhas eva Me ry Pree rs 8 So er) OUT ai ray epeee vee . tree rye Cee ee hea ay pee yey ek eee gd ‘ ' , Cos vata s ae vey ts tan yauce ‘ : . by ‘ yay a 4 \ Cee yay : vs veg ' pope ny pipe eee Met eye ve } ' rey . ve ye p : ma sayy veyed ra 7 yy Pelsatey ‘ phoyeeg'e ‘ pvAv : . versed why ; oy yyy rerypegy ue ; Rah ee are ety ' Pid eps y haty es Ya te Pe 4 H vfs } ta a } a3 ' ‘ : : : ' ) wo aa bud et ‘ i i i tidied i ‘ ; ioe va i ; . : ' ° int oa ted aay , i . i ow Cr ee . no ; ‘ ; oid De ei Peas Sak ba ‘ Pine SVG faba ii he dab (oe Pe te ca de tite : “ Pi bane RE a pa dad at i fide Cee Re is ee cae A SU Cnn ‘ et a Ga sha Raed A i hice ha bal aa Oe Be tae st bed abt i arene be Wr} dptel 2 Yok i Pe tee ; rededebig wndidd he Jy duadiay ee nan sig laa iE trsia beens i eee ea ae fae pebue ‘ a) OEY dap od abl wa a a ek U Sr ee ras ita bead Vit abd Cae ee Ce te tr ae Neri Be Wears at 6 tate bod Dibba date er whe oh Ba Ge nd hd $i ef Rete bad Phat Ue eee i ar ay , See a ee Vo katid Won ae H ha robe wae bite Pa A ce a Dee ob ; “ PAC La We a are Arar Meee Scared Oe) Bibra walk Qs . ' : H H Hiner ie Ne iow tie pret Vip eda ce te ee ae . ree btn’ ‘ ine tiga 4 FIND heed ta Pa a Vai hash Pb Geis bob’ Le tlibaiag Ped a eh ade ewe bd oe Cc ee a DS badd Wad due eae Ne bksahy, hayes yh ad rae DA Ta tides bet 4 Poselaege 320 UF hea PN Nb bdtiebed Pa kgs ya ww deer hide Dew Pia Be te is eae Pare er gee eee a ar Barts Blre if ae ta o foe . PEP to) Jk a eked Cen) A eb tte Me ee ae Coen CY ay ee Pe a Lag La obeys das ine wide is Le Hd ge meme ye ye AG wie wt ba cate e ett net rt Meare ae tree De eee rae Cee are aedave wpe tes ca Oe a ee a al He eid Maat men awa bard eel Taged ‘ vel PC ee rE dee ed Be he iow i yeetbood a He gw We die Fegan det iaty kai hed ern . pb apy 9 tel a he ae ope ad Fapi ek lg regen the et tee OR De Bae aged aye denne ts eae ond We 8 oe ee Ce ee ee eT, ee eee bebe bode denotes Se A a ce Ce ee CO Bee ater Sent ic lbel EAM ete ode sae eh Sagi Gag oy ape Beam nda Db ae be ah hee SSC CT eT er reer gh beth 16 ah : WA Be dP lbibady we aaet POR PRL MR YE Me NLR PICA ne Pore a 3 U8) Cap Pe vb Hv be bg at arg EL Dp be ae bid pehev aed fob toad SO rae Cyto eV Har ke ad Be A oa Cee en er A | Vet Soak a eee ael dpe res peiL sb ABIE, ap baNDE eueelPoake de Pants Pet vetie as Pees Mt Gad cee Ber Ge esi oe TU a eT RT Peek de Fi Seo mtd tae Calli, de ded Begse so koa et ae Ua Pane geek tye ba hnap apa ih ae aQgAt pny ome ne OE eG Carpe ee ar ar are gm Sata Ve er Pee an heath Fa MME HRS cigritenarty ping the f Pe wee IT OR | URC CMC Mert a er ec Bd cob bur yo ye IMIG Pa sete as rea ee) eee ar UUme ae aCe ee Be yo ee) Be OT Ee mE ee bee bee ate Peek a aber wd Madi ih ot aa Ae oe erred a EG ate et never bide Gpl “ a a) hme wk € ioe eee aetksthe lo ee we a LC ee ee ee ee Ce ee She “- Me ” sie Le ee 8 aed er Berm dee o Whe meaty Ce Oe eC) Cal Wee Ue oT ee ROE PA OOM OL ee dee etd hob ad “ * FE wel mene re ee ; See ie rt Perc Be hide wk ff Mar te are tates prearrerT pyre et tonkyad Gabba bab’ i wk Fit * Meta thuregiie key eA deli Ce ae a ee BO te Ram eee alle Pe A ott Wom i teeny hist Cb ee ee he ee bee Sab a OT Oa hh ga LP EO De idiphdd at ds edb a Aegbil adept al Cay WEG haste age Dabok dak weed ines ipeend trad au Pere renin wtey vt ia ct bE 8 eh Be MR i ee oh Nh wa aed CCC An SR se ee a) Cate One eee iene HO EMP ap eat ar wees ' . Pie ean A is Dr er a a ibeadve Pa (wee PE rn bgp ee te Se TF MSRIE gy ebnae ne i Wipe de eg b beta, ora te bo ve wee aa ae rte Se | iva PR Ve badd eR OM teak goby peat HO EAD avd ga Gbethy dE britete Bog Uk de ie dante eae emit Batre. 1d oo TORO E MMe yards s Cet de ae een re RM EE EAL Ae he te hte me Fab ime Cod on cette Mane ad ha Dope att an tnd bese yas rele I ak oa Giger ped ever ats Pert Pe Net a a | thot eit po er HR Bre biti bdde Wok WED Aa be Beare ye ceeds Gfsig yt oe a ne Creer at Bebe Peowse ne fo ee PED tek Pee eda eg CO oe rg Phat ee deh? Fe bene ai Pap qea 2 pares ably Vike Co rT HOD be Eh a Pe gee ot cee iue set eae Via Ghd ede Pa the lO et Senda PE we Td ea ee dG aE Een dae doa ed CU ‘ (ere al web eoter bb tage POE EE et ab ae CE EEO RRC a ae a UO OR me BE ite ti Tee Jee dey isi ec Fae CO esa er eo ‘ feed Pe Ee Rd Sor en ee De ee gabe ee ee o eet wt heb om bn ba be nowt ve era ede fra ert Oe MRL RUke aT ett TT Ee eh ee ote , te re brred O64 wes ‘ dette OR Oe EC ee sew wa uel Par aeit Beh Sanna hPiiem githhiad We a de ated CP OE MEE I GE ee a tere ADE Tb tele tdi poe ta 1) de Oe ee ec ae erat ba Oe UR Fa Otel 6 Re Ra tet a ied atidt, Corie et oe de id ous QR wor eee Pur Woe Bac Ga A Par Ta | oe ‘prorbaie . Te OC Ce eC Tf ied hohe ot A CT baa Ao (aaa bage Cad ta Dano ‘ Ee Re OLR ad Ree nak Deeks ood aed Wore ned bath a iad eM eth opin bane, Man Pr de Gey Dd cond a Seb Salleh aide bok tn dpe G Ua! EIT a veh Mote ea ear ee fai a ae i er Otbetivayet CRC ey Ser CY IRR OL “|! PLL CT aa Oe Oe a ST AT re eT eee Me aif oa ‘ Abe ee ah e Dt tak bad SL ce eC aa POE TD ie EI ES Bed ed eid corrara er) Cana bewas Peenene ee D dab uid ied oO Ae Lobedyad de Ole (CHL RAIA CO a ee inden, os ChE ME ak whea Se ee ee ee eR che dite SUED Avil A aks aD ghetto eid Tonih Oe _ BIOLOGICAL BULLETIN Marine Biological Laboratory WOODS HOLE, MASS. a ceanian Instiige SX i. \ A ~ JUN 30 1915 Editorial Statt Na as E. G. Conxiin—Princeton University. Jacques Lors— The Rockefeller Institute for Medical Research, Gerorce T. Moore— Zhe Missouri Botanical Garden. , : T. H. Morcan— Columbia University. W. M. WxHeEeELer—Aarvard University. E. B. Wirtson— Columbia Universzty. Managing Loitor Frank R. Liririe— Zhe University of Chicago. VoLUME XXVIII. WOODS HOLE, MASS. ‘JANUARY TO JUNE 1015 TT P THE NEW ERA ‘LAN CONTENTS OF VOLUME XXVIII. No. 1.. JANUARY, IQI5. (usta hs) nitvaronio; Development in Nereis. 0)... . LILLIE, FRANK R. Sperm Agglutination and Fertilization...:.. Wireman, H. L. Observations on the Spermatogenesis of the Gall- LUM, [DSO ARO GPU REGO (MAGI) <3 ea a A a a cle SweEzy, OLive. Egg Albumen as a Culture Medium for Chick MRISSUUCIS Sooke: SOU e ee CBee .a Gn 8 AS eet a i gia ga came a WERBER, E. I. The Influence of Products of Pathologic Meta- bolismion the Developine NeleostOuwny ee No. 2. FEBRUARY, I9Q15. Lores, Leo. An Early Stage of an Experimentally Produced Intrauterine Pregnancy and the Spontaneous Parthenogenesis Oihe Hoos mmuike Quary oj the Guinea Pigs) sean. GARREY, WALTER E. Some Cryoscopic and Osmotic Data...... LoEB, JACQUES. Concerning Brachet’s Ideas of the Réle of Mem- ORAM eR OLMGAMOM AU NHCHEN ZOOM seme ey tae a ee Just, E.E. An Experimental Analysis of Fertilization in Platy- ORCUS GGL OLDS je oe Sie pee a Nea rN OMEE LIne Oa eke tire Sk Maks Nel LINDON ODWiNe ES POKoGysts aman Ajnelid ses. 4.0 alee No. 3. Marcu, Io15. SUTHERLAND, G. F. Nuclear Changes in the Regenerating Spinal Cord of the Tadpole of Rana clamitans................... Ricuarps, A., AND Woopwarp, A. E. Note on the Effect of X- IROCVOOD: Ol, TG RIMTAUES oe hs ols) PAB ols os ee Bl GLASER, OrTo. Can a Single Spermatozoén Initiate Development hip. LAUTAN OE RON se Se, AY eh LA REE PE Shs OR an LD Curtis, MAyNiE R. Studies on the Physiology of Reproduction UO HUE IDO GORE IPODS” DLA G ENE ian ou a We eee No. 4. APRIL, IQ15. Wootsey, CARRIE I. Linkage of Chromosomes Correlated with Reduction in Numbers Among the Species of a Genus, also Wahine nS pectesroj. the Locustid@s: 28s. el: 111 140 149 154 iy 1V CONTENTS. - SHULL, A. FRANKLIN. Periodicity in the Production of Males in LSC HOR S.”’ Cre D es GIS bo oo Go oc aoc Linton, Epwin. Note on Trematode Sporocysts and Cercarieé in Marine Mollusks of the Woods Hole Region............... Purers, C.F. An Experimental Study of the Behavior of Amph- pods with Respect to Light Intensity Direction of Rays and Metabolism..... At eR One Ma TA ye ny Ui aba She dscns: Ewine, H.E.. A ‘Case of Persistent Melonisi ae ae ae eee LILLIE, PRANK IR: Siudves of Fertilization ee No. 5. May, 1915. Moore, ARTHUR R. On the Rhythmical Susceptibility of Develop- ang Sea Urchin Eggs to Hypertonic Sea Water..........-.% LILLIE, RALPH S. On the Conditions of Activation of Unfertilized Starfish Eggs under the Influence of High Temperatures and Fatty Actd (SOuUutows oc. a See oe BupincTon, R. A., AND HARVEY, H. F. Duivision Rate in Ciliate Protozoa as I eens by Thyroid Constituents . SHELFORD, V. E., AND Powers, E. B. An Experimental Study of the Movements of Herring and Other Marine Fishes......... No. 6. JUNE, 1915. Seventeenth Annual Report of the Marine Biological Laboratory. . Haraitt, Cuas. W. Regenerative Potencies of Dissociated Cells Of FIV CPOMCEUSC. 1 oxy. Wie Lhe se ee eee SPURGEON, CHARLES H. The Eyes of Cambarus setosus and Cam- barus DEUUWCIdUs oooh hos Ck Ss CE eee Morcutis, S., Hower, PAaut E., anD Hawk, P. B. Studies on Liussues of Fasting Animals... .s55)) hee ee ee McInpoo, N. E. The Olfactory Sense of Coleoptera... ....-.%-: 187 Shah - Aarine sBioloatcal aboratory : ee woons HOLE, MASS. ‘Laue, Frank R. - Sperm Aaglatination and ee 18 ae ‘Observations. Om the Spermatogenesis of the | Gall Py Dr ee erinacet i Mayr he : Swezy, oe ee Albumen as: a Culterre Medium efor ; a i Cndee Tissue. 7 NM Ee See 4 an Wenner, B i The Influence ae Pee of Pathologic. al stg Riri oe r < Awe Ct Metabolism on es Coe? Teleost ne ea a Te ion = SR ors pins vars MonTHLY BY THE: “MARINE. BIOLOGICAL LABORATORY. ie: PRINTED. AND ISSUED BY “THE NEW ERA PRINTING COMPANY _ LANCASTER, PAW " yn ie p > A ve y8 Rea te et es Bis anus \ ee ? Po ae “AGENT. FOR Grear BRITAIN SRT NU Acext FOR ‘GenwAny WILLIAM WESLEY Aone cm R. PRIEDLANDER es - ee Essex Shes, Strand he ‘s Sa : Bei: Ny W. ae a Bye ee eet London W. a Recreate ee yep Carlstrasse, Be i Oe De soem ce Gents. Fer Volurie (6 ca Ske oe By “Entered October 10, 1902, at Lanéaster, Pa., as sec ond-class ‘matter, ett eh Oh “under « Act ‘of Foueres: of aay 16, 1834. ; Vol. XXVIII. January, Tors. No. 1 SIOLOGICAL BULLETIN INITIATION OF DEVELOPMENT IN NEREIS. D, 13, US. (From the Marine Biological Laboratory, Woods Hole, Mass., and the Physiological Laboratory, Howard University School of Medicine, Washington, D. (E>) If any apology be needed for merely adding to the long list of eggs susceptible to agents of artificial parthenogenesis it may be suggested that initiation of development in annelids possesses some interest since annelid ova respond only with difficulty to agents that induce development.1 We need but recall the case of Chetopterus (Loeb, ’orb, Lillie ’o2, Allyn), of Amphitrite (Loeb, ’o1a, Scott), of Podarke (Treadwell), of the Pacific Nereis (Loeb, ’13), of Polynoe (Loeb, ’07, ’08) and of Nereis limbata (Fischer, Lillie, 11, Loeb, 120, ’13)? to justify this statement. In all these eggs differentiation without cleavage is far easier to obtain than development closely simulating the normal. Among diverse agents few only will give cleavage in Chetopterus (Allyn, Loeb and Wasteneys). The case of Thalassema stands almost alone among annelids in giving development which is, according to Lefevre, to a surprising degree like the normal. Another instance among annelids of differentiation with cleavage arti- ficially induced is worthy of note. This report, however, on the initiation of development in Nereis by heat has, I think, special significance. The results here offered require an interpretation which concerns the funda- mental theory of parthenogenesis and fertilization. EXPERIMENTAL. Certain preliminary experiments date from 1913. The ex- periments given here were performed during June, July, and 1 Bullot claims to have produced normal development in Ophelia with artificial means. * Loeb’s experiments with Nereis (’12) were apparently incomplete. I 2 BAe. JUST: August 1914, at the Marine Biological Laboratory, Woods Hole. The majority of the experiments deal with the effect of heat on the Nereis egg. Under A these experiments are described. Under B are described experiments with KCl. A. THe EFFECT OF WARMING ON THE INITIATION OF DEVELOP- MENT IN NEREIS. Methods.—At first all sea-water used was heated, usually not beyond 75° C., to destroy any spermatozoa possibly present, cooled, and vigorously shaken before the experiment. But this is unnecessary, as my observations showed. I have kept Nereis eggs in sea-water during the cool days of June for thirty-six hours without even jelly secretion. During several seasons I have never found eggs spontaneously developing in sea-water, although eggs occasionally extrude part of their jelly. More- over, in not a single uninseminated control in ordinary sea-water was a developing egg ever found. In many experiments in - addition to the uninseminated control a batch of eggs from the same animal as those warmed was inseminated. It was thus clearly proved that the eggs subjected to warming are in no wise abnormal. For fear of contamination, the needless inseminated control was discarded in the later work. For a given experiment the following procedure was adopted: A small flask or a large test tube with a measured quantity of sea-water was placed in a large beaker of sea-water. This was warmed over an alcohol flame and the temperature kept constant by the use of thermometers in the flask and in the beaker. The eggs were generally from one female; if from several small ones, they were mixed so that the inseminated or uninseminated con- trols and the warmed eggs were always the same. The eggs in the initial experiments (see below) were either from females cut in the warm sea-water or they were put in the warm sea-water dry; 7. e., from a thoroughly dried female which was pricked to cause the escape of eggs. Eggs were also subjected to heat after washing by changing the sea-water several times during various intervals of time. By means of a capillary pipette measured quantities of eggs were transferred after exposure at varying intervals to five or to one hundred c.c. of ordinary sea-water. The Pic. on INITIATION OF DEVELOPMENT IN NEREIS. 3 experiments: were performed during the morning and afternoon following the evening that the worms were captured. A few experiments were performed during the evening of capture. The point to be emphasized is that washing in sea-water so modifies the eggs that they do not respond readily, or at all, to parthenogenetic treatment. THE EXPERIMENTS. The experiments with heat may be divided into four groups as follows: I. The initial experiments in which the eggs were cut from the animals while in the warm sea-water. 2. The experiments with dry eggs. 3. The experiments with eggs in warm “serum.” 4. The experiments with washed eggs. 1. THE INITIAL EXPERIMENTS. In the initial experiments, worms were cut in 5, I0, 25, 50, anc nOOnc cy Of sea-waten at BO, 32-4138 43447 35) 307_C., the: worms removed and the eggs exposed for from five to fifteen minutes. The following experiments selected from a number give the details: (a) July 22,1:45 P.M. A female put in 100 c.c. of sea-water at 31° C. swims actively without discharging eggs. At 1:50, the temperature is 35° C. Eggs are cut out, the worm removed. Ten samples of eggs are removed to 5 c.c. of ordinary sea-water AsmiOllows: "2.08902 NOr 2 259 2):35)) 2:45, 2255 3:05, 4:15, 3:25. Bea aes At times the temperature rose to 36° C. and once to 36.5° C. Many eggs at the time of removal from the warm sea-water exhibited membranes standing off at an unusual distance, others were darker than normal, and a few had disintegrated. Later experiments showed that these changes are due to exposure at too high temperature. Even five minutes exposure at 37° C. will bring them about. The jelly is formed in the warm water, and often at 35° C. or above it is dissolved and disappears. This may be shown by examining eggs in India ink ground up in 4 B. EU ust: sea-water. Many developing eggs are devoid of jelly hull, but the cortical changes are complete. One hour and ten minutes after exposure, some eggs are in “blister” cleavages; that is, the protoplasm is irregularly budded. One hour and thirty minutes to two hours after the change to ordinary sea-water, among all gradations of cleavage-like patterns are some normal two and four-cell stages. The next day, Nos. 3 to 10 showed some real cleavages and a small per cent. of ap- parently normal swimming forms. Many are beaded or blis- 7 tered, some are unsegmented “‘swimmers,’’ and some two and four-cell swimming forms. Some eggs remain in the germinal vesicle stage. (b) June 23, 11:00 A.M. A female placed in 50 c.c. of sea- water at 35.5° C. is rendered immobile but does not shed. Eggs are cut out at 11:02, the worm removed. Six samples of eggs are taken as follows: 11:18, 11:25, 11:34, 11:40, 11:45, 11:50. Many eggs on removal from the warm sea-water show the jelly formed. The membranes after jelly formation are still a little farther from the eggs than in normal fertilization. Many eggs remain in the germinal vesicle stage with the cortex intact. 1:45 P.M. Fairly normal cleavages in Nos. I to 4. June 24, 8:30 A.M. Swimming forms are found in the dishes. By far the best are those in Nos. 2 and 3. The optimum time of exposure, therefore, lies between twenty- three and thirty-two minutes. Later experiments showed that the optimum exposure at 35° C. is at or near twenty-five minutes. (c) June 24, several experiments were run at various tem- peratures. Those at 35° C. confirmed the findings of the previous ones. Temperatures ranging from 30° C. to 31° C. give no results; regardless of the length of exposure the eggs remain in the ger- minal vesicle stage. The following experiment of June 24, at 33° C. is typical of a number of repetitions at this temperature: (d) June 24, 11:15 A.M. A female placed in 25 c.c. of sea- water at 33° C. swims actively without discharging eggs. Eggs are cut out at 11:15 and samples taken at five-minute intervals up to 11:50. The temperature is practically constant. The samples taken are masses of eggs with the cortex wholly or (in INITIATION OF DEVELOPMENT IN NEREIS. 5 earlier ones) partially broken down. The cytoplasm is normal in color and the membranes normal. 1:35 P.M., many eggs are in cleavage. 4:00 P.M., many eggs are in late cleavage. June 25, 9:00 A.M. The dishes show a good per cent. of very fine ‘swimmers.’ The cleavage seems almost normal. Thirty-five minutes’ exposure gives by far the highest per- centage of swimming forms. As in all the experiments of this group, some eggs remain in the germinal vesicle stage with cortex intact. If eggs be warmed in “‘egg-water’’ (sea-water charged by eggs that have remained in it for several hours) the results are no different. To sum up, we find that eggs of Nereis cut out in warm sea- water and exposed to temperatures ranging from 33° to 35° C. develop with cleavage which is closely similar to the normal. Some eggs remain in the germinal vesicle stage. For the best percentage of swimming forms the optimum exposure at 35° is twenty-five minutes; at 33°, is thirty-five minutes. 2. EXPERIMENTS WITH Dry EGGS. Many of the experiments with dry eggs were run with the washed egg series. In the majority of cases eggs from one female thoroughly dried on clean filter paper were received in a dry watch glass.1 These eggs were divided into two lots; one lot warmed in sea-water at the given temperature and the other washed by changing the sea-water several times, allowed to settle, and after draining placed in the warm sea-water. A large number of experiments was made with dry eggs, in the attempt to determine the quantitative relations early found to control the number of eggs developing. Thus, with smaller quantities of warm sea-water every single egg quickly forms jelly and at least ninety-eight per cent. cleave, but with larger quan- tities of warm sea-water the percentages are lower. As Miss Allyn found for Chetopterus cleavage appears to in- 1 With dry eggs one must be careful for the mere drying will initiate changes as Ihavefound. Eggs left on filter paper for from five to twenty minutes form jelly, a small per cent. cleave and a few swim. 6 De 135 AUIS. terfere with the further development. I have never been able to get more than twenty per cent. of these eggs to reach the swim- ming stage. If one could determine definitely the quantitative relations this percentage might be increased. From the observa- tions it appears that the optimum amount of warm water used varies; it depends upon the bulk of the eggs. While best results are got with small quantities of water, it is possible to use too little—three c.c. for instance, for the eggs for a large female. Jelly formation and cleavage are induced but swimming forms are less numerous than in the case of ten c.c. for about the same bulk of eggs. The following are typical experiments of this group: (a) July 16, 10:30 A.M. Eggs from a dry female in a dry . watch glass are divided into two lots; one lot washed, the other transferred to 5 cc. of sea-water at 34° C. Samples out as follows: at 10:20, 10:50, and at five-minute intervals thereafter to 11:20. 2:00 P.M. Uninseminated control, no change. Every single warmed egg had formed jelly: all have formed polar bodies. At least half of these are in cleavage stages. July 17. All dry eggs in some stage of cleavage, many of which are normal; some swimming forms in many of the dishes even after forty minutes’ exposure. (b) July 17, 9:58 A.M. Eggs from a dry female divided into two lots. Lot A in 5 c.c. of sea-water; Lot B in 20c.c. of sea-water. Both exposed to 33° C. 10:01 jelly formation. Eight samples taken as follows: 10:05, 10:11, 10:16, 10:22, 10:27, 10:33, 10:38, and 10:43. Lot A gave at least 95 per cent. of cleavage and a percentage of swimming forms in all dishes beginning with No. 3 (the 18- minute exposure). Lot B gave 75 per cent. of cleavage and best swimmers for 24, 29, and 35-minute exposures. (c) Other experiments showed that the highest per cent. (100 per cent.) of jelly formation and of cleavage (98 to 99 per cent.) is in the smaller quantities of sea-water—5, 6, and 10 c.c. —whatever the temperature;a fewswim. With larger quantities of sea-water at the various temperatures more eggs remain in the germinal vesicle stage. The lower exposures give most normal-looking swimming forms—trochophores scarcely to be INITIATION OF DEVELOPMENT IN NEREIS. 7 distinguished from the normal either while living or in sectioned material. The higher exposures give more abnormal swimming forms. For comparisons I have selected the following tables from my notes to show the percentages of cleavage and of “‘swimmers”’ obtained with eggs from worms cut in warm sea-water and with dry eggs. Itis apparent at once that while there is no appreciable difference in the percentages of swimming forms after warming either the ‘‘cut out” or the dry eggs, there is a marked difference in the percentages of cleaving eggs. This is the case in all the experiments. _ July 12. Two females cut up at 9:55 A.M. in separate flasks of sea-water at 35° C. gave the following results: Female No. fr. Sample Taken Cleavage. Swimming Forms. IN@: 2 10:06 65% Po % sigice vo 10:13 50 4 rare 10:15 55 5 Wael 10:20 67 5 Semtsiy 10:25 60 9 Be KG) 10:30 72 3 re Pati 10:35 | 47 | 7 eS 10:40 \ 68 4—abnormal re IO:45 | 62 very abnormal LO T0:50 33 | very abnormal Female No. 2. Sample Taken Cleavage. | Swimming Forms. No. 1 10:07 66% | 3% ey es 52 IO:14 42 8 Dy eis I0:16 33 I2 ata vd: TO:21 AT | I4 Stns 10:26 44° 3 ect) I0:31 74 no) iene 10:36 24 2—abnormal oer) IO:41 81 | 5s— “* crs OFda| 10:46 32 very abnormal oF 6G) IO:51 17 very abnormal July 20. Dry eggs in 30 cc. of warm water at 35° C. gave the following results: Exposure, Cleavage. Swimming Forms. I8 minutes 90% 1% 23 se oe I5 % 30 ce ce Io % 35 ce GG 1% 8 1B Be USI. With dry eggs one may obtain 100 per cent. cleavage; with the eggs cut from worms in warm water one never gets more than 81 per cent. the average being very much lower as the figures given above show. With both kinds of eggs 20 per cent. swimming forms is the maximum, the optimum exposure for the various temperatures used being the same. Experiments show that the use of warm “‘egg-water’’ does not improve the results. 3. THE EXPERIMENTS WITH SERUM EGGs. The effect of warming Nereis eggs in the body fluids was studied with difficulty mainly because of the scarcity of body fluid in Nereis. As Lillie has pointed out this worm is little more than a bag of eggs. The amount of blood present is negligible and unavailable for. warming experiments. I therefore adopted the method used by Lillie—that of cutting up spent females. In his study this juice gave results comparable to the perivisceral fluids in Arbacia.. For an experiment I minced as many spent females as I could get, using a small quantity of sea-water; the juice thus obtained is designated as ‘‘serum.” While I think that my experiments with this serum are conclusive I wish to point out that Nereis is not the most favorable form with which to establish the fact of serum inhibition—certainly this is true for the method I used. It may be stated at the outset that as Lillie found for both Nereis and Arbacia I have found repeatedly that the “serum” of Nereis quite definitely inhibits fertilization. Further- more, just as definitely does the serum inhibit initiation of de- velopment with warming. I cite experiments to give the details: . (a) July 15, 10:55. Eggs from one fine large female previously dried are divided into four lots. Eleven spent females are finely minced to procure twenty drops of ‘‘serum.’’ Ten drops of the “serum”’ is added to each of two dishes containing 3 c.c. of sea- water; eggs added to both. One lot is warmed at 34.5° C.— Lot A; Lot Binseminated. Samples of A are taken at five-minute intervals up to 11:35. 2:00 P.M., I per cent. of cleavage in . both lots. Next day no swimming forms in either. Eggs from the same female, Lot C, warmed in sea-water and Lot D, in- seminated, develop. INITIATION OF DEVELOPMENT IN NEREIS. 9 During the afternoon of July 15 this experiment was repeated with the’same results. (b) July 22. Eggs warmed at 34° C. in serum plus sea-water (serum from the bodies of seven spent females cut up in two c.c. of sea-water): 10 drops plus three c.c. sea-water, 10 drops plus five c.c. sea-water and Io drops plus ten c.c. sea-water. Eggs exposed for twenty-five minutes. Less than one per cent. developed in any dish. During August these results with serum eggs were verified. The highest per cent. of swimming forms obtained was one per cent.; this was with a very dilute serum. Not only do the eggs fail fo cleave but fail in the great number of cases even to form jelly. In some cases the development of eggs inseminated in serum was farther advanced than the serum warmed eggs. Since in the case of the initial experiments the worms were cut up in sea-water, it may be that failure of a percentage of eggs to cleave is due to the inhibition of the escaping blood and tissue juice. With the dry eggs cut quickly on the dry watch glass this escaping juice cannot so easily contaminate the eggs. 4. THE EXPERIMENTS WITH WASHED EGGs. In Platynereis sea-water definitely destroys the fertilizing power of the egg. Even minute quantities of sea-water will render the egg incapable of cleavage although the spermatozoa may pene- trate. Moreover, if the eggs of one female remain in a small quantity of sea-water, 5 c.c., for instance, for thirty seconds their fertilizing power is lost. And yet in nature, inseminated eggs begin to be laid in many cases five or six seconds after copulation (see Just,’14). In Nereis, therefore, it was thought that washings in sea-water by frequent changes through several hours might act as the sea-water does in such a surprisingly short time on Platynereis eggs. During the June Nereis run, then, as many experiments as possible were conducted to determine the ‘‘fertilizable”’ period by inseminating at intervals eggs that had remained in sea-water with and without frequent washings. Lillie has shown for Arbacia eggs that the capacity for being fertilized decreased with the decreased secretion of fertilizin. He finds for Nereis also very much the same relationship. Io Be Ey JUS. Without going into details, it may be said at the outset that the egg of Nereis gradually loses its power of being fertilized and eventually reaches the condition of the Platynereis egg where in- semination induces maturation only. I cite a single experiment. June 28, 9:10 P.M. Dishes of eggs Nos. 1, 2, and 3 were set aside. The next day at 2:10 P.M. each dish of eggs was drained and divided into two lots—A and B. Lot A in each case was inseminated in the water-which had stood over the eggs for seven- teen hours. Lot B of each dish was inseminated in fresh sea- water. No eggs in either lot of No. 1 developed beyond matur- tion. In Lots A and B of Nos. 2 and 3, .1 per cent. or less went as far as the two-cell stage. Some eggs in all the dishes were in the germinal vesicle stage. No trochophores were found. Eggs were frequently tied in bags of filter paper and placed in a beaker under running water for twelve hours. In other cases they were washed by changing the water at odd times during the day. It was found that eggs differ greatly with respect to the time that they must remain in sea-water before they lose their fertilizing power, but it may be clearly proved that washing or staling of Nereis eggs renders them incapable of being fertilized. This stage may be reached after three hours in seawater (cf. Just, ’12): This varying susceptibility proved very disappointing because I had suspected, not, of course, the degree of susceptibility present in Platynereis, but perhaps such as could be expressed more definitely. Because of these results with washed and stale egg insemina- tion, when the warm sea-water experiments were continued during the July Nereis ‘“‘run”’ I was certainly unprepared for the results obtained. The following experiments are typical of a large number performed almost daily during the July and August “TcEb have fe (a) July 11, 9:50 A.M. Eggs from one female divided into two lots; one lot put in sea-water. This lot transferred from the sea-water to warmed sea-water (35° C.). The eggs form jelly in the warm sea-water and make a mass which has to be shaken to obtain samples. 10:55, many have formed jelly and maturated, but most retain jelly with germinal vesicle intact. Some of INITIATION OF DEVELOPMENT IN NEREIS. iti these eggs again subjected to heat; no results. July 12. Very few, I in 1000, swimming. (b) July 15, 3:20 P.M. Eggs cut out and washed, put in 6 c.c. of sea-water warmed at 35°C. Samples taken at five-minute intervals for forty minutes. Next day: Majority are in germinal vesicle stage, at least seventy-five per cent. Less than one per cent. swimming. ~ (c) July 16, 9:35 A.M. Eggs washed ten times evening before and five times during this morning. Two series: A inseminated, B in warmed sea-water at 3:50 samples taken (ten in all) at five- minute intervals. Uninseminated control. July 17, 1:30 P.M. No development in uninseminated control (few have cytolyzed). Inseminated eggs show that few have formed jelly (ten to fifteen per cent.). One per cent. have cleaved and some of these swim. Of the warmed eggs at least ninety-five per cent. are in the germinal vesicle stage with cortex intact. Less than one per cent. have formed polar bodies. ° : (d) July 16, 10:30 A.M. Eggs from a dried female divided in two lots; one lot washed in 100 c.c. of sea-water by changing the water four times. 10:40 A.M. In warmed sea-water, 34° C. Samples out at five-minute intervals for sixty minutes. 2:00 P.M. At least ninety per cent. in the germinal vesicle stage, small per cent. form jelly and divide. Next day, none swim. I was tempted to discredit my June experiments after the first of these findings. I could only convince myself after running series after series of washed and dry eggs along with eggs cut out directly into warmed sea-water. Most workers in inseminating eggs obtain the sexual products in separate dishes, and add sperm. Such procedure succeeds admirably with Nereis giving one hundred per cent. of cleavage. But if eggs be cut out of Nereis in sea-water, divided in two lots, and washed once or twice, one lot being inseminated and the other warmed we get the surprising result that while every single inseminated egg develops, few of the warmed go beyond maturation. If the water over the eggs be changed a few times in ten minutes, ninety per cent. warmed in sea-water fail even to maturate. This must mean that the egg of Nereis is so susceptible to 2 E. E. JUST. sea-water that warming fails after washing although fertilization is still possible. If fertilization be impossible (as in stale eggs) warming also produces no effect. Washed or stale eggs warmed in sea-water charged by eggs that have remained in it for some time do not fare any better than those subjected to warmed sea-water; as in the first and second series of experiments this ‘‘egg water’’ makes no difference in the results. I think that these facts are incontrovertible. Washing or even residence in sea-water for a short time interferes seriously with the effect of heat in initiating development. Study of insemination of dry and washed eggs was made. Apparently there is a difference here of response to the spermato- zoon. The dry egg is more irritable, jelly formation being ex- tremely rapid. This is true of dry eggs inseminated in small quantities of sea-water. This behavior recalls that of Platy- nereis. These results, moreover, might suggest that our methods are much too crude in the study of these extremely sensitive cells— the egg and the spermatozoon. Summing up we may say concerning the effects of warming on the eggs of Nereis: (1) That while eggs cut out of worms in the warm sea-water form jelly and divide in large numbers, a small per cent. swimming, some remain in the germinal vesicle stage. (2) That at least ninety-eight per cent. of the dry eggs form jelly almost all of which cleave: twenty per cent. become trochophores closely re- sembling the normal. (3) That eggs in “‘serum’”’ fail to develop except in very small numbers. (4) That washed eggs even after but two or three washings develop if at all in small numbers. B. Errect or KCl In THE INITIATION OF DEVELOPMENT. According to Fischer the eggs of Nereis after treatment with KCI will go through cleavage and produce trochophores. Lillie (11), however, could not get the eggs after KCl treatment to go beyond maturation. During three seasons this had been my experience. This summer I studied the effect of KCl on washed and unwashed eggs. If the eggs be washed two or three times before exposing to INITIATION OF DEVELOPMENT IN NEREIS. 13 the action of KCl every egg maturates but never more than one in a thousand swims. If the eggs be allowed to remain in sea- water from two to twelve hours with frequent changes of sea - water the results are about the same. Dry eggs subjected to KCl treatment maturate, cleave once or twice, and produce, in one experiment at least, seven per cent. of swimming forms made up of unsegmented two and four-cell ‘swimmers.’ THE EXPERIMENTS. 5, 10, 15, 20 and 25 per cent. 2.5M KCl were used. It was found that 15 per cent. 2.5M KCl in sea-water gave the best results. Typical experiments follow: (a) August 12, 10:54 A.M. Lot A: Eggs from two females cut in 80 c.c. of 20 per cent. 2.5M KCl at 10:54. Lot B: Eggs from one dry female put in 3 c.c. of 20 per cent.2.5M KCl. At 11:00 jelly formation in both. Samples of eggs taken from A and B as follows: IN[O SSRs, 43: Anes to) SR cece aeecl aio aco (ont aie II:00 INOS 2 eae See cece 2 cavics vse Gin G) oe een ceaT me II:10 INTO ssi SaaS etOho a ERP nae As coe aay onde ir giey INO NG spel ol orane tous epee eip oly craps THSAS INTO SOS a ares reateaic cu eh ee in ES oe eue eta Wat 6332) INOSRO RG ce ereetet ctor neice: nie untshav enous enone tes II:40 IN OMB Preece eee a ease ye Niaseuatey aweoeuteercnres II:50 INO SIS ei eles erstea rope e ines atenauet svageen ayeneme 12:00 August 13, 12:00 M. Dry eggs of August 12 (Lots A and B). All maturated; some cleavage-like processes and some swimming forms after twenty minutes’ exposure or more. Highest per- centage (five) of swimmers after fifty minutes’ exposure. These are unsegmented, two and four-cell swimmers. (b) August 12, 12:05 P.M. Water changed three times on eggs during three hours and then placed in 20 c.c. of 20 per cent. 2.5 M KClinsea-water. Samples taken at five-minute intervals up to 1:00 P.M. August 13, 1:00 P.M. Washed eggs of August 12, all matur- ated; I in I,000 swim. Experiments during August 13 and 14 with fifteen per cent. 2.5 M KCl gave about the same results. (c) August 15, 9:30 A.M. Two females quickly cut in I0 c.c. 14 Ee Ee. jUSI: of 15 per cent. 2.5 M KCl and removed. Samples taken as follows: INHER: 3 guaien is oe Soe eon Dee aaie T0335 Be A EG Nae tan RR tI tom aN eG 8 10:50 OME Bish daalasoh tna etaeee ye Pues at a arene ea II:05 BS Te hs Be Can The Se ds ae UA ee I1:20 BOAR RUSS. Wes Pap Btls Riots Nant ol RP Petes I1I:40 Sabet OVER ROM PSEA cee SEN ST ne BON CS ETOCS Can 12:00 FSD) EE ss ie “altar vah ouch ue rare ake Re ete Pek Meneame 1:25 INOS Wc lacce Myke etts ole soem 4 per cent bere Retet ae MEN SIO ENan a a tron es Tae Mie pares ee Hecmerihhd Ne ak ae ahs. Kiger DPa aah ae ype BAB ah detae atin ter eee he ee CNT) tee tea a Me RMR Se oe EU a re CN Dee Star ne BREE M6 hse ap near haere He rch Ge Deltas ts Hh fewer tee ee EER ect cass Bee: es (d) On August 17, eggs were washed by changing the water six times in five hours then subjected to 15 per cent. 2.5 M KCl in sea-water, samples being taken at sixty minutes and thereafter at five-minute intervals. One tenth per cent. (.1%) was the best result after sixty minutes in the KCl sea-water. (e) Combination of KCI with Heat..—An experiment of last summer was repeated except that dry eggs were used and the minimum exposure, five minutes, of the series the only one tried. The protocol follows: August 17, 1:25 P.M. Dry eggs are put in 10 c.c. of 15 per cent. 2.5 M KCl in sea-water for five minutes; jelly formation almost at once. Eggs are then placed in 50 c.c. of sea-water at 35°C. Four lots removed to 100 c.c. of sea-water as follows: INTO Hane We Bee ee ae ot weds aie rae a I:55 SP IDE Ay oh Ath ce TN Aon a Renee tee 2:00 eet AR ach ARR trie ebee Sane tia Can 2:05 th) WANA shone Se ee Cet tee ee AO LO The next day, at 10:30 A.M., the percentages of swimming forms, largely unsegmented found were as follows: INOS ase (25 minutes’ exposure to heat)...... 3 per cent. pune See (30 a “ ier ahah) eet itis ep ae, re Bue ae (35 i ie Bie Rahs heres eyes DR ra. (aay tne tae (40 eS ss Daie lint) Muniuaweas DIGG nit it 1 Tt will be recalled that Allyn used a combination treatment of KCl and heat on the egg of Cheloplerus with rather different results from those mentioned here with Nereis. Her method however, was different. ———— INITIATION OF DEVELOPMENT IN NEREIS. 15 It appears, therefore, that with KCl, and KCl and heat, washed and unwashed eggs alike will maturate, but that the dry eggs alone respond with cleavage or the production of swimmers. DISCUSSION. In the egg of Nereis Lillie discovered a substance, fertilizin, which has the property of agglutinating Nereis sperm. This sub- stance may be detected in the water in which the eggs have re- mained for a short time. If, however, the eggs be washed by changing the water two or three times the fertilizin is no longer secreted in detectable quantities, 7. e., there is not enough to agglutinate the sperm. Such eggs are none the less fertilizable by sperm, giving off at the time of insemination more fertilizin, all of which is then utilized or completely thrown off during the cortical changes. It therefore follows that at the time of shed- ding the egg is laden with free fertilizin ready for secretion. This conclusion is supported by additional facts. In the first place I have pointed out above that the dry egg or egg in small quan- tities of sea-water is hyper-irritable—that is, if jelly formation may be taken as index. If one inseminates the eggs of Nerevs dry or in small quantities of sea-water the jelly formation is extremely rapid. Jelly formation is correspondingly slow in washed and stale eggs. The breeding behavior noted night after night for several seasons is significant: freshly shed eggs at the surface of the sea excite numbers of males to shed their sperm around the shedding or recently spent female. Lillie’s experi- ~ ments (Lillie and Just) on this sperm shedding reflex, moreover, prove that the egg loses fertilizin once in the sea-water. The “dry’’ and ‘“‘washed”’ eggs of my experiments, then, are physio- logically different: the dry egg has all its available fertilizin content, the washed egg has secreted part of this substance. Lillie has shown that the eggs of Nereis will not fertilize in the tissue juices of the animal; my experiments show also that the body juice of spent females inhibits fertilization. Unlike the washed egg, the ‘‘serum” eggs possess fertilizin but its action is inhibited. But it is on the basis of experiments on Arbacia that Lillie has developed the fertilizin theory as an explanation of the me- 16 Dy By juSi. chanism of fertilization. Without going into details it may be said that in Arbacia it is found that the egg secretes a substance, fertilizin, whose presence is capable of quantitative determination and which is necessary for fertilization since first, eggs washed free of it are no longer capable for fertilization; second, fertilized eggs no longer secrete it; and third, eggs after membrane forma- tion with butyric acid are not capable of fertilization and do not give off the substance. The perivisceral fluid of Arbacia, moreover, produces an inhibiting effect on fertilization preventing the action of fertilizin on the egg. My results with warming Nereis eggs parallel to a striking degree these facts brought out in the studies of fertilization in Nereis and Arbacia (Lillie, *12, ’13a, ’130, 14). Eggs washed free of the bulk of fertilizin will not develop however long the warming treatment lasts; serum inhibits the artificial initiation of - developmental processes; only the dry eggs with their full content of fertilizin when suddenly shocked with elevation of temperature respond with jelly formation and cleavage. It would seem, there- fore, as I have suggested for Platynereis, that fertilizin is just as essential for artificial initiation as for normal fertilization. The difference seems to be that for artificial initiation more fertilizin is required. Further attempts at Woods Hole this summer to induce artificial parthenogenesis in Platynerets strengthen this belief; a percentage of Platynereis eggs will fertilize in small quantities of sea-water; the same bulk of eggs in the same amount of water fail to respond when subjected to warming. If, therefore, as Loeb (’12a) says, ‘‘fertilization is primarily and essentially artificial parthenogenesis’’; or if ‘‘a theory of fertilization must also be a theory of parthenogenesis at least for the phenomena common to both”’; and if “similarly a theory of fertilization must be consistent with the facts of parthenogen- ~ esis” as Lillie ('14) suggests; these experiments, we are forced to conclude, make another link in the chain of evidence which supports the theory that fertilization is essentially a process of the egg. The spermatozoon initiates the development of the egg, as does warming, through the activation and the binding of the fertilizin. a PO ee gee ee ——— ee ee a a INITIATION OF DEVELOPMENT IN NEREIS. 17 LITERATURE CITED. Allyn, H. M. "12 The Initiation of Development in Chetopterus. BurioL. BULL., 24. Bullot, G. 704 Artificial Parthenogenesis and regular Segmentation in an Annelid (Ophe- liad. Arch. Entw.-Mech., 18. Fischer, M. H. ’03-~«Artificial Parthenogenesis in Nereis. Am. Jour. Physiol., 9. Just, E. E. 712 ~The Relation of the first Cleavage-plane to the Entrance-point of the Sperm. BIOL. BULL., 22. "14 Breeding Habits of the Heteronereis Form of Platynereis megalops at Wood’s Hole, Mass. Bror. BULL., 27. Lefevre, G. ’o7 ~©=Artificial Parthenogenesis in Thalassema mellita. Jour. Exp. Zool., 4. Lillie, F. R. ’o2 ~©Differentiation without Cleavage in the Egg of the Annelid, Chetopterus pergamentaceus. Arch. Ent.-Mech., 14. *tr Studies of Fertilization in Nereis, t and 2. Jour. Morph., 22. "12 Studies of Fertilization in Nereis, 3 and 4. Jour. Exp. Zool., 12. 713 Studies of Fertilization, 5: Jour. Exp. Zool., 14. ’13b The Mechanism of Fertilization. Science, N. S., 38. ’14 Studies of Fertilization, 6. Jour. Exp. Zool., 16. Lillie, F. R. and Just, E. E. 13 Breeding Habits of the Heteronereis Form of Nereis limbata at Wood’s Hole, Mass. BIOL. BULL., 24. Loeb, J., Fischer, M., and Neilson, H. ‘ora Arch. f. d. Ges. Physiol., 87. Loeb, J. ’o1b Experiments on Artificial Parthenogenesis in Annelids (Chzetopterus) and the Nature of the Process of Fertilization. Am. Jour. Physiol., 4. ’07 Ueber die Allgemeinen Methoden der kunstlichen Parthenogenese. Pflti- ger’s Archiv, 118. 708 Ueber die Entwicklungserregung unbefruchteter Annelideneier (Polynoe) mittels Saponin und Solanin. Pfliiger’s Archiv, 122. *12a Heredity in Heterogeneous Hybrids. Jour. Morph., 23. ’12b The Comparative Efficiency of Weak and Strong Bases in Artificial Par- thenogenesis. Jour. Exp. Zool., 13. 13. «Artificial Parthenogenesis and Fertilization. The University of Chicago Press. Loeb, J. and Wasteneys, H. 12 Fertilization of the Eggs of Various Invertebrates by Ox-serum. : Science, 36, No. 921. Scott, J. W. *06 The Morphology of Parthenogenetic Development of Amphitrite. Jour. Exp. Zool., 3. Treadwell, A. L. 702 Notes on the Nature of Artificial Parthenogenesis in the Egg of Podarke obscura. Bio. BULL., 3. SPERM AGGLUTINATION AND FERTILIZATION. FRANK R. LILLIE. In a recent paper on “Cluster Formation of Spermatozoa Caused by Specific Substances From Eggs’”’ Loeb (’14) has pre- sented a criticism of my theory of fertilization (Lillie 136 and 14), based on observation of the California sea-urchin Séron- gylocentrotus purpuratus. My own observations were made on Arbacia punctulata of Massachusetts, and it would appear that part at least of Loeb’s criticism was due to certain differences in the two forms, for he has now stated (Loeb, 1914), p. 318, foot- note) that the ‘‘cluster formation” of the spermatozoa may find its explanation ‘“‘on the assumption of an agglutination at least in the case of Arbacia,”’ as I maintained; it is therefore not a ““tropistic reaction”’ as he thought probable from his observa- tions in California. This was one of the chief differences of opinion. A second one was in regard to the source and sig- nificance of the substance in the fluid of egg suspensions that caused such agglutination; Loeb maintained that it was merely the dissolved chorion (1. e., jelly layer) of the egg, and that after this was removed the eggs no longer produced the agzglutinating substance, and yet were capable of fertilization; whereas my contention was that the agglutinating substance was a secretion of the egg soaked up by the jelly, as by a sponge; that the eggs produced it for a certain length of time after the removal of the jelly, and lost their power of fertilization after they ceased to produce it. These criticisms cut at the foundation of my theory. Inas- much as the correction of the tropistic interpretation of agglutin- ation is given only in a footnote to another paper, and no cor- rection of the source of the agglutinating substance has yet appeared, it is incumbent on me to consider the criticisms carefully; at the same time I wish to take the opportunity to explain certain points that appear to be open to misinterpretation, and to record some new observations. 18 SPERM AGGLUTINATION AND FERTILIZATION. 19 I. ‘‘CLUSTER-FORMATION’’ VERSUS AGGLUTINATION. The phenomena exhibited by sperm suspensions of Arbacia with which we have to deal are of four distinct types, which it is essential to distinguish sharply: (1) activation; (2) aggregation; (3) agglutination, (4) mass-coagulation.! (1) That the activity of spermatozoa is affected by substances in the sea-water requires noargument. The subject is discussed in study V (Lille, 1913a, pp. 519-532). (2) Aggregation of spermatozoa may be brought about by tropistic reactions. In my paper on the “Behavior of Spermatozoa,’ I devoted a great deal of attention to such ageregation phenomena and the distinction from phenomena of agglutination (1913, pp. 532-548 and pp. 551-552). Among other things I pointed out that aggregation as a tropistic phe- nomenon implies a gradient,? and that the spermatozoa never adhere, however crowded they may be; there is no observable physical change of the spermatozoa and the slightest agitation suffices to disperse them again. Such tropistic phenomena may be exhibited in response to CO, and other acids (Nereis), or certain constituents of egg secretions, to mention only chemotaxis. (3) Agglutination of spermatozoa on the other hand requires no gradient, and the spermatozoa adhere physically to such an extent that the agglutinated masses may be preserved intact in killing fluids; its degree is a function of the concentration of the agglutinating medium, and is also different in different species. Agglutination is non-toxic, not limiting the life of the sper- matozoa; it is reversible, its duration depending on the concen- tration of the agglutinating medium; it cannot be repeated if the reaction is complete, at least within the time limits of my ex- periments, even though the spermatozoa remain motile;’ finally motility of the spermatozoa is a prerequisite to a decided reac- 1 The reaction here referred to is a lethal phenomenon. It possibly involves cytolysis with subsequent adhesion of the spermatozoa. 2Tt is important to notice that the spermatozoa of suspensions may produce gradients through their own activities. Thus I pointed out that autogenous aggre- gation reactions in sperm suspensions of Nereis arise from the positive chemotaxis of the spermatozoa to their excreted COs:, giving rise to very striking phenomena (Lillie, 1913a, pp. 519-521 and pp. 538-540). It is conceivable that such a tro- pistic phenomena is involved as a part factor in the agglutination phenomena under discussion. 3 Glaser (1914) also comes to this conclusion. 20 FRANK R. LILLIE. tion; evidently because the physical change on which the reaction depends is not sufficient to cause adhesion except when the spermatozoa positively collide These six criteria definitely define the phenomenon. Agglutination is positively distinct from aggregation. It is an entirely different biological phenomenon. The two may, however, be exhibited simultaneously, as when a drop of egg secretion of Arbacia is injected into a sperm suspension of the species. In such a case the spermatozoa exhibit positive chemo- taxis to one constituent of the egg secretion, and are agglutinated by another (the fertilizin). The separateness of these two sub- stances was maintained in my first publication on the subject and demonstrated by repeated experiments (see Lillie, 1913a, p. 549, and 1914, pp. 545-546). (4) The phenomenon of mass-coagulation is, on the other hand, a lethal irreversible phenomenon. It may be exhibited in response to various agents, such as KOH, NaOH, salts of lanthanum and cerium,” etc., and in some cases the secretions of the eggs of other species or their blood. Hitherto I have not adequately defined this phenomenon as distinct from the agglutination phenomena, though in my last study (1914), I noted the distinction (p. 541). The phenomenon is essentially lethal, but not all destructive agents exhibit it; thus acids, so far as I have observed, destroy the spermatozoa without causing mass coagulation. The phenomenon is irreversible, and this suffices to distinguish it from true agglutination, even if no other criterion were available. However, it exhibits quite a different aspect from agglutination; in the latter the sperm masses tend to take on a spherical form; if originally elongated they contract into balls or break up into smaller masses which become spherical, thus offering considerable resemblance to a phenomenon of surface tension, as Loeb notes. The peripheral spermatozoa are in violent movement until the time of reversal. In the mass-coagulation reaction, on the other hand, there is no such surface tension effect, strands anastomose 1 Loeb argues that the necessity of movement on the part of the spermatozoa for the appearance of this phenomenon removes it from the category of true agglu- tination; but this seems to me to be a purely arbitrary criterion. 2 My attention was called to the action of the salts of these metals by a letter from James Gray of Cambridge University. SPERM AGGLUTINATION AND FERTILIZATION. 21 wit4 other strands and form a net-work and the movements of the spermatozoa soon cease. The substances of egg secretions, which I have hitherto called hetero-agglutinins, belong to this category, in some cases at any rate. Though I will not assert that there is no such sub- stance as a hetero-agglutinin in the real sense of agglutination, yet the substance in Arbacia blood, or egg secretions, the effects of which on Nereis sperm I have previously studied, should be regarded as a toxic rather than an agglutinating substance, having the mass coagulant action. As I stated in my last paper, p- 541,.1t produces a permanent coagulum in Nereis sperm sus- pensions; “‘in this respect the action differs from the iso-agglu- tination, which is without toxic effects.” | We must keep firmly in mind the distinctions between aggre- gation (tropisms) agglutination, and mass-coagulation. Agglu- tination, with which we are particularly concerned, is distin-: guished from aggregation by the facts that it occurs in the absence of a gradient, it involves physical adhesion, and cannot. be repeated if the reaction is once complete; it also is characterized by a high degree of specificity! From mass coagulation it is. distinguished by the facts (1) that it is non-toxic, (2) reversible,. (3) dependent on motility of the spermatozoa. Agglutination occurs so far as I have observed with certainty only in response to egg-secretions of the same species. For description of the phenomena of agglutination of sperm by egg-extractives of the same species, I must refer to my previous paper (Lillie, 1913a@); the phenomenon in Arbacia is a true agglu- tination in the sense defined, not a tropistic reaction, nor yet a mass coagulation. Loeb has admitted this for Arbacia, and I would therefore venture to suggest the probability that the phe- nomenon which Loeb has described in Strongylocentrotus and termed “ reserve as a possible tropistic reaction, is also true agglutination, which differs only quantitatively from Arbacia and Nereis. The cluster formation,’ which he interprets with some 1 Loeb admits that the “cluster formation”’ exhibits a high degree of specific- ity. It is therefore inconsistent to interpret the reaction, as he also does, as a “possible tropistic phenomenon”’ because such phenomena so far as we know do not exhibit specificities of this kind. Agglutination phenomena, on the other hand, as is well known, commonly exhibit equal specificity of a similar kind. Ze, FRANK R. LILLIE. conditions under which it occurs, in response to egg secretions of the same species, its character, reversibility, and the specificity of the reaction are identical with Arbacia. It is apparently, however, less pronounced, and therefore not so readily recog- nizable of itself as an agglutination phenomenon. Even the ‘‘apparent surface tension phenomena’”’ which Loeb describes for the clusters—‘‘Short streaks or cylinders contract into spherical masses, the above described clusters; and long cylinders break up into a series of small clusters’ —are the same as I previously described for Arbacia (1913a, pp. 550-551). Loeb’s interpretation of the “‘cluster-formation”’ as a possible tropistic reaction confuses the two sets of phenomena—viz., aggregation (a true tropistic phenomenon) and agglutination— which sperm suspensions may exhibit to the egg-sea-water of its own species. But the aggregation (tropism) can take place only when there is a gradient from the secretion to the sperma- tozoa. This is realized under the conditions of my experiment of injecting a drop of egg-sea-water into a fresh sperm suspension beneath a raised cover slip; in such a case the two phenomena take place simultaneously viz.; aggregation in the form of a ring around or in the introduced drop (depending on concentra- tion), and agglutination. These two phenomena are produced by two constituents of the egg-sea-water, as I have already maintained. For the study of the aggregation phenomena therefore it is desirable to employ an agent which has no agglutinative action. This I did in an extensive series of experiments by the method just referred to (1913, p. 533 ff.). To illustrate:—a drop of a 4/100 dilution of a saturated solution of CO» in sea-water injected into a sperm suspension of Nereis in sea-water mounted beneath a raised cover-slip is marked within a few seconds by the formation of a ring of active spermatozoa within the margin of the intro- duced drop, and separated from the general sperm suspension by a clear zone nearly free of spermatozoa 1.5 to 2 mm. in diameter. I interpreted the ring formation as a positive reaction to the attractive substance (CO, and acids generally) ; the spermatozoa follow the gradient from the suspension into the drop containing CO» a certain distance, 7. e., up to a certain concentration, and SPERM AGGLUTINATION AND FERTILIZATION. 23 are there arrested. The proof of this interpretation is found in the fact that, if increasing concentrations of CO, are used, the ring forms outside the drop and becomes progressively wider, 1. €., the migration ceases at a distance from the center which increases with COs or acid concentration (see 1913a, pp. 536-538). Loeb suggests that the ring formation with a clear external zone around it is “‘an indication that the spermatozoa are negatively chemotropic to the strong egg-sea-water, and possibly positively chemotropic to the more diluted egg-sea-water, or to the dense collection of spermatozoa in the ring.’ The latter suggestion is of course untenable as a primary cause, for the ring-formation is precisely the phenomenon to be explained. It is also unnecessary to assume any negative tropism; the ring formation is due to a limitation of the positive movement by concentration. This is fully discussed in the paper referred to above, but Loeb does not allude to the discussion. 2. THE SOURCE OF THE AGGLUTINATING SUBSTANCE. Professor Loeb has also taken issue with me on the question of the origin of the agglutinating substance. He regards his experiments as proving that the substance which causes the ‘cluster formation” is not formed in the egg but in the chorion; 1. é., in the layer of jelly which surrounds the egg. On the other hand I regarded it (and still hold to the opinion) as a secretion of the egg; with which the jelly of course becomes saturated. Loeb’s observations again were on Strongylocentrotus and mine on Arbacia. ‘The issue is a real one even though the chorion is itself a secretion of the egg in earlier stages. Loeb’s conclusion was based on his observation that if the chorion be dissolved off in dilute hydrochloric acid in sea-water, the naked eggs transferred to sea-water produce no detectable amounts of the agglutinating substance any more, whereas the acid sea-water contains it in large quantities. My conclusions were based on the observation that when eggs of Arbacia are deprived of jelly (chorion) by shaking, or a prolonged series of 1 Glaser (1914) also agrees substantially with me: ‘‘the agglutinating substance is located in greatest abundance in the jelly and the eggs also contain this material,”’ p- 371. 24 FRANK R. LILLIE. washings, they still continue to produce the agglutinating sub- stance in sea-water, though in much diminished quantity; in my full paper, which Loeb had not the opportunity of consulting, I gave series of measurements on this point (1914, pp. 532-538); I also pointed out that in immature ovaries containing many primary ovocytes, but some mature eggs, the quantity of agglu- tinating substance produced was relatively very smell (1914, p- 530), and I therefore suggested that the substance was secreted by the eggs at the time of maturation and was soaked up by the jelly as by a sponge. The eggs, however, continue to produce it after maturation, as I shall show. The immature eggs have as thick a chorion as the mature eggs; therefore the agglutinating substance cannot be merely dissolved chorion. I recognized the possibilicy of the view expressed by Loeb, investigated it as fully as possible at the time, and rejected it. Since Loeb’s paper has appeared, I have repeated his experi- ments and found my former observations and conclusions con- firmed in all respects: Experrments.—The optimum concentration of HCl for removal of jelly without injury to eggs was found to be 50 c.c. sea-water - 1.4 ¢.c. N/to HEI. 1.2'c.c. N/1o H@lin Soic:c. sea=water didmon fully remove the jelly, and 1.6 c.c. caused too much injury to the eggs evidenced by heavy agglutination and later cytolysis. In an experiment of July 17, 1914, the three above concentrations were used. The complete removal of the chorion in the inter- mediate concentration was demonstrated by observation of the eggs in a thick suspension of India ink in sea-water; even the minutest traces of adherent jelly can readily be detected by this method, but it was all gone. The eggs were then washed as follows: 10.11 A.M. 42/6 c.c.; 10.40 51/5 c.c.; 10.58 50/4 c.c. The supernatant fluid was then tested and found to be free from sperm agglutinating substance; thus furnishing proof that all originally contained in the jelly had been washed out. At 11.20 the supernatant fluid was poured off leaving only 5 c.c. in the tube. The eggs were allowed to settle, and at 11.25 the super- natant fluid was tested and gave a 9-10-second agglutination reaction with fresh sperm suspension. Thus these eggs entirely deprived of jelly by HCl are producing agglutinating substance. SPERM AGGLUTINATION AND FERTILIZATION. 25 At 4.25 P.M. the eggs were washed again 5/0.7 c.c. and the new fluid gave a I4-second reaction. The next morning the same eggs were washed again 5.5/1 c.c. The new fluid gave a 6-7- second reaction. These results may be expressed in a different way: thus in an experiment of July 20, a series of eight successive washings of eggs deprived of jelly by acid sea-water represented a dilution of the agglutinating substance contained in the acid sea-water remaining with the eggs of 12,700,800 times. But the acid solvent itself was negative at 1/800 dilution: it was of 400 agglutinating power. In other words, after the removal of the jelly the eggs themselves had produced a sufficient quantity of the agglutin- ating substance to account for the tremendous difference; and they were still producing it. These eggs without jelly are fertilizable, as Loeb states, but only 37 per cent. segmented in a heavy insemination of the first day in the experiment of July 17, and only a small part of these developed to the ciliated stage, none of which were normal, most being stereoblastulae and incapable of farther development. The result is entirely similar to that described in my last paper (study VI, ’14) for the fertilization of eggs deprived of jelly by shaking and subsequent washing. The same experiment was repeated on July 18, 20 and 21, with identical results: the eggs from which jelly i@entirely removed by HCI continue to produce the sperm-agglutinating substance (fertilizin) so long as they live, but their capacity for development after fertilization is much reduced. In all experiments at least three concentrations of acid were used, and in each experiment it was observed that when the con- centration was sufficient to dissolve the jelly there was a good deal of agglutination of the eggs, and in the later washings a great many eggs broke down liberating their pigment. As I have previously shown, broken-down eggs liberate a substance (anti-fertilizin) which neutralizes the sperm agglutinating action of the fertilizin. Therefore, when a sufficient percentage of the eggs are breaking down, the production of sperm-agglutinating substance (fertilizin) by intact eggs may be entirely masked. I have no intention of disputing Professor Loeb’s observations 26 FRANK R. LILLIE. for Strongylocentrotus. But they merely prove either that Strongylocentrotus sperm is not so delicate an indicator as Arbacia sperm, or that the method employed by Loeb was inadequate to detect small quantities of fertilizin. In Avrbacia the eggs con- tinue to charge the sea-water with sperm-agglutinating substance after complete removal of the jelly, whether by shaking and repeated washings, or by HCl; and the substance continues to be formed as long as the eggs remain fertilizable and living, no matter how often the eggs are washed. The eggs of Arbacia secrete the substance as I previously maintained. It is not merely the ‘dissolved chorion.” It might possibly be objected to this conclusion that the con- tinued appearance of the agglutinating substance in egg suspen- sions in sea-water after removal of the chorion indicated merely previous adsorption of the substance of the chorion. But the indefinite continuance of its production is inconsistent with the idea of a mere secondary removal of an adsorbed substance. The idea is also inconsistent with the fact that Nereis eggs have no jelly at the time laying, but produce a similar sperm agglu- tinating substance. In this form the jelly also is secreted by the egg after insemination. Finally if it can be shown that the jelly of immature eggs is entirely devoid of the sperm agglutinating substance, my position that this substance is a later secretion of the egg is rigorously proved. As noted above I maintained the probability of this view in my previous paper (Study VI). This summer my first experiments were undertaken to investigate this point anew. Fortunately the season was late, and not a single Arbacia was ripe when I began work (June 8). This applied to males as well as females: so it was impossible at first to secure ripe sperm as indicator. I therefore made extracts of immature ovaries to be kept for subsequent testing from three females (1, 2, and 3, June 8). June 11 extracts of ovaries in sea-water were made from females 4, 5, and 6: numbers 4 and 6 contained only ovo- cytes; No. 5 had a large number of ripe ova in addition. On June 16 extracts I-6 were tested with Arbacia sperm suspension: I, 2, 3, 4, and 6 were absolutely negative; no agglutination. No. 5 gave a strong agglutination reaction lasting about one minute. SPERM AGGLUTINATION AND FERTILIZATION. 27 It is highly improbable that the agglutinating substance had been destroyed in five of the six, and retained in the only one (No. 5) of the extracts which was made from ovaries containing some ‘ripe ova. So far as these observations go, the jelly of immature ovocytes is free of agglutinating substance. Again on June 15 I made extracts from ovaries of three females in two of which ripe ova were practically absent, the third bad a few. Tested the same day the two former extracts had no sperm agglutinating properties; the third gave slight agglu- tination. The females appeared to mature slightly earlier than the males, so that for these experiments I was forced to use rather thin sperm suspensions (mixed more or less with immature spermato- zoa), which were probably not as delicate indicators as one could wish. However the difference between the ovaries containing ripe ova and those without was perfectly distinct. Later when fully ripe males could be had all ovaries contained ripe ova. The following observation also tends in the same direction: June 27, 1914—Three females were selected, of which number 1 was the ripest attainable, the eggs flowing freely out of detached ovaries, and very few ovocytes occur; numbers 2 and 3 were the least mature attainable; number 2 had very few detachable ova, mostly late ovocytes with a sprinkling of ripe eggs; number 3 had quite a few detachable ova with a large proportion of ripe eggs. The ovaries of all three were cut up equally, and sea-water added to each to make 10 c.c. When the ova and ovaries had settled they stood at 1.5 c.c. in I, at 1.3 c.c. in 2, and 1.5 c.c. in 3. After five hours, tests of the agglutinating strength of the super- natant fluids were made with clear fresh sperm. No. I gave a 10-second reaction at 1/800 dilution. No. 2 gave a 6-second reaction at 1/10 dilution. No. 3 gave a 7-second reaction at 1/40 dilution. Thus No. 1 is 80 times the strength of 2 and 20 times the strength of 3. In general the fertilizin production is proportional to the ripeness of the ovaries. There is not the slightest doubt in my mind about the demon- strative character of these observations. The appearance of agglutinating substance in the jelly of Arbacia eggs is secondary, 28 FRANK R. LILLIE. and takes place probably at the time of breaking down of the germinal vesicle. Loeb’s contention that the agglutinating substance is merely dissolved chorion therefore does not hold for Arbacia. With this his argument against my fertilizin theory also falls: ‘‘More- over if it should turn out that the substance which is responsible for the cluster formation is identical with the substance which Lillie calls ‘‘fertilizin,’’ which is very likely the case, Lillie’s theory becomes untenable, since this substance does not, in all probability, originate from the egg, but from the chorion and since there is, as we have seen, no connection between the presence of this substance and the power of the eggs of being fertilized” (pp. 136-137—Loeb, -'14). In this statement Loeb sums up the essentials of his criticism; since I have shown that “‘cluster formation” is true agglutination (which Loeb now admits), and that the agglutinating substance (my fertilizin) is not dissolved chorion but a true secretion of the eggs which continues to be produced after the chorion is removed, the entire stated criticism becomes ineffective. There 7s a connection between the presence of this substance and the power of the eggs of being fertilized: the substance can first be demonstrated at the time that the power of being fertilized first arises, viz., after breakdown of the germinal vesicle; it can be demonstrated as long as eggs retain the power of being fertilized, whether the chorion be removed or not, and it disappears ab- solutely after fertilization, as I showed in my previous paper (study VI, p. 553, 1914). 3. OTHER CRITICISMS. Another objection raised by Loeb is that ‘‘the supernatant sea-water of the eggs of Strongylocentrotus franciscanus will not induce cluster formation of the sperm of Strongylocentrotus purpuratus: yet the latter sperm fertilizes the eggs of francis- canus,”’ from which he argues that the fertilizin of Strongylo- centrotus franciscanus can not be necessary for the fertilization of itseggs. Anerror in logic is involved here; agglutination of sperm is merely an indicator of the presence of a certain substance, which is none the less present in franciscanus even if purpuratus SPERM AGGLUTINATION AND FERTILIZATION. 