' ' ’ sore . . bu Wat tere te WD Nee pepe ht ae or nee
7 ’ # ‘ ‘ 4 ’ - ' eer) oenew A Soe ere te Se doen vam
. 1 ¥ ane ig to. er eee 2) Stee ee ee ee 2 ee oe re ey fh ace wes
' t i y ¥ ‘ ‘ ‘ ‘ ’ tapes tad TOP ee deka om Oe eT Ce er
' . . i . 7 7 “7 ‘ ry 4 + tered nee te bg hede e Mi Mea ee eM em
' be 4 . » §itg a 1h. 4 ‘ ss on * CP eer? ey ee T5989 suo te
8 . ' 4 : ' ‘ j tat “ HN Oe ete tt bel ehh: Hime bath Wee PTD RL tag em ert ew
L , ’ v if N Oe ee De eS ee ie Pore re on
* ‘ ' ‘ . 4 < 7 4 ar] s= et a Cee ee her er ee wv tee
‘ ” . . « : ‘ tee ey Vor 4 we 4 eee er ee ee
— e . oa - wo ’ ) year, e eee een er Ome eer re Ee
‘ : ' oa ' ' 3 ' ‘ eh ee ee ee ee eed ee Pate res van
, SAR SH , agi ‘ VU DP Bed FRE yee dee tee hohe oe aot n 8 Ged OHH Fy tad Memn cm ah ot phe
' A . Pee ee ee en ee Oe Cree ee yi ey mete
’ ’ r " ‘ ' ' ‘ an name ere sae tod Py 41% 4 bot ohm et
: y ‘ . ' ro ee err) FE He ee Dm eg Lu ape ete ARPA att g
. - . ' . 1 ‘ , ‘ ’ 4 ‘ soy Reeth oe tow Ate « tame) ep he de te bebe amid eA Fhe
. ‘ ' . ' m4 Va Pee ee ee ee ‘ Fh he Heme gy it ea agyetby: Ameen re
' ' ; x, iA ‘ tere «wey eer pep en itt wyeehe eee eee et
‘ cd , , x ' . . . . Hy © , in We ® oF Bred de he rae ba ’ CCRC Rots
. . ot ai ce ‘ ‘ ' ‘ . ’ a4 ree ate a +! rs a CD) tee ee Tr
oe o4 bean ' . vpae ar eee Ch er See eee wr errr]
, 7 ¢ , ‘ er) o, Ot Wes treme Cer See ee eer a
’ i i we ' REE. Say Cee mores 9 alge
' , é a0 -abed> We, vee c 4 engi: os
‘ . . ' ‘ pin a fe rr . Me ED OO OH My Pett
é . ‘ r iaae wd PR ety etme cable Heth A,
’ 3 t* + thes He Ee rei gt pleathabrn eng
. : . ‘ ’ wf i acs + Wier uy ‘ ‘ Pa
’ Fi 1.4 ws : Pree rere Aree ae Seine ree | fos +!
' A . ‘ ‘ oot Cady Ge a eee ghey are Lenore
‘ ’ 7 rf 0 2 : wit eee ' 1 ’ its Spero rain
: ‘ ’ va ran ea ¢ ae de 4 nea
' > 4 oa ‘ ’ ra ieee 8 . OW setts 10 NB Unpibew Oh ne eee eee Mott ne
; ite ; oa an fee tie Perera foi var mer ee oe
” Y . 7 : bes ed te bee oh ipNgihgs statheoane le
. bd 1 ’ sue sa ory ran) Pres yead shes Hats lity
' 5 ‘ 4 rv | an th) . eer ‘ v8 tye ‘ samy Ve we ae ney ere Magen’ "
' + . 2 : ry ‘ * bare . . 7 ' Byer ee tte tpt,
t . ‘ i ene Y st oe ne oon ates Lan em retest *
F ‘ on 4 “se % ted OP A emt tnt tgs
; oe ‘ ae <a , sy P : vs ra heh dot oe ae hed k An Pea etenet
: ‘ ’ ‘ ee ee ap hveee . "
7 } 7 ia in 2 ‘ won rf 8 . ‘ ‘* r . Wi h stghuve mo ery seearie
‘ 4 > bo Fy ee ee er etait . shim ky epee
4 : - 7 . ° ve eny “yee . ate cage ia vie
A . - ‘ n rans F Cs ee ee er, ee t Bed teeny
re ta Geo a ‘ ‘ ree ee sth pay? er ree)
i : j be ae 7 pay 7” et arenas
. ‘ oe 4 vate ew aya of . » pte tis
: . aye ' ‘ eH . c- ' eee wee 4 yt . sete
. r Fl ; wae \ ‘ Vor hey ewes oe Cre Pore
48 ’ ”
‘
a : P '
‘ ‘ ‘ +4 ‘ ‘
‘ 1 : ae 14
ws A ‘ ‘ ‘ “ ‘
. “ ‘ ‘
‘ ‘ ‘ ‘
'
‘ ’ ,
’
ee ‘ 7 i
‘ ’
4
‘ ‘
‘ '
. ‘ te
' 1
i '
. |
’
i r+ . “4
‘
, had
. ‘ ” ‘ ho we +
. i ’ oi te ’ 7
‘ rag , .
ot ‘ . :
a sf : in : . . Ft een yge ”
- i ’ teve ' ‘ ‘ cn ay . wee oe 44
‘ . ' ° ot Parr .
« wh ek ack woe
‘ "a ‘ sl } Aare * Us
‘ oa As ‘ j whe
. fi id er oe , 4te
‘ . ee es | ig oven tte . wPrre
‘ rn ‘ ‘ ¥4 +0 ‘ "s wdiagtee
. Dye oth Go dO te
‘ . é ‘ ; . ie "
‘ i wae ; =. ‘ . .
hae ; ‘ ' ‘ ; eo % out 4
* x ' . eae ’ <a ete
; P > ‘ ’ ' ae) ' P a) iy wer
: de . ’ “pee on vi ren 1 7 .
’ $ oad , ‘ ry gine cs «neta
. - . ' i Y fae Pa i ur sate Wr trae
' . ' , we . ‘ a a a i ‘ wed aw inl -# 68 Ole
4 nn ‘ fer 4 . ‘ ta ? . lone
A * ’ . toes + ¢ , . ‘ ed ’ ‘
’ ' . ‘ .* ~ ? . . . ee ob
rere wo . noe a4 sone woe “ . ‘
. + , ' ' o%u ” , § !
: , i oo oe Vaud ’ its ‘ ae eC Ce ee tore aw ete yrd eta whnatats
, ’ ‘ 4 ate ‘ ceeds ch aeee pie otters
: . ’ , + ’ ’ . 7 is a4 toed fi dre 4 ate sonnet +90 o
: : ; > eae rs ai ’ . - a ‘ ¥
. . my t ¢ ’ 7 se . Ls ‘ .
, eye , + +e re sea . an)
: ’ . . 2 ‘vd vane
ae ¢ . ah € a ab ‘ Hi Oy
. ws . 1 an ve me rece) or ha .
2 ere oa on see
+ - . ‘ ‘ ‘ , re
. . ' ' ' ‘ ‘ . wy oe twa ate tere
ee r > wo sp tp cent od teeters 4 ativan
‘ ‘ wee ' mers) ~ 4 Oe " “ hey saree Pere res PC ee te
’ . . oa . . ‘ ‘ ‘ ‘ . ‘o ie oot ee ri +e eo ete wets Se OF eee de EN AEONN hd OE fe Gi odewinrgsstasgl
. ' ore 1 + +e bet tte nay A sar Ore, ar rerr Srey iene er ae
‘ io ‘4 ‘ ‘ oe -o ” ebe +) same vy ODE BETH eae wereaneN pars ; oaax
ae ‘ a ' . Woetee 4 is aa fd Pepe ee ee ee ee rd Fan dt oe gh
oe poe ' ‘ 1 eee on ‘ % . a4 ae " ee od Et ad le 904 900 Onn ms tO 8 oe tied
‘ . 5 ees ' “4 ’ * ‘ pure ‘ dae We GW AEE e ly om ee le Hiv y ppemianitinee
‘ : * ep ld . oe . ‘ ‘ te “eens ” ert ee re Se etse Try ie eee Ieee Ory
‘ . . + , . ' ‘ tee 4 . eeote tee, e & fe ore mw tig W peidnwwate ortere
a 4 , ‘ * «8 vot ieyen Peee re eee the etme ee howe
. ‘ ' . ve e- ' oe ‘ eae * ’ a5 as . Pe eee ed SPO bm A GO a ee BI GA eas
. ® ' oe ra ee » tire & tae ’ 4400 ORF whi ot san ae SHUT O Mh Hee ed rd men
= ‘ 14 pao Va tee ‘ ot 2 Wed ee wa RRA RE A hl ed
¢ oe ‘ wu bas a4 oe ' ‘ 1.900) Ww) ecb #8 he ' eae Be Asa yaa ood ee lod De a
’ ' ' Pe er . ef . taoes sey ee eo ere |
, ‘ ac) re | te ‘ + ‘ ‘ Ca , . he Le. eee eee Pee Ce Pe)
> ae , so 4 -4 Coreen eC SU
“ ees ‘ ¥ ‘ . onenwe ae deen ete tnad tiene my
“ ‘ ' be o4e ' ‘ ' soe eo Maar :
> fiver gad : ; ; io. he ROLES Aw OA APD Pe of 88
. ‘ “ i 1 'y “ . 1 Odaee comarca
he ' ‘ re ee af . ‘ ou teal see 14 phbepe er Cuan tetetal an
5 ’ . : roe hgnee tig ree
‘ ‘ , , ane is tos ' Cee We eee seems eae
ee ee ' i + . aes . _> ”
8 OP eR ee ee J ’ ef see ror ne * ’ o . . ’ . 0 ek ute ee
Poteet! : 5 . . oo Pa Te ee eo Vedat play veg tered ‘ nade
ouedbag> ane ‘ toa t ° poalere ‘ . ‘ f+ ‘ toes re ee ee ayhe au Oo nt ote Se ee od ee
etme Pa ay a 3 ‘ . re F, , ‘ re oo ee ee oe ee) « © OE ow Ged ene
ft0 eB tanh pone pu ' ' i Pe Ha ‘ thon ' . a) ” ’ tiete de “er Oo oh gre rsae .
ae houate ‘ ’ oe ts gee Ce Pe ee ae re ee ee Pee va OPT eee
woe . . sete ee ’ to. oa ae pte eh ee ee ee err ce ne sine
oa 5-4 1 over . re a oe) hd a oe eee miata ep ONE ae ee ee ee eee ae “ne
' ‘ oon oe ne . oo oy i rae on ad fy Cee eery i eer ey eer eee oo
: . . . see . ” oe + . te ee er ae re ee Th eel
’ ‘ ' oeta . a ce ed . wee viet oe oe fa Ly; *# hea ere ie cee Pere er ae rr verre rer) er ie te ee ete en Te
' . . eee oe p65 8) 2 A “ere WPT rat pee 6 pee Sod ew dee ater emp o. Te
or ta ee toa weler se ¢ ’ shrew tine oe Ce oe Eee orient) Some eee cee .
: one i.8 ‘ ‘- oe “ cdi dee tw te bowed eo oe, ee | ee ae eet RL Ree
F m whee rear rate ti Bavliaie i“ fern the
Fi 4 re bo We pee . iene ‘ eee ya 4 be we woot we rn Fs 10 pte 5
ve oh ’ . . “* ‘ . vor - v an aree Rants a 4) Pe ermhartow base om NN het) ata
* Sa fe 7 < ee a 4 eee eee | Pe ee oe “ gterger id On
; , . 4 ‘ . i i a Cn cr a ee Meg Opatar ade etn ate 45 QeQateres
on 28" BRL ‘ tye onme ven ' ot be pte oe, “toe “a0 “ le ee a ee ee ee
¥- 7 es Py be e+ oe ce ee ee ere er ee i re so wer ae Ahm > he eed &
7 <i ’ we ’ weer * * ‘ + . “* perme aed Pe) aren rote aetaw Cad
, ot . _ 4 a) i sig ata t ieee ’ ‘ orl bade halved) so Oe acme red eee dt Oe
‘4 ’ - ’ Cry epee . ee ‘ Oba ae we mie teeta lem wie Orbe ehale
of ‘ + * we ‘ 4a 88 “oF ba pee whe i tr SC oe eee ern ey I Orerrety pat
’ tae ee - tee ot rho De em at oho oC eee te Pe er eee ee)
. . ” . pete poet gi oa Ce en ry eaters er eee were mere eT re) pea segregate
. bi ma eek eee we teteew ue . eutmwrdae
at ‘ ' ’ oe oe 4 se sites Siacs Ht ee , Hy Bad 4d Uke a time rere
v4 ’ reer . ar) ‘ Cary nr eee | at) v@r erm 8 senate Le Lid ear hehdanat ot bbe tebed ee CPt eet ee
‘ ‘ . ae te oa Beep anay » s-ateree ngs wie ar oo ae ousera
A a ‘ or tem ‘ ‘ ates aonwes ’ areraiesa eS a ieee l) eee oat
eb oaabon woe oe eo fc ra a6 ” tue a4 @ eed oissiaieptans alte 2
he beh & Pe bat yee te tena 019 Wg wr octaned gee 24,
; ea Dore oe seme Pe ee yorre woe mwe meters goede pee
aie f Pere pee) v4 1 bee od ‘eo 0nn cae oe Pere rt Le
‘ +b be tae . oie hoe * . = ote stecgee arae Tasteurad bebe dod m
ty “ aes ae ae vere ie See ae ee | mae weed one SRO a Ae Oe OPA es! Ee Te ite Mant he
’ ' seb bathe than : mae 4 eee ee Pe ee ee ere er ee) eo ene aL eo ey
' . ’ ; Pov eohg lege @ dy ‘ bie 8 sack dew) Oe ae Pare tees) Oper betes * A rasardi rd 0 ee aoe Pre ee ee ee
7 7 ‘ re .-* ron ° . Cee Pe ee Cer er eo) FEOTE ONE OY cote
one ‘ | te will A ae? ume reane oe ow ate ern Peek aha dete eae ote rd
’ ry . ‘ eve reer ee me ery ee | or ’ ‘ aoe wee wee Te tener tts ATR ere ee mero
‘ ‘ A “8 ' " ree ee erect ae ee Oks ee Se Pe er ee eo eee es aT
‘ ’ ‘ ' s ve sor 8 + te ober aene vow Or erartheds les thas deud Siar 4) ah
‘ - ' ‘ aes 1 7 Phe se eae RaOCor tt Cer ee vm s
e+ ‘ ‘ a4 . oi 4 . . . ue -¢ te Ooowretes ae ee 4 Oh ore 4 twheore ae Ateeunre ae OMe Pe eee? Pee Pe ce oe
: ‘ .- , oe tas oe as ee ee ee Pc ee ee Re ee ee ce ms iwi “f * Lente bby Cha
. i . -% ww er he Cr i ee ae ee re eee ee “ tour as we
P os oe ee on oe tee Oh ones) a HHO Od oe eee yore ia shee eee avidsacd ind
ca ‘ ‘ arare ee Se ee Tee) widow aw ete ewanee
, 4 ' 4 ,* . dae, 7 i ee ee ee aa Orde
. r , re ‘ so hee ee@e . Renna ' et er ee Or
" * sae * ’ - notre bey it *o8 bebe avac Me le Le er re eee lL
oF ’ ’ of . on “* ue rad de Wnetee eo new bed de dediee ind baked be he se
“: ; tre i xe te a= ae ‘ Sia ni aa aaah ae Sacra “ 7 aioe
< ' . . ' soe. . ‘ oe eae ase dre wilde Lcd “ |
' . or ‘ ' , ' ‘ OR ed RAR A TAL Wihon lea ag
’ , ’ hig eee . 6 ? a #19 09 se Be ” enters’ an eee
whe 4 ’ 1 ’ , . o+ ‘ woweaedel de peti ot boat itt
. ’ ‘ a oem “. ty ee ee ot een ey peste 6 ste aes
' so oe ‘ rn toe ’ ee ee a) br boda @ 4 ep e teen @ Ee ee ee
‘ . ' . so ’ erg@ets Pe ase ‘ones 9 OO, OF Wore wert Ot ew® iaktg ode
14 ' © ory eer e dyna vee Overraare ven So era ee Ee oe CoE eee core ee oe
‘ o oe hae » > 1 . . ' . sti rtaen . bean oo ot me ‘ eure OP er eee er 2) ey 1 Oe te wan
’ ‘ ee > ‘ or a) . ‘ ae 4 one er gia aces re ee 6 . ot aterse bee Went mw dey ee ee) 1 ae el ibe
' hee ie « te ew ew we ey ee ' re ee Oe Oe ee et ce bee icegusucere avers eed

¢

e-— 7
= =
=

|
|

Si
| r t

THE
BIOLOGICAL BULLETIN

PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY

Editorial Board

DAVID W. BISHOP, Carnegie Institution of
Washington

HAROLD C. BOLD, University of Texas
FRANK A. BROWN, JR., Northwestern University
JOHN B. BucK, National Institutes of Health

LIBBIE H. HYMAN, American Museum of
Natural History

J. LOGAN IRVIN, University of North Carolina

V. L. LOOSANOFF, U. S. Fish and Wildlife
Service

C. L. PROSSER, University of Illinois

BERTA SCHARRER, Albert Einstein College of
Medicine

FRANZ SCHRADER, Duke University
WM. RANDOLPH TAYLOR, University of Michigan
CARROLL M. WILLIAMS, Harvard University

DONALD P. COSTELLO, University of North Carolina
Managing Editor

VOLUME 118
FEBRUARY TO JUNE, 1960

Printed and Issued by
LANCASTER PRESS, Inc.

PRINCE & LEMON STS.
LANCASTER, PA.

li

Tue BrioLocicAL BULLETIN is issued six times a year at the
Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Penn-
sylvania.

Subscriptions and similar matter should be addressed to The
Biological Bulletin, Marine Biological Laboratory, Woods Hole,
Massachusetts. Agent for Great Britain: Wheldon and Wesley,
Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London,
W.C. 2. Single numbers $2.50. Subscription per volume (three
issues ), $6.00.

Communications relative to manuscripts should be sent to the
Managing Editor, Marine Biological Laboratory, Woods Hole,
Massachusetts, between June 1 and September 1, and to Dr.
Donald P. Costello, P.O. Box 429, Chapel Hill, North Carolina,
during the remainder of the year.

Second-class postage paid at Lancaster, Pa.

LANCASTER PRESS, INCz, LANCASTER, PA.

CONTENTS

No. 1. FEBRUARY, 1960
PAGE
BAGNARA, JOSEPH T.
Tail melanophores of Xenopus in normal development and regenera-

BECKMAN, CAROLYN, AND ROBERT MENZIES
The relationship of reproductive temperature and the geographical
pmecio: Eve maine woodborer Limnoria tripunctata.....:......... ne et)
Boyp, Cart M. :
The larval stages of Pleuroncodes planipes Stimpson (Crustacea,

Ee Ne NCICIAC) OK ee Ce ee ea oe ie IVA
Crisp, D. J., AND B. S. PATEL
eaeramicime cycle in Balanus balanoides L.... 2. oe ee 31

Davis, HARRY C.
Effects of turbidity-producing materials in sea water on eggs and larvae
Crem | Venus, (Nercenaria) mercenaria). . oo) o0sbe se Se kek A8
EPPLEY, RICHARD W., AND CHARLES C. CYPRUS
Cation Pe ietion and survival of the red alga, Porphyra per lores, in

MEE MILe EOMCCIILicited sea water... lute ee cee Fe A ee ee 5D
GESCHWIND, |. I., M. ALFERT AND C. SCHOOLEY
The effects of thyroxin and growth hormone on liver polyploidy...... 66

GREGG, JAMES H.

Surface antigen dynamics in the slime mold, Dicty ostelium discoideum. 70
HANLON, Davin P.

The effect of potassium deficiency on the free amino acid pattern of the

muscle tissue of protein-maintained Fundulus heteroclitus............ 79
Hoz, GEORGE G., JR.

Structural and functional changes in a generation in Tetrahymena.... 84
KOHLER, KURT, AND CHARLES B. METz

Rear the Sea Urchin, sperm surtace ss, 035 ee ee ee 96

MILBURN, NANCY, ELIZABETH A. WEIANT AND KENNETH D. ROEDER
The release of efferent nerve activity in the roach, Periplaneta ameri-
mn OX ActS OF ue COPpPuS CardiaCum. 0 oa ees oe ek eee ‘ig Ml
OHTSUKA, EIJI
On the hardening of the chorion of the fish egg after fertilization. IIT.
The mechanism of chorion hardening in Oryzias latipes...........'... 120
PASSANO, L. M.
Low temperature blockage of molting in Uca pugnax................ 129

iil

IV CONTENTS

TrAvis, Dorotny F.
The deposition of skeletal structures in the Crustacea. I. The histology
of the gastrolith disc skeletal tissue complex and the gastrolith in the
crayfish, Orconectes (Cambarus) ‘virilis:dagen-—_Decapoda.. . sys
Topp, Mary-ELIZABETH, AND PAuL A. DEHNEL
Effect of temperature and salinity on heat tolerance in two grapsoid
crabs, Hemigrapsus nudus and Hemigrapsus oregonensis............

No. 2. APRIL, 1960

ALLEN, KENNETH, AND J. AWAPARA

Metabolism of sulfur amino acids in Mytilus edulis and Rangia cuneata.
CostTLow, JOHN D., Jr., C. G. BooKHOUT AND R. MONROE

The effect of salinity and temperature on larval development of Sesarma

cinereum (Bosc) reared in the laboratory, .75.. =... 2
CosTLow, JOHN D., JR., AND C. G. BOOKHOUT

The complete larval development of Sesarma cinereum (Bosc) reared

inc the laboratory is. 675. aioe seein ee i ee
DEHNEL, PAUL A.

Effect of temperature and salinity on the oxygen consumption of two

intertidakicralys; e235 be ees bes ee aie ee at eee
DURAND, JAMEs B.

Limb regeneration and endocrine activity in the crayfish.............
GEORGE, J. C., AND C. L. TALESARA

Studies on the structure and physiology of the flight muscles of birds.

9. A quantitative study of the distribution pattern of succinic dehydro-

genase in the pectoralis major muscle of the pigeon.................%
GOLDSMITH, Mary HELEN M., AND HowarpD A. SCHNEIDERMAN

The effects of oxygen poisoning on the post-embryonic development

and behavior of 4 chalcid*wasp.’... .40 3 ea ee
JOHANSEN, KJELL

Circulation: in the -haghish, Mysxine clutinosa 14.3]. 4. =. ae
RIEGEL, J. A., AND L. B. KIRSCHNER

The excretion of inulin and glucose by the crayfish antennal gland....
RIZKI, NE TOM.

The effects of glucosamine hydrochloride on the development of Droso-

pluila melanogaster t-te ee See adi sdhs Access & ohn et eee
ROSENBAUM, ROBERT M., AND CARMEN ROLON

Intracellular digestion and hydrolytic enzymes in the phagocytes of

PIAMAETATIS: Fo PF cancye, Sosa PES Ek aw ESO we ing 0 ee
SHIRODKAR, M. V., A. WARWICK AND F. B. BANG

The in vitro reaction of Limulus amebocytes to bacteria. .......:.:2%
TELFER, WILLIAM H.

The selective accumulation of blood proteins by the oocytes of saturniid

moths... eG Ed cil Fae cab nena ae oa ae rss 2e ive Dacre ere cea
TELFER, WILLIAM H., anD L. Davin RUTBERG

The effects of blood protein depletion on the growth of the oocytes in the

Cecropia mothe i vuk eee 2 cick Ss ee lhe Renee

|

150

i73

183

203

2S

250

262

269

289

296

308

31S

324

338

CONTENTS

No. 3. JUNE, 1960

Brown, F. A., Jr., W. J. Brett, M. F. BENNETT AND F. H. BARNWELL

Magnetic response of an organism and tts solar relationships..... .
Brown, F. A., Jr., H. M. WEBB anp W. J. BRETT

Magnetic response of an organism and its lunar relationships.........
FINGERMAN, MILTON, AND WILLIAM C. MOBBERLY, JR.

Investigation of the hormones controlling the distal retinal pigment of

fae eee Me NC TTVONEUCS HF 5 Fcc. fy se oie Wyse Sk le ea ee Me die ie
F6yN, BJORN

ae ecurimiaciitance mi Lllya sown ee a ee ee ee a
FREEMAN, JOHN A.

Influence of carbonic anhydrase inhibitors on shell growth of a fresh-

eeeteomatieee iy sa NeterOstroplia. (isc. oe ee ee eo he bad

—}> GOREAU, THOMAS F., AND Nora I. GOREAU

The physiology of skeleton formation in corals. III. Calcification rate

as a function of colony weight and total nitrogen content in the reef coral

Mn IMReOla Ga. (Minnaecuis) 2 vais tS nko a ee ee ek
Lynn, W. GARDNER, AND JAMES NORMAN DENT

The action of various goitrogens in inhibiting localization of radioiodine

meee cuyrord and thymus glands of larval tree toads.......52...0.%..
Metz, CHARLES B.

Investigation of the fertilization inhibiting action of Arbacia dermal

SR DETTE | J SL Ce gis Se iiss, oh peti ik ean gs i ieeae aca Mey ne eas ge EAN
SPIEGEL, MELVIN

Rmeemenanees Ii -Cevelopment.(.. hee aed eas Ce ho Shee we
TRIPLETT, EDWARD L., AND SUSANNE BARRYMORE

Tissue specificity in embryonic and adult Cymatogaster aggregata

PP ON ase PASIAN ta LOIN. (3 ik ee rele Mae Med es Ges
TRIPLETT, EDWARD L., AND SUSANNE D. BARRYMORE

Some aspects of osmoregulation in embryonic and adult Cymatogaster

marr road other empiotocid fishes: oo. se. 6 jee wie in eee ee

oS)

407

412

419

430

439

451

463

if ae

A.

>~

5 7.0572

Davip W. BIsHoP, ae Institution of V. L. Loosanorr, U. S. Fish and Wildlife |

| Washington Service

HaRoLp C. ‘Bop, University of Texas. C. L. PROSSER, University of Illinois :
FRANK A, BROWN, JR., Northwestern University BERTA SCHARRER, Albert Einstein College of
JouN B. Buck, National Institutes of Health Medicine —
Lipsie H. Hyman, American Museum of FRANZ SCHRADER, Duke University.
Natural History WM. RANDOLPH TAYLOR, University of Michigan
ae LOGAN IRVIN, University of North Carolina CARROLL M. WILLIAMS, Harvard University
' DONALD P. COSTELLO, University of North Carolina
Managing Editor

Na
Ae

Oh SARE

BIOLOGICAL BULLETIN

PUBLISHED BY fas
THE MARINE BIOLOGICAL LABORATORY |

Editorial Board

FEBRUARY, 1960

Printed and Issued by

LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.

aman AES A Cay , Pera elke (oy (Number I.
K Y Ny L/ ‘e ; ; bf j - } Lf ' \

\

PMP >>

: —~ q A . en S|
f « : 3 — |

The Biological Bulletin, accepts papers on a variety of subjects of biological interest. In -
general, however, review papers acon ahewa written at the specific invitation of the Editorial
Board), short preliminary notes and papers which describe only a new technique or method
without presenting substantial quantities of data resulting from the use of the new method cannot
be accepted for publication. A paper will usually appear within three months of the date of
its. acceptance., a eeTE CK |

The Editorial Board requests that manuscripts conform to the requirements set below; —
those manuscripts which do not conform will be returned to authors for correction before they |

\

are refereed by the Board.

“#4. Manuscripts. Manuscripts must be typed in double spacing (including figure. legends,
foot-notes, bibliography, etc.) on one side of 16- or 20-lb, bond paper, 83 by 11 inches. They

f

should be carefully proof-read before being submitted and all typographical errors corrected _

legibly in black ink. Pages should be. numbered. ‘A left-hand margin of at least 1} inches |
should be allowed. el, re 3

2. Tables, Foot-Notes, Figure Legends, etc. ‘Tables should be typed on separate sheets and
placed in correct sequence in the text. Because of the high cost of setting such material in type,

authors are earnestly requested to limit tabular. material as much as possible. Similarly, foot- _

notes to tables should be avoided wherever possible. If they are essential, they should be indi-
cated by asterisks, daggers, etc., rather than by numbers. Foot-notes in the body of the text.
should also be avoided unless they are absolutely necessary, and the material incorporated into
the text. - Text foot-notes should be numbered consecutively and typed double-spaced on a'sepa-

rate sheet. Explanations of figures should be typed double-spaced and placed on separate sheets.

at the end of the paper. e : M4
3. A condensed title or tunning head of no more than 35 letters and spaces should be included. _

4, Literature Cited. The list of papers cited should conform. exactly to the style getina

recent. issue of The Biological Bulletin; this list should be headed LITERATURE CITED,
and typed double-spaced on separate pages. ee x et !

5. Figures. ‘The dimensions of the printed page, 5 by 73 inches, should be kept in mind in ; |

preparing figures for publication. [lustrations should be large enough’so that all details will be.

clear after appropriate reduction. Explanatory matter should be included in legends as far as ~

- possible, not lettered on the illustrations. Figures should be prepared for reproduction as line

cuts or halftones; other methods will be used only at the author’s expense. Figures to be repro-_
duced as line cuts should be drawn in black ink on white paper, good quality tracing cloth or
blue-lined coordinate paper; those to be reproduced as halftones should be mounted on Bristol

Board, and any designating numbers or letters should be made directly on the figures. AJ]
figures should be numbered in consecutive order, with no distinction between text- and plate-

figures. The author’s name’should appear on the reverse side of all figures, as well as the desired

reduction. ; 4 TAY

6. Mailing: Manuscripts should be packed flat ; large illustrations may be rolled in a mailing

tube. All illustrations larger than 83 by 11 inches must be accompanied by photographic

reproductions or tracings that may be folded to page size. —

Reprints. Authors will be furnished, free of charge, one hundred reprints without covers.