29 sperm does not reveal it; it may nevertheless be activated by purpuratus sperm and this is the essential point in the theory.! Agglutination of sperm is of no significance except as indicator. As I pointed out in my previous paper, binding of the fertilizin by sperm receptors, 7. ¢., the chemical reaction, is a thing entirely distinct from agglutination; if such binding causes a certain kind of physical surface change of the spermatozoa of suspensions of a certain minimum concentration, they agglutinate; otherwise not. Agglutination is a valuable indicator that enables us to make certain analyses, and that is all. The same principle of fer- tilization may hold in the entire absence of sperm agglutination. Another objection in which Loeb supports the possibility of superposing fertilization on parthenogenesis will be dealt with in a separate paper. My contention in this case is that the possi- bility of such superposition always rests upon incompleteness of the parthenogenetic reaction; if the fertilization reaction be complete, whether by parthenogenesis or insemination, it cannot be repeated. Everybody admits that eggs fertilized by sperm cannot be refertilized; it is a logical impossibility that eggs ‘‘fer- tilized” by parthenogenetic reagents should be refertilized. The problem of the apparent contradiction involved in Loeb’s and Herbst’s contention of superposition works out in the manner indicated. A study of this problem by one of my students will appear soon. Loeb cites as a farther difficulty of my fertilizin theory, which he says I have not considered, ‘‘that in addition to the membrane forming substance still another, namely a correcting agency, is necessary for causation of the development of the egg.”” Though 1 Loeb states (1914, p. 135): ‘‘If the phenomenon of cluster formation were in- separably associated with the power of the eggs of being fertilized, we should expect that sperm should only be able to fertilize the eggs of a species if the egg-sea-water of the same species caused the cluster formation of the sperm.’”’ I have never maintained that agglutination (‘‘cluster formation’’) is inseparably associated with the power of the eggs of being fertilized, but merely that a certain substance pro- duced by the egg is a necessary factor in fertilization. In some cases this substance (fertilizin) produces agglutination of the sperm of its own species, and this reaction furnishes an indicator of its amount, when present, or of its absence. In other cases such an indicator is lacking: I do not find that supernatant sea-water of the eggs of the starfish (Asteria forbesii), for instance, agglutinates its own sperm; but I have evidence, to be published elsewhere, that the mechanism of fertilization may be explained in the same way as in Arbacia. 30 FRANK R. LILLIE. I cannot accept this statement of the problem, I have nevertheless taken into consideration the fundamental fact, to which Loeb alludes, in the full account of my experiments, which appeared after Loeb’s paper was in press. The fundamental fact is simply — that the fertilization process in some cases can be divided in two sharply marked stages. This is perhaps most simply and convincingly shown by my own experiment (Lillie, 1911) of removing the spermatozoon from the egg of Nereis after it had already induced the cortical changes, with the result that the developmental phenomena came to a standstill before the first cleavage. I cannot agree with Loeb that the second stage in- volves a factor corrective of an excess action of the factor of the first stage. I think it is probable that we have a progressive process readily capable of resolution into two stages. In my complete paper (Lillie, 1914) I considered the pea phase of fertilization with reference to the new theory, and may refer the reader to the discussion there given (study VI, pp. 582-584). Here it is only necessary to point out that the ‘fertilizin’’ theory is at least as well adapted to account for the two stages as the “‘lysin”’ theory. 4. CONCLUSION. I may be allowed to emphasize the essential features of my theory with some added light thrown by the work of this summer. The fundamental conception is that all agencies initiating development of the egg do so by the same means, viz., activation of an ovogenous substance, which I have termed fertilizin. This conception brings fertilization and parthenogenesis under one conception. I further assumed that such activation in the case of fertilization was caused by union of a constituent substance of the spermatozoon (the sperm receptors) with the fertilizin, and that the activation expressed itself by consequent union of the fertilizin with certain egg substances (the egg receptors). The reaction was thus conceived in terms of the Ehrlich side- chain theory, and was represented diagrammatically accordingly. That certain chemical combinations form an essential feature of the fertilization reaction cannot be open to doubt. I have not previously taken into account the consideration that the SPERM AGGLUTINATION AND FERTILIZATION. 31 4 occurence of such reactions, taking place, as they must, across the egg membrane, is dependent on physical conditions of the membrane, especially its permeability to the substances con- ‘cerned. In speaking, as I did, of five blocks to the fertilization reaction, I was concerned only with the chemical reactions in- volved. There may be other blocks of a physical nature. Indeed these were much in evidence in the fertilization of Asterias, which I studied in the first part of the summer, and shall report on elsewhere. Another important consideration is that the reaction must also be dependent on environmental conditions such as temperature, ionic constitution of the medium (see Loeb, - ’14b), etc. Blocking of fertilization may also arise from such causes. Continuing the exposition of the theory; I identified the fertil- izin of Arbacia with the substance found in the fluid of egg sus- pensions which causes agglutination of sperm suspensions of the same species. This phenomenon cannot possibly be lacking in significance, for it furnishes direct evidence of a combination of egg and sperm derivatives; the phenomenon itself is not con- cerned in fertilization, for a single spermatozoon may fertilize an egg. Neither does the absence of such agglutination in other species affect in the least the conclusion that may be drawn from Arbacia: because we may have a combination of egg and sperm derivatives without any sperm agglutination. The agglutination is incidental, the combination is the essential thing. The fertilizin theory in its essential aspects is not dependent on the identification of fertilizin and sperm agglutinating sub- stance. I believe in their identity; but if it were proved, as Loeb has sought unsuccessfully to do, that the agglutinating substance is not essential for fertilization, the fertilizin theory would still not be attacked in its essence. The conception that initiation of development is essentially a phenomenon of activa- _ tion would still stand in opposition to theories of external agents acting directly by corrosion (cytolysis), or coagulation, or what not. The egg could still be regarded as a self-contained system with no more than the usual environmental relations. It is only from this point of view that the complex phenomena of parthen- ogenesis and fertilization can be united in a logical whole. 32 FRANK R. LILLIE. The theory of the identity of fertilizin and sperm agglutinating substance rests upon a considerable body of ascertained facts (see study VD), and it gives us at once a point of attack and a working hypothesis of considerable value. I have been able to show for instance that the origin of the capacity of the egg for being fer- tilized can be understood on this basis; that the cessation of fertilization capacity can also be so understood; and that the physiological sterility (prevention of polyspermy) of fertilized eggs is readily explained by the neutralization of the fertilizin by a substance (anti-fertilizin) demonstrably present in the egg. On the other hand the theory does not postulate that the fertilizin of all forms should agglutinate sperm of its own species. There may be many forms in which the union of the sperm re- ceptors with fertilizin does not produce such physical changes of the spermatozoa as to lead to agglutination. In those cases in which agglutination does occur we have a reaction very useful in analysis; but it cannot be too strongly emphasized that the agglutination itself is to be regarded merely as an indicator of the essential reaction. SPERM AGGLUTINATION AND FERTILIZATION. 33 LITERATURE. Glaser, Otto. ’14 A Qualitative Analysis of the Egg-Secretions and Extracts of Arbacia and ; Asterias. Biot. BULL., Vol. 26, pp. 367-386. Lillie, Frank R. ’r112 Studies of Fertilization in Nereis—I. Cortical Changes in the Egg. II. Partial Fertilization. Journ. Morph., Vol. 22, pp. 361-391. III. and IV: Jour. Exp. Zool., Vol. 12, pp. 413-474. ’12 The Production of Sperm Iso-agglutinins by Ova. Science, N. S., Vol. XXXVI, pp. 527-530. ’13a Studies of Fertilization. V: The Behavior of the Spermatozoa of Nereis and Arbacia with Special Reference to Egg-Extractives. Journ. Exp. Zool., Vol. 14, pp. 515-574. ’13b «=The Mechanism of Fertilization. Science, N. S., Vol. XXXVIII., pp. 524-528. ’14 Studies of Fertilization—VI. The Mechanism of Fertilization in Arbacia, Journ. Exp. Zool., Vol. 16, PP. 523-590. Loeb, Jacques. "14a Cluster Formation of Spermatozoa Caused by Specific Substances from Eggs. Journ. Exp. Zool., Vol. 17, pp. 123-140. ’14b On Some Non-specific Factors for the Entrance of the Spermatozoon into the Egg. Science, N. S., Vol. XL., pp. 316-318. \ OBSERVATIONS ON THE SPERMATOGENESIS OF THE GALL-FLY, DRYOPHANTA ERINACEI (MAYR). H. L. WIEMAN, ZOOLOGICAL LABORATORY, UNIVERSITY OF CINCINNATI. INTRODUCTION. Dryophanta erinacet is one of the gall-producing Hymenoptera having two generations in the year: males and females in the spring, and females alone in the fall. The fertilized, eggs of the bisexual generation produce females exclusively, while the un- fertilized eggs of the female generation produce both males and females. The material for this study was obtained on April 13, 1914, from galls occurring on white oak trees (Quercus alba) in the neighborhood of Cincinnati. The galls are smooth spherical protuberances on the bud-scales, several millimeters in diameter. Usually one, but occasionally two, and sometimes three galls very close together are found at the end of a single twig. Twigs bearing galls were cut off, brought into the laboratory, and the cut ends inserted in sand moistened with water. Ten days later males and females emerged from the galls, and continued emerg- ing for two weeks. Copulation took place immediately after emergence. These galls supplied all stages of developing males and females from the late larva to the imago. For fixation Petrunkewitsch’s fluid was used; penetration being facilitated by making a longi- tudinal incision through the body wall. In some cases ovaries and testes were dissected out, but better results were obtained when the organs were left in situ and the entire animal sectioned. Embedding was done in rubber-paraffin, and serial sections cut 10 w in thickness. Two methods of staining were used: safranin and light-green, and iron-haematoxylin with or without counter- stain. After dehydration the stained sections were cleared and ‘For assistance in collecting and preserving material I wish to acknowledge my indebtedness to Dr. Annette F. Braun. 34 SPERMATOGENESIS OF THE GALL-FLY. 35 mounted in euparal. Euparal offers several advantages over balsam as a mounting medium. Thus its use obviates running stained sections through absolute alcohol; since sections may be transferred directly from 95 per cent. alcohol to euparal. Next the index of refraction of euparal is low 1.483. And lastly euparal dries quickly, so that sections may be studied at the end of twelve to twenty-four hours after mounting, without danger of injury. OBSERVATIONS. The testes of the late larva and early pupa show primary spermatocytes at the end of the growth period as large polygonal cells having a reticulated nucleus containing a poorly defined nucleolus, often of a bipartite character (Fig. 1). The nucleolus does not take the safranin stain as deeply as the chromosomes, and thus differs markedly from the chromosome nucleolus of the primary spermatocytes of many Hemiptera. A true primary spermatocyte division does not occur. Instead, a small mass of cytoplasm free of chromatin is constricted off, forming the so-called polar body. Preparation for this sup- pressed or abortive division begins with a change in the outline of the cell, the spermatocyte assuming a pear shape (Figs. 2, 3 and 4). From the narrow end of the cell and forming the stem of the pear, extends a short filar process. At the base of this process, which at first glance suggests the tail of a spermatozoon, is often. found a light basic-staining spherule which may or may not be a centrosome. While these changes are taking place in the cyto- plasm the nucleus undergoes a slight contraction and the chro- matin passes through a series of transformations terminating in the formation of chromosomes (Fig. 4). The next step in the process is somewhat uncertain and there may be some question as to seriation. It seems that after the chromosomes are completely formed, they become massed in clumps at one side of the nucleus, and from these masses distinct loops extend toward the opposite side of the nucleus (Figs. 5 and 6). The cell shortens, the filar process becomes less distinct (Fig. 6), and a portion of the cytoplasm is constricted off (Figs. 7 and 8). As this is taking place the nuclear membrane appears very irregular in outline but seems to remain intact. Inside the 36 . H. L. WIEMAN. nuclear area the chromosomes are in the form of single rods whose free ends extend toward the polar body. There is every appear- ance to indicate a resistance of the part of the chromosomes against a tension pulling toward the polar body. Distinct spindle fibers are not to be seen, but the cytoplasm contains a reticular structure which may represent a poorly developed spindle. The polar body is quickly cut off from the cell to which, however, it may remain attached for a considerable length of time (Fig. 15). The free polar body of Fig. 8 belongs to a cell in an adjacent sec- tion. Polar bodies cut in various planes are frequently seen in the spaces between spermatocytes at this time (Figs. 8 and 16) and throughout the second spermatocyte division. The complete absence of polar bodies in cysts containing cells with the chro- mosomes in the looped condition of Figs. 5 and 6 makes it almost certain that the looped stage precedes that of Figs. 7 and 8, in which the chromosomes show free ends. Preparations for the second spermatocyte division follow very rapidly. After the formation of the polar body, the second sper- matocyte rounds up; the knot of chromosomes separates into distinct, short, thick, curved rods, 12 in number (Fig. 9). In the cell figured here, a late prophase, the nuclear mem- brane is fairly distinct. Details of spindle formation were not observed. Figs. 10 and 11 show characteristic side-views of spindles at metaphase. The chromosomes seldom lie in one plane so that counting even in polar views is a difficult matter. In such views, as in Figs. 12 and 13, 12 chromosomes can be counted with considerable accuracy in the majority of cases. A characteristic late telophase is shown in Fig. 14 which re- sembles to a striking degree a somatic mitosis, and strongly suggests that the chromosomes have been divided longitudinally. In later stages of this division (Fig. 15) the chromosomes become packed into dense compact masses, so that it is impossible to determine the number of constituent chromosomes in the daughter groups. When reconstitution of the nuclei occurs (Fig. 16), these masses break up'into slightly bent rods of ragged outline. In cross section these rods appear as dots of which 12 can often be counted. Counts of the daughter groups of chro- mosomes made in this way are not very satisfactory, since one.is SPERMATOGENESIS OF THE GALL-FLY,. 37 never sure that a cross-section includes all of the rods or that a single rod has not been cut more than once. The spermatids formed by this division seem therefore to be equal in size and chromatin content, and all of them develop into spermatozoa. There is no evidence of a heterochromosome or chromatoid body passing undivided into one of the spermatids. By the end of the second spermatocyte division all of the polar bodies are detached and show signs of disintegration, fragments being frequently seen in the intercellular spaces giving the ap- pearance shown in Fig. 16. The relatively distinct outline of the chromosomes seen in this last figure persists for but a short time and is completely lost in the young spermatids. Figs. 17 and 18 are early stages in the transformation of the spermatids into spermatozoa. Such in brief is an outline of the main features of development of the germ cells in the male of Dryophanta from the growth period to the spermatids. There is but one true maturation division—that of the second spermatocyte. The first spermato- cyte division is indicated by the pinching off of a small quantity of chromatin-free cytoplasm which forms the so-called polar body. DISCUSSION. Doncaster in his studies of the gametogenesis of the gall-fly, Neuroterus lenticularts, arrived at certain conclusions which may be considered at this point. This species of Hymenoptera has a similar life-history to that of Dryophanta. Thus according to Doncaster the female generation emerges in April from galls formed during the preceding summer and immediately lays eggs in oak buds (species?). Early in summer the galls appear from which males and females emerge. After copulation the female lays eggs in the tissues of young leaves at the side of a small vein. From the galls resulting, females emerge in the following spring. As in Dryophanta, therefore, the fertilized eggs of the bisexual generation develop into females; while the unfertilized eggs of the female generation produce both males and females. Doncaster found that the first spermatocyte division is abor- tive—a small portion of the cytoplasm being constricted off as the polar body. This is followed by a resting stage which resembles 38 H. L. WIEMAN. the metaphase of a true division, but is distinguished from it by the persisting nuclear membrane and the position of the chromo- somes at one end of the nucleus near the broad end of the cell. No nuclear division takes place but the nucleus becomes oval in shape and the chromosomes generally contract to form a compact mass lying across its center. Insome cells at least this chromatin mass seems to divide—one half passing to each side of the oval nucleus. The chromatin may finally disperse and give rise toa condition resembling the first spermatocyte in which the chro- matin has begun to appear. ‘“‘Possibly the division of the chro- matin inside the nucleus, which occasionally seems to occur, is the persistent remnant of a true nuclear division, or it may be compared with the ‘intranuclear karyokinesis’ described by Kostanecki in the parthenogenetic eggs of Mactra”’ (p. 93). Toward the end of the rest stage the chromatin becomes grouped in the form of large elongate granules or small bands having a more or less meridional arrangement under the membrane. The second spermatocyte division in Neuroterus is a true mi- totic division in which the haploid number of chromosomes, 10, appears on the spindle to be equally divided between the daughter cells. There is also a small stained body lying outside of the spindle which passes undivided to one of the spermatids. In the spermatogonia and in mitotic figures of nerve cells in the developing nervous system Doncaster finds the halpoid number of chromosomes, 10, but in mitoses of immigrant meso- derm cells the diploid number, 20. The eggs layed by the females of the bisexual generation undergo two maturation divisions; leaving 10 chromosomes for the female pronucleus. The spermatozoon brings into the egg 10 chromosomes, and 20 chromosomes appear on the cleavage spindles. The parthenogenetic eggs of the female generation may be divided into two groups: Those which undergo matura- tion and develop into males; those which omit the maturation divisions and develop into females. In the first group 10 chro- mosomes are found in the cleavage divisions; in the second group 20. Since any female produces only one kind of egg, there are male-producing females and female-producing females. Mitoses in the nervous system of all females show the diploid number of chromosomes. SPERMATOGENESIS OF THE GALL-FLY. 39 Returning now to Dryophanta I should like first to consider the stage represented in Figs. 5 and 6, which I believe corresponds to the second spermatocyte resting stage mentioned by Don- caster in Neuroterus. The figures at first glance suggest the synapsis stage of other insects, but in view of other facts it is difficult to interpret the condition as a fusion of chromosomes. Earlier stages such as the prophase shown in Fig. 4 display the same number of chromosomes as appears in the-second spermato- cyte division, 12, which is assumed to be the haploid number approximately. Since there is no evidence in Dryophania of an intra-nuclear division of these 12 chromosomes into two groups, a true synapsis at this time would be equivalent to a second ‘“reduction.’”’ A more probable interpretation of this “looped stage’’ and one that is warranted by a close study of the sections is that the limbs of a loop are the halves of a chromosome that has undergone a temporary and incomplete splitting. With the next step in the process, the formation of the polar body, the split disappears and the chromosomes have every appearance of being single, solid rods (Figs. 7 and 8). The latter condition might of course be brought about by breaking of the loops at the middle, but in that event one would expect to find twice as many single chromosomes as loops. Such is not the case, for the number of unsplit chromosomes is the same as the number of loops so far as could be determined. Reversing the seriation at this point would of course change the interpretation offered here; but the main reason for placing the looped stage before the other, as has been mentioned above, is that there is no evidence of polar body formation at this time. And to this may be added the fact that the outline of the cell at the looped stage as shown in Fig. 6 represents an intermediate condition between that of Fig. 4 in which there can be no question about polar bodies being absent, and Figs. 7 and 8, in which the polar bodies certainly are present. An actual resting stage, if one occurs at this time, must be of very short duration. The second spermatocyte division follows very quickly after the formation of the polar body. Fig. 9 represents a prophase of this division in which the chromosomes are surrounded by an intact nuclear membrane. The spindle 40 H. L. WIEMAN. area of the second spermatocyte is rather distinctly marked off from the rest of the cytoplasm (Fig. 10) and suggests that the . nuclear membrane disappears very slowly. Polar views of the metaphase display, as nearly as could be determined, 12 chromosomes, presumably the haploid number. It would seem that each chromosome is divided quantitatively by a longitudinal splitting; although it must be remembered that attempts at verifying this conclusion by studying the constit- uents of the daughter groups are not satisfactory owing to the tangled condition of the chromosomes. ; I find nothing resembling the small stained body which in Neuroterus according to Doncaster passes undivided to one of the spermatids. As Wilson has observed this body is of the same nature as the chromatoid body seen in the growth-period and spermatocyte-division of Pentatoma. The chromatoid body is of rounded form, dense and homogeneous consistency, and after double staining with haematoxylin or safranin and light green is at every stage colored intensely blue-black or brilliant red, precisely like the chromosomes of the division period or the chro- mosome-nucleoli of the growth period. Nevertheless Wilson finds that the body is neither a chromosome nor any kind of a chromosome and takes no visible part in the formation of the spermatozoa. In the transformation of the spermatids it wanders far into the sperm-tail and is at last cast off altogether. I have not yet had opportunity to study the maturation phe- nomena of the egg in either generation of Dryophanta, but obser- vations confined to individuals of the bisexual generation point to general conclusions which differ somewhat from Doncaster’s views regarding the chromosomal relations in the alternate generations. In the material at my disposal spermatogonial divisions are not abundant enough to determine the number of chromsomes. While mitoses abound in the somatic cells of male larvae and pupae, it is difficult to find good clear metaphases; but wherever counts were possible, the number found was 12 (Fig. 19). In the follicle cells of the ovary I have found it less difficult to count the chromosomes. Figs. 20, 21 and 22 are drawings of metaphase plates of such cells in which the numbers are re- spectively 13, 14 and 13. SPERMATOGENESIS OF THE GALL-FLY. AT In the somatic cells of both males and females one occasionally finds mitotic figures concaining a much larger number of chro- mosomes, but such cases are in the nature of exceptions and no - one would contend that they represent an average condition. If there is such a thing as constancy in the number of chromo- somes in the majority of somatic cells, the constant is in the neighborhood of 12 in both males and females of the bisexual generation. Because this is the number of chromosomes found in the second spermatocyte division, I2 is assumed to be the ap- proximate haploid number. Now in any case where an accurate count is difficult or impossible in the somatic cells, it is always possible to determine with certainty that the number is very much less than the expected diploid number 24. In view of the fact, that in the honey-bee it is said that the somatic mitoses show a very much higher number of chromosomes than occurs in the gonial cells, somatic mitoses should not be used as a safe and reliable method of determining the diploid number. There may however.be some significance in the fact that a large number of somatic cells of both males and females of Dryophanta contain a number of chromosomes that approximates the number found in the dividing spermatocyte rather than a multiple of this number. Any definite statement regarding the origin and significance of this condition must await examination of the maturation and cleavage spindles of the egg. However, the facts at hand do suggest that the males and females of the bisexual generation of Dryophanta develop from eggs whose chromosomes have under- gone reduction in maturation. The slightly large number of chromosomes found in the females somatic tissues may or may not be of significance, but if sex determination has its basis in the chromosomes, a difference in the method of distribution of the chromosomes in maturation may explain why some of these eggs develop parthenogenetically into females and others into males. In a recent paper Nachsheim has summed up in a general statement the results of investigations dealing with sex-de- termination in Hymenoptera as follows: “Die Mannchen der Hymenopteren entstehen aus unbefruchtete Eiern, die zwei 42 H. L. WIEMAN. Richtungskérper abgeschniirt und eine Reduktion ihrer Chromo- somenzahl erfahren haben. Sie besitzen also nur ein Chromo- somensortiment, das miitterliche, und infolgedessen muss in der Spermatogenese die Reduktionsteilung unterbleiben. Die Weib- chen der Hymenopteren besitzen beide Chromosomensortimente, also die diploide Chromosomenzahl in ihren somatischen Zellen, da sie aus befruchteten Eiern ihre Entstehung nehmen oder—bei den Blatt- und Gallwespen—zwar ebenfalls aus unbefruchteten Eiern, aber aus solchen, die den Reifungsteilungen ihre Chromo- somenzahl nicht reduciert haben; entweder findet in diesen Eiern iiberhaupt nur eine Reifungsteilung statt, oder beide Reifungs- teilungen sind Aquationsteilungen. Der zweite Richtungskorper kann also. . . . an Stelle der Spermatozoons treten, d.h. der zweite Richtungskérper bringt in Verbindung mit der Eikern dasselbe Geschlecht hervor wie der Eikern in Verbindung mit einem Spermakern”’ (pp. 220-221). My findings in the somatic chromosomes of Dryophanta raises ~ the question as to whether females of the bisexual generation are produced parthenogenetically from eggs that do not undergo reduction in maturation. An examination of maturation stages in the egg is necessary to decide this point and material for this purpose is being collected at the present time. LITERATURE CITED. Doncaster, L. : ’r011 Gametogenesis of the Gall-fly, Neuroterus lenticularis (Spathegaster baccarum). Parts I. and II. Proc. Roy. Soc., B., Vols. 82 and 83. Nachsheim, H. ’r3 Cytologische Studien tiber die Geschlechtsbestimmung bei der Honigbiene (Apis mellifica). Arch. f. Zellfschg. Bd. 11. Wilson, E. B. 713, A Chromatoid Body Simulating an Accessory Chromosome in Pentatoma. Biot. BULL., Vol. 24, 1913. 44 H. L. WIEMAN. EXPLANATION OF PLATES. The figures are camera drawings made at table level with Zeiss apochromatic objective, 1.5 mm. and compensating ocular, 12. There has been some reduction in reproduction. , PLATE I. Fic. 1. Primary spermatocyte at the end of the growth period. Male pupa. FIGS. 2, 3 AND 4. Primary spermatocytes undergoing changes in outline pre- liminary to the formation cf the polar body. Fics. 5 AND 6. Primary spermatocytes having chromosomes in the form of loops or split rods. Fics. 7 AND 8. Stages in the cutting off of the polar body. Fig. 8 contains a second polar body belonging to a cell in a neighboring section. Fic. 9. Prophase of the second spermatocyté division showing 12 chromosomes. Fic. 10. Side view of the second spermatocyte spindle at metaphase. BIOLOGICAL BULLETIN, VOL. XXVIII. PLATE |. H. L. WIEMAN. omy 46 H. L. WIEMAN. PLATE II. Fic. 11. Side view of the second spermatocyte spindle at metaphase. Fics. 12 AND 13. Polar views of the second spermatocyte spindle at metaphase showing 12 chromosomes. Fic. 14. Second spermatocyte spindle at late anaphase showing a free polar body near the upper end of the cell. Fic. 15. Second spermatocyte at telophase with a polar body attached to the upper daughter cell. Fic. 16. Early spermatid, reconstruction of the nuclei. Polar body fragments near the upper cell. FIGS. 17 AND 18. Stage in the transformation of spermatids into spermatozoa. Fic. 19. Metaphase chromosome group in the mitosis of a developing wing, showing 2 chromosomes. Young male pupa. FIGS. 20, 21 AND 22. Metaphase chromosome groups of ovarian follicle cells, showing 13, 14 and 13 chromosomes respectively. Late female larva. PLATE Il. BIOLOGICAL BULLETIN VOL. XxVII. H. L. WIEMAN. EGG ALBUMEN AS A CULTURE MEDIUM FOR CHICK IUSSUWIE,, OLIVE SWEZY. Egg albumen as a culture medium for chick tissue in vitro has received but scant attention from experimentalists, in spite of the fact that it forms the natural medium, in part at least, of the embryo chick. Ina recent series of experiments, however, results have been obtained which show that all the usual mani- festations of cell activity, noted by various observers in other culture media, were to be met with in cultures made from egg albumen, and have, I believe, demonstrated satisfactorily its: entire adaptability to that use. These experiments were carried on in the laboratory of Prof. S. J. Holmes, to whom my thanks are due for his kindness in giving advice and assistance through- out the course of the work. The technique followed has been that outlined by Burrows and Carrel, modified to suit the different conditions under which the work had to be carried on, using embryos varying in age from twenty-four hours to fourteen days. Of these it was found that the most successful results were obtained from embryos of from ten to fourteen days growth, though all showed considerable activity. Fragments of all the organs of the body, including the brain and spinal cord, were used, but the most active growths were obtained from the heart. Several series of preparations were made by cutting up the entire embryo into minute particles in a small amount of Ringer’s solution and egg albumen, stirring and shaking these rapidly for a few minutes and then placing a small drop of the mixture on the slide and sealing in the usual way. By this process cultures could be made containing but a few or even single cells. The medium used has been egg albumen alone or mixed with- varying proportions of egg yolk, -Ringer’s solution and extract of muscle tissue. Egg yolk proved entirely unsatisfactory because of the impossibility of seeing what was taking place within it. The best results were obtained from egg 47 48 OLIVE SWEZY. albumen alone and with mixtures of albumen and muscle tissue extract, the latter being prepared from embryo chick tissue and added to the albumen either before or after making the culture. Egg albumen coagulates to a more or less firm consistency and thus gives one of the conditions apparently requisite for the growth and activity of the tissue cells. Owing to the viscosity of the albumen, considerable care is necessary in handling the specimens when it becomes needful to transfer the culture to a fresh medium, the usual method of procedure being to cut away the old albumen with a sharp knife. When, as is frequently the case, the outgrowth seemed to be mainly on the surface of the glass, and thus could not be trans- ferred in the usual way without the loss of the greater part of the growth, another method was used. Inverting the cover glass the albumen was removed with forceps and pipette, several changes of Ringer’s solution successively placed over the culture and, after removal of this, a fresh drop of albumen was added to the culture and it was again sealed up. The latent period, before the beginning of activity of the culture, lasted from half an hour to several days. Usually, in good preparation, active amceboid movements began within half an hour after being put on the slide. At that time along the border of the tissue could be seen the elongated, outpushing cells forming a fringe along what was before a clear cut outline, with a few scattered cells lying at some little distance from the main mass. These cells displayed very active amoeboid movements that are less common in the older cultures though still present to some extent. When these cells are chilled or disturbed they contract and become rounded. Ona number of cultures groups of cells showed long clear processes extending outward, some- times branched, with the ends breaking up into short filaments. These were in all cases cultures which included portions of the brain or spinal cord from a four-day chick. An attempt was made to photograph one of these cultures but the length of time necessary was sufficient to chill the slide and, on examination, it was found that the processes had all been retracted. Subse- quent incubation had no effect on the culture, though disinte- gration did not take place for several days. In all the cultures EGG ALBUMEN AS CULTURE MEDIUM FOR CHICK TISSUE. 49 these processes disappeared, were retracted apparently, in the course of fifty to seventy hours and no further evidences of them were seen. In the preparations made by shaking up the finely cut embryo with Ringer’s solution, a greater or less number of single cells were found. In the course of a few days these were greatly increased in number with a distinct massing together. of the cells, usually along the outer border of the drop of albumen. Owing to accidents of various kinds these were not carried along far enough to show the tissue formation noted by Carrel. The most marked instance of tissue formation was that appar- ent in a culture made from the heart of a fourteen day chick, which, at the end of twenty days was encircled by a new forma- tion five times the diameter of the original piece of tissue. This new formation was several cells in thickness and composed of fusiform and polygonal cells, sometimes massed together, forming a network, or in other places showing distinct cell boundaries. Among these cells many showed division figures at various stages. Around the outer margin of the mass of cells and extending nearly three-fourths of the entire distance around it, the cells had taken on a different character. Here they had become flat, thin and- elongated in a direction parallel with the margin of the circle. This formation was several cells in thickness with the cells closely matted together and forming a distinct boundary that was conspicuous without the aid of alens. The remaining one-fourth of the margin was occupied by cells actively pushing outward. To test the effects of cold on the growth of the tissues, the embryo was sealed up in a stender dish containing Ringer’s solution and placed in the ice box of the refrigerator with the temperature but a few degrees above zero, Centigrade. The first of these was used the second day and behaved like normal tissue. Most of those kept in the refrigerator for a number of days became infected with bacteria. The longest period of cold storage which gave successful cultures was four days, from Jan- uary 31 to February 4. One half hour after making the cultures from this embryo the cells were moving out in an active condition in four out of the sixteen cultures made. The subsequent history of these cultures was the same as that of unrefrigerated tissue. The longest period during which tissues have been kept alive 50 - OLIVE SWEZY. without any evidences of necrobiosis has been ninety-three days, and in the majority of these cases death has been caused by in- fection with bacteria or molds or other accidents, and, not, apparently, by any lack of vigor in the tissues themselves. This, in general, seems to be true of most of the cultures which appear to be in a thriving condition after the second day or third day, and especially where renewals of the culture medium have been frequent, and precautions have been taken to avoid tearing or otherwise injuring the tissues. However disintegration fre- quently takes place from no apparent cause. Egg albumen presents some difficulties when a stained prep- aration from the culture is desired, on account of its avidity for stains. In the first stained preparations made it was impossible to distinguish the outlines of the cells, and the study of the specimen seemed a hopeless task. This difficulty was later over- come by the following methods: the cover glass was inverted and placed on the mouth of a vial containing a quantity of osmic acid. The mouth of the vial was small enough to be completely covered by the cover glass and yet not touch the preparation. After fixing in this manner for ten minutes the cover glass was placed in a stender dish containing distilled water and left for a number of hours. Frequent agitation and changes of the water removes the greater part of the albumen, leaving the tissue adhering to the glass, which may then be put through the alcohols and stained in the usual way. With this method very clear preparations may be obtained. ZOOLOGICAL LABORATORY, UNIVERSITY OF CALIFORNIA, BERKELEY, CAL., October 13, 1914. THE INFLUENCE OF PRODUCTS OF PATHOLOGIC METABOLISM ON THE DEVELOPING TELEOST OVUM. KE. I. WERBER, DEPARTMENT OF BIOLOGY, PRINCETON UNIVERSITY. In his recent work on pathological human ova, after careful sifting of anatomical evidence, Mall! arrives at the conclusion that the failure of large numbers of ova to develop normally is to be traced to diseases of the uterus. According to his view, which is supported by obstetrical and gynecological data, diseases of the uterus are the primary cause of the faulty implantation of the ovum. This in turn makes proper nutrition of the developing embryo impossible thus leading to various degrees of malforma- tions by arresting development. The deformed embryo is eventually aborted after it has exhausted its inadequate means of subsistence in the uterus. Full-term monsters would be born from such deformed embryos if they were not hindered in their further development by starvation. According to this theory, - therefore, an apparently healthy ovum discharged into a diseased uterus fails to develop normally owing to its defective implan- tation. Mall studied largely pathological ova of the first two months and the interpretation of the numerous cases described by him seems justified. Practically all pathological ova of the early months studied by him as well as by other investigators, exhibited the condition of faulty implantation, so that it is not unwarranted to regard this condition as the direct cause of monstrous develop- ment. A consideration of some instances of arrested, defective or even monstrous development found after full-term birth would suggest, however, that there must be also some other factors 1 Mall, F. P., ‘‘A Study of the Causes Underlying the Origin of Human Mon- sters.” Journ. of Morphology, Vol. X1X., 1908; “The Pathology of the Human Ovum” in Keibel-Mall ‘‘Handbook of Human Embryology,” 1910. 5st 52 E. I. WERBER. which primarily interfere with normal development. Such defects as rudimentary development of one or both eyes, con- genital absence of both arms, hydrocephalus, possibly also cases of congenital deafness, to mention only a few that are well known to occur, can, in the writer’s opinion, hardly be traced to defective implantation. The results of investigations in experimental teratology by Panum,! Dareste? and more recently by Stockard’ and Bardeen* would seem to suggest that some physico-chemical factors may be at work in a great number of cases of pathological development. These factors may in some instances be the primary cause of terata, while in other cases they may be only secondary contributing causes. The experimental teratologists subjected developing ova in very early stages to changes in the physico-chemical nature of the environment and found that various monstrosities could be produced under these conditions. It was impossible, however, for them to control the results of experimentation, as they could not predict the type of monster which would result from the employment of the same factors. The experiments of Stockard, where a more or less definite monstrosity—cyclopia or monoph- thalmia—appeared with considerable certainty in a large per- centage of embryos developing in magnesium chloride or alcohol solutions, mark a distinct progress in this field of inquiry, because they paved the way towards experimental control of monstrosities occurring in nature. To the writer Stockard’s work suggested the possibility that the monstrosities met with in higher animals and man may to a certain extent be due to the influence of injurious substances found in the circulation under pathological conditions. While this hypothesis could not be applied to bacterial toxins on account of insufficient knowledge, it seemed that some substances thrown into the circulation in various metabolic diseases may be re- 1Panum, ‘‘Entstehung der Missbildungen,”’ 1860. 2 Dareste, ‘‘Recherches sur la production de monstrosites,’’ Paris, 1891. 3 Stockard, C. R., ‘‘ The Artificial Production of a Single Median Cyclopean Eye in the Fish Embryo by Means of Seawater Solutions of Magnesium Chlorid,” Arch. f. Entwmech., Vol. XXII., 1907; ‘‘The Influence of Alcohol and Other Anzs- thetics on Embryonic Development,”’ Am. Jour. of Anat., Vol. X., I910. 4 Bardeen, C. R., Jour. of Experimental Zool., 1907; Am. Jour. of Anat., Vol. XI. TELEOST OVUM. 53 sponsible for pathological development. Thus the etiology of defective or monstrous development would be traced to the pathological metabolism of the mother or possibly even of the father. For, as Bardeen! has shown, a normal, healthy ovum of the toad, if fertilized with sperm which had been injured by exposure to the action of X-rays, will give rise to a deformed embryo. With this idea in mind the writer conducted during the summer of 1914 experiments on eggs of Fundulus heteroclitus. The eggs of this fish are easily obtained at Woods Hole and are excellent material for experimentation. The investigations on the fish eggs are of a preliminary character, and were undertaken to ascertain the influence of some toxic substances occurring in pathological metabolism on the developing egg. The number of these substances being rather large while the spawning season is limited to a few weeks, it was impossible to try more than a few of the chemicals. Urea, butyric acid, lactic acid, sodium glycocholate, acetone and ammonium hydroxide were tried as to their effect on the development of fertilized eggs. Definite results were so far obtained only with butyric acid and acetone. Ten c.c. of a 1/12—1/14 molecular solution in 50 c.c. of sea water was found to give the greatest number of monsters when butyric acid was used. The eggs were submitted to the action of this solution for 20 hours after they had reached the eight-cell or sixteen-cell stage, 7. e., 3 to 3% hours after fertilization. While under this procedure numerous monstrosities were at first ob- tained, the method failed almost completely in later experiments. I therefore employed developing eggs in the first stages of division (2- and 4-cell stages) when many monstrosities were produced even after a sojourn of thirty hours in the butyric acid solution. But it seems to me that the reason why the method failed with the eggs in more advanced cleavage stages was that the time of exposure was too long, as very many eggs were dead by the end of that treatment, and that with an exposure of 10 or 15 hours better results would have been obtained. There is, however, as important difference in the effect which 1 Bardeen, C. R., Jour. of Experimental Zool., 1907; Am. Jour. of Anat., Vol. XI. 54 E. I. WERBER. this toxic substance has upon developing eggs in the first and second or in the third and fourth divisions. In the former case anterior hemiembryos, dwarf embryos with deformities of the eyes or of the otic vesicle, and malformations of the most extreme kind were predominant, while in the latter deformities of the eye such as cyclopia and monophthalmia, etc., were mostly observed. In either case, however, there were very few embryos in which only the nervous system was affected. In most of the deformed embryos all organ systems were more or less involved in the malformation. Similar results were obtained with acetone in sea water, varying in concentration from 20-50 c.c. of a molecular solution in 50 c.c. of sea water. In this mixture the eggs remained from 24-72 hours from the eight-cell or sixteen-cell stage. In every case great numbers! of monsters similar to those already mentioned were produced. The monstrosities in both series of experiments with butyric acid and acetone being essentially alike it will not be necessary to describe separately the deformities produced by each. Cyclopia and asymmetric monophthalmia were found to occur rather abundantly. There were also some cases of asymmetric monopththalmia in which an open orbit was found on the side lacking the eye. It is of some interest to note in this connection that the eyeless orbit in such cases is usually closed on the outside by periorbital tissues. The anatomy of the head of such embryos may probably reveal some interesting conditions. Other cases of asymmetric monophthalmia were found in which an apparently free eye had developed on the yolk-sac at a considerable distance from the embryo. Probably the most striking of the results obtained in this investigation were some eggs in which nothing could be observed but an eye. In only one case this eye seemed to be perfectly developed, while the other solitary eyes had ‘“‘coloboma’’-defects, the fissure of the chorioid still being patent. Only a few (five or six) of these malformations are recorded, but in spite of their rare occurrence they are very sig- nificant from the standpoint of experimental embryology. At the 1 No attempt was made to ascertain the percentage of the deformities found in these experiments, this part of the work being deferred to later investigation. TELEOST OVUM. 55 present time it is, obviously, impossible to account for the occur- rence of these remarkable cases. However, it is hoped that an anatomical investigation of early stages in the development of eggs subjected to the influence of the environmental modi- fications used in these experiments, may give at least a clue as to what may have happened in the development of these eggs. Practically all other known deformities of the eye such as total blindness, or presence of lenses only, or presence of supernumerary lenses were frequently found. To the student of the physiology of development the occurrence in these experiments of large numbers of anterior hemiembryos which seem to be closely analogous to those obtained by mechan- ical means by Roux,! Endres,? Morgan? and K. Ziegler‘ will be of special interest. As will be pointed out soon the formation of the hemiembryos in these experiments may also possibly be due to similar factors. A great number of embryos were hydrocephalic and so far as could be determined it is reasonable to expect that an anatomical investigation may reveal in some deformed embryos oedematous conditions, also herniae and other mechanical obstructions which played a part in their formation. Striking abnormalities of the heart and blood-vascular system were found in all malformed embryos with the exception of those which showed only median cyclopia. Some were entirely devoid of the heart, while other possessed an exceedingly delicate tube in its place which was practically straight and of about the size of the intestinal blood vessels in a normal embryo of a corre- sponding stage. The rate of the heart beat varies with the degree of the abnormality of the organ, and is, as a rule, very slow in all monstrous embryos. The range of variation in the develop- ment of the blood vessels is very wide. There may be merely blood islands scattered on the yolk-sac, rudimentary, imperfectly connected, or in some instances more or less normal vessels. 1 Roux, W., ““Gesammelte Abhandlungen zur Entwicklungsmechanik der Or- ganismen,”’ II., 1895. 2 Endres, H., ““Anstichversuche an Froscheiern,’”’ Sitzber. d. zool.-bot. Sektion d. schlesischen Ges. f. vaterlandische Kultur, 1894. 3’ Morgan, T. H., “The Formation of the Embryo of the Frog,’’ Anat. Anz., 1894. 4 Ziegler, K., “Zur Postgenerationsfrage,’’ Anatomische Hefte, Vol. LXVI., 1902. 56 E. I. WERBER. Twins were found only in a few cases and only once were true ‘‘Siamese’’ twins observed. They were much deformed, had one common heart and only vestigial eyes. Several eggs were recorded, in which an anterior duplicity had developed. In one of these latter cases the components of the duplicity were totally blind, hydrocephalic, their hearts were very delicate, the blood vessels rudimentary and the yolk-sac was covered with dense networks of richly pigmented blood islands. These monstrous embryos hatch only very rarely, most of them dying after the development has reached the stage in which the remant of the yolk-sac is in the normal embryo converted into the anterior body wall. As far as could be determined from the embryos in toto it is the enormously large (oedematous?) peri- cardia that mechanically obstruct the formation of the ventral body wall. The correctness of this interpretation will be tested by microscopic sections of these embryos. The mechanism of the formation of the described monsters can at this time not even be definitely suggested. The observation was made that the yolk-sac in all extremely malformed embryos shows a marked decrease in size as compared with that of normal eggs of the corresponding stage of development. The greater the degree of injury inflicted on the embryo the smaller the yolk- sac. It is not impossible that the chemicals used in these experi- ments indirectly bring about this decrease in the size of the yolk-sac. For it was noticed that the chemicals used in these experiments softened the egg-membrane considerably, a fact which suggests an increase in the permeability of the egg. Owing to both increased permeability of the germ-disc cells and to internal osmotic pressure of the yolk-sac, an escape of sub- - stance from the yolk-sac might have been caused, which, being forced out at different points of the yolk-sac, might have frag- mented the germ-disc. Many eggs were observed in which this fragmentation of the germ-disc was very evident. Some parts of this ruptured germ-disc may be so badly damaged as not to beable to develop further, while the remaining fragments, even if they are very small, may still give rise to various monsters, hemiembryos, dwarfs or even toa solitary eye. Or possibly the decrease in size of the yolk-sac of malformed ova may point to TELEOST OVUM. 57 elimination of both yolk-sac and germ substance as an effect of the solvent action of the chemicals to which the eggs were ex- posed. Whatever the mechanism involved in the production of -the recorded pathological ova may be, at the present time, it can hardly be more than conjectured. It will be the object of future investigations to find a satisfactory answer to this open question. There seems to be a close similarity between these cases where parts of the germ-disc are apparently lost through elimination of some kind and the production of hemiembryos by mechanically injuring one of the blastomeres of the developing frog’s egg, as described by Roux, Morgan and other investigators. The writer intends to continue this work on the teleost eggs as well as on the amphibian and hen’s eggs. He also hopes that he may in the near future secure adequate facilities for carrying on experiments on the influence of the toxic substances of patho- logical metabolism on the development of the mammalian em- bryo. The plan of this work would be to mate animals in which metabolic disturbances had been produced experimentally. A complete description and analysis of the results obtained in the investigation reported here will be published at an early date. The writer takes pleasure in acknowledging his indebtedness to Professor C. R. Stockard of Cornell University Medical College with whom he on several occasions had discussed some phases of the work, and from whom he has received valuable suggestions regarding preservation of material. MARINE BIOLOGICAL LABORATORY, Woops Hote, Mass., September 3, 1914. 5; Sy é 3 ee ag Be “ Ly gH es ae OF oie ed _ Marine Biological aborat } “5 ue a nae -woons HOLE, “MASS. AS Wath et - Fepevary, 101 a CONTENTS RYE a be. ee oe an. ail eee Produced Lntrauterine Pregnancy and See the: Spontaneous Parthenogenesis ee 4 the Eggs. 21: the Ovary ae the Coe We Wio eg 6 8 ale bie wae wet ee eee ee ey os oe 4 “Hon. re Spores tm an Annelid.... een ge? - [oe Brothits Ties a the Réle™ or Rei pane Formation om lee: gs fe Ape LEuacrineiae: Pe of Fire gation In Platynerets MOZUOPS.. se vtNy - PumnisHED Marine? BY THE “MARINE. BIOLOGICAL LABORATORY. ae ae ee Si age “PRINTED! ‘AND ISSUED BY sestees us “THE NEW ERA PRINTING COMPANY POEs CS re tae LANCASTER, BAD es Me Loans. PoveAl iy mel, AGENT FOR Gama BRITAIN ue Res shot ee aie) Acent, FOR Germany : WILLIAM WESLEY ne ahs RR PRIEDLANDER & SON - = we See bei BO BOETNE 8. Essex Street, Strand eG ey Dre: ~ Berlin, No Wo ; a Londen, We C rae MESS Carlsirasse, 17 fee = “under Act of Congress of J aly 16: 1894) iby Se mab shins? Cone ce a RAPA UY © ant Volenee Vie February, 1915. No. 2. PolOLOGICAL BULLETIN ee ee AN EARLY STAGE OF AN EXPERIMENTALLY PRO- DUCED EXTRAUTERINE PREGNANCY AND THE SPONTANEOUS PARTHENOGENESIS OF THE EGGS IN THE OVARY OF THE GUINEA PIG:! LEO LOBB. The observation on which I wish to report is of great interest from several points of view. It explains the negative result of our former attempts which aimed at producing experimentally an extrauterine pregnancy in the guinea pig. It contributes to the understanding of the mechanism of the sexual cycle and it makes certain my previous conclusions, which formerly had only been probable, concerning the fargoing parthenogenetic develop- ‘ment of ova in the ovary of the guinea pig, conclusions which our previous studies had made very probable. Ina great number of previous experiments we made incisions in various parts of the uterus of the guinea pig and at different times after copulation.? Under these circumstances it certainly must often have happened that fertilized eggs left the uterine cavity. But_extrauterine pregnancy did in no case take place under such circumstances. Even after ligation of the fallopian tubes we were not able to observe the occurrence of an extrauterine pregnancy. This latter observation is in accordance with some experiments of Mandl and Schmidt.’ It was of interest to determine what was the fate of the ova which left the lumen of the uterus and passed into the peritoneal cavity after fertilization. An observation 1 From the pathological laboratory of the Barnard Free Skin and Cancer Hos- pital, St. Louis. 2Leo Loeb and John W. Hunter, University of Pennsylvania Medical Bulletin, Dec., 1908. 3 Archiv f. Gynaecol., 56, 1898. 59 u“ f DS 60 LEO LOEB. which we made in the course of our continued experiments serves to clear up this point. Two days, sixteen hours after copulation, incisions were made into the uterus of a guinea pig. The weight of the animal at the time of the operation was 550 grams. ‘The incisions were longi- tudinal and extended through both horns of the uterus up to near the point of juncture with the tubes. Besides the longi- tudinal incisions a number of transverse incisions into the uterine wall were made. Eighteen days after copulation uterus as well as one of the ovaries was taken out for examination. The ovary was cut into serial sections. Small follicles in the early stages of development, as well as other follicles in early stages of con- nective tissue atresia were found. In addition there were many follicles in the last stages of follicular atresia. There were also present several young corpora lutea, the center of which was partly filled out by connective tissue, while the center of the cavity had not yet been organized by connective tissue. There were furthermore present corpora lutea in an early stage of retro- gression, as well as yellow bodies, completely atretic corpora lutea. These findings correspond to an ovary about three days after ovulation. Microscopic examination of those parts of the uterus which had not been incised during the operation showed cylindrical surface and glandular epithelium with numerous mitoses in the glandular ducts. The fundi of the glands are somewhat smaller. In the lumen of the uterus there are some polynuclear leucocytes, a greater number of which are found in the ducts of the glands. In the connective tissue of the mucosa as well as in the surface epithelium the presence of several small round cells is noted. There are very few mitoses in the connective tissue of the mucosa which is rich in nuclei. These findings correspond to a condition of the uterus about 3-3% days after copulation. Near the tubal end of one of the uterine horns, not far from the usual situation of the ovary there was a small nodule. This nodule was cut in serial sections, and its structure is best ex- plained by referring to the illustrations. Fig. 1 shows the position of the embryo. (a) It lies in the neighborhood of the fallopian tube. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 61 \ (b) Some distance from the embryo we see the cut wall of the uterus. -(c) In the direction towards the tube we see the musculature of the uterus, in the opposite direction the epithelium with the Fic. 1. Low power. a, embryo; 0, Fallopian tubes; c, everted walls of the uterus. A more detailed explanation of the figures is found in the text. FIGS. I, 2, 3, 4 and 6 are from microphotographs. Fig. 5 from a drawing. glands and the connective tissue is visible. In the detached part of c the glands have the character of mucous glands. At this place the mucosa of the uterus is everted as a result of the incision. If we follow on further sections the position of the placenta which surrounds the embryo proper, we find that at some distance from the embryo proper it dips into the peritoneal side of the uterus at a place above the beginning of the incisions, where therefore the uterine lumen is still intact, and it even penetrates into a fissure of the musculature of the uterus. Further downwards the embryonal placenta extends to the peritoneal tissues of the upper part of the incised uterus. Fig. 1 of course represents only one section while the description which we just gave is based on a study of a number of serial sections. The egg embedded itself 62 LEO LOEB. evidently in the connective tissue between the upper end of the uterus and the lower end of the tube and its derivatives pene- trated still deeper between the musculature of the uterus in the direction from the peritoneal side. The character of the embryonal structures and their relation to the surrounding tissue are more clearly shown on Fig. 2. 6 is Fic. 2. Thedeveloping embryo; somewhat higher magnification. a, embryonal structure (neural canal?); 6, Fallopian tube; c, giant cells of the embryonal placenta; d, cuboidal cells of the embryonal placenta surrounding cavities; h, hemorrhages in the surrounding connective tissue. the fallopian tube. a is the embryo, which is surrounded by placental structures c and d and other similar not especially designated structures. In the periphery of these structures are — found extensive hemorrhages into the connective tissues and these are in turn surrounded by strands of connective tissue and by blood vessels. The entire region between the tubes and the outer hemorrhagic zone is filled out by embryonal placenta. The embryo proper corresponds to a developing guinea pig at a stage directly following the formation of the germ layers. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 63 A points to a central structure, which probably corresponds to the Anlage of the neural tube. Under the abnormal conditions under which the embryo must develop, the various embryonic structures are evidently somewhat distorted. Fig. 3 shows the Cc a d i i Fic. 3. The embryo proper, higher magnification. a, mitosis; c, surrounding giant cells; d, a structure which perhaps corresponds to the placental cavities lined with cuboidal cells. f central part of the embryo at a higher magnification. A points to the same cavity as a in Fig. 2. The cell designated by a is seen in the process of mitotic division. Other embryonal cells also divide mitotically at various places. Surrounding the central parts of the embryo we find epithelial structures arranged in layers adjoining as is shown on Fig. 2. Giant cells c surround the embryo at various places on Fig. 3 in a similar way as seen on Fig. 2. A larger number of giant cells are also found at a some- what greater distance from the embryo. These giant cells are arranged typically around cavities, which are lined by smaller cuboidal cells. Don Fig. 2 points to such a cavity lined with such cuboidal cells. Perhaps also the canal d on Fig. 3 corresponds to such a cavity. The small cuboidal cells often proliferate and 64 LEO LOEB. their proliferation leads to the formation of papillary excrescences into the cavities. These excrescences fill sometimes a great part of these cavities. Mitoses often appear in these cuboidal cells. Fig. 4 shows such a placental structure at a higher magnification. a d | Fic. 4. A placental cavity lined with cuboidal cells.: c, giant cells; d, cuboidal cells lining a cavity and forming papillary excrescences; c’, a giant cell penetrating into the surrounding connective tissue; v, blood vessels; , hemorrhages in the con- nective tissue; w, experimentally misplaced uterine epithelium. d points to a cavity lined with cuboidal cells. The cuboidal cells form papillary proliferations into the lumen. The cavity bulges into the surrounding tissue at d1. The cavity is surrounded on several sides by giant cells c and these giant cells protrude into the cavity and divide it into two parts. These giant cells have the power to penetrate farther into the surrounding tissue inde- pendently. Ci represents such a giant cell, which penetrates into the surrounding’ fibrous tissue. Surrounding this placental structure we find connective tissue in which there are many hemor- thages h. U represents a cavity lined with uterine epithelium. v represents a blood vessel. Fig. 5 represents a drawing of a similar placental structure. D represents the cavities lined with cuboidal cells, and partly filled with the proliferated cuboidal EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 65 cells. C are the giant cells surrounding the cuboidal cells. F is fibrillary connective tissue in which there are many hemorrhages h. There is nowhere a formation of adecidua. visa blood vessel. Fic. 5. A typical placental structure. The various letters have the same sig- nificance as in Fig. 4. As we have already seen on Fig. 4, the giant cells penetrate deeper into the tissue, independently of the small cuboidal cells. They prefer especially the neighborhood of blood vessels, pene- trate the walls of the latter and replace the endothelial cells. Blood vessels thus changed are of course thereby weakened, and they are no longer as well able to resist to the full extent the blood pressure, and thus hemorrhages into the tissue, as so frequently seen, result. Fig. 6 shows two vessels v. Giant cells c have advanced up to the lumen of these vessels and substitute the endothelial cells. In ci also there lies a giant cell in the tissue. At many places there are hemorrhages h in the connective tissue. In the periphery of the upper half of the section, connective 66 LEO LOEB. tissue surrounds the structure. d points to a cavity filled with small cuboidal cells. These findings will have to be interpreted in the following way. At the time when the incisions were made into the uterus, namely two days and sixteen hours after copulation, the ova had probably already left the tube and had reached the upper part of the uterine cavity. At this time one or more of the ova left the uterine cavity through the incision into the uterine wall Fic. 6. Placental embryonal giant cells penetrate into the vessel wall. v, blood- vessels; c, giant cells substitute vascular endothelial cells; c’, a giant cell lying in the host tissue; 4, hemorrhages in the host connective tissue; d, cuboidal placental cells of embryonal origin. and one of the ova passed around the outer side of the upper end of the left uterine horn, and embedded itself in the connective tissue between the tube and the upper end of the left uterine ‘horn. A part of the embryonal placenta in the course of de- velopment penetrated farther into the musculature of the uterine horn. ‘The fertilization of this ovum had in accordance with the EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 67 general view concerning the time of fertilization of the guinea- pig ovum already taken place at the time of the operation. We excised the nodule fifteen days eight hours after the incisions had ‘been made. Our description of the embryo clearly shows that under the existing abnormal conditions the development of the ovum was greatly retarded. The embryo is still alive and even growing, as the mitoses, which were found at various places, indicate, but the embryo is found to be at a very much earlier stage of development than one would expect eighteen days after copulation. The embryonal placenta also is only very incompletely developed. While the normal placenta of the guinea pig shows a complicated structure at this period of development, in our case the embryonal placenta consist solely of layers of small cuboidal cells, which usually line cavities, and produce papillary excrescences pro- jecting into the cavities. On the outer side of these cavities _ there are giant cells. The giant cells penetrate also independently into the surrounding connective tissue and substitute walls of blood vessels, and thus contribute to the hemorrhages which we find so frequently. Cuboidal cells as well as giant cells are growing actively by mitosis—the latter however to a lesser degree. The surrounding host tissue remains passive. The embryonal tissue is surrounded by fibrillar connective tissue containing the ordinary connective tissue cells. There 1s no- where an attempt at the formation of a decidua on the part of the host tissue. These observations are in entire accord with our former ex- perimental findings from which we concluded that in the guinea pig solely the connective tissue of the uterine mucosa is able to produce decidua in response ‘to artificial stimuli, as cuts and foreign bodies while the fallopian tube, peritoneal and other con- nective tissue are unable to do so.! These additional observations again prove the similarity 1m the mode of action of the artificial stimuli leading to the formation of a decidua on the one hand and of the ovum on the other hand. Ina similar manner as the artificial stumula were not able to call forth a 1Leo Loeb, Zentralblatt fiir Physiol., Bd. XXIII., No. 3; Journal Am. Med. Association, Vol. LIII., p. 1471, 1909. 68 . LEO LOEB. formation of decidua in the peritoneal connective tissue, the ovum is likewise unable to do so. These observations furthermore clear up the fate of the ovum in cases in which it is not able to develop normally in the uterine wall. Frequently a fixation of the ovum does not take place in such abnormal cases, especially on the smooth peritoneal epi- thelium. In other cases however the ovum fixes itself and begins to develop in the connective: tissue without however finding the necessary decidual reaction on the part of the surrounding con- nective tissue. In such cases the development of the embryo proper as well as of the embryonal placenta is very much re- tarded as compared to the normal development; furthermore the embryonal differentiation also remains incomplete and we may assume, that after some time the growth ceases and the em- bryonal structure is substituted by host connective tissue in a similar manner as in the ovary of the guinea pig. We find there- fore in the guinea pig no or only a very much retarded and in- . complete development of the ovum outside of the uterus. This is in all probability due to the fact that the host tissue is not suited to receive the ovum and to supply it with the necessary food stuffs. In this case the host tissue behaves passively in contradistinction to the uterine mucosa. This conclusion agrees — with the fact that we find a general parallelism in the ability of the uterine mucosa to produce decidua or deciduomata and to permit a normal development of the ovum. As I have previ- ously shown, various experimental interferences, as for instance extirpation of the corpora lutea or of the ovaries, have approxi- mately to the same extent an inhibiting influence on the develop- ment of deciduomata and of pregnancy. We may thus conclude that the ability of the host connective tissue to produce a decidua in a normal manner is of significance for the normal development of pregnancy. We see therefore that in the guinea pig the ovum does either not develop at all outside of the uterus or in case an extrauterine fixation of the ovum should take place, the development is much retarded and soon comes to a standstill. As our present and 1 As we shall later especially emphasize, the same holds good in the case of the parthenogenetic development of the egg in the ovary of the guinea pig. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 69 especially our previous observations concerning the partheno- genetic pregnancy in the ovary of the guinea pig demonstrate,} the development of the embryonal placenta preponderates re- latively very much over that of the embryo proper, probably because as I have already suggested, in contact with the host tissue the derivatives of the ovum produce mainly the placental structures. This is very marked in the case of the partheno- genetic development in the ovary of the guinea pig, where in typical cases under those conditions placental structures are found exclusively and only exceptionally the embryo proper begins to develop. These observations explain apparently very well the fact, that while in the guinea pig a further going development of the ovum is possible after extrauterine fixation, in man a complete extra- uterine development is not an infrequent occurrence. Our findings suggest as one of the causes for this difference in occur- rence of extrauterine pregnancy in man and guinea pig, the fact that in the case of man the host tissue offers a more suitable soil than in the case of guinea pig; while as we saw in the latter the development of the decidua in response to various kinds of stimuli takes only place in the connective tissue of the uterine mucosa, in the case of man the connective tissue of various pelvic organs and even the appendix is able to produce decidua as many observations show. In accordance with this interpretation a number of observers actually reported the development of a decidua in the fallopian tube in cases of tubal pregnancy. It is very probable that in tases of tubal pregnancy in which a decidua was not found in the tube, we had to deal with stages in which the chorionic wandercells had already penetrated deep into the host tissue and thus gradually destroyed the decidua; in a similar manner in the case of the guinea pig it can readily be seen that the wandercells of the embryonal placenta destroy a greater part of the decidua. It is very probable that from a certain stage of embryonal development on, the decidua is no longer in- dispensable as far as the continued existence and further develop- ment of pregnancy are concerned. 1 Roux’s Archiv, Bd. XXXII., p. 662, 1911; Zeitschrift f. Krebsforschung, 11. Bd.» 2. Heft, 1912. 70 LEO LOEB. Our observations are also of interest from another point of view. We know that under ordinary circumstances the corpus luteum remains longer preserved in pregnancy than in the non pregnant animal. Pregnancy prolongs the sexual cycle. We may now inquire into the cause of the prolongation of the life of the corpus luteum during pregnancy. Several years ago I pointed out, that the growth of the embryo might perhaps directly or indirectly prolong the life of the corpus luteum during preg-. nancy.! | Now we find in our case a small embryo as well as an embryonal placenta developing outside the uterus. Notwithstanding this fact a new ovulation had taken place about three days pre- viously and accordingly the corpora lutea of the preceding sexual cycle which had been terminated at the time of the last ovulation were degenerated. This observation proves that a developing embryo including embryonal placenta is in itself not sufficient to protect the corpus luteum from degeneration, and to prevent a new ovulation. It -is possible that the maternal placenta is concerned in the pro- longation of the life of the corpus luteum either alone or in con- nection with the embryo, which latter as our further experiments have shown, prolongs noticeably the life of the experimental placentomata (deciduomata). Indeed experiments which I carried out some time ago have shown that the development of deciduomata without the development of an embryo is able to prolong the sexual period; while normally the sexual cycle in the guinea pig has a duration of from 15 to 18 days, it lasts from 20-30 days after production of deciduomata.2. Whether as a result of these experimental interferences also the life of the corpus luteum is prolonged will have to be still further investigated. Our observations are furthermore of significance for the inter- pretation of certain structures, which I found in about 5 per cent. of the ovaries of young guinea pigs.’ In as much as these structures become absorbed after a certain time and are substituted by connective tissue, these structures 1 Zentralblatt f. Physiol., Bd. XXIV., Nr. 6; Medical Record, June 25, 1910. 2 Leo Loeb, BIoLoGicAL BULLETIN, Vol. XXVII., July, 1914. 3 Arch. f. mikrosk. Anatomie, Bd. 65, 1905; Roux’s Archiv, Bd. XXXII., p. 662, 1911; Zeitschrift f. Krebsforschung, 11. Band; 2. Heft, 1912. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 71 must in fact occur more frequently than the direct findings suggest. We have to deal with formations which resemble closely structures of the embryonal placenta, and they originate in ovarian follicles. They are either well preserved or are found in the process of retrogression and in the end are substituted by connective tissue. In two cases I was able to find besides em- bryonal structures proper, for instance the Anlage of the nervous system. It had been known previously and I myself had de- scribed processes which had to be interpreted as the first seg- mentations of eggs in atretic follicles which in consequence of’ the abnormal conditions under which they took place followed as might have been expected an abnormal course.!. The interpre- tation that we have to deal merely with the disintegration of the ova can be excluded with certainty. Such an interpretation would be contradicted by the regularity of the divisions. Fur- thermore we may find in these various segments either nuclei or the remnants of nuclear spindles and I was able to observe the simultaneous presence of a mitosis in each one of the two such segments. These segmentations also are found chiefly in the ovaries of the young guinea pigs. A somewhat furthergoing formation of the first segments in ovarian eggs has recently been described in armadillo by Newman.’ ' In all these cases we have merely to deal with the first parthe- nogenetic segmentations of the ovum, while our observations in the ovary of the guinea pig prove a much furthergoing develop- ment leading to the formation of embryonal placenta and of embryos in the stage of the germ layers within the ovary. It is of course natural, as I emphasized on a former occasion, that under these abnormal conditions the processes of development cannot follow an altogether normal course, and it was therefore tLeo Loeb, “On Progressive Changes in the Ova in Mammalian Ovaries, Journal of Medical Research, Vol. VI., 1901. Arch. f. mikrosk. Anat., Bd. 65, 1905. 2H. H. Newman, BIOLOGICAL BULLETIN, XXV., p. 52, 1913. It may be espe- cially emphasized that our interpretation of the placental and embryonal structures found by us in the ovaries of guinea pigs does in no way depend on the interpreta- tion of those changes in the ova within the ovaries of the guinea pig which in common with previous authors we held to be early abnormal segmentations of ova, while a number of other investigators interpreted them as of a degenerative character. There can be no doubt about the presence of further developed embryonal structures in the ovaries of guinea pigs. 72. LEO LOEB. desirable that a confirmation of our interpretation of these ovarian structures should be obtained. The findings which we have just communicated offer the desired confirmation. In our new ob- servations we have also to deal with embryonal structures found in the peritoneal connective tissue and developing in an abnormal situation without being aided by the host tissue through the formation of a decidua. We have of course to consider the fact that in the ovary the limitation of space is still more marked than in the connective tissue on the outer side of the fallopian tube and of the uterus. In both cases the placental structures preponderate over the embryonal ones proper; in both a retardation in the de- velopment is found and a preponderance of certain placental struc- tures. Such favored structures are the layers of cuboidal cells, lining cavities, forming papillary excrescences into these cavities and surrounded at the periphery by giant cells which latter pene- trate in both cases into the surrounding tissue, especially around the blood vessels, the wallsof which they may perforate, thus giving rise to hemorrhages. The identity of both formations, namely of the experimentally produced extrauterine pregnancy which we have just described and of the embryonal structures developing parthenogenetically in the ovary becomes quite evident, when one compares the microscopic sections of both of these formations. The microphotographs and the drawings also show the similarity. The similarity of the embryonal structures proper becomes clear through a comparison of Figs. 2, 3, and 6 in the former communication (Zeitschrift fiir Krebsforschung),) and of Figs. 2 and 3 in the present communication. The similarity of the placental structures is made evident through a comparison of drawings 1,2 and 4 inthe Archiv f. mikrosk. Anatomie,? of the Figs. 10, 12, 14 and 15 in the Zeitschrift fiir Krebsforschung with Figs. 4 and 5 of the present article. On several of these former figures there were also represented the relations of the wandering giant cells to the blood vessels and the hemorrhages resulting therefrom. Our new observations render 1t therefore certain that a fargoing parthenogenetic development of ova takes place in the ovaries of a 1 Loc. cit. 2 Loc. cit. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 73 relatively large number of guinea pigs, leading in the first place to the formation of placental structures, in some cases however also to the formation of embryos in the stage of the germ layers. We have discussed the possible causes for this parthenogenetic develop- ment on another occasion.1 We have perhaps to deal with a development which is caused by changes in the circulation and in the exchange of gases at the time and in consequence of the rupture of follicles. Such an explanation would be in accordance with the fact that the first segmentations of the ovum in the ovary of the guinea pig are found especially in atretic follicles, that the segmentations set in with beginning atresia and then gradually progress. Now we know that the atresia of follicles is more marked, than at any other time, at the time of ovulation.2 In this connection it is especially worthy of notice that the first segmentations of the ova in the ovary as well as the furthergoing parthenogenetic development, which leads to the formation of embryonal and placental structures, is preferably found in the ovaries of young animals. The latter, however, occurs occasionally also in some- what older guinea pigs. In such cases we may perhaps have to deal with structures which originally developed in younger an- imals, which then however had remained stationary for a longer period of time. We have still to discuss the significance of these structures for the interpretation of certain pathological formations, namely the embryomata and the chorion epitheliomata of the female germ gland. The large majority of pathologists assume in agreement with the suggestion of Bonnet and Marchand that these patho- logical structures take their origin from misplaced blastomeres and not from the parthenogenetically developing ovum. As I formerly emphasized? our observations make it very probable that such pathological formations originate from parthenogenetic- ally developing ova. They are therefore the ‘descendants”’ and not the “‘brothers”’ of the organism in which they originate. We may assume that in certain cases the parthenogenetic develop- 1 Leo Loeb, Proceedings Am. Philosophical Society, Vol. L., p. 228, 191. 2 Leo Loeb, Journal of Morphology. 3 Zeitschrift f. Krebsforschung, loc. cit. 74 LEO LOEB. ment of ova leading to these pathological structures begins only after birth. In a similar manner as we saw that embryonal placenta as well as the embryo proper can develop from the parthenogenetically segmenting ovum and that the embryonal placenta can be formed without the simultaneous development of the embryo proper, thus chorion epitheliomata may originate in the ovary without any accompanying embryonal structures proper. In other cases however there develop mainly the em- bryonal structures proper or certain of their parts. This conception of these structures explains the fact that they are mainly found in the germ glands. On the other hand, there exists no reason, why we should expect that aberrant blastomeres should mainly be found and develop at this place. Furthermore I have never been able in the many hundreds of ovaries of guinea pigs which I have examined microscopically to find a structure resembling a misplaced blastomere. ; We still have to explain why these structures are occasionally also found in the male germ glands and especially, why teratomata occur also, although less frequently, at other parts of the body, outside of the germ glands. As far as their occurrence in male germ glands is concerned, it might be explained by the fact that in a certain number of cases cells of both sexes may be found in the same individual, that therefore true hermaphroditism occurs. That this is not so rare an occurrence as has been assumed has recently been shown by L. Pick.1 We have perhaps also to consider the possibility that at a certain stage of development also the male germ glands are capable of developing in a similar manner as the ova. However at the present time there exist no facts supporting such an hypothesis. We know furthermore that in the course of embryonal de- velopment the germ cells migrate. It is therefore conceivable that occasionally one of their number may follow a wrong path and thus give origin to the formation of the teratomata outside of germ glands. While we are thus able to explain the origin of these structures on the basis of a parthenogenetic development of ova we do not intend to deny the possibility that under certain conditions irregularities in the embryonal development may lead 1 Cited from a review in the Miinch. med. Wochenschrift, 1913. EXPERIMENTALLY PRODUCED EXTRAUTERINE PREGNANCY. 75 to the transformation of blastomeres or of remnants of not fully differentiated embryonal tissues into teratomata, an hypothesis which would be in accordance with the finding of misplaced blastomeres by W. Roux in the course of the embryonal develop- ment of amphibian eggs. SUMMARY. 1. It is possible in the case of guinea pigs to produce experi- mentally the first stages of an extrauterine pregnancy. 2. In a similar manner, as in the case of guinea pigs experi- mental interferences of various kinds are not able to call forth the production of deciduomata in the connective tissue outside of the uterine mucosa after the discharge into the circulation of the sensitizing substance which is secreted by the corpus luteum, the developing ovum is unable to call forth a decidual reaction. 3. Under the conditions produced by us experimentally the development of the embryo is very much retarded and will in all probability come to a standstill after some time. Neither does the embryonal placenta develop in an entirely normal manner, although quantitatively the embryonal placental structures preponderate considerably over the embryonal proper. It is very probable that the lack of the decidual and of the typical blood vessel reaction on the part of the host connective tissue is the cause of this abnormal development. In man an extrauterine decidua can develop and accordingly here a fully developed extrauterine pregnancy is not rare. As we have shown pre- viously the effect of the extirpation of the corpora lutea on the formation of the decidua and on the development of pregnancy is approximately parallel. This is an additional fact which renders probable the significance of the decidual reaction for the complete development of the extrauterine pregnancy. The decidual reaction is at least one of the conditions which has to be considered in this connection. 4. Notwithstanding the presence of a young, developing em- bryo in the extrauterine connective tissue a degeneration of the corpora lutea and a new ovulation took place in the ovary. This proves that the persistence of the corpora lutea during pregnancy does not depend upon a substance secreted by the embryo; it is 76 LEO LOEB. probable that the growth of the decidua perhaps in combination with the growth of the embryo prolongs directly or indirectly the life of the corpora lutea during pregnancy. Thus far ex- periments, which we have carried out in order to decide this question, have shown that the presence of living and growing deciduomata prolongs the sexual period; furthermore that preg- nancy prolongs the life of the deciduomata. Further investiga- tion will decide whether or not these effects are exerted indirectly by means of the corpus luteum. 5. Our experiments render it certain that the structures which we found in a considerable number of guinea pigs and which we formerly interpreted as early stages of parthenogenetically developed pregnancies in the ovaries of guinea pigs really represent a relatively far going parthenogenetic development of ova which may lead to the formation of embryos in the germ layer stage which however usually leads merely to the formation of an em- bryonal placenta probably as a response of the developing ovum to the influence exerted by the contact with the surrounding host tissue. We show furthermore the significance these findings have for the interpretation of the teratomata and chorionepi- theliomata of the germinal glands. 6. The embryonal wander cells destroy outside as well as within the wall of the uterus bloodvessels of the surrounding host tissue in the ovary as well as in the peritoneal connective tissue and they thus cause hemorrhages in the surrounding host tissue. SOME CRYOSCOPIC AND OSMOTIC DATA! WALTER E. GARREY. Subsequent to the publication in 1905 of data on ‘‘The Osmotic Pressure of Sea Water and the Blood of Marine Animals, etc.,” (1) ‘the author has had occasion in the course of his other investiga- tions, to make numerous determinations of the freezing point of various sea waters, solutions and bloods; this method having been used to check up other methods of obtaining solutions of known osmotic pressures. Someof the data thus acquired have been correlated, and although somewhat fragmentary, they are published in hopes that they may facilitate the work of other biologists. The determinations have been made with the Beckmann apparatus and a differential thermometer, which could be read accurately to 0.005° C. When it is remembered that the de- pression of the freezing point (A) of a gram-molecular solution of a non-electrolyte is (theoretically at least) 1.85° C. below zero, that this depression corresponds with an osmotic pressure of 22.4 atmospheres (at 0° C.), and that the osmotic pressures vary di- rectly with the depression of the freezing point, it is seen that the osmotic pressure of any solution may be calculated from the simple formula: osmotic pressure = 22.4 a. A/I.85. SEA WATERS. Sea waters are not solutions of absolutely fixed chemical com- position, nor have they a constant concentration. While the ratios of certain salts are quite constant, there are other variations such as the content of absorbed oxygen and carbon dioxide and even of the fixed carbonates. J. Loeb (2) has called attention to the fact that the free alkalinity, 7. e., the number of HO ions, is distinctly higher in the sea water at Woods Hole than at Pacific Grove. 1 From the Physiological Laboratory of Washington University, St. Louis. 77 78 WALTER E. GARREY. The figures for the depression of the freezing point (A) given in Table I. indicate the wide range in concentrations in sea waters of different localities. AREER IE Sea Water from: A—°C. Observer. Reference. INaplessar een sascee =2.20 Bottazzi |Arch. ital. de biol., 1897, XXVIII., 61. Arcachon a eiaeion eo. —1.89 Rodier /|Trav. des Lab. d’Archachon, 1899. Pacific Grove, Cal.....| —1.925 | Greene /|Bull. U. S. Bureau Fisheries, 1904. XXIV., 429. Pacific Grove, Cal.....| —I.90 Garrey |BroLt. BULL., 1905, VIII., 257. Woods Hole: ......... —1.81 . Bio. BULL., 1905, VIII., 257. Beaufort, N. C........ —2.04 ae IQII. Helsolandmernee eerie —1.90 Dakin Bio-Chem. Jour., 1908, 269. In the Kattegat....... —1.66 a 5 ; Open BalticSea....... —1.30 af Kiel harbor.......... —1.093 kg In the following sections further details obtained by the author, by means of the cryoscopic method, are given for sea waters of some American localities. (a) Woods Hole.—Determinations made during the summer of 1904 have been previously reported (loc. cit., pp. 258-259) show- ing the freezing point to be slightly variable between — 1.805 and —1.84° C. The average of determinations made the latter part of July of six different years gave an average A = — 1.81° C,. with which, as will be seen from succeeding data, the following solutions are isosmotic: Sodium chloride, 0.52 m; Magnesium chloride, 0.29 m., cane sugar, 0.73 m. ‘‘Van’t Hoff’s solution,” made from m/2 stock solutions, had a freezing point of — 1.84° C., this is so slightly in excess of the concentration of Woods Hole sea water that it may be considered isosmotic with it. This “Van’t Hoff’s solution’? was made up from half molecular solu- tions according to the formula given by J. Loeb, (3) viz: 100 mole- cules NaCl, 2.2 molecules KCl, 1.5 molecules CaCl, 7.8 molecules MgCl, and 3.8 molecules MgSO,. The traces of bicarbonate and phosphate were omitted from the solution, but when added in optimum amounts (e. g., I c.c. N/20 NaHCOs3 per 100 c.c. solution, as in the procedure of Loeb, p. 35), the solution becomes exactly isosmotic with Woods Hole sea water. (b) Pacific Grove-—Green in 1904 made freezing point deter- CRYOSCOPIC AND OSMOTIC DATA. 79 minations of the Pacific Grove sea water and found that A= — 1.924° C. Garrey in 1905 made determinations, obtaining a slightly lower value for A viz., —1.905° C. On the basis of either of these figures, it is seen that the sea water in this locality is about 5 per cent. more concentrated than at Woods Hole and that a correction for this amount must be made if the osmotically equivalent solutions are to be calculated from the figures given in the previous section (a). | (c) Beaufort, N. C.—Working in the laboratories of the U. S. Bureau of Fisheries during the summer of 1911, the author made the following observations of the freezing point of sea water obtained at different localities in that vicinity and under different conditions as described in Table II. TABLE II. IQII. Beaufort, N. C. [X= 'AC, Remarks. June 8 |Open sea, outside “Sea Buoy’’ | —2.043 ee ~ IWharivof U.S:F.C.... 5.2.2. —1I.987 |I1:40 a.m.—tide low, N.E wind, previous showers. ; Whart-of ULS:F.C.. 2. 65... —2.015 |4:00 p.m. tide high July 19 |Wharf of U.S.F.C........... — 2.038 “Open sea, at ‘‘Sea Buoy’’.....| —2.03 0S BO Neri, 1S I9Coo 650 ¢oeb suo —2.07 |9:15 a.m., tide low He ES INAne rate (OAS) ) aha Cag. eeeaa ions era eres —2.06 |3:00 p.m., tide high Weare SO SATEY SOUT Gey acy eet eeusne ia eke ool —2.073 HO ee Soe SIN lovee t AGES she Craniiy Geena retrey ere ara —2.079 |Tide low, strong south wind Soe, © WS IBM ob oo oooocdodocuDE —2.05 - S INWinevar WESHIKCss 6 Gc obeacuCs —2.052 |Tide low ““"\Newport River, at ‘‘Cross FROCK SA erates here oyeteet: —1.707 From these figures it is seen that the open sea water off Beau- fort hasa A = — 2.04° C., and is 12 per cent. more concentrated than at Woods Hole. (d) Diluted Sea. Water.—The constant necessity for the use, in biological investigations on marine forms, of diluted sea water and corresponding concentrations of pure salts, has led the author to make the determinations found in Table III. Various dilutions of Woods Hole sea water were made and the freezing points determined. In most cases the densities at these dilutions have also been determined by the pycnometer method. The figures given in the table are all from actual determinations made 80 WALTER E. GARREY. by the author; when not given they may be approximated by interpolation.! TABLE III. Dilution : Densities of Sea ; Woods Hole Distilled NOG: Water Dilutions NaCl eae Sea Water \ + { Water at 21.59 C. (Ref. ae Seiko sown. c, cm. c. cm, H,O at 21.5° C.) Undiluted fo) —1.8I 1.02426 3.04 Sy Ces nS Gx —1.54 2.6 Tae Ae Fae —1.35 2.275 662.55 285 > —1.20 2.00-+ 60 “ Owned —1.09 I.81 So) Olan —0.915 I.0123 1.58 AS is ie es —0.82 I.4 Ae) Oo) ™ —0.73 I.0096 I.21 Ri 0 @g 0% —0.64 I.07 33% “ 662% “ —0.61 1.008 T.02 SIA wiles GS an —0.505 I.00 3 OMe x HO. 3 —0.547 1.0073 0.91 Paes Gemma ts —0.460 T.0062 0.76 gx0) 90 80 “* —0.37 1.0046 0.60 TOES: 90 “* —0.187 1.0023 0.30 CANE SUGAR. Attention should be directed to a fact to which Jones (4), Morse and Fraser and Berkeley and Hartley have called atten- tion, viz., that cane sugar solutions show osmotic pressures considerably in excess of what theory would lead one to expect. Loeb has shown the importance of this fact for biological work (5). From purely theoretical considerations one would expect a molar (gram-molecular) solution to show an osmotic pressure only slightly in excess of that of Woods Hole sea water. Loeb found that it caused a shrinkage of the eggs of the echinoderms even of the Pacific, and his experiments caused him to select 6/8 m. cane sugar as the proper concentration for the development of Stron- gylocentrotus purpuratus. The osmotic pressure of Woods Hole sea water by calculation from the freezing point is 21.9 a. (at o° C.), a figure which is almost identical with that obtained by 1 The determinations of Gerlach for NaCl and KCl (Chemiker-Kalender, 1914, I., p. 261) and of Schiff for MgCl and CaCl (zbid., p. 265) show, that, for concentra- tions of solutions of the magnitudes with which we are dealing and in which these salts are present in sea water, the densities are a linear function of the concentration. A plat of our determinations shows the same to be true for both densities and freezing points of dilutions of sea water. CRYOSCOPIC AND OSMOTIC DATA. 81 calculation for 0.75 gram molecular solutions of cane sugar, using the measurements of Berkeley and Hartley. Some of our de- terminations of the freezing point of solutions of cane sugar illustrate their peculiar osmotic behavior.? For a gram molecular solution of cane sugar (342.2 grams per liter of solution) we found A = — 2.775° C.; for 34 mol. (256.6 grams per liter) A = — 1.855°, — 1.86° C.; and for % mol. (171.1 grams per liter) A = — 1.15°, — 1.155° C. For these three solutions the theoretical depression of the freezing point would be to — 1.85°, — 1.387° and — 0.925° C., respectively. Comparison of these figures shows how much in excess of the theoretical osmotic pressure, that of these solutions really is. Morse and Fraser have pointed out that the correspondence with the theoretical expectations is greater, if ‘‘weight normal”’ solutions are used, 7. e., if the substance is present in a liter of the solvent, instead of this volume of the solution. This does not account, however, for the full amount of the discrepancy found. To illustrate this: It was found in our experiments that in making a gram-molecular solution by dissolving 17.11 grams of cane sugar in 50 c.c. of the solution (15° C.), it was necessary to add only 39.4 c.c. of distilled water; A was — 2.775° C. Had 50 c.c. of solvent been used to make the corresponding “weight normal”’ solution, A would have been — 2.187° C.3 This figure exceeds the theoretical A (— 1.85° C.) by 0.337° C., which is probably to be accounted for by hydration of the sucrose molecule (Callendar (6)). In the figures given above it is to be noted that the A of .75 mol. solution of sucrose (— 1.855° C.) is that which theory expects of agram molecular solution, and its osmotic pressure lies between that of the sea water at Woods Hole and Pacific Grove. By 1 The measurements of Berkeley and Hartley were made using other concen- trations. The original figures of these workers as also those of Morse and Fraser et al. are given in the “‘ Physikalisch-chemische Tabellen,’’ Landolt, Bornstein and Roth, 4th ed., Table 179, p. 787. Their original papers are referred to, zbid., p. 790. 2 The sugar used in these experiments was free of all reducing sugars and had been twice recrystalized from glass-distilled water with subsequent drying in vacuo. 3 Calculation of the freezing point of a molecular ‘‘weight normal”’ solution, based upon Morse’s figure for the observed osmotic pressure (24.8 a. —o° C.) gave a slightly lower figure, viz.: A = — 2.048° C. 82 WALTER E. GARREY. extrapolation we obtain the following figures for the concentra- isosmotic with sea water of: tion of sucrose; I. Woods Hole 2. Pacific Grove 3. Beaufort 0.73 m. (A = — 1.81° C.) = 0.765 m. (A = — 1.90° C.) = 0.81 m. (+) (A = — 2.01° C.) SALT SOLUTIONS. (a) Sodium Chloride—In addition to the freezing points of solutions given in Table III., the following have been deter- TABLE IV. Nee een Made by /\ = OC, Remarks. 0.65 L. —2.255 0.65 Gupracic: —2.35 0.65 (EG, 229 Co —2.50 Madein a flask standardized to I5° C. 0.60 pls —2.11 0.58 G. —2.03 Isosmotic with Beaufort Sea water 0.54 G. —1.90 Isosmotic with Pacific Grove Sea water 0.54 L. —1.895 0.52 G. —1.81 Isosmotic with Woods Hole sea water. 0.50 L. —1.735 0.50 L. —1.74— 0.50 G. —1.745 0.50 F. =1.75 0.50 L. —1.765 0.50 G. —1.745 0.444 G. —1.54 mined by the author on solutions made up by different com- The concentrations chosen were somewhat petent workers. TABLE V. MgCl, Concentra- tion.1 0.50 molecular 0.36 of 0.35 oy 0.31 0.30 0.29 0.10 [NC (C- —2.845 —2.03 —1.985 —1.895 —1.85 —1.815 —0.495 Remarks. Isosmotic with sea water at Beaufort. Isosmotic with sea water at Pacific Grove (or 0.32 m. according to Greene’s determination). solution). Isosmotic with sea water at Woods Hole. Dissolved in 100 c.c. of distilled water (not of 1 Concentration referred to volume of solution, not of solvent. CRYOSCOPIC AND OSMOTIC DATA. 83 to either side of those isosmotic with the sea waters of our coast laboratories. (b) Magnesium Chloride.—In solutions of this salt some of the molecules are dissociated into three ions, which accounts for the fact that the osmotic pressure is greater and consequently the depression of the freezing point is lower, than that of equi- molecular solutions of sodium chloride. The following commonly employed solutions have been tested (cf. Table V.). ANIMALS. In addition to data previously published by the author (loc. cit., p. 263), several determinations have been made on the blood of animals of the waters of the American coast,'and inland rivers. (a) Limulus polyphemus.—In the work referred to it was shown that under experimental conditions the blood of this animal, like that of other marine invertebrates varies to conform in concentration to that of the external medium. We have since found that this is true in the natural habitat of these ani- mals, thus at Woods Hole (1904) the blood of Limuli depressed the freezing point like sea water, to — 1.82° C. At Beaufort, N. C., July 20, 1911, the water of the Fisheries “pound” froze at — 2.03° C. The blood of four Limuli taken from this water, in which they had been kept for several weeks, showed the fol- lowing freezing points, respectively, — 2.025°, — 2.03°, — 2.04° and — 2.35° C. In the case of another Limulus captured at “Cross Rocks” in the Newport River, near Beaufort, September 8, I91I, the blood A = — 1.71° C. while the water at that place depressed the freezing point to — 1.707° C. Such readings taken from animals under natural conditions established the absolute identity of osmotic pressure of the external and internal media despite the differences in their composition. (b) Elasmobranchs.—A shark seven feet in length (not identi- fied) was captured in the Fish Commission nets at Beaufort, July 27, 1911; the sea water froze at — 2.02° C.; cryoscopic readings of the blood from the heart and portal vein, withdrawn immediately after death, were identical, within 0.01° C., and showed A to be — 2.182° C., which again is practically identical with that of the Beaufort sea water taken in the neighborhood of the fish trap on that date. 84 : WALTER E. GARREY. The A for this elasmobranch is larger than for those tested at Woods Hole by the author and later by Scott (7). The figures indicate an adjustment to the greater concentration of the sea water at Beaufort, a fact which is also borne out by determina- tions made on the blood of ‘sting rays,’”’ the blood of four of which at Beaufort gave A = — 1.98°, — 2.04°, — 2.03°, — 2.07° C., respectively. These depressions are not greater than that of the sea water from which the animals were taken although both the author and Scott found a slightly greater depression for the blood of the dog fish (Mustelus canis) of Woods Hole than for the sea water of the laboratories; this water is, however, somewhat less concentrated than the water outside the heads. (c) Marine turtles—The defibrinated blood of three species of marine turtles! caught at Beaufort in 1911 was frozen and the A thus determined for each individual is as follows: Chelonia mydas A = —0.675° C. Cclpochelys kempi A = — 0.687°, — 0.70°, — 0.70° C. Caretta caretta A = — 0.69°, — 0.69°, — 0.685° C. In the cases of two carettas obtained at Woods Hole in 1913, the A found was identical with that given above. These depressions (A) are, in all cases, greater than those obtained by Bottazzi for ‘‘Thalassochelys caretta’” (A = — 0.61°) although it is worthy of mention that the waters from which our animals were taken were, if one can judge from Bottazzi’s writings, less con- centrated than that from which his specimens were obtained; if any adjustment to aqueous media were to take place it would be in the direction opposite to that indicated by the above figure. It is certainly true, however, that the blood of fresh water and land turtles shows a depression of the freezing point which is distinctly less than that of marine turtles; Bottazzi found A for Emys europa = — 0.463° to — 0.485° C., while for Pseudemys elegans of the Mississippi Valley we obtained a depression in which A = — 0.48° C. These figures are so much below those obtained with the blood of marine turtles that they would seem to indicate the possibility of some degree of adjustment to the concentration of the external medium; on the other hand it is a 1 The author is indebted to Mr. Hay for the identification of these animals. CRYOSCOPIC AND OSMOTIC DATA. 85 fact that the land turtles do not show a more concentrated blood than do those which live mainly in fresh water. This point was put to the experimental test upon the marine turtles, Colpochelys kempi and Caretta caretta; the last figure in the previous data given above for each of these species was ob- tained, with the blood of a specimen which had been kept for two months in a tank containing fresh water. There was ab- solutely no change in the concentration of the blood of these individuals, and we feel justified in concluding that adjustments of the nature of those under consideration do not take place in these forms at least not within the duration of our experiments. (d) Fresh Water Fish.—Preliminary to a study of the effects of osmotic and saline media upon fresh water fish (soon forth- coming) it was desirable to know the osmotic pressure of the blood of forms taken from the Mississippi river. The following list contains some forms peculiar to this region on which no data have hitherto been given. 1. Polyodon spathula = — 0.492°, — 0.486°, — 0.50° C. 2. Scaphirhynchus platyrhynchus = — 0.505°, — 0.507°, — 0.503° C. 3. Lepidosteus osseous (L.). (‘Gar’) = — 0.487° to — 0.52° C. 4. Amzia calva (L.) (land locked) = — 0.508° C. 5. Catostomus teres = — 0.51° to — 0.52° C. 6. Perca fluviatilis = — 0.498° to — 0.51° C. Fresh water ganoids are seen to have blood which is identical in concentration with that of fresh water teleosts. All have blood less concentrated than that of any of the marine fishes and it is conceivable that in the case of these animals some adjustment to environment has taken place; such adjustments are known for marine fish as has been shown by the author (1) and others (Fréd- ericq, Bottazzi, Dekhuysen (8), Dakin, Joc. cit.). REFERENCES. I. Garrey, Walter E. 705 Brov. BULL., VIII., p. 257. 2. Loeb, J. ’r3 ‘‘Artificial Parthenogenesis’’ Chicago, p. 34 3. Loeb, J. P. 35, loc. cit. 4. Jones, H. C. ’93 ~Zeitschr. f. Physikalische Chemie.—XII, pp. 110 and 529; XII., p. 623. 86 WALTER E. GARREY. | 5. Loeb, J. P. 130, loc. cit. 6. Callendar 208 Proc. Roy. Soc., A, 1908, 80, 466. 7. Scott, G. G. 13 Annals N. Y. Acad. Sciences, 1913, XXIII., p. 1 et seq. 8. Dekhuyzen, M. C. Arch. Neerland, Sc. Exact. et Nat., Ser. 2, 10, 121. Note.—For other literature consult the papers by Garrey! and by Scott;’ also Bottazzi, Ergebnisse der Physiologie, 1908, VII., p. 162; cf. also Table I. of this paper. CONCERNING BRACHET’S IDEAS OF THE ROLE OF MEMBRANE FORMATION IN FERTILIZATION.! JACQUES LOEB. 1. A recent publication by Brachet? seems to make it necessary to discuss once more the relation between membrane formation and development. The writer had shown in 1895 that if oxygen is completely withdrawn from the fertilized sea urchin egg no development is possible, while the moment oxygen is admitted the development can begin again. As he had suggested in 1906 and _as has since been proved by O. Warburg, and H. Wasteneys and the writer, the entrance of the spermatozo6n into the egg of the sea urchin increases the rate of oxidations in the latter (by 400 or 600 per cent). The entrance of the spermatozoén causes also a membrane formation which is very marked in the fresh egg and is generally less marked or may appear to be absent if the egg has been lying in sea water for a day or more. It has been shown, moreover, that the artificial production of a membrane in the unfertilized egg by butyric acid has the same influence upon the increase of the rate of oxidations as the entrance of the spermatozo6n. These and other facts seemed to support the view of the writer that an alteration of the surface of the egg, which usually but not necessarily results in a membrane forma- tion, is an essential feature of the development of the egg. More recent experiments by Warburg? have made it very probable that the process of oxidations in the sea urchin egg is a case of catalysis by iron, which is confined mainly if not ex- -clusively to the surface; and this fact, in connection with the data mentioned above, seems to indicate that the process which under- lies membrane formation in the unfertilized egg may consist in bringing about or rendering possible the iron catalysis which is 1 From the Rockefeller Institute for Medical Research, New York. 2 Compt. rend. l’ Acad. d. sc., CLIX., 642, 1914. 3 Warburg, Zeztschr. f. physiol. Chem., XCII., 231, 1914. 87 88 JACQUES LOEB. responsible for the sudden increase in the rate of oxidations after artificial or natural membrane formation. Since fertilization by sperm is accompanied by a membrane formation and followed by the same increase in the rate of oxidations as is artificial membrane formation by butyric acid, it is probable that the alteration of the surface (underlying membrane formation) is also the cause for the increase in the rate of oxidations in the process of natural fertilization. 2. The writer has time and again stated that the formation or non-formation of the fertilization membrane is of only secondary importance; what matters are the physicochemical changes which underlie the membrane formation and which are responsible for the sudden rise in the rate of oxidations of the sea urchin egg after artificial or natural membrane formation; and which may even occur when for some reason the fertilization membrane is modified or when its formation is entirely suppressed. There is no doubt that in the writer’s first experiments with the purely osmotic method, the fertilization membrane was often very indistinct or in some cases even completely lacking, while nevertheless the enormous increase in the rate of oxidations and development to the pluteus stage ensued.1 . It is possible to modify the surface of the unfertilized egg in such a way that if it is later fertilized by sperm the abnormal character of the membrane formed, or the abnormal conditions of the surface, may lead to the death of the egg. The writer described such a case in 1909.2, When the unfertilized eggs of Strongylocentrotus were treated for five minutes with a hyper- alkaline solution of NaCl (50 c.c. m/2 NaCl + 1.0 c.c. N/1o NaOH) and then transferred to normal sea water to which sperm was added, the eggs were all fertilized but apparently without membrane formation, though in reality probably with a tightly fitting membrane. They all segmented but perished in the blastula or gastrula stage. When, however, the eggs were not fertilized immediately after the treatment with alkali but after 1 The literature of the subject can be found in the writer’s recent book on “Artificial Parthenogenesis and Fertilization,’’ Chicago, 1913. ? Loeb, “‘ Die chemische Entwicklungserregung des tierischen Hies,’’ Berlin, 1909, jh Tey MEMBRANE FORMATION IN FERTILIZATION. 89 they had been in the sea water for one hour or more, a more normal membrane was formed and the eggs developed into plutei. Why did the eggs only live to the blastula or gastrula stage when they were fertilized immediately after the alkali treatment? Should this have had something to do with the abnormal char- acter of the membrane which was formed when the egg was fer- tilized immediately after the alkali treatment? Were the cells pressed by the membrane which was too tight, and did this pressure kill them if prolonged? If this were the case, a tearing of the membrane should save the life of the egg. It would be of interest to try this experiment. 3. Ina recent number of the Comptes rendus de Il’ Académie des Sciences, Brachet has published an observation which may or may not be similar to the one just mentioned. He found a year ago that if the eggs of the sea urchin, Paracentrotus lividus, at Roscoff, are put for two hours in contact with sperm of Sabdellaria alveolata (which cannot fertilize the eggs) and if they are after- wards fertilized with sperm of their own species, they develop without apparently forming a fertilization membrane. From this Brachet concludes that the formation of a fertilization membrane is not necessary for development, a conclusion which will surprise nobody who is familiar with my first experiments on artificial parthenogenesis, or who has ever fertilized eggs which have been lying in sea water for several days. Moreover, Brachet observed that the sea urchin eggs which are fertilized with sperm of their own species, after two hours’ treatment with the sperm of Sabellaria, die at the time of gastrulation. The prolonged treat- ment of the eggs of Paracentrotus with the sperm of Sabellaria seems therefore to have a similar effect as the short treatment of the egg of Strongylocentrotus with the alkaline NaCl solution in my experiments. 4. The deductions which Brachet draws seem, however, difficult to reconcile with each other. We stated already that he assumes that the eggs of Paracentrotus after two hours’ treatment with the sperm of Sabellaria form no fertilization membrane after fertilization with their own sperm. Yet, he states further that these eggs die in the gastrula stage for the reason that they cannot hatch; for if he shakes the eggs and thereby destroys ‘‘la couche 90 JACQUES LOEB. corticale’”’ the larve can hatch’and are now able to develop into plutei. The only membrane, however, which can prevent the eggs from hatching is the fertilization membrane, and it is im- possible to harmonize the two statements of Brachet, first, that these eggs have no fertilization membrane and, second, that the gastrulae cannot hatch unless the membrane of the egg is pierced. Professor Goldschmidt, to whom I showed Brachet’s paper sug- gested that Brachet probably means by “‘couche corticale”’ the hyaline membrane (Herbst’s “‘Verbindungsmembran’’) which surrounds the blastomeres and that he assumes erroneously that this hyaline membrane forms a continuous layer around the blastula in the same way as the fertilization membrane does. This is, however, not the case since the hyaline membrane par- ticipates in the process of segmentation and forms a distinct layer around each individual blastomere, but not a continuous envelope around the whole blastula. Brachet’s observation is intelligible on the assumption that the egg after it has been treated with the sperm of Sabellaria forms a very tightly fitting membrane when it is fertilized with its own sperm and that this membrane must be torn by shaking the egg in order to allow the blastula to hatch (or to escape from being killed by the mechanical pressure of the tightly fitting membrane?). Brachet found also that it is possible to sub- stitute for the shaking of the egg a treatment with butyric acid, which as he assumes also tends to remove the obstacle to the hatching. This may be correct, but unfortunately he draws the further conclusion that the butyric acid treatment must have the same effect upon the unfertilized egg as upon the fertilized egg which has previously been treated with the sperm of Sabel- laria. Leaving aside the fact that the unfertilized egg has no membrane, it has been shown that the butyric acid treatment raises the rate of oxidations of the unfertilized egg about 400 or 600 per cent., while acid does not increase, but, on the contrary, - lowers the rate of oxidations in the fertilized egg. Moreover, the writer has shown that if a fertilized egg is treated with butyric acid, in the same way as is required for inducing artificial parthe- nogenesis, the fertilized egg is not injured, while the inducing of a membrane formation by butyric acid in the unfertilized egg leads MEMBRANE FORMATION IN FERTILIZATION. OI to the rapid death of the latter, if it is kept at room temperature and if it does not receive a second treatment either with a hyper- tonic solution or lack of oxygen. This case was fully discussed by the writer in a recent paper.! It is therefore not justifiable to conclude that the action of butyric acid on the unfertilized egg must be identical with the action of the same substance on a fertilized egg, treated beforehand with the sperm of Sabellaria. Should it be possible that Brachet’s ‘‘couche corticale”’ is the chorion or the “jelly”? which surrounds the unfertilized egg? But this jelly is normally dissolved when the egg is fertilized. It might be conceivable that the sperm of Sabellaria causes a harden- ing and a contraction of this jelly which protects it against being dissolved by the sperm of the sea urchin and that subsequent shaking or a subsequent treatment with acid destroys this jelly. But granted this were the case, it would be erroneous to use experiments on an artificially altered chorion to draw conclusions upon the réle of membrane formation in fertilization or artificial parthenogenesis. The writer wonders how Brachet (or Herlant) are going to harmonize the following well-established facts with their views. If the eggs of Strongylocentrotus purpuratus are treated with hypertonic sea water for about two hours, they form in most cases no membrane and nothing happens to them except that a certain percentage of them begin to divide very regularly into 2, 4, 8, possibly 12 or 16 cells and then stop. Such eggs are to all appearances in the resting stage and live as long as the other unfertilized eggs if nothing is done to them. If they are fertil- ized by sperm each blastomere forms a special fertilization mem- brane and now each blastomere develops into a blastula or into a pluteus, according to the size of the blastomere. They also develop into plutei if an artificial membrane formation is called forth with the aid of butyric acid. The writer is inclined to explain this phenomenon by assuming that the treatment with the hypertonic solution called forth two effects, one of which was a peripheral change resulting in an increase in the rate of oxidations. This effect is, as the writer has shown, reversible 1 Loeb, “Weitere Beitrage zur Theorie der kiinstlichen Parthenogenese,’’ Arch. f. Entwckingsmech., XXXVIII., 409, 1914. 92 JACQUES LOEB. and was possibly reversed while the eggs were in an early stage of development. It seems to the writer impossible to reconcile these observations with the purely morphological views of Brachet or Herlant. | Brachet (like Herlant) tries to explain the phenomena of artificial parthenogenesis and fertilization without any consider- ation of the striking chemical processes that accompany fertiliz- ation and artificial membrane formation. He reverts to that standpoint of the pure morphologist which Sachs, in his papers on ‘“‘Matter and Form in Plants”’ characterized as ‘ ism.” This standpoint disregards the sources of energy in life phenomena and treats morphological changes as if they required no source of energy. It seems to the writer that the fact of the necessity of oxygen for development, the fact that mere mem- brane formation (both by butyric acid or by a spermatozoén) raises the rate of oxidations 400 or 600 per cent, and the fact that the amount of rise is identical in both cases, are so striking, that ‘empty formal- these facts cannot be ignored in a theory of the rdle of membrane formation in the development of the sea urchin egg. The writer has always considered the changes underlying the membrane formation as the essential factor in the initiation of development, while he considered the formation of a fertilization membrane only as a welcome but not essential indicator of the chemical changes in the surface of the egg; afact which Brachet, on account of his disregard for the chemical processes, has entirely over- looked. Brachet, from his purely morphological standpoint, erroneously assumes or makes it appear as if I considered the formation of a visible membrane as the only and essential act in the initiation of development. AN EXPERIMENTAL ANALYSIS OF FERTILIZATION IN PLATYNEREIS MEGALOPS. BE. BE. JUS#. Study of the breeding habits of Platynereis megalops revealed the fact, as has been pointed out (Just, 14), that insemination takes place in the body cavity of the female and that although egg laying begins often but five seconds after copulation, the eggs will not fertilize when artificially inseminated after exposure to the action of sea-water. It is this failure of sea-water insemina- tion that forms the basis of the present contribution to the analysis of fertilization in Platynereis. In order clearly to inter- pret the phenomena of sea-water insemination a study of the morphology of the normal fertilization was made (see Just, ’15a). The experiments undertaken for the analysis of fertilization in Platynerets come under three heads: A. Conditions of successful insemination. B. Cross fertilization with Nereis. C. Artificial parthenogenesis with various agents. B and C are taken up mainly because they supplement results under A. A. CONDITIONS OF SUCCESSFUL INSEMINATION. During the summer of 1911, I was studying the maturation and fertilization of the Platynereis egg for comparison with those processes in Nereis. The methods of insemination used with Nereis, cutting out the eggs and sperm in sea-water, gave no cleavage. Various trials with the utmost care, using diverse methods never gave cleavage. Not until August 24, 1911, did I chance to find that normally insemination takes place in the body cavity of the female (cf. Just, ’14). I. Observations on Eggs Inseminated in Sea-water. If eggs and sperm be cut out of Platynereis and mixed in sea- water, the phenomena of maturation, sperm attachment, and 93 04 E. E. JUST. copulation of the germ nuclei may be readily followed; but such eggs do not segment nor do they ever develop into swimming forms. The Living Egg. If insemination be made in a suspension of India ink ground up in sea-water, the jelly formation may be easily followed: it differs but little from the cortical outflow observed in eggs nor- mally laid. All eggs, however, do not secrete this jelly; of these, some remain in the germinal vesicle stage and others go through maturation with all or part of the cortex intact. As in the normally inseminated egg (see Just, 15a) no cone is present. More often than in the normally laid egg a broad plateau of cytoplasm marks the point of sperm attachment. The sperm, from one to six, are attached to the membrane above this raised cytoplasm or near it. Maturation proceeds about as in the normal egg. At matura- tion stages slightly later than in the normal egg, the sperm may be found in the egg. It moves forward with aster formation. The pronuclei meet, remain apposed for a short time, separate, and fade from view. This is not true of all eggs; for apparently, those in the germinal vesicle stage or in maturation stages with cortex intact never engulf the sperm. Moreover, in many eggs that are in maturation with the cortical layer gone, one cannot find sperm. These eggs never divide. At first, 1911, I thought that this behavior of the egg was due to injury of the worms. Its sig- nificance became clear only after the discovery of the normal method of egg-laying. The Sectioned Egg. During four seasons eggs have been preserved at three and five minute intervals upward to two hours after insemination in sea-water. Study of the sectioned eggs confirms the findings of the study of living eggs. Many eggs remain ovocytes with sperm attached or not. Those that go through maturation do so with or without jelly formation. Eggs that form jelly are likewise of two classes: those in which sperm are found to have penetrated and those in which no sperm are found. FERTILIZATION IN PLATYNEREIS MEGALOPS. 95 I have not been able so far to determine any structural differ- ences in the ovocytes with and without sperm attached. In the case of the eggs that maturate with the cortex wholly or partially intact, the spindle may be abnormal. In most cases if it reach the periphery of the egg it does so at a point practically devoid of cortical cytoplasm. Or again, it may lie parallel to a tangent of the egg membrane. Those sections which reveal the sperm within the egg are in the minority. It appears from experiments several times repeated during the four seasons of study that the penetration of the sperm depends upon the amount of sea-water used. If the eggs be inseminated in a large quantity of sea-water or washed (by changing the water several times) very few eggs form jelly. With less water more form jelly. Eggs inseminated quickly in small quantities of sea-water are capable of engulfing sperm. The history of the penetration as known may be briefly given. One finds sperm external to the egg at different stages. How it gets into the egg I cannot yet state with certainty although this point has received most careful study for three years. Material has been prepared in every way possible to demonstrate the early penetration. So far I have not found the sperm entering the egg as a slender thread like that in the normal egg. It can be easily demonstrated in the endoplasm. On one slide of the 1911 series, for instance, I counted twenty sperm heads with their asters lying near the centre of the egg. The sperm head remains for a longer time than in the normal egg a black knot with a long drawn out thread extending to the single aster. A second aster has never been found. The germ nuclei copulate but the eggs never cleave. Various stages are found from sixty to one hundred twenty minutes after insemination—sixty minutes after cleavage in the normal egg. The pronuclei after apposition gradually separate and degenerate as discrete nuclear masses. Many eggs show only one chromatin mass in process of degeneration; doubtless, these are eggs which sperm do not enter. The sections of such eggs closely resemble those of Nerezs eggs from which the sperm have been removed (see Lillie, ’12). I have repeatedly made observations on living eggs inseminated in sea-water and on sections. I have yet to find a single cleaving egg. 96 [Dra 195, {USite Two hours after insemination the eggs exhibit cytoplasmic stratification; the oil drops later fuse to form one at the vegetative pole. Twelve hours after insemination the conditions are the same; there is never a swimming form among these eggs. 2. Nature of the Inhibition to Development. It may be very clearly shown that sea-water is responsible for the lack of cleavage by the method of “dry insemination.” If males and females dried on filter paper be cut up separately and the drops of eggs and sperm thus obtained be mixed with sub- sequent addition of sea-water, a percentage of the eggs always cleave and develop into normal trochophores. I have kept larvae from such dry inseminations until they were seven mm. long with thirty or more segments, few differing from normally laid eggs. There is doubtless an optimum time after mixing for the addition of sea-water, but any time upward to two minutes gives results. The following is an example: August 3, 1912. To determine the time interval after mixing dry eggs and sperm before adding sea-water. Per Cent. Water Added. of Cleavage. Res AENOMCE Sie ts SPiyh ese ey nea ela tector Peak ee one uen unica cae Renee 60 2. Nive Seconds ahterncic cos -pone eh eiuera are toeeces ci onben ee Rrereene 50 3. ensecondsrahtereame. telecine 90 4: Twenty: secondsiaiters sac yee e nee ee eee ei 45 Practically, as soon as eggs and sperm are mixed, sea-water may be added. I have not been able to add sea-water quickly enough after mixing to prohibit cleavage. If the eggs are allowed to stand two minutes the majority are plasmolyzed by the addi- tion of sea-water. | The amount of sea-water that will permit fertilization has been repeatedly determined: July 28, 1912, 9:45 P.M. Experiment to determine the maxi- mum amount of sea-water that permits fertilization. Males and females are thoroughly dried on clean filter paper. A male and a female placed in each of the eight perfectly dried clean watch glasses. Sea-water added as follows: FERTILIZATION IN PLATYNEREIS MEGALOPS. 97 NG ues Bikte How o onde la Ope IgewLe gle OmiaIod oS DOE LD I drop OG py te Side bralete 6/60 Bide 0 Slocproloiphowolp ae 0 CHES cen 2 drops OG i Reh eB ERAN O TS AEA EGS He Rime om Celen S ° Aine aha ete BCS Sl Sad eR er ee Goo RHO © > Tete easy ART: CO ee ea emi oer bo Gia cee Pho Ral el a Ga fat eee neh co BCU RC RRR hie BO yale AME VAP) Mi coat en cr alot, Qonra'l iva ace parka, Sirk oh. TN ay (Si. 4h OL is a No Ana eon tt coho tN: 0 Pig Din Cate eae ad IO c.c. Gs Ta Le oie Baril TL all een ibe, Segemioha n:6 b-GiiG.0\c- Ob CAC SiR no sea-water. The worms were then cut up and flooded with sea-water, later transferred to fresh sea-water in finger bowls. Nos. 1, 2, 3 and 8 gave cleavage; a per cent. of normal trocho- phores was found the next morning. In dishes 4, 5, 6 and 7 not an egg divided, no swimming forms developed. No single observation in the whole work was made as often as this; the results are wonderfully precise. As I shall show later the experiment quoted was conducted under the optimum con- ditions, and yet it shows the inhibiting effect of such a surprisingly small quantity of sea-water. All other observations show two drops of sea-water for each worm to be the maximum that will permit normal fertilization. Inno case have I got cleavage where two and one-half drops of sea-water for each worm (2. eé., five drops to two worms) were used. While the same pipette was used to secure equal drops, the worms, females particularly, vary in size. I have usually taken the average females for these experi- ments. Such an animal, as found by actual count in three cases, has about 11,000 eggs. There is enough variation, however, in the size and weight of the worms to make impossible any law concerning the lethal amount of sea-water. I believe, never- theless, that there is an optimum time for the addition of sea- water—equal to the time the sperm are in the female in normal insemination; and an optimum amount of sea-water—about as much as the worms will take up after thorough drying. The results of these inseminations over a period of four seasons prove clearly that sea-water except in minute quantity is fatal to fertilization. Does Sea-water Injure Egg, Sperm, or Both? Three explanations of the failure of Platynereis eggs to cleave after insemination in sea-water are possible: 98 By 1d, just (a2) Both eggs and spermatozoa are injured by the sea-water. (b) The sperm alone are injured by the sea-water. (c) The eggs alone are injured by the sea-water. The failure of the eggs to go beyond maturation may be due to the injurious action of the sea-water on both eggs and sperm alike. It would seem reasonable to assume that for internal insemination both cells need the perivisceral fluids. It might be difficult to conceive how this adaptation in Platynereis could have taken place acting on one only of the sex elements. As both eggs and spermatozoa are protected by body fluids in normal insem- ination, so both are exposed to the lethal action of sea-water. Embryologists are all careful when inseminating eggs of forms in which insemination normally taken place in the sea not to con- taminate the dishes containing ova with the animal’s tissues or fluids. Lillie (130, ’14) has shown why this is essential. I have, however, repeatedly with success fertilized Nereis eggs dry (see Just, 150) doubtless because the body fluid of Nereis is practically negligible. And the case of Platynereis is similar to that of Nereis; in this smaller worm there is no more fluid; the female is a mere locomotor ovary, although the male does have a small amount of fluid and a great number of corpuscles. The second possibility is that the sperm alone are injured by the sea-water. Injury to the sperm through transference from the male’s body fluid to sea-water, however, cannot be due to difference in osmotic pressure. For as Frédericq has shown, and Garrey since for the Woods Hole region, the osmotic pressure of invertebrate body fluids is about the same as that of sea-water. Moreover, Platynereis sperm in sea-water as far as I could de- termine exhibit none of the effects experimentally produced by Koltzoff on various sperm cells including those of Nereis (dumer- iii?) through treatment by various salt solutions or those con- ditions described by de Meyer with hypotonic and hypertonic solutions. Insome other way, then, the sperm must be assumed to be weakened but still capable of partially fertilizing the egg as the Hertwigs, Gemmil, Budington, Dungay, etc.,haveshown. And indeed my Platynereis slides of sea-water inseminated eggs show similarities to the figures by Lillie of the penetration of injured sperm in Nereis; in Platynereis, however, the germ nuclei develop FERTILIZATION IN PLATYNEREIS MEGALOPS. 99 a little farther. Steinach long ago, later Walker (’99, 11) and Hirowaki have shown that in mammals the prostate secretion is necessary for fertilization. Sea-water, then, might injure the " sperm and hinder fertilization by destroying a supporting medium necessary for fertilization. (On this point, cf. Gemmil’s ex- periments.) Finally, a third explanation is possible: the egg alone is injured through sea-water treatment. The egg, in this case, may be dependent on a substance in the female’s body or on some se- cretion of its own necessary for fertilization. Both egg and sperm may need body fluids but sperm may be hardier, egg less resistant. The seasons of 1912 and 1913 were largely given over to ex- periments to determine which possible explanation is valid for Platynereis. In 1914, many of these experiments were repeated. And I may say at once that the explanation must come under the third head as shown by the following experiments. The Experiments. The plan of the experiments is briefly as following: Males and females were cut up separately in dishes of clean sea-water. The bits of tissue were carefully removed, the dish of eggs being handled with utmost care to prevent unnecessary agitation. The eggs and sperm suspensions were filtered after having remained in sea-water for varying lengths of time. Sexual products treated thus are designated “washed eggs’’ and “washed sperm.’” Males and females were thoroughly dried on filter paper or clean sheer linen. The males were cut up in dried clean watch glasses; the females were cut up in the same way or pricked when 1 That the resistance of eggs and sperm of both Nereis and Platynereis is unequal would seem probable from the following: If to a Nereis sperm suspension janus green be added the fertilizing power of the sperm is in no wise impaired; or if the dye be added to sea-water the living males absorb it readily without any injurious effect on the sperm. The same quantities of the dye in sea-water is toxic to the egg before or at insemination. Eggs taken from a female Platynereis that has been swimming in a janus green-sea-water solution that is not toxic to the males or their sperm will not fertilize. Cf. also action of nicotine on Strongylocentrotus sperm and eggs as observed by the Hertwigs. 2 Several methods were used for ‘‘washing”’ sperm and freeing them of sea- water, among others that of centrifuging at high speed for six minutes. These were all abandoned for the method here described. 100 E. E. JUST. most of the eggs that escaped were collected in dry watch crystals. Bits of tissue were always removed. Such eggs and sperm are ‘‘dry eggs’’ and “‘dry sperm.” For a given experiment eggs and sperm were mixed and after an interval of time varying from five to sixty seconds flooded with sea-water. Four kinds of inseminations were made: Washed eggs X washed sperm. Washed eggs X dry sperm. Dry eggs xX dry sperm. Dry eggs xX washed sperm. The experiments fall into two groups: “A.M. inseminations” —made the morning after the worms were captured; and “P.M. inseminations’’—made during the evening of capture. The following table gives a summary of results: TABLE I. Eggs. Sperm. 5 Group. Development. Wrashedieee ree WSN. scan adde ASIN BHoGl IPSIMos Go p5400 None. WRISINEG!s bo 500006 Dry A.M. and P.M.......... None. DD iny dere mert: DD) reyanrarein eta ieee A.M.and P.M.......... Cleavage and larve. ID vere trsasretattcnsisy Washed......... AIM oy oe Sones eoeanee None. IDSA auch mabe aOR Wiashediae ssn oe [PAINT Ie kee iraen ene Cleavage and larve. Washed eggs, inseminated with dry or washed sperm, never reach cleavage stages nor do they ever produce swimming forms. I have commented above on the dry egg X dry sperm series. These eggs cleave and later produce normal larvae. Washed sperm X dry eggs of the A.M. group (1912) did not yield cleavage or swimming forms. The worms do not thrive well in the laboratory. The practise, therefore, of conducting experiments the morning after capture has been since I912 prac- tically abandoned. The only test for the vitality of the worms is copulation—a test the very nature of which precludes experiment. Doubtless, therefore, this set of experiments gave no results because the animals were not fit. Study of sections of eggs normally inseminated and laid as early as 5 A.M. shows a large percentage in the germinal vesicle stage. I have made counts in dishes of living eggs to show at the later cleavage stages the pro- portion of eggs still inthe germinal vesicle stage. For example, FERTILIZATION IN PLATYNEREIS MEGALOPS. IOI August 8, 1912, 2 P.M., six hours after laying of 10,851 eggs (from one female) six per cent. were still in the germinal vesicle stage. Other counts of living eggs and of sections show higher percentages. Every egg laid the night of capture cleaves. Dry inseminations, day or night, at best never give more than ninety per cent. of cleavages. The poor quality of the animals after several hours in the laboratory may account for the failure of the dry eggs X washed sperm A.M. group to cleave. But since the dry eggs X dry sperm A.M. series gives cleavage, I am rather inclined to believe that the method used was poor: for instance, the filter paper then used was too soft allowing the loss of most of the spermatozoa or too much water was left when the dry eggs were added. The results with dry eggs X washed sperm, P.M. group are wonderfully uniform and show conclusively that the sea-water, at least for the exposures used, has no harmful effect on the sperm. The method used is simple. As soon as possible after capture one to three males are cut up in from 8 drops to 20 c.c. of sea-water and allowed to stand upward to twenty minutes. (The sperm are active after having been in sea-water for twelve hours.) The sperm suspension is then filtered. I used a very hard filter paper. This paper was then tilted and thoroughly drained until under the lamplight the glistening water was thoroughly absorbed. A dried female was cut up on the filter paper or pricked and the eggs thus procured rolled over the paper to reach the sperm left behind or caught in the pores of the filter. The whole was then put in a dish of clean sea-water. It would be tedious to cite the individual experiments. They show conclus- ively that dry eggs inseminated with washed sperm develop in normal fashion. Now since, as has been shown above, there is a minimal amount of sea-water that will permit fertilization, dry eggs _ ought to fertilize if put on the filter paper before all the water has been absorbed. Such indeed is the case. Moreover, dry eggs put in two drops of thin sperm suspension develop. From a suspension made by cutting up one or more males in sea-water two drops are taken. Dry eggs put in this cleave and next morning swim. 102 E. E. JUST. This observation led to a series of experiments (during 1913 and 1914) designed to ascertain whether or not the density of the sperm suspension is a factor in the fertilization of Platynereis. These experiments prove in general that the number of dry eggs added to sperm suspensions that develop depends upon the density of the suspension. The denser the suspension the larger the number of trochophores. Moreover, for dense suspensions the minimum amount of sea-water permitting fertilization appears to be slightly higher than for thin suspensions. Cleavage is directly a function of the chances of the spermatozoa reaching the egg before the fertilizing substance is lost. The time of flooding with sea-water after insemination is also important for the highest percentage of cleavage. But these factors cannot be expressed with mathematical exactness. Some points, particularly with reference to inseminations with dense suspension need further experiments to determine their signi- ficance. That the egg when exposed to the action of sea-water quickly loses something necessary for fertilization must be the conclusion drawn from these experiments with washed or unwashed eggs. Even thirty seconds residence in sea-water, as repeatedly proved, is sufficient to inhibit cleavage in every single egg. If dry eggs from a single female be put in five cubic centimeters of sea-water and thoroughly drained as soon as they settle they will not de- velop after insemination although this procedure may take but a half minute. The egg alone is affected by sea-water; the fer- tilizing power of the sperm is not affected by exposure to sea- water. 