Additional copies:may be obtained at cost; approximate prices will be furnished by the Managing —

Editor upon request.

THE BIOLOGICAL BULLETIN |

THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and
Lemon Streets, Lancaster, Pennsylvania.

Subscriptions and similar matter should be addressed to The Biological Bulletin, Marine —
Biological Laboratory, Woods Hole, Massachusetts, _ All subscriptions expire with the December

issue and are renewable prior to.the next succeeding February issue. Single numbers, $2.50.
Subscription per volume (three issues), $6.00.

Communications relative to manuscripts should be sent to the Managing Editor, Marine
Biological Laboratory, Woods Hole, Massachusetts, between June 1 and September 1, and to
Dr. Donald P. Costello, P. O. Box 429, Chapel Hiil, North Carolina, during the remainder of
the year. | .

Second-class postage paid at Lancaster, Pa,

a

BACK ISSUES

The Marine Biological Laboratory
needs many numbers of

THE BIOLOGICAL BULLETIN
to replenish its stock, greatly de-
pleted after meeting the needs of
individuals and libraries during the
last 25 years. Persons willing to
contribute any numbers, volumes,
or runs may send them, express col-
lect, to the
Marine Biological Laboratory,

Woods Hole, Massachusetts

ST
Fak
Ho a

SOE aa
Bie! ate

WAVELENGTH IN MILLIMICRONS

400 450 500 550 600 650 700 750 800 650 900 950

Didymium transmission curve showing resolution
obtainable with Spectronic 20

Jom. ROTO-CELL

. for converting Spectronic 20
into an inexpensive spectro-
photometer for control,
research, teaching, etc.

9085-C.

ROTO-CELL. A double cell in swivel action
carrier with water-jacketed housing, for rapid
scanning at controlled temperatures.

Readily interchangeable with the single
place sample holder of Spectronic 20.

Finger tip rotation of knob introduces
either cell chamber into light beam and pro-
vides dark current check.

9085-C. Roto-Cell, Thomas, with double 1
ml cell of Vycor brand glass, with cover; with
Oem heht path as. |: ce eee eee 80.00

9084-E. Spectrophotometer - Colorimeter, B. & L.
Spectronic 20. A compact, direct reading, versatile
instrument, simple to operate. Range 375 to 950
mmu, band width 20 mmu. Reads in transmission
and optical density. Complete outfit for rapid
spectral scanning and plotting of absorption curves,
consisting of blue and red-sensitive phototubes
and filter, voltage stabilizing transformer and
Thomas Roto-Cell with double 1 ml cell for 10 mm
light path. For 115 volts, 60 cycles, a.c....381.50

A one-piece, double 1 ml cell, 10 mm light path,
with liquid-cooled carrier

Provides instantaneous interchange within the
imstrument of sample and blank for speed in
plotiing absorption curves

9084-E.

Copy of Bulletin 121 sent upon request.

A.H.T. CO.

PHILA., U.S.A

LABORATORY APPARATUS

ARTHUR H. THOMAS COMPANY

More and more laboratories rely on sin

Laboratory Apparatus and Reagents

VINE ST. AT 3RD ¢ PHILADELPHIA 5, PA,

Vol. 118, No. 1 February, 1960
TE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

fee ME LANOPHORES OF XENOPUS IN NORMAL DEVELOPMENT
AND REGENERATION *

JOSEPH f BAGNARA

Department of Zoology, University of Arizona, Tucson, Arizona

A peculiar group of light-sensitive melanophores occurs in the ventral fin of
Xenopus larvae. Bles (1905) noticed that these chromatophores are contracted
during the day, but expand at night, darkening the tail markedly. More recently,
Bagnara (1957) has shown that the melanophores of the tail fin are directly photo-
sensitive and refers to this response as the “tail darkening reaction.” His sug-
gestion that a photochemical system is involved has been supported by Van der
Lek, De Heer, Burgers and von Oordt (1958) who point out that the degree of
melanophore response is a function of light intensity.

Because of the above observations and in view of the report that the numerical
density of tail fin melanophores is only slightly influenced by the hypophysis
(Bagnara, 1957), it seems likely that the presence or absence of light, not the
pituitary, is the primary effector of tail melanophore response. The chromato-
trophic hormone (CTH) of the hypophysis exerts some influence on these cells,
however, as was shown by Thing (1952) who used the tail fin melanophores for
assay of such hormone preparations.

It is curious that these melanophores are restricted to the distal half of the
ventral fin (Bles, 1905; Bagnara, 1957). Apparently, the tip of the tail provides
an environment favorable not only for the establishment of a light-sensitive mech-
anism, but also for melanogenesis. That the caudal portion of the ventral fin is
generally chromogenic is suggested by our previous observation (Bagnara, 1957)
that guanophores, which are never present in the tail of normal Xenopus larvae,
develop in the distal portion of such tadpoles which have been deprived of their
hypophyses.

The present study concerns the responses of tail fin melanophores of Xenopus
to changes in illumination during the course of normal development and regenera-
tion, and compares these melanophores with those in other areas of the larvae.

MATERIALS AND METHODS

The larvae of Xenopus laevis used in this investigation were reared from eggs
obtained from natural spawning of our adult stock which are kept in outdoor tanks.

1 Aided by grants G-4048 and G-6231 from the National Science Foundation.
1
Copyright © 1960, by the Marine Biological Laboratory

MITHSONIAN |
NST ITUTION MAR1' = 1960

2 JOSEPH T. BAGNARA

In a few cases, however, ovulation was induced with chorionic gonadotrophic
hormone. ‘Throughout the larval period, powdered nettle was employed for feeding.
Experiments were carried out in a temperature-controlled room at temperatures of
21° C. during the spring and summer and 24° C. during the fall and winter.
Illumination was from north-facing windows.

During some of the experiments, larvae were subjected to darkness for periods
of thirty minutes to several hours. Thirty minutes is approximately the minimal
time interval for achievement of the tail darkening reaction (Bagnara, 1957; Van
der Lek et al., 1958). Dark treatment was carried out in a closed drawer in a
darkened room. After appropriate intervals of exposure to darkness, the larvae
were immersed immediately in 25% formalin. Such treatment insures rapid fixa-
tion and prevents melanophore contraction during the fixing process. The same
method of fixation was employed in all experiments in which degree of expansion
was to be observed. For long-term preservation, the larvae were removed to
10% formalin.

For the tail regeneration experiments, young Xenopus larvae of stages 51
through 54 (stages of Nieuwkoop and Faber, 1956) were used because older
animals usually conclude their metamorphosis before tail regeneration is com-
pleted. Tails were cut off with iridectomy scissors while the larvae were hovering
in a stationary position. The cuts were made just anterior to the pigmented area.
In this way, enough tail remained to allow the larvae sufficient mobility for feeding
and for respiration. Care was taken to use only tadpoles on which none of the
pigmented area of the fin remained.

In order to evaluate the degree of melanophore response, the melanophore
index (M. 1.) of Hogben and Slome (1931) was employed. This index is general
enough to permit analysis even of types of melanophores which differ slightly from
one another in form.

RESULTS

I. Normal development of the fin melanophores

Tail fin melanophores begin to appear at stage 47 (Fig. 1). They are first seen
immediately adjacent to the somitic area at the very tip of the tail. The pigmented
region gradually increases in size so that at stage 51 the distal third of the tail
fin contains many melanophores. By stage 52 the pigmented area has increased
to cover the entire caudal half of the tail. The anterior boundary of the pigmented
region forms a clear line of separation from the proximal portion of the ventral
fin which contains no melanophores at all.

In their early developmental stages, the young tail melanophores are elongated
and sparsely branched. As differentiation proceeds, new branches are added
radially so that the expanded melanophores appear stellate. During stages of
expansion, melanophores of the tail fin differ from those of the head and trunk
in that their processes are thin and much more heavily branched. As tail pig-
mentation proceeds during the early larval period, new melanophores are added
at the edge and tip of the fin. Actually, there is a gradient in degree of differen-
tiation; thus, under expanded conditions, melanophores immediately adjacent to
the somites are fully developed and stellate while those at the edge are elongated
(Fig. 2). Between these two areas exist partially differentiated melanophores

TAL MELANOPHORES OF XENOPUS 3

Ficure 1. First appearance of tail fin melanophores at stage 47. Normal aquarium illumi-
nation. Magnification: x 10.

Figure 2. Ventral fin of stage 53 larva during the tail darkening reaction. Melanophore
differentiation gradient from somites at left to fin edge at right can be seen in these expanded
melanophores. Magnification: x 50.

with an intermediate number of branches. This suggests either that new melano-
phores are proliferating at the edge of the fin or that here unpigmented melanocytes
are coming under the influence of a melanogenic substance which is distributed
along a gradient from the somites to the ventral fin edge.

II. Normal development of the tail darkening reaction

Although not as striking as that of fully differentiated melanophores, the re-
sponse of differentiating melanophores is discernible almost from their first appear-
ance. Under normal room illumination, young melanophores are not able to
assume the punctate form of the fully developed tail melanophores (Fig. 3) ;
instead, they appear elongate with relatively few visible branches (Figs. 1 and 2).
After exposure to darkness for thirty minutes or more, however, their branches
‘become more obvious and often secondary branches can be seen (Fig. 2). Fully
‘formed tail melanophores seen during the tail darkening reaction are stellate and
“possess secondary and tertiary branches (Fig. 4).

IIL. Regeneration of tail melanophores and. the tal darkening reaction

Approximately four days after tail amputation, a new fin can be seen forming
-at the end of the regeneration blastema and at the cut edge of the ventral and

4 JOSEPH T. BAGNARA

Ficure 3. Mature melanophores of ventral fin in full contraction after bright illumination.

Magnification: < 100.
Figure 4. Mature melanophores of ventral fin in full expansion during the tail darkening

reaction. Magnification: x 100.
Ficure 5. Fifth day of tail regeneration. Note melanophores streaming from the blastema

at the line of amputation. Magnification: X 7.
Figure 6. Seventh day of tail regeneration. Melanophore streaming much more obvious.

Magnification: X 7.

os

TAIL MELANOPHORES OF XENOPUS 5

dorsal fins. At this time, the formation of melanophores begins in the new ventral
fin near the base of the somitic blastema. During the next 6 or 7 days, these
melanophores increase in number and appear as a stream of cells stemming from
the regeneration blastema (Figs. 5 and 6). These melanophores are present only
in the ventral fin and remain excluded from the area anterior to the original cut.
Thus, the proximal area retains its unpigmented state. During the course of
tail regeneration, tail melanophores gradually appear in the ventral fin, first making
their appearance adjacent to the newly forming somites. After the tenth day of
regeneration, the somitic core of the regenerating tail appears as a long spike to
which is attached a generous ventral fin. Melanophores are abundant in the upper
half of the fin near the developing somites, but are lacking in the lower half near
the edge. After three weeks, tail regeneration is practically completed and the

Ficure 9. Melanophores typical of the head of mature larvae under normal illumination.
Magnification: xX 100.

new ventral fin is copiously supplied with melanophores. However, a clearly
defined margin is present along the ventral edge of the fin (Figs. 7 and 8) which
is permanently devoid of melanophores. Moreover, the melanophores at the edge
of this margin are completely differentiated and there is no indication of any
newly forming melanophores which might ultimately invade the clear margin. The
unpigmented border extends along the complete length of the regenerated ventral
fin except for the extreme proximal portion of the pigmented zone. This is the
point at which the stream of melanophores was first seen in connection with the
regeneration blastema. Apparently at the start of regeneration, a sufficient stim-
ulus for melanization allows the complete reconstitution of pigmentation in this
localized region. Presumably, in later stages of regeneration, the melanizing

Figure 7. Unpigmented area at edge of ventral fin at twenty-first day of regeneration.
Slight tail darkening reaction is evident because larva was preserved immediately after removal
from a murky aquarium. Magnification: x 10.

Ficure 8. Larva similar to that in Figure 7. This larva, however, was placed in the dark
before fixation; thus, the regenerated fin melanophores are fully expanded. Magnification: x 10.

6 JOSEPH. 1) BAGNARA

stimulus is not sufficiently strong for complete repigmentation and thus the clear
marginal area is formed.

Just as in normal development, the fully differentiated melanophores in the
regenerating fin are light-sensitive (Figs. 7 and 8). Under normal illumination,
the melanophores are punctate in appearance; but after exposures to darkness of
thirty minutes or more, they expand and assume a stellate form. Differentiating
melanophores in the tail regenerate are also light-sensitive just as they are during
normal development (Fig. 2).

IV. Observations on other melanophores

In contrast to the melanophores of the tail fin which expand upon exposure
to darkness, the large epidermal melanophores of the head (Fig. 9) contract after
similar treatment. This reaction was observed empirically by Bles (1905), but
no quantitative estimate of the response was recorded. In the present experiments,
it was observed that the average M. I. for head melanophores of 64 tadpoles
which were kept under normal illumination was 4.7, while the M. I. for 74 tad-
poles fixed after exposure to darkness was 3.3. This response to changes in
illumination is not as marked as that shown by the melanophores of the tail fin
but it is clearly visible even macroscopically on the tadpoles which, except for
the tail, appear to blanch after exposure to darkness.

DISCUSSION

The fact that light-sensitive melanophores develop in the fin of Xenopus larvae
poses several questions of embryological and physiological significance. First of
all, what mechanism allows these peculiar light-sensitive melanophores to form
only in the distal half of the fin, leaving the proximal portion completely free of
this type of chromatophore? Stevens (1954) has shown that the posterior trunk
tailfold areas of Xenopus neurulae are a good source of melanophores; thus the
entire ventral fin has an adequate reservoir of melanoblasts upon which to draw.
Realization of the melanophore potential is another matter, however, and may be
dependent, as Twitty and Niu (1954) have suggested for urodeles, upon migra-
tory responses of melanophores to environmental stimuli. It is possible, there-
fore, that the proximal area of the ventral fin does not provide an “attractive”
stimulus for the migrating melanoblasts and that instead such cells migrate and
complete their differentiation in the more “attractive” distal portion of the fin.
Because of the peculiar melanophore distribution and because of the lack of other
varieties of pigment cells in the tail, it is hard to explain the tail pigmentation
patterns on the basis of physiological antagonisms between migrating chromato-
blasts as has been suggested by Twitty and Niu (1954).

As an alternative explanation, it can be supposed that melanoblasts invade both
the proximal and distal areas of the fin, but for reasons as yet unknown, only the
distal portion of the tail provides a medium favorable for melanin synthesis.
Wilde (1955) has strongly suggested that special metabolism of phenylalanine
is required for normal differentiation of neural crest elements; thus, one might
reason that prevalence of phenylalanine in the distal region may provide the
stimulus for tail fin melanization. Furthermore, a gradient of this substance or
some other melanogenic substance flowing outward from the somite to the edge

TAIL MELANOPHORES OF XENOPUS 7

of the ventral fin would possibly account for the apparent differentiation gradient
observed in the fin melanophores during normal development. Consistent with
this hypothesis, the unpigmented margin along the edge of the regenerated fin
may result from either a gradient which is too weak at its periphery, or from a
diffusion blockage which does not allow penetration of the melanogenic substance
to the edge of the fin.

Another fundamental question which suggests itself is concerned with the
fact that not only are the tail melanophores extraordinarily light-sensitive but
their responses are just the reverse of those of melanophores of other areas. The
present observation that epidermal melanophores on the head and dorsal surface
contract as a result of dark exposure is known for other amphibian larvae: for
Ambystoma, Babak (1910) and Laurens (1917) and for Taricha and Rana,
Bagnara (unpublished). To our knowledge, however, the very marked melano-
phore expansion, such as is seen in the tail of Xenopus, is unique among Amphibia.
What mechanism allows the tail melanophores alone to possess this capacity is
an enigma. Apparently the distal tail fin provides an environment which is favor-
able not only for melanogenesis but also for the development of a light-sensitive
mechanism.

Whether the dorsal melanophores of Xenopus larvae are directly stimulated
by light has not been determined; however, one is led to think that this is so
because of the observations by Laurens (1917), who showed that the dorsal
melanophores of blinded Ambystoma larvae behave toward light like those of
intact animals. His additional observation that melanophore expansion after re-
moval to light occurs much faster than the original melanophore contraction brought
about by dark-treatment suggests that a photochemical reaction is involved in this
system just as for the tail fin melanophores. Although light sensitivity of the
dorsal melanophores is apparently a common thing among amphibian larvae, it
should be emphasized that the reaction is relatively subtle, with changes of approx-
imately 2 units of M. Il. compared to a change of about 4 M. I. units for Xenopus
tail melanophores.

Coincident with the unusual behavior of the tail melanophores of Xenopus is
the unique form exhibited by these chromatophores. In an expanded state, their
very thin branches lead one to suspect that they have less melanin than the thick
melanophores on the dorsal surface. Perhaps this is a reflection of the apparent
low sensitivity of the tail melanophores to the chromatotrophic hormone. Under
normal lighting conditions, when the dorsal melanophores are expanded under
influence of CTH, the contracted state of the tail melanophores implies that they
are not sensitive to the circulating level of hormone. If the amount of hormone
is raised by injection of CTH (Thing, 1952), the tail melanophores expand,
strongly suggesting that the threshold level of response to CTH is higher for the
tail melanophores than for those on the dorsal surface. That the tail darkening
reaction is independent of hypophyseal stimulation was shown by Bagnara (1957)
who pointed out that it occurs equally well in normal larvae, in hypophysioprivic
larvae and in excised tails.

It is interesting to note that the response of the tail melanophores to light
appears fairly early during the differentiation of these cells. Possibly it may even
be present in melanoblasts before the synthesis of melanin. The reaction does not
appear to be very strong in young melanophores; however, due to the relatively

8 _ JOSEPH T. BAGNARA

small amount of melanin in these young cells, it could be fully developed and
not appear so. It is significant that the melanophores of the regenerated tail can
carry out the tail darkening reaction. This seems to indicate that the tail has
regenerated completely, not only morphologically but physiologically as well.

SUMMARY

1. The peculiar light-sensitive melanophores in the ventral fin of Xenopus larvae
first appear about stage 47. Gradually, pigmentation increases in the distal half
of the fin, but the proximal portion remains free of melanophores throughout the
larval period. New melanophores first become visible at the fin edge, with an
apparent gradient of melanophore differentiation from somite to edge. The young
melanophores are thin and elongated, but as differentiation proceeds they add new
projections and become stellate. Even the youngest melanophores exhibit some
degree of tail darkening in the absence of light, but the strongest response is dis-
played by the fully formed melanophores.

2. Four days after tail extirpation, melanophores are seen in the regenerating
ventral fin. Pigmentation returns to all of the ventral fin except for a margin
along the fin edge, with a “somite-to-edge” gradient again apparent. The new
melanophores are of the typical tail fin type and are fully light-sensitive.

3. Melanophores on the dorsal surface differ markedly from those on the tail.
The former are heavily pigmented and are apparently more sensitive to the chro-
matotrophic hormone than the latter. Dorsal melanophores contract upon exposure
to darkness and expand in the light. They are less reactive to changes in illumina-
tion than the tail fin melanophores.

LITERATURE CITED

BaBak, E., 1910. Zur chromatischen Hautfunktion der Amphibien. Pfliiger’s Archiv, 131:
87-118. ;

BAGNARA, J. T., 1957. Hypophysectomy and the tail darkening reaction in Xenopus. Proc.
Soc. Exp. Biol. Med., 94: 572-575.

Bugs, E. J., 1905. The life history of Xenopus laevis Daud. Trans. Roy. Soc. Edinburgh, 41:
789-822.

Hoesen, L., AnD D. Stome, 1931. The pigmentary effector system VI. The dual character
of endocrine coordination in amphibian colour change. Proc. Roy. Soc. London, Ser.
B, 100: 10-53.

LAuRENS, H., 1917. The reactions of melanophores of Amblystoma tigrinum larvae to light
and darkness. J. Exp. Zool., 23: 195-205.

Nrieuwkoop, P. D., AND J. Faper, 1956. Normal Table of Xenopus laevis (Daudin). North
Holland Publishing Co., Amsterdam.

STEVENS, L. C., 1954. The origin and development of chromatophores of Xenopus laevis and
other anurans. J. Exp. Zool., 125: 221-246.

Tuinc, E., 1952. Melanophore reaction and adrenocorticotrophic hormone. III. The use
of tadpoles of Xenopus laevis as test animals for melanophore hormone. Acta Endo-
crin., 11: 363-375.

Twitty, V. C., anp M. C. Niu, 1954. The motivation of cell migration studied by isolation
of embryonic pigment cells singly and in small groups in vitro. J. Exp. Zool., 125:
541-574.

VAN DER LEK, B., J. DE Heer, A. C. J. Burcers ANnp G. J. von Oorpt, 1958. The direct
reaction of the tailfin melanophores of Xenopus tadpoles to light. Acta Physiol.
Pharmacol. Neerlandica, 7: 409-419.

Wipe, C. E., Jr., 1955. The urodele neuroepithelium II. The relationship between phenyl-
alanine metabolism and the differentiation of neural crest cells. J. Morph., 97: 313-344.

ieee A LIONSHIP OF REPRODUCTIVE TEMPERATURE AND
fee GEOGRAPHICAL RANGE OF THE MARINE
WOODBORER LIMNORIA TRIPUNCTATA?*

CAROLYN BECKMAN AND ROBERT MENZIES

Lamont Geological Observatory, Columbia University, Palisades, New York

The objective of the study was to determine the influence of temperature on
populations of the wood-destroying isopod Limnoria. The genus, which is eco-
nomically important, contains about 16 described species (Menzies, 1957; Menzies
and Becker, 1957; Pillay, 1957). Geographically, the species of the northern
hemisphere have been divided into boreal, L. lignorum; temperate, L. quadripunc-
tata; temperate-tropical, L. tripunctata; and tropical, 5—7 species.

The species we have worked with, Limnoria tripunctata, is found around the
world. It is a hardy species which survives and reproduces under laboratory con-
ditions. Our interest has been population growth at temperatures which might
be encountered by the species in nature. On the basis of our results, we believe
that we can explain the known distribution of the species in terms of the effect of
temperature on population growth.

The specimens studied were collected through the courtesy of Mr. Thomas P.
May, manager at the International Nickel Company testing laboratory at Wrights-
ville Beach, North Carolina, where the species is particularly abundant. The ani-
mals were air-shipped to the laboratory in containers which were packed so that
oxygen was not excluded.

PROCEDURE

The animals received at the laboratory were removed from the infected wood,
and sorted for size and sex. For each culture, 25 males and 25 non-gravid females
were placed in a dish containing sea water and a piece of presoaked pine. In all
three experiments a total of 1050 animals was used. Non-gravid females were
selected in order to keep each culture equivalent to the next. The use of females
with eggs in different stages of development would have permitted errors in defining
the reproductive temperature range and population growth at the various tem-
peratures. In order to permit the animals to establish burrows in the wood, each
culture was allowed to stand one week at room temperature (20—-24° C.). Salinity
was held within tolerable limits by addition of distilled water. Tolerable limits,
30-39%o , were determined by preliminary experimentation. Salinity data for the
three experiments are presented in Table I.

Seven cultures were prepared for each experiment in the manner described
and one of each was put into a controlled temperature box and kept for a period
of 66 days. This time duration was selected in order to allow for the production
of young, but not their maturation. The cultures were kept at approximately

1 Contrib. No. 395.

10 CAROLYN BECKMAN AND ROBERT MENZIES

0, 5, 10, 15, 20, 25 and 30° C. Usual temperature variation in the cultures was
+1°C. Short-term larger variations did occur as a result of equipment and power
failure. These occasional failures resulted in a gradual return to room tempera-
tures. For over 99% of the time, temperature variation was within £1° C.
The temperature means from the twice daily record are shown in Table I.

IDABEE AL

Salinity of culture water at the start and finish of each experiment

Experiment I Experiment II Experiment III
Original salinity June, July, Aug. Nov., Dec., Jan. May, June, July
35.7% 34.1% 35.0%o
Temperature End salinities
a 37.0 %e 34.2 %o 34.5 %o
5 36.8 34.1 S57)
10 SZ 34.0 33.6
15 37.9 33.6 3 5ul
20 38.1 33.8 33.8
INS 37.6 33.8 SouZ
30 38.4 325 34.6

Mean temperatures during each 66-day period

Experiment I OSE: Sal Ge DOVE AMS aICR | 20 7ExE ise: 29°31
June, July, Aug.

Experiment I]
Nov., Dec., Jan. 0° C. aes a HOS: 15.17C.2|-20°:€.. | 25, 1" Oke

Experiment III
May, June, July LIC: 62.G Ian. ator es 91°C. |. 24° EX 29 sseee

The measurements taken of the young at the end of the 66-day period are
recorded in size class, based on pleotelson width. The size classes used were taken
from Johnson and Menzies (1956).

RESULTS

The results of three experiments, each of 66 days duration, are presented in
tabular form (Tables II and III). These are the number of young present at
the end of each experiment, the largest size attained by the young, adult mortality
and gravidity of surviving females. From these data the population change in
66 days has been determined (Fig. 1).

Cultures kept at 0° C. showed no reproduction and the highest adult mortality
(86-100% ). At O° C. the animals were immobile, did not feed and starvation is
the probable cause of death.

The 5° C. cultures also showed no reproduction and a high adult mortality
(50-92%). The animals were feebly motile and showed slight feeding, but here
again starvation is the probable cause of death.

The 10° C. cultures showed no reproduction and a lower mortality (0-64%).

LIMNORIA TEMPERATURE EFFECTS 11

In two experiments gravid females were present; however, only eggs were found,
and no larva or young were produced in the 66-day period. The animals were
sluggish but did feed.

The 15° C. cultures showed a reproductive rate between 0 and 6 young pro-
duced per female. We are confident that a female produces more than one brood

TABLE II

Degrees
Centigrade......... 0° Se 10° 154 20° 252 30°

Adult mortality—Number of individuals in each culture of
50 animals that died in 66 days

Experiment I
June, July, Aug. 46 29 6 8 8 if 7

Experiment II
Nov., Dec., Jan. 50 46 32 20 al 26 Sy

Experiment III
May, June, July 43 28 0 + 2 0 32

Number of young present at the end of 66 days

Experiment I
June, July, Aug. 0 0 0 Loy 302 457 254

Experiment II
Nov., Dec., Jan. 0 0 0 0 43 64 9

Experiment III
May, June, July 0 0 0 20 45 73 0

Experiment I
June, July, Aug. a as == 2 3 3 3

Experiment II
Nov., Dec., Jan. nee aaa = a y 2 2

Experiment III
May, June, July a ss = y, 3 3 —

* Pleotelson Width Size Classes after Johnson and Menzies (1956).

Class Width in mm.
2 .28—.33
3 .35-.40

during her lifetime and therefore a 15° C. temperature permits population growth
to occur. Adult mortality was 640%.

The 20° C. cultures showed a reproductive rate between 2 and 12 young per
female. The average rate was 5 young per female and growing population is
evident. Adult mortality varied between 4 and 54%.

i

Degrees
(OA

CAROLYN BECKMAN AND ROBERT MENZIES

Experiment I

June, July, Aug.

TABLE III
Gravidity at the end of the experiments

Experiment II

Nov., Dec., Jan.

Experiment III

May, June, July

No. No. No. No. No.
P P
females | Temiles | gevid | Cn ee ees
3 0 0 0 0 0 4
9 3 33 1 0 0 12
21 9 43 14 0 0 D5
16 9 56 16 0 0 22
PAA lf 3 16 0 0 23
22 7 32 18 1 6 2S
20 2 10 7 0 0 0
500
400
\
300
200
100
f-
os \
hee EN
a x
O “as
en :
@. Lee =
eae OO
65" 10 5 eo 25 0
TEMPERATURE °C.

No.
females
gravid

WOO RP Oo OS ©

POPULATION CHANGE IN 66 DAYS

FicureE l.

Population change at various temperatures after 66 days.

Per cent
gravid

LIMNORIA TEMPERATURE EFFECTS 13

The 25° C. culture showed a reproductive rate between 3 and 18 young per
female. The average rate was eight young per female, or about 1.6 times that of
the 20° culture. The adult mortality rate was 0-52%.

The 30° C. culture showed a reproductive rate between 0.3 and 10 young
per female. The average rate was three young per female. The adult mortality
varied between 14 and 64%.

The most obvious feature of the three experiments is the lack of population
growth below 10° C. The animals do not feed at these temperatures and eventu-
ally die of starvation as their fat reserve is depleted. The greatest population
increase occurred in the cultures kept at 25° C. This might be considered the
optimal temperature for population growth.

Adult mortality was highest at temperatures below 10° C. and at 30° C.
Mawatari (1950) records 2-6° C. as the temperature lethal to adults of Limnoria
lignorum (actually L. tripunctata). Our results also agree with the findings of
Kampf (1957), who reports maximum population growth for Limnoria tripunctata
from the Mediterranean to occur at 24° C. He also reports a shortening of the
life span at higher temperatures which may account for the higher mortalities

apoo -C.
INHERENT SEASONAL RHYTHM

Kampf (1957) reported a cyclic occurrence of gravidity for Limnoria kept under
reasonably constant aquarium conditions. Unfortunately, his populations were
mixed cultures of two species, L. carinata and L. tripunctata. Nevertheless, the
low proportion of the former in his culture probably means that the cyclical record
of gravidity is real for L. tripunctata. Coker (1923) reports that there is little
winter gravidity found in Beaufort, North Carolina, Limnoria during the winter.
Our observations of samples shipped from Wrightsville Beach, North Carolina,
confirm this. The fact that culturing the specimens in the laboratory at tempera-
tures approached only during the summer in the field still results in low gravidity
and low young production suggests that we are dealing with an endogenous
seasonal gravidity period which does not appear to be subject to immediate modi-
fication by change of temperature, and a reproductive rate imposed on the former
which is subject to modification by change of temperature. It will be of the
greatest interest to determine whether a seasonal endogenous reproductive period
occurs in populations of the same species collected from a tropical locality where
seasonal temperature fluctuations are almost absent.