3. The Nature of the Fertilizing Substance. The fertilizing substance once lost cannot be restored. If washed eggs be mixed with an extract obtained by crushing dry eggs in one or two drops of sea-water and dry sperm added, cleavage does not result. I lay no stress on this, however, for it seems to me that such an extract might yield anything. The presence of various substances in the sea-water or the lowering of the temperature of the sea-water does not prevent or restore the loss of this substance. FERTILIZATION IN PLATYNEREIS MEGALOPS. 103 KOH.—Eggs were teased out of the female directly into sea- water plus KOH in various proportions. Or, eggs from dried females were placed in the solution. After remaining from thirty seconds to two minutes in the alkaline sea-water the eggs were inseminated dry and flooded with sea-water. In other cases in- seminations were made in the solutions. Washed eggs were similarly treated. Whatever the method alkaline sea-water never gave cleavage. (Cf. sections on cross fertilization and artificial parthenogenesis. ) Hypertonic and Hypotonic Sea-water.—Egegs, both washed and dry, were treated with 244 M KCI + sea-water as follows: I. 1 drop 23-M KCl + to drops of sea-water. 2. 2 drops “ Bt verte Th Se enum on SEES aie gies pire Diese Sg eee ca iL 'S 7) Oa age Gi Sie Gh ein peat pens er ce S OxiOiie Ay ape nes hatte Bs Dry sperm were added at once and the dishes flooded with sea-water after five minutes. Or, after treatment for varying number of minutes the eggs were inseminated dry. The eggs developed no farther than with KCl treatment alone (see beyond); they form jelly and maturate. Hypotonic solutions used similarly gave no cleavage. Ether.—The following table is a summary of the experiments with ether: Eggs. Solutions Used. Exposure. Inseminations. Washed, .3 to .6 per cent. I to 5 minutes dry; in the solution. Dry, oe oe oe be oe oe Teased, sé ee ee 6eé “é ce “Teased”’ eggs are those got by cutting up the female in the ether-sea-water. , A few eggs form jelly and maturate after the ether treatment. Compared with sea-water inseminations, ether cuts down the per cent. of maturations. According to R. S. Lillie (12) star- fish eggs resistant to fertilization may be rendered normal by ether in low concentration. In Platynereis the condition is different. The egg is not rendered resistant to fertilization by the action of sea-water; it is weakened through loss of something 104 ES Ee aust by the sea-water since it combines but feebly with the sperm The ether as in Aséerias renders the Platynereis egg irritable since as shown by the low percentage of maturation more fertilizing substance must be secreted. KCN.—Inseminations made with washed or dry eggs during or after treatment with KCN (1 per cent. KCN and sea-water made in various proportions) gave only maturation. But the eggs will maturate in KCN alone while in the solutions. (Cf. Allyn on Chetopterus.) CaClz.—Newman found that CaCl. inhibits fertilization in Fundulus through a precipitation effect. I thought that in somewhat the same way calcium chloride might through action on the cortex inhibit the loss of the fertilizing substance in Platy- nereis. M/2 CaCl, added to sea-water in different quantities does not inhibit the loss of the substance since after the calcium chloride treatment the egg does not fertilize. Cooled Sea-water.—Sea-water was cooled to 10.5° C. and dry eggs after 30, 60 and 90 seconds’ treatment in 5 c.c. were insemi- nated at this temperature or after the cooled water was pipetted off. In some experiments the female was kept at the low tem- perature for several minutes before the eggs were cut out. 5 C.c. of sea-water were used in each experiment. The eggs never cleave, but more form jelly and maturate than controls insemi- nated in ordinary sea-water. This would seem to indicate a slowing down of the secretion. The effect of cold is just the op- posite of the effect of ether. Unfortunately, only few of these experiments were made. Perhaps they should be repeated at lower temperatures. Concerning the nature of this substance, some of my earliest notes are of interest. After insemination in sea-water I found some time later (forty minutes in one case) ‘‘sperm dancing above the eggs.’’ In 1914, I found the sperm of sea-water insemination active after twelve hours. One does not find this after dry in- semination, even with excess of sperm. Sperm in the dishes of successfully inseminated eggs are profoundly changed. Study of the movements of Platynereis sperm reveals the circular swim- ming of echinid spermatozoa, as shown by Buller, Gemmil, Winslow, and others (see also Dewitz, Ballowitz, etc.). They FERTILIZATION IN PLATYNEREIS MEGALOPS. 105 finally become quiescent through lack of oxygen! in various positions without orientation. After dry inseminations they come to rest, as can be seen after flooding the dishes, definitely oriented and not in haphazard arrangement. Clustered among the jelly hulls, their heads point toward the eggs. On occasions, I believed that I demonstrated the agglutination of the sperm by sea-water in which the eggs had been lying. The evidence is not clear-cut and more recent attempts have failed. The egg charged sea-water, however, does activate the sperm. I wish to point out the serious difficulties experienced in the series of sperm agglutination experiments. In the first place, twenty ‘‘large’”’ dried males (two and one half centimeters long) do not yield enough sperm and body fluid to make up a drop as large as a drop of dry sperm from a very small Nereis. Then again the thickest suspension got is largely made up of blood cor- puscles. I have never succeeded in procuring a ‘milky sus- pension’’—the admixture of corpuscles and body fluid giving always a pinkish mixture. And finally, one cannot always get twenty or more males necessary to make up even this thin sperm suspension. Repeated efforts, therefore, extending through two seasons have not been marked with very positive results. With Nereis sperm, the case is indisputable. If water in which Platynereis have laid eggs be taken it is found to have an agelu- tinating effect on Nereis sperm. Thus: August 18,1914. At10:15 P.M., ten females laid eggs in six c.c. of sea-water each. After five minutes some of this water was drawn off—z2o c.c. in all. Nereis sperm suspensions were made up fresh at 10:20, 10:30, 11:00 and 11:05. A drop of the sperm suspension was mounted on a slide under a raised cover slip. A drop of the water taken from the dishes of eggs was injected beneath the cover slip. Under the microscope, the quiescent sperm appeared at first intensely active, then rushed together and formed agglutinated masses among others still free-swimming. 1 This fact was brought out in 1913 when I was repeating some old observations on echinoderm spermatozoa. While experimenting with the sperm of Thyone in janus green solutions, I noted after some time had elapsed that cover-slip prepara- tions showed that bacteria present previously bluish in color had changed to a decided red. Later observations proved that as the dye was reduced in bits of tissue under the cover slip the sperm quieted down in various positions. 106 1B, 185 USI. The same experiment succeeds if one uses the water from dishes in which uninseminated eggs have remained for a few minutes. Washed eggs do not cause agglutination of Nereis sperm; water charged «by normally inseminated eggs or uninseminated eggs retains its power of agglutinating Nereis sperm after twelve hours at least, the reaction coming on more slowly. The freshly charged water acting on fresh sperm suspension gives a clear-cut and beautiful reaction. It may seem far-fetched to argue that the fertilizing substance lost by Platynereis eggs when exposed to sea-water is agglutinin or fertilizin as discovered by Lillie in Nereis and Arbacia because the washed egg, no longer fertilizable by its own sperm, can not sufficiently charge the sea-water- to agglutinate Nereis sperm. Yet I believe this is the case precisely. The agglutination of Nereis sperm by Platynereis egg-water is correlated with jelly formation in Platynereis by Nereis sperm. In sea-water in- seminations, Nereis spermatozoa are almost as effective as those of Platynereis. Added to this is the difference in behavior of Platynereis sperm in egg charged sea-water, in sea-water insemin- ations, and in dry inseminations. The evidence may be scant, but it seems to me sufficient to indicate that the substance lost which is necessary for fertilization is identical in nature with the fertilizin of Lillie. B. Cross FERTILIZATION WITH NEREIS. I have mentioned (Just, 14) the fact that it is generally taken for granted that reciprocal crossing of Nereis and Platynereis is the rule. This led me to attempt cross fertilization. Cross fertilization never produces segmentation or development though it may induce the maturation process. Of the methods used in echinoderm hybridization—those of Loeb, Tennent,! etc.: (1) high temperature; (2) treatment with fresh water; (3) treatment with alkalis; (4) allowing the eggs to stand; and (5) polyspermy—all were tried except the first. Since the eggs of Platynereis are normally inseminated in the body cavity and therefore with little sea-water, I tried “‘dry 1Dr. Tennent in 1912 very kindly communicated to me at length his latest methods in echinoderm hybridization. FERTILIZATION IN PLATYNEREIS MEGALOPS. 107 inseminations’: 2. e., Nereis males were cut up dry and a drop of the sperm without the addition of sea-water added to eggs of Platynereis cut up dry. Inseminations were made in a variety of ways as the following table of method shows: TABLE II. SUMMARY OF INSEMINATIONS MADE IN IQII, I912, I913, AND I9I4 Platynereis sperm on Nereis egg. I. Few sperm in sea-water. Fresh eggs in sea-water. 2. Dense sperm suspension. 3. Few sperm in sea-water. Stale eggs in sea-water. 4. Dense sperm suspension. ; 5. Few sperm, dry, Fresh eggs dry. 6. Heavy insemination dry. . Few sperm, dry. Stale eggs washed. ~w 8. Heavy insemination, dry. Reciprocal crosses of Platynereis eggs and Nereis sperm were made. “Stale eggs”’ are eggs that have stood in sea-water for several hours. ‘‘Stale eggs, washed”’ are stale eggs on which the water has been changed several times. These experiments were made repeatedly during four seasons. The sperm of Platynerets has practically no effect on the egg of Nereis whether fresh or stale, dry or in sea-water. In one ex- periment (1911) I got jelly formation in a few eggs. This ex- periment later repeated (1913) gave no result. If Nereis eggs be inseminated with Platynereis sperm during the evening of capture they show no change the next morning. Inseminated with Nereis sperm twelve hours after insemination with Platy- nereis sperm, the eggs develop normally if anything in greater numbers than such stale eggs in ordinary sea-water do. Nereis sperm will cause Platynereis eggs to form jelly, the per cent. of eggs thus responding depending upon the amount of sea-water used and the density of the sperm suspension. But in general many of the eggs fail to form jelly.or go through matura- tion. Many that maturate do so with the cortex partially or wholly intact. Sections of these eggs preserved at three minute 108 EOE. just intervals after insemination have been studied. The sperm does not enter; or, if it enters must disintegrate early for I have never found sperm nuclei in these preparations. — ! Clearly, then, one may not use the eggs of these worms in- discriminately. C. ARTIFICIAL PARTHENOGENESIS. The following agents have been used in an attempt to bring about artificial parthenogenesis in the egg of Platynereis megalops: Centrifuging, KCI, NaOH, KOH, HNOs, lath, 7. Warm sea-water. ONES Ae Cie The eggs were cut out of the worms in sea-water centrifuged; subjected to varying quantities of salt, alkalis, or acids for dif- ferent lengths of time; or warmed in sea-water for from five to thirty minutes at 35° C. . These methods gave polar body forma- tion, cytoplasmic changes, fusion of the oil drops, and finally chromatin disintegration in the animal hemisphere. The eggs never cleaved. Study of the literature reveals the fact that the clearest cases of artificial parthenogenesis closely simulating the normal in cleavage and in larval development are of those eggs that have formed one or both polar bodies when shed: the echinids, for example, and the asteroids. Other eggs shed in the germinal vesicle stage like those of Polynoe (Loeb ’08), Amphitrite (Loeb "oI; Scott.) Nereis (Lillie ’11), etc., give only differentiation without cleavage or incomplete cleavage. Loeb and Wasteneys’ work on Chaetopterus with ox serum as well as Miss Allyn’s on the same egg with heat are exceptions. The great exception to the general statement made above is Thalasema (Lefevre) where it appears with single substances, acids mostly, normal development is closely simulated. On the whole, however, ovocytes yield less readily to parthenogenetic agents than mature ova. FERTILIZATION IN PLATYNEREIS MEGALOPS. 109 Mathews’ experiments (’o1) on Asterias may in this connection be cited. He found that when the eggs of this starfish were got while still in the germinal vesicle stage shaking would produce development only after the eggs had remained in sea-water until maturation was gone through with. Sea-water acts as a first stimulus and mechanical shock induces further development. So R. S. Lillie (08) on the same egg finds that its responsiveness to momentary elevation of temperature as a means of producing artificial parthenogenesis ‘‘ varies greatly at different periods in the life of the egg.” ‘“‘The most favorable period is some little time (10 to 20 minutes) before the separation of the first polar body.” Reasoning thus, I thought that I might carry Platynereis eggs through maturation with one agent and then through cleavage with another. Eggs were, therefore, treated with KCl, KOH, and NaOH in sea-water for various lengths of time and then subjected to heat, shaking, and centrifugal force. In no case did I procure cleavage although the first agent in each case caused maturation. With Nereis, on the other hand, KCl and subsequent warming in sea-water induces development (see Just 7150). It is interesting to note that eggs subjected to heat in the minute quantities of sea-water that permit fertilization do not develop beyond maturation. Apparently, the conditions for successful artificial initiation of development are more exacting than those for successful insemination. We may conclude, then, that the results of attempted cross fertilization and artificial parthenogenesis are harmonious with those of sea-water insemination, so far as cleavage is concerned, in their negative results. The fundamental questions are: (1) the significance of the sea-water insemination and (2) the extent to which the results with Nereis sperm and with parthenogenetic agents are capable of like interpretation. DISCUSSION. Any analysis of fertilization must deal with the phenomena from the point of view of heredity or of initiation of development. Considered as the process of initiating development, fertilization may be divided into the stages of insemination, sperm pene- tration, and germ nuclei copulation. As Lillie has repeatedly 110 E. E. JUST. pointed out! experimental evidence must be amassed testing the meaning of each of these stages. lets I. Concerning insemination, as Lillie has shown, the egg plays an important part through the production of agglutinins.? For both Arbacia and Nereis it has also been shown that chemotaxis plays a part in insemination. (Lillie, ’12, ’13a, ’13b, and ’14). I believe that Platynereis belongs to this class. I may, how- ever, be permitted again to point out the great difficulty attending the use of Platynereis eggs on this phase. All the phenomena are extremely rapid, the reactions must be very nice. The material is unfavorable for any intensive study of agglutination and che- motaxis. When one stops to think of the extremely precise reactions of the eggs, one gets a hint of the task. The carrying over of the smallest drop of sea-water above the maximum to eggs from vigorous females within the shortest time after capture will prohibit cleavage in every egg. . To answer the general question whether or not eggs secrete substances that activate the spermatozoa, I believe forms whose eggs are inseminated normally in sea-water should be used. So far as Platynereis is concerned, agglutination or not, chemotaxis or not, the egg must lose a substance or substances when in sea-water whose presence is necessary for fertilization. 2. Study of the normal fertilization of Platyneretis indicates that as in Nerets the egg plays the active réle in the penetration of the spermatozoon for it actually draws in the passive spermatozo6n. After sea-water treatment I have not, as mentioned above, found the early stages of penetration in eggs fixed at three minute intervals after insemination. Either the sperm penetration is unlike that after normal insemination or penetration takes place with extreme rapidity. In the later stages of penetration it is 1 Lectures to classes in embryology, Woods Hole, Mass. 2 Apparently Buller did not realize that he obtained iso-agglutination of sea- urchin sperm, although he speaks of the sperm forming “‘balls’’ and although the phenomena of agglutination were well known at that time. Landsteiner the year before had secured sperm agglutinating sera. Nougouchi’s work on Nereis sperm ~ is of interest: he demonstrated agglutination with snake venom. The experiments of Schiicking, von Dungern, de Meyer, and others are well known. An observation of Walker’s (’10) is likewise worthy of mention—the agglutination of the sperm of the rat when mixed with the seminal vesicle secretion of the same animal. Chemotaxis of sperm has been demonstrated for mammals—see for instance, Low. FERTILIZATION IN PLATYNEREIS MEGALOPS. III clear that the spermatozoa behave in abnormal fashion even granting that I may have overlooked the amphiaster. The evidence seems to indicate that after sea-water treatment the egg lacks the power to engulf the sperm. However, whatever the method of penetration one point is beyond contradiction: these washed eggs never cleave. The observations agree with those of Lillie (14) who notes * that some unpublished observations in the case of Nereis show that “‘if the cortical changes be induced by artificial means there is a brief period in which insemination of the eggs may be followed by penetration of the spermatozo6n, but without causing cleavage of the egg.’’ Miss Allyn found that after KCl treatment of the egg of Chetopterus, the spermatozo6n may enter but its behavior is not normal. Kite (quoted from Lillie ’14) finds that sper- matozoa injected into star-fish eggs never give cleavage. ji In these cases, the interpretation must be that the ‘‘fertiliz- able”’ condition of the egg has been destroyed through loss of fertilizin before insemination. In the same way sperm may pene- trate unripe eggs as Hempelmann has shown for Saccocirrus (so too, von Hofsten for Otomesostoma and Shearer for Dinophilus gyro- ciliatus). Two yearsago I found that eggs from Nereis limbata just before transformation into the heteronereis phase would not fer- tilize with active sperm either from the nereis or heceroneris form. Moreover, eggs from metamorphosing worms kept for several weeks in the laboratory although apparently ripe would not fertilize on insemination during the dark of the moon. At full moon, sometimes but a few days later, eggs from the same animal would fertilize and develop into larve which were kept for weeks. We may assume in these cases that the fertilizin is either absent or is unavailable. Penetration, therefore, may take place before the fertilizable period is reached as well as after it has been passed, but the egg is not capable of fertilization. 3. Apposition of the germ nuclei of Platynereis after sea-water insemination may ensue, but never cleavage. After the loss of the fertilizing substance, then, the normal fertilization process may be closely simulated even to the point of the copulation of the pronuclei but development never goes beyond this point. In short, the normal fertilization process demands at the very Tpn2 E. E. JUST. outset the fixation by the spermatozo6n of the escaping fertilizin. This takes place in Platynereis almost instantaneously (see page 93) but brief though this phase may be it cannot be omitted. The experiments with Nereis sperm and agents of artificial parthenogenesis demand explanation. Eggs such as those of echinids used in cross fertilization (Loeb, Tennent, Baltzer, Herbst, etc.) or in artificial parthenogenesis when subjected to treatment are so subjected with their substances intact. They are normally shed in sea-water for insemination and the sea-water does not for some time destroy their fertilizing power. Platy- nereis eggs when subjected in sea-water to foreign sperm or to various agents have lost something through the action of sea- water. This very ‘‘something’’ is necessary for artificial par- thenogenesis and, moreover, as shown above (for Nereis also) must be present in greater quantity than necessary for fertiliza- tion. I am emboldened further to suggest that eggs normally inseminated in the ovocyte stage yield to parthenogenetic agents only with difficulty because they lose fertilizin at the impact of the first stimulus—chemical treatment, shock, etc. Sperm alone, in most cases, are strong enough by fixation of the fertilizin to carry such eggs through their dual phase—maturation and fertilization. Whether by sperm, then, or by artificial agents, the initiation of development is fundamentally the same.1 The egg plays the leading rdéle; it needs but to have its fertilizin ac- tivated in order to develop. The observations on Platynereis were rendered less difficult because of the study of the maturation and fertilization in Nereis. For this study I was fortunate to be able to supplement my own slides with two series lent me by Professor F. R. Lillie. It isa genuine pleasure here to acknowledge my further indebtedness to him for his many suggestions and for his stimulating interest in the Platynerets studies begun at his suggestion and under his direction. MARINE BIOLOGICAL LABORATORY, Woop’s Hote, Mass. 1] think that Martin Jacoby’s experiments support this view. He found (Bizo- chem. Zeit., 20, 333-335) that serum from rabbits into which eggs had been injected showed an increased power to stimulate parthenogenetic development of the eggs. He also found (ibid., pp. 336-343) that an enzyme which may be extracted from sperm and from eggs after sperm penetration may be got from parthenogenetic eggs. FERTILIZATION IN PLATYNEREIS MEGALOPS. 113 LITERATURE CITED. Allyn, Harriett M. I2 The Initiation of Development in Chaetopterus. BioL. BULL., 24. Budington, R. A. 932 The Influence of Magnesium Chloride on the fertilizing Potential of Sperma- tozoa. Science, N. S., 35. Buller, A. H. R. 200 «6h fertilizing Process in the Echinoidea. Report, British As. Ad. of Sci. 202 Is Chemotaxis a Factor in the Fertilization of the Eggs of Animals? OQ. J. M.S., 46. Dungay, N. S. ’13, A Study of the Effects. of Inquiry upon the fertilizing Power of Sperm. BIOL. BULL., 25. Frédericq, L. -%04 Sur la concentration moleculaire du sang: et des tissus chez les animaux aquatiques. Arch. de Biol., 20. Gemmil, Jas. F. 700 ©On the Vitality of the Ova and Sperm of certain Animals. Jour. Anat. and Phys., 34. Garrey, W. E. 204 Osmotic Pressure of Sea-water and of the Blood of marine Animals. BIOL. BULL., 7. Hempelmann, F. ’12 Die Geschlechtsorgane und -zellen von Saccocirrus. Zoologica, Heft 69. Hirokawa, Waichi 09 Ueber den Einfluss des Prostatasekrete und der Samenfliissigkeit auf die Vitalitat der Spermatozoen. Biochem. Ztschr., 19. Jacoby, M. ’r0 Ueber das Verhalten der Sperma- und Eienzyme bei der Befruchtung und ersten Entwicklung. Biochem. Ztschr., 26, 336-343. Just, E. E. ’14 Breeding Habits of the heteronereis form of Platynereis megalops. BtOL, BULL., 27. ’r5a The Morphology of the normal Fertilization in Platynereis megalops. Jour. Morph., in press. *15b = Initiation of Development in Nereis. Biov. BULL., 28. Koltzoff, N. K. 09 ~©60Studien iiber die Gestalt der Zelle, ii. Arch. f. Zellforsch,, 2. Landsteiner, K. ’99 «©6Zuur Kennteris der spezifisch auf Blut kérperchen Wirkenden Sera. Cent. f. Bak., 25. Lefevre, G. 202 ~=6Artificial Parthenogenesis in Thalassema mellita. Jour. Ex. Zool., 4. Lillie, F. R. Ir Studies of Fertilization in Nereis. I. The Cortical Changes in the Egg. Il. Partial Fertilization. Jour. Morph., 22. ’72 III. The Morphology of the normal Fertilization. IV. The Fertilizing Power of Portions of the Spermatozoon. Jour. Ex. Zool., 12. ’13a V. The Behavior of the Spermatozoa of Nereis and Arbacia with special Reference to Egg-extractives. Jour. Ex. Zool., 14. 114 E: Esujust: °13b The Mechanism of Fertilization. Science, N. S., 38. ’r4 Studies of Fertilization, VI. The Mechanism of Fertilization in Arbacza. Jour. Ex. Zool., 16. Lillie, R. S. 08 Momentary Elevation of Temperature as a Means of producing artificial Parthenogenesis in Star-fish Eggs and the Conditions of its Action. Jour. Ex. Zool., 5. ’72 Certain Means by which Star-fish Eggs naturally Resistant to Fertilization may be rendered normal and the physiological Conditions of this Action. BIOL. BULL., 22. Loeb, J., Fischer, M., and Neilson, H. 70x Arch. f. d. Ges. Physiol., 87. Loeb, J. 708 Ueber die Entwicklungserregung unbefruchteter Annelideneier (Polynoe) mittels Saponin und Solanin. Pfliiger’s Arch., 122. Loeb, J., and Wasteneys, H. ’12 ©Fertilization of the Eggs of various Invertebrates by Ox Serum. Science, 36. Liéw, Otto 202—’03 + Die Chemotaxis der Spermatozoen in weiblichen Genitaltract. Zitz. der Kaiserlichen Math-Naturwissen. Classe. 111-112. Mathews, A. P. 701 Artificial Parthenogenesis produced by mechanical Agitation. Am. Jour. Phys., 6. de Meyer, J. "11 Observations et Experiénces relatives a l’action exercée par des extraits d’oeufs et d’autres substances sur les spermatozoides. Arch. de Biol., 26. Schiicking, A. ’03 Zur Physiologie der Befruchtung, Parthenogenese, und Entwicklung. Arch. f. d. Ges. Physiol., 98. Steinach, E. ’94 Untersuchungen zur ver gleichenden Physiologie der mannlichen Geschlechts- organe insbesondere der accessorichen Geschlechtsdriisen. Arch. f. d. Ges. Phys., 56. Walker, Geo. ’99:«€Cf Beitrag zur Kenntnis der Anatomie und Physiologie der Prostata nebst Bemerkungen iiber der Vorgang der Ejaculation. Arch. f. Anat. und Physiol. ’10 The Nature of the Secretion of the Vesiculi Seminalis and of an adjacent glandular Structure in the Rat and Guinea Pig, with special Reference to the Occurrence of Histone in the Former. Johns Hopkins Hosp. Bull., 21. "11 The Effect on Breeding of the Removal of the Prostate Gland or of the Vesiculi Seminalis or of Both, together with Observations on the Condition of the Testes after such Operations on the White Rats. Johns Hopkins Hosp. Rep., 16. ‘ SPOROCYSTS IN AN ANNELID.1 EDWIN LINTON, WASHINGTON AND JEFFERSON COLLEGE, WASHINGTON, PA. In the summer of 1910, while at work at the United States Fisheries Biological Station, Woods Hole, Mass., I was told by -Dr. Gilman A. Drew that what were supposed to be cercarize had been noticed at different times associated with the annelid Hydroides dianthus Verrill among material being used for study at the Marine Biological Laboratory. Acting upon this suggestion I examined a large number of these serpulids on several dates in August of that year. Although much of the material was examined very minutely, the worms having been removed from the tubes, teased, and everything that even remotely resembled a sporocyst further examined, neither sporocysts nor cercariz were found. In the following summer I secured two lots of these sporocysts from this annelid. For the first lot, July 15, I am indebted to Dr. Drew, and for the second, July 21, to Miss Margaret Morris. In each case the single annelid was lying in a dish of sea water, and in the bottom of the dish there were a large number of sporo- cysts. These sporocysts were found to contain cercarize in various stages of development but no rediz. As they lay free in the sea water the sporocysts were for the most part white, or bluish translucent white. In some of them there were varying amounts of orange pigment of similar appearance to the abundant pigment in the annelid. They were short and thick, bluntly rounded at the ends, and more or less arcuate. In some cases they were curved until the ends almost touched each other. Many of the second lot were orange yellow, also many of them were actively contractile. A frequent change of shape was that from the characteristic short, blunt-pointed sub-cylindrical form to a fusiform shape with elongated and slender-pointed 1 Published by permission of Commissioner of Fisheries. 115 116 EDWIN LINTON. ends. In this condition they were sometimes straight and some- times arcuate (Fig. 1). In each sporocyst there were tailed cer- Fic. 1. Contraction shapes of sporocysts, life. carie along with various stages of developing cercariz, from globular balls of cells, 0.04 millimeter in diameter, to cercarie, 0.5 millimeter or more in length (Fig. 2). The anterior portion Fic. 2. Sporocyst showing cercarie in different stages of development. Cam- era lucida sketch of stained and mounted specimen. Actual length 1.04 millimeter. of these cercariz is sub-cylindrical and slightly tapering at the anterior end. It is marked off from the elongated tail portion by a constriction, which, in the mature cercariz, is at about the anterior fourth of the entire length. In other words, the tail, which is forked at the end, is about three times the length of the body. Along the dorsal aspect of the body in a few instances a longitudinal row of exceedingly slender spines was noted. As this cercaria resembled very closely a cercaria which I have found in the scallop (Pecten trradians), I recorded in my notes that it was likely that these spines, as in the cercaria from the scallop, are remnants of a fin-like membrane, and that the type represented by this cercaria is evidently near that of Cercaria cri tata L> Val. SPOROCYSTS IN AN ANNELID. 117 On July 19, 1914, through the kindness of Dr. E. J. Lund, I had the opportunity of examining another lot of these cercariz from this same annelid. Some of these were observed to be covered with an exceedingly thin hyaline membrane which be- comes constricted at frequent intervals, the constrictions ulti- mately being the only part of the membrane that is visible. The cer- cariz from Hydroides, as was the case with those from the scallop, exhibit great activity, but the nature of their movement is different. Instead of a characteristic pecking motion of the anterior end, the cercarie from the annelid, occasionally, after lying mo- tionless for a time, perform exceedingly rapid wriggling movements. The an- terior end of the body is provided with a short, retractile boring appa- ratus, shown protruded in Fig. 3. In addition to the various stages of developing cercariz, other structures were observed in these sporocysts. With transmitted light these appeared to be granular, but with reflected light, or, with high magnification, they appear to consist of minute oil drop- lets, at least in part. In some cases they were distributed rather uniformly near the surface, in others they were massed in the central region. The number of these sporocysts is very great. In the first lot it was estimated that there were between 900 and 1,000 sporocysts in the dish with the annelid. After the worm had been lying for a few minutes in a dish of clean sea water to which it had been Fic. 3. Cercaria with fin-like crest. The thin investing mem- brane is indicated at the poste- rior end. Length of body 0.17 millimeter, length of tail, 0.52. Fic. 4. Anterior end of a cer- caria with a crest of slender, cilia-like spines. transferred, a number, 25 or more, of sporocysts made their appearance on the bottom of the dish. The worm was then _ 118 EDWIN LINTON. placed in corrosive-acetic and afterwards sectioned. The sections show an immense number of sporocysts (Fig. 5). For the most Fic. 5. Transverse section of Hydroides dianthus showing sporocysts embedded in the body wall. a, intestine. part they are distributed ventrally in the inner portion of the body wall, although a few lay among the muscles near the exterior, and a few in the epidermis. If I interpret the sections correctly, the sporocysts escape from the ventral side of the serpulid, where the body wall is comparatively thin, and where the sporocysts are in greatest numbers. There is considerable variation in the size of the sporocysts. The largest noted was 0.70 millimeter in length and 0.28 milli- meter in diameter; the smallest 0.17 in length and 0.10 in di- ameter. In like manner the cercarie varied in length, but the length of 0.12 millimeter for the anterior portion, and 0.36 for the tail, or 0.48 millimeter for the whole length, is not far from the usual length of a mature cercaria. One cercaria, living, had the following dimensions: Length of body 0.17 millimeter, breadth 0.04; length of tail 0.52, breadth 0.02. What were interpreted to be striated muscle fibers were noticed in the tails of living cercarie (Fig. 3). These fibers extend diagonally backward and inward from the exterior to the median line. They were about 0.0017 millimeter in diameter, and what appeared to be cross striations were plainly visible with a Zeiss D objective. Under an oil immersion lens their resemblance to striated muscle was evident. é OF THE _ Marine ‘Biological ‘Laboratory / _ Woops HOLE, MASP. vy ‘A Ae See eO RS earls eC Oo UNEAREH, 1915 Tara oN oS i tion in the Domestic Fowl. es tee Pine a Sutnertann, G. Woo. | Nuclear hinge) TH ibe amen =: ee : pe es Cord we ade Oe : BME utele Baccus Ve at - clamitans idee Be aa) beeen. : os 4 Ricans, Ao AND a Note on the Effect - ne Radiation on : phony aap: AS EL Fertilizin Udeties Hebe Pakcuyuatar tae dude IE “Can a Siele “Spermatozoon Initiate Development on. Arbacta? a SESS ea USiidies on the Physiology 154 ea e 5 a ey i “3 Pobisuten Mowry ‘BY THE "MARINE BIOLOGICAL. LABORATORY Py sae i a cy PRINTED AND isSUED BY “THE NEW ERA. PRINTING COMPANY LANCASTER, PA. t 4 ane ron GREAT Berea: : ae FOR GERMANY. “WILLIAM WESLEY ~- R. FRIEDLANDER Soe 8 SON Nag ns SO a) OO SOI, 1h 28 biel See Rigs s one pee is ee Berlin N. We ay: o es ae W. C. RE siya a ae co Che Cane Tr : a ae Me i Single, Numbers, 75 Gents. Per Volume 6 numbers) cae 00. Tider Entered Oetober 10, 1902, ‘at Lancaster, Pa., as second- clase mutter, under Act of Congress, of J Aly, 16, 1894. a Te ah ey xe Bet Vol. XXVIII. March, 1915. No. 3 PPIOLOGICAL BULLETIN NUCLEAR CHANGES IN THE REGENERATING SPINAL” aati CORD OF THE TADPOLE OF RANA CLAMITANS."’ GEORGE FRED SUTHERLAND. CONTENTS. I. Statement of the Problem...........-----2-..-2-eeee eects eet eeeee IQ Il. Material and Methods....>......... 1 se OM TERN ES, 3c SOAP ana 121 Tit, OPSAAVAOMS. obo nGcecu ss ode oc od Go dn GS bacco ocsopogHoeawan op UO ObS 122 I. Degenerative Changes after an Operation..........-..---+-+-+--:- 123 2. Enlargement of Nuclei.............-.0 eee ee tee ete 129 3. Temporary Partial Closing of the Spinal Cord............------- I29 A, (Call IDIOM 65 oconneccounboogeooccouan ace scos boon GHD e DoE DG I31 IW) IDYSGURSH elite Gio 6 ables voocs Ob baw Udo Be mine obiG ou 5.b dln o'a cs lolgum pl ciceo mrcro- 134 1. Amitosis and Fragmentation.............-.-. eee eee eee eee 134 2. Appearance of Leucocytes......-- 1... 222+. eee eee ett 135 3. Temporary Closing of the Spinal Cord.............-...--.++-05. 136 4. Rate of Division. Amitosis vs. Mitosis.......-....+++++-++-++--- 137 Wi, QWMINIAT) Goon oc coos oancn oon geo uGaddh ode ooguogn sD oooooSECoadGaS 138 Wil, Ihilbiliqereeyplinys 500 aco cocade anon be godaguadandsanoudoocospevKd Dour 139 I. STATEMENT OF THE PROBLEM. The present paper gives the results of an histological study of the early stages of regeneration in the spinal cord of the frog tadpole, Rana clamitans. It deals especially with the degen- erative nuclear changes immediately following the operation, and the phenomena of nuclear division in the formation of the new organ. Fraisse (1885) studied these stages in several vertebrates in order to discover the origin of the regenerated tissues, and pre- sented the following conclusions which may be used as a basis for a further detailed study. “‘t. Sowohl bei Amphibien wie bei Reptilien sind verletzte Gewebe nur im Stande, wiederum gleichartig Gewebe zu erzeugen. Die Leukocyten iibernehmen bei der Gewebsbildung nur die 1Contribution from the Zoological Laboratory of the University of Illinois, No. 37. 119 I20 GEORGE FRED SUTHERLAND. Function der Ernahrung; ausserdem nehmen sie zerfallende Gewebsproducte auf und assimiliren dieselben, um sie an anderen Orten wieder zu deponiren. Niemals werden sie selbst zu fixen Gewebszellen, weder in der Bindesubstanz noch sonst wo. ‘“‘2, Sammtliche der in Frage kommenden Gewebe der Am- phibien und Reptilien sind im Stande, sich zu regeneriren; entweder direct aus ihren Elementen, oder aus einer Matrix, so lange diese Matrix unverletzt ist. Als Matrix fiir die Epidermis ist das Rete Malpighii, fiir das centrale Nervensystem das Epithel des Centralcanales, fiir die Muskulatur die Muskel- kérperchen zu betrachten. ‘““2, Zuerst regeneriren sich Epithel und Bindegewebe; beides scharf getrennt, urspriinglich aus gleichartigen Zellen bestehend, die sich spater differenziren.”’ There remains the further problem of the stages in the process by which the old organs at the cut surface replace their lost parts. Two distinct kinds of changes take place in this process, (1) degenerative and (2) regenerative. First the injured cells at the cut edge degenerate. Then follows regeneration proper, or the formation of the new organ from the remaining elements of the old. There are three ways in which regeneration proper might take place. (1) The cells at the cut edge of each organ by dividing might extend outward, and in time form the completed organ: (2) the cells in front of the cut edge might wander back- ward; and (3) the cells in front of the cut edge might divide in situ and push backward the more distal cells. These possible methods of regeneration will be made clearer by a diagram of that part of the hollow neural tube extending forward from the cut (Fig. 1). If (1) (division of cells at the cut edge) were the method of regeneration, we should find after the operation that ANTERIOR 1 Cur 1] 4 Fic. 1. Diagram, explained in the text. the cells at the cut surface A, or from A to C, are dividing rapidly while from C to B about the normal number of cells is dividing. NUCLEAR CHANGES IN RANA CLAMITANS. I21 If (2) (migration of cells) were the method, we might find no dividing cells at all, but should expect to find that the cells from B to A or possibly only from D to A are turned with their long axes parallel to the longitudinal axis of the spinal cord as if moving toward the cut end. If (3) (division of more anterior cells in situ) were the method, we should expect to find dividing cells all the way from B to C or possibly concentrated in a growing zone ED. The present paper aims to give an account of the nuclear changes, both degenerative and regenerative, involved in the formation of the regenerated spinal cord. II. MATERIAL AND METHODS. Serial sections were made of tadpole tails killed after various regeneration periods. This enables one to follow the process from stage to stage. But to get uniform results from this method and eliminate individual variations, one must take tadpoles as nearly alike as possible at the start, operate on all at the same time, keep them under uniform laboratory conditions and make sections of several individuals at each stage. On October 12, 1913, seventy tadpoles of Rana clamitans, varying in length from 30 to 60 mm., were brought into the lab- oratory. Two days later they were put into individual finger bowls, and forty-four medium sized individuals (32-40 mm. in length), chosen to constitute the main series, were grouped by twos or threes. Those of each group were as nearly alike as possible and each group was treated as a unit in the time of operation, killing, etc. The finger bowls were placed side by side on a table some distance from the windows so that uniform conditions of temperature, light, etc., were insured. None of the tadpoles was fed during the course of the experiment, and none died from the effects of laboratory conditions. On October 15, the first operations were performed. Each tadpole was transferred from the finger bowl to a paraffin block and approximately one fourth of the tail was removed, with a sharp scalpel, at right angles to the plane of the tail. The animal was returned to the finger bowl and the removed part put into Gilson’s killing fluid. At the end of the period of regener- ation, the animals were again taken out onto the block and the 122 GEORGE FRED SUTHERLAND. regenerated tail plus a second fourth of the normal tail was removed and put immediately into Gilson’s killing fluid. The times of killing were as follows: normal, immediately after the operation, I, 3, 54%, 9%, and 14 hours, and 1, 2, 3, 4, 6, 8, 9, Io, 12, 14 and 16 days after the operation. Usual methods of technique were followed. Delafield’s hematoxylin and acid fuchsin stain the nuclei blue and the cytoplasm pink, but do not distinctly bring out cell boundaries. For the most part sections were made in the sagittal plane. III. OBSERVATIONS. The study was confined to the histology of regeneration in the spinal cord, since a preliminary examination showed that this organ of all those in the tail was best adapted for a study of the present problem. Fig. 2 shows by a sagittal section the spinal Fic. 2. Sagittal section through a part of the normal! tail, showing the spinal cord and its relation to the surrounding tissues. mc, spinal cord; cc, central canal; ntc, notocord; ct, connective tissue; pc, pigment cell. (330 diameters.) cord, and its relation to the surrounding tissues. Fig. 3 shows a transverse section of the spinal cord alone. It is a hollow tube which distally is formed of a single layer of cells. The nuclei are very near the inner border of the cells so that there is a wide outer zone of cytoplasm but practically no inner cytoplasmic NUCLEAR CHANGES IN RANA CLAMITANS. 123 zone. At this stage in the development of the tadpole, the cells near the distal end of the spinal cord show little differentiation. Fic. 3. Transverse section through the normal spinal cord, showing the nuclei and the outer cytoplasmic zone. cc, central canal. (890 diameters.) 1. Degenerative Changes after an Operation. When a tadpole’s tail is removed the old notocord extends out beyond the other tissues, and the connective tissue between the notocord and spinal cord:’is usually broken so that the spinal cord bends dorsally as in Figs. 7 and 8. A transverse cut through the tail leaves the various organs at the cut surface in contact with the surrounding medium, the water in which the tadpole lives. Sections of tadpoles killed immediately after the operation, show the direct effect of the cutting (Figs. 4 and 5). Many. Fic. 4. Transverse section through the end of the spinal cord immediately after the operation, showing deeply-staining nuclei. cc, central canal; 2m, normal nuclei; dn, deeply-staining nuclei. (920 diameters.) nuclei and cells are broken and irregular in appearance and may be loosened or torn apart from each other. The injured nuclei at the cut edge and extending forward with decreasing frequency, are homogeneous in appearance and take a deep haematoxylin stain. Undoubtedly some of the nuclei are cut, and this accounts for the irregularity in shape of a good many. But a good many others, also staining deeply, are rounded and smaller than normal nuclei. These may be either normal nuclei which under the 24 GEORGE FRED SUTHERLAND. stimulus of the operation are contracted or compressed, or cut nuclei which have rounded off. These deeply-staining nuclei, whether rounded or irregular in shape, are smaller than normal Fic. 5. Sagittal section through the side of the spinal cord immediately after the operation, showing the deeply-staining nuclei at the cut end. dn, deeply- staining nuclei. (920 diameters.) nuclei, so it may be that the chromatin, which stains deeply, is condensed on account of the loss of achromatic material. The same assumption is borne out by the somewhat different appearance of nuclei in the tadpoles killed one hour after the operation (Fig. 6). Some are rounded as before; others are angular or slightly hour-glass shaped, with rather dense cyto- plasm extending out from the corners. If parts of the nuclear Fic. 6. Sagittal section through the spinal cord one hour after the operation. This shows the “‘contracting’’ nuclei. cc, central canal; dn, deeply-staining nuclei; nn, normal nuclei. (920 diameters.) membrane were held by the cytoplasm while the nucleus as a whole decreases in volume either by contraction or loss of achro- matin, the nuclei might present such an appearance. Moreover there are gradations from hour-glass-shaped to normal nuclei NUCLEAR CHANGES IN RANA CLAMITANS. 125 and corresponding gradations in size and depth of stain. In cases of this sort there are often vacuoles or cytoplasm between the nuclei as if the latter had shrunken, whereas in the normal cord, the nuclei are so close together that no cytoplasm can be seen between them. These facts indicate that normal nuclei become deeply staining nuclei by contraction or by loss of achro- matic material. This “contraction” of nuclei seems to be caused by contact with the water or killing fluid, or the succession of the two, as well as by direct injury from the scalpel, for other nuclei which are in contact with the exterior only through the central canal show this phenomenon. In some cases, the end of a nucleus nearest the central canal is deeply stained and contracted while the other part is normal (Fig. 4). The question immediately arises, why does not the water or other external factor enter the open neural tube and cause the contraction of the inner parts of practically all nuclei in the spinal cord? It is probably because of the presence in the tube of some substance which prevents the ready admission of external fluids, though capillarity would have a similar effect. Since the sections show very little structure within the central canal, this content must be liquid or semi- liquid. However, in a number of sections there is a rather long narrow band of cytoplasmic material which may be the more solid part of a semi-liquid substance coagulated by the killing reagent. There are other evidences of the presence of such a liquid. The sections from two of the tadpoles killed one hour after the oper- ation show a coagulation of the outer surface of the blood plasma covering the wound, but over the spinal cord this coagulating process is delayed. The most plausible explanation seems to be that some cerebro-spinal fluid (compared by Barfurth to the cerebrospinal fluid of mammals) exerts an outward pressure which breaks through any slight hardening of the plasma at this point. Perhaps transference of the animal to a medium of different density, the killing fluid, aids the outburst. Sections of another tadpole killed at one hour show the presence of this coagulated plasma over the end of the spinal cord as well as over other parts of the tail. The outward pressure of a fluid would tend to push out into 126 GEORGE FRED SUTHERLAND. the blood plasma any free elements such as the injured and de- generating nuclei with very little cytoplasm and hence little connection with other cells; and when this fluid breaks through, some of these nuclei may break off and float away. At one hour after the operation, broken and small rounded nuclei are seen in betes en Fic. 7. Sagittal section through the spinal cord and the surrounding region one hour after the operation, showing irregularly shaped, deeply-staining nuclei in the end of the spinal cord and in the coagulated plasma layer. dn, deeply-staining nuclei; pl, plasma layer; cf, connective tissue; utc, notocord. (1,100 diameters.) the end of the spinal cord and extending out into the hardened layer of the plasma, giving evidence of some force acting outward at this time (Fig. 7). Other evidences will be mentioned in describing the stages at which they appear. NUCLEAR CHANGES IN RANA CLAMITANS. 127 Three hours after the operation there are fewer of the angular nuclei than at one hour and more of the round deeply-staining nuclei. The latter vary from the size of similar ones in the earlier stages down to fragments. Moreover some of the larger of these seem to be in the process of fragmentation, that is, ap- pearances indicating stages in direct division are seen. The gradation in size and depth of stain at one hour from normal nuclei nearly to rounded ones, and the gradation down to frag- ments at three hours, as well as the appearances of fragmentation, make it fairly clear that normal nuclei just in front of the cut edge may contract, become rounded, and fragment. This must be a degenerative process. Even finer intermediate steps are seen in preparations of later stages. Sections of one individual at this period appear very much like those immediately after the operation. The deeply-stained nuclei are similar, and the spinal cord is not covered either by epidermis or plasma, so that a recent outbreak of the cerebro- spinal fluid must have taken place. In this case a second contact - with the exterior has again started the degenerative process. At five and a half hours the spinal cord is entirely covered by the thickened plasma layer, in which is a group of fragmenting globular nuclei. In one preparation at this time, the epidermis has closed-in over the entire wound, and there is a series of stages in the degeneration of nuclei. Some are only slightly smaller and darker than normal nuclei; others have the angular appearance characteristic of nuclei one hour after the operation, while still others are round and fragmenting. At this stage there is another evidence of the presence of a cerebrospinal fluid. The plasma covering the end of the spinal cord is pushed outward, making a knob-like extension of the central canal similar to that shown in Fig. 8. This did not appear in earlier stages either because not enough cerebrospinal fluid was present, or because the plasma layer had not coagulated sufficiently to resist the outward pressure of this fluid. Of the two preparations of tadpoles killed after a nine and a half hour interval, one shows the epidermis and plasma covering all the wound except the neural tube; the other shows this part also covered. In the former, the sides of the neural tube are 128 GEORGE FRED SUTHERLAND. separated as if by a recent outburst of cerebrospinal fluid, and deeply-staining rounded and fragmenting nuclei are seen. In the second preparation, the deeply-staining nuclei are all small Fic. 8. Sagittal section through the end ot the spinal cord fourteen hours after the operation. This shows the epidermal layer, the plasma layer, and the knob- like extension of the central canal, caused by the outward pressure of the cerebro- spinal fluid. ep, epidermis; cc, central canal; pl, plasma layer; ntéc, notocord. -(330 diameters.) and fragmentary. In other words no more nuclei seem to be starting to degenerate. At fourteen hours, plasma and Soins cover the spinal cord though the plasma is pushed outward by the cerebrospinal fluid (Fig. 8). There are nuclear fragments in the cord and de- generating nuclei in the plasma. Another preparation of the same period shows the nerve cord still open to the exterior, as well as the nuclear appearance of an earlier stage. At twenty-four hours, only a few of the nuclei are slightly smaller and darker than the normal. At this time there appear near the end of the spinal cord, granular leucocytes containing pigment granules and fragments which closely resemble the fragments of degenerating nuclei. It may be that the leucocytes appear at this time and dispose of nuclear fragments. After one day, the degenerating nuclei are too rare to be significant. The degenerative process which the foregoing facts seem to show, may be indicated diagrammatically as follows: Cells directly cut —> broken nuclei — rounded nuclei —> fragments —> disposed of by outbreak of cerebro-spinal fluid, or by leucocytes. Cells just in front of those cut —> angular nuclei —> rounded nuclei—> fragments — > disposed of by leucocytes. NUCLEAR CHANGES IN RANA CLAMITANS. 129 2. Enlargement of Nuclet. A few preparations of the spinal cord soon after the operation show plainly that the nuclei near the end, but just in front of the deeply-staining nuclei, are larger than those of the normal cord. The long axes of nuclei close to the edge were measured and com- pared to nuclei of the same preparation which are some distance forward in the old tissue (Table I.). Immediately after the TABLE I. - a Nuclear Length Nuclear Length Difference in Time of Regeneration. Close to Edge. in Front of Edge. Length. INonmall es feriensisra wae cies 7.9 7.5 4 Immiediatelyyo. 5-2. 22 fase 10.5 7.8 Def TMEV Ulca ss siteises ey ave ni eoetioletecavayemere II.3 9.1 2.2 QPMOULS EEA ome 8.6 Toe 1.3 SG ROOULS Oia cas iota shaveroucidi suckers I2.8 10.5 Does QaSvWOuUnsiecs jas aries eesve wee ances 8.1 8.0 .I PAV OUTS tarties ks ear eee 7.6 8.3 —.7 GEA OR Ge SOO Bones 8.2 6.6 r.6 DECAY Sheurerie ee spice header caus a) cues Sccbe 8.1 8.0 .I UGLADY Sie eta estate ers el seaatnrs 8.2 8.9 —.7 AWAY Shaver sie tetka ei siaeeed wb eine 8.4 8.2 B (Sy GENYES i, Sattler loners mien Re 10.7 10.4 3 Explanation.—Each measurement recorded here is the average of the measure- ments of 9 or 10 nuclei. These were recorded in terms of the spaces of the ocular micrometer, but since one space was equal to approximately one micron (.955), the measurements were not transposed. operation and in the very early regeneration stages, the nuclei near the end are larger, but the difference decreases until after nine and a half hours it is hardly significant. This enlargement might be preparatory to normal division or it might be a swelling which is a degenerative change preliminary to fragmentation Since this size difference is greatest at the very beginning and decreases during the first day until it is no longer significant, and since mitotic divisions are not seen in numbers until the third day, the enlargement is probably an early stage in nuclear degeneration. 3. Temporary Partial Closing of the Spinal Cord. After the degenerative process is complete and the deeply- staining nuclei have disappeared, the end of the nerve cord starts to close over. By the first day, the nuclei in the end of the cord 130 GEORGE FRED SUTHERLAND. have begun to pull apart, stretching out the connecting cyto- plasm (Fig. 11). In general they extend toward the opposite Fic. 9. Sagittal section close to the edge of the central canal, showing a row of cells, not quite at the end, extending across the central canal. Other sections of the series show that the end of the cord is still open. cc, central canal; rbc, red blood corpuscles; Jc, leucocyte. (920 diameters.) wall of the central canal, thus narrowing the opening at the end. Some sections show pseudopod-like cytoplasmic extensions of the cells into the central canal as if closing were to be produced Fic. 10. Sagittal section through the new spinal cord six days after the opera- tion. bv, blood vessel; mit, mitotic figures; tc, notocord; ct, connective tissue. (330 diameters. ) > -shep 6 gE vy S © ten it i ov Zz I “+ -sfkep 8 &¢ 29 8c 6 V £e vy 4¢ & %% z & Teer gg 61 £61 Zz *+-+skep 9 £j 1 9g $ £ & g & poe, 8 £z SI Zz *-shep V Tg 39 to) 61 it *+-+shep € p08 $ ¢ @ | 8 | i posse 8 8 £ £ (0) £1 (See 8 2 AO) it o t t I : z i : g 0) ue fa hoe 29syAb) v1 2. : i g 8 ? (0) ue z siq oe ey & & E 41 (0) we Zz “say SS lo) £ £ £ £ £ £ £ i £ = se é see ae ee ee al yee cee boy ee cel aes a © ee a i T i Or ad e psweiau| ss £ £ () e € |-+ jewi0N H 9 H iS} 1S) = Re Se >xX a i @o| “PLOD | “PIOD | *STENPIA) -smry, won 3 I *amssly, MON, ‘ansst], PIO 8 B 3 53/ MON Ur) PIO UL) -IPUl | -erouesoxy B 5 % q ea 8 5 B B SISO} | S9SOITP | JO “ON, ‘T] F1avL 134 GEORGE FRED SUTHERLAND. 10 (six days’ regeneration) shows the cells in one part of the cord stretched out to such an extent that vacuoles are left between the 50 40 30 20 10 12345678 91011 1213 1415 16 Fic. 12. Curve giving the number of mitotic divisions in that part of the spinal cord within 3 mm. of the edge. Beyond 3 mm. the mitoses are scattered. An abscissa represents the period of regeneration and the corresponding ordinate gives the average number of mitoses found in the individuals killed at the end of that period. cells. It is during the period from four to sixteen days that most of the increase in length takes place, by active proliferation and migration of cells. IV. DISCUSSION. 1. Amitosts and Fragmentation. Fraisse in describing the regulative process at about two days after the operation, says; ‘‘Bereits friiher machte ich darauf aufmerksam, dass am Wundrande eine starke Auswanderung von Leukocyten stattfindet, und dass diese es sind, welchen vor allen Dingen die Bildung des homogenen, lymphartigen Saumes, welcher zuerst die wunde bedeckt, zuzuschreiben ist. Das Riickenmark geht nun an meinen Schnitten bis dichte an diesen homogenen Saum heran, und die Elemente, welche es zusammen- setzen, lassen sich immerhin noch nach 24 Stunden auch an NUCLEAR CHANGES IN RANA CLAMITANS. 135 diesem Saum von einander trennen, dann aber tritt eine bedeu- tende Wucherung von Kernen auf, und zwar scheint dieselbe auszugehen von den sogenannten Kérnern,! deren Inhalt véllig homogen und stark lichtbrechend erscheint. Durch Picrocarmin werden diese Elemente ebenfalls stark tingirt, und nun sieht man an diesen nahezu gleich grossen Kérnern Kerntheilungen, ohne dass jemals eine Spur von karyokinetischen Figuren con- statirt werden konnte, in der Weise auftreten, das der Kern oder die Korner sich in der bekannten Weise schuhsohlenformig einschntiren, und dass dann aus beiden Halften Elemente gleicher Art hervorgehen. Nicht nur eine einmalige Einschniirung glaube ich beobachten zu kénnen, sondern auch eine mehrfache, so dass der Kern sich bei diesem Process nicht nur in zwei, sondern auch in mehrere Stiicke theilen kann.” Fraisse discusses further the evidence that the nuclei from the end of the spinal cord, which are found in the lymph-like border, divide amitotically. This agrees with the present observations. But he is satisfied to show that direct division does take place. So far as my preparations show, there are few evidences that the nuclei which divide amitotically afterward become normal nuclei. In some of the preparations of stages at which the deeply-staining nuclei have almost disappeared, there are a few nuclei which stain only slightly darker than the normal ones, and at this time there are no stages between these and the fragments. These few slightly darkened nuclei may, then, be forming normal nuclei again. All other evidence points towards the conclusion that at successive stages, these deeply-staining nuclei become smaller and smaller as if fragmentation or repeated direct division, is taking place. The conclusion from these facts is that nuclei which have only started to degenerate may perhaps return to the normal condition, but that nuclei that have gone so far as to divide amitotically are destined to fragment. 2. The Appearance of Leucocytes. Barfurth (1891), working on the regenerating spinal cord of the frog larva at forty-six hours and at three days, makes the 1 K6rnern-nuclei of the gray substance, which are not present in the distal region of the spinal cord. I 26 GEORGE FRED SUTHERLAND. following statement: ‘‘ Die unterste Theil des regenerirten Medul- larrohres beherbergt in seinem Innern und zwischen seinen Epithelzellen zahlreiche fettig degenerirende Leukocyten; viele kleine und grosse Fetttropfen, die man hier iiberall findet, fiihre ich ihrem Ursprunge nach auf solche zerfallene Wanderzellen zurtick. Ausserdem finden sich hier auch viele Pigmentkérnchen» die wohl bei der regressiven Metamorphose der zerfallenden Leukocyten entstehen (Pigmententartung).” Barfurth figures the spinal cord of a larva of Triton cristatus after the sixth day of regeneration, in which these leucocytes and fat drops are shown. His figure is very similar to Fig. 11, which shows a section of a tadpole killed twenty-four hours after the operation. Both Fraisse and Barfurth mention particularly the presence of leucocytes in the early regeneration stages, but in the present study, leucocytes were not found in large numbers. Up to the end of the first day, none at all were seen close to the spinal cord. The earliest stage mentioned by Barfurth is that after a forty-six hour regeneration period, and this probably accounts for the different interpretation he gives of the origin of the “Fetttropfen’”’ or fragments. If these fragments are followed back into earlier stages in my sections, they become larger and larger and are seen to be identical with the degenerating nuclei. To be sure, the leucocytes when they first appear in the spinal cord region contain what might be called fat drops, but is it not more reasonable to suppose that the leucocytes which are present at this time dispose of the fragments of injured spinal cord nuclei? 3. Temporary Closing of the Spinal Cord. Barfurth describes the closing of the spinal cord at three days by means of cytoplasmic extensions of the cells, such as were seen in the preparations used in the present study. ‘Der sich wieder ansammelnde Liquor cerebrospinalis driickt nun auf die neugebildeten, noch wenig resistenten untern und seitlichen Theile des Rohres, und treibt sie kolbenartig auseinander. Die Zellen passen sich einstweilen durch ihre Lagerung diesem Druck an und behalten spater diese Lage noch eine Zeit lang bei.’’ Bar- furth mentions this as a temporary closure of the spinal cord, so NUCLEAR CHANGES IN RANA CLAMITANS. 137 his later preparations evidently show the cord again open. The regenerated spinal cord at sixteen days has almost reached its maximum length, but it is not yet closed. Whether or not the completely regenerated spinal cord is open at the end or closed as in the normal tail cannot be answered by the present study. 4. Rate of Division. Amitosis versus Mitosis. Durbin (1909), in analyzing the rate of increase in length throughout the regenerative process in the tail of Rana clamitans, distinguishes four periods. ‘The operation was followed by an interval of low rate, succeeded by one of rapidly increasing rate, then by one of rapidly decreasing rate and finally an interval in which the rate gradually approaches zero. The first low period is explained by a combination of two factors, (a) the shock of the injury, and (6) the formation of a cap of embryonic cells which is to serve as a basis for the more active regeneration. The second or period of rapidly increasing growth is the one in which prac- tically all the cells in the new part are undifferentiated and rapidly dividing. The third and fourth periods are explained by the appearance of differentiation, which lessens the number of dividing cells.” Fig. 12, based on the number of mitotic divisions in the spinal cord, shows these same periods. The initial period of low rate covers the first two days; that of rapidly increasing rate includes ~ the third to ninth days; the period of rapidly decreasing rate extends from the tenth to sixteenth days, and the period of gradually decreasing rate, though not covered in the present work, would undoubtedly extend on from about sixteen days. In the light of this histological study, a somewhat different inter- pretation might be given to the initial period. It is during these first two days that degeneration of the injured cells is taking place. Though at this time a cap of undifferentiated cells is being formed over the wound, the spinal cord does not participate in the formation of this cap, nor is any such cap formed at the end of the spinal cord. Since the spinal cord cells in this part of the tail are so slightly differentiated, the new cord is formed from the old without the separation of a group of special embryonic cells. 138 GEORGE FRED SUTHERLAND. The similarity of the rate curves based on a counting of the mitotic divisions with that based on the amount of tissue formed at each period, seems to be significant. It shows that the rate of tissue formation is closely correlated with the number of mitotic divisions. Considering amitosis, this may be interpreted in one of two ways—(1) either the number of amitotic divisions is similarly correlated with the rate of growth so that the total number of divisions both mitotic and amitotic, gives the same form of curve as the mitotic divisions alone, or else (2) amitotic divisions are not numerous enough to be significant. The former explanation is improbable. The nuclear conditions producing mitotic division are probably different from those producing amitotic division. Different cells in the same region may divide by different methods, but it is very improbable that the conditions producing one form of division would increase and decrease in influence at the same rate and the same times as those producing the other form. Moreover, in the present study, no examples of direct division,were seen exceptin the degenerating, fragmenting nuclei. This similarity of the rate curve of mitotic divisions to the rate curve of growth is evidence, other than the negative observational evidence, supporting the view that amitotic division is not important in the formation of this organ by regeneration. V: SUMMARY. 1. The regenerating spinal cord of the frog tadpole has been studied histologically in order to learn the mechanism, or the stages in the process, by which the new cord is formed from the old. 2. During the first day after the operation, injured nuclei in the end of the spinal cord degenerate. ‘There is first a decrease in size, by contraction or loss of achromatin, and then a frag- mentation of these degenerating nuclei. The fragments may be carried away either by the outbreaking of a cerebrospinal fluid or by leucocytes which appear at this time. These fragments are parts of disintegrated spinal cord nuclei and not of leucocytes. 3. From the second to the sixth days there is a temporary partial closing of the neural tube, probably by migration of the cells near the end. ee NUCLEAR CHANGES IN RANA CLAMITANS. 139 4. The new cord is formed by the cells of the old cord near the cut edge, by mitotic division and migration. 5. The number of mitotic divisions at different periods is proportional to the rate of regeneration at those periods as de- termined by Durbin. Amitotic division, if it occurs, is not important in the formation of the regenerated organ. 6. There is no observational evidence from this study that amitotic division does occur in normal regenerating spinal cord cells. This work was carried on under the direction of Dr. Charles Zeleny. His suggestion of the problem, and constant interest in its progress are sincerely appreciated. VI. BIBLIOGRAPHY. Barfurth, D. ’91 Zur Regeneration der Gewebe. Archiv fiir mikroskopische Anatomie, Bd. 37, Pp. 406-491. ’03 Die Erscheinungen der Regeneration bei Wirbeltierembryonen. Handbuch der vergleichenden und experimentellen Entwickelungslehre der Wirbel- tiere. Bd. 3, Teil 3. Boring, A. M. 705 Regeneration in Polychoerus caudatus. Part II. Histology. Jour. Exper. Zoél. Vol. 2, No. 3. Child, C. M. 706 Contributions toward a Theory of Regulation. I. The Significance of the Different Methods of Regulation in Turbellaria. Arch. f. Entwicklungsm. der Org., Bd. 20, p. 380. Durbin, M. L. 2709 ©=An Analysis of the Rate of Regeneration Throughout the Regenerative Process. Jour. Exper. Zoél., Vol. 7, No. 3. Fraisse, P. 785 Die Regeneration von Geweben und Organen bei den Wirbelthieren, besonders bei Amphibien und Reptilien. Kassel und Berlin, 1885. Morgan, T. H. 700 Regenerationin Planarians. Arch. f. Entwicklungsm. der Org., Bd. 10, p. 58. ’or Regeneration. New York. Morgulis, Sergius : ’10 Is Regeneration a Repetition of the Ontogenetic and Phylogenetic Proc- esses? Amer. Nat., Vol. 44, p. 92. Stevens, N. M. ?o0r Notes on Regeneration in Planaria lugubris. Arch. f. Entwicklungsm. der Org., Bd. 13, p. 396. 701 Regeneration in Tubularia mesembryanthemum. Arch. f. Entwicklungsm. der Org., Bd. 13, p. 410. ’07 Histological Study of Regeneration in Planaria simplicissima, Planaria maculata, and Planaria morgani. Arch. f. Entwicklungsm. der Org., Bd. 24, DP. 350. NOTE ON THE EFFECT OF X-RADIATION ON FERTILIZIN.! A. RICHARDS AND A. E. WOODWARD. The observations of one of the writers that x-rays would produce changes in the activity of certain enzymes suggested that these rays might perhaps be effective in bringing about changes in the action of cell extractives, particularly of fertilizin, the descriptions of which by Lillie and Glaser appeared at the beginning of the summer. The opportunity was presented to test this suggestion during the past summer at Woods Hole, since the one of us carried on studies on the effect of x-rays on some marine eggs and the other continued the work on fertilizin of Arbacia begun by Glaser. This note gives a summary of the results. It is realized by the writers that the study is by no means a complete one, but it is believed by them that publication is justified in view of the facts that the experiments give clear evidence on the main point under investigation and that there is at present no prospect of opportunity for their further work on the problem. } In taking up these experiments the writers felt that if it should be shown that x-radiation influences the activity of the cell extractive called fertilizin, that fact would be of interest from several view points: (1) without regard to the nature of fertilizin or its role in the fertilization of the egg, it is a substance derived from the eggs which has the property of being definitely modified by those external agents of which experimental use may be made; (2) in cell extractives, of which fertilizin is an example, there is a basis for the action of x-rays upon living cells, and doubtless the marked effects of the rays upon tissues is partially due to such action; (3) the modifiability of its activity by radiation is an interesting property of fertilizin; (4) this property may serve as a point in determining the relation of fertilizin to enzymes. 