DISTRIBUTION, TEMPERATURES AND POPULATION GROWTH

From the data presented, it would be a likely prediction that Limnoria tri-
punctata should not occur in localities where the sea water temperature is below
10° C. for most of the year. On the Atlantic coast the species is known to occur
as far north as Cape Cod, Massachusetts. Here the yearly average sea water
temperature is about 11° C. For five months of the year, however, the tempera-
ture rises above 15° C. and thus there is time for the populations to reproduce.
The existence of the species is probably very precarious at this locality since
monthly averages are below ten degrees for five months of the year. While signs

14 CAROLYN BECKMAN AND ROBERT MENZIES

of Limnoria damage are extensive in the area, more often than not the burrows
are empty and finding infested wood is sometimes a difficult job. North of Cape
Cod, at Eastport, Maine, for example, the sea water temperature reaches 10° C.
for only three months of the year. In such a situation we would not expect
Limnoria tripunctata and to our knowledge the only species found there is Limnoria
lignorum.

On the Pacific Coast of North America the species is not known as far north
as Friday Harbor, Washington. Here the mean temperature of the surface sea
water rises to 10-11° C. for only three months out of the year. During the re-

ome
ATURE

OPTIMAL REPROOUCTIVE TEMPERATURE

i

nn
on

> DEGREES CENTIGRADE TEMP

8 53 &
1

oN bk 7@ Zs

EASTPORT wooOOS HOLE CHARLESTON MIAMI GUANTANIMO CRISTOBAL
me. MASS. ss FLA. CUBA cz.

ATLANTIC COAST

OPTIMAL REPRODUCTIVE TEMPERATURE

)

nn 2 @ ©@
—> DEGREES CENTIGRADE TEMPERATURE

sa) iia Ale Sel tll Be aif eh |

FRIDAY HARBOR CRESCENT CITY SAN FRANCISCO SAN DIEGO BALBOA
WASH. CALIF CALIF CALIF, Cz.

°
I

PACIFIC COAST

LATITUDE DEGREES NORTH

50 4 40 3 30 25 20 15 ike) 5 te)

Ficure 2. Geographical distribution, mean monthly surface sea water temperatures plotted
at approximate longitude and reproductive temperature limits (the horizontal lines connected
by the sloping block): The shaded areas represent the period of time that the temperature is
within the reproductive range of the species. The minimum reproductive temperature, 14° C.,
is estimated from the average of the three experiments to be at this temperature. At this point
one young would be produced for each adult.

maining months it is lower than 10° C. This would give the species three months
to feed, and no period of time suitable for reproduction. Southward, the species
is also absent at Crescent City, California. Here temperatures rise above 10° C.
for ten months of the year. The maximum temperature is 13° C. and this occurs
for only two months of the year; 15° C. is not reached seasonally.

The species occurs along the Pacific Coast from San Francisco, California, to
Mazatlan, Mexico, and probably as far south as Panama. At San Quentin, San
Francisco Bay, lethal temperatures do not occur seasonally and temperatures favor-

LIMNORIA TEMPERATURE EFFECTS 15

ing population growth occur for seven consecutive months of the year. Seasonal
temperature distribution, reproductive temperature limits for the species, and the
established geographic distribution of Limnoria tripunctata in North America are
shown in Figure 2. The remarkable coincidence between the geographic range
of the species and the seasonal occurrence of temperatures favorable for reproduc-
tion leads us to conclude that temperature is a major factor controlling the distri-
bution of the species.

The reason for the restriction of a species to a given geographic range has
been of interest to zoologists and zoogeographers for many years. Temperature
means, maxima and minima have been examined as parameters of possible control.
The duration of a given range of temperatures has also been considered (Menzies
and Hedgepeth, unpublished data). Experimentally determined physiological lethal
limits for short periods of time are usually too wide to account precisely for the
distribution of a marine species. Our examination of the effect of temperature
on population growth of a single marine species suggests that the geographic
distribution of this species is closely tied to the prevalence of temperatures favor-
ing reproduction. The minimum duration of a favorable temperature necessary
would depend on the growth rate of the species. Doubtless the duration of un-
favorable temperatures for survival is also important. In terms of our experi-
mental data it is difficult to explain the survival of populations of Limnoria tri-
punctata at Woods Hole, where it is necessary for the species to maintain itself
for three months of the year at mean temperatures below 5° C. Our data suggest
that this is a remote possibility ; nevertheless, the species is precariously established.

The figures were drawn by Carolyn Peppin and Don Robinson. This re-
search was sponsored by the U. S. Navy, Office of Naval Research, Contract
Nonr-266 (41).

SUMMARY

1. Experimental results show that Limnoria tripunctata will feed at tempera-
tures ranging from approximately 10° C. to 30° C. The reproductive temperature
range appears to be from about 15° C. to 30° C. and the greatest population in-
crease is in the neighborhood of 25° C. Excessive mortality results at 30° C.

2. Gravidity and hence number of young produced depends upon season and
does not appear to be immediately modified by the presence of favorable repro-
ductive temperatures.

3. The experimental results agree well with the known geographic range of
the species.

LITERATURE. CITED

Coxer, R. E., 1923. Breeding habits of Limnoria at Beaufort, N. C. J. Elisha Mitchell Sct.
Soc., 39: 95-100.

Kamer, W. D., 1957. Uber die Wirkung von Umweltfaktoren auf die Holzbohrassel Limnoria
tripunctata Menzies (Isopoda). Zeitschr. f. Angewandte Zool., 44: Drittes Heft,
359-375.

Jounson, M. W., AND R. J. Menzies, 1956. The migratory habits of the marine gribble
Limnoria tripunctata Menzies in San Diego Harbor, California. Biol. Bull., 110:

54-68.

16 CAROLYN BECKMAN AND ROBERT MENZIES

Mawatrarl, S., 1950. Biological and industrial study of marine borer problem in Japan.
Studies on the aquatic animals of Japan (1): 9-12, 45-124. In Japanese with English
summary.

Menzies, R. J., 1957. The marine borer family Limnoriidae (Crustacea, Isopoda). Bull.
Mar. Sci. Gulf Caribb., 7: 101-200.

Menzigs, R. J., AND G. Becker, 1957. Holzzerstorende Limnoria Arten (Crustacea, Isopoda)
aus dem Mittelmeer mit Neubeschreibung von L. carinata. Zeitschr. f. Angewandte
Zool., 44: Erstes Heft, 85-92.

Pitiay, N. K., 1957. A new species of Limnoria from Kerala. Umiv. Kerala Trivandrum, V,
No. II, series C, 149-157.

fey AL SEAGES OF PLEURONCODES PLANIPES STIMPSON
(ehUSiACEAY DECAPODA;,GALATHEIDAE)

CARE My BOYD

Scripps Institution of Oceanography,1 University of California, La Jolla, California

The zoeal larval stages of a galatheid crab were found in large numbers (up to
42,000/1000 m.*) in plankton tows taken at several stations of the Marine Life
Research program of the California Cooperative Oceanic Fisheries Investigations
near Magdalena Bay, Baja California, Mexico. Since these unknown larvae were
found in an area where the galatheid Pleuroncodes planipes was extremely
abundant, it seemed likely that they were the zoeal stages of that species.

METHODS

The larvae were tentatively identified as P. planipes. This identification was
later confirmed by two methods: (1) the first larval stages were hatched from
the eggs of the adult P. planipes, (2) one specimen of the fifth larval stage was
caught in the process of molting to the juvenile stage, and this juvenile could be
identified as P. planipes. Several attempts were made to follow the succession
of larval molts in the laboratory. Hatching the larvae from the egg was relatively
easy, and was achieved by several methods. However, no larvae survived longer
than eight days, and all died before they had molted once. Plunger jars and
still-water aquaria were used with variations of temperature, aeration and food
(Platymonas, Chlamydomonas and Coccolithophores). The plunger jar method
and a procedure described briefly by Ray (1958; p. 501), involving rotating flasks,
apparently failed because of the constant vigorous tumbling of the larvae.

Adult females carrying eggs have been found in the plankton from December
through March. Gravid females kept in the laboratory hatched larvae during most
of that period and the same is probably true of females in nature. As larvae of all
stages are found in the plankton from January to July, it has proved impossible to
determine rates of development.

DESCRIPTION OF THE LARVAE
Stage I (Figure 35)

First antenna (Fig. 1). This appendage has a long, unsegmented base with
two distal branches. These, the exopodite and endopodite, are continuous with
the base. The former bears six setae, the latter, a long plumose spine.

Second antenna (Fig. 6). The antenna consists of a base, continuing into
a long medial prickly spine, and a scale. The scale bears about eight plumose
setae and is continued as an apical spine, prickly at the terminal end. On the

1 Contribution from the Scripps Institution of Oceanography, New Series.

bi

18

CARL M. BOYD

Ficure 1.
FIGURE 2.
FIGURE 3.
Ficure 4.
Ficure 5.
FIGURE 6.
FIGURE 7.
Ficure 8.

I2 IS

First antenna, Stage I.
First antenna, Stage II.
First antenna, Stage III.
First antenna, Stage IV.
First antenna, Stage V.
Second antenna, Stage I.
Second antenna, Stage II.
Second antenna, Stage III.

I4 IS

Ficure 9. Second antenna, Stage IV.

FIGURE
FIGURE
FIGURE
FIGURE
FIGURE
FIGURE

Second antenna, Stage V.
First maxilla, Stage I.
First maxilla, Stage II.
First maxilla, Stage ITI.
First maxilla, Stage IV.
First maxilla, Stage V.

YY
YY WAP ANE

ee

Ficure 16.
Ficure 17.
Ficure 18.

Ficure 19.
Ficure 20.

LARVAE OF PLEURONCODES PLANIPES

\
/

Mi, py 2 Loy
LEP .

Ky

Second maxilla, Stage I.
Second maxilla, Stage II.

Second maxilla, Stage III.

Second maxilla, Stage IV.
Second maxilla, Stage V.

O
a‘ \

\
if
PIV IPLL

Ficure 21.
FicurReE 22.
FicureE 23.
Ficure 24.

24

First maxilliped, Stage I.

First maxilliped, Stage II.

First maxilliped, Stage III.

First maxilliped, Stage IV and V.

19

20 CARL M. BOYD

dorsal side of the inner flagellum is a smooth spine which, though small in this
stage, increases in size in later stages.

Mandible. The mandible is a simple toothed process lacking a palp.

First maxilla (Fig. 11). The endopodite is a simple unjointed structure
bearing three spinose setae and two smooth setae. The endopodite does not change
in number of setae but only in size in the next four stages. The exopodite is
split into a basipodite and a coxopodite. These generally add setae in the later
stages. In Stage I, the basipodite bears five heavy setae. The coxopodite has
SEVeM) Sctae:

Second maxilla (Fig. 16). As in the endopodite of the first maxilla, the
endopodite of the second maxilla does not change the number of setae in any of
the stages. Terminally on the endopodite there are four setae; three spinose and
one naked. <A group of three spinose setae is located subterminally. The basipodite
lobes each bear four spinose setae, as does the distal lobe of the coxopodite. The
proximal lobe of the latter carries six spinose setae. The scaphognathite carries
only four plumose setae along its margin. Its posterior end is produced into a
long plumose projection. 3

First maxilliped (Fig. 21). The maxilliped is well developed in the first larval
stage. The appendage consists of a coxopodite with one seta on its interior
margin, an unsegmented basipodite, an endopodite with five segments, and an
exopodite with two indistinct segments. The basipodite has ten spinose setae on
its interior margin. There are four setae terminally on the exopodite. The five
segments of the endopodite each bear spinose setae, but in differing numbers. The
fourth, second, and first segments have two setae, and the third but one seta. There
is also one fine, smooth seta coming off from the joint of the distal (fifth) segment.

Second maxilliped (Fig. 25). The second maxilliped has much the same
structure as the first. The endopodite of the second maxilliped, however, has
only four segments; the third is elongated. The coxopodite has no setae; the
basipodite has three spinose setae. The terminal (fourth) segment of the endop-
odite has four spinose setae and one fine, smooth seta. There are four setae
terminally on the exopodite.

Third maxilliped (Fig. 29). The third maxilliped is rudimentary in Stage I;
it lacks setae and is poorly developed in musculature. The appendage consists
only of a basipodite and an exopodite; the endopodite is not present.

Abdominal segments. There are five segments, plus the telson in Stages I
and II.

Telson (Fig. 36). The telson of the first larval stage has seven projections
on each side. The first (outermost) is a spinose extension of the telson, and is
unjointed at its base. This spine has 6-10 thorns distally. The second projec-
tion is a fine, hair-like, plumose seta. The next five are similar in that they are
all plumose setae. In addition to the plumes, each seta carries thorns on its inner
and outer margins.

Along the posterior margin of the telson between the plumose setae, several
small thorns occur. In Stage I larvae there are usually one to three thorns
between each of the inner five setae. Generally the number of thorns increases be-
tween the setae more medially located, so that there are usually two to three thorns
between setae #6 and #7. In the medial cleft of the telson, one to three thorns

a3

Ficure 25.
Ficure 26.
FicuReE 27.
Ficure 28.
Ficure 29.

LARVAE OF PLEURONCODES PLANIPES Dil

Second maxilliped, Stage I.
Second maxilliped, Stage IT.

Second maxilliped, Stage III and IV.

Second maxilliped, Stage V.
Third maxilliped, Stage I.

Figure 30.
Ficure 31.
Ficure 32.
Ficure 33.
FicurE 34.

Third maxilliped, Stage II.
Third maxilliped, Stage III.
Third maxilliped, Stage IV.
Third maxilliped, Stage V.
Telson, Stage IV.

22 CARL M. BOYD

may occur on each side. A tuft of short fine hairs occurs further medially. Each
tuft is composed of seven to ten hairs. The numbers of these thorns and hairs
are variable and are seldom identical between the right and left halves of the
telson of any individual.

Stage II (Figure 37)

First antenna (Fig. 2). This appendage is much the same as in Stage I, but
here three small hairs appear toward the distal end of the base. These hairs are
on the location of an ultimate segmentation appearing fully in Stage III.

Second antenna (Fig. 7). The antennal scale has gained two more plumose
setae, giving it usually ten. A new spine has appeared on the ventral side of
the base. This spine is not segmented from the base, but is common with it.

Mandible. No change from Stage II.

First maxilla (Fig. 12). A seta has been added to the basipodite. The coxop-
odite has remained unchanged.

Second maxilla (Fig. 17). The endopodite, basipodite and coxopodite are
unchanged. The scaphognathite has added two marginal setae, giving a total
of six.

First maxilliped (Fig. 22). There are now seven terminal setae on the
exopodite.

Second maxilliped (Fig. 26). There are now seven terminal setae on the
exopodite.

Third maxilliped (Fig. 30). This appears to have become a functional organ;
its musculature has developed and there are six setae on the terminal end of the
exopodite.

Telson (Fig. 38). The primary change in the telson of Stage II larvae is
the addition of one plumose seta, increasing the number of projections to eight.
In this stage, however, the third projection does not have thorns on its margins,
but only plumes.

The arrangement of thorns occurring between setae is similar to that of Stage I.
In Stage II, however, there are neither hairs nor thorns in the medial cleft.

Stage III (Figure 39)

First antenna (Fig. 3). The segmentation initiated in Stage II has been com-
pleted, setting the exopodite off from the base. ‘Three hairs originate at the joint.
A hair appears on the exterior lateral margin of the base. The combined length
of the new terminal segment and the setae is longer than the spine of the endopodite.

Second antenna (Fig. 8). The antennal scale now usually has 11 setae. The
spine on the inner flagellum has increased in length and now has a terminal hair.

Mandible. No change.

First maxilla (Fig. 13). The basipodite has eight setae, an increase of two
setae over Stage II]. The coxopodite is unchanged.

Second maxilla (Fig. 18). The distal lobe of the basipodite now has seven
setae. The proximal lobe has six. The coxopodite is unchanged. The exterior
margin of the scaphognathite now has nine setae, plus one seta located on the
interior margin of the posterior end.

First maxilliped (Fig. 23). There are now eight setae on the exopodite.

EARVAE OF PREWURONCODES PLANIPES

Ficure 35. Larva, Stage I. FicureE 37. Larva, Stage II.
Ficure 36. Telson, Stage I. Figure 38. Telson, Stage II.

28)

24 CARL MM. "BOYD

Second maxilliped (Fig. 27). The exopodite has added one terminal seta,

giving a total of eight.

Third maxilliiped (Fig. 31). The only change in the third maxilliped is the
development of a small bud on the basipodite.

Abdominal segments. There are six segments, plus the telson in Stages III,
IV and V.

Telson (Fig. 40). The changes in the telson between Stages IJ and III are
radical. There are typically now nine projections on each half of the telson.
The first (outermost) is a short naked spinose extension of the body of the telson.
The second is a plumose hair-like seta. The third is a short heavy plumose seta.
The fourth is the major projection of the telson; it is an extension of the telson
and is not jointed at its base. It is covered on its interior and exterior margins
by many small thorns. The fifth through the eighth projections are plumose setae
bearing thorns on their lateral margins. The ninth projection is a small plumose
seta lacking thorns. This innermost seta is not always present. It may be lack-
ing on one or both sides of the telson.

Thorns are found abundantly between the fourth through the eighth projec-
tions. Between projections #4 and #5, zero to three thorns may occur. Usually
three to five thorns are found between each of the fifth through eighth projections.
Between the eighth projection and the medial line up to four hairs may be present,
in addition to the ninth projection. The occurrence of these hairs is erratic.

On the dorsal surface of the telson, two pairs of very fine setae are found. The
posterior pair is near the base of the seventh projection. The second pair is in
line with the first pair, but anterior to it.

Uropods. ‘The inner uropods are present only as poorly developed plates;
they carry no setae. The outer uropods have from 8 to 11 plumose setae. The
setae do not carry thorns. Their number is variable and may differ between the
right and left sides. The tip of the outer uropod is not produced in an extended
spine, but is short and naked. Small accessory setae commonly occur on the
ventral surface of the outer uropod. These are not at the margin but are based
somewhat anteriorly of the posterior margin. Generally there are two on each
uropod, one between setae #1 and #2 (counting from the distal end) and the
second between setae #4 and #5.

Stage IV (Figure 41)

First antenna (Fig. 4). The exopodite has gained three pairs of setae along
its inner margin. In addition, the endopodite has developed a joint at its base.

Second antenna (Fig. 9). The scale usually now has 17 plumose setae. The
spine on the inner flagellum has now increased in length so that it is much longer
than the flagellum.

Mandible. The mandible at this stage has a small bud on the anterior margin
which will become a palp in Stage V.

First maxilla (Fig. 14). The basipodite has 11 setae. The coxopodite has
10-11 spinose setae.

Second maxilla (Fig. 19). The lobes of the basipodite and coxopodite have
added setae, but their absolute numbers are variable. The exterior margin of the
scaphognathite now has about 25 setae; the posterior interior margin has six setae.

LARVAE OF PLEURONCODES PLANIPES

Ficure 39. Larva, Stage III.
Ficure 40. Telson, Stage III.
Ficure 41. Larva, Stage IV.

26 CARL M. BOYD

First maxilliped (Fig. 24). The only change from Stage II and III is in the
addition of a plumose seta on the exterior margin of the third segment of the
endopodite.

Second maxilliped (Fig. 27). No change from Stage III.

Third maxilliped (Fig. 32). The exopodite has added two setae, giving now
a total of eight. The bud on the basipodite has somewhat increased in size from
Stage III, but is still rudimentary.

Thoracic legs. Thoracic legs are not visible on Stage IV larvae from a lateral

view. They are present, however, as poorly developed appendages under the

carapace.

Telson (Fig. 34). The telson in this stage is similar in aspect to that in
Stage III, but it has added several plumose setae. The total number of projec-
tions may vary from 10 to 12, with the deletions occurring in the innermost setae ;
i.e., setae #11 and #12 may individually or both be absent. Typically there are
11 projections. Setae #9, #10, #11, and #12 are plumose but do not bear thorns.
Projection #1 appears to be jointed and is no longer common with the body of
the telson.

On the dorsal surface of the telson there are now five pairs of fine, hair-like
setae. Three pairs have been added anteriorly to those present in Stage III.

Uropods. Inner uropod: this now has from 10 to 17 plumose setae. Three
tufts of hairs occur medially from the innermost seta. A fine, hair-like accessory
seta is found on the dorsal surface of the distal end of the inner uropod. Outer
uropod: the extremity of this branch is now extended as a long spine bearing
many thorns along its lateral margins. From 13 to 21 plumose setae are present
on the posterior margin of this branch. Several fine hairs are found medially of
the innermost seta. Four accessory setae are again found on the ventral surface
of the outer uropod. These commonly occur between setae #1 and #2, #4 and
#5, #9 and #10, and #14 and #15, starting from the outer seta.

Stage V (Figures 42-44)

First antenna (Fig. 5). The exopodite has added more medial setae, bringing
the total to about 22. The endopodite has elongated, but is still shorter than the
exopodite. The base now appears to be very lightly segmented into three parts.

Second antenna (Fig. 10). The number of plumose setae on the scale has
increased to usually 19. The inner flagellum is reduced to a very small size and
the spine which formerly grew upon it is as long as the antennal scale.

Mandible. The palp on the anterior margin is well developed, though un-
segmented.

First maxilla (Fig. 15). The basipodite has about 15 setae and the coxopodite
has about 13 setae.

Second maxilla (Fig. 20). The basipodite and coxopodite have added setae.
The scaphognathite now has about 43 plumose setae on its exterior margin. Its
posterior interior margin now has eight setae. The extreme posterior spine has
been replaced by three plumose setae.

First maxilliped (Fig. 24). Unchanged from Stage IV.

Second maxilliped (Fig. 28). The only change from Stage III and IV is in
the addition of a smooth seta on the exterior margin of the third segment of the
endopodite.

Se _

LARVAE OF PLEURONCODES PLANIPES yA)

Third maxilliped (Fig. 33). At this stage the endopodite appears. However,
it is soft, poorly musculatured and unsegmented. On its interior margin is a
plumose seta and on its exterior margin, distally, is a fine smooth seta.

Thoracic legs. The thoracic legs are easily visible in a lateral view of the
Stage V larvae. The first is chelate, the fifth is quite small and hidden behind the
fourth leg.

Pleopods. There are now four pairs of unequally bifid pleopods present.

Telson (Fig. 43). The telson of this stage has reached its full development
and variability of spine count is reduced. There are now commonly 12 pro-
jections. Of the inner eight projections, only #5 through #8 have thorns as
well as plumes on their margins. One to five thorns are found on the margins
of the telson between each of projections #4 through #9. No thorns are found
medially of the ninth projection, but up to three hairs may occur between each of
the setae between the ninth projection and the medial line. The five pairs of
fine dorsal accessory setae of Stage IV are unchanged in Stage V.

Uropods. Inner branch: the inner branch now has from 18 to 21 plumose
setae. Several hairs are found medially of the innermost setae. The fine setae
occurring on the dorsal surface of the distal tip of the inner uropod in Stage IV
is still found in this stage. Outer branch: the characteristic form of the outer
uropod has not changed from that of Stage IV. The tip of the uropod is still
drawn out as a spine and still bears thorns on its lateral margins. In this stage
there are from 20 to 23 plumose setae on the posterior margins of the outer uropods.
Fine, accessory setae occur ventrally between #1 and #2, #4 and #5, #9 and
#10, and #14 and #15.

THE LARVAL STAGES
The five larval stages may be separated by the following key:
A. Uropods absent

1. The posterior margin of the telson has five plumose setae on

each side; the third maxilliped is rudimentary. SG
Beetme telson has six plumose setae on each side; the third
maxilliped is functional. SAGs
B. Uropods present
1. The inner uropods are naked ; pleopods are absent. SAG LT
2. The inner uropods have setae; pleopods are absent. Sue. Ly
3. Pleopods are present. lbs Gy BE

The sizes (total lengths) of the larvae in each stage were as follows (range and
mean) :

Number
Stage Minimum Mean Maximum measured
I 2.6 mm. 2.7 mm. 2.8 mm. 8
II | 3.4 3:6 8
III 4.1 4.6 5.0 7
IV a ep | 6.5 11
V j pe 7.6 7.8 11

All the larvae measured and figured were from the Magdalena Bay area.

28 CARL M. BOYD

Ficure 42. Larva, Stage V.
Ficure 43. Telson, Stage V.
Ficure 44. Larva, Stage V.

LARVAE OF PLEURONCODES PLANIPES 29

Lebour (1930) notes that Galathea dispersa may omit either Stage IV or V,
but that there are usually five stages. It is not known if all P. planipes larvae
pass through each of the five stages before molting to the juvenile stage; one or
more stages may be by-passed. However, the measurements given above indicate
that this is infrequent.

DISCUSSION

All except one of the developmental features of Munida and Galathea larvae
as described by Lebour (1930; p. 177) are also found in P. planipes: the pleopods
do not appear as small knobs in the third stage in P. planipes, but appear without
previous indication in the fifth stage. Plewroncodes larvae share the following
traits with larvae of Munida, but not with those of Galathea: the endopod of the
first maxilla has only one segment; in Stages I and II the lateral-most spine of
the telson of both Munida and Pleuroncodes is longer than any of the posterior
margin on the telson; a large dorsal spine occurs on the posterior margin of the
sixth abdominal segment in Stages III, IV and V. Pleuroncodes thus appears
to be more closely related to Munida than to Galathea.

The larval stages of five species of the genus Munida have been described.
G. O. Sars described the larvae of M. sarsi Brinkman in 1889. Johan Huus (1934)
found that these were morphologically identical with the larvae of M. tenuimana
M. Sars, with the exception of some minute hairs in the medial cleft of the telson.
Lebour (1930) described the larval stages of M. rugosa (Fabricius) and later
(1931) set up a key to distinguish the galatheid larvae of the Plymouth area.
M. rugosa is, however, the only species of the genus Munida covered by this key.
(Lebour described the larvae as M. banffica; Zariquiey (1952), however, has
synonymized this species with M. rugosa.) Rayner (1935) described five larval
stages of M. gregaria (Fabricius) and M. subrugosa (White) but was unable to
distinguish the larvae of the two species. The adults of these two species are
limited to the southern hemisphere and along the Eastern Pacific they do not occur
north of 41° 30’ S. (Matthews, 1932; p. 470).

Larvae of Pleuroncodes planipes can be easily distinguished from the described
larvae of European munids by the aciculate second antennae and the elongated
postero-lateral spines of the carapace of the latter. They are, however, very much
like the larvae of the New World munids described by Rayner (1935). The chief
difference is the lack of serration on the lateral margins of the rostrum of the
larvae of M. gregaria and M. subrugosa. The rostrum of each of the stages of
P. planipes, except the first, is strongly serrated. The fifth stage of the Munida
species has some serration on its rostrum, but not to the degree found in P. planipes.
Rostral serration probably will not hold good in the future as a valid characteristic
setting off the genus Pleuroncodes, for Gurney (1942; p. 255) mentions an un-
identified galatheid larva from Bermuda which had a serrated rostrum. No species
of Pleuroncodes is known from that area.

Both Huus (1934) and Lebour (1930, 1931) use color and arrangement of
pigmented areas as an aid to specific identification. All color fades, however,
from specimens kept longer than a few weeks in either alcohol or formalin. No
color pattern was noted in P. planipes for that reason.

The area off Magdalena Bay (24° 40’ N., 112° W.) appears to be peculiar
as far as P. planipes larvae are concerned. Within a 50-mile radius of this bay,

30 CARL M. BOYD

densities up to 42,000/1000 m.* have been observed and are not uncommon.
Outside this area, but well within the usual distributional area of adult P. planipes,
the larvae, as sampled with a meter net from 140 m. to the surface, have an average
density of 5/1000 m.* (based on 22 tows taken between January and July; range
0-26/1000 m.?). On this evidence it would appear that the Magdalena Bay area
is a center of breeding. This is somewhat unusual in that adults are considered
epi-pelagic and are distributed over a very wide range, from about 16° N. to 36° N.
and from the littoral to several hundred miles off-shore.

The author wants to express his thanks to Dr. Martin W. Johnson for the
help he has given me in this study. The author is also grateful for the assistance
of Dr. E. W. Fager and Mrs. Margaret K. Riedel in criticism and preparation of
this paper.

SUMMARY

The larval stages of a galatheid decapod crustacean, Pleuroncodes planipes,
have been identified, described and figured. There are five stages. First stage
larvae were hatched from eggs carried by adults; no larvae molted under labora-
tory conditions. Larval morphology indicates that the genus Pleuroncodes is more
closely allied to the genus Mumida than to Galathea. The larvae occur in high
densities (42,000/1000 m.*) in the region of Magdalena Bay, Baja California,
Mexico.

LITERADURE Chirp

GURNEY, Ropert, 1942. Larvae of Decapod Crustacea. 306 pp. The Ray Society, London.

Huus, JoHan, 1934. Zur morphologisch-systematischen und biologischen Kenntnis der
nordischen Munida-Larven (Crustacea Decapoda). Bergens Museums Arsbok, Natur-
videnskapelig rekke, 1934: 1-32.
Lesour, Marie V., 1930. The larvae of the Plymouth Galatheidae. I. Munda banffica,
Galathea strigosa and Galathea dispersa. J. Mar. Biol. Assoc., 17: 175-188.
Lesour, Marie V., 1931. The larvae of the Plymouth Galatheidae. I]. Galathea squamifera
and Galathea intermedia. J. Mar. Biol. Assoc., 17: 385-390.

MatTtTHews, L. Harrison, 1932. Lobster-krill; anomuran crustacea that are the food of whales.
Discovery Reports, 5: 467-484.

Ray, D. L., 1958. Some marine invertebrates useful for genetic research. In: Perspectives
in Marine Biology, edited by A. A. Buzzati-Traverso, pp. 497-512.

Rayner, G. W., 1935. The Falkland species of the crustacean genus Munida. Discovery
Reports, 10: 209-245.