1 Contribution from the Marine Biological Laboratory at Woods Hole, and from the Zoology departments of the University of Texas (No. 123) and of the University of Michigan. 140 EFFECT OF X-RADIATION ON FERTILIZIN. I41I The methods used in these experiments were largely those of Lillie and Glaser. The solution used as a standard was prepared according to the method of Lillie by “adding to a certain number of ‘dry’ eggs, double their volume of sea-water, and with occa- sional slight agitation allowing ten minutes to elapse. At the end of this time the ova were precipitated by 100 revolutions of the centrifuge and the supernatant fluid, a clear, golden liquid in the case of Arbacia,’ was decanted, (Glaser, 14a). The agglutination of fresh sperm in suspension by fertilizin in both control and radiated solutions, was tested by the unit concentra- tion method of Lillie, of which he says, (130) ‘The agglutination reaction of the sperm in the presence of this substance (Arbacia fertilizin) is, as noted in previous studies, reversible, and the intensity and duration of the reaction is a factor of the concen- tration of the substance. The entire reaction is so characteristic that it was possible to arrive at a unit by noting the dilution at which the least unmistakable reaction was given. This was fixed at about a five or six-second reaction, which is counted from the time that agglutination becomes visible under a magnification of about 40 diameters until its complete reversal. The unit is so chosen that a half dilution gives no agglutination of a fresh 1 per cent sperm suspension.” Further details are given in his recent paper (’14, pp. 526-528). One can best observe the details of the reaction with the low power of the microscope. The sperm suspension is mounted under a cover glass and the drop of fertilizin added at the edge of the suspension by means of a pipette. The entire process is observed through the microscope, and the time elapsing before the complete reversal of the reaction is carefully noted by means of a stop-watch. Thus it was possible to determine the degree of activity of a given sample of fertilizin, and by comparing radiated and non-radiated solutions, to measure the effect of the radiation by x-rays. Another possible method of studying the effect is suggested by the fact that fertilizin can be used to bring about the parthen- ogenetic development of Arbacia eggs, the so-called auto-parthen- ogenesis. The efficacy of fertilizin before and after radiation in bringing about auto-parthenogenesis is a measure of the action of the radiation on it. 142 A. RICHARDS AND A. E. WOODWARD. In all the experiments with sperm it has been our policy to use only data from clear cut reactions in which the beginning and the end of the agglutination were definitely marked. Precautions were taken to see that the sperm suspension was fresh and clean. Lillie has shown that both of these factors are important, for an old suspension becomes inactive and the presence of impurities such as blood acts as an inhibitor of the reaction. Previous experience (Richards, 14) has shown the radiations to be of three kinds in relation to their effect on enzymes depending on duration, intensity and distance of the object from the x-ray tube; namely, accelerative, non-effective, and inhibitive. Under the conditions which usually prevailed in these experiments, a short exposure, of about 2 minutes, is accelerative; an exposure of about five minutes is non-effective; and one of longer duration becomes inhibitive. In view of these facts, similar exposures of fertilizin were made and the resulting activity tested as already explained. In a preliminary experiment on July 8 the following figures were obtained as the average of a number of readings of the time elapsing before the complete reversal of the agglutination reaction after short and long radiation of fertilizin. The fertilizin solution used was about 2 per cent. standard strength (in this early ex- periment the strength was not accurately determined, but it is not strictly necessary under the conditions of this test that it should be known exactly). For the control, non-radiated solu- tion the average reaction time was 32 seconds; for the 2-minute radiation the average time was 33 seconds; and for the 15-minute minute radiation it was 23% seconds. This solution was then diluted to one-half and these figures obtained : Control, 19 seconds; 2-min. radiation, 20 seconds; 15-min. radiation, 16 seconds. This experiment is incomplete and the differences lie nearly within the limits of variation, but they suggest definitely that the short radiation rendered the fertilizin slightly more active (that is, enabled it to hold the sperm in agglutination longer), and the long radiation caused it to be less active than the control. More decisive data would have been given had the dilutions been continued to unit concentration, a fact which led to the adoption of that method in subsequent experiments. EFFECT OF X-RADIATION ON FERTILIZIN. 143 In another experiment (July 14) a 1/50 dilution (2 per cent. standard) of Arbacia fertilizin was used. It was separated into four parts, of which one (Sc) was kept as a control solution, one (S2) was radiated 2 minutes, one (S5) five minutes, and the last (S7) seven and a half minutes. The results of these solutions when tested for their agglutination time at successive dilutions to unit concentration are given in the following table. %Sc means control solution diluted to one-half; 14 Sc, diluted to one- fourth, etc. The difference between two successive reaction times is marked d. Unit concentration is indicated by the asterisk(*). TABLE I. Succes-| Reac- Succes-| Reac- Succes-| Reac- Succes-| Reac- sive Di-| tion | Valuelcive Di-| tion |Value|sive Di-| tion |Valuelcive Di-| tion | Value lutions.| Time, | f 2: |lutions.| Time. | Of 2 |lutions.| Time. | Of 2: | lutions.| Time. | of 2: Se. |34 sec. S2 |37 sec. S5 /34 sec. S7 |29 sec. 2 Sc. |22 sec.| 12 4 S2 |23 sec.| 14 | 4S5 |27 sec.| 7 3% S7 |22 sec.) 7 ¢ Se. |17 sec.| 5 %S2 |15 sec.| 8 | +Ss5 |19 sec.| 8 487 |17 sec.| 5 4 Se. |r0 sec.| 7 $ S211 sec.| 4 | $S5 |12 sec.| 7 %S7 |12 sec.| 5 Te Sc. |4-5 sec.*] 5 |a6S2/ 7 sec.| 4 | asS5 |5 sec.*| 7 | isS7/8 sec*| 4 gx S2 |4 sec.*!| 3 sz $7 |0 sec. Inspection of this table shows that the activity of S2 was increased by the short radiation, for five dilutions were required to reduce it to unit concentration, whereas that state was reached in four dilutions in the other three solutions; also the full strength of this solution held the sperm in agglutination longer than did that of the control, 37 against 34 seconds. In other words, Sc was 800 units agglutinating strength, S2 was 1,600 units, S5 and 57 were each a little over 800 units, and much below 1,600 units strength. (Lillie, ’14, p. 527.) The number of dilutions required in S5 was the same as in Sc and the sperm were agglutinated the same time by both solutions. This is in line with the previous experience that a radiation of about five minutes’ duration under the conditions of these experi- ments is non-effective. However, these figures give an additional fact of possible significance which has not been entirely confirmed by other experiments either on fertilizin or on enzymes such as pepsin. If d represents the differences between the number of seconds required for the reversal of the reaction by successive dilutions, its value in S5 is practically a constant, 7; but in Sc and 144 A. RICHARDS AND A. E. WOODWARD. S2 it begins as a large number and decreases rapidly: in Sc its successive values are 12, 5, 7 and 5, while for S2 they are 14, 8, 4 and 4. In S7 the values of d are smaller and decrease more slowly, being 7,5,5 and4. Thissuggests that the laws governing the agglutination reactions by the various solutions are of dif- ferent character. But in as much as this interesting result has not been generally obtained it is not possible to attach special importance to it at this time. It is given merely as suggestive. _ The data in the case of S7 indicate that the activity of the fertilizin was decreased although the number of dilutions was the same as in the control, because the number of seconds required for the reversal of the reaction at unit concentration was much larger than is usual; yet at a further dilution no reaction was obtained. Also the undiluted solution did not hold the sperm in agglutination as long as in the control. Furthermore, it may be significant that the value of d for S7, as indicated above, are smaller than in the case of the other solutions. Subsequent experiments along the same line gave similar results. They show clearly that radiation by x-rays is capable of changing the activity of fertilizin, and in general agree with previous work that weak radiation is accelerative and strong inhibitive. Some of our experiments were performed during the latter part of the summer at the end of the breeding season and there were irregularities in the results, but it is believed that these irregularities may be attributed to the unsatisfactory condition of both sperm and eggs at this season of the year and that the statement above gives the true effect of radiation on fertilizin. Also during the latter part of the summer the writers tested the effect of x-radiation on fertilizin with regard to its power of inducing auto-parthenogenesis. Due to the near end of the breeding season these results are not entirely trustworthy, but they agree fully on one point, namely, that the radiation effects changes in the capacity of fertilizin to induce parthenogenesis. On August 10 a sperm agglutination experiment was performed which possibly throws some light on the irregularity of the auto- parthenogenesis and at the same time makes the auto-partheno- genesis test doubtfully applicable for the radiation problem. This experiment gave data showing that the radiation effects EFFECT OF X-RADIATION ON FERTILIZIN. 145 wore off when the fertilizin had stood for some time. If this is true in general it must follow that, since the fertilizin must stand in the parthenogenesis experiments, there would be irregularity in the results. The only tests of the effect of x-radiation on Asterias fertilizin were made on July 28, when the fertilizin was divided into four portions, as usual. One was kept for a control, one radiated two minutes, one five minutes, and the fourth fifteen minutes. The fertilizin was then put on mature Asterias eggs, which were allowed to stand two hours in the solution. They were then rinsed with sea-water and treated with hypertonic sea-water (50 c.c. sea water + 8 c.c. 2.5 M. NaCl) for thirty minutes, washed again with sea water, and allowed to stand for 12 hours. All four lots of eggs showed parthenogenetic development, and those treated with fertilizin which had been radiated 2 minutes had a much larger percentage of cleavages than either the control or the others. Several times Arbacia fertilizin was similarly subjected to x-rays and then tested for its auto-parthenogenetic effect on fresh Arbacia eggs. The experiments are not satisfactory, because in most cases eggs from the same females gave abnormal results when tested in other ways. The following summarizes the more interesting experiments. Percentages were obtained by counting about 200 eggs. TABLE II. Experi- Experi- Experi- Experi- Experi- ment I. ment II. ment III. | ment IV. ment V. (25) ee(B8l eal b8| oe losl ae |Se 22/43/28 /43|e8|/42/28|45|28)43 OG | s2t Vea | sul og| sun, oe!] saul or) sa Os gS O> aS Os a Os 5s Os a BS |O5 | SS OF | BE /OR | SOR | BE | OF Ol]xs ©) || es QO} +128 oO} xs © |] +2 28 os ae ge A [eae POEL ebay eae Leet See eee Sperm control 23.8 46.6 Fertilizin control (unradiated) |30.2| 0 | 9.5] 0 | 5.2] I |20.2] .5 115.5 Fertilizin 2 min. radiation.....]24.3 |few ]1I4.1| I |10.8| 3 |13.5] 0 |28.5| 2 Fertilizin 5 min. radiation....j21.6/ “‘ |17.9| 0 | 8.4] o {10.5} oO {87.2 Fertilizin 15 min. radiation...]17.5| 0 |15.6| .5 | 9.5] 0 |13.6] o {52.1 Since the effect of x-radiation on fertilizin seems to be similar to its effect on enzymes, it is of interest to note the fact that the 146 A. RICHARDS AND A. E. WOODWARD. efficiency of the agglutinin contained in fertilizin, like pepsin (Euler, p. 132) varies with the square root of the concentration. If the efficiency is measured by the number of seconds the sperm remain agglutinated, and the concentration is measured by units of strength, the curves in Figs. 1, 2, and 3 are obtained for the readings of July 14, August 10, and August II, respectively. The average is shown in the dotted line of Fig. 4. If an equation is worked out for this curve, we obtain y? = 11x where y repre- sents the efficiency and x the concentration. ‘This equation is plotted as a solid line on Fig. 4. In the higher dilutions, of which a greater number of values were averaged, and where readings could be made more accurately, the curves coincide very closely. In the less dilute portion the coincidence is not so marked, but is still within the limits of experimental error. The writers are not now able to offer an opinion as to whether or not fertilizin has the character of an enzyme. ‘The coin- cidence, however, in the behavior of this substance, when treated by x-rays, to that of true enzymes, is indeed striking. While the nature and composition of fertilizin are as yet unknown, it is a cell-extractive which is capable of undergoing changes under the action of experimental agents such as radia- tion by x-rays. Possibly it, or its forerunner, exists in the egg in combination. Among the other constituents of Arbacia eggs, this substance stands as one which, at least in solution in sea water, is able to bring about certain reactions on the part of sperm, and these reactions are subject to experi- mental modification. This justifies the inference that this sub- stance or perhaps some similar one within the egg may be capable of undergoing modification in its relations to the various intra-cellular activities. In this modification we may look for the seat of part of the changes which are brought about in living tissues and especially egg cells by radiation. The Hertwigs, Packard and others have shown that the chromatin of such cells is affected, and there is good evidence that the cytoplasm as well is influenced. Changes in their activity have also been demonstrated in the case of enzymes. These experiments add still another to the list of substances which are affected by the action of x-rays. It is = Oe ee hee eee EFFECT OF X-RADIATION ON FERTILIZIN. 147 probable that fertilizin is simply one example of a group of sub- stances which may be the object of such action (but an example which may be studied). It is to be noted that these experiments render untenable the conclusion of the Hertwigs, that chromatin is the chief and perhaps exclusive seat of the effects of radiation upon eggs. Fertilizin is a substance doubtless without mor- phological representation in the structure of the egg; yet it may suffer considerable modification from x-ray treatment. REFERENCES. Euler, H. ’r2 General Chemistry of the Enzymes. New York. Wiley. Glaser, Otto *13, On Inducing Development in the Sea Urchin (Arbacia punctulata) together with Considerations on the Initiatory Effect of Fertilization. Science, N. S., 38, p. 446. : ’14a A Qualitative Analysis of the Egg-Secretions and Extracts of Arbacia and Asterias. Biot. BuULL., XXVI. ’14b On Auto-Parthenogenesis in Arbacia and Asterias. Biot. BULL., XXVI. Lillie, F. R. ’73a Studies of Fertilization. V. The Behavior of Spermatozoa of Nereis and Arbacia with Special Reference to Egg Extractives. Jour. Exp. Zool., 14. ’13b The Mechanism of Fertilization. Science, N. S., 38, p. 524. ‘t4 Studies of Fertilization. VI. The Mechanism of Fertilization in Arbacia. Jour. Exp. Zool., 16. Richards, A. 14 The Effect of X-Rays on Certain Enzymes. Amer. Jour. Physiol., 35. “| aLvid (bbAIOL) —b aYNdId GHYVMdO0OM ONY SGYVHOIY "GJINA SAITIL NI NOLLVYINGINOD ‘Q3ANILNOD NOWWNIINTNDY SONOS JO YSAWAN NOLWIGVY NiWe TOULNOD "IHAXX “10A ‘NILA11NG IVvOINOTOIS es =ehy ue i, ALVId (0h nY) - S BUND GuyMaoomM GNy, SQYVHOIY NOILVYLNSINOD OAANINOD NOMWNILMDNY SANOI3S 40 SaaWAN : " a le ae NOIMVIOVE Nie TOYLNCO "INAXX "10A ‘NIL377NG 1VvOID0I018 GYYMGOOM GNY SGYVHOIY NOILWSLLNIINOD =z S, = jos} G3 DB gS B S 0) ase) = G> S GS 3 Ss 0S = a = < 2 8 08 eer eae i 2 Gl See ae u Ty ¢ se SITY NOILVIGVY NIW 2 TOYLNOD MW9nv) — 2 synod "WW ALW1d / “HIAXX "10A NILZTING TV9I907101I8 v1 “AL dLvId —ysundis GYVMGOOM ANY SGYVHOIY “NOHVELLNIINOD + 4 b SANNIINOD NOILWNILATIDY SONOD3S 40 YSBWAN (11) 4) NOILWNGS JO SAYNI te a SOVYIAV “IHAXX *10A ‘NILaTIpa 1v9I501019 CAN A SINGLE SPERMATOZOON INITIATE DEVELOP- MENT IN ARBACIA?! OTTO GLASER. During the summer of 1913 while making the camera lucida tracings on which I have based my comparisons between the volumes of the unfertilized and fertilized ova of Arbacia,? it became necessary, in order to prevent rotation on the part of the eggs, and the consequent necessity of readjusting the focus, to employ very attenuated suspensions of sperm. The result of the highest dilutions used in these experiments, however, gave an unforeseen result since the appearance of the fertilization mem- branes was either very much delayed, or failed entirely to take place. This observation suggested the idea of a mass effect of the spermatozoa, and the possibility that this might play a rdéle in normal fertilization. At that time I had already made observations which had con- vinced me that the fertilization membrane in this egg is not formed de novo, but is preformed in the unfertilized egg, and simply rendered visible by changes occurring at the time of impregnation.’ The mechanism through which the fertilization membrane becomes visible will be dealt with in detail at another time; for the present it is sufficient to say that the absorption of water plays an important réle. It occurred to me therefore that the prevention of this absorption and perhaps the prevention of fertilization itself might be possible even with the employment of more concentrated suspensions of sperm, if the eggs were first treated with Ca. As a matter of fact, it was either difficult or impossible to fertilize eggs so treated. The spermatozoa were active enough, but failed to enter, and fertilization membranes did not appear ‘The following protocol is typical: In a small watch 1from the Marine Biological Laboratory at Woods Hole, and the Zodlogical Laboratory of the University of Michigan. 2“*The Change in Volume of Arbacia and Asterzas Eggs at Fertilization,’ Bio- LOGICAL BULLETIN, Vol. XXVI, pp. 84-01. 3“*On Inducing Development in the Sea-Urchin (Arbacia punctulata), together with Considerations on the Initiatory Effect of Fertilization,’ Science, Vol. XXXVIII., pp. 446-450. 149 I50 OTTO GLASER. crystal, 4 volumes of fairly dense egg-suspension in sea-water-+2 volumes n CaCls. After two minutes washed in sea-water. 12.26 insemination moderate. 12.28 o fertilization membrane. UDO) a 12.24. i bb bc Control normal. All eggs with fertilization membranes in 3-5 minutes after insemination. 100 per cent. cleavage. Hundreds of eggs examined in both control and experiment. In Ca-eggs very few divisions. In connection with these experiments I noticed that insemi- nation with great excesses of sperm frequently led to results at variance with the above, for fertilization membranes appeared about the majority of the eggs despite the use of Ca, and these eggs developed. This experience strengthened my belief, not only in the validity of the Ca-experiments, but also in the cor- rectness of the original idea, namely that the number of spermato- zoa that come into contact with the egg may make a difference. Encouraged by this result, I diluted a sperm-suspension until only the faintest trace of opalescence remained. Several drops of this attenuated fluid were then drawn up into a medicine dropper of medium size and expelled quantitatively. If the dropper, which of course remained infected with sperm, was then used to agitate eggs in a small quantity of sea-water by carefully drawing the water in and expelling it several times, it was found that very soon a few spermatozoa had attached themselves to every egg. In an optical diameter, 4 to 5 sperm could easily be distinguished, but I awaited further changes in vain, despite the fact that the spermatozoa seemed to have reached the eggs, exhibited the usual amount of activity, and were potent in 100 per cent. of the cases when applied in larger quantities to eggs of the same lot. The following experiment is illustrative: 12.17 insemination with infected pipette. 12.18 o fertilization membranes. 12.19 O ty i 12.20 Toei 222 2 -3C WW NN DEVELOPMENT IN ARBACIA. I51 Control normal; all eggs with fertilization membranes in 3 to 5 minutes. Experimental eggs examined at irregular intervals throughout the day, but no increase in the number of membranes. Whether the appearance of a fertilization membrane, and im- pregnation itself will fail to take place in other eggs under similar conditions cannot be predicted, and is perhaps even improbable. With the eggs of Arbacia punctulata however I repeated these tests so often that I cannot doubt the correctness of my obser- vations, and I therefore fail to understand Kite’s! claim that he succeeded in calling forth a fertilizatiot membrane in this egg by means of a single spermatozoén. I imagine that his method involved factors whose importance was unsuspected, since he says: ‘The real difficulty with this type of experiment is not the size of the spermatozo6n, but the fact that when four or five are injected into the egg-jelly, they usually swim out and away from the egg. This necessitates the making of many injections in order to get a single spermatozo6n to attach itself to the vitelline membrane and start the reaction.’ The ‘‘making of many injections”’ very likely involves touching the vitelline membrane ’ an equal number of times, which recalls an experiment men- tioned in my earlier paper? in which fertilization membranes were induced by surrounding the eggs with large numbers of minute infusoria. Observation indicated a continuous bombardment of the ova. A quantitative relation between the rate of appearance of the -membrane and the agencies, spermatozoa, normally calling it forth is really no more surprising than the efficacy of Ca as an inhibitor. Since now sea-water of sufficient hypotonicity will of itself call forth membranes’ one may expect the exact reverse of the Ca-experiments if one immerses the eggs briefly in hypotonic solutions. Such ova, if not submerged too long so that the 1G. L. Kite, ““The Nature of the Fertilization Membrane of the Egg of the Sea Urchin (Arbacia punctulaia),’’ Science, Vol. XXXVI., pp. 562-564. 2 Science, loc. cit. 3 In my preliminary communication (Science, loc. ctt.) I considered the method of “inducing”’ a fertilization membrane in Arbacia by means of hypotonic sea-water new. Schiicking however described this procedure in the year 1903. (Arch. f. d. ges. Physiol., Vol. 97, p. 85.) The same method was used on Arbacia eggs by McClendon in 1910. (American Journ. Physiol., p. 246.) 152 OTTO GLASER. appearance of the membrane would have to be attributed to the hypotonic treatment itself, should be capable of fertilization by means of the sperm-infected medicine dropper. Actually under these circumstances fertilization with only 4 to 5 spermatozoa visible in the optical equator is possible in a considerable number of eggs. PROTOCOL. In a watch crystal 3 volumes of sea-water+3 volumes of dis- tilled. Added 1 volume of an egg-suspension in normal sea- water. At the instant’ when the first indications of membrane ‘initiation’? were noticeable added 3 volumes of “double sea- water,’’ 7. e., sea-water whose volume had been reduced one-half by boiling. By means of a sperm-infected pipette every egg was provided with 4 to 5 spermatozoa. In a series of microscopic fields the number of undivided eggs was later compared with the number that had divided. The results were: Experiment I. Experiment II. Undivided. Divided. Undivided Divided. 7 2 4. I 4 3 3 2 4 4 12 2, 7 2 18 (0) I 2 6 5 6 (0) 6 to) 2 4 7 4 5 I 8 (6) 5 4 8 5 5 2 14 4 3 2 T4 3 § 3 12 3 5 3 13 Ir 6 5 8 2 2 6 6 3 4 3 7 I 13 2 6 I 6 3 ako) I | 8 iD Motalen 72 46 189 45 Per cent.61 39 81 I9 Controls: Normal eggs-++usual amount of sperm = 100% Fertiliz- ation. Eggs treated as above-++usual amount of sperm=100% Fertilization. This is in sharp contrast with the earlier experi- ments in which the operations were carried out at the same DEVELOPMENT IN ARBACIA. 153 dilutions but without the brief fore-treatment with hypotonic sea-water. Results which harmonize with these but prove less satisfactory on account of injuries to the eggs can be gotten by the use of heat. In this case one might think of a parthenogenetic effect, but in Arbacia at least, it is not easy to confuse the usual parthenogenetic cleavage with normal two or four-cell stages. It is very easy to misunderstand these experiments and to draw wrong conclusions» There is no more doubt in Arbacia punctulata than in any other form that a single spermatozoon is sufficient to carry out the biparental effect. Furthermore the experiments with dilute sperm do not in anyway enable us to prejudge what would happen in another egg under similar con- ditions nor do they warrant the inference that the initiation of development by a single sperm is impossible in Arbacia ova deprived of their superficial coverings. I feel very sure of this however: In Arbacia the appearance of the fertilization membrane after insemination is a sign that the egg investments have allowed the sperm to pass through. This passage has been possible because the coverings have changed. The change depends on a synchronous softening and absorption of water, the latter having consequences as the result of which the membrane becomes vis- ible. Inasmuch as the becoming visible of the membrane is a reliable index of fertilization, and one of the consequences of fertilization is the division of the ovum, we may say that the initiation of development by a single spermatozo6n in this case is impossible because a single sperm cannot effect those changes in the egg-coverings which will permit it to reach the protoplasmic surface film that lies beneath. The situation is exactly as though the entrance to a room were blocked by a barrier which a single man could not break down, although a group of ten might. Once broken down, any one of the men could cross the threshold, but for the opportunity of doing this, the services of the others would be needed. With this analogy in mind, the statement that a single spermatozoon cannot except possibly under special conditions, fertilize the normally invested egg of Arbacia punctu- lata, would appear to agree with the facts. ZOOLOGICAL LABORATORY, UNIVERSITY OF MICHIGAN, November 12, 1914. STUDIES ON THE PHYSIOLOGY OF REPRODUCTION IN THE DOMESTIC FOWL. XII. On AN ABNORMALITY OF THE OVIDUCT AND ITs EFFECT UPON REPRODUCTION.! MAYNIE R. CURTIS. In a recent paper Pearl and Curtis (1914) have shown that when the passage of an egg through the oviduct is prevented by surgical interference with the duct the sex organs pass through their normal reproductive cycles. The oviduct functions to the level where the passage is interrupted and the egg is then.returned into the body cavity. The eggs thus set free may be absorbed without . causing any serious disturbance in metabolism. In a paper still in press (Curtis and Pearl) it has also been shown that congenital or acquired obstructions to the oviduct may occur without arti- ficial interference and that the results in such cases are the same as in the former cases. The following case was recently brought to our attention by Mr. J. C. Hawkes, Poultryman at the Maine Agricultural College poultry plant. A year and a half old Rhode Island Red bird was killed for meat. She was well grown, in good flesh and in every respect- was perfectly normal in appearance. When an incision was made to remove the viscera a full sized membrane shelled egg slipped into the opening. Mr. Hawkes then kindly turned the bird over to us for examination. The eggs and egg membranes shown in Fig. I were all removed from the body cavity of this bird. These represented every possible stage of absorption of the egg from a normal membrane shelled fresh egg to the collapsed empty membranes shown in the fourth line of the figure. Some of the eggs and some of the empty membranes were free in the body cavity. Some were 1 Papers from the Biological Laboratory of the Maine Agricultural Experiment Station No. 76. 154 PHYSIOLOGY OF REPRODUCTION IN DOMESTIC FOWL. 155 partly or entirely enclosed by peritoneum. In several instances two eggs or an egg and a bunch of membranes were walled off together. The last line of the figure shows collections of empty membranes enclosed in peritoneum. These peritoneal covered masses were attached by suspending stringslor folds of peritoneum. The large mass at the right end of this line contains a very large number of these empty membranes. A larger view of it is shown in Fig. 2. The second line from: the bottom of Fig. 1 shows collapsed empty egg membranes of which some are single and some two or three tightly packed together. The three top lines of the figure show eggs in various stages of resorption. One was a normal fresh egg in a single egg membrane. Ten had evidently been normal eggs but at the time of autopsy they con- tained a homogeneous mixture of yolk and albumen which had lost the gelatinous character of fresh egg albumen. Each of these eggs was enclosed in a single egg membrane. The other four eggs were double eggs. These eggs were much like the double eggs (ovum in ovo) described by Parker (1906), Patterson (1911) and by many other writers. (The appended bibliography is ‘supplementary to the one given by Parker 1906.) The eggs of this sort described in the literature had all been laid. Most of them have had shell on one or both of the con- centric components. The double eggs found in the body cavity of this Rhode Island Red hen had no shell on either the enclosed or enclosing egg. The nature of the contents of the double eggs differed in each of the four cases. In one both enclosed and -enclosing egg contained yolk. The yolk and albumen of the enclosing egg were somewhat mixed, although they did not yet constitute a homogeneous fluid. In fact the currents or streams of yolk could be seen in the clear albumen through the semitrans- parent egg membrane. The yolk and albumen of the enclosed egg were still more distinct although the-yolk membrane had already ruptured. The enclosed egg was about the size of the normal egg and the enclosing egg (the third egg in the top line) was the largest egg found in the body cavity. A second double egg was composed of a normal sized enclosed egg which had apparently contained the normal egg parts. The contents had, however, been reduced to a homogeneous brownish-yellow liquid 156 MAYNIE R. CURTIS. much thinner than fresh egg albumen. The enclosing egg was only slightly larger than the egg it enclosed and it seemed probable that a second egg membrane had been received directly around the first on its passage back up the duct. A third of the double eggs had two closely applied egg membranes as in the preceding case but the enclosed egg was itself a double egg. The inner egg in this series was a small “‘witch”’ or “‘cock”’ egg con- taining a little yolk not enclosed in yolk membrane and a small amount of normal fresh albumen. The outer egg contained only normal fresh albumen. The other double egg was even more remarkable in character as it consisted of a concentric series of four enclosed eggs. The inner one, like the inner egg just described, contained a little free yolk enclosed in normal albumen. Each of the successive enclosing eggs contained only normal albumen. This whole egg was not larger than a normal hen’s CBSE These peculiar double formations indicate that an egg did not always pass up the duct in time to get out of the way of a succeeding egg. In case an egg met another yolk it might become enclosed in a double egg or it might change the direction of the incoming yolk If the yolk was ruptured and a part remained in the duct it might furnish the nucleus for a “cock” egg which might then become enclosed in a succeeding egg. Apparently the direction of peristaltic movements became at times much disturbed, as the last double egg described must have passed up and down the duct several times before it was finally extruded into the body cavity. : The visceral organs of the bird were in normal condition. There was a little slightly oily yellowish serous fluid bathing the viscera. The peritoneum was very slightly thickened but other- wise normal. The ovary was normal with a normal series of enlarging yolks and resorbing follicles. It was apparent that the bird was in the midst of a normal reproductive period and was backing membrane shelled eggs into the body cavity and re- sorbing them with great rapidity. The oviduct (Fig. 3) was perfectly normal from the funnel mouth to the posterior end of the isthmus. Here the tube abruptly ended blindly at D. There was no shell gland or vagina. ee PHYSIOLOGY OF REPRODUCTION IN DOMESTIC FOWL. 157 The oviduct ligaments were continuous to the posterior end of the body cavity. That is the tube ended in the fold of enclosing peritoneum while the fold continued to the posterior end of the body cavity. The heavy bands of smooth muscle in the ventral ligament (see E, Fig. 3) continued to the end of the body cavity —several centimeters beyond the end of the tube. The tube rounded off smoothly at the posterior end and the ligament behind did not present the slightest indication that it had ever contained any oviduct tissue. It seems probable that the duct had never extended any farther than at present. From the embryonic history of the oviduct it is evident that if the actively growing point of a duct should cease at an unusually long distance anterior to the cloaca a blind oviduct of this form might result. The development of the oviduct according to the account given by Lillie (1908) begins on the fourth day of incubation as a groove- like invagination of a strip of thickened peritoneum on the surface of the Wolffian body or embryonic kidney. The lips of this groove fuse on the fifth day so as to form a short tube open an- teriorly to the body cavity and ending blindly posteriorly. The open end of this tube becomes the ostium tube abdominale or funnel‘mouth of the oviduct. The posterior end grows backward between the strip of thickened peritoneum and the Wolffian body. It normally reaches the cloaca on the seventh day. The growing point is always a short solid wedge of cells. The duct receives its lumen a short distance anterior to this. On the twelfth day of incubation the primordium of the shell gland is distinctly visible as an expansion of the lower end of this tube. The most probable explanation of the abnormality of the oviduct found in the case described is that in early embryonic development (probably on the sixth or seventh day of incubation) the backward growth of the primordial oviduct stopped per- manently while the differentiation of the part already formed continued in the normal manner. As in other cases where the passage of the egg is prevented the sex organs passed through their normal reproductive cycles; the oviduct functioned as far as the point where the passage was interrupted; the eggs were then returned to the body cavity and resorbed. The number of eggs and empty egg membranes found 158 MAYNIE R. CURTIS. in this fowl which was apparently in a perfectly normal physical condition show that a bird possesses very great power of resorp- tion of its own proteins from the peritoneal cavity. Such resorption does not necessarily cause metabolic disturbances. BIBLIOGRAPHY. Brown, J. T. Encyclopaedia of Poultry, Vol. 1, p. 137. Brown, M. C. ’10 Freak Eggs. Poultry Digest, Vol. 1V,.No. 11, p. 5. Curtis, M. R., and Pearl, R. Studies on the Physiology of Reproduction in the Domestic Fowl. X., Further Data on Somatic and Genetic Sterility. In press. Hargitt, Chas. W. ’99 ©SSome Interesting Egg Monstrosities. Zool. Bull., Vol. II., pp. —. ’12 Double Eggs. Amer. Nat., Vol. XLVI., pp. 556-560. Henneguy, L. F. *rr Ocuf complet de poule inclus dans un autre oeuf complet. Compt. rend. Soc. de biol. Par., Vol. LXX., pp. 779. Lillie, F. R. 708 The Development of the Chick. New York. Parker, G. H. 706 Double Hen’s Eggs. Amer. Nat., Vol. XL., pp. 13-25. Patterson, J. T. *11 A Double Hen’s Egg. Amer. Nat., Vol. XLV., pp. 54-50. Pearl, R., and Curtis, M. R. Studies on the Physiology of Reproduction in the Domestic Fowl. VIII., On Some Physiological Effects of the Ligation, Section or Removal of the Oviduct. Jour. Exp. Zool., Vol. 17, pp. 395-424. Pick, E. W. ’11 Egg Abnormalities. Poultry World, Vol. 7, pp. 495. fs phate MAYNIE R. CURTIS. * ? EXPLANATION oF PLATE Fic. 1. Eggs and egg membranes removed from the Island Red fowl. ie BIOLOGICAL BULLETIN, VOL. XXVIII. PLATE |. MAYNIE R. CURTIS. IKE, Is 162 MAYNIE R. CURTIS. EXPLANATION OF PLATE II. Fic. 2. Natural size photograph of the large peritoneal covered mass of egg membranes shown at the lower right hand corner of Fig. 1. This is cut across and opened back to show its composition. Fic. 3. Photograph (greatly reduced) showing the oviduct of the bird from which the eggs in Fig. 1 were taken. A =funnel; B=albumen secreting region; X =isthmus ring; C =isthmus; D=blind end of the oviduct; H=mass of smooth ; muscle in ventral ligament posterior to the end of the oviduct. PLATE Il. BIOLOGICAL BULLETIN, VOL. XXVIII. IRHEGs Do 3- Eines MAYNIE Re CURTIS. woons HOLE, ‘MASS. eS e eyo Vol. XX VIII. April, 1915. No. 4. SAIOLOCICAL BULLETIN — 4 yea yan ee ® \Wea oo 2.28 P.M. 96.5% 68% 61.5% 18.5% 16.3% 235 U/TO% » 2.30 P.M. 46.5% 18% 8.5% a Ye, A. 1/64%.. 2).22) PN. 0.5% o% 0% © % 5. 1/256%. 2.34 P.M. fo) o% 0% 6. 1/1024% 2.36 P.M. to) 0% 7. 1/40906%..| 2.39 P.M. not fertilize at all. It should of course be noted that the sperm suspensions used were diluted 300 times in the actual insemination (10 c.c. = 300 drops of sea-water, and one drop sperm added). Thus time is an important factor in the fertilizing power of sperm dilutions from 1/4 per cent. down. The matter cannot be stated with great accuracy, but we can say in general (referring to Table II.) that sperm suspensions from 1/300 per cent. to 1/1200 per cent. lose their fertilizing power nearly completely in 64 minutes; from 1/2400 to 1/24,000 in 32 minutes; from 1/30,000 to 1/60,000 in 16 minutes; from 1I/120,000 to 1/240,000 in 6 Table III. shows loss of fertilizing power of higher concentrations by a different method. C. We are now in a position to understand the principal reason why the curves of successive half or quarter dilutions of a I per cent. sperm suspension run off so rapidly. The reason is that the preparation of the series requires time, 10 to 20 or more minutes. In the early experiments of this kind the significance of brief periods of time was not recognized, and so no time records were kept; but I have 13 curves with accurate time records. Of these I reproduce only two (Fig. 4). For this experiment (Sept. 5), (1) 6.6 c.c. of I per cent. sperm was prepared, 9.30 A.M. (2) 4 c.c. of I was transferred to a Syracuse watch crystal and 4 c.c. of sea-water added (= 1% per cent.). (3) 4 c.c. of 2 was transferred to crystal 3 and 4 c.c. sea-water added (= 4 per cent.) and this was continued to 12 numbers (finished at 9.39.30 A.M.). The suspensions 2-12 was then divided in two equal amounts of 2 c.c. each, making series A and B.. To each crystal of series A 2 drops of a I per cent. egg-suspension were minutes. STUDIES OF FERTILIZATION. 243 added at 9.47, that is, 17 minutes after the 1 per cent. sperm was prepared. Twenty-eight minutes later (10.15 A.M.) two drops of the same egg suspension were added to series B. The only difference between series A and B is the time factor. The per- centages of cleavages were counted for both series, and the plotted results given the curves. Considerable loss of fertilizing power On Tiy O 2h Use ye Aly Shiny Ok, Zor nO ON nd Oly a Te Too has occurred in series B as compared with series A. Now, if we compare these time intervals with those given in Table II. we see that, in the curve of series B the last number, which is a 1/2" or 1/2048 per cent. sperm suspension, loses its fertilizing power completely in 36 minutes (7. e., from 9.39.30 to 10.15.30) which corresponds very well with the rate of loss of fertilizing power in a 1/2400 per cent. sperm suspension. This agreement is rather closer than usual; in some cases the series of 14 dilutions ran out at higher concentrations in about the same time; but in no case, I think, did they require more time. This suggests some possible stimulating effect of the successive changes which causes the spermatozoa to lose their fertilizing power more rapidly than under the time factor alone. Gemmill (1900) observed that the duration of vitality of spermatozoa of sea-urchins and limpets. tested by their move- ments or by the fertilizing capacity, varies greatly “ according to the amount of sperm used in proportion to the volume of sea- 244 FRANK R. LILLIE. water in which it was shed.”’ ‘‘Whenasmall quantity of sperm was mixed with a large quantity of sea-water, the duration of vitality of the spermatozoa is short, but when the converse pro- portions are used, it is greatly lengthened.” ‘‘ By taking sperm from a sea-urchin and mixing it in different vessels with different quantities of sea-water, one obtains sets of spermatozoa, which will retain their vitality for a rising series of terms, e. g., 8, 12, 16, 24, 48 and 72hours. For the longest term, the proportion of sperm- atic fluid to sea-water should be not less than 1 to 10.” Gemmill was thus dealing with the same phenomenon with which we are concerned. He gives, however, no exact quantti- tative data and relatively few experiments were performed. He attributes the results to (1) greater activity of the spermatozoa, and consequent earlier exhaustion in the more dilute-suspensions and (2) to dilution of the “spermatic fluid”’ by which he sup- poses the spermatozoa to be nourished. 5. Other Factors in the Fertilizing Power of Sperm Suspensions. In the large number of experiments carried out to test the fertilizing power of sperm suspensions the general form of the curves is remarkably constant. Some, however, are quite ir- regular, and it was never possible to get exactly the same curve in the repetition of any experiment. A few of the irregularities may conceivably be due to error, as for instance the accidental presence of some toxic substance in one of the dishes of a series, though painstaking care was used to avoid such sources of error. The failure to obtain exactly the same curve in different experi- ments is no doubt also due in part to the natural variability of dif- ferent lots of eggs and sperm. In an attempt to discover the sources of variation and error, the effect of egg concentration, 7. e., the absolute quantity of eggs in a given bulk of a sperm suspension of given strength was tested. On the whole the effect of egg-concentration was found to be relatively small within so wide a range that it cannot be regarded as a large factor in the variability of the curves; because the egg-concentration of the curves was always below the point where it was demonstrably a limiting factor. Tests were made of sperm suspensions ranging from 1/62.5 per cent. to 1/8,000 STUDIES OF FERTILIZATION. 245 per cent. But it was only from about 1/500 per cent. down that any considerable effect was observed within the range of egg concentration employed. - The method of the experiments tabulated (Table IV.) may be given for 1/500 per cent. sperm as it was the same for the others August 31: A quantity of 1/500 per cent. sperm suspension was freshly prepared 10.54.30 A.M., 2 c.c. of this was then placed in each of seven crystals (1-7). From 10.56.30 to 10.59 A.M. eggs were added as follows: to 1, one drop of a 1.75 per cent. egg-suspension, to 2 two drops, to 3 four drops, to 4 eight drops, to 5 sixteen drops, to 6 one c.c., to 7 twoc.c. The numbers in the table give the percentages of segmented eggs. The tests with 1/1,000 per cent. and 1/2,000 per cent. sperm were made with the same egg suspension. For the tests with 1/4,000 per cent. and 1/8,000 per cent. sperm a 3.3 per cent. egg-suspension was used. Thus for each series the egg-concentration is approxi- mately doubled in successive numbers of the series (in No.7 = 64 times No. I). TABLE IV. EFFECT OF EGG-CONCENTRATION ON THE FERTILIZING POWER OF SPERM SUS- PENSIONS. Sperm Egg-suspensions. Suspen- SOS it 2. Sy. 4. | Ss 6. Ts t/500% | 100 99 99.5 07 ay n93s5 82.5 56 1/1000%| 97.5 94.5 93 7Go5 1 9 ASE 76.5 83-5 1/2000% 96.5 83.5 75 WBa5 | 42.5 36 32 1/4000% 719.5 66.5 42.5 47-5 | 72.5 16 4.5 1/8000% 46.5 52 80? 30.66 Do WES 7.5 The percentages of fertilization fall off in each of these sperm suspensions with increase of egg-concentration, and the amount of falling off increases in general with the dilution of the sperm. There was certainly no numerical deficiency of spermatozoa in the highest egg-concentrations; the reason for the falling off therefore appears rather obscure, and as it is not involved in the present problem, I shall not discuss it here. But as the egg- concentration employed in any of the preceding experiments did not exceed that of column 3, and the same egg-concentration 246 FRANK R. LILLIE. was always employed throughout any experiment, it is obvious that the effect to be attributed to the egg-concentration employed in the preceding experiments is very small. III. Discussion. Within a wide limit of egg-concentration the important factors in fertilizing power of sperm suspensions are: (1) concentration, (2) time. A third factor, which is not of equal significance to the other two, is the given variability of the reproductive elements. Such variability attaches of course both to ova and spermatozoa; in general it will affect only absolute values for given combina- tions, and not at all the relative values found in any single experi- ment. Moreover, as it is a chance factor, it will tend to be eliminated in a series of determinations. Fortunately both eggs and spermatozoa of Arbacia are relatively very constant mate- rials if care be taken to wash the eggs thoroughly, and if the factors of concentration and time are fairly constant for the sperm. For the eggs these two latter factors are of such slight importance within the given limits as to be practically negligible. The significance of the concentration factor for the fertilizing power of sperm is of course obvious without discussion. We therefore turn to the time factor. . The most significant aspects of the time factor are, first, the unexpectedly rapid rate of loss of fertilizing power of sperm suspensions, and second the increase of rate of loss with dilution. There are but two ways of explaining these facts: either (1) the motility of the spermatozoa is quickly reduced in sperm suspen- sions to such an extent that they cannot bore into the egg or (2) the spermatozoa lose some substance essential for the fertilization reaction. The following are the objections to the first alternative: (a) Microscopical examination lends it no support; I have repeatedly — observed, that fertilizing power of sperm cannot be expressed either in terms of motility, or of success in penetrating the jelly of the egg and coming in contact with the membrane. In the experiments on successive half dilutions (p. 238) I kept records, in several series, of the numbers of spermatozoa in the jelly of unfertilized eggs, and found in some cases an average of 9 STUDIES OF FERTILIZATION. 247 spermatozoa visible in the jelly, or on the membrane of certain lots of eggs none of which had fertilized; this could not be more than half of the spermatozoa in association with such eggs; and other observations made immediately after insemination demon- strated the high degree of motility of spermatozoa of entirely barren sperm suspensions. These observations contrast in the most striking manner with the fact that not a single spermatozoén can be seen in the jelly of eggs fertilized with highly dilute fresh sperm suspensions, where, nevertheless, nearly every egg may be fertilized. (b) Penetration of the egg is not solely a function of motility of the spermatozoén. Penetration follows, as a matter of fact, after the fertilization reaction has begun, and it is due to the inception of such reaction, not the reverse as is commonly assumed.2 In Nereis, as I have previously described, penetration does not take place until 45 to 50 minutes after insemination and the initiation of the fertilization reaction. The facts described in this paper show that in Arbacia no penetration takes place unless the sperm has started the fertilization reaction; if this does not take place, the spermatozoén remains external, however active it may be. And if it does occur the initiating spermatozo6n is speedily engulfed by the egg. (c) It is not easily understood on this theory why dilute sperm suspensions should lose their fertilizing power more rapidly than 1 Glaser’s experiments (1915) would bear the interpretation that, in those cases of normal insemination described by him in which fertilization does not occur except in the presence of several spermatozoa for each egg, the time factor which I have just described was operative. In other words that the majority of the spermatozoa in question had lost their receptors. But in the absence of exact data as to age and concentration of the sperm suspension, it cannot be asserted that this is the correct interpietation although I obtain exactly the same results in my time series (p. 238). My dilution experiments prove beyond a doubt that a single spermatozoon suffices for the whole process of fertilization under optimum conditions (defined on p. 233). Glaser’s experiments, however, raise the question whether the efficacy of heavy insemination in the case of a stale sperm suspension is due to mass action, or to the survival of a small percentage of effective spermato- zoa? So far as I can see this question can not be answered on the basis of our present information. 2 Spermatozoa may penetrate into unripe ovocytes in some cases, as has been noted by several observers; in such a case there is no fertilization reaction. In the present experiments the unfertilized eggs were not penetrated by the spermatozoa. 248 FRANK R. LILLIE. more concentrated suspensions;! the relative freedom from CO, and other sperm excreta should favor a longer continuation of their motility in the dilute suspensions rather than the reverse. (d) Moreover, in general the results of recent fertilization studies such as the antagonistic action of sperm suspensions of different phyla, inhibition of fertilization in the presence of blood of the species, or in the absence of certain ions (Loeb, 14), or again the sterility in certain self-fertilizations, and finally the inability of spermatozoa to penetrate fertilized eggs, unite in demonstrating the relative lack of significance of motility as such. | We come therefore to the conclusion that the individual spermatozoa im suspension tend to lose their fertilizing material, so that an increasing proportion of these spermatozoa become absolutely in- effective whatever their motility. This conclusion is in agreement with all the data of the foregoing experiments, and seems to be the only one competent to explain the results. The following questions arise: (1) Whether the loss of this substance by the sperm is a mere process of diffusion or an active secretion? (2) Can the substance be recovered from the fluid of the suspension, or can its presence in the fluid be demonstrated in any way? As regards the first question: In the case of the ova we know that the external jelly-covering is loaded with sperm-agglutinat- ing substance which diffuses into the sea-water continuously. It is theoretically possible, at least, to apply a similar conception to the spermatozoén, although no such covering is demonstrable. The more rapid: loss of fertilizing power in the greater dilutions would be consistent with this interpretation. From this point of view we would have to regard the sperm head as covered superficially with a layer of fertilizing material, like the phos- phorus on a match. Such a conception is by no means im- possible. On the other hand the fact that dilutions reached by a series of successive half-dilutions from I per cent. lose their 1Gemmill (1900) observed the same phenomenon and concluded that the more rapid exhaustion of spermatozoa in dilute suspensions is due to dilution of a hypothetical nutritive medium which keeps the spermatozoa of concentrated sus- pensions in a vigorous condition. This explanation comes back to the principle of loss of motility, so far as it relates to fertilizing power. STUDIES OF FERTILIZATION. 249 fertilizing power more rapidly than the same dilutions made in one stroke, indicates that successive stimulation hastens the loss, which therefore appears more in the nature of a secretion or a discharge than mere diffusion. The source of the substance must ultimately be the sperm cell itself, and it is quite possible that, as in the case of the egg, there is both a superficial layer and an internal supply. It must be admitted that the data are inadequate to answer this problem. The statement of the problem can therefore serve only to bring out the resemblance between the spermatozo6n and the ovum in respect to the existence of a fertilizing substance in each, the fertilizin in the case of the ovum and the sperm receptors in the case of the spermatozo6n, and also the possible resemblance in respect to the disposition of the substances in each. It certainly is an interesting parallelism that both cells contain a substance necessary to fertilization, which may be lost in the sea-water. The most interesting and crucial question of course concerns the possibility of detecting this lost substance in the fluids of the suspensions. If such a substance actually occurs in the fluid it should have the property of fertilizing ova; unless it can be detected by this property, we have no other indicator for it. So far I have not been able to make even a beginning on this problem. As is well known a number of experimenters have attempted without success to derive a fertilizing medium from spermatozoa. It has been suggested by Loeb that the reason for the failure to secure an extract of spermatozoa that will fertilize is that the motile power of the spermatozo6n is needed to carry the effective substance into the egg. But it may equally well be that the methods hitherto employed have been too brutal; the substance may well be too labile to withstand extraction by ether, etc. My results strongly suggest, if they do not prove, that such a substance must be present in the fluid of sperm suspensions of Arbacia, and they therefore suggest other methods for securing it for testing. We must bear in mind that it can form only an extremely small proportion of the entire spermatozo6n, as proved by morphological considerations alone, and that it must 250 FRANK R. LILLIE. be superficial in position and easily detached as proved by its effectiveness before the spermatozo6n penetrates. Extracts of the entire spermatozo6én must contain numerous other substances which may neutralize its effectiveness. The difficulty of the investigation as shown by my experiments is that it is liberated only very slowly in concentrated suspensions and that its amount in dilute suspensions would presumably be too slight to be effective. Some means can probably be devised for liberating it in concentrated sperm suspensions and freeing it of the spermatozoa for testing. Finally I may point out that the conclusion that spermatozoa lose a substance necessary for the exercise of their fertilizing power is consistent with my own point of view of the mechanism of fertilization as well as with Loeb’s. From my point of view the spermatozoén loses its receptors, viz., the substance that activates the fertilizin of the egg; from Loeb’s point of view the spermatozoon loses its lysin, the substance that corrodes (cy- tolyzes) the egg. My previous experiments had shown that eggs produce a certain substance in sea-water (fertilizin) which is necessary for their fertilization; fertilized eggs no longer produce this substance and are incapable of fertilization. Both eggs and spermatozoa therefore contain substances, more or less liable to loss, which are necessary for fertilization. The mechanism of fertilization cannot possibly, therefore, be regarded in the simple manner postulated by Loeb’s theory. The existence of parthenogenesis demonstrates the efficacy under given condi- tions of the egg-substance alone; we must therefore regard the spermatic substance essentially as an activator of the fertilizin of the egg. LITERATURE. Gemmill, James F. 700 On the Vitality of the Ova and Spermatozoa of Certain Animals. Journ. of Anat. and Physiol., Vol. 34 (N. S., Vol. 14), pp. 163-181. Glaser, Otto. 713. On Inducing Development in the Sea-urchin, together with Considerations. on the Initiatory Effect of Fertilization. Science, N. S., Vol. XX XVIII, Pp. 446-450. 14 The Changein Volume of Arbaciaand Asterias Eggs at Fertilization. BioL. BULL., Vol. XXVI., pp. 84-01. OE a STUDIES OF FERTILIZATION. 251 "75 Cana Single Spermatozo6n Initiate Development in Arbacia? Bio. BULL., Vol. 28, pp. 148-152. Lillie, Frank, R. 713) The Mechanism of Fertilization. Science, N.S., Vol. XX XVIII, pp. 524—- 528. ’r4 Studies of Fertilization VI. The Mechanism of Fertilization in Arbacia. Journ. Exp. Zoél., Vol. 16, pp. 523-590. Loeb, Jacques. ’r4 On Some Non-specific Factors for the Entrance of the Spermatozo6n into the Egg. Science, N. S., Vol. XL., pp. 316-318. ex ae Th acs oe ay) oe > : : \ Ste ‘ es Fy See 4 % ban SA dries ee Ww hy ae EY ot ead ge: j Oe the 5 Rivitnitcat Susceptibility ee ES ies 0 ime ee e Rae Deweloping Sea. Urchin ae to Be - : me Ay oe pees eas c & PARES, Sea’ Water. ss ne vend Nene eo a hae ao oe ‘Starfish Bees ee ae é 5 oe cS » the Influence of fiigh, Cie a ae . tures and Fatty Acid Solutions. 2 ‘ : Division Rate in Ciliate Poco pe en ee i a, Thyroid Constituents O , Say, i pS P oy An Experimental. Study Of the eee: fae are 3 : Movements of Herring and Other He ee oe Gas < War “tne. Fishes er oee (ESS Nie aan aot) ose - Pusuisnep Monee RY THE MARINE, | BIOLOGICAL, LABORATORY ee cans Gee eae 2, “PRINTED. AND. ISSUED MaMa? THE NEW ERA PRINTING COMPANY Gon Se Shh ie ss ‘LANCASTER, PAL co UR RGR Aes SE be Pd ee acco AGENT FOR GERMANY. “R FRIEDLANDER & SOHN 2B Essex ‘Streei, Strand fee ae oe Bee NN. we a A aa “ | ae 3 Lanion, We C : cs & : oe | Carkstrasse, aa ‘Singie lies 75 Cents. Per Volume (6 numbers), $3.00 : Vol. XX VIII. May, 1915. No. 5. mOLOGCICAL BULLETIN oan Institon~ ON THE RHYTHMICAL SUSCEPTIBILITY OF DE- fae “ oN < VPLOPING SEA URCHIN EGGS TO foo i an ee 0 1915 HYPERTONIC SEA WATER. ( Mie ag ae x ry) ie 4 ARTHUR RUSSELL MOORE. tational Mus? . (From the Biological Laboratory of Bryn Mawr College and the Marine Biological Laboratory at Woods Hole.) In a recent communication M. Herlant! attempting an analysis of Loeb’s method of artificial parthenogenesis concludes that (1) the fatty acid treatment gives rise to the rhythmical activity of the centrosome but never to normal divisions of the egg; (2) the treatment of the eggs with hypertonic sea water causes the formation of accessory asters and is necessary to complete the causes for normal division; (3) the optimum results are obtained by applying the hypertonic solutions at certain intervals after fatty acid treatment, viz., 30 and 70 minutes and possibly 115-120 minutes, while with 40-50 and 95-100 minute intervals marked minima are shown. As to the first generalization, Herlant ignores the fact that in Strongylocentrotus and Arbacia the fatty acid treatment alone may cause normal segmentation. If the eggs of S. purpuratus are kept at a low temperature (5°-10°) after acid treatment alone, they divide regularly and may reach the morula stage.’ In Arbacia eggs, normal segmentation may take place after acid treatment without subsequent treatment with the hyper- tonic solution, but does not asa rule proceed beyond the two- cell stage. On the other hand hypertonic treatment alone may cause Arbacia eggs to segment and develop into swimming larve, while it brings about only early segmentation stages in the eggs of Strongylocentrotus. . 1M. Herlant, Comptes Rendus de l’ Academie, T. 158, p. 1531. 2]. Loeb, ‘‘ Artificial Parthenogenesis and Fertilization,’’ p. 76. 253 254 ARTHUR RUSSELL MOORE. Herlant apparently assumes that the action of the hypertonic solution must be subsequent to the acid treatment, that to be effective such action must occur in a certain phase of the rhyth- mical activity of the centrosome. Since in Strongylocentrotus purpuratus treatment of the eggs with the hypertonic solution may precede that with acid sea water by as much as a forty-eight hour interval with the result that normal parthenogenetic larve are formed,! Herlant’s contention is not justified. Furthermore, treatment of the eggs with small quantities of KCN or depriving them of oxygen, may replace hypertonic treatment after fatty acid. Now lack of oxygen or the repression of oxidations does not cause aster formation, but on the contrary suppresses it. In fact it has even been shown that in Arbacia the first steps in development induced by the acid treatment may be reversed and the egg returned to its resting stage with its original possi- bilities of fertilization, simply by withholding oxygen from such an egg or by treating it with KCN In view of these facts, it seems evident that Herlant’s conclusion that the hypertonic solution is a necessary factor in artificial parthenogenesis because it controls aster formation, does.not hold. As to the rhythmicity in effectiveness of the hypertonic treat- ment which Herlant found in his experiments, it seemed possible that the relation between the time spent in normal sea water after acid treatment and the time in the hypertonic solution might have a bearing upon the question. Six years ago the present writer found, in working with the eggs of Strongylocen- trotus purpuratus, that if the exposure to normal sea water in such an experiment be lengthened, the subsequent treatment by hypertonic sea water must be shortened to secure optimum results.2 Repetitions of the experiment at Woods Hole during the past summer, however, have indicated that the relation does not exist for Arbacia. Table I. shows the results of ‘dividing a lot of Arbacia eggs, after fatty acid treatment, into three parts which remained in normal sea water, 5, 25 and 90 minutes respectively, before being put into hypertonic sea water. In each case a portion 1J. Loeb, Journ. Exp. Zool., vol. 15, p. 201. 2 J. Loeb, Science, N. S., Vol. 38, p. 740. 3 J. Loeb, ‘‘ Artificial Parthenogenesis and Fertilization,” p. 96. RHYTHMICAL SUSCEPTIBILITY OF SEA URCHIN EGGS. 255 of each lot was removed from the latter solution after 17%, 20, 25, 30, 35, and 40 minutes, and allowed to develop in normal sea water. The percentages given in the table show the degree of blastula development in each culture. Repetitions of the experiment showed no significant variation in the optimum exposure to the hypertonic solution with changes in the time the eggs remained in the normal sea water after acid treatment. SABE le LO? = BOP = 2e® Eggs Remained in Percentages of Eggs Develop into Blastule After Exposure to Hypertonic Normal Sea Water, Sea Water for After Butyric ; Acid Treatment 17% Min. 20 Min. | 25 Min. 30 Min. 35 Min. 40 Min. 5 minutes...... — 1% | 5% 13% 24% 18% DIST ae = 8% | 18% 24% 20% 8% QOm me ip is fous — — | 1% 3% 1% In order to determine if the eggs of Arbacia punctulata which had been treated with acidulated sea water, showed a rhythmicity in sensitiveness to the hypertonic solution, such as Herlant’s experiments with the eggs of Paracentrotus lividus indicated, the following experiments were carried out. The eggs of several sea urchins were collected, treated with sea water made acid by the addition of 2 c.cm. N/1Io butyric acid to 50 c.cm. of sea water. After remaining in this solution for from 2 to 24% minutes the eggs were transferred to normal sea water. At the end of 5- minute intervals lots were removed to finger bowls containing hypertonic sea water [50 c.cm. sea water + 8 c.cm. 24 M (NaCl + CaCl, + KCl)]. After remaining in the hypertonic sea water for 25 minutes the eggs were put into normal sea water and allowed to develop. The percentages of advanced morule or non-swimming blastule were determined by counting random fields. The following table (Table II.) gives a typical result, showing optimum effects when the eggs were put into the hyper- tonic solution 40, 60, 90-100 and 115-125 minutes after acid treatment. The rhythmical character of the result is obvious from Curve I., where the ordinates indicate the percentage of larvee formed, while the abscisse indicate the time which elapsed between acid treatment of the eggs and their exposure to hyper- tonic sea water. ARTHUR RUSSELL MOORE. 256 soynurpy SST SVI Cer "II @AunD Sor S6 Sg SL U01}e}USUISOS P4IGL UO01}e}JUIUISOS puoses ‘] aAuaAD UO}e} UNOS 4ST RHYTHMICAL SUSCEPTIBILITY OF SEA URCHIN EGGS. 257 TABLE II. T° = 19°— 21°. Hypertonic exposure =25 minutes. Time in normal sea water after acid (IRE VOMEM, Ao bo ode 5 15 25 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 Blastule per cent.... — 5 1623 301I QI012 9 5 6 7 91316 12 5 4 Time in normal sea water after acid treatment......... II5 120 125 130 140 150 165 175 185 Blastule per cent. ... 8 1 & A O ACTIVATION OF UNFERTILIZED STARFISH EGGS. 269 At 30° the proportion of eggs forming membranes and develop- to larval stages is higher than at 29°, although considerable variability is still shown. Five series of experiments were performed at this temperature. Table II. summarizes the results of four of these.!_ Each lot of eggs was favorable, matura- tion and development to larve after sperm-fertilization taking place in nearly all. Table II. gives the approximate proportion of eggs forming membranes and developing to blastule after exposure for the periods given in the first column. TABLE II. 30°. Proportion of Eggs Forming Fertilization-membranes and Larve. Duration of Ex- June 7. June 8. June re. June 13. ‘unapnes. [os | | | ; eee | Larve. nae Larve. eee | Larve. | Membranes.) Larve. Up to | | 3 ida, (0) (0) (0) (o) O 0) Aim, ee, 596| Oo (0) (0) O | .@ Galil a |P5— 2095) 0 fo) (0) (0) (0) Ki | o 6m. |ca. 20%| 0 0) ) | “i. |ca. 30%) ° 0 | GQ, 39)| © <1% fe) Sims ia, SOI) O len, 19%) © | 5-10% (0) 9-10 m.|ca. 100% ca. 2-3 |20-25%| 0 40-50%| oO I0-15% (0) | A || | I2 m. 50-60% ca. I % 70-80% (o) 20-25% <1% I4-1I5 m.|ca. 100% ca. 5% |80-90%| 2— 3%| >90% 10-15% 17-18 m. | ca. 90%| 5-10% | 30-40% |ca. 5% 20-21 m. <50%| 5-10% | ca. 40% ca. 30% 24m. | | | 30-40% |30-40% 28 m. | ca. 40% ca. 40% 30 m. | ca. 10%| <5% | ; 34 m. | ca. 15-20%| <1% It will be noted that in four out of the five series at 30° an exposure of 8 to 10 minutes was required to cause membrane- formation in IO per cent. or more of the eggs; in the fifth series (June 7) 5 minutes was sufficient. The proportion of eggs developing to larvae was comparatively low in all series; the optimum exposure lay between 24 and 28 minutes in the only series (June 13) in which the proportion of larve was consider- able. With longer exposures membranes become fewer and there is a rapid decline in the proportion of eggs forming larva. 1Tn the remaining series the longest exposure was 10 minutes, at which about two thirds of the eggs formed membranes and a small number developed to larve. 270 RALPH 5S. LILLIE. At 31° the-conditions become more favorable and with the proper times of exposure practically all mature eggs form fertiliza- tion-membranes, and in favorable cases the great majority develop to larve. Four series of experiments were performed at this temperature; in one of these (August 28) only about half the eggs underwent maturation, and with 15 minutes’ exposure (approximately the optimum) only Io to 15 per cent. of all eggs formed larve. In the other three series the eggs were normal. The proportions of eggs forming membranes and larve in these series with the different times of exposure are given in Table III. TasBLe III. eiiee Proportion of Eggs Forming Fertilization-membranes and Larve. Duration of Exposures June 8. June 12. June 13. in Minutes. : Membranes, Larve. Membranes. Larve. Membranes, Larve. I-2 m. (0) @ (0) 24 m. Oo (0) O 3m. Co) fo) ca. 10-15% ca. 5% (0) 33 m. 30- 40% ca. 50% 0 4m. | Few (<1%) o 70-— 80% 60- 70% Co) 5m.| 10-15% ) ca. 90% >95% Oo 6m. ca. 20% (0) S90% ca. 100% fo) 8 m. 80-90% ca. 2-3%| > 90% | ca. 1% | ca. 100% ca. 1% Io m. ca. 90% 15-20% | ca. 100% | 20-30% | ca. 100% ca. 20% I2 m. ca. 90% 40-50% ca. 100% ca. 60% I4-I5 m. 70-80% 40-50% ca. 90% 80-90% I7-18 m. ca. 50% ca. 40% ca. 75-80%| 50-60%; 20-21 m. 15-20% Bt 5% 40-50% |ca. 10-15% 25-30 m. ca. 20% (0) At this temperature an exposure of 3 to 4 minutes is required to cause membrane-formation in Io per cent. or more of the eggs; exposure must be prolonged to ca. 8 minutes before any eggs form larve; 14 to 15 minutes is the approximate optimum. In the series of June 12 this optimum was not reached. At 32° a larger number of experiments were performed than at any other temperature, and their results show a decidedly greater uniformity than at lower temperatures. With the optimal times of exposure (from 7 to 8 minutes) the proportion of larve yielded by normal eggs is always high,—usually over go per cent. This is illustrated by Table IV., which summarizes the results of six successive series performed during June at a a ACTIVATION OF UNFERTILIZED STARFISH EGGS. 271 time when starfish eggs were unusually abundant and favorable. On account of the relative completeness of my observations at this temperature, the general results of these experiments will be described in some detail. ‘The exposure required for membrane-formation is about half that at 31°. An exposure of 3 minutes typically forms mem- branes in all normal eggs, and one of 2 minutes is usually sufficient to produce this effect in a minority and sometimes in a majority of eggs. From 3 minutes on the conditions remain normal for membrane-formation until the exposure is prolonged to 12 or 15 minutes, after which in a certain proportion of eggs membranes tend to separate imperfectly or even fail to form. In most series exposures longer than 15 minutes were not used, since eggs so treated never form larve; in one series, however, eggs were exposed for 27 minutes, at which exposure nearly half failed to form membranes. This decline in membrane-formation when exposures are prolonged beyond a certain maximum is general for all temperatures (cf. also the series at 30°, 31° and 33°); the fact is interesting since it indicates that the process is not a direct effect of the high temperature but constitutes an active response —probably in the nature of a secretion—on the part of the egg. The optimum exposure for inducing complete development at 32° varies between 6 and 8 minutes, and with this exposure the great majority of normal eggs cleave and develop to larval stages. Many of the gastrule and Bipinnariz thus obtained are appar- ently quite normal and swim freely at the surface of the water. The rate of development is, however, always slower than that of sperm-fertilized eggs; relative slowness of development seems in fact to be a constant peculiarity of parthenogenetically acti- vated eggs.! Exposures well above the optimum are followed by imperfect or delayed cleavage and failure to develop beyond early stages. Table IV. gives a summary of the results of the six successive series referred to above. The approximate proportions of eggs forming free-swimming larve are given; the conditions of mem- brane-formation have already been sufficiently described. 