Sars, G. O., 1889. Bidrag til Kundskaben om Decapodernes Forvandlinger I]. Lithodes—
Eupagurus, Spiropagurus—Galathodes—Galathea—M unida—P orcellana—(Nephrops).
Archiv. for Mathematik og Naturvidenskab, 13, 1889-1890: 132-201.

ZARIQUIEY ALVAREZ, Ricarpo, 1952. Estudio de los especies Europea, del Gen. Munda Leach
1818. Eos; Revista Espanola de Entomologia, 28: 143-231.

fame UEPING CYCLE IN BALANUS BALANOIDES L:

Ds CRISPVAND Bs se PATEL
Marine Biology Station, Menai Bridge, Anglesey, N. Wales

Darwin, as early as 1854, noticed that a number of operculate cirripedes often
moulted their skin, which included the exuviae of prosoma, appendages, penis,
oesophagus, rectum and the inner lining of the mantle. He also stated that the
frequency of moulting was related to the speed at which the barnacles grew.

Recently Costlow and Bookhout (1953, 1956) studied the frequency of moulting
in relation to growth in recently settled cyprids of Balanus improvisus and Balanus
amphitrite, but, since their observations were limited to a certain length of time
after settlement, no accurate and detailed information is available regarding the
seasonal fluctuations in the moulting rhythm and the factors controlling it.

The following investigation was therefore carried out to study the moulting
and the factors influencing it in the boreo-arctic species Balanus balanoides L.

METHOD

Suitable groups of Balanus balanoides L. were obtained for the investigation on
the shells of living Mytilus edulis growing on the piles of Bangor Pier in the Menai
Straits, North Wales. In such a position, with a good current of water flowing
past, both mussels and barnacles obtain abundant food and grow well. As soon
as possible after collection all barnacles other than B. balanoides were removed.
Usually 10-20 mature individuals were left on each shell. Care was taken not to
perforate the shells of any specimens which were to be used in the experiment,
since even a slight puncture, which can readily be made at the basal attachment,
may lead to the death of the individual in a few days. During the first few days
individuals which died were removed from the group and were not considered in
the experiment.

Each group, comprising 20-30 individuals, was kept in 6-inch diameter glass
crystallising dishes and the water was changed every day. It was not convenient
to simulate the semi-diurnal tide ; however, it was found beneficial for the barnacles
to be out of water for about 8 hours each day. The dishes therefore were filled
at night and emptied each morning and the cast skins present were counted.
Sometimes during the breeding season pools of sperms were seen, probably indicat-
ing that copulation had occurred. Occasionally, during the period of liberation,
swarms of nauplii were noted. However, if liberation occurred soon after dark,
the larvae did not swarm to the side of the dish but were often devoured by the
parents by the following morning. Large numbers of chocolate-red faecal pellets,
containing much larval pigment and undigested chitinous appendages, were pro-
duced in consequence of this cannibalistic meal. These served to indicate that
liberation had taken place. The presence of sperm pools and the evidence of libera-
tion of nauplii were carefully noted whenever they occurred.

31

32 DU J~GRISPSAND, Be Soren bie

In some of the experiments the barnacles were fed liberally on Artemia larvae,
and in others no food was given, a change of water being made only once a day.
Both the fed and the starved animals were maintained at a series of temperatures.
One series was kept in a thermostatically controlled refrigerator at 3=1° C,,
a second in a thermostatically controlled cool cabinet at 5 to 7° C., a third set in
a cool basement, the temperature of which differed by only 1-2° C. from that of
the mean sea water temperature, and a fourth in a laboratory where the temperature
remained considerably higher, ranging between 15-21° C.

In order to study the influence of the parasite Hemiomiscus balani on the moult-
ing rhythm, infected specimens of 6b. balanoides L. and of Elminius modestus
Darwin were collected from Brixham, S. Devon, where a high level of infection
occurs (Southward and Crisp, 1954). Infected specimens of B. amphitrite var.
denticulata Broch and B. perforatus Bruguiere were collected from the warm water
docks at Swansea and from the coast of the Gower Peninsula, respectively. For
these experiments animals were isolated in dishes and fed on Artemia larvae;
subsequent examination revealed whether they were infected.

SEASONAL VARIATION IN THE MouULTING RHYTHM

Darwin (1854) in his monograph on Cirripedia mentioned that he was informed
by a Mr. Peach that off the coast of Cornwall barnacle exuviae were most abundant
in April-May and again in September. The specimens sent to Darwin included

eee ao nen
TEMPERATURE
Ee

INES

‘CASTS PER DAY PER BARNACLE

mu - i e Shy i
ocr DEC [JAN | FEB [MAR | APR | MAY [JUN | JuL | AUG [SEPT [OCT [NOV [Otc [JAN F MAR_| APR [MAY JUN | Juc[AuG [sept [oct [Nov | Dec [JAN
1954-55

1955-56 1956-57

Ficure 1. Seasonal variation in the moulting rhythm of groups of B. balanoides L. collected
at approximately fortnightly intervals during the years 1954-57. Periods when fertilisation
is taking place are shown by dark’ rectangles, and periods when liberation takes place are
shown by clear rectangles.

B. balanoides L., B. perforatus Bruguiére and Chthamalus stellatus Poli, but it is
difficult to say to which species these exuviae belonged. Darwin also quotes Mr.
Thomson’s observation that 20 specimens of B. balanoides kept alive for 12 days
shed 21 casts; presumably one must have moulted twice and the rest once. How-
ever, no information is given regarding the condition under which they were main-
tained.

The seasonal differences in the frequency of moulting of B. balanoides over a
period of two successive years were determined from the moulting frequency of
groups collected at short intervals, roughly once a fortnight, after which a new
batch was substituted for the previous batch. It was found unnecessary to feed
this species artificially, since for these short periods the moulting rate was not

~

MOULTING OF BALANUS BALANOIDES 33

appreciably influenced (see below and Table III). The moulting rates of such
groups are presented in Figure 1 as per cent casts per day, together with tempera-
tures both of the sea and of the cool basement where the groups were kept for
observation.

It will be clearly seen from Figure 1 that the rate of moulting, which was
about 8-12% per day in September—October, fell dramatically by November—
December after the natural population had become fertilised. During December—
January animals almost ceased to moult. The period of anecdysis before the animals
resumed moulting varied slightly with the individual and its environment. There-
after the moulting rate increased gradually from 3-5% per day during late Febru-
ary—March, to a maximum value of 8-12% per day by late April-May. How-
ever, it dropped slightly by June-August, but increased thereafter till breeding
activity started.

The seasonal variation of temperature did not correspond in any way with
changes in the frequency of moulting, for, though the temperature remained fairly
high during summer, the frequency of moulting in both years diminished in mid-
summer and rose again by September and October.

Loss oF PENIS

Crisp and Patel (1958) reported that in B. balanoides only, the first cast after
the period of anecdysis contained all tissues of the penis, separated by an abscission
layer of new cuticle, and that this phenomenon took place whether the particular
animal had been fertilised or not. Figure 2 shows the appearance of the cast skin
of this species at different times of the year and the opaque appearance of the penis
(Fig. 2b) in the first cast after the end of anecdysis is due to the tissues remaining
in it.

During the year 1957-58 measurements of the size of the penes of from 10
to 15 cast skins were taken every day. The individuals were of the same approxi-
mate age group, mostly exceeding two years; no young individuals were used.
During the period of anecdysis, measurements of penis length were made from
individuals removed from the experiment. The results are shown as a series of
mean values in Figure 3. A new penis gradually developed during the period of
summer growth, reaching its maximum length before the onset of the next breeding
season (Fig. 2a), and was subsequently lost after the period of anecdysis as
described above.

Figure 4 illustrates the moulting behaviour after fertilisation had occurred
and shows that all individuals gave out this particular type of cast skin once, and
once only, after anecdysis; subsequent moults were normal, a stump being left in
the position of the penis (Fig. 2c).

In order to confirm that the animals did not give off their penes as a result
of laboratory conditions, a number of specimens were collected from the shore
during the second week of January, 1957, and the penis length measured. Figure
5 shows the distribution of penis length throughout the population. The results
show a clear bimodality on account of there being two types of individuals, those
which had lost and those which had retained the penis. By distinguishing between
those which had lost and those which had retained the penis it was possible to

34

Dire | WCRIS ESN DB oneal ea

Ficure 2. Photographs of the cast skins at different times during the year, showing the
seasonal changes in appearance of the penis. (a) September-November: Cast with the skin
of fully developed penis. (b) January-February: First cast after the period of anecdysis
with all tissues of the penis shed in the penis skin, rendering it opaque. (c) March-April:
Post liberation cast with remains of egg masses, scarcely any penis present. (d) May—August:
Summer cast with developing penis stump.

show that the loss of penes occurred earlier in barnacles growing at high water
mark than those growing at mean tide level and low water mark (Table 1). This
appeared to be due to the fact that breeding activity started first in those settled
high up on the shore (Crisp, 1959) and which were therefore correspondingly in
advance of those lower down in casting the penis.

To investigate whether fertilisation was a necessary prelude to the loss of the

(b)

(d)

MOULTING OF BALANUS BALANOIDES 35

1s
=
=
=
PS
S10
=
2
=
5
°

SEP OcT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG

Figure 3. Seasonal changes in the development of the penis, measured from the cast skins.
E], mean values from a highly bimodal distribution of lengths due to some individuals having
shed the penis. The mean size of those retaining the penis is given thus @. Note the
shrinkage in size after fertilisation, and eventual loss of the organ.

penis, mature animals collected at the end of October and early November were
isolated in dishes to prevent the occurrence of copulation and fertilisation of eggs.
They were maintained at temperatures corresponding closely with those in the
sea. Not only the individuals which already had fertilised egg masses, but also

TOTAL CASTS

ANECDYSIS
ECDYSIS WITH
PENIS TISSUE

NORMAL

w
w
>
ra)
UO
WwW

MEAN NUMBER OF MOULTS PER BARNACLE (CUMULATIVE)
FERTILISATION

1e)

DEC JAN FEB MAR

Ficure 4. Moulting rate of 36 specimens of B. balanoides L. at 7° C. between 8.XII.1957
and 29.111.1958 showing how, after the period of anecdysis, the first moult contains all tissues
of the penis and the subsequent moults are normal. @, total number of all casts shed. ©, casts
containing penis tissue, shed in January and February, one only per barnacle. FJ), normal casts
shed after the middle of February.

36 De}. CRISP"AND BS we AT EL

60

% STILL RETAINING PENIS

ie)
fo} I 2 3 4 5 6 7
PENIS LENGTH IN MM.

Ficure 5. Distribution of penis lengths in a natural population of about the same year
group on January 11th, 1957, showing clear bimodality on account of there being two types
of individuals, those having lost and those still retaining the penis.

those which were still unfertilised gave off the cast with penis tissues (Table IT).
This indicates that fertilisation of itself had no marked effect on the loss of the penis.

B. balanoides could also be prevented from becoming fertilised by keeping
them at fairly high temperatures (15°-21° C.) throughout the summer and autumn.
These unfertilised animals continued to retain a normal penis, but if they were
subsequently transferred to a temperature of 6° C. and kept cool for several weeks,
they entered the breeding condition, became fertilised, and showed the typical pat-
tern of anecdysis and loss of penis. Others were maintained at 5° to 7° C. through-
out the summer and autumn, but, on account of abnormal laboratory conditions
not fully understood, reached maturity considerably later than those kept in the
cool basement or those growing normally in the field. These animals, although
kept in the cold for a long period, retained a normal penis until breeding had taken
place. After having bred, all individuals gave off their penis with the first cast
following anecdysis. Furthermore, in some experiments one set of barnacles were
starved and another fed; this did not influence in any way the loss of the penis.

It is clear therefore that the loss of the penis is not directly dependent on ferti-
lisation nor on maintaining the animals at low temperature, nor on the availability
of food. It appears to be a part of a normal physiological cycle, in which the
gonads as a whole undergo recession, the loss of the penis being accompanied by

TABLE [|

Percentage of fertilised population, collected from different tide levels, which had still
retained the penis on January 11th, 1957

Level % with penes
H. W. 10.5
iY toes We 20.4

LW, 37.8

MOULTING OF BALANUS BALANOIDES 37

TABLE II

Percentage of animals which lost the penis after the period of anecdysis when
; maintained at 7° to 12° C.

Percentage which gave off casts

Number of specimens used an gv
SPECIMENS USE containing penis tissues

Date animals were removed

for examination
Animals without Animals with Animals without Animals with
egg masses egg masses _ egg masses egg masses
9.1.1958 188 25 42.5 28.0
Meet OS f 14 12 50.0 50.0
21.11.1958 30 36 97.0 94.0

shrivelling of the vesiculae seminalis and degeneration of testes. This takes place
only after they have experienced such conditions as allow the gonads to reach
their full development, and under normal circumstances for them to become
fertilised. |

This phenomenon was observed only in this species. Crisp (1954), however,
had reported in Balanus porcatus (da Costa) a shortening in the length of the
penis after fertilisation, followed by gradual lengthening during the summer,
reaching a maximum length just before copulation.

Mou.tinGe RATE OF YOUNG BARNACLES

Costlow and Bookhout (1953, 1956) studied the moulting rate of recently
settled cyprids of Balanus amphitrite niveus and B. wmprovisus under varying con-
ditions of light and food till they reached the adult stage.

During the present investigation the moulting rate of recently settled spat of
B. balanoides was studied.

bh

WwW

=
=
=
oa
<
=
w
«
w

NUMBER OF MOULTS (CUMULATIVE)
I)

° = 10 15 20 25
AGE IN DAYS

Ficure 6. Moulting rate of young spat of B. balanoides L. of known age when kept
atele 3G, andatee Non

38 DJ? CRISPJANDEBE See AEE,

Cyprids which had settled on inert plastic plates and metamorphosed were
brought into the laboratory and kept in dishes. The water in the dishes was stirred
mechanically and Chlamydomonas sp. were offered every day as food. Since satis-
factory conditions could not be guaranteed in the laboratory, these plates were
returned to the sea. At the same time another batch, which had been put out
for settlement, was brought into the laboratory, and was placed in a dish; the
water was stirred and the animals fed as before. Several plates were thus kept
in rotation between the laboratory and the sea, a few plates being kept continuously
in the laboratory. The number of exuviae was recorded each day.

Figure 6, drawn from the sets of results, illustrates the cumulative number of
casts shed during the first few weeks after settlement. Metamorphosis of cyprids
occurred after 24 to 36 hours and the apodemes of the paired eyes, the bivalve
carapace and the exuviae of the thoracic limbs were shed (Crisp and Stubbings,
1957). No further ecdysis occurred for at least 5 days. The time of the first
adult cast varied considerably, but on average it took place between 9 and 10
days after metamorphosis. This was longer than the corresponding period for
B. amphitrite (Costlow and Bookhout, 1956), in which the first cast occurred
between 3 and 5 days after metamorphosis. Thereafter B. balanoides moulted
approximately once every four days at 18° C. and about once every six days
at 12° C. From Figure 6 it will be clearly seen that the moulting rate of the
rapidly growing spat was at least double the maximum rate of moulting of the
adult over the same period of the year (Fig. 1).