1 This has been my uniform experience since I began studies of this kind, and apparently the experienceis general. This suggests.strongly that the spermatozoon contributes to the egg material which is utilized in normal development. 22 RALPH S. LILLIE. TABLE IV. Ray?) SORE Proportion of Eggs Forming Free-swimming Larvz. eeeaaey June 12 June 13. June 18. June 24. June 25. June 26. I-3 m. 0) 0) 0 (o) Ce) Co) 4m. Ba, u% KG, | ets G=B% || a= BIA || Ga, 3% Co) & io. ca. 3- 4% 2— 3% | 15-20% | 25-35% | ca. 50% | 10-15% 6 m. ca. 35-40% | 20-30% | 55-60% | 60-70% | 80-90% | 25-35% 7 OD, 70-80% =90% ca. 60% | 50-60% 8 m. >90% ca. 95% | Ca. 957% | 90% | 25-35% | 80-90% ro m. 85-90% 50-55% | 75-85% | 50-60% | < 5% | 80-90% I2 m. 15-20% | 25-35% (0) ca. 20% I5 m. fo) <1% 18, 22, and 27 m. ) | (0) These results may safely be regarded as typical. Six other similar series were carried out at this temperature. In two of these the eggs were unfavorable or the treatment was applied too late. In the four others—two in early June and two in late August—the results were similar to the above, although fewer eggs formed larve; the optimum exposures ranged from 6 to 8 minutes, with respectively 20, 20, 40 and 50 per cent. of mature eggs forming larve. It will be noted that the optimum exposure is approximately 8 minutes in five out of the six series in Table IV. Different lots of eggs vary somewhat in the duration of this optimum; thus in the series of June 25 half of the eggs formed larve with only five minutes’ exposure and the optimum was 6 minutes, and on June 26 the eggs showed almost equally good development with the 8- and the 10-minute exposures. In the majority of series, however, there was a well-defined optimum at 7 or 8 minutes. The physiological effects following exposure to 32° vary in a constant and highly characteristic manner with the duration of the exposure. Eggs exposed for a period insufficient to induce membrane-formation show no apparent change on return to sea-water and later break down without development. Such eggs, however, can be shown to have undergone some internal change similar in kind to that following longer exposures; thus if later they are again exposed to 32° they are found to require, in order to induce favorable development, a shorter exposure ACTIVATION OF UNFERTILIZED STARFISH EGGS. Dis than previously untreated eggs (cf. below, p. 288). Exposure for 3 to 4 minutes induces typical membrane-formation in all eggs, followed, however, not by cleavage and further develop- ment but by irregular changes of form, fragmentation, and eventual breakdown. With somewhat longer exposures (4 to 5 minutes) membrane-formation is followed by symmetrical cleav- age in a certain proportion of eggs; and the proportion of such cleavages, and also their approximation to the normal in rate and character, show a progressive increase with increasing length of exposure up to the optimum of about 8 minutes. With still longer exposures the response again becomes unfavorable, and eventually the eggs entirely fail to develop and even to form membranes. We have here an apparent reversal of the rule enunciated by Loeb with reference to the action of membrane- forming agents on the sea-urchin egg. ‘‘A relatively brief exposure to a cytolytic agent leads only to membrane-forma- tion, while a longer exposure causes cytolysis.’ In the star- fish egg a relatively brief exposure to warm sea-water (one just sufficient for membrane-formation) is followed by an early cytolysis, while a longer exposure results not only in membrane- formation but in an approximately normal development; still longer exposures again cause cytolysis without development. This rule applies to the action of cytolytic substances like fatty acid, as well as to high temperatures (cf. below, p. 282). To illustrate the effects of exposures of different duration on cleavage the following record is given (Table V.) describing the condition of the eggs about four hours after exposure to 32° for the times given. It will be noted that with brief exposures (3 to 4 minutes) membrane-formation is typical, but the eggs are unable to cleave normally and undergo irregular change of form followed by breakdown. As the time of exposure increases, an increasing proportion of eggs cleave, until the optimum (6 to 7 minutes) is reached at which cleavage approaches the normal in rate and character, and the great majority develop to larval stages. Over-exposure (10 minutes) is again followed by failure of cleavage and development. Similar observations were made 1“ Artificial Parthenogenesis and Fertilization,” 1913, p. 8. 274 RALPH S. LILLIE. in experiments at other temperatures; in all cases the exposure which induced the largest proportion of regular cleavages was found to correspond with that at which the largest proportion of eggs formed larve. In the series of Table V. the optimum, 6 minutes, is somewhat shorter than usual; on June 24 the optimum of cleavage was found at 7 minutes, and on June 26 at 8 minutes, with 10 minutes somewhat less favorable. RABE Ve JUNE 25. 32°. Condition of Eggs 4 Hours after Exposure, and Proportion Time of Exposure. of Eggs forming Larve. TH SME T SS Oa e ear Mealy clay TS ete ts Great majority are unchanged; a few have membranes. No larve. Dee Tha cS leeway ee aaa Most eggs have typical membranes and are irregular or amceboid in form; a few show irregular cleavages. No larve. BMA EYL ste sa eerepeece elope Antaris Almost all eggs have membranes and exhibit irregular forms; a few have cleaved symmetrically. Ca. 5 per cent. form larvae. AISMISP OO ieueeter en EN ots Gis eecL as 0 Marked improvement over Experiment 3: most eggs have cleaved, and many are in regular 4- and 8-cell stages. Ca. 50 per cent. form larve. ReeOMIIs 5 cea ens cies ymeeen eee Almost all eggs are cleaved; cleavages are more regular and advanced than in Exp. 4; 16-cell stages are frequent. 80-90 per cent. form larve. On 7m Bae a aetna eee eae The proportion of regular cleavages is also high, but rather less than in Exp. 5. Ca. 60 per cent. form larve. iat an beaietea her tyeteneosste: Uo. aie eb Cleavages are fewer and less advanced than in Exp. 6. Ca. 25-30 per cent. form larve. SD LOM cc eee eee eee Great majority are uncleaved; many are irregular in form or fragmented. Larve are few: < 5 per cent. EMME Qe Tihs ean canieaetaaye eer Almost none have cleaved. The eggs are largely irregular or with small surface-vesicles detached. While an exposure just long enough for membrane-formation is insufficient by itself to induce normal cleavage and develop- ment, it is possible, after forming membranes in this way, to make the eggs cleave and develop to larval stages by subjecting them to a second treatment with warm sea-water, or by after- treatment with fatty acid (m/260 butyric acid in sea-water), hypertonic sea-water, or cyanide (7/1000 KCN in sea-water). These effects will later be described in detail (cf. Tables XIII., XVIL.). ‘| | ACTIVATION OF UNFERTILIZED STARFISH EGGS. 275 It is interesting to note that the effects produced by weak fatty acid solutions (7/260 butyric acid) on unfertilized starfish eggs also vary with the time of exposure in a manner closely similar to that just described. Brief exposure causes membrane-forma- tion followed by irregular change of form and breakdown without development, while longer exposure induces not only membrane- formation but cleavage and development to larval stages; still longer exposure is again unfavorable. Eggs in which membranes have been formed by the minimal exposure to fatty acid may be made to develop by the above forms of after-treatment. The effects of the two agents, warm sea-water and weak butyric acid solution, seem in fact to be identical in every essential particular, and the one may be substituted for the other without altering the effect on the egg (cf. Tables XIV. to XVII.). Experi- ments showing this parallelism will be described in detail later. There is in fact every indication that the underlying physio- logical process which enables the egg to continue normal develop- ment is of the same nature as that which induces simple mem- brane-formation, the only difference being that the duration of the process must be considerably longer in the second case than in the first. The temperature-coefficients of both effects indicate the same, as will appear below (cf. Table X.). The possible nature of this process will be discussed in the concluding section of this paper. Treatment with sea-water at 33° gives similar results to those above described, except that the times required to produce a given physiological effect are only a little more than half as long as at 32°. An exposure of from one to one and a half minutes is needed to call forth membrane-formation in the majority of eggs. Four series of experiments with normal eggs were performed at this temperature, and in every series the great majority of eggs formed larve with the optimal times of exposure. In these series the earlier transfers from the warm sea-water to normal temperature were made at half-minute intervals. The results are summarized in Table VI. In all of these series the proportion of favorably developing eggs is high with the optimum exposures of 414 to 5% minutes. The series of June I5 is unusual in that nearly all of the eggs 276 RALPH S. LILLIE. form larvee with exposures varying in length from 4 to 6 minutes. The optima seem to be less sharply defined when the eggs are in the best of condition, probably because then the power of regulatory adjustment to environmental variations is at its maximum, and slight deviations from the optima are auto- matically corrected. TABLE VI. Bon. Time of Exposure. 1 iS} al eon a ees NIE NY NIH PE, BEEEESE 6 m. I2 m. I5, 18, 21 and 25 m. Proportion of Eggs Forming Larve (Blastule and Gastrule). June 9. 10) (0) Few (<1%) IO-15% ca. 50% 80-90% 60-70% 40-50% 20-25% ca. 2-3% June zo. June 15. (0) (0) Very few (<0.1%) | (0) a, 2%, << 1% ca. 10% CG. 8% 50-60% 20-30% 80-90% ca. 90% 90% ca. 90% 90-95% ca. 95% | 90-95% >90% | ca. 90%. 75-85% 70-80% 10-15% 15-20% (o) (0) (o) June 17. (0) Oo 25-30% 55-65% 90-95% ca. 95% 70-80% 30-40% At 34° the majority of eggs form membranes with one minute’s exposure, and 30 seconds is sufficient for a minority. A few eggs form larve after 2 minutes’ exposure; the optimum is 3 to 4 minutes; longer exposure is injurious. Table VII. summarizes Time of Exposure, 2m. Ny QdikU1 diH BE NIKO bi Ww HHESH ESS aS On Proportion of Eggs Forming Larve. TABLE VII. 34%. June ro. June 5. to) (0) gn, 2%, (0) 25-35% ) 50-60% 20-30% 65-75% ca. 90% 70-80% ca. 90% 50-60% ca. 90% : ca. 90% 5-10% ca. 10% cay 2% <1% (0) (0) 7, 8, 10 and 12 m. (0) August 26, August 27. eal eae ACTIVATION OF UNFERTILIZED SATRFISH EGGS. PTT the results of four series of experiments at this temperature. The August eggs were less favorable than the June eggs. Two similar series at 35° and one at 36° were carried out in June at a time when starfish eggs were unusually favorable. The results were similar to those at 33° and 34° except that the physio- logically equivalent exposures were shorter. At 35° an exposure of 30 seconds induces membrane-formation in many but not in all eggs, and one of 45 seconds in practically all. At 36° 15 seconds is sufficient to form membranes in about half the eggs and 30 seconds in all. Longer exposures eventually interfere with membrane-formation; thus after 6 to 8 minutes at 36° mem- branes failed to form or were imperfect in 40 to 50 per cent. of eggs. Table VIII. gives the proportions of eggs forming swim- ming larve in these experiments. The transfers from warm to normal sea-water were made at first at intervals of fifteen seconds. TaBLeE VIII. 35° and 36°. Proportion of Eggs Forming Larve. Time of Exposure. |—_ aa June 15 (35°). June 16 (35°). June 17 (36°). Tne, BO (0) (e) (0) asl” (0) ) 20-30% Im. Gi, UGG fo) 85-90% ay ale 57 ca. 5% ca. 4-5% 95% I m. 30” 80-90% 35-45% 70-80% i Io ARY >95% 70-80% ca. 10% 2m. ca. 90% =90% none free 24 m. ca. 50% 40-50% oO Bere CO, B%, 5-10% (0) 33 Mm. ) 0 Oo 4m.—-I0om (0) () fo) The rapid decrease in the optimum exposures as the tempera- ture rises is to be noted; the optima are respectively 1144 to 2 minutes at 35°, and I to 144 minutes at 36°. VARIATION WITH TEMPERATURE IN THE RATE OF THE PROCESS UNDERLYING ACTIVATION BY HEAT. The foregoing results show that the process, whatever its nature, which is initiated in the mature unfertilized starfish egg by temperatures of 30° to 36°, and which brings the egg into a condition to form membranes cleave, and develop, proceeds 278 RAPLH S. LILLIE. twenty or twenty-five times as rapidly at 35° as at 30°. Thus at 30° the minimum exposure for membrane-formation is about IO minutes, and at 35° 30 seconds or less; similarly at 30° the minimum exposure needed to induce even a few eggs to form larve is about 20 minutes, at 35° it is about I minute; at 30° the optimum exposure is ca. 30 minutes, and at 35° ca. 1% minutes. For each temperature it is possible to assign a definite length of exposure which produces a definite effect on the egg. The manner in which these times of exposure vary at different temperatures may be seen by reference to Table IX. Here are TABLE IX. s Minimum for ae Temperature Membrane- Minimum for Optimum for Larve. | Maximum for Larve. and Series. fOCmnOne arve. 29° (June 10) I4 m. AS ian. 30° (June 13 8-10 m. 12-18 m. 24-28 m. an® Mee 5 intl, 8 m. 12-14 m. 20 m. June 12 3 m. 8 m. ; Ce 33 m. 8 m. I4-I5 m. 21 m. (<25 m.) 32° (June 12 25 mM. 4m. 8 m. (June 13) 2m. 4-5 m. 8 m. I2 m. (I10 m.) (June | 2m. 4m. 6 m. Io m. (<12 m.) (June 26)) 2 ig). 5 m. 8-10 m. I2 m. 33° (June 9) 2m. 3m. 4m. 7m. (June 10), Im. 24 m. 43-6 m. IO m. (June a Im. 3m. 4-6 m. 8 m. (<12 m.) (June 17)) Im. 3 m. 5-6 m. Io m.+ 34° (June 10) 30 sec. 134m. 3-33 m. 5 m. (<6 m.) (June 15) I m. 24 m. 3-4 m. 53 m. (<6 m.) 35° (June 15) 30 sec. Im. |r m.30/—r m. 45” 3 m. (<3% m.) (June 16) 30 sec. i fia, eR” ca.2m . 3m. (<3% m.) 36° (June 17) I5 sec. As! ( S30”) Im.-rm.15” | ca.2m. (<23 m.) tabulated the observations made in all of those series of experi- ments in which a large proportion of eggs formed larve,—in which, therefore, the conditions may be regarded as essentially normal. In the series at 29° few eggs formed larvee; at 30° only one series out of five gave a considerable proportion of larve (ca. 40 per cent.) with ca. 30 minutes’ exposure; in all of the other series in the table, except one at 31°, the great majority of eggs— usually over 90 per cent.—formed larve with the optimal exposures. In the first column is given the least time of ex- posure required for membrane-formation in a significant pro- ACTIVATION OF UNFERTILIZED STARFISH EGGS. 279 portion of eggs—1Io per cent. or more; in the second column the least exposure at which any eggs (> I per cent.) formed larve; in the third the optimal time of exposure; and in the fourth the longest observed exposure at which any eggs (> 1 per cent.) formed larve. If the several observed durations at each temperature are averaged, the following results are obtained (Table X.); the values are given in approximate terms rather than strict arith- metical averages, to emphasize the fact that the precise durations vary to a certain degree, even in normal eggs. There is, however, for each temperature a well-defined modal duration of exposure for producing a definite physiological effect such as membrane- formation or complete activation. TABLE X. APPROXIMATE TIMES OF EXPOSURE REQUIRED TO PRODUCE THE FOLLOWING EFFECTS AT DIFFERENT TEMPERATURES. Formation of Minimum for Optimum for Maximum for Temperature, Membranes. Larve. Larve. Larve. Bow ca. 12-14 m. 20-25 m. 30-40 m. 30° 8-10 m. ca. 18 m. | ca. 28 m. ?>30 m. Biles ca. 4m. ca. 8m. ca. 15 m. 21-25 m. Boe ca. 2 mM. 4— 5 m. 7—- 8m. IO-12 m. Bee ca. I m. 24-- 3m. 43-53 m 8-10 m. 34° 30” to I m. Ga. 2) ms 3-34 Mm. ca. 5 m. 25° ca. 30” I-I}t m. 13-23 m. ca. 3 m. 36° Gan TY” 30-45” I-1I¢ m. ca. 2 m. Bin 30-35 sec. 38° ca. 20 sec. It will be noted (1) that for each temperature there is a minimum effective exposure which induces membrane-formation followed by failure to develop and early breakdown; and (2) that an exposure of approximately twice the minimum for membrane-formation is required to enable even a few eggs to develop to larval stages, and an exposure of three or four times this minimum to enable development to proceed normally in all eggs; and (3) that if the exposure is prolonged to about one and a half times this optimum the eggs are again incapacitated from further development. The fact that the ratios of the dura- tions required to produce these several effects are approximately the same at any one temperature indicates that a single process 280 RALPH S. LILLIE. of a definite kind forms the determining condition of all. This process is peculiar in undergoing marked acceleration by slight rise of temperature; it is also clear, from the fact that an effective exposure must last for a certain minimal time at any temperature, that the process must proceed to a definite stage before the egg is rendered capable of continuing its development to advanced stages; if the process is arrested before its completion, only the earlier developmental changes can be carried out (membrane- formation, early cleavage or change of form); if, on the other hand, it is allowed to proceed too far, injurious conditions arise which eventually prevent all development; a sufficiently pro- longed exposure to high temperature renders the egg incapable even of membrane-formation. | In endeavoring to form some consistent conception of the nature of this process the following facts have to be considered. It exhibits a high temperature-coefficient: from fifteen to twenty times the duration of exposure is required to induce membrane- formation at 30° as at 35°; the ratios between 29° and 34° and between 31° and 36° are the same. At each temperature the proportionate durations of the minimum, optimum, and maxi- mum exposures for forming larve are approximately the same. In other words, the critical change underlying simple membrane- formation is affected by temperature in the same way as that underlying complete activation of development: 7. e., the pro- portionate increase in velocity by rise of temperature is the same in both cases, a fact which can only indicate that one funda- mental process—and not two—is concerned in producing both effects. If we assume that the above proportionate increase in velocity prevails through a rise of 10°, a Qio value of from 225 to 400 is indicated, as against the 2 to 3 characteristic of chemical reactions in homogeneous media. Activation by heat thus 1 The temperature-coefficients of the rate of cytolysis of sea-urchin eggs and of the duration of life of sea-urchin larve and of Tubularia stems at temperature of 25° to 40° show similarly high values. In these cases the direct effect produced on the cell by the high temperature is probably of the same kind as that underlying the above activation-effect; this appears to be a change in the colloids of the plasma- membranes, leading to an increase of permeability. (See below, p. 296.) Such a change if not reversed within a certain time results in cytolysis. In the unfer- tilized starfish egg temporary increase of permeability involves activation. For ‘data on the temperature-coefficients of cytolysis and heat-death, cf. J. Loeb, srt wel ACTIVATION OF UNFERTILIZED STARFISH EGGS. 281 depends on some critical change in the egg which does not begin until a temperature of about 29° is reached, but which undergoes very rapid acceleration with further rise of tempera- ture. The liquefaction of gels by heat seems to be the only relevant process which shows these characteristics. The change in viscosity preceding the gelation of a gelatine sol undergoes very rapid acceleration with lowering of temperature, within a few degrees of the temperature of gelation. The inverse process, melting of gels, has a similarly high temperature-coefficient (cf. below, p. 295). In general the facts suggest that the direct effect of the high temperature is to cause a change in the colloidal system of the egg, of such a kind as to render possible a chemical interaction between substances which in the normal condition of the resting egg are kept apart. *This restraining condition may be some physical barrier like a membrane, impermeable to the diffusion of the substances concerned, or it may be a certain state of electrical polarization of the general cell-surface, as suggested below (p. 299). It is also important to note that the activation-process may be arrested by a return of the eggs to sea-water at ordinary temperatures, and renewed after an interval without interfering with its effect. A reversibility of the physico- chemical change forming its basis is thus indicated. It should further be noted that cytolytic agents like butyric acid not only have the same general physiological effect as brief warming, but that the relations between time of exposure and physiological effect produced are the same in both cases. Some process which is affected similarly by these two dissimilar agents is thus to be sought. In the following section the results of experiments with weak butyric acid solution are described in greater detail. EFFECTS OF EXPOSURE TO ButTYRIC ACID SOLUTION FOR DIFFERENT PERIODS. As already stated, treatment of starfish eggs during the early maturation period with weak solutions of butyric acid in sea- water (7/260) produces the same effects as temporary warming, Archiv f. d. ges. Physiologie, 1908, Vol. 124, p. 411; A. R. Moore: Quarterly Journal of Experimental Physiology, 1910, Vol. 3, p. 257; Arch. f. Entwicklungsmech., 1910, Vol. 20, pp. 146, 287. 282 RALPH §. LILLIE. and the time-relations of the exposures necessary for these effects are closely similar with both methods. Table XI. summarizes the results of five series of experiments with separate lots of eggs. The eggs were exposed at normal temperatures (20° to 22°) to an 2/260 solution of butyric acid in sea-water (50 c.c. sea-water plus 2 c.c. n/10 butyric acid), and portions were transferred to normal sea-water at the intervals named. The approximate proportion of mature eggs developing to free- swimming larve (blastule and gastrule) is given. RhApre XU: N/260 Butyric ACID.) Proportion of Mature Eggs Developing to Larve. Time of : : Reporte: Series 1 (Aug. 31.) ERS Series 3 (Sep. 1). |Series 4 (Sep. 2.) ee I m. (0) (0) I or 2 larve t blastula <1% 2m. 2 or 3 larve fo) GG, Wf ca. 1% 2-3% A ia, <1% <1% ca. 4-5 % I- 2% 20-30% 4m. <1% ca. 1% ca. 10% 5-10% 55-60% 5m. ca. 1% ca. 10% IO-15% 20-30% | 75-85% 6 m. ca. 5% ca. 50% 20-30% 30-40% 80-90% 7m. 20-30% 70-80% 40-50% 20-25% 35-40% 8 m. ca. 50% 65-75% 50-60% 15-20% 20-30% IO m. 80-90% 65-75% ca. 60% ca. 1% 10-15% I2 m. 30-40% 25-35% 40-50% t blastula (0) I5 m. <1% Gn, WG 20-30% (0) (0) The close parallelism between these experiments and those of warming to 32° or 33° will at once be noted. With brief exposure there is the same simple membrane-formation followed by break- down without development; as the exposure is prolonged there is a progressive increase in the proportion of favorably develop- ing eggs up to an optimum; then follow a decrease and eventual failure to develop. More detailed observations show that the rate and regularity of cleavage show a corresponding steady improvement up to an optimum which is again followed by a decline. } The following observations show the condition of the eggs in the second series of September 1, at about four hours after the treatment with butyric acid (Table XII.). The optimum time of exposure shows somewhat more vari- ability in these series than is usually the case with exposure to ACTIVATION OF UNFERTILIZED STARFISH EGGS. 283 warm sea-water (32°); in all five, however, the optimum lay between five and ten minutes.! There is thus an approximate constancy in the time of exposure required to induce complete development with solutions of this concentration. Probably TABLE XII. N/260 Butyric AcID. Condition of Eggs 4 Hours after Treatment, and Proportion Time of Exposure, forming Larve. TWIP eee ane ae er eae wane All eggs have membranes; most are irregular or amoeboid in form; none are cleaved. No larve. DAIS Osa DECOR OLR OTe Similar to r m. lot, but a few eggs (ca. 2-3 per cent.) are in the 2-cell stage. No larve. B TODS sy carceb CRI ETERS Generally similar to the 2 m. lot, but the cleavages are more numerous (ca. 10-15 per cent.), mostly 2-cell with a few 4-cell stages. Very few larve (< I per cent.). Gl, Sai dps ie ee sets dem ree ear Cleavages are more numerous and advanced; 40-50 per cent. are cleaved, mostly 2 and 4-cell, with a few 8-cell stages. Larve still few (ca. I per cent.). pla Meeet oyes cane sesalece iy a eapaiseaie Cleavage is more advanced than in the 4 m. lot; ca. 50 per cent. are cleaved, largely 8- and 16-cell stages. Ca. Io per cent. form larve. Gaderer pree se oe ey eee RN Most eggs are cleaved (ca. 70-80 per cent.), many in 16- to 32-cell stages. Ca. 50 per cent. form larve. GT IODA hoeee eect Ree tv er a Almost all eggs are cleaved (90 per cent. or more), many in normal-looking 16- to 32-cell stages. 470-80 per cent. form larve. RTI Re in ee ne as i tlre Similar to 7 m. lot; most eggs are in 16- to 32-cell stages. 65-75 per cent. form larve. MOpMbey tre caepewe a aseaw eke ee eee alec Cleavages are fewer and less advanced; ca. 70-75 per cent. are cleaved, mostly 4- and 8-cell stages. 65-75 per cent. form larve. TD! Weoley echelons OeRCRe Ve ee Te Nee ace Comparatively few cells are cleaved; ca. 10 per cent. are in 2- or 4-cell stages, largely irregular; the rest uncleaved. 25-35 per cent. form larve. TES) SRG 3:3 dye sai cae Tee ane oes pace Almost all eggs remain uncleaved, and many show the beginnings of surface-disintegration. Few form larvee,—ca. I per cent. Controls: Unfertilized eggs disintegrate without membrane-formation or de- velopment. Nearly all sperm-fertilized eggs develop to larve. 1 This variability may be due partly to the fact that on account of the lateness of the season and consequent scarcity of ripe starfish the eggs used in these experi- ments came from fewer animals; thus in Series I, 4, and 5, eggs from only one star- fish were used in each case, and in Series 2 and 3fromthree. Inthe earlier experi- ments with warm sea-water the mixed eggs from several animals were used in each series. 284 UNIEIZIEL Sj ILIDLILIIS, an inverse relation exists between the concentration of fatty acid and the time of exposure required to produce a given effect. Systematic experiments to determine the character of this rela- tion have not yet been carried out, but there are some observa- tions bearing on this question. In several of my experiments in the summer of 1912 starfish eggs exposed for only one minute to acetic or butyric acid of ca. /176 concentration (6 c.c. 1/10 acid plus 100 c.c. sea-water) formed a large proportion of larvee.! Lyon observed some years ago that the exposure required to induce parthenogenesis in Arbacia pustulata by means of weak solutions of HCl in sea-water decreased with increase in the con- centration of acid up to a certain point.2, The minimum exposure to n/260 butyric acid required to form membranes is very brief in starfish eggs. Experiments last summer showed that while 10 seconds was insufficient to form membranes in more than a few eggs (ca. 10 per cent.), with 20 seconds all formed membranes, followed by the typical irregular changes of form and breakdown. After one minute’s exposure to 2/260 butyric acid an occasional egg may form a blastula; yet in the series showing the shortest optimum exposure of any performed last summer (No. 5, Sept. 2) at least 3 minutes was required to enable any considerable proportion of eggs to develop to a larval stage. The parallelism between the effects of high temperature and of weak fatty acid solutions indicates that the two agents act by producing the same kind of change in the egg-system. More detailed experiments to determine the influence of concentration as well as time on the action of this and other cytolytic substances remain to be carried out, and their results will probably throw further light on the nature of this change. EFFECTS OF MEMBRANE-FORMATION BY HEAT oR Fatty ACID COMBINED WITH AFTER-TREATMENT BY THE SAME AGENT. The fact that a longer treatment with the membrane-forming agent produces the same effect as a short treatment combined with after-exposure to hypertonic sea-water or cyanide suggests that a suitable after-treatment with the membrane-forming 1Cf. Journal of Experimental Zoélogy, 1913, Vol. 15, pp. 41, 42. 2Lyon, American Journal of Physiology, 1903, Vol. 9, p. 310. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 285 agent itself should have a corrective effect similar to that exerted by the agents just named. If the effect of the initial or mem- brane-forming treatment is to cause a partial activation which requires later to be completed by the after-treatment, we should expect it to be a matter of indifference (within certain limits of time) whether the activation is completed in one stage—e. g., by a continuous warming to 32° for 8 minutes—or in several; development ought to follow equally well if the eggs are returned to sea-water after an exposure just sufficient for membrane- formation, and afterward again exposed to the same treatment for an appropriate length of time. Experiment shows that it is in fact possible to substitute for the after-treatment with hyper- tonic sea-water or cyanide a brief exposure either to warm sea- water or to 2/260 butyric acid. We have here clear indication that the essential changes produced in the egg by after-treatment with an agent like hypertonic sea-water are not qualitatively different from those caused by the first or membrane-forming treatment, but serve simply to renew and bring to its completion a process which has been initiated by the first treatment but prematurely arrested by the early return to normal sea-water. According to this conception the whole activation-process is unitary in nature and does not consist of two qualitatively distinct and mutually complementary processes, as Loeb has maintained on the basis of his experiments with sea-urchin eggs. The following series (Table XIII.) illustrates the effects of treating eggs, in which membranes have been formed by 3 minutes’ exposure to 32°, a second time with sea-water at 32° for 4. minutes; the second exposure was made at varying intervals after the first, ranging from 9 minutes to nearly 4 hours. TABLE XIII. AFTER-TREATMENT WITH SEA-WATER AT 32°. June 24. Eggs from several starfish were exposed, about 35 minutes after removal from the animals, to sea-water at 32° for 3 minutes (11.12-11.15 A.M.), and then returned to sea-water. Part of these eggs were left permanently in sea- water for control; the rest were again exposed to 32° for 4 minutes, successive portions being thus treated at 10-minute intervals until well after the separation of the second polar body. The condition of the maturing eggs at the time of the second treatment is indicated in the first column. 286 RALPH S. LILLIE. Control lots of eggs were exposed (for purposes of comparison) to 32° for single continuous periods ranging from 2 minutes to 10 minutes. After-exposures to 32° at Follow- Results (Condition of Eggs ca. 4 Hours Later, and ' ing Times after First Exposure. Proportion Forming Larve. I. Control: no second ex- iS) DOSUTE Aj sere All mature eggs have membranes but are uncleaved and largely irregular in form. None form larve. - 9 Mm. (II.24—11.28) (no polar bodies at 11.28)... Marked contrast to control; almost all eggs are cleaved, largely to 32- or 64-cell stages. Ca. 70-80 per cent. form larvae many of which swim at the surface of the water. . IQ m. (11.34-11.38) (first polar bodies be- ginning to separate at Tet? 3'8)\ agence meade eee cetas Cleavage is less advanced than in Experiment 2, and a minority are uncleaved. Somewhat fewer larvee (ca. 65-75 per cent.). . 290 m. (II.44-11.48) (first polar bodies in all CEES) ane MIWA) oooncoese Cleavages are fewer and less advanced than in Ex- periment 3. Most eggs are uncleaved. Larve fewer (ca. 30—40 per cent.). 2 20) me) Gr54—T1 58) (all with first polar bodies, none with second at 11.58)........ Contrast to Exp. 4; great majority are uncleaved and largely irregular; a few 2- and 4-cell stages present; few form larve (ca. 5 per cent.). . 49 m. (12.04—12.08) (ca. 50 per cent. have second polar body at TOS Nise cen cee ar oeerees Similar to Exp. 5 but with fewer cleavages. Larve also are fewer (ca. 2-3 per cent.). > 5G) sin, (12.14-12.18) (all eggs have second polar bodies).......... Nearly all are uncleaved; largely irregular or frag- mented. Almost no larve (only one feeble blas- tula seen). . Th. gm. (12.24-12.28)...All are uncleaved but irregular forms are fewer. No larve. » Gs B lls (AAGO=BOB) ooo coc Similar to Exp. 8. Controls with one exposure to 32°: eggs exposed 7 minutes continuously (11I.12— II.19) gave ca. 90 percent. larve. With 4 minutes’ exposure few eggs (ca. 2-3 per cent.) formed larve. Controls of unfertilized and sperm-fertilized eggs were normal. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 287 Experiments similar to the above were performed with pre- liminary exposures to 32° of 2, 3 and 4 minutes, followed by after-exposure to 32° as above for 4 minutes (in one series for five minutes), all of which gave the same general result. Appar- ently it is a matter of indifference whether the second exposure to 32° follows immediately after the first or at an interval, provided that the second exposure takes place before the separa- tion of the first polar body. After this event there follows a decided and rapid decline in the favorability of the response to the after-warming treatment, and after the separation of the second polar body after-warming is apparently quite ineffective. As I described in my former paper on this subject, the suscepti- bility to parthenogenesis by temporary continuous warming always undergoes marked and rapid decrease at the time of the maturation-divisions... The above decline in the response to after-warming is evidently the same phenomenon. A similar decrease in the susceptibility of the eggs to sperm-fertilization also takes place at about the same time, although this decrease is not so pronounced as in the case of parthenogenesis; thus it is usually possible to fertilize a certain variable proportion of starfish eggs (not all) after maturation has been complete for some hours.” The fact that the general responsiveness of the egg to any activating agent undergoes a sudden decline at the time of separation of the polar bodies suggests either that some material necessary to development is then lost, or that a refrac- tory state conditional on some other kind of change (possibly a change in the plasma-membrane) then develops. As already pointed out, the fact that sperm-fertilization is possible (although less favorable) at a time when the egg fails to respond to the parthenogenetic treatment suggests that some definite material playing an important part in development is introduced into the egg by the sperm. This is also indicated by the general fact that sperm-fertilization induces a more favorable development than artificial activation. It may be that this material is the Same as some substance lost from the egg at the time of the maturation-divisions. Further research has to decide between these possibilities. 1 Loc. cit., 1908, p. 400. 2 IL GG, (icy 185 Alike 288 RALPH S. LILLIE. In the above described experiments the total optimum period of exposure to 32° is about the same (ca. 7 to 8 minutes) whether the exposure is continuous or in two stages. No doubt it would be possible to increase the number of stages to three or more, especially if lower temperatures (31° or 30°) were used, but no experiments of this kind have so far been attempted. Appar- ently what is essential is that the critical process begun by the warming should continue, at the given temperature, for a certain definite length of time, sufficient presumably to allow some critical chemical interaction to proceed to its completion. It is interesting to note that a preliminary warming which is too brief in itself to cause membrane-formation may nevertheless have the effect of shortening the period of after-warming neces- sary to cause complete development. In one experiment the preliminary exposure to 32° was only 2 minutes, a time insufficient for membrane-formation in more than very few eggs (ca. I per cent.); these eggs, however, when again exposed to 32° for 4 minutes, gave a considerable proportion of larve (5 to 10 per cent.); while eggs exposed to 32° for 4 minutes without any previous treatment formed membranes, but none developed to larve. A continuous single exposure of 6 minutes gave 25 to 35 per cent. of larve; this exposure was well below the optimum of 8 to 10 minutes at which 80 to 90 per cent. formed larve. This effect of the four minutes’ after-exposure on eggs which otherwise showed no external change indicates that membrane- formation is not in itself a critical event, but simply an expression of a partial initiation of the general developmental process: i. €., a partial activation has been accomplished, enabling the egg to carry out a few of the early steps in development. Since brief exposure to weak fatty acid solution has the same physiological effect on the egg as brief warming, it would appear that the essential change produced in the egg-protoplasm by either form of treatment is the same; if so, after-treatment with warm sea-water should have a similarly favorable effect on eggs in which membranes were formed by fatty acid. The following series of experiments shows that this is the case (Table XIV.). The eggs, after membrane-formation by butyric acid, were after- treated with warm sea-water (32°) for periods ranging from 2 ACTIVATION OF UNFERTILIZED STARFISH EGGS. 289 to I2 minutes. For comparison part of the eggs were after- treated with hypertonic sea-water and cyanide. TABLE XIV. n/260 ButTyrRic ACID WITH AFTER-TREATMENT WITH SEA-WATER AT 32°. August 24. Eggs from one starfish were used. These eggs were not very favorable and a rather small proportion underwent maturation. They were exposed, about 45 minutes after removal, to 7/260 butyric acid solution for one minute and then returned to normal sea-water. Twelve to sixteen minutes later they were aiter-treated as follows, with the results indicated. ieee Mee Results iatavss ener ee oe Proportion rt. None (control treated with butyric acid AONE) ernest Typical fertilization-membranes in all mature eggs; later the eggs assume irregular forms and break down. None form larve. 2. Hypertonic sea-water (250 c.c. Ss. w.+40 c.c. 2.5 m NaCl) for SO MTEUIA ee eyn ee ester Markedly favorable effect: most mature eggs are cleaved to ca. 32-cell stage. 20-30 per cent. form larve. 3. M/tooo KCN in sea- water for 30 min. ....Eggs cleave as in Experiment 2. Ca. 25-30 per cent. form larve. 4. Sea-water at 32° for 2 LTT TA es EU EWG Fah ne wae After four hours most eggs are irregular and un- cleaved; a few are cleaved. Very few form larvee (< I per cent.). 5, QO° tigi @ imi, ocaono0s Like Exp. 4, but more eggs are cleaved. Few larve, —1I per cent. or less. OG, 32° ior Al WA. Gono ane Cleavages are more numerous than in Exps. 4 and 5. Ca. 5 per cent. of mature eggs form larve. Fo BEC TOr FR WMDs 5 a5 o00cc Ca. 20-30 per cent. are cleaved. Ca 5 per cent. form larve. So BO? tow © WM boo cop ad Cleavages are more numerous than in Exp. 7. Ca. 40-50 per cent. larve. OQ. BAe tow G/ whe. bs oonoece Most eggs in 16- to 64-cell stages. 50-60 per cent. form larve. 10. B@° iow 8 imine. po dono06 Like Exp. 9, but somewhat less favorable. Ca. 50 per cent. form larve. II. 32° for 10 min.........After five hours few eggs are cleaved and cleavages are less advanced than in Exps. 9 and 10. Ca. 10-15 per cent. form larve. TA, QO nohe wy inshngs 6! og oo oe Almost none have cleaved after five hours. Practic- ally none form larve (one blastula seen). 290 RALPH S. LILLIE. For comparison eggs were exposed to 32° without previous membrane-formation for 4, 5, 6, 7, 8, and 10 minutes; the optimum exposure was 8 minutes at which 50-60 per cent. of the mature eggs formed larve. A sperm-fertilized control also yielded numerous larve. After-exposure to 32° for the proper time thus greatly increases the proportion of favorably developing eggs. No marked im- provement is seen until the duration of after-exposure reaches four minutes; with longer exposures the proportion of eggs forming larve shows progressive increase up to an optimum at about seven minutes; a decline then follows; an exposure of 10 minutes effects only slight improvement, and one of 12 minutes appears ineffective. Similar results, differing slightly in detail in different series, were obtained in eight other series of experi- ments. In general, after the preliminary membrane-formation by one minute’s exposure to 2/260 butyric acid, the time of exposure to 32° required for optimal development was found to range from 5 to 7 minutes; one minute’s exposure to 2/260 butyric acid appears thus physiologically equivalent to warming at 32° for the same or a somewhat longer period. After-treat- ADATEILID OVE n/260 BUTYRIC ACID WITH AFTER-TREATMENT WITH SEA-WATER AT 34°. August 27. The eggs from one starfish were used. The eggs were few in number, but the majority showed normal behavior. They were exposed to 1/260 butyric acid for one minute and then returned to sea-water. Later (within 20 minutes) portions were exposed to hypertonic sea-water, cyanide, and warm sea- water as indicated. After-treatment. Results. Te NOnen(COntLOl) peer Typical membrane-formation, followed by breakdown of almost all eggs. One blastula found. 2. Hypertonic sea water POTS O} TS Prose spoon ere oes 35-45 per cent. of the eggs form larve. 3. n/t1000 KCN for 30m. . Ct, 50 per cent. of all eggs form larve. Ay Bao nope in Tb Spo Goa 6 Only a few eggs form larve: < I per cent. a Syl Noe A Me Sb 6 bin 68 oo Marked improvement: 20-30 per cent. form larve. (Gs Gal NOW B WMNbNes oo bond 6 Larvee are fewer than in Exp. 5: ca. 20 per cent. 7. 34° for 4min...........Few eggs form larve: < I per cent. Go GVO Oe Ginins 55 oboe + ae Most eggs fail to divide; none form larve. Warming at 34° without previous membrane-formation: Eggs were exposed to 34° in the usual manner for 2, 3, 4, 5,6, and 7 minutes. The best development resulted from the 2- and 3-minute exposures, with respectively 25-35 per cent. and 35-40 per cent. of eggs forming larve; with the 5-minute exposure only 5 per cent. formed larvee. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 291 ment with sea-water at 31° and at 34° was also tried; the results were the same except that the after-exposure required at 34° was only a half to a third as long as at 32°, and at 31° about twice as long. The following series at 34°(Table XV.) is typical. These results show that the effective duration of after-exposure at 34° is about one third of what it is at 22°22 aie Bins We ESE results were gained with after-exposures of 8 to 10 minutes. The temperature-coefficient of the physiological change resulting from the after-warming treatment is thus evidently of the same order as in the case of simple warming without previous mem- brane-formation. This of course is not surprising, since un- doubtedly the same process is concerned in activation by heat whether this is preceded by another treatment or not. It is thus plainly a matter of indifference, as regards the effect AUgsieis OVAL. BotH MEMBRANE-FORMATION AND AFTER-TREATMENT BY n|/260 Butyric ACID September 6. The eggs from one starfish were used; these were few in number, but almost all (ca. 90 per cent.) showed normal maturation, and in the sperm-fer- tilized control almost all formed larve. The eggs were exposed for one minute to n/260 butyric acid and returned to sea-water; part were left in sea-water as control; the remainder were again placed, 18 minutes later, in n/260 butyric acid, from which portions were returned to normal sea-water at the intervals indicated. These eggs developed as follows: Results (Condition of Eggs after 4 Hours and fter- t 5 A : ASSES Sie ee Proportion forming Larvz). TeNonen(control) pene eee All show typical membrane-formation followed by irregular change of form and breakdown in nearly all eggs. Only one larva seen. 2. N/260 Butyric acid: 2 TLIO eS alone ToeWe were A few eggs are cleaved. Ca. 10-15 per cent. form larve. 3. Butyric acid: 4m. ......Cleavages more numerous and more regular than in Exp. 2. Ca. 40-50 per cent. of eggs form larve. 4. Butyric acid: 6 m....... Cleavages still more numerous: Most eggs form larve (70-80 per cent.). 5. Butyric acid: 8m....... Like Exp. 4, but fewer eggs form larve (50-60 per cent.). 6. Butyric acid: 10 m...... Cleavages are fewer and slower. 25-35 per cent. of eggs form larve. 4. Butyric acid: 12 m...... Cleavagesare stillfewer. Ca.10 percent. of eggs form larve. 8. Butyric acid: 15 m...... Practically none are cleaved. No larve. 292 RALPH S. LILLIE. produced by this form of after-treatment, whether the membrane- formation is induced by heat or by fatty acid; in either case warming for a few minutes completes the process of activation and enables the eggs to develop favorably. Precisely the same effect is gained by after-exposing eggs, in which membranes ~have been formed by either method, to weak solutions of fatty acid for a brief period; the effects of such treatment are in all respects similar to those of after-warming. This is illustrated by the following experiment (Table XVI.). It is clear that in the time-relations of its action as well as in its other characteristics, this form of after-treatment resembles closely that with warm sea-water. It is also possible to treat the eggs first with warm sea-water and then after-treat with butyric acid solution; precisely the same results follow as in | the experiment just described. Thisis illustrated by the following series (Table XVII.). ? TaBLeE XVII. BRIEF EXPOSURE TO 32° WITH AFTER-TREATMENT BY 7/260 BuTYyRIc ACID. September 7. The eggs from one starfish were used; eggs were few but appar- ently normal, over 90 per cent. showing normal maturation, and sperm-fertilization resulting in a large proportion of larve. The eggs were exposed to sea-water at 32° for 3 minutes, then returned to sea-water at normal temperature, and 16 minutes later placed in 2/260 butyric acid solution, from which they were again returned to sea-water after the times indicated. After-treatment. Results. To Nome (@2a° jie BS. i, alone) ee ae cere No development; only a small proportion form mem- branes. 2. N/260 butyric acid: DIN Se Servite eueeskeiene eacnmaren aus All form membranes but few are cleaved after three hours. Ca. 5 per cent. form larve. 33, Biba BXCNGIS 4) Ts 5 sao oc A large proportion (50-60 per cent.) are cleaved after three hours. More than 50 per cent. form larve. Al, sie PAS ACIS © 7, 555000 Most eggs are cleaved after three hours. 70-80 per cent. form larve. Be Bltyricacid @oms sane In contrast to Exp. 4, few eggs are cleaved after three hours, and only 1-2 per cent. form larve. 6. Butyric acid: 10m...... No eggs cleave within three hours. None form larve. The favorable effect of this after-treatment is evident. It will be noted that the three minutes’ exposure to 32° was in- sufficient for membrane-formation in most eggs; but the effect ACTIVATION OF UNFERTILIZED STARFISH EGGS. 293 of this preliminary treatment is seen in the fact that an after- exposure of only 4 minutes was sufficient to induce development to larval stages in more than half of the eggs. After-exposure to butyric acid solution has the same favorable effect when the preliminary warming is sufficient to form membranes in all eggs; in a second similar series on September 12 the eggs were exposed for 4 minutes to 32° and all mature eggs thus treated formed membranes; without any after-treatment almost none (less than I per cent. formed larve, but with an after-treatment of 4 to 8 minutes with 2/260 butyric acid favorable development took place in a large proportion of eggs. GENERAL DISCUSSION AND CONCLUSION. The interchangeability of the treatments with warm sea-water and butyric acid solution indicates that both agents produce their effect by inducing the same kind of change in the egg- system. This change is evidently of a ‘‘releasing’’ kind, and initiates the sequence of developmental processes; these, once started, continue automatically to their conclusion. Probably their most distinctive peculiarity is the highly specific character of the chemical transformations that take place. From the food contained as reserves in the egg, or taken in from the surround- ings, the developing germ builds up the specific compounds which form the structural basis of the organism; this synthetic process, in the case of the chief structure-making compounds, the proteins, undoubtedly starts—as in the constructive metabolism of the adult animal—with the amino-acids, which are recombined in the specific manner predetermined by the chemical organiza- tion of the germ. Bodies of the most highly specific and indi- vidualized physical and chemical properties are thus built up and laid down in definite positions as development proceeds. Their properties and their spacial disposition determine at any time the character of the transformation undergone by the building material which is being incorporated. According to this conception it is the chemical specificity of these substances that determines the specific character of development in the more evident or morphological sense,! 7. e., why the egg gives 1 Reichert’s work on the crystal-forms of haemoglobin and other complex compounds from different species of animals and plants constitutes perhaps the 2904 RALPH S. LILLIE. rise to an individual of the same species; and we must therefore be prepared to find among the earliest chemical changes asso- ciated with development, interactions of a specific kind— . e., specific in the sense in which the interaction of antigen and anti- body is specific—between complex substances already present in the egg. There is now definite experimental evidence that such reactions do in fact constitute an essential part of the fertilization- process. Specific substances which apparently unite in fertiliza- tion (since after fertilization they are no longer demonstrable) are present in the unfertilized mature egg; one of these (‘‘fer- tilizin’’) may be largely removed from the egg by washing, and when this is done fertilization is prevented. If such specific unions are essential to fertilization, we must conclude that the specific substances concerned in this process are in some way kept from interaction in the resting mature egg, and that the activating agent removes this hindrance to interaction. The question which I wish briefly to discuss in this section relates to the nature of this inhibiting condition, and the manner in which the activating agent effects its removal. The nature of the effects following exposure of unfertilized eggs to temperatures of 30°-35° indicates clearly that activation does not depend on simple acceleration of some chemical process, é. g., oxidation, which is already proceeding in the egg, since in this case the temperature-coefficient of the activation-process would presumably show the usual value of Qi = 2-3. It is also evident that heat-coagulation is not concerned, since these temperatures are too low, and the readiness with which the activation process can be arrested by cooling and renewed by a second warming shows that its basis is some effect which is com- pletely reversible by change of temperature. These character- istics, high temperature-coefficient and reversibility with change of temperature, are however shared by the typical melting and gelation (sol-gel transformation) exhibited by solutions of gela- best evidence of this. The morphological characters of crystals and crystal-aggre- gates varies with their chemically specific (‘‘species-specific’’) character in a definite and constant manner. It is fair to assume that the influence of these compounds in determining organic structure depends largely on the kind of aggre- gates they form. Cf. Reichert: Science, 1914, N. S., Vol. 40, page 649. 1Cf. F. R. Lillie, Journal of Experimental Zoology, 1914, Vol. 16, p. 523. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 295 tine, agar, soaps, lipoids and other hydrophilous colloids. The relations of temperature to this process show in fact a close resemblance to those described above for the activation-process. One striking peculiarity of melting and gelation is that both processes take place gradually; when (e. g.) a gelatine sol is brought below the gelation-temperature and the conditions are then kept constant, the actual solidification takes place only after the lapse of a considerable period of time. The time re- quired to reach the gelation-stage decreases rapidly as tempera- ture is lowered; thus Levites found that a gelatine sol kept un- disturbed at 26° took 26 hours to gelatinize, at 25° only 11 hours.! The first observable change in the solution is an increase in viscosity; this continues until the system sets; the setting repre- sents the end-stage of the whole process, whose course can thus be traced by successive viscosity-determinations. Gelation is thus equivalent to a progressive increase in viscosity to a final stage at which the ordinary fluid mobility is lost.2. It is found that above a certain temperature the viscosity of the hydrosol undergoes no change with time; but if the temperature is lowered a critical point is eventually reached below which the viscosity undergoes steady increase (at a rate dependent on temperature, presence of salts, reaction) until gelation occurs. The rate of this increase in viscosity (i. e., of the gelation-process), An/At, shows a high temperature-coefficient. With a 1 per cent. gelatine solution Schroeder? obtained the following values for the viscosity at 21°, 24.8°, and 31° at different intervals after bringing the warm gelatine solution to the temperature of observation: Viscosity Observed at Interval. 210% 24.89. Bro. ETI ATA pene ey ate mies aus iss, Cae 1.83): T.05 I.41 HG) Wahea sy cle osteo eH GI eRe 2.10 I.69 I.41 IG AUGDUT Ae cnc ceeee eee Rc ee 2.45 1.74 I.42 OMIM Rea yokes haat Arle 4.13 1.8 1.42 OOM recy Sie seve uees alteate 13.76 1.9 I.42 Thus while at 31° the viscosity undergoes no change with time, 1Levites, Kolloid-Zeitschrift, 1907, Vol. 2, p. 211. 2 Cf. Schroeder, Zeitschrift fiir physikalische Chemie, 1903, Vol. 45, p. 75; Levites: loc. cit., p. 209; Freundlich, “‘Kapillarchemie,’’ 1909, pp. 416 ff. 3 Schroeder, loc. cit., p. 88. 296 UNLESS, LULL, at the lower temperatures there is a steady increase. If we take comparatively short time intervals, e. g., Io minutes, we Bilis = Th foe 10 ey == 1405; 10 find that the value of Ay/At at 21° ( = 0.062 ) is about seven times greater than at 24.8° ( = 0.009 ) ; In other words, a difference of about 4° increases the average rate of the gelation-process from six to seven times. What is true of the gelation-process is also true of the inverse degelation or melting process, whose rate increases at a similarly rapid rate with rise of temperature above the critical maximum at which the system remains permanently in the gel state.! In starfish eggs the rate of the activation-process, at tempera- tures between 30° and 36°, shows a similar proportionate increase with a given rise of temperature, as will be seen by reference to Table X.; 7. e., the temperature-coefficients of the two proc- esses, gel-sol transformation, and activation of the egg under the influence of high temperatures, are similar in their order of dimensions; thus a rise of 4° shortens the time of exposure necessary to cause membrane-formation or development by six to ten times. On the assumption that some specific chemical interaction is the essential change in the initiation of develop- ment, such a result indicates that the rate of this interaction is dependent, in the case of parthenogenesis by warming, on the rate of some process involving either degelation or decrease in the viscosity of some portion of the colloidal system of the egg. This is as much as can be inferred on the basis of these facts alone. If we also take into account the other methods by which mem- brane-formation and activation can be induced, we are led to the further inference that this colloidal change affects chiefly if not exclusively the surface-layer (cortical zone or plasma-membrane) of the egg. Thus typical membrane-formation can be induced by brief treatment with pure isotonic solutions of neutral salts.? 1On account of the hysteresis of the gelatine system, the melting temperature is typically several degrees higher than the solidification-temperature; it is also higher after the gel has stood some time than immediately after solidification. Cf. Pauli (Pascheles): Archiv f. d. ges. Physiologie, 1898, Vol. 71, p. 336. 2R.S. Lillie: American Journal of Physiology, 1910, Vol. 26, p. 106. The fact CAGE ete tie ee aa ACTIVATION OF UNFERTILIZED STARFISH EGGS. 297 whose action is certainly superficial, as well as by substances like fatty acids, weak bases, and lipoid-solvents, which readily penetrate the plasma-membrane. ‘Those neutral salts of sodium and potassium which are the most effective in inducing mem- brane-formation, iodides and thiocyanates, are also the most effective in lowering the melting points of protein gels and in promoting water-absorption by such gels.'. Such facts suggest that the salts act in a way similar to that of high temperatures, i. e., by furthering degelation of surface-structures or absorption of water in the surface-layer of the egg. The effect of such an increase in water-content would be to increase the general permeability of this region, since according to the experiments of Bechhold, Ruhland, and others? the permeability of gels to diffusing substances, especially to colloids, is a direct function of their water-content. High temperature, according to this interpretation, acts like other parthenogenetic agents, by increasing the permeability of the surface-layer,—this effect resulting directly from some change in the nature of a degelation or decrease in the viscosity of the colloidal system in this region. Apparently. the immediate effect .of this change is to allow a chemical interaction to“take place between substances which in the normal resting state of the surface-layer are kept apart. The general fact that identical physiological effects may be produced by lipoid-solvents, and by substances which appear to alter the membrane by interacting chemically with its constituents,’ indicates that the integrity of the plasma-membrane as a semi-permeable partition is the essential factor in preserving the resting condition of the egg.* that this action can be prevented by anesthetics confirms the view that it depends on an increase in the permeability of the plasma-membrane: cf. my recent paper in the Journal of Experimental Zoology, 1914, Vol. 16, p. 591. 1 Cf. Pauli (Pascheles): Archiv f. d. ges. Physiologie, 1898, Vol. 71, p. 333; Levites: loc. cit.; Pauli and Rona, Beitriéige zur chemischen Physiologie u. Pathologie, 1902, Wolk, By Ds Ala 2 Bechhold u. Ziegler, Zeitschr. f. physik. Chem., 1906, Vol. 56, p. 105; also, “die Kolloide in Biologie u. Medizin,’ 1912, p. 48. Ruhland: Biochemische Zeitschrift, 1913, Vol. 54, p. 59; Freundlich, Kapillarchemie, pp. 515 seq. 3 When membrane-forming substances act by combining chemically with egg- constituents, it is to be expected that the rate of action will vary with temperature in accordance with the chemical temperature-coefficient. Cf. the experiments of Loeb and Hagedoorn, ‘‘ Artificial Parthenogenesis and Fertilization,’ page 146. 4Cf. my paper, Amer. Journ. Physiol., 1911, Vol. 27, p. 289. 298 LUNIEI2IEL Sip IED, Hence it is a matter of secondary importance in what manner this semi-permeability is temporarily destroyed, provided that the condition of increased permeability lasts long enough—not too long—and is not associated with irreversible changes making recovery impossible. It is presumably during this stage of increased permeability that the above specific interaction takes place; this process requires time, and its rate will be a function of the rate at which the two interacting substances can come together; this second rate will be a function of the viscosity or gelation-state of the protoplasmic system at the site of inter- action,—hence its dependence on temperature, as seen above. When this critical interaction has taken place, there follows at once the characteristic change of physiological activity normally resulting from fertilization; membrane-formation and the other events preparatory to cell-division occur and the developmental process proper is initiated. How far development proceeds, however, depends on the degree of completion of the primary specific reaction; hence for complete activation the exposure to the membrane-forming condition must have a certain minimal duration, and in case the preliminary exposure is insufficient some after-treatment may be necessary to complete the process. This after-treatment may be of the same kind as the preliminary membrane-forming treatment, or it may be of entirely different kind—e. g., hypertonic sea-water, cyanide, an anaesthetic, etc. But there seems to be no need of assuming that its direct physio- logical effect is qualitatively different from that of the membrane- forming agent.1 It merely renews and brings to completion a process already initiated by the first treatment. Comparative study of the conditions of both normal and 1 The above experiments are a sufficient justification of this contention. But they do not explain why, for instance, after-treatment with cyanide, which by itself does not induce membrane-formation in starfish eggs (cf. Journal of Experimental Zoology, 1913, Vol. 15, p. 38), is so effective. Clearly the condition of the egg after membrane-formation is altered so that the activation-process may then be influ- enced by agents which previously had no effect upon it (as cyanide, alcohols, or hypertonic sea-water in brief exposure). Sensitization to these agents seems to be involved in the process of membrane-formation, but the basis of this effect can not be defined at present. There is, however, no necessary inconsistency between these facts and the conception that the activation-process is essentially unitary in char- acter in the above sense. The case of hypertonic sea-water offers certain special problems, which are partly discussed below. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 299 artificial activation ought to yield data from which by elimina- tion the essential factors common to the two processes may be determined. Judging from the data available at present, the most general common feature appears to be the initial increase in permeability.t It is not yet clear, however, how this change can be the means of initiating the specific interaction assumed. The substances which interact are assumed to be present in advance in the egg; how is their interaction prevented by the existence of a semipermeable surface layer? The connection between change of permeability and activation is probably indirect; and the analogy to stimulus and response in the general stimulation-process of irritable tissues still seems the best adapted to throw light on this question.? In stimulation an electrical depolarization of the plasma-membranes of the irritable elements is apparently the critical event; in some way this change enables the characteristic response of the irritable system to take place. Similarly in the initiation of development in the unfertilized egg. The agents which induce membrane- formation in eggs have typically a depolarizing action on irritable cells like muscle-cells—4. e., cause a negative electrical variation.’ Such a change appears to result whenever surface-permeability is increased; and it seems therefore probable that this depolariza- tion, as such, is what enables the union of specific substances— the first step in activation—to take place. We may assume that one of the interacting substances is situated immediately beneath the electrically polarized surface-film of the egg, that it is a negative colloid, and that its tendency to unite with some amboceptor-like body also present in this region is compensated by the electrostatic attraction between it and the layer of 1 Cf. my paper just referred to. Inarecent paper Gray confirms McClendon in finding a temporary increase in the electrical conductivity of sea-urchin eggs immediately after sperm-fertilization. Cf. Gray, Journ. Mar. Bicol. Ass., 1913, Vol. r0, p. 50; McClendon, American Journ. Physiol., 1910, Vol. 27, p. 240. 2 T have discussed this analogy in more detailin the paper above cited (footnote 2, p. 296); also in the Journal of Experimental Zoology, 1913, Vol. 15, p. 23. 3 For the action of cytolytic substances in producing local negative variation, cf. Straub, Archiv f. exp. Path. u. Pharm., 1902, Vol. 48, p. 1; Zeitschr. f. Biol., 1912, Vol. 58, p.'251; Henze: Arch. f. d. ges. Physiol., 1902, Vol. 92, p. 451; Hermanns: Zeiischr. f. Biologie, 1912, Vol. 58, p. 261; Allcock; Proc. Roy. Soc., B, 1906, Vol. 77, p-. 267; Journal of Physiology, 1906, Vol. 33, p. xxviii; Evans, Zeitschr. f. Biol., 1913, Vol. 59, Dp. 397. 300 RALPH S. LILLIE. positive ions immediately external to the egg-surface. De- polarization would then permit interaction to take place. Such a conception, while in a sense diagrammatic, helps at least to explain how a non-specific agency, provided it only alters sufficiently the boundary-layer of the egg, can be the means of initiating such a highly specific process as development. The discussion of this question can hardly be considered complete without some reference to the case of hypertonic sea- water. As Loeb has shown, exposure to this agent forms a supplementary treatment which is remarkably favorable with some eggs, especially sea-urchin eggs. This treatment seems to occupy a special position among the parthenogenetic agents. It may either precede or follow the membrane-forming treat- ment,” and in some way it puts the egg into a condition which is favorable to subsequent development; this action seems quite independent of the nature of the membrane-forming or activating agent, and so far it has received no satisfactory explanation. Loeb has shown that a purely physical abstraction of water is not the only factor concerned; a chemical factor, apparently involving oxidation, is essential; free oxygen must be present during the treatment, and the effective times of exposure vary at different temperatures according to the chemical temperature- coefficient.2 Some hypothesis as to its mode of action seems required; and I suggest the following, which is consistent with the foregoing point of view, and has not, to my knowledge, yet been put forward. 1 The inorganic analogy would be, e. g., the interaction between solution and metal at the surface of the plate in a battery when the circuit is closed. While the battery is at rest (with open circuit), interaction between (e. g.) sulphate ions and zinc is prevented by the polarization at the surface of the zinc plate. The tendency to this ionic interaction is compensated by the polarization, the zinc ions being held back by the negatively charged plate. Similarly, mutatis mutandis, with the reactions at the cell-surface, or other surfaces (adsorption-surfaces) within the cell. The facts of stimulation afford in general strong evidence that the chemical proc- esses in the living cell are largely dependent on changes in the electrical polar- ization of the limiting membranes. Cf. my paper in the Journal of Biological Chemistry, 1913, Vol. 15, p. 237. Also, for a more general discussion of this ques- tion, the article entitled ‘“‘The Physico-chemical Conditions of Stimulation,” in the Popular Science Monthly, 1914, p. 579. 2 Cf. Loeb, “‘ Artificial Parthenogenesis and Fertilization,’ Chapter 11; Archiv fiir Entwicklungsmechanik, 1914, Vol. 38, p. 409. 3 “* Artificial Parthenogenesis and Fertilization,’’ Chapter 11. ACTIVATION OF UNFERTILIZED STARFISH EGGS. 301 It is to be assumed that the activation-process—as the earliest step in development, an essentially constructive process—in- © volves syntheses of some kind. Now the intracellular as well as other organic syntheses consist as a rule, in the union of two or more molecules, with loss of water, to form larger molecules,—as in the formation of fats from glycerol and acids, of starch and glycogen from sugar, of polypeptides and proteins from amino- acids, etc. In order to account for the readiness with which these condensations occur in cells, it seems necessary to assume that the protoplasm is the seat of energetic dehydrations, prob- ably in certain localized situations (possibly at membranes or other adsorption-surfaces). The artificial enzymatic synthesis of triolein from glycerol and oleic acid has been found to take place readily only when water is removed as completely as possible from the reacting mixture.’ Hence the synthesis of fats by enzyme action in cells is intelligible only on the assump- tion that in the region of their formation there is energetic abstraction of water or dehydrolysis. Certain biological facts indicate that partial removal of water from cells is favorable to syntheses of the above kind. According to Overton, plasmolysis of plant-cells furthers the formation of starch in chloroplasts.” Butkewitsch also finds that the formation of starch in the amylase-rich cortex of certain plants (Sophora, Robinia) is promoted by placing in strong sugar-solutions (10-20 per cent. dextrose and saccharose).2 The observations of Pavy and Bywaters and of Rubner on the formation of glycogen by yeast cells in strong sugar solutions constitute probably a further instance of the same phenomenon.? In general loss of water will 1Cf. the papers of Pottevin: Comptes rendus de l Académie, 1903, Vol. 136, p. 1152, and 1904, Vol. 138, p. 378; Taylor, Journal of Biological Chemistry, 1906, Vol. 2, p. 87; Hamsik, Zeitschr. f. physiol. Chemie, 1909, Vol. 59, p. 1; Armstrong and Gosney, Proceedings Roy. Soc., Ser. B, 1914, Vol. 88, p. 176. 2 Overton, Vierteljahrsschrift d. naturf. Ges. in Ziirich, 1899, Vol. 44, pp. 131-2. 3 Butkewitsch, Biochem. Zeitschr., 1908, Vol. 10, p. 314; of. pp. 336 seq. 4Pavy and Bywaters, Journal of Physiology, 1907, Vol. 36, p. 149; Rubner, Archiv fiir Physiologie, Suppl. 1912, p. 252, and zbid., Vol. for 1913, p. 244. Pavy and Bywaters found that in pure dextrose solutions the deposition of glycogen in yeast cells increased rapidly with increase in the concentration of dex- trose up to an optimum. In 2 per cent. solutions there was little effect; in 4 per cent., 8 per cent., and 16 per cent. solutions there was a rapid progressive increase in the quantity of glycogen laid down in the cells toa maximum of over 13 per cent. 302 RALPH 5S. LILLIE. be favorable to—since it will supplement—the action of any dehydrating mechanism; and it is possible that in the sea- urchin egg after membrane-formation the intracellular dehydra- tion-processes are by themselves not quite energetic enough to effect the syntheses necessary for initiating development, but become so when supplemented by the action of the hypertonic sea-water; 7. e., this agent has the effect of reducing the con- centration of water at the locus of the reactions sufficiently to enable syntheses to take place which otherwise are impossible under the conditions. It is significant that cell-division is started in the sea-urchin egg by simple membrane-formation, but fails to continue,—just as if there were some failure in the supply of the necessary constructive materials; partial abstrac- tion of water rectifies this condition. Since oxygen is necessary to this corrective process, we may assume that the syntheses belong in part to the class designated by Schmiedeberg! as oxidative syntheses. From this general point of view the action of hypertonic sea-water becomes in a measure theoretically intelligible and ceases to be merely a detached empirical fact. Certain avenues of experimental approach to the problem are also suggested. SUMMARY. §1. The effects following exposure of maturing unfertilized starfish eggs to high temperatures (29-36°) vary in a constant manner with the times of exposure as follows. Below a certain minimal duration of exposure to any given temperature (e. g., 32°), no visible change is produced in the egg; slightly longer exposures induce the formation of typical fertilization-mem- (as compared with about 5 per cent. under normal conditions) ; in more concentrated solutions there wasa decline. They also found that too long exposure to a favorable solution (10 per cent.) was unfavorable; thus yeast incubated in ro per cent. dextrose for 25 hours showed an increase in glycogen-content from 4.84 per cent. to 11.66 per cent.; four hours later there was a decline to 9.33 per cent. These facts show a suggestive parallel with the effects of hypertonic sea-water on sea-urchin eggs; here also there is no effect until a certain minimal osmotic pressure is reached; with further increase in osmotic pressure there is a rapid increase in favorability up to an optimum; still further increase is unfavorable. Also for a favorable con- centration there is at any temperature a definite optimum time of exposure. 1 Cf. Schmiedeberg, Archiv f. exper. Pathologie u. Pharmakologie, 1893, Vol. 31, p. 281. Se ACTIVATION OF UNFERTILIZED STARFISH EGGS. 303 branes, but the eggs fail to cleave and soon break down without development; in order to induce favorable development an exposure of three to four times the minimum for membrane- formation is required (e. g., 7-8 minutes at 32°); more prolonged exposures are again followed by failure to develop. 2. Between 29° and 38° the times of exposure required to produce these effects decrease very rapidly with rise of tempera- ture; on the average a rise of 1° approximately halves the exposure required for a given physiological effect (such as membrane-formation, or complete activation, or heat-inactiva- tion). The activation-process thus exhibits a characteristically high temperature-coefficient (Qi9 = 200-400). 3. The effects of exposure to weak butyric acid solution (7/260) vary with time of exposure in a similar manner,—brief exposure causing membrane-formation followed by breakdown, longer exposures causing cleavage and development to larval stages, and still longer exposures causing cytolysis without development. 4. The inference is that the same process is initiated in the eggs by exposure to warm sea-water as by fatty acid solution. This process must proceed to a certain stage in order that activation may be complete; if arrested too soon (brief exposure) only partial activation (membrane-formation followed by break- down) results. 5. Eggs in which membranes are formed by minimal exposure to warm sea-water or 2/260 butyric acid, followed by return to sea-water, may be made to develop favorably by a second treatment with either warm sea-water or fatty acid solution, as well as by after-treatment with cyanide-containing or hypertonic sea-water. A favorable after-treatment may thus be of the same kind as the membrane-forming treatment. 6. The temperature-coefficient of activation by high tempera- tures is of the same order as that of the melting of gels or the decrease in the viscosity of gelatine solutions. The above high temperatures thus probably act by producing degelation-effects in the surface layer of the egg; increase of permeability, with consequent depolarization, is the result of this change. 7. A new hypothesis of the mode of action of hypertonic sea-water is put forward. DIVISION RATE. IN SCILIADE~ PROMOZOAR AS INFLUENCED BY THYROID CONSTITUENTS? ROBERT A. BUDINGTON AND HELEN F. HARVEY. INTRODUCTION. In the very numerous studies which have been made to ascer- tain the effect of thyroid tissues and extracts on growth and differentiation, the material employed, whether used as a food for large organisms or as a component of a medium in which to breed smaller forms, has been taken in a very large proportion of instances, if not always, from some mammal, e. g., cow, horse, or sheep. This has been the case even though the animal under observation may have been a mammal, a bird, an amphibian, or a protozoan. Assuming that the doctrine of evolution is a fairly probable hypothesis it is only a natural if not necessary corollary that each of the several organs involved, as well as the organism as a whole, has experienced its own successive changes, its own evo- lutionary modifications. Variations, ‘‘continuous’” and ‘‘dis- continuous,’ have occurred in internal as well as in external organs, and these variations have involved the physiological value of the organs concerned, as well as their anatomy; so that, of glandular tissues, for example, the composition of the output has undergone phylogenetic changes, so to speak, during the process of descent of one phylum from another. It is a prior. improbable, of course, that the chemical composition, and conse- quent stimulating potency, of the thyroid secretion is the same throughout the entire vertebrate phylum. Apparently the earliest experimentation along the line with which this paper deals was carried out by Nowikoff ('08), who found that one effect of putting sheep thyroid into the medium in which Paramecium was living was to cause it to divide more rapidly than normally. Recently, Shumway (14) has published a paper in which he verifies Nowikoff’s contentions. Both these investigators, how- ever, employed mammalian thyroid; and, since our results agree 1Hrom the Department of Zoology, Oberlin Colleg>. 304 DIVISION RATE IN CILIATE PROTOZOA. 305 with theirs, the question for mammalian thyroid, at least, seems fairly well settled. Nowikoff’s work suggested to us the query whether or not his results could be taken as widely significant.!_ Our purpose has been, therefore, to add to the known facts along this line by ascertaining the influence of glands taken from each of the five main subdivisions of vertebrated animals, so far as they or sub- stances derived from them, affect certain protozoa; and, using division-rate as an index, to thus get a line on the comparative physiology of this gland. MATERIALS AND METHODs. Perfectly fresh thyroid glands were taken from the fresh-water sucker (Catostomus teres), the frog (Rana pipiens), the turtle (Cistudo carolina), the chick (Gallus domesticus), and the cat (Felis domestica), dissected as cleanly as possible from surround- ing tissues, and then dried by moderate heat; each was then ground to powder in a mortar, and the material then kept in vials till used. In the instance of the mammalian gland, fatty tissue was present in such amount that this was dissolved away by repeated washing in ether to bring the gland to such condition that it could be finely pulverized. In supplying thyroid material to protozoa in this form, we endeavored to avoid any alteration in its character such as might result in the making of glycerin or alcoholic extracts. This seems a point which should be rather carefully guarded. The forms employed for experimentation were Stylonichia and Paramecium. ‘Yo familiarize ourselves with a method of hand- ling such organisms, and also for the purpose of securing indi- viduals whose ancestry would be known to us, we first carried isolated “‘wild”’ forms through a considerable number of genera- tions (in the case of Stylonichia, seventy-four); we employed depression slides, kept in a moist chamber, each slide carrying four drops of bacterial hay infusion made up in the manner of that used by Woodruff (’05) in much of his work. The “wild”’ specimens were taken from ordinary laboratory cultures, but the particular individuals used in any given experiment were taken from the pedigreed lines descended from a single parent. The 1 Qur work was completed before Shumway’s article appeared. 306 ROBERT A. BUDINGTON AND HELEN F. HARVEY. protoplasm of the line treated with thyroid was identical with that of the control carried beside it. The procedure in any experiment was this: two protozoa of common parentage were isolated, each in four drops of the same culture medium. To one of the slides was added a minute mass of pulverized gland, which would thus influence the protozoan either as a food, or as a factor in the environing medium so far as this acted as a solvent.! The actual amount of each pulverized gland thus used was small, and a like amount of each was deter- mined as closely as possible by careful subdivision of a slightly larger mass on a clean paper surface. To weigh out the powder would give no more equal amounts, inasmuch as the glands are so invaded by vascular and connective tissue that any moiety taken might easily contain more or less of other than glandular material. A slight amount of fresh hay infusion was added to each slide each day, and the experiment continued six days or more. The results given in this account are limited to those obtained during the first six days only; to keep track of the off- spring of even a single protozoan longer than this is extremely difficult, as many know. The effect of each different gland was tested by three trials. Circumstances were such that it was not always convenient or possible to run experiments with all five different thyroids at one time, so a control was carried along beside the gland-fed individual in each case. This assured that the same conditions of every sort attended both experimental and control lines, no matter when the observations were made. If any circumstance favored or interfered with either, the same was true for the other. EXPERIMENTAL FINDINGS. The following tables show the exact results, so far as number of individuals resulting from division of the original one goes, this rate of cell division being the only index of thyroid effect at present ready for presentation. The ciliate used in the first series of experiments was Stylonichia; in the second and third series we used Paramecium. While the evidence is too limited to permit any rigid conclusion of the kind, the data at hand seem 1 Shumway states in his recent paper, loc. cit., that more or less of the material thus offered Paramecium is actually ingested and digested. DIVISION RATE IN CILIATE PROTOZOA. 307 to indicate that Paramecium is rather more susceptible to thy- roid ingredients than is Stylonichia. RESULTS FROM USE OF FISH THYROID. First Experiment, Second Experiment, Third Experiment, No. of Individuals. No. of Individuals. No. of Individuals. Control. |Thyroid-fei.) Control. |Thyroid-fed.| Control. |Thyroid-fed. Ist day T I I I I I 2d day 4 2 I 2 I I 3d day 6 6 3 2 6 6 Ath day 15 15 5 7 TA 16 5th day Dy Ba 7 50 18 BE 6th day 43 49 15 90 52 73 RESULTS FROM USE OF AMPHIBIAN THYROID. First Experiment, Second Experiment, Third Experiment, No. of Individuals, No. of Individuals. No, of Individuals. Control. |Thyroid-fed.| Control. |Thyroid-fed.| Control. |Thyroid-fed. Ist day I I I I I I 2d day I it I 2 I 2 3d day 6 5 2 4 8 18 Ath day 7 Tit 8 9 16 76 5th day 7 12 a0) 30 25 127 6th day 12 30 12 (op 36 243 RESULTS FROM USE OF REPTILIAN THYROID. First Experiment, Second Experiment, Third Experiment, No. of Individuals. No. of Individuals. No. of Individuals. Control. |Thyroid-fed.} Control. |Thyroid-fed.} Control. |Thyroid-fed Ist day I I it I I I 2d day 4 4 4 5 I 2 3d day 8 Io 8 23 3 a Ath day 8 Io 8 47 I5 42 Sth day 8 21 14 148 23 80 6th day 8 38 15 362 50 276 RESULTS FROM USE OF AVIAN THYROID. First Experiment, | Second Experiment, Third Experiment, No. of Individuals. No. of Individuals. No. of Individuals. Control. | Thyroid-fed. Control. |Thyroid-fed.} Control. |Thyroid-fed. Ist day I I I I I I 2d day I 4 if 2 2 I 3d day 2 9 3 7 8 16 4th day 2 a0) 4 45 23 32 5th day 4 15 15 63 59 141 Cult Gene A 24 35 243 QI 399 FIsH. 450 308 ROBERT A. BUDINGTON AND HELEN F. HARVEY. RESULTS FROM USE OF MAMMALIAN THYROID. First Experiment, | Second Experiment, Third Experiment, No. of Individuals. No. of Individuals. No. of Individuals. Control. |Thyroid-fed.| Control. |Thyroid-fed.{ Control. |Thyroid-fed. Ist day I I I I I I 2d day 13 I5 4 8 2 2 3d day 16 18 6 18 9 18 Ath day 18 20 IO 48 34 590 5th day 30 132 I4 60 56 129 Gt hidays ees 306 29 use) i G0 487 For convenience in seeing at a glance the relation which held between the experimental lines and the controls during the use of any particular kind of thyroid, the data for the three experi- ments with each thyroid have been averaged, the controls for the same averaged, and the results placed in curve form. These follow: ° ° ° ° ° ° ° ° ° 1n ° 1n (o} Ye} to) In + ise) Sp) N NQ H 4 STeNprAIpuy 1 An erroneous conclusion is rather easily drawn from these charts, for at first glance it appears that the potential of the gland increases by steady gradation from the fish up to the mammal. When figured as percentage increases of the experi- mental over the control lines, it is found that such is not the case. Data pertaining to this relation are now being collected. Day 1 those treated with fish thyroid; Showing number of individuals at end of each CHART I. 309 DIVISION RATE IN CILIATE PROTOZOA. *prorAyy ueryiydor yqIM peindses }[Nso1 SUIMOYS 4Nq ‘I JIeYD Se IOUULUT oUIeS UT pd}10[q 9 S v € Zz I Aeq "€ LAVHD om 3 A. = Sie A = a n *pIorAy} uerIqiydure UDATS SEM STENPIATPUT J9}SIS OA\} JO oUO UAYM qMsor SUIMOYS JNq ‘I JIeYD Sse IoUUeUT oUTeS UI poz10[g °Z LUVHD 9 s v € Zz I Aeq ‘NVIGIHdNY os oot OST 002 s[enprArpuy oSz oo€ oS€ ooV ROBERT A. BUDINGTON AND HELEN F. HARVEY. 310 yy s: 9S W[NSeI Sur oys 4nq ‘I y1eYD se JeuUeU aUTeS UT poq}o[g “S LUVHS oS oor OSI 00d sfenprarpuy oSz2 oo€ oS€ oov os _- 4yNser Surmoys IUIeS UT p9}0TgG t O} oS oor oSTt 00g oSz oo€ oSé oov oS STeNPIATpUy DIVISION RATE IN CILIATE PROTOZOA. 3II A curve plotted from the averages of all five of the different experimental lines and drawn beside a curve portraying the averages of the control lines for the same periods, each figured day by day for the six days, represents, in a manner, a generalized curve of the effect of vertebrate thyroid constituents on protozoa, as based on our data. This curve takes the following form: 300 iS) On (o} Individuals 150 Too 50 CuaArt 6. Acurve plotted from the averages of Charts 1, 2, 3, 4 and 5, and thus representing a composite of results obtained from use of thyroids from each of the five main classes of vertebrates. DISCUSSION. The number of papers which have hitherto been published along this line seems to be limited to those already mentioned; so that conclusions have to be drawn from a comparatively meager literature, and to gather largely around the investigator’s 312 ROBERT A. BUDINGTON AND HELEN F. HARVEY. own experience. There is essential agreement between the findings of Nowikoff, Shumway, and ourselves as to the constant effect of thyroid ingredients in increasing the division rate of protozoa beyond the normal, at least so far as Paramecium is concerned. The work of others, notably Gudernatsch (12, ’13), on the feeding of amphibian embryos, seems to indicate that the effect there observed is mainly one of acceleration of differentiation of tissues in the growing organism; at least this is the interpretation given their findings. West (14) has verified certain features of Gudernatsch’s results. A similar betrayal of specialization in function would, of course, not be possible within the limits of a unicellular organism. It seems entirely probable, however, that intra-cellular modifications of the Paramecium protoplasm does accompany its feeding upon and living in a medium which, among other things, brings it hurriedly to its most crucial ex- perience, self-division. The fact that rapid fission of thyroid-fed Paramecia is accompanied by their increased activity and trans- parency, and by smaller size,! indicates that very important in- ternal modifications doubtless occur. Careful study of protozoa exposed to exigencies of this sort should be made. If cell division in protozoa is to be compared with anything in the life history of metazoa, it should certainly be considered be- side the early development of the metazoan egg. If the egg has already advanced to the proportions of an embryo or larva, and the precocious differentiation of tissues and organs in such is under consideration, the question arises: Is this differentiation at all explained in the same terms as is protozoan cell division, or does it involve the same basic factors? It seems to us that this query may very possibly be answered in the affirmative, for the reason that the sprouting out of legs from the tadpole and establishment of other organs characteristic of the adult, is surely not due to mere unusual division of labor among the young cells generally acting as little more than unit components of the infant tadpole body; but that these latter have been provoked (by thyroid ingredients?) to abnormally rapid division, probably 1 Shumway mentions these alterations to occur in thryoid-fed Paramecia, and we have found such to be practically always observable. DIVISION RATE IN CILIATE PROTOZOA. 313 with accompanying abnormally small size, and that entirely normal differentiation has set in among cells which have been de- rived by the abnormally early (rapid) multiplication of their _ ancestors. We would suggest, therefore, that there may be, at bottom, not any great difference between what shows itself in Guder- natsch’s work as differentiation, and the result which shows itself as cell division in an animal where differentiation, so far as it exists, can assert itself only intra-cellularly, and thus in a very obscure manner. It is certainly entirely unnecessary to dwell upon the obvious fact that the more nearly adult a metazoan animal is, the more difficult it becomes to even suggest parallelisms which may exist between it and unicellular organisms; so that, to discuss the numerous physiological effects which have been obtained from feeding thyroid tissues to various vertebrata, or from grafting and transplantation experiments, or to examine the studies of conditions in higher types provoked by pathological thyroid growth and disease, is quite beyond the scope, if not impossible in connection with the subject, of this paper. SUMMARY. The conclusion to which the foregoing experimental results point is that thyroid ingredients, no matter from what class of vertebrates the gland be taken, produce essentially the same result when given to ciliate protozoa (Paramecium and Sty- lonichia) as a food or as a factor in the medium in which they live, viz., increased division rate. : The tissue which has hitherto been used in experimental work along this line has, we believe, always been taken from mammals. We think it safe to say that, no matter how far apart taxonomi- cally, or how distantly related phylogenetically the “higher”’ and “‘lower’’ members of the vertebrate phylum may be, certain physiological qualities in the thyroid glands are constant and similar in all. Sufficient difference exists between the potential of the thyroid secretion of one vertebrate class and that of another, so that, if studies of the normal value of this gland are being made, glands 314 ROBERT A. BUDINGTON AND HELEN F. HARVEY. from the same class, if not from the same genus and species of animal as the one under observation, may well be employed. BIBLIOGRAPHY. Gudernatsch, J. F. 12 Feeding Experiments on Tadpoles, I. Arch. f. Entwick., XXXV., 3. ’13. Feeding Experiments on Tadpoles, II. Am. Jour. Anat., XV., 4. Nowikoff, M. ’08 Die Wirkung des Schilddrusenextracts auf Ciliaten. Arch. f. Prot., XI., 2. Shumway, W. "14 Effect of Thyroid on Division Rate of Paramcecium. Jour. Exper. Zool., SOWA 5 Be Woodruff, L. L. 705 An Experimental Study on the Life History of Hypotrichous Infusoria. Jour. Exper. Zool., II., 4. West, P. A. ’14 Experiments in Feeding Tadpoles. Science, N. S., XX XIX., 106, p. 918. AN EXPERIMENTAL STUDY OF THE MOVEMENTS OF HERRING AND OTHER MARINE FISHES! VICTOR E. SHELFORD AND EDWIN B. POWERS. VES Albay ereCO Ye HANS Voto Var5 oie areal onanG 6 dl odo cho ctoleintced Mae amar oech opie eo. Orersiecs Site aais Ores Buns ite VaterialtandsWMiethod sh ci sxe ctcnsae Gots ata euse daly caanarscteigiau ts Bit 7 Tepe sStO CKO MtiGlle ais ele sccm mye scp elcens czas stellt to Speanvien eat tay epeancr Sect te tenes capes Bitly By TMS WAU? Soyo Oy Oi WAS GUATON, as oon0doo0accebovovouupvoecoge 317 III. The resistance of fishes to contamination and decomposition products.. 320 IV. The reactions of fishes to chemical conditions in sea-water............ 323 i Comebiooms aincl imvenaorls OF GeWChYs co cnacucsoccoacbcvuccavsguouce 323 5. AVERY NES CAE DIRS Soe at keene! Sen etna eet auronan Pmt RneRic te roi coral stiches fa anus its 325 Beeb GROLEMY SUG Ee ee eracecsicyecs kc ete. ean car Schad ccs Aveeno Ue ee Pan tal ee 320 AmSalinityvzandsallkalimnityaandeacicityes ieee ier rien tie erie 326 icy OGRE Wiser Beto en ORE CCRC ORS HES ER Ce heer ten ore Ceaae cen este ran deescnctne Oke 331 Wo Sumimemy ancl Drscission OF Comeliisiome, 55500cccncvnc0ccgacccucc0s 332 Wis Acknowledgements and! Bibliography..-..-755..22.2-225-4++++5s5-5: 333 I. INTRODUCTION. The general problem of increasing the supply of any species of fish or any other aquatic food animal, or of maintaining such species against extensive catch and against pollution of waters with sewage and the waste products of manufacturies, is very complex. The older methods of study are as important now as ever. The study of the food of an animal, its relation to its natural enemies, and its breeding habits still must receive their proper share of attention. In addition to these we now know that attention must be given to the chemical condition of the water, its effect on the movements and migrations and general health of the animals. Likewise it is especially important to study the physical and chemical conditions in which the animals breed and to look especially into the matter of the preservation of the natural breeding grounds. It is well known that one of the reasons for the depletion of the white-fishes in Lake Michigan is the destruction of their breeding grounds by the addition of sewage, saw-dust and other refuse to the water, which has settled on the breeding grounds and rendered them uninhabitable 1 Contribution from the Puget Sound Marine Station. 315 316 VICTOR E. SHELFORD AND EDWIN B. POWERS. by the lowering of the oxygen content and covering the surface with materials which bury and tend to smother the eggs during development. The number of individuals of a species is never any greater than the breeding grounds can support. Finley (13) has shown that the number of prairie chickens in certain counties of Illinois is directly proportional to the area of breeding grounds. Likewise the senior author (Shelford, ’11) has shown that in a series of ponds at the head of Lake Michigan, food fishes are absent where their food is greatest in quantity because the breeding conditions are absent, due to the covering of the bottom with the decaying food of fishes. It is especially note- worthy that the food of the youngest fishes is especially abundant in ponds where the best food fishes cannot breed. This is not due to the failure of young fishes to destroy the small crustacea, because the same principle holds for ponds in which there are as many crustacea-eating fishes in stages suitable for food fishes as in stages suitable for only non-food fishes. The economic justification for the study of the movements of fishes is two fold. First experimental studies are concerned with the question of the conditions which the fishes select or reject when presented with two or three kinds of water to which they have free access under experimental conditions. Their importance in this connection is based upon the fact that so long as we are concerned with conditions which the fishes habitually encounter in nature, the selections or rejections represent in a general way the physiological character of the fishes and as a rule conditions which fishes reject are detrimental if continued for a long time. Thus, as we shall see later, fishes turn away from water contain- ing hydrogen sulfide and we will show further that they die very quickly when exposed to only a small excess of this gas in the sea water. Here then the fish is so constituted that its behavior and safety are intimately linked. Of course there are exceptions to this rule and it does not hold when we are concerned with changes in conditions which are not commonly encountered in nature. Thus we learn something of the conditions that are probably deleterious to the animals without either killing them or breeding them continuously under the modified conditions. The second justification lies in the fact that we can learn by MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 317 such experiments the effect of advancing civilization and industry upon the presence or absence of a species in any locality. The movements of the fishes must be known as well as the cause -therefor, before we can intelligently approach the question of capturing them in quantities. Il. MATERIAL AND METHODS. 1. Stock of Fishes. The material used in these experiments was chiefly the fry of the herring (Clupea pallast Cuvier) 6 cm. (2% in.) common in Puget sound. The fry were caught on July 2 and were kept in a float car anchored ina good tide and until July 22 practically none of the fish died except during the first few days when those probably injured in catching were the chief victims. A few soles (Lepidopsetta bilineata Ayres) secured on July 4 at Fisherman’s bay, Lopez Island, were keptin the same car. A few young hump- back salmon (Oncorhynchus gorbuscha Wal.) 7 cm. (234 in.) long were secured at sea, through the courtesy of Dr. E. Victor Smith on June 30, at Turn Island and were not used after July 8 as they did not appear to be in normal condition after that date. A single Cottid (Oligocottus maculosus Girard) was used in killing experiments. The soles and herring appeared to be in essentially as good condition at the end of the period of work as at the beginning. 2. The Water Supply of the Station. Experiments were run in which both fresh and salt water were used. Thus it is necessary to consider the character of both. The frésh water in use during the summer of 1914 was supplied by the village of Friday Harbor and came from deep wells. Owing to the rocky character of the ground in the vicinity, it was impracticable to bury the pipes and the temperature varied greatly with the weather, night, day, etc. The highest tem- perature noted was 24° C. The water contained an excess of gas which escaped in a cloud of bubbles when it was withdrawn from the tap. This was neither oxygen nor carbon dioxide and gave no odor which points to the conclusion that it was nitrogen. The water was distinctly alkaline to phenolphthalein, free carbon 318 VICTOR E. SHELFORD AND EDWIN B. POWERS. dioxide being wanting. The half bound carbon dioxide was 24.2 c.c. per liter and the fixed 28.6 c.c. The oxygen was less than 0.5 c.c. per liter (for methods see Birge and Juday, ‘11, pp. 13-21). Such water is unsuitable for biological purposes and was used in these experiments only after aération by running it slowly over an inclined board ten inches wide and four feet long. After this aération the oxygen content was, at 13° C., 4.9 c.c. per liter and the excess of other gas was removed, but the water still remained alkaline. The sea water supplied at the station building was pumped from a depth of about four feet below mean tide. It was retained in a wooden tank, being pumped twice per day, in the evening and in the morning. Upon standing in the tank the temperature rose from 11° to nearly 15° on warm days. The oxygen was determined by the Winkler method. In no case was the sea water from the tank or from the bay from which it was pumped, saturated with oxygen even in samples collected at the surface. The only surface collection made that showed saturation according to the tables of Fox (see Murray and Hjort, "12, p. 254) was from the strong tide rips off Point Caution at 5:30 P.M. Collections from the bottom of sandy shores among Ulva were super-saturated. , Chlorine was determined by titrating with silver nitrate. It usually amounted to about 16.93 grams per liter. It was usually a little higher in water from the tank than in water collected from the sea. The determination of carbon dioxide was made by the method in common use in fresh water. The sea water was titrated with 3'5 normal solution of sodium carbonate, with phenolphthalein as an indicator. The water was usually ‘acid in reaction indicating about 1.7 c.c. per liter of free carbon dioxide. The half bound and bound carbon dioxide as indicated by the method used by Birge usually amounted to 25.3 c.c. per liter each. There was considerable uniformity in the results of such titrations and while the method is not especially accurate the lack of oxygen common in the water would indicate an excess of free carbon dioxide over that commonly reported for sea water. The correctness of these figures is further suggested by the slight alkalinity of the water taken from the vicinity of green alge and containing an excess of oxygen. MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 319 Hydrogen sulfide is very commonly present in sea water when decomposition is taking place. This was determined by titration with iodine which was the only method we were equipped to employ. It is never present in any quantity in freely circulating waters. The highest records are for collections made near the bottom under Ulva, where the odor is often quite distinct. On account of the probable presence of other substances which may absorb iodine the determinations may be slightly too large (Birge and Juday, ’11). MABLE I: THE DISSOLVED GASES OF THE SEA WATER ABOUT F RIDAY HARBOR, WASHINGTON. DATA IN C.C. PER LITER. Date. Place. Hour. Tide. Collected. | COg.| Og. | HoS.| Temp. Point Caution 5:30 P.M.| Low, in | Surface 5.6 7/23 | N.E. Brown’s Id. |10:10 A.M.| Low, out | Surface 1.76] 4.9 |.187| 11.6 7/25 Do. 11:10 A.M.| Low, out | Surface |1.64| 4.6 |.237 10.7 4/25 Do. 7:15 P.M.) High, Surface [1.91] 4.6|.268] ro.5 7/26 | S. Brown’s Id. 10:45 A.M.| Low, in | 8’ under |o.00| 9.2 |.536 10.5 Ulva 7/23 Do. 12:00 M. | Low, in Do. 0.00 |10.8 |.536| 13.2 7/25 Do. 12:00 M. | Low, in | 18’ Do. |o.00| — -339| 13.2 7/25 | Station dock tr:10 A.M.| Low, in | Surface [1.86] 5.2 -149|] 11.6 7/25 Do. 12:45 P.M.| Low, in | Surface |3.10| 4.8 -205 7/26 Do. 9:30 A.M.} High, out} Do. 1.81] 4.2|.205|} 10.6 7/26 | Tap-pumped at 6:30 A.M.) Med. low| 4’ deep |1.76| 4.7 .223 j It will be noted from a study of the table that the water from Point Caution where the tide has full sweep is the only water saturated with oxygen at the surface. In other places the sea water at the surface is about 1 c.c. less than the amount given by Fox (see Murray and Hort, ’12, p. 254). Aérating the sea water increased the oxygen. The water from the tank did not seem to have been modified by standing for sixteen hours or more. On the whole there must be much decomposition in Puget Sound waters. There was no constant difference between the water from outside and inside the side of the island which encloses Friday Harbor. The COs is a little higher except at low tide in the sample taken near the Ulva; the oxygen remains about the same. The hydrogen sulfide does not average appreciably higher. The explanation for the alkaline character of the water under the Ulva is that the plants take up the CO, and give off oxygen and thus remove the excess which occurs in other localities. 320 VICTOR E. SHELFORD AND EDWIN B. POWERS. The absorption of oxygen in connection with the development of the hydrogen sulfide probably prevents any very great excess of CO, from accumulating (Lederer ’12). III. THE RESISTANCE OF FISHES TO CONTAMINATION AND DECOMPOSITION PRODUCTS. It was not possible to try the resistance of the fishes (Wells, 13) to the effect of the lack of oxygen either separately or in combination because no means of removing it was at hand. It was possible only to add gases to the water. Hydrogen sulfide and carbon dioxide were used. 1. Herring (Clupea pallasit Cuvier). Hydrogen sulfide is extremely poisonous to the fishes (Weigelt, 03). In the first attempted gradient experiments where the water at one end contained only a little of the gas the fishes turned on their backs in two or three minutes when the one inflow was showing 8.3 c.c. per liter and the other was pure sea water. This happened in spite of the fact that more than half of the time was spent in the end with least H2S. The experiments were performed in the manner described by Wells. When placed in a solution of 7.6 c.c. per |. the herring gasped after 1 minute and 45 seconds, turned over after 5 minutes, and were apparently all dead in 6 minutes. In carbon dioxide of about 20 c.c. per |. the herring showed evidence of loss of equilibrium after three minutes. Some of them sank to the bottom after 12 minutes. After 39 minutes to 62 minutes herrings turned on their sides on the bottom, resting in this position for a time and then swimming more nearly normal for a time again. One died after 102 minutes, the others after 159 minutes’ exposure. The oxygen was about 5.5 c.c. per |. and varied directly with the amount of tank COs: used, indicating that the carbon dioxide contained much oxygen. When carbon dioxide and hydrogen sulfide were used together the carbon dioxide was about 30 c.c. per 1. and the hydrogen sulfide 2.9 c.c. per 1. The amounts were controlled with some difficulty and thus the experiments are not alike in the matter of concentration. Herring were much stimulated at the beginning. ain, MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 321 After I minute and 30 seconds there was a general loss of correla- tion of movements. At the end of four minutes all the herring were dead. Thus we note that the combination of hydrogen sulfide and carbon dioxide is exceedingly deadly. In alkaline partly aérated fresh water herring showed loss of equilibrium in from 10 to I4 minutes. They nearly succumbed and then recovered a few times, the first one dying after 30 minutes and all being dead in 44 minutes. 2. Soles (Lepidopsetia bilineata Ayres). In the hydrogen sulfide (7.6 c.c. per 1.) the soles showed some signs of loss of equilibrium at the end of one minute. In 5 minutes they were on their backs. After 13 minutes they had revived again. They were nearly dead after 16 minutes and all dead at the end of 24 minutes. In the carbon dioxide (20 c.c. per 1.), after 45 minutes one sole gasped, which was the first sign of any disturbance and one turned on its back after 54 minutes. For three hours this was repeated at intervals and each gasping time was followed by recovery. In the combined carbon dioxide (30 c.c. per 1.) and hydrogen sulfide (2.9 c.c. per 1.) the soles lost equilibrium after 2 minutes and 30 seconds. In I1 minutes they were barely alive and in 13 minutes were dead. In fresh water the soles showed stimula- tion at the end of 3 minutes. They died in from 48 minutes to one hour. 3. Cottid (Oligocottus maculosus Girard). One fish of this species was added from curiosity but the results were sufficiently surprising to record. In the hydrogen sulfide the cottid seemed unaffected until the end of 6 minutes, after the herring were all dead. It breathed heavily after 16 minutes. The fish was alive at the end of three hours when it was returned to running salt water, and allowed to recover, after which it was used in the carbon dioxide experiment, with similar results. In the combined carbon dioxide and hydrogen sulfide it was not visibly affected and in fresh water there was no evidence of any disturbance. These fishes were seined from the sandy bottom among the Ulva, coming in with numbers of the small soles. 322 VICTOR E. SHELFORD AND EDWIN B. POWERS. 4. Summary. We note that on the whole the presence of a quantity of carbon dioxide in the water affected the fishes less than a smaller amount of hydrogen sulfide. The combination of hydro- gen sulfide and carbon dioxide was most rapidly fatal. Since decomposition yields CO: and consumes oxygen and is accom- panied by the production of hydrogen sulfide which is also accom- panied by the consumption of oxygen, it is reasonable to suppose that on a bottom from which vegetation is absent and decom- position actively takes place a fatal combination of lack of oxygen, and presence of hydrogen sulfide and probably carbon dioxide can develop quickly. Considering the fishes tested we note that the herrings were most sensitive. They were sharply marked off from the bottom species which are resistant to a marked degree. This resistance is in a very general way associated with the habitat preference of the species. Still the marked resistance of the small cottid is not quite explicable on this or any other basis. The importance of factors which kill fishes is greatest in the early stages for two reasons. First the small size of the eggs and embryos makes the ratio between volume and surface smallest and thus any substance in solution will reach all parts of the organism at a more rapid rate. Secondly the inability of the eggs and embryos to move about makes them the easy victims of any adverse conditions that may occur. The eggs of the herring are deposited on the bottom. Nelson mentions rocks only (Marsh and Cobb, ’10, p. 46) and rocks are usually swept fairly clear of organic matter and the water well aérated down to the depth of one fathom where the fishes breed. If this means that sandy bottoms of bays are avoided it probably means the avoidance, during the breeding, of water high in hydrogen sulfide (see table) which would be fatal to the eggs and small herring fry to a greater degree than to those studied, which were 6 cm. long. Sensitiveness to hydrogen sulfide is a matter of much importance from the standpoint of the suitability of a given arm of the sea for herring and the influence upon fishes of contamina- tion of the shores with refuse from the land. Carbon dioxide is not high in such shallow water on account Pe MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 323 of the presence of so many green plants. Carbon dioxide is probably more important in connection with movements of the fishes than in the matter of restricting their breeding places. IV. REACTIONS OF FISHES TO CHEMICAL CONDITIONS IN SEA-WATER. 1. Conditions and Methods of Study. The experiments were performed in a gradient tank. Water of two kinds was used in the experiments. One kind was allowed to flow into one end at a definite rate and another kind into the other end at the same rate. It flowed out at the middle at the top and at the bottom so that the two kinds of water met at the center. The outflow at the center did not of course prevent the mixing of the two kinds of water and thus the middle section, equal to one half or one third of the tank was a gradient between the two kinds of water. The tank used in these experi- ments was 122.3 cm. by 15 cm. by 13 cm. deep. The front wall was of plate glass and a plate glass top was used at times. Water was allowed to flow in at both ends at the same rate (usually 600 c.c. per minute) through tees the cross bars of which con- tained a number of small holes. The cross bars of the tees were at the center of the ends of the tank behind screens. The drain openings were located at the center near the top and in the bottom. The outer openings of the drain tubes were at the level of the water in the tank. The water flowed in at the ends and drifted toward the center and flowed out through the drains. We found no evidence that fishes react to the slight current thus produced. Since each half of the tank held about 9 liters, it required 15 minutes to fill it or to replace all the water in one of the halves. The tank was enclosed under a black hood. Two candles (electric lights being wanting) were fixed in the rear and above the center of the two halves, 7. e., above a point midway between the screen partition and the center drain. The light was 15-20 cm. above the surface of the water which was 13 cm. deep. The room was darkened during the experiments which were observed through openings in the hood above the lights or through the glass side late at night. Fishes do not usually note objects separated from them by a light. 324 VICTOR E. SHELFORD AND EDWIN B. POWERS. Water differing as little as possible from that in which the fishes usually live was used for control readings. Controls were observed and the conditions in the two ends of these were the same either because the water introduced at the two ends was alike or because no water was run into either end (standing water). In the controls the fishes usually swam from end to end in a rather symmetrical fashion, and thus by comparing these move- ments with those occurring when the fishes encountered differ- ences in water, we are able to determine the reactions of the fishes to the differences. Various kinds of water were used at one end as follows: (1) water with varying amounts of carbon dioxide added; (2) water with oxygen added; (3) water with hydrogen sulfide added; (4) fresh water. When the difference between the solutes at the two ends of the tank was not great we found by chemical tests that the central portion of the tank was a gradient between the characteristic waters introduced at the two ends. Usually the end thirds were essentially like the inflowing water. When the difference in con- centration was great the region of the gradient was propor- tionally longer and the ends with the inflowing concentrations correspondingly shorter. When the difference in concentration was very great the entire tank was gradient. For an experiment a fish was placed in a dish containing enough water to barely cover it and set above the tank. When all was in readiness the fish was emptied into the center of the tank. Marks on the sides divided the tank into thirds. The fish nearly always swam back and forth, apparently exploring the tank. The movements of the fish were recorded graphically as‘shown in Chart I. For this purpose sheets of ruled paper were used. Four vertical double rulings corresponded to the thirds and two ends of the tank. Distance from right to left was taken to represent the length of the tank, vertical distance to represent time and the graphs drawn to scale. The width of the tank was ignored. THe graphs on the following pages are copies of the originals. Before or after the experiment, the headings of the sheets were filled with data regarding the kind, size, and previous history of the fish, the conditions in the tank, concentration of MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 325 the solutes and other significant data. The fish was observed continuously for twenty or more minutes. In order to maintain a constant flow the water was introduced into the tank by means of siphons from cans on the top of the hood with a 74 cm. head. Connected with one of the two cans was an inclined plank trough 420 cm. by 25 cm. for the purpose of aérating water before it entered the can if so desired. By the method just described it is possible to obtain unusually accurate data on the factors influencing the movements of fishes. According to Marsh and Cobb ('07) a great difficulty in the herring fishery is the erratic movements of the fish. Schools may visit a bay for three or four years, in succession, and then without any apparent reason, avoid it for a season or two _or altogether. Bertham (97) noted a possible relation between the abundance of these fishes and weather and suggests that climatic cause may have more to do with the failure of some branches of the fisheries than is generally believed. He attri- buted the failure of the fisheries of Cape Benton to the occurrence of severe east and northeast storms during the running season. It is not clear what the effect of such storms may be, but they chiefly affect the dissolved content of the water. Johnstone, ’08, page 246, says that it is now nearly certain that the shoaling migrations of the herring of Europe are to be associated with the salinity and temperature of the sea. It is évident from these experiments that acidity and alkalinity are more important than salinity and the solution of the problem will come from a careful study of the reactions of fishes along with a similar study of hydrographic conditions. 2. Reactions to Temperature. These fishes are remarkably sensitive to differences in tem- perature. We obtained good reactions with a difference of 0.6° C. in the length of the tank. Fair reactions were obtained with differences of 0.5° C. and since the fishes often turned around near the center it appears that they recognized a difference of 0.2° C. In graph 1, Chart I., we show the reaction of fish in a gradient of 0.6° C. (compare with graph 2—control). The fish was taken from sea water at 10.9° and the experiment performed 326 VICTOR E. SHELFORD AND EDWIN B. POWERS. at 12.8° and 13.2°. It will be noted that the fish showed a prefer- ence for the higher temperature. Eleven experiments were performed with herring and in seven cases the fishes showed a preference for the warmer water and in three cases for the colder. One did not show any marked preference. The differences were too slight to be of great significance in determining whether the fishes move into warmer or colder water but show a great sensi- tiveness. Thus temperature may play an important rdle in the movements of fishes. It will be noted by reference to the graph, that the fish moved into the colder water several times as if trying out the entire tank and then turned back periodically from the colder end. In the control where there was no flow or difference in tempera- ture the fish turned back from both ends at times but by chance as shown by other controls, turned a little more often from the end corresponding to the cold end of the experiment due perhaps to difference. 3. Hydrogen Sulfide. The animals turned back sharply from all concentrations not great enough to cause intoxication as shown in graph 3, Chart I. (compare with control graph 4). In this experiment the hydrogen sulfide was only 4.5 c.c. per |. and the fishes avoided it sharply and after trying out the tank turned about at a point where the concentration could not be more than one tenth of that at the ‘treated water end or about equal to that under the Ulva on the south side of Brown’s Island (p. 319). This experiment is typical of several and the fishes are thus seen to be able to orient with reference to an increase in the solute and to turn back from it very sharply. The control (graph 4) to this experiment is symmetrical, there being turning from both ends in equal number. It shows the reaction of the fishes when no stimuli are encountered in the tank. 4. Reactions to Salinity, Acidity and Alkalinity. As noted above, the fresh water of the laboratory was from deep wells and not good for biological work. It was alkaline, containing no free carbon dioxide, 24.2 c.c. per |. half bound and ihe MOVEMENTS OF HERRING AND OTHER MARINE FISHES. 327 28.6 c.c. per |. of bound carbon dioxide. There was a deficiency of oxygen and what was present was probably due to leaky pipes. It was only 0.5 c.c. per liter. It contained a large excess of some odorless gas which escaped in bubbles and was probably nitrogen. This water was aérated by running it over a board 420 cm. long, into the siphon bucket. This reduced the gas in excess to air saturation and raised the oxygen to 4.8 c.c. per liter. In the experimental tank the difference between the density of the fresh and salt water was so great that the fresh extended nearly to the opposite end at the top with very little mixing and the salt water occupied a corresponding place on the bottom. Thus there was a sharp gradient from top to bottom, but a very imperfect one from end to end. To avoid this siphons were inserted which withdrew water from each side near the bottom at a point one third the length from the salt end and from near the top at the same distance from the fresh end. This was found not to remedy the difficulty sufficiently and so a screen incline which extended from bottom at the salt end to the height of 8.5 cm. at the fresh end. Above this was another screen which was 8.5 cm. at the salt end, and which ran up to the surface of the water at the fresh end. This enclosed the fish in an inclined cage 8.5 cm. deep at the salt water end and 5.0 cm. deep at the fresh end. The fish moved back and forth in this at a distance of about 4 cm. from the lower screen. The gradient of salinity between the acid sea water and the alkaline fresh water was essentially as perfect as shown in the accompanying Fig. 1. By AS 4,41 ve Me Ss 4 Fic. I. Fic. 1. Showing the distribution of salinity in terms of grams of chlorine per I. in Roman and oxygen content in c.c. per 1. in italics; Al, alkaline; N, neutral; Ac, acid. consulting this figure it will be seen that the oxygen content was essentially the same throughout. ‘The salinity corresponded to 10.561 grams of chlorine in the salt water end to 6.45 grams in the’ fresh water end. The acidity to phenolphthalein reached 328 VICTOR E. SHELFORD AND EDWIN B. POWERS. 1 TG ee 3 Hos 4 5 Hons 6 v8 ae I = SS — == poss ‘ = a= \ == 1 = =} ———— == a — =e 1 eel ee == == SS — = = = = =.5 =ce == =I ————— i= Sane = a = == ' —— - = eS = -- == ; == == 1 = =) 2) i 25 a= Eee — ae (= Se =i = eer ; == == ——— —— = =" See ecg 25 iS SS — Ze Peet 0 ESE == : Sis ————— = 52) == = == 2 — z Ee 22 (22 2.2 = : = }) See SS SE Se UME == z2 SS SSeS = = 2 eS a bi ae a ee == : == SS = = = ——— = = == —= == 1S = == ian! = 5 Gant = — =e =e = t RSIS Ail hese met = —— eed ==) =e FE= = ; Seas nealeil = {ss SS S25 => =a == ee = 1 9 2 Ce wi 2 SF: 22) = == =e So) = SS BS eee == z= 32S NS Se = <= = (2: S55 == Se SS SS eee == = == i A == Unmet = =; = SSS Se S=— SEZ == _—— = 5 ae aaa Be c Bs _—S— ———— ——— ails NE a S02 SSE == = == A 22 SS SS SS 22 1S == = = = == 1 == ae es f — 2 i == i =S wen3: —— == - SE H = i Se = ' == ae == 2S = ‘ == : == ' == wel = aye == = f == ray! m= |@22 22 == H == 1 =e ' == ne == == = i = 1 == a0 == ' = 2 ee =s == == == ty Sees == 4 ae === == == 7 = = 4 SS i a= ' == | SS 22 a= : f= = H == 22 4 paper ily = == = = phe = == roe Sa ==} == =15=- =15= = i S= SS S5— Sz = = ' =e tte aR a 1 SVS ‘ == 2s Spe H = i Sue ash Sew Ee iy) aL Sue == ae = Des US eee ea a= jee =e =a) z = A Se Wt Sie 9 SS) ES Se == > SS = Vi See oe == a == == = =i ' Shee uM == = == Ste == = : n= at =a ail 22 (Sea = == =o = ' == net xs 8 == == 22 = PS 0 eiepere Sa (== 2 = == ——— 1 _ 1 = ft ore eo = at == = 1 = bh 0 SOS ) Fe Sie ====2 4 | = a =.= Wh eS Soa. GaSe 0S. et Va Sas = Sue SSS =f (== | i SW = SS o a=) Ss | = =a D a 1 t a= = == SS SS 2 SS ae == a == ese 22 sae 55) | ee Sasa} 6S SSS = == =e 0 == 22 S]( 0 2s SS = SS 52) See = 5) , 55 2225 22 J Ne = 2 eee —— =e My) wets S25 Se =— = 2) == Sue — i ft alts = a == } £2 z= —— = = , == — ae! — Z2 22 DS = aS 2: Se ae 5) Se l= Se 7 SS — Sk 7 SS — sk | fh eIKES = an SS aa ' aS ‘ == == ‘ =s= = SS eS ==s 9 ssf == see fs | aa = =i Sis Cares Cn =S ===. ! ' TSS = a = == -- 1 =e /= = = ——— Sue a= mt = i i =y= 1 Ss ( 4 = = == o So (een) = hg Son. = 1 ' == = p =s Se ' = Ss — == ' , a =2S00 2264 == ; j J) Be —— ' araee= = Ss aaa a { —— ae Hoan i 7 ———a —SS" ot f Sonat OD ee Be Roane Deeb OouO POPS 0 0552 AII (D) Oenae SOLIS a ooo s 05s So aemeaN ae aocsceU Mer soo Gnd op le os Fer AT4 (c) Individual and sexual VAT a tLLOTS Si pn ee Oe Ee een ae A418 Siro AS) gin sk bo 0 ohne 0 yb eid or Dae NORTE oie Ga a chioneeapeoie cio eee Go) 3 ccika etosare hoa 7c 420 (@) Dsxiearall GuANCHURS. oo 6c (Pe pome ne Bap se Reo eS Reo Das R cae oes eee 420 (OD) Thartennell SienCHwe. . soon eodaaedeee oles ued mens Nebo Gaede ste R a 420 Experiments to determine the location of the olfactory oreans see a) 420 The olfactory sense of Harpalus pennsylvanica.......-.++-+---2+2+ 2227s A28 (a) Effects with antennee UIE! Otte aiem seh ab eaduime danse >e2ssedeo0 430 (b) Effects with elytra and ebavers) UIUC! Ott; wo ono caer ooee area yp sce A31 (c) Effects with elytra and wings pulled off and pores on legs covered with PART > oc cad oro pn odo Seles Scola Sheets Bo.c oe eRe re 432 The olfactory sense of Harpalus caliginosus......-------++++2s seer trees 433 (Gy ibabectsiwithyancvennes pullediofi 2). 0. 433 The olfactory sense of Epilachna borealis.....---------- ++ ++ -22 ese s ste A34 (a) Effects with antenne pulled off........----------++s sets s scenes 434 (b) Effects with elytra and wings plilledtotie te aae cee 435 The olfactory sense of Chaulcognathus pennsylvanica.........--------+--- 436 (@mbtectsswitheantennas pulledt of yoo x. oy 5) Sol ea as 436 The olfactory sense of Passalus cornutus.....---+-++-+-+++2+s 0 soos a 436 (a) Effects with antenne pulled Off eR eae ea ees te mam Ons clo ors, A e-or5 0 437 The olfactory sense of Cotimis nitida.........---+- +++ ene 437 @) Effects with antenna pulled off... .. 2. 2+ +32 - 2H ae A437 The olfactory sense of Euphoria Sepulohralis. «2 aj.6 2 sai A437 (@\Etects withvantenne: pulled othe) -)-7.° 5-5 438 The olfactory sense of Cyllene robini@......--- +++ 2222 ttt tts 438 @) Effects with antenne pulled off.....-.-.------2 0223s ae 438 (b) Effects with elytra and wings plilled\ oft sclerotic 438 The olfactory sense of Leptinotarsa ro-lineala.....--- +--+ +000 439 (@)) Bifects with antenne pulled off2.-...--.-- 4... --) os. 440 (b) Effects with elytra pulled off and wings cut off...2....--:2-.---.-- 440 (c) Effects with elytra pulled off, bases of wings glued and pores on legs covered withivascliness ae seco: = cis noone 441 The olfactory sense of Epicauta marginata.......------+++s sess ttt tes A42 (a) Effects with antenne cut off......... es BRU ane am ge Mt 2s 443 (@ywEtectstwithvartemnce pullledl oft 9-55. 5 3 oc 17. Megalodacne heros 102 120 383 605 Colydiide...... 118. Coxelus guttulatus 93 50 LA® || BAS Rhyssodide .....|19. Clinidium sculptile I31 4o | aie Cienyicceane ak 20. Cucujus clavipes 107 104 aR || sa Mycetophagide .|21. Mycetophagus punctatus 165 135 379 679 Dermestide..... 122. Dermestes marmorata 90 80 570 740 Inistenidan. a. |23. Hister depurator 74 80 240 403 Trogositide..... 24. Tenebroides castenea 96 Lis 200 505 Eilateniciaen «alae: 25. Elater apicatus III 130 365 | 606 Buprestide ..... 26. Melanophila longipes 186 IIo Dis || Bird Lampyride..... 27. Lucidota californica it 7G) 123 AS || Sz Telephoride...../28. Chaulcognathus pennsylvanica | 308 157 445 | 910 \29. Podabrus comes 341 157 280 | 778 Malachide...... 30. Collops bipunctatus 120 IOI 248 | 469 Wucanidaesneeee 31. Platycerus quercus 158 160 536 | 854 32. Passalus cornutus 203 184 782 | 1,169 Scarabeidz..... 33. Canthon levis 62 180 724 | 966 34. Cotinis nitida 162 39 934 | 1,135 35. Euphoria sepulchralis 90 36 Org i) Wao 36. Osmoderma scabra I27 182 782 | I,091 Cerambycide....|37. Orthosoma brunneum 79 207 982 | 1,268 i38. Callimoxys fuscipennis 87 175 BuO | 97/2 39. Cyllene robinie IIo 40 620 779 Chrysomelide...!40. Leptinotarsa ro-lineata II5 130 A476 721 IBiuchicdaersa ae ee 41. Bruchus pist 65 137 352 554 Tenebrionide....|42. Tenebrio molitor 66 I45 462 673 43. Uloma impressa 80 im ge) 361 560 IMIGIONGBES se Gooe 6 '44. Cysteodemus armatus 133 39 172 45. Epicauta marginata 157 94 504 755 46. Epicauta pennsylvanica 125 Too 440 665 Rhipiphoride....'47. Myodites scaber mo || AGO) || Bee) |). Fon Rhynchophora: | Rhynchitide. .'48. Rhynchites bicolor 84 148 480 412 Otiorhynchide |49. Cephus latus 41 IIL 382 | Sod Curculionide .|/50. Zygops seminivius 65 116 BO2n oS 49- 12— UZO= || - 27 Z= Variation 341 310 982 | 1,268! 1 The total number of pores of apterous species are not included. 420 N. E. MCINDOO. of the same species from 661 to 785 with 720 as an average. It is thus seen that the females of Harpalus have a few more pores than the males, while the males and females of Leptinotarsa have the same number of pores. ; STRUCTURE. In the preceding pages it has been shown that most of the variations in regard to the disposition of the olfactory pores are slight. In the following pages it will be shown whether or not this is true for the structure of these pores. (a) External Structure. When examined under a low-power lens, the olfactory pores may be easily mistaken for hair sockets from which the hairs have been removed. When more carefully observed under a high-power lens, a striking difference in external form is usually seen, but sometimes it is difficult to distinguish the pores from hair sockets. The pores appear as small bright spots when a strong transmitted light is used. Each bright spot has a dark boundary or pore wall (Plate I., Fig. 1, PorW).1 Near the center of this boundary is a transparent spot, the pore aperture, which may be round, oblong, slit-shaped, or club-shaped. On the legs the pore apertures may be round (Fig. 2, PorAp), oblong (Fig. 3, PorAp), slit-shaped or club-shaped (Fig. 1, PorAp). On the elytra and wings they may be round or oblong (Figs. 4 to 8). The hair sockets (Figs. 1 and 2, PorWHr) are generally smaller than the olfactory pores and the pores of the hypodermal glands (Figs. 1 and 2, PorWGl) are easily dis- tinguished from the hair sockets and olfactory pores by their small size. (b) Internal Structure. All the olfactory pores studied are more or less flask-shaped structures. They are of three general types. In the most common type, as found in Uloma, the mouth of the pore (Figs. 9-12, Mo) is flaring and the sense cell (Fig. 12, CS) lies in the lumen of the appendage outside the pore cavity. The chitinous 1 All figures, except Text-figs. 1, 2, and 3 are numbered consecutively on Plates I. and II. , ‘ THE OLFACTORY SENSE OF COLEOPTERA. A2l cone (Fig. 9, Con) never occupies more than one fourth of the pore cavity and usually much less (Fig. 12, Con). The cone always stains less deeply than the surrounding chitin, and it is common to see a hypodermal secretion (Figs. 9 and 10, Hyp) inside the pore cavity. The sense fiber (Fig. 9, SF) pierces the cone, and the chitin between the pore aperture and the cone, and it ends in the bottom of the pore aperture or pit (Figs. 9-12, P) with its peripheral end exposed to the air in the pit. The second type of pores is found in the legs of Orthosoma (Figs. 13-15), although the pores in the elytra (Fig. 21) and wings (Fig. 31) of the same beetle belong to the first or most common type. The chitinous integument of the legs of Ortho- soma is thicker than that of the legs of any other beetle examined. Instead of the sense cells (Fig. 13, SC) lying in the lumen of the legs outside the pore cavities, in this type they lie inside the pore cavities. When the chitin forming the wall of the pore is not thick enough to protect the entire sense cell, the wall of the pore projects flange-like (Fig. 14, Fl) into the lumen of the leg. In Fig. 14 only about one third of the sense cell (SC) is shown. Studies of the olfactory pores in various hymenopterous insects made by the writer have shown that the sense cells begin to differentiate at the time when the chitin is beginning to be formed. From this fact, it is quite probable that the sense cells found in the second type of pores have not migrated into the pore cavities, but they now remain in approximately the same position as when the chitin was being formed. The third type of pores is found in the legs of the lady beetle, Epilachna borealis. Instead of the chitin over the external end of the pore being depressed to form a pit, it is elevated dome-like above the surface of the leg. In the center of the dome lies the pore aperture (Fig. 16, PorAp). All the pores in the tro- chanters and most of those in the tibiee (Fig. 17, PorAp) are of this type. Sometimes in the tibia is found a pore whose aperture is on a level with the surface of the tibia. The apertures of all the pores in the elytra (Fig. 18, PorAp) and wings (Fig. 19) of this beetle are on a level with the surfaces of the appendages. As already stated, the olfactory pores of beetles are more or less flasklike as a rule, but there are many variations among 422 N. E. MCINDOO. them. ‘They may be inverted flask-shaped as found in the legs of Epilachna (Figs. 16 and 17) and in the wings of Passalus (Fig. 20). ‘Some have the shape of a flask without the neck (Figs. 9, 10 and 12). Some are long and slender like fingers or test tubes (Figs. 11, 18, 19 and 21). Their sizes also vary much. The length of a pore always depends on the thickness of the chitin. The diameters of the pores of a small beetle (Fig. 25) may be as large, or even larger (Figs. 9 and 10) than the diameters of the pores of a large beetle (Figs. 13-15). A chitinous cone is always present, although it may sometimes: be almost indiscernible. It invariably has the same shade of coloration (Fig. 17, Con) as the remaining chitin (Fig. 17, Che) which is formed after the insect has emerged into the imago stage. This is the first time that the writer has been able to determine definitely the formation of the cones. In all the hymenopterous insects studied by the writer, the chitinous integument is prac- tically developed when the insects emerge, but in most beetles only about one third of the chitin is formed when the insects emerge. Since this is true the hypodermal cells are still large and they are rapidly secreting a substance which forms new chitin. Their external ends stand in contact with the chitin, and when no chitin is present they send processes into all holes or cavities in the chitin. Thus the hypodermal cell (Fig. 23, HypC) at the mouth of each olfactory pore sends a process into the pore. Since the sense fiber has entered the pore aperture before the cone is formed, the latter is formed at the external end of the pore around the sense fiber. When the chitinous integument (Fig. 17) is fully developed no hypodermal processes run into the pores and the hypodermal cells are very small. The sense cells are always spindle-shaped (Figs. 12, 13, 16-19 and 23, SC). Only occasionally is an entire sense cell seen in a cross section, because the entire cell seldom lies in the same plane as that of the section. More entire sense cells may be seen in longitudinal sections, but even in these the cells are usually eut in two. Entire sense cells were best seen in the oblique sections through the peduncles of the elytra of Passalus and Epilachna. The nucleus (Figs. 13 and 23, SC Nuc) of the sense cell is always THE OLFACTORY SENSE OF COLEOPTERA. 423 conspicuous. It may be darker (Fig. 13, SCNuc) or lighter (Fig. 23, SCNuc) in color than the cytoplasm in the cell. The nucleoli (Fig. 23, SCNuc) are also conspicuous. Smaller sense cells may be seen in the sections through the proximal ends of the trochanters and through the proximal ends of the tibia. These (Fig. 17, SC:) belong to tactile hairs (Fig. 7, JBkAe In the sections through the legs and elytra of Epilachna, gland cells (Fig. 17, GIC) are plainly seen in the hypodermis (Hyp). These are equally as large as the olfactory sense cells, -but they are quite different in structure. The diameters of the pores of the glands (PorGl) are slightly smaller than those of the hairs (PorHr), and they are much smaller than those of the olfactory pores (Por). The morphology and physiology of these gland cells will be given in another paper. The shapes of the external ends or tops of the pits depend on the shapes of the pore apertures when seen in superficial views. That is, they are round, oblong, slitlike or clublike. The interna! ends or bottoms of the pits are always round. The pore aperture, proper, is the round opening leading from the bottom of the pit to the external end of the pore. This aperture is closed by the peripheral end of the sense fiber. The shapes of the pits in cross sections, therefore, depend on the directions in which the micro- tome knife passes through the pits. The most common shape of a pit in cross section is that of an urn (Fig. 9, P). Pits includ- ing the pore apertures may be likened to round funnels, or to funnels slightly flattened, or to funnels considerably flattened, or to funnels so flattened that their tops would be club-shaped. In spiders the pits are slits which pass entirely through the cuticula. The sense fibers enter the pore apertures at the bot- toms of the slits. The pits or slits in spiders, therefore, may be likened to funnels considerably flattened. When just emerged into the imago stage the pits (Fig. 9, P) in the legs generally extend about one-third the distance through the chitin, but when the chitin is fully developed, the pits extend perhaps from one fifth to one eighth the distance through the chitin. In all the figures showing two shades in the chitin, the darker one (Fig. 17, Ch,) represents the chitin formed at the time when the 424 N. E. MCINDOO. insect emerges from the last pupal stage, and the lighter one (Che) represents the chitin formed after emerging into the imago stage. As already stated, instead of the olfactory pores of the lady beetle, Epilachna, having pits, the chitin over each pore in the legs is elevated domelike above the surface of the leg. The olfactory pores (Figs. 24 and 25) in the legs of the two blister beetles, Epicauta marginata and Epicauta pennsylvanica, have only indications of pits. Their pore apertures are therefore on a level with the surface of the legs. The olfactory pores in the legs of the potato beetle, Leptinotarsa to-lineata, have shallow pits (Fig. 26, P). All four just enumerated species have hypo- dermal gland pores distributed over the entire body except the wings. These pores are perhaps most abundant on the elytra, but they were never seen on the peduncles of these appendages, and it is quite probable that the secretion from their glands never covers the olfactory pores found on the wings and on the peduncles of the elytra. Judging from the gland pores, the hypodermal glands in the legs of Epilachna are more highly developed than are those of the other three species. The gland pores (Figs. I, 2 and 28, PorWGl) on the legs of Eptilachna lie on all sides and even among the olfactory pores, but in the legs of the other three species the gland pores never lie near the olfactory pores. When examined under a low-power lens the legs and elytra of Epilachna appear wet, and many small yellow flakes may be seen on them. The wet appearance is certainly due to the secretion from the hypodermal glands and the flakes | are the remains of the secretion after it becomes dry. Thus in Epilachna there seems to be a direct correlation between the olfactory pores and the gland pores. Since the pore apertures in the legs lie above the surface of these appendages, the secretion from the hypodermal glands runs away from the pore apertures instead of into them. Such a device enables both sets of organs to function normally without the one hindering the other. In the legs the sense cells always lie in a blood sinus (Figs. 16 and 17, BlSin) some distance from the muscles (Fig. 28, M). The nerves (J) are easily seen and branches (VB) are given off which run to the sense cells (SC). The neurilemma (Fig. 17, THE OLFACTORY SENSE OF COLEOPTERA. 425 Neu) of the nerve is usually distinct. In the cross section of a nerve, the nervous substance appears more or less netlike and nuclei, probably neuroglia nuclei (Fig. 17, NeurNuc), stand out conspicuously in the network. The trachea (Figs. 16, 17 and 28, Tr) and nerves (N and NB) are firmly suspended by the connective tissue whose nuclei (Con7TNuc) are seen only occa- sionally. The lumen of the leg at the proximal end of the tibia of Epilachna seems to be divided into two chambers by a mem- brane (Fig. 17, Hyp:) which resembles hypodermis. This structure has never been seen before by the writer and nothing can be said about its function. The hypodermis (Fig. 18, Hyp) beneath the olfactory pores in the peduncles of the elytra is much thicker than elsewhere. It usually contains all the sense cells (SC), but in the elytra of Passalus the hypodermis is thinner and since the sense cells are so large and so numerous there is not enough room for all of them in the hypodermis. For this reason only a few of them lie among the hypodermal cells and the remainder of them lie in the lumen of the peduncle between the hypodermis and nerve. As usual they are surrounded by blood. In only one instance was the writer able to trace a sense cell all the way from the pore aperture to the nerve. Fig. 23 represents this sense cell connecting with the pore aperture (PorAp) and with the nerve (NV). The trachea (Tr) lies by the side of the nerve. A large nerve (Fig. 29, N) and a large trachea (77) run through the radial plate (RP) of the peduncles beneath the olfactory pores. From the nerve many branches are given off which connect with the sense cells. The hypodermis (Fig. 22, Hyp) beneath the olfactory pores in the wings is usually much thicker than elsewhere, but it does not contain the sense cells (SC). These cells lie in a blood sinus (Fig. 22, BlSin) between the hypodermis (Hyp) and the trachea (Tr), nerve (NV) and nerve branches (VB). In the wings it is usually difficult to trace a sense fiber all the way to the pore aperture, but in oblique superficial sections this is easily done (Fig. 30). A large nerve and a large trachea run into each wing. These divide so that a smaller nerve and a smaller trachea run through each main vein. The largest trachea (Fig. 31, Tr) runs” 426 N. E. MCINDOO. through the subcosta (Sc) while the largest nerves (VV) pass through the veins bearing the olfactory pores. The nerve and trachea run directly beneath the sense cells (SC) and from the nerve pass off many branches which connect with the sense cells. In the costa (C) and subcosta (Sc) where there are no sense cells, only a few nerve fibers can be seen. In the preceding pages it has been shown that there are many variations in the structure of the olfactory pores of beetles, and that these organs are very similar to those of hymenopterous insects. On the basis of the location of the pore apertures in the integument, the olfactory organs in beetles are intermediate between those of spiders and those of Hymenoptera. EXPERIMENTS TO DETERMINE THE LOCATION OF THE OLFACTORY ORGANS. Since it is now generally believed that the olfactory organs of beetles are borne by their antenne, these appendages of many individuals were pulled off. From one to seven days later, the mutilated insects were tested with odors. In the preceding pages it has been shown that the olfactory pores of Coleoptera are located on the peduncles of the elytra, on the wings and on the legs. In order to ascertain if these structures receive odor stimuli, the elytra, wings and legs were mutilated. One or more days later these mutilated beetles were tested with odors. In all the experiments with unmutilated and mutilated beetles, 434 individuals have been tested. These belonged to I1 species representing eight families. In order that the behavior of the mutilated beetles would be correctly interpreted, the behavior of unmutilated beetles under experimental conditions was first studied. Since it was not desired to ascertain the relative sensitiveness of males and females, both sexes were used indiscriminately. To determine the relative sensitiveness of unmutilated and mutilated indi- viduals under conditions which permitted of their close observa- tion, triangular experimental cases were employed. These were made of three narrow wooden strips, two of which were five and the third four inches long, each strip being half an inch thick. Wire screen served as a bottom and glass as a top for the case. THE OLFACTORY SENSE OF COLEOPTERA. 427 The apices and bases of these cases rested on two supports above a rigid table near a window. No screen was used to prevent the beetles from seeing the observer because they never showed any responses to the movements made by the observer. ' The following sources of odors were used for determining the reactions of the beetles in the experimental cases; chemically pure essential oils of peppermint, thyme, and wintergreen; parts of plants—leaves and stems of pennyroyal (Hedoma pulegi- oides?), and of spearmint (Mentha spicata); decayed matter —parts of decayed beetles (Harpalus pennsylvanica). All these substances were kept in stoppered vials of the same shape and size. The leaves and stems of the pennyroyal were dried, but they still gave off a strong odor when the vial was uncorked. The leaves and stems of the spearmint were fresh and they did not emit as strong an odor as did the other substances used. Beetles were killed and were torn to pieces. The pieces were put into a vial and after two or three days they emitted a foul and sickly odor. A beetle was carefully placed into one of the experimental cases. When first put into the case the insect usually wandered about for several minutes, but finally it became quiet. The insect was tested with the above odors only when it had become perfectly quiet, without the antennz being moved in the least. The stopper of a vial was quickly removed and the vial was gently and slowly placed under the experimental case directly beneath and within one half inch of the individual being tested. When all of these precautions are taken, a normal beetle generally responds to anyone of these odors within 60 seconds, but when all the reaction times are counted, it is seen that several of them failed to respond within 60 seconds. If a beetle when tested fails to react to an odor within 60 seconds, the response may be regarded as negative, and when it reacts to an odor within 60 seconds, the response may be called positive. As a control, an empty and odorless vial was now and then placed under the insects in the same manner. If by chance a beetle moved while the control test was being made, its behavior was different from that observed when odors were used. Only the first responses have been recorded and in all cases where there was the least 428 N. E. MCINDOO. doubt as to whether the insect moved for any reason other than the olfactory stimulus, such movements were never recorded. The reaction time was counted in seconds. With an ordinary watch the minimum time which can be definitely recorded is two seconds, although many of the individuals responded to some of the odors much more promptly. Owing to this source of error, the average recorded time is probably double what it should be in the cases where all the responses for the same insect were prompt. An intermission of 10 minutes elapsed between any two tests in the same experimental case. Each individual was tested only once with the same odor. In recording the responses the term “‘vibrated’’ is used to describe the rapid movement of the antennz or legs up and down: or from side to side. When this movement is slow, these ap- pendages are described simply as having “moved.” When the antenne, legs or mouth parts are moved so that they are quickly bent at their articulations, they may be described as being ‘“worked.’’ When at rest a beetle usually lies flat on its thorax and abdomen, so the word ‘“‘arose’’ means that the insect gets up and stands on its feet. In the averages of reaction times the probable error is presumably high. It has not been calculated since slight differences in reaction times are not considered as Significant in the discussion of results, All anthropomorphic terms are put in quotation marks. CARABID. Tue OLFACTORY SENSE oF Harpalus pennsylvanica. Many ground beetles (Harpalus pennsylvanica) were caught under flat stones in a corn field near the laboratory. As soon as brought to the laboratory, 25 of them were placed singly into the experimental cases. As they were being placed into the cases, some of them discharged a substance, presumably from the anal glands, which gave off an odor similar to that from formic acid. Confined in these cases, they sought the dark corners of the cases and did not wander about much inside the cases unless irritated. When half hidden in the dark corners, they rarely responded to odors, so it was necessary to keep them out of the corners while they were being tseted. The longer they remained THE OLFACTORY SENSE OF COLEOPTERA. 429 ha in the light and the more they were handled, the more satisfactory they were to experiment with. Owing to this kind of behavior, this species and several others used responded more slowly to odors a short time after being caught than they did a few days after being kept in confinement. This fact will explain why some unmutilated beetles just caught respond to odors more slowly than they do two or three days later after having had their antenne pulled off. The following are the responses of this eround beetle to the odors from the six different substances and the average reaction times in seconds. Oil of peppermint: 5 moved away quickly. rt worked legs. 5 vibrated antenne. I kicked quickly. 4 arose quickly. I vibrated antenne and legs. 4 moved slightly. _ I vibrated legs. 2 moved antenne and legs. rt jumped slightly. Reaction time 2 to 10 seconds, average 3.6 seconds. Oil of thyme: 6 moved away quickly. I arose slowly. 5 moved quickly. I vibrated antenne. 5 moved slightly. — I moved backward slowly. 2 worked antenne. rt worked legs. 2 moved antenne and legs. t did not respond. Reaction time 2 to 60 seconds, average 8.5 seconds. Oil of wintergreen: 5 moved away quickly. rt stroked antenne. 5 moved slightly. I vibrated antenne and legs. 4 moved away slowly. I worked antenne. 3 moved quickly. rt worked legs. 2 vibrated legs. I did not respond. 1 arose slowly. Reaction time 2 to 60 seconds, average 16.4 seconds. Leaves and stems of pennyroyal: Io moved away quickly. I moved away slowly. 6 moved slightly. I vibrated legs. 3 vibrated antenne. 1 did not respond. 3 worked antenne. Reaction time 3 to 60 seconds, average 21.1 seconds. Leaves and stems of spearmint: 5 moved away slowly. 2 did not respond. 5 moved slightly. t worked mouth parts. 3 moved antennee and legs. I vibrated legs. 2 moved away quickly. I vibrated antenne and legs. 430 N. E. MCINDOO. 2 worked antenne. I moved antenne. 2 jumped slightly. Reaction time 3 to 60 seconds, average 21.8 seconds. Parts of decayed beetles: 7 moved slightly. I moved legs. 5 did not respond. t worked antenne. 4 moved away quickly. I vibrated legs. 3 moved away slowly. I vibrated antenne and worked 2 jumped slightly. mouth parts. Reaction time 5 to 60 seconds, average 28.1 seconds. The general average reaction time of the 25 beetles tested to the six odors is 16.5 seconds. As a possible reason why one fifth of the individuals tested failed to respond to the odor from the decayed beetles is that these insects probably do not respond to decayed matter unless they are hungry. The 25 beetles tested were put into a wooden box four inches wide, seven inches long and two inches deep. One half inch of moist earth was also put into the box. The beetles soon buried in the earth and from that time on they appeared quite “at home.” The box was put into a table drawer where it was more or less dark. About twice each week water was poured upon the earth and the beetles were fed earthworms and various insect larve. They drank some of the water and always greedily ate the food given to them. Up to the time of this writing (Jan. 15), 24 of these beetles have died. These lived from 18 to 180 days with 61 days as an average. All the beetles confined in the laboratory have not been fed since Oct. 15, but they have been given water once or twice a week. A few of the dead beetles when removed from the box had been partially eaten, but these insects were never seen fighting one another. While collecting this species in the corn field, a dead one was now and then found. (a) Effects with Antenne Pulled Off. The antenne of 25 Harpalus pennsylvanica were pulled off at their bases. These insects were then put into a wooden box similar to the one containing the unmutilated individuals just described. This box, also containing moist earth, was placed into the table drawer. The beetles appeared normal in all respects for they drank and ate as greedily as the unmutilated THE OLFACTORY SENSE OF COLEOPTERA. 431 ones and buried in the earth as usual. Seven days later they were placed singly into the experimental cases and were tested with the six odors as usual. They wandered about in the cases. slightly more than did the unmutilated ones, but when tested they gave similar responses and reacted just as promptly. Their reaction times are as follows: Oil of peppermint, 2 to 15; seconds, average 3.8 seconds; oil of thyme, 2 to 25 seconds,. average 4.7 seconds; oil of wintergreen, 2 to 25 seconds, average 6.9 seconds; leaves and stems of pennyroyal, 3 to 50 seconds, average 14.4 seconds; leaves and stems of spearmint, 3 to 60 seconds, average 34.9 seconds. Ten failed to respond to this odor. Parts of decayed beetles, 3 to 60 seconds, average 32 seconds. Eight failed to respond to this odor. The general average reaction time of the 25 beetles tested to the six odors is 16.1 seconds. Up to the time of this writing (Jan. 15), 23 of these beetles have died. They lived from 19 to 171 days with 58 days as an average. (b) Effects with Elytra and Wings Pulled Of. The elytra and wings of 25 Harpalus pennsylvanica, just collected from the cornfield, were pulled off at their articulations.. These mutilated insects were then put into a third box, similar to the two already described. The box was kept in the table: drawer with the others. On the following day after mutilating: the beetles, they were placed singly into the experimental cases: and were tested with the six odors as usual. They seemed normal in all respects except they were extremely restless. Their responses to odors were similar to those of unmutilated ones, except they were slower. Their reaction times are as follows: Oil of peppermint, 3 to 45 seconds, average 10.7 seconds; oil of thyme, 5 to 50 seconds, average 10.2 seconds; oil of wintergreen, 5 to 60 seconds, average 18 seconds. Two failed to respond to this odor. Leaves and stems of pennyroyal, 5 to 60 seconds, average 29.2 seconds. Seven failed to respond to this odor. Leaves and stems of spear- mint, 5 to 60 seconds, average 24.7 seconds. Four failed to respond to this odor. Parts of decayed beetles, 5 to 30 seconds, average 13.4 seconds. The general average reaction time of the 432 N. E. MCINDOO. 25 beetles tested to all six odors is 17.7 seconds. These mutilated insects lived from 2 to 21 days with 9 days as anaverage. All the time they were confined in the small box, they drank, ate, and buried in the earth normally, but many times one was seen biting the soft dorsal portion of the abdomen of another. With the elytra and wings removed, the abdomens were unprotected and many of them shrank considerably in size before the beetles died. Some of these beetles were certainly killed on account of the dorsal sides of their abdomens being bitten, because nearly every one found dead had been entirely eaten except the chitinous parts. In the other two boxes as already mentioned, only occasionally was a dead beetle found that had been eaten. (c) Effects with Elytra and Wings Pulled Off and Pores on Legs Covered with Vaseline. The elytra and wings of 18 Harpalus pennsylvanica were pulled off at their articulations. Four days later the trochanters, femurs and proximal ends of the tibize of these mutilated beetles were covered with a vaseline-beeswax mixture, consisting of three fourths yellow commercial vaseline and one fourth beeswax. An hour after the legs had been vaselined, the beetles were placed singly into the experimental cases and were tested with the six odors as usual. Most of them were comparatively quiet, but a few were extremely restless. Their responses to odors were not pronounced and were slow, otherwise they were similar to those of unmutilated beetles. Their reaction times are as follows: Oil of peppermint, 3 to 60 seconds, average 19.5 seconds. Three failed to respond to this odor. Oil of thyme, 3 to 60 seconds, average 12.5 seconds, ‘Two failed to respond to this odor. Oil of wintergreen, 3 to 60 seconds, average 18.7 seconds. Four failed to respond to this odor. Leaves and stems of pennyroyal, 5 to 60 seconds, average 38.6 seconds. Nine failed to respond to this odor. Leaves and stems of spearmint, 3 to 60 seconds, average 32.9 seconds. Seven failed to respond to this odor. Parts of decayed beetles, 4 to 60 seconds, average 22.1 seconds. Two failed to respond to this odor. The general average reaction time of the 18 beetles tested to all six odors is 24.1 seconds. Confined in a box similar THE OLFACTORY SENSE OF COLEOPTERA. 433 to the other three already mentioned, these mutilated beetles drank, ate and buried in the earth normally, but they were less active than unmutilated ones. It was common to see them biting the dorsal sides of the abdomens. Before they died several of their abdomens had shrunk considerably in size. When found dead several of them had been entirely eaten except the chitinous parts. Counting from the time the elytra and wings were pulled off, they lived from 5 to 21 days with ro days as an average. THE OLFACTORY SENSE OF Harpalus caliginosus. Eight ground beetles (Harpalus caliginosus) were caught under flat stones. They were tested with the odors from only the three essential oils. In behavior, they were comparatively quiet. When tested, many of them moved away quickly; a few vibrated the antenne, and a few moved their legs. Their reaction times are as follows: Oil of peppermint, 2 to 10 seconds, average 4.4 seconds; oil of thyme, 2 to 8 seconds, aver- age 4.1 seconds; oil of wintergreen, 2 to 8 seconds, average 4.1 seconds. The general average reaction time to all three odors is 4.2 seconds. The antennz of these beetles were pulled off and the insects were then kept in a small box containing earth in the table drawer. (a) Effects with Antenne Pulled Off. Eight days after the antenne of the eight preceding Harpalus caliginosus had been pulled off, the remaining six live ones were again tested with the same odors in the usual way. Their responses were similar to those given before they were mutilated, but were not so pronounced. When tested with the oil of thyme, one beetle rubbed a hind leg on an elytron for a half minute. Their reaction times are as follows: Oil of peppermint, 3 to 25 seconds, average 12.5 seconds; oil of thyme, 4 to 60 seconds, -average 14.3 seconds. One failed to respond to this odor. Oil of wintergreen, 10 to 35 seconds, average 22.5 seconds. The general average reaction time to all three odors is 16.4 seconds. These mutilated beetles were quite inactive and sometimes scarcely moved when touched with a pencil. They did not eat as greedily as before being mutilated. They lived from 2 to 65 days with 18 days as an average. 434 N. E. MCINDOO. COCCINELLID&. THE OLFACTORY SENSE OF Epilachna borealis. Many lady beetles (Epilachna borealis) were caught on pumpkin vines in the corn field. When brought to the laboratory, they were put into a large glass jar near a window. The jar was II inches tall and 9 inches in diameter. It was covered with cheese-_ cloth. Since this lady beetle feeds upon the leaves of pumpkin and of allied plants, several pumpkin leaves were put into a wide- mouthed bottle containing water. The bottle with contents was then put into the jar. The beetles soon found the leaves and from that time on, they appeared ‘‘at home” as much as they do in corn fields on pumpkin leaves. They were regularly provided with a fresh supply of food. Occasionally they were seen copulating. On the following day after being caught, 18 of them were removed from the jar and were put singly into the experimental cases. When mechanically irritated they draw in the antennz and legs, usually eject a small drop of yellowish liquid from each femoro-tibial articulation, and feign death. They may lie apparently lifeless for several moments and when tested with odors they may or may not respond. Owing to this peculiar behavior, they were unsatisfactory to experiment with and their average reaction times are slower than might be expected. They were extremely quiet and when tested they generally moved away slowly. They often vibrated the antenne and mouth parts, and sometimes the legs. Their reaction times to the odors from the three essential oils are as follows: Oil of peppermint, 2 to 55 seconds, average 12.4 seconds; oil of thyme, 2 to 20 seconds, average 6.8 seconds; oil of wintergreen, 3 to 60 seconds, average 22.2 seconds. Three failed to respond to this odor. The general average reaction time to all three odors is 13.8 seconds. Sixteen of these insects were muti- lated for other experiments. The seventeenth lived only 3 days and the eighteenth is still living at this writing (Jan. 15). (a) Effects with Antenne Pulled Of. The antenne of 25 Epilachna borealis, just caught, were pulled off at their bases. A small drop of yellowish blood exuded from THE OLFACTORY SENSE OF COLEOPTERA. 435 each wound. On the following day the beetles were tested with odors. Asa rule they were so inactive that they appeared life- less. If touched while moving they feigned death and remained inactive for several moments. When tested with odors most of them worked the mouth parts; some moved away slowly; a few vibrated one or more legs, and some failed to respond. Their reaction times to the odors from the three essential oils are as follows: Oil of peppermint, 2 to 60 seconds, average 18.6 seconds. Three failed to respond to this odor. Oil of thyme, 2 to 60 seconds, average 38.7 seconds. Fourteen failed to respond to this odor. Oil of wintergreen, 3 to 60 seconds, average 35.1 seconds. The general average reaction time to all three odors is 30.8 seconds. Up to the time of this writing (Jan. 15), 15 of these mutilated beetles have died. They lived from I to 96 days with 22 days as an average. (b) Effects with Elytra and Wings Pulled Off. The elytra and wings of 10 Epilachna borealis were pulled off at their articulations. A small drop of yellowish blood exuded from each wound. A liquid of the same color is also present throughout the elytra and in the veins of the wings. On the second day after being mutilated, the four remaining live beetles were tested as usual. They were very quiet, but appeared normal in all respects except they responded to odors more slowly than unmutilated ones. Their reaction times to the odors from the three essential oils are as follows: Oil of peppermint, 10 to 60 seconds, average 25 seconds. One failed to respond to this odor. Oil of thyme, 5 to 60 seconds, average 33.5 seconds. Two failed to respond to this odor. Oil of wintergreen, 7 to 60 seconds, average 35.5 seconds. Two failed to respond to this odor. The general average re- action time to all three odors is 31.3 seconds. Up to the time of this writing (Jan. 15), I of these beetles has died. Counting the 7 mutilated beetles that died, they lived from 2 to 3 days with 2 days as an average. 436 N. E. MCINDOO. TELEPHORID#. THE OLFACTORY SENSE OF Chaulcognathus pennsylvanica. Many fireflies (Chaulcognathus pennsylvanica) were caught on goldenrod (Solidago). They were put into a cage 20 inches long, 16 inches tall and 12 inches wide. The sides and top of the cage were cheesecloth while the ends and bottom were wood. The cage was kept in the light near a window and a fresh supply of goldenrod was constantly kept in the cage. On the goldenrod in the cage, these insects appeared quite ‘‘at home.’’ Twenty- five of them were tested with the odors from the three essential oils. When tested most of them moved away quickly; a few vibrated antennz; a few vibrated legs, and a few arose slowly. They were extremely restless at all times. In the cage they copulated as freely as they do out-of-doors. Their reaction times are as follows: Oil of peppermint, 2 to 12 seconds, average 2.6 seconds; oil of thyme, 2 to 10 seconds, average 3 seconds; oil of wintergreen, 2 to 10 seconds, average 3 seconds. The general average reaction time to all three odors is 2.8 seconds. They lived from 3 to 7 days with 3.2 days as an average. (a) Effects with Antenne Pulled Of. The antennz of 27 Chaulcognathus pennsylvanica were pulled off at their bases. A day later only three were alive. When tested these three responded as promptly as unmutilated ones. The general average reaction time to the odors from the three essential oils is 2.8 seconds. Counting all 27 beetles, they lived from I to 5 days with 1.3 days as an average. LUCANID. THE OLFACTORY SENSE OF Passalus cornutus. Four stag beetles (Passalus cornutus) were removed from rotten stumps. While being tested with odors they were compara- tively quiet and responded promptly. Their most common response was to draw in the antenne and to move away slowly. The general average reaction time to all six odors is 3.2 seconds. The antenne were pulled off at their bases. A small drop of THE OLFACTORY SENSE OF COLEOPTERA. 437 blood exuded from each wound. The beetles were kept in a small box filled with moist rotten wood. (a) Effects with Antenne Pulled Of. Two days after pulling off the antennae, the four preceding mutilated beetles were again tested with the same odors. They were more quiet than before being mutilated. Their responses were just as prompt but were less pronounced than before they were mutilated. Their most common response was to work the mouth parts and to move away slowly. The general average reaction time to all six odors is 3.3 seconds. They lived from 4 to 20 days with 12.5 days as an average. SCARABAID. THE OLFACTORY SENSE OF Cotinis nitida. One lamellicorn beetle (Cotinis nitida) was tested with the six odors. The most common response was to stretch out its head, and to move its antenne and front legs. Once it drew in the antenne and moved the front legs. The average reaction time is 8 seconds. The antenne were pulled off at their bases. A small drop of blood exuded from each wound. (a) Effects with Antenne Pulled Off. A day after pulling off the antenne, the preceding Cotinis nitida was again tested with the same odors. It responded as promptly as before being mutilated. The most common re- sponse was to work the mouth parts and to move away slowly. The average reaction time is 8.3 seconds. It lived 12 days after being mutilated. THE OLFACTORY SENSE OF Euphoria sepulchralts. Five lamellicorn beetles (Euphoria sepulchralis) were caught on goldenrod (Solidago). While being tested with the odors from the three essential oils, they were extremely restless. They generally moved away slowly and drew in the antenne when tested with an odor. The general average reaction time is 3.6 seconds. After the antenne had been pulled off at their bases, the beetles were put into the cage described on page 436. 438 N. E. MCINDOO. (a) Effects with Antenne Pulled Of. A day later the five preceding mutilated insects were again tested with the same odors. They were quiet and their responses were similar to those before being mutilated, except, of course, there were no antennal movements. The general average re- action time is 4.3 seconds. These beetles lived from 9 to 42: days with 20 days as an average after being mutilated. CERAMBYCID&. THE OLFACTORY SENSE OF Cyllene robinie. Eighteen wood-boring beetles (Cyllene robinie) were caught on goldenrod (Solidago). While being tested with the odors from the three essential oils, they were extremely restless. When tested, most of them moved away quickly; a few arose quickly, and a few vibrated the antenne. The general average reaction time is 5.4 seconds. These beetles were confined in the cage described on page 436. They were regularly given a fresh supply of goldenrod. They seemed ‘‘at home”’ and copulated as freely in the cage as they do out-of-doors. They lived from I to 17 days with 10.4 days as an average. (a) Effects with Antenne Pulled Of. Eighteen more Cyllene robinie were collected from goldenrod. Their antennz were pulled off at the bases. A small drop of blood exuded from each wound. These beetles were placed into the cage with the unmutilated ones. Two days later the 15 remaining live ones were tested with the odors from the essential oils. They were very quiet and their responses were similar to those of unmutilated individuals, except as a rule they were more prompt. The general average reaction time is 3 seconds. In the cage it was common to see the unmutilated and antenneless cerambycids copulating. The former were very active and flew out of the cage whenever the door was opened, but the latter seldom flew and they were not so active. The mutilated ones lived from I to 11 days with 5 days as an average. (b) Effects with Elytra and Wings Pulled Off. Eighteen more Cyllene robinie were collected. Their elytra and wings were pulled off at the articulations. A small drop THE OLFACTORY SENSE OF COLEOPTERA. 439 of blood always exuded from each wound caused by the elytron being pulled off, but only occasionally was blood seen where a wing had been pulled off. A day later when tested with the odors from the three essential oils, these beetles were compara- tively quiet and they appeared normal in all respects except in their slowness in responding to odors. Their reaction times are as follows: Oil of peppermint, 2 to 30 seconds, average 7.1 seconds; oil of thyme, 3 to 20 seconds, average 8.9 seconds; oil of wintergreen 3 to 55 seconds, average 13.4 seconds. The general average reaction time to all three odors is 9.8 seconds. In the cage with the other beetles, these mutilated ones were as active as the unmutilated cerambycids and they were often seen copulating with each other, and with the unmutilated and antenneless ones. They lived from I to II days with 4.2 days as an average. CHRYSOMELID. THE OLFACTORY SENSE OF Leptinotarsa ro-lineata. Forty-five Colorado potato beetles (Leptinotarsa 1o-lineata) were collected in a potato patch near the laboratory. While 25 of them were being tested with the six odors, they were com- paratively quiet as a rule, but five were so restless that they were discarded and others were used. Their responses were similar to those of Harpalus pennsylvanica, described on page 429. Their reaction times are as follows: Oil of peppermint, 2 to 7 seconds, average 3.3 seconds; oil of thyme, 2 to 5 seconds, average 3.1 seconds; oil of wintergreen, 2 to 12 seconds, average 5 seconds; leaves and stems of pennyroyal, 4 to 60 seconds, average 26.7 seconds. Six failed to respond to this odor. Leaves and stems of spearmint, 2 to 60 seconds, average 25.6 seconds. Seven failed to respond to thisodor. Parts of decayed beetles, 5 to 60 seconds; average 27.9 seconds. Seven failed to respond to this odor. The general average reaction time of the 25 beetles tested to all six odors is 15.4 seconds. These insects were confined in a cage in the light near a window. This cage is 30 inches long, 30 inches high and 4% inches wide. A\ll six sides are wire-screen. A fresh supply of potato plant leaves was constantly kept in the cage. The beetles confined in this cage on the potato plant 440 N. E. MCINDOO. leaves appeared “‘at home”’ just as much as they do in potato patches. They ate the leaves, copulated and laid eggs as usual. Up to the time of this writing (Jan. 15), 28 of the 45 beetles have died. These lived from 14 to 151 days with 69 days as an average. , (a) Effects with Antenne Pulled Of. Twenty-nine more potato beetles were collected from the potato patch. Their antennze were pulled off at the bases. A small drop of blood exuded from each wound. These insects were put into the wire-screen cage with the unmutilated ones. Two days later the 23 remaining live ones were tested with only the odors from the three essential oils. All of these beetles were quite inactive and three failed to respond when tested. These three also failed to respond when touched with a pencil. For this reason they were discarded. The general average reaction time of the 20 beetles tested is 3.5 seconds. Asa rule these mutilated insects appeared normal in all respects several days after having the antennz pulled off, because they ate, copulated and were as active as ever. They lived from 2 to 140 days with 38 days as an average. (b) Effects with Elytra Pulled Off and Wings Cut Off. Thirty-one more potato beetles were collected. Their elytra were pulled off at the articulations and the wings were cut off as closely as possible to the articulations. A small drop of reddish or yellowish blood exuded from each wound. The heavy veins, extending from the base of the wing to where the wing folds, contain most of the blood found in these wings. The elytra are also filled with blood. The amount of blood in them gradually diminishes from the base to the distal end. A day after being mutilated 25 of these insects were tested with the six odors. They were apparently normal in all respects except in their slowness in responding to odors. They were as active as un- mutilated ones and eight were extremely restless. Their re- sponses were similar to those of unmutilated beetles, except they were not pronounced. Their reaction times are as follows: Oil of peppermint, 2 to 40 THE OLFACTORY SENSE OF COLEOPTERA. AAT seconds, average 7.8 seconds; oil of thyme, 2 to 15 seconds, aver- age 4.8 seconds; oil of wintergreen, 3 to 60 seconds, average 21.1 seconds. Five failed to respond to this odor. Leaves and stems of pennyroyal, 5 to 60 seconds, average 32.2 seconds. Ten failed to respond to this odor. Leaves and stems of spearmint, 3 to 60 seconds, average 29.8 seconds. Eight failed to respond to this odor. Parts of decayed beetles, 3 to 60 seconds, average 30.4 seconds. Eight failed to respond to this odor. The general average reaction time of the 25 beetles tested to the six odors is 22.7 seconds. In the wire-screen cage with the other potato beetles already tested, these mutilated ones appeared normal, because they ate normally and copulated as much as usual. Since the soft dorsal sides of their abdomens were unprotected, many of them soon began to sink, so that by the time a beetle died, the abdomen had shrunk to about one-fourth its original size. Up to the time of this writing (Jan. 15), 29 of these 31 mutilated insects havedied. They lived from 3 to 140 days with 52 days as an average. (c) Effects with Elytra Pulled Of, Bases of Wings Glued and Pores on Legs Covered with Vaseline. Twenty-nine more potato beetles were collected. Their elytra were pulled off at the articulations. Two days later the upper surfaces of the bases of the wings of the 26 remaining live ones were covered with liquid glue. Since the olfactory pores extend a considerable distance from the base of the wing along the radial vein, the glue applied probably did not cover more than 90 per cent. of the pores on each wing. Three hours after apply- ing the glue, the trochanters, femurs and proximal ends of the tibiz of these beetles were covered with the vaseline-beeswax mixture. An hour still later the insects were tested with the six odors. They were as active as unmutilated ones and appeared normal in all respects except in their responses to odors. Their responses were never pronounced and seldom prompt. Their reaction times are as follows: Oil of peppermint, 3 to 60 seconds, average 10.7 seconds. One failed to respond to this odor. Oil of thyme, 3 to 60 seconds, average 9 seconds. One failed to respond to this odor. Oil of wintergreen, 5 to 60 seconds, 442 N. E. MCINDOO. average 35.9 seconds. Eleven did not respond to this odor. Leaves and stems of pennyroyal, 3 to 60 seconds, average 35.2 seconds. Twelve did not respond to this odor. Leaves and stems of spearmint, 5 to 60 seconds, average 42.6 seconds. Four- teen failed to respond to this odor. Parts of decayed beetles, 5 to 60 seconds, average 40.3 seconds. Fourteen failed to respond to this odor. The general average reaction time of the 26 beetles tested to the six odors is 29 seconds which is twice the reaction time of unmutilated potato beetles to the same odors. When the reaction times to the odors from only the three essential oils are considered, these mutilated insects responded only one fifth as rapidly as did the unmutilated ones. In the wire-screen cage with the other potato beetles already tested, they were apparently normal as long as they lived, because they ate and copulated as usual and were always as active as the unmutilated ones. Before they died their abdomens shrunk considerably in size. Up to the time of this writing (Jan. 15), 28 of the 29 have died. These lived from 2 to 151 days with 61 days as an average. MELOID. THE OLFACTORY SENSE OF Epicauta marginata. Twenty blister beetles (Epicauta marginata), commonly known as the “old-fashioned potato bugs,’ When mechanically irritated, they fold the antenne and legs ’ were caught on clematis. against the body, usually eject a small drop of amber-colored liquid from each femoro-tibial articulation, and feign death. On account of this behavior, they were unsatisfactory to experi- ment with. When put into the experimental cases, some of them lay apparently lifeless for almost a half day. In this state they never respond to any odor, and after becoming as active as usual, they may or may not respond to odors. When tested with the odors from only the three essential oils, a general average reaction time of 13.9 seconds was obtained. Two of them failed to respond to each of the oils of peppermint and wintergreen. These insects were confined in the cage described on page 436. They were regularly provided with a fresh supply of clematis. In this cage on the clematis they seemed ‘‘at home,’’ but they flew out at every opportunity. THE OLFACTORY SENSE OF COLEOPTERA. 443 They copulated as usual. They lived from 11 to 40 days with 27.6 days as an average. (a) Effects with Antenne Cut Off. _ Eight more Epicauta marginata were collected. Their antennze were cut off at the bases. A small drop of amber-colored blood exuded from each wound. Seven days later the two remaining live ones were tested with the odors from the three essential oils. The general average reaction time is 5 seconds. All these beetles were abnormal in behavior. They lived from 1 to 8 days with 3.4 days as an average. (b) Effects with Antenne Pulled Of. The antenne of 12 more Epicauta marginata were pulled off at their bases. A small drop of blood exuded from each wound. When tested with the odors from the essential oils three days later, the eight remaining live beetles gave a general reaction time of 5.9 seconds. They were less abnormal in behavior than those with the antennz cut off. They lived from 2 to 13 days with 5.5 days as an average. (c) Effects with Elytra and Wings Pulled Of. The elytra and wings of nine Epicauta marginata were pulled off at their articulations. A small drop of blood exuded from each wound. When tested with the odors from the essential oils two days later, the seven remaining live beetles gave a gen- eral reaction time of 25.7 seconds. Two of them failed to re- spond to each of the oils of peppermint and wintergreen. These mutilated insects appeared normal in behavior and in confine- ment they copulated as usual. They lived from 2 to 14 days with 8 days as an average. Tuer OLFACTORY SENSE OF Epicauta pennsylvanica. Twenty-five blister beetles (Epicauta pennsylvanica) were caught on golden rod (Solidago). This species has the same habit of feigning death when mechanically irritated as has Hp- cauta marginata. When tested with the odors from the essential oils, they gave a general average reaction time of 11.5 seconds 444 N. E. MCINDOO. which is only one-half as rapid as the reaction time of the same species devoid of antenne. Three failed to respond to the oil of peppermint, one to the oil of thyme and two to the oil of wintergreen. A common response was to vibrate the legs. They were placed into the cage with the other species of blister beetles. They were regularly provided with a fresh supply of goldenrod. In the cage they appeared normal, and they copulated as much as usual. They lived from 2 to 25 days with 11.2 days as an average. (a) Effects with Antenne Pulled Of. The antenne of 30 Epicauta pennsylvanica were pulled off at their bases. When tested with the odors from the essential oils three days later, the 22 remaining live beetles gave a general reaction time of 5.3 seconds. They were only slightly abnormal in behavior. They lived from 2 to 25 days with 8.7 days as an average. (6) Effects with Elytra and Wings Pulled Off. The elytra and wings of 21 Epicauta pennsylvanica were pulled off at their articulations. A small drop of blood exuded from each wound. Blood was also seen in the distal ends of the elytra. When tested with the odors from the essential oils two days later, the 17 remaining live beetles gave a general reaction time of nine seconds. One of them failed to respond to the oils of thyme and wintergreen. These insects appeared normal in confinement with the other blister beetles. They copulated as usual. They lived from 1 to 33 days with 10.7 days as an average. A summary of all the preceding experiments to determine the location of the olfactory organs in beetles is best presented in a tabulated form. The following table is such asummary. Since a comparison of the behavior of unmutilated and mutilated in- sects alone is not always a safe criterion for judging the general behavior of mutilated beetles, the behavoir of the mutilated beetles recorded in this table is based mostly upon a comparison of the longevities of unmutilated and mutilated individuals of the same species. A “+ ”’ after a figure in the last column means that all the insects used in the experiment have not yet died. The longevity is based only on those that have died up to the time of this writing (Jan. 15). THE OLFACTORY SENSE OF COLEOPTERA. divers: JL 445 SUMMARY OF EXPERIMENTS TO DETERMINE THE LOCATION OF THE OLFACTORY ORGANS IN COLEOPTERA. Average Reaction Time. 3| Average Length Ey Average Le 6 ; nS) of Life in Species Se ee ae lace ce eel, Cees Odors. oe | os | Sec. Sec. A Days. Harpalus Unmutilated. Normal in be- DAMS ACEH: | IANO co noo so Gade ao oar Ge O.§ || TO5 | as 61.0+ Antenne pulled off. Normalin De AVM OTE Se ihe e ialav ede saceranane Sot | mOsw || Bs 58.0+ Elytra and wings pulled off. \Slightly abnormal in behavior.| 13.0 | 17.7 | 25 9.0 ‘Elytra and wings pulled off and pores on legs covered with vaseline. Slightly abnormal MINE AO Ioan ahs. cm, sinclar eqenar ene 16.9 | 24.1 | 18 10.0 Harpalus Unmutilated. | Normal in be- caliginosus. IOEINAUGI Cd Gin Rectete piece ements 4.2 8 |Used below Antenne pulled off. Slightly abnormal in behavior....... 16.4 6 18.0 Epilachna Unmutilated. _ Normal in be- borealis. IN AWAO Tey dete cues tens). ote ss, 4s Gee 13.8 18 | Used below Antenne pulled off. Slightly abnormal in behavior....... 30.8 AS 22.0+ Elytra and wings pulled off. Slightly abnormal in behavior) 31.3 4 2.0+ Chaulcognathus |\Unmutilated. Normal in be- PELMOSSMTCOEE. || WENO, cocoscasnoadcecoou™ 2.8 25 Baz Antenne pulled off. Slightly abnormal in behavior....... 2.8 3 I.3 Passalus Unmutilated. | Normal in be- cornutus. IDERGKOIES oro. ec rote Ol creer ee 3.0 Bho 4| Used below Antenne pulled off. Slightly abnormal in behavior....... 3.0 3.3 4 12.5 Cotinis nitida...|Unmutilated. Normal in be- INEINVAVONES oitib.5s5 ar Cho CE RENE ROR Ge RCI 5.0 8.0 1 | Used below Antenne pulled off. Normal ime belasylOte ayes nies s+ <0 wos 5.6 8.3 I 12.0 Euphoria Unmutilated. Normal in be- sepulchralis. IGEN AOS 5 Gd ose biol iene RE 3.6 5 | Used below Antenne pulled off. Normal in behavior saeco. osisa ce. 4.3 5 20.0 Cyllene robinie..|Unmutilated. Normal in be- MAW Oren Verne ss oleic aes 5.4 18 10.4 Antenne pulled off. Slightly abnormalin behavior....... 3.0 15 5.0 Elytra and wings pulled off. Slightly abnormalin behavior! 9.8 18 4.2 446 N. E. MCINDOO. Average ay NeaeioNn Biases = Z Average Length Sage Experiment and Behavior of For For E 4 oh ene ee? Insects Tested. Three Six |aa Capt. Odors. | Odors. oe 2s Sec. Sec. Bi Days. Leptinotarsa Unmutilated. Normal in be- ro-lineata. RAVAGE case ee cole e eae Bn | ws o4) || 2s 69.0 + Antenne pulled off. Normal in behavior sscee eae: 3.5 20 38.0 Elytra pulled off and wings cut off. Normal in behavior...| I1.2 | 22.7 | 25 52.0+ Elytra pulled off, bases of wings glued and pores on legs covered with vaseline. Nor- maleinebehavloissaeieneenon 18.5 29.0 | 26 61.0+ Epicauta Unmutilated. Normal in be- marginata. NAVIOK ties. cane oe eee 13.9 20 DL) Antenne cut off. Considerably abnormal in behavior....... 5.0 2 3.4 Antenne pulled off. Slightly abnormal in behavior....... 5.9 8 5.5 Elytra and wings pulled off. Slightly abnormalin behavior] 25.7 7 8.0 Epicauta Unmutilated. | Normal in be- pennsylvanica. HaviOrss o..0 ne eee or II.5 25 Rit, Antenne pulled off. Slightly abnormal in behavior....... 5.8 22 8.7 Elytra and wings pulled off. Normal in behavior........ | 9.0 ie 10.7 A summary of the preceding table shows the following: After the antenne were pulled off, four of the 11 species tested were normal and seven were slightly abnormal in behavior. After the elytra and wings were pulled off one species was normal while four were slightly abnormal in behavior. After the elytra were pulled off and the wings were cut off, the one species tested was normal in behavior. After the elytra and wings were pulled off and the pores on the legs were covered with vaseline, the one species tested was slightly abnormal in behavior. After the elytra were pulled off, the bases of the wings glued and the pores on the legs covered with vaseline, the one species tested was normal in behavior. Four unmutilated species responded to odors more slowly than did the same species after the antenne had been pulled off. This is explained by the fact that most beetles are more or less “‘ timid” THE OLFACTORY SENSE OF COLEOPTERA. 447 for some time after being caught, and some feign death. Asa rule the longer they are confined and the more they are handled, the more satisfactory they are to experiment with. Five species without antennz responded to odors as promptly as did the same species unmutilated. Two species without antenne responded to odors more slowly than did the same species unmutilated. Since these were abnormal in behavior and judging from the reaction times of the other nine species with antennz pulled off, it is only reasonably to attribute the slow reaction times of these two species to their abnormal condition caused by the antennz being pulled off. The six species so mutilated that most of their olfactory pores on the elytra and wings were prevented from functioning responded from two to five times more slowly than did the same species unmutilated or with the antennez pulled off. The two species so mutilated that most of their olfactory pores on the elytra, wings and legs were prevented from func- tioning responded from two to six times more slowly than did the same species unmutilated or with the antenne pulled off. From all the preceding results, it seems that the antenne do not carry any of the olfactory organs, while the olfactory pores found on the peduncles of the elytra, on the dorsal surfaces of the wings, on the trochanters, tibia, sometimes on the femurs and tarsi, and perhaps on the mouth appendages, are the true olfactory organs in beetles. SUMMARY. In making a comparative study of the olfactory pores in beetles, 50 species belonging to 47 genera and representing 34 families were used. A group of pores is always present on the peduncle of each elytron. It lies on the dorsal side of the well- exposed radial plate. The number of pores on a pair of elytra varies from 12 to 310. Asarule, the more pores in the group the smaller they are and the closer they are together. Of the 47 winged species examined, 11 have only one group of pores on each wing, 21 have two groups on each wing, 12 have three groups on each wing, and 3 have four groups on each wing. These groups are always located on the dorsal surface. Only occasionally are a few scattered pores found on the ventral side 448 N. E. MCINDOO. of a wing. When one or two groups are present, they lie on the radius. When three groups are present, all three may lie on the radius, or two may lie on the radius and the third on the media. When four groups are present, one lies on the subcosta, two on the radius and one on the media. The largest group on the radius usually extends nearly all the way to the fold of the wing and sometimes all the distance to the fold. The number of pores on a pair of wings varies from 130 to 982. There are usually two groups of pores at the proximal end of each trochanter. Sometimes a pore is found at the proximal end)'of the femur. It is common to find a few pores at the prox- imal end of each. tibia; and sometimes pores are found in the tibial spines and on the tarsi. The number of pores on all six legs varies from 49 to 341. In regard to water beetles, the better the legs are adapted for locomotion in water, the fewer pores they have. The smallest winged species (Coxelus) examined has 273 pores, which is the smallest number of all the species, and the largest species (Ortho- soma) has 1,268 pores which is the largest number of all the species examined. The apterous species have more pores on the legs than usual. Asa rule, the smaller the species, the fewer its pores and the larger they are, comparatively speaking. As a rule, there are no generic and specific differences, except variations in number of pores, the amount of variation depending on the sizes of the individuals compared. There are no individual and sexual differences other than slight variations in number of pores. The pore apertures or pits are round, oblong, slitlike or club- shaped. On the elytra and wings they are always round or oblong. On the legs they have all four of the enumerated shapes. The spindle-shaped sense cells of most beetles lie in the lumens of the appendages outside the pore cavities, but in the legs of Orthosoma the sense cells lie inside the pore cavities. A small chitinous cone is always present. It is formed by the hypodermal cell at the mouth of the pore after the insect has emerged from the last pupal stage, and at the same time when the chitinous integument is being considerably thickened. The sense cells are fully developed when the insect emerges into the imago stage. The sense fiber pierces the cone and the layer of chitin between THE OLFACTORY SENSE OF COLEOPTERA. 449 the pore aperture and cone, and it enters the bottom of the pore aperture or pit where its peripheral end comes into direct contact with the outside air. In Hymenoptera the sense fibers enter the pore apertures which are almost on a level with the external surface of the chitin. In Coleoptera, with a few exceptions, the sense fibers enter the bottoms of pits which lie in the chitin one third (at time of emerging into imago stage) the distance from the external surface. In the legs of the lady beetle, Epilachna borealis, instead of the chitin which surrounds the pore apertures being depressed, it is elevated so that the pore apertures lie in the center of domes above the general surface of the legs. In the legs of the blister beetles, Epicauta marginata and E. penn- sylvanica, the pore apertures lie on a level with the surface of the legs. In the legs of the potato beetle, the pore apertures lie at the bottoms of shallow pits. All four preceding species have hypodermal gland pores over the entire body, except the wings. These pores in the lady beetle are perhaps the most highly developed. They lie on all sides and even among the olfactory pores on the legs. In the other three species they are less highly developed on the legs near the olfactory pores and none is found very close to an olfactory pore. This correlation between the hypodermal gland pores and the olfactory pores is certainly a means of preventing the secretion from the gland cells from running into the pore apertures. A large nerve and a large trachea run into oe elytron and wing. In the peduncle of the elytron they run through the radial plate just beneath the group of olfactory pores. Branches from the nerve are given off which connect with the sense cells. The large nerve and trachea passing into the wing soon divide so that a smaller nerve and a smaller trachea run through each main nerve. The largest trachea passes through the subcosta, and the largest nerves pass through the veins carrying the ol- factory pores. These nerves give off branches which connect with the sense cells. The sense cells wherever found are always surrounded by blood. In the experiments to determine the location of the olfactory organs, 434 individuals were tested. These belonged to I1 species representing 8 families. After the antenne were pulled 450 N. E. MCINDOO. off, 4 of the 11 species tested were normal and 7 were slightly abnormal in behavior. After the elytra and wings were pulled off I species was normal while 4 were slightly abnormal in behavior. After the elytra were pulled off and the wings were cut off, the 1 species tested was normal in behavior. After the elytra and wings were pulled off and the pores on the legs were covered with vaseline, the I species tested was slightly abnormal in behavior. After the elytra were pulled off, the bases of the wings glued and the pores on the legs covered with vaseline, the I species tested was normal in behavior. Four unmutilated species responded to odors more slowly than did the same species after the antenne had been pulled off. This is explained by the fact that most beetles are more or less “timid” for some time after being caught, and some feign death. Asa rule, the longer they are confined and the more they are handled, the more satisfactory they are to experiment with. Five species without antennz responded to odors as promptly as did the same species unmutilated. Two species without antennz responded to odors more slowly than did the same species unmutilated. Since these were abnormal in behavior and judging from the reaction times of the other 9 species with antennz pulled off, it is only reasonable to attribute the slow reaction times of these two species to their abnormal condition caused by the antenne being pulled off. The 6 species so mutilated that most of their olfactory pores on the elytra and wings were prevented from functioning responded from 2 to 5 times more slowly than did the same species unmutilated or with the antenne pulled off. The two species so mutilated that most of their olfactory pores on the elytra, wings and legs were prevented from functioning responded from 2 to 6 times more slowly than did the same species unmutilated or with the antenne pulled off. From all the preceding results, it seems that the antenne do not carry any of the olfactory organs, while the olfactory pores found on the peduncles of the elytra, on the dorsal surfaces of the wings, on the trochanters, tibia, sometimes on the femurs and tarsi, and perhaps on the mouth appendages, are the true ol- factory organs in beetles. THE OLFACTORY SENSE OF COLEOPTERA. A5I DISCUSSION. Since the writer (’14c) has already written a complete review of all the literature available concerning the sense of smell in insects, only a brief discussion is necessary in this paper. Hicks (’57) says that the olfactory pores in Coleoptera are arranged in long rows along the subcostal nerves. The same author (’59) states that in Coleoptera these organs are highly developed and occur in numerous groups on the subcostal vein, mostly at the widest part, but are also scattered along it to the fold of the wing. In Carabus they are found on veins other than the subcostal. In many beetles the pore is overarched by a hair, which probably protects the organ. He could distinguish no sexual differences in these organs, except the pores are slightly larger in the females, due to their greater size. Hicks (’60) first found the olfactory pores on the legs of beetles. The present writer has never seen a hair overarching an olfactory pore. Hochreuther (12) seems-to be the first to study the internal anatomy of the olfactory pores in beetles. Since he used only Dytiscus marginalis and perhaps because he did not have enough sections through these organs, he failed to understand their anatomy. He states that each dome-shaped organ is located at the bottom of a chitinous flask, the mouth of which communicates with the exterior. Instead of the peripheral end of the sense fiber coming into direct contact with the air in the flask, it apparently stops just beneath the chitinous dome at the top of the organ. His terminal strand (Terminalstrand) may be the same as the hypodermal secretion forming the cone described by the writer. Hochreuther found a few of these dome-shaped organs on the epicranium near the margin of the eyes, nine on the proximal end of the first antennal segment, two on the distal end of the second antennal segment, a few on the dorsal side of the labrum, a very few on the dorsal side of the mandible, several on each maxilla, about 18 on the first four segments of the front legs, about 10 on the first three segments of the middle legs, and a few on the trochanters of the hind legs. He evidently did not examine the wings. Thus according to Hochreuther these organs are rather widely distributed. Since the peripheral ends of the sense fibers do not come into contact with the outside air, 452 N. E. MCINDOO. but connect with the tops of the domes, he suggests that they receive some kind of mechanical stimuli, although he performed no experiments to determine their function. Lehr (14), resuming the search for sense organs in Dytiscus marginalis where left off by Hochreuther, found dome-shaped organs on the elytra and wings. He found three main groups in identically the same places as described by the present writer. The number of pores in the group on the elytron varies from 130 to 150. The two main groups on the radius (his subcosta) of the wing are large, but he did not count the pores in them. He found a fourth group, consisting of about 30 pores, on the ventral side of the costa near the base of the wing. He also found a few scattered pores on the dorsal side of the costa just distal to the fold of the wing, a few on the second cubitus, and a few irregularly scattered along the full length of the media. Lehr has described the anatomy of these organs almost identically as seen by the present writer, but it seems that he has not correctly interpreted some of the structures. He seems to think that each sense cell is surrounded by another cell, but the latter cell is perhaps nothing more than coagulated blood and the portion of it extend- ing into the pore is certainly a hypodermal secretion forming the cone as described in the preceding pages. His neurilemma nuclei are perhaps hypodermal nuclei. He is able to trace the sense fiber through the cone, but he has not recognized the small opening through the dome. ‘This is not surprising, because the pores in the wings as so small that the openings or pore apertures are never noticed unless first seen in the largest pores in the legs or mouth parts. In the thinnest sections, the chitin forming the dome is so thick as compared to the diameter of the pore aperture that the aperture appears only as a streak slightly lighter than the other chitin in the dome. Lehr has nothing to say about the physiology of these organs. In experimenting with mutilated beetles, Hauser (’80) seems to be the only one who has taken their longevity into considera- tion. And even he has not kept an accurate record of their behavior and longevity. He claims to have studied the behavior of beetles before and after the removal of the antenna. When the antennze were removed he ascertained that many beetles THE OLFACTORY SENSE OF COLEOPTERA. 453 soon became sick and died, while others lived thereafter for many days. When tested with odors, most of the beetles without antennz failed to respond, but Hauser states that Carabus, Melolontha and Silpha still responded to odors, although more slowly. PorPI MPPg PeorP] P| a ' Fic. 3. Antennal organs of the water beetle, Dyliscus marginalis, copied from Hochreuther (1912). A, small tactile hair (Sinneshaar) from first segment of an- tenna, total preparation (Fig. 1 from Hochreuther), X 330; B, portion of Fig. 12 from Hochreuther, showing four small sense bristles (Sinnesborsten) from proximal end of second segment of antenna, X 265; C, longitudinal section (Fig. 48 from Hoch- reuther) through a hollow pit peg (hohlen Grubenkegel), X 470; D, longitudinal section through a small massive pit peg (massiven, grubenstandigen Zapfen) and two pore plates (kelchférmige Organe), X 590. This drawing is a combination of Figs. 32 and 58from Hochreuther. Only the pore plates (PorPl) are taken from Fig. 58. Hochreuther gives a drawing of only one perfect pore-plate organ, or cup- shaped organ, and it is from the maxillary palpus. CM, cup-shaped membrane; HPPg, hollow pit peg; MPPg, massive pit peg; PorPl, pore plate; SB, sense bristle; THr, tactile hair. See page 456 for other abbreviations. A54 N. E. MCINDOO. For the purpose of judging whether the antennal organs are better adapted anatomically than the olfactory pores for receiving odor stimuli, the former organs (Text-fig. 3, p. 453), of Dytiscus marginalis have been copied from Hochreuther (12). This work of Hochreuther is a comprehensive study of the morphology of all the chitinous sense organs of Dytiscus. Since it is perhaps the latest and certainly the best study on the antennal organs of beetles, these organs shall be briefly described. Each of the 11 segments in the antenna of Dytiscus carries a number of sense organs. The farther from the base of the antenna the more numerous they are. The distal half of the antenna is covered abundantly with sense organs, while the proximal half is sparingly covered with them. The first and second segments are well provided with slender tactile hairs (Text-fig. 34, THr) which have been called Sensilla trichodea by Schenk. These hairs are also found on all the other appen- dages and even on the head, thorax and abdomen. Two groups of sense bristles (Text-fig. 3B, SB), called Sensilla chetica by Schenk, lie at the proximal end of the second segment. These hairs are also common on most of the other appendages, on the head, thorax and abdomen. All segments, except the first one, are well provided with small massive pit pegs of the thick-walled type (Text-fig. 3D, MPPg). All segments, except the first one, are only sparingly provided witha second type of pit pegs. This one is the hollow or thin-walled type (Text-fig. 3C, HPPg). Only about six of these were found on each segment. Besides being found on the antenne, both types of pit pegs are common on all the mouth parts, on the mesothorax, around the spiracles, on all the legs, and on the sexual apparatus. Pit pegs have been called Sensilla coelloconica by Schenk. All segments, except the first two, are abundantly supplied with the cup- shaped or pore-plate organs (Text-fig. 3D, PorPl). For both antenne they are estimated between 4,500 and 5,000. These organs are also common on the palpus of the first maxilla. They were first studied by Nagel on the antenne and maxillary palpi of Dytiscide. In the honey bee Schenk has called them Senszlla placodea. Of the five antennal organs of Dytiscus, only the hollow pit pegs are regarded by Hochreuther as probably ol- THE OLFACTORY SENSE OF COLEOPTERA. 455 factory in function. If they really act as olfactory organs, then the mouth parts, thorax, legs and sexual organs must aid in receiving odor stimuli. Hochreuther considers the antennze more important as appendages for carrying organs for receiving mechanical stimuli rather than those receiving chemical stimult. According to various authors the antennal organs of different beetles vary only slightly. The antennal organs of Dytiscus are also similar to those of the honey bee. In both of these snsects the tactile hairs are of the same type. The Forel flasks and pit pegs of the honey bee are two types of pit pegs which are perhaps rudimentary, because the tips of the hairs do not come to the exterior of the chitin. The massive pit pegs, hollow pit pegs, and the sense bristles of Dytiseus are certainly nothing more than three types of tactile hairs. The hollow pit pegs compare closely with the pegs of the honey bee, except the pegs have thinner chitin at’ the tips. This is probably on account of more acute sense of touch in the honey bee. The pore-plate organs of the honey bee and the cup-shaped organs of Dytiscus are also quite similar. ; One or more of the antennal organs of every insect studied have been called olfactory organs, and it is possible that most of these organs may be fourd on other appendages, besides the antenne, as already seen in Dytiscus. In conclusion it seems beyond a doubt that none of the antennal organs of beetles shown in Text-fig. 3 serves as an olfactory organ, and that the olfactory pores are well adapted anatomically for receiving odor stimuli, because the peripheral ends of their sense fibers come into direct contact with the external air. LITERATURE CITED. Hauser, Gustav. *80 Physiologische und histologische Untersuchungen iiber das Geruchsorgan der Insekten. Zeitsch. f. wiss. Zool., Bd. 34, Heft. 3, pp. 367-403, with 2 pls. Hicks, J. B. ’57 On a New Organ in Insects. Jour. Linn. Soc. London, Zool., Vol. 1, pp. 136-140, with r pl. ’e9 Further Remarks on the Organs Found on the Bases of the Halteres and Wings of Insects. Trans. Linn. Soc. London, Zool., Vol. 22, pp. 141-45, with 2 pls. 760 On Certain Sensory Organs in Insects, Hitherto Undescribed. Jbidem, Vol. 23, pp. 139-153, with 2 pls. 456 N. E. MCINDOO. Hochreuther, Rudolf. 212 Die Hautsinnesorgane von Dytiscus marginalis L., ihr Bau und ihre Ver- breitung am Ké6rper. Zeitsch. f. wiss. Zool., Bd. 103, pp. I-114. : Lehr, Richard. ’t4 Die Sinnesorgane der beiden Fliigelpaare von Dytiscus marginalis. Zeitsch. f. wiss. Zool., Bd. 110, Heft. 1, pp. 87-150, with 45 text figs. MclIndoo, N. E. ’t4a The Olfactory Sense of the Honey Bee. Journ. Exp. Zool., Vol. 16, no. 3, April, pp. 265-346, with 24 text figs. ’t4b The Olfactory Sense of Hymenoptera. Proc. Phila. Acad. Nat. Sci., Vol. 66, pp. 294-341, with three text figs, and 2 pls. : ’14c The Olfactory Sense of Insects. Smithsonian Misc. Collec., Vol. 63, no. 9, Nov. (Publication 2315), pp. I-63, with six text figs. EXPLANATION OF PLATES I. AND II. All figures including: Text-figs. I and 2 are from camera lucida drawings made at the base of the microscope. Figures 1 to 8 inclusive and 22 on the plates are enlarged 465 diameters. All the remaining figures on the plates, except the dia- grams 28, 29 and 31, are enlarged 580 diameters. ABBREVIATIONS. WE SCM esse blood sinus. BMn ee ee basal margin of elytron. Ge otan seie costa. (Gl etn blaerccohg chitin. Chie aera chitin formed before insect emerges into imago stage. Ghia meee oaaheeees chitin formed after insect emerges from last pupal stage. (Gls Sani aes chitinous membrane of pore plate. GME IBN acres: cup-shaped membrane of tactile hair on antenna. : ; COne ne ates chitinous cone. Conds cun = 2 connective tissue. ConT Nuc. ...nucleus of connective tissue. Cx AR ee coxa. Y ASN eae J oe aks femur. TE Dre trac is opti, bes flange of olfactory pore. EO bate pe ete te: where wing folds. GIO Aeon Pee. gland cell. EAP ORS ir sae hollow pit peg on antenna. Ui i Reeton eee ae hair. EBV Ds 3 tic a6, hypodermis. YE bihen eave ears membrane resembling hypodermis which divides the lumen of proxi- mal end of the tibia of Epzlachna into two chambers. FA PEG ees hypodermal cell. HypNuc..... hypodermal nucleus. IES. oa ob dioe hypodermal secretion. MESS A casein: muscle. VID) nye eco muscle disk. MCUs netets media THE OLFACTORY SENSE OF~COLEOPTERA. A57 WWPIP LASS 5 5 Be small massive pit peg on antenna. INS cic Been eR CONE nerve. IN(IB} 3 Seep nerve branch. UNI U hte variety ee neurilemma. NeuNuc..... nucleus of neurilemma. NeurNuc....neuroglia nucleus. Bre Rats ctisicts: boli pit of pore. IACPAMay's baa pore aperture. POvV.Gline oh. 4 pore of gland. UZ OVEL IA eee pore of hair. LOY Zieh eererene pore plate on antenna. J ENOV? Ls ne OR pore of olfactory organ. OTe a ale. pore wall. PorWGl.....pore wall of gland. PorWHr.....pore wall of hair. Je elaine radius. REIS oe ees radial plate. AES beste Uyak small sense bristle on antenna. Oars sense cell. IS Gress sie avon sense cell of tactile hair SS GG tks ae sense cell group. SXCINICS bo p00 sense cell nucleus. SCNucl......sense cell nucleolus. SG oie eee subcosta. Sal Eleca Bue peor subcostal head. SV AL aay cee sense fiber. TPG ted ost tarsus. TE Die ag Sea Me tibia. TbSp........tibial spine. BELPER ohh lS tactile hair. IE? soe EEG trachea. TrNuc.......Nucleus of trachea. IR Oe eee es trochanter. THEO OMe), 3 aysus a groups Nos. I to 6 of the olfactory pores. (ie, eee ONT location of scattered pores on ventral side of wing. ) Bal @oicdcoc location of scattered pores on tibia. 458 N. E. MCINDOO. PLATE I. Fic. 1. Six of the eight olfactory pores (PorW) in group 6 on inner surface of right hind leg of Epzlachna borealis; also one hair (Hr), one hair socket (porWHr) and two hypodermal gland pores (PorWGl). Fic. 2. Two olfactory pores (PorAp), five hairs (PorWHr) and 19 gland pores (PorWGl) on outer surface at proximal end of right hind leg of Epilachna. Fic. 3. Five olfactory pores from tibial spine of Epicauta marginata (same as shown in Text-fig. 2G). Fic. 4. Group r of olfactory pores on peduncle of elytron of Epilachna (same as shown in Text-fig. 1B). Fic. 5. Seven of the olfactory pores in group 2 on wing of Epilachna. Fic. 6. Eleven of the olfactory pores in group 3 on wing of Epilachna. Fic. 7. Ten of the olfactory pores in group 4 on wing of Epilachna. Fic. 8. Four of the five olfactory pores on ventral side of wing of Epilachna. Figs. 5 to 8 represent some of the pores as shown in Text-Fig. 1C. Fic. 9. Olfactory pore from trochanter of Uloma. Fic. ro. Olfactory pore from tibia of Uloma. Fic. 11. Three olfactory pores from elytron of Uloma. Fic. 12. Olfactory pore and sense cell from wing of Uloma. Fic. 13. Olfactory pore and sense cell from trochanter of Orthosoma (cut slightly obliquely). Fic. 14. Olfactory pore and about one third of sense cell (SC) from trochanter of Orthosoma, showing pit (P) and flange (Fl). Fic. 15. Olfactory pore from tibia of Orthosoma. Fic. 16. Oblique section through trochanter of Epilachna, showing anatomy of leg. It was cut in such a manner that no muscles are shown in the section and that the nerve (VV) is severed in two places. Fic. 17. Cross section through proximal end of tibia of Epilachna, showing anatomy of leg at this place. The gland pore (PorGl), hair pore (PorHr) and sense cells (SC), belonging to the tactile hairs (Hr) were taken from two other sections, and the gland cell just beneath the gland pore was taken from the other end of this section. Fic. 18. Four olfactory pores and a small portion of hypodermis from elytron of Epilachna. The material used for Figs. 17 and 18 was from an old adult beetle that had been confined in the laboratory nearly all summer. Fic. 19. Four olfactory pores, sense cells and nerve (NV) from wing of Epilachna. . ne BIOLOGICAL BULLETIN VOL. XXVIII. N. E. MCINDOO DEL. 460 N. E. MCINDOO. PrAnm ite Fic. 20. Three olfactory pores from wing of Passalus. Fic. 21. Six olfactory pores from elytron of Orihosoma. Fic. 22. Cross section through wing of Orthosoma, showing anatomy of wing beneath olfactory pores. Fic. 23. Olfactory pore from elytron of Passalus, showing sense cell (SC) connected with pore aperture (PorAp) and with nerve (NV); also hypodermal cell (HypC) that forms the cone (Con). Fic. 24. Olfactory pore from trochanter of Epicauta marginata. Fic. 25. Olfactory pore from trochanter of Epicauta pennsylvanica. Fic. 26. Olfactory pore from trochanter of Leptinotarsa ro-lineata. Fic. 27. Three olfactory pores from tibial spine of Epicauta marginata. The material used for Figs. 24 to 27 had been treated with caustic potash. Fic. 28. Transverse-longitudinal diagram of proximal end of trochanter be- longing to right hind leg of Epilachna, showing internal anatomy of leg and super- ficial view of hairs, hair sockets, gland pores and olfactory pores. The four pores at the right belong to group 6 and the three at the left belong to group 5. Fic. 29. Oblique transverse-longitudinal diagram of portion of peduncle belonging to Epilachna, showing internal anatomy of radial plate (RP), innervation of olfactory pores and a superficial view of a few of the pores in group 1. The transverse portion of the diagram passes through the radial plate in the direction of the line marked “‘a”’ in text Fig. 1B. Fic. 30. Oblique superficial view of olfactory pores on wing of Epilachna, showing sense fibers (SF) connected with pore apertures (PorAp). Fic. 31. Transverse-longitudinal diagram of portion of wing belonging to Orthosoma, showing internal anatomy of wing, innervation of olfactory pores and a superficial view of a few of the pores on radius (R) and media (Me). BIOLOGICAL BULLETIN, VOLe XXVille PLATE ll. N. E. MCINDOO DEL. 4 he Ni hh Ay na Ninn (a pre ti iM 4)