_ Factors INFLUENCING EcpysIs
a. Feeding

Groups of B. balanoides were maintained at several temperatures, some being
fed on Artemia larvae and other groups having the water changed only. Those
groups which were given Artemia were observed to be beating the cirri intermit-
tently throughout the day and producing dark red faeces. The starved groups,
on the other hand, were active for a short time after the water in the dishes had

PABrE Uhl

Moulting rate of B. balanoides under varying conditions of temperature and feeding.
The animals were collected on July Ist, 1957

Condition Rate of moulting

Number of

animals

available Fed or Mean First 10 days after First 60 days after

starved temperature collection collection

46 fed oC 0.10 casts per day 0.13 casts per day
47 starved 1D ae 0.10 casts per day 0.065 casts per day
29, fed 70°C 0.12 casts per day 0.12 casts per day
21 starved US 0.10 casts per day 0.06 casts per day
27 fed | oC 0.05 casts per day 0.046 casts per day
29 starved 6.0" C. 0.05 casts per day 0.032 casts per day
ay fed 30°C 0.05 casts per day 0.03 casts per day
wy| starved 0.05 casts per day 0.03 casts per day

MOULTING OF BALANUS BALANOIDES oy)

CASTS CONTAINING
PENIS TISSUE

“= SPERM POOL
LIBERATION ‘

S|
°
ce)
a
=
a
wi
o
wm)

CASTS CONTAINING
PENIS TISSUE

STARVED

MEAN NUMBER OF MOULTS PER BARNACLE (CUMULATIVE)

JUL AUG SEP OocT NOV DEC JAN FEB MAR APR- MAY JUN JUL
1957-58

Figure 7. Moulting rate of groups of B. balanoides L. showing the effect of starvation
at 5-7° C. Sperm pools indicate that fertilisation was taking place. Casts containing penis
tissues are shown by oblique lines and liberation by small arrows.

been changed, but remained quiet for the rest of the day except for very occasional
eating.

As can be seen from Table III, both fed and starved batches moulted at approxi-
mately the same frequency immediately after collection and for the following ten
days. Thereafter the fed barnacles continued to moult at the same or at a slightly
decreased rate, but the starved barnacles showed a progressive reduction in fre-
quency but did not cease to moult altogether. These long-term changes are shown
in Figure 7 for animals kept at 5 to 7° C. It will be noticed that after anecdysis
the moulting rate was low, but it recovered much more rapidly in those individuals
which were fed. These results suggest that the amount of reserve food available
is important in determining the moulting rate. The results in Table III show
that at higher temperatures there is an even more marked difference between fed
and starved individuals due, no doubt, to the greater loss of reserve food at the
higher metabolic rate.

b. Temperature

Southward (1955a), who studied the influence of temperature on the cirral ac-
tivity of B. balanoides, reported that the rate of cirral beats increased linearly with
the increase in temperature from 3 to 20° C. The rate of beating, however, fell
with a further rise in temperature, 31 to 32° C. being lethal.

Similarly (Fig. 7) the moulting rate of barnacles fed on Artemia increased
linearly with the rise in temperature from 3 to 20° C. The moulting rate of
starved specimens, on the other hand, increased from 3 only up to 12° C. and
then diminished with a further rise in temperature. If, as seems likely, the rate
of moulting is influenced by the food reserves available, it would not be surprising
that with increased metabolic rate at higher temperatures the reserves would fall

40 Dale GRISPAND - BSP AtT EE

sufficiently to reduce the moulting rate. After a sufficient period the effect of loss
of reserves might more than counteract the normal increase in moulting rate with
temperatures, so that the moulting rate actually fell with rise in temperature.

Both fed and starved specimens moulted at the same frequency at the lowest
temperature (Table III, Fig. 8) ; the reason for this might be the fact that at 3-4°
C. those which were offered food remained quiet and fed very little, while those
which were starved did not lose reserve metabolites at any appreciable rate at low
temperatures.

Southward kept his specimens only for a very short time at 25 to 30° C. and
was able to measure their cirral beat at these temperatures; however, both fed and

+15 MOULTING RATE

4 (FED)
2 7
(a)
= CIRRAL ACTIVITY ay Ss
Z
7 oie) S °
2 a
x ba
Ww)

a 3 e
WwW <
a 6-05 WwW
v ao
a MOULTING RATE 2
5 (STARVED) .

ro) re)

oO 5 10 15 20 25 30
TEMPERATURE IN DEGREES CENTIGRADE

Ficure 8. The effect of temperature and feeding on the moulting rhythm of B. balanoides L..
and the effect of temperatures on cirral activity (from Southward, 1955a).

starved specimens during the present series of experiments died if kept for more
fiauethincetdays ate2o. 10.27) ©:

c. Breeding

The majority of recently fertilised animals abruptly ceased to moult for some
six to eight weeks (Fig. 9); this might be regarded as an adaptation to prevent
recently oviposited eggs, which are not yet hardened, from being shed with a cast.
However, after anecdysis, the still gravid barnacles resumed moulting, especially
if they were fed (see Figure 7) and were found to have cast the skin with a torn-off
mantle lining, due to the pressure exerted by the egg masses. The latter had by
this time become hard and were pressed well up against the mantle lining of the
parietes and basis. Occasionally only the exuviae of the appendages and prosoma
were shed, the mantle lining being retained and shed later at the time of the libera-
tion of the nauplii hatched out from the egg masses.

It will be seen from Figure 9 that the barnacles which were kept from being
fertilised by being artificially isolated continued to moult when the fertilised popu-
lation had ceased. Nevertheless they showed a sharp fall in moulting rate by the
end of November. This suggests that the drop in the moulting rate may be due
to a basic physiological rhythm, like that controlling the loss of the penis and

MOULTING OF BALANUS BALANOIDES 41

degeneration of the gonads, and that the sudden onset of this process in fertilised
barnacles may be due to the stimulus of copulation or oviposition.

d. Effect of emersion on moulting

As Darwin (1851) observed, the skin cannot be shed except when the barnacle
is immersed in water. The period of emersion of organisms which grow between
low water neap and high water neap tide levels never exceeds about 12 hours. In
a humid environment intertidal barnacles may be kept out of water quite healthily

UNFERTILISED
(ISOLATED)

IWNGIAIGNI Yad AVG Yad 1SVD 7.

OCT NOV DEC JAN FEB MAR APR
SEASON

Figure 9. Moulting rate of isolated specimens not carrying embryos and of those with
fertilised egg masses, measured over the same period. Fertilised specimens show an immediate
drop to zero; unfertilised specimens continue to moult freely in mid-November, but show a
fall in late November at the end of the normal breeding season.

for several days. It is thus possible to ascertain whether the frequency of moulting
is influenced by an abnormally long period of emersion.

Four groups of barnacles growing under nearly identical conditions on settle-
ment plates on a raft were brought into the laboratory. Here they were kept
under a damp cloth for varying periods of time. The total number of casts shed
when the barnacles were introduced for a period of 24 hours into aerated sea water
was recorded. In Table 1V the aggregate percentage of casts (per 100 barnacles)
is shown against the total number of days kept in the laboratory. The letter “D”

42 Di MCRISP AEN DIB Snbeawt eae

TABLE IV

Moulting rate in relation to the treatment of various groups of barnacles; letter ‘‘D”’
indicates the days when the specimens were kept dry, on the other days
they were kept in sea water at 12° C.

(1) (2) (3) (4) (Soni (6) (7)

Numb f
Group parndclesased 1st day 2nd day 3rd day 4th day Toes after
in experiment wy

a 382 D D D 26.2% 100

b 373 D D 22.0% 35.4% 132

C 370 D 14.3% 20.2% 28.6% 106

d 453 3.1% D D 35% 150
Mean = = = = 30.9% (7.7 © per day)
x2 = 9.98

indicates the days when the groups were kept dry, and no casts were therefore
emitted. Column 7 gives the total number of casts for the whole period.

It is clear that moulting proceeds steadily throughout the experiment and is
not delayed significantly as a result of the barnacle being out of water. The num-
bers of casts found after 4 days in the different groups, a, b, c and d, which were
treated differently give a value for y” of 9.98 when compared with the expectation
of 7.7% per day per individual. This is slightly higher than the value of x7 = 7.8
for the 5.0% significance level. Since, however, there was no consistent trend in
the moulting rate in relation to the emersion period, the significance of the x?
value was probably due to intrinsic variations in the rate of casting of the four
groups, a, b, c and d; such variations are frequently observed between groups of
individuals from apparently similar habitats. The data of Table IV, rearranged
to show the moulting rate in relation to the time kept out of water, are displayed
in Table V. The rates of casting are closely uniform and the x? test applied to
the differences was found not to be significant (,? = 6.26 for 3 degrees of freedom).
The number of casts produced in a given time is therefore not altered significantly
by emersion up to a period of four days.

TABLE V

Moulting rate in relation to the total time out of water

(1) (2) (3) (4)
Barnacle days .
Numb f days Moult t
te ee Beet ou (hos ot barnadie: Number of casts (% cubis Seen

4 1528 100 6.55%

3 2478 215 8.69%

2 740 53 717%

1 1566 120 7.68%

MOULTING OF BALANUS BALANOIDES A3

It was often noticed that the rate of moulting for the first day or two was
higher than the average for the whole period. This may have been partly due to
the individuals having been out of water since the previous high tide. Possibly
also the changed conditions in the laboratory caused a temporary rise in the rate
of moulting. However, this small initial rise had no significant influence on the
average rate taken over two or three weeks.

e. Tidal periodicity

In many marine organisms, especially in lamellibranchs and in annelids, the
effect of moon and tide on the breeding cycle had been well established (Korringa,
1947.) However, Crisp and Davies (1955) reported that Elminius modestus
bred at any time irrespective of the tidal cycle.

20 1eaDOC@DOC ODOC ODOC Ee )OC@

CASTS PER DAY PER BARNACLE

JUN AUG
1957

Ficure 10. Average moulting rate of the groups of B. balanoides L. collected at monthly
intervals and kept at sea temperature over the spring and neap tide periods (shown by moon
phases). Stippled blocks, average moulting rate over the spring tide period; clear blocks,
average moulting rate over the neap tide periods.

%o

Churchill (1917-18) studied the effect of the lunar cycle on the moulting rhythm
of the blue crab Callinectes sapidus, but on the basis of his results dismissed the
popular belief prevailing that both moon and tide had a marked effect on the
moulting of crabs as being a folklore superstition. However, Wheeler (1937) in
Anchistioides, Nouvel and Nouvel (1939) in the mysid Praunus flexuosus and
Nouvel (1945) in Leander serratus found that the largest numbers of exuviae were
shed on the day corresponding to the largest tide. However, their observations
were made only over a single lunar period and may therefore be fortuitous.
Kinne (1953, 1959) found no tidal rhythm in the moulting of the amphipod
Gammarus duebeni. Tidal rhythm in moulting does not therefore seem to be of
general occurrence in the Crustacea.

+4 De jC RISE VAN DEB ese ead Bile

During the present investigation experiments were carried out to test if any
tidal rhythm could be found in barnacles. Each tidal cycle was divided into
two periods, approximately midway between the largest and the smallest tides,
one period corresponding to the spring and the other to the neap tides (shown
by moon phases in Figure 10). The average moulting rate of groups of animals
collected at monthly intervals was determined for pairs of successive neap and
spring periods. The difference between the average for neap and spring periods
was tested by applying Student’s “t” test and found to be insignificant. Ecdysis
therefore occurred at approximately regular intervals, provided barnacles were
immersed, irrespective of the tidal cycle.

{. Effect of tidal level

Animals were collected from three different tidal levels and their moulting
rate for the first ten days was recorded. It can be seen from Table VI that there
was no significant difference in the moulting rhythm. Though Southward (1955b)
did record specimens (B. balanoides) collected from low water mark beating

TABLE VI
Moulting rate of B. balanoides, collected from different tide levels, for first 10 days

Casts per day per individual
Pe ee WSLE Temperature
H. W. M. T. L. W.
22nd Jan. S10 7G. 0.03 0.04 0.05
29th Jan. 8-10° C. 0.025 0.03 0.04
28th Feb. 8-10° C. 0.06 0.04 0.04
5th Sept. 14-16° C. 0.110 0.10 0.12

markedly faster than those collected from higher up, when he repeated the experi-
ment on the following day with the same specimens he found that the differences
had disappeared. Neither moulting nor cirral activity appears to be influenced
by tidal level.

g. Influence of light on moulting rhythm

It has been claimed that light has an inhibitory action on the moulting rhythm of
Crustacea (Nouvel, 1945). However, Costlow and Bookhout (1956) did not find
any significant differences in the moulting rate of the series of young spat of B.
amphitrite niveus reared under different conditions of illumination.

During the present series of experiments the influence of light on the moulting
rhythm of adult barnacles was investigated. Groups of barnacles were collected
from the shore and maintained under three different conditions. One batch was
kept under a light-tight box kept in continuous darkness. A second batch was
kept in a continuously illuminated, well ventilated box, illuminated by a 25-watt
bulb. The third batch was exposed to natural variations of light conditions, 12
hours light during the day and 12 hours darkness at night. All were kept at the

MOULTING OF BALANUS BALANOIDES 45

same temperature of 10° C. No significant difference in moulting rate was found
over a period of 50-60 days.

h. Influence of parasite Hemioniscus balani

The majority of the species of British barnacles was found to be capable of
harbouring the isopod Hemioniscus balani in the mantle cavity, in the space nor-
mally occupied by the fertilised egg masses. Crisp (1954) found this parasite
causing castration of the female gonads in Bb. porcatus (da Costa) and later a
similar effect was reported in Elminius modestus (Crisp and Davies, 1955).

The effect of the parasite on the moulting rhythm was therefore investigated
using infected B. balanoides, Elnunius modestus, Balanus perforatus and B. amphi-
trite var. denticulata. It will be seen from Table VII that the presence of the
parasite had no significant effect on the moulting cycle of the host; the intermoult
periods of infected and uninfected barnacles were not significantly different. How-
ever, as with fertilised individuals which had eggs in the mantle space, the speci-

TABLE VII

Frequency of moulting of specimens infected with Hemioniscus balani and of uninfected
specimens collected from the same locality

Mean intermoult period
Percentage BCE SEO
Number of : >
Species Temperature animals used ceed week
in experiment Balagns Individuals
with Individuals
Hemioniscus not infected
Balanus balanoides 12% 422°C. 35 57% 10.9 9.0
Elminius modestus ae oe Oe 10 50% 7.3 13
Balanus perforatus Ise Zee. 20 ST, 6.0 6.4
Balanus amphitrite Wore, 2G) 100 6% 10.3 9.8

mens harbouring the parasite were found to shed exuviae with torn-off mantle
linings. On no occasion was the parasite rejected with the cast, except in the
large barnacle B. perforatus. This species was observed to reject the parasite
enclosed within the casts, and this may account for the normally very low degree
of parasitism present compared with that found in other species in the same
vicinity.

SUMMARY

1. The seasonal variations in the frequency of moulting of boreo-arctic species
of the operculate barnacle Balanus balanoides L. were studied during the years
1954-57 by observing groups of barnacles kept in the laboratory at temperatures
corresponding to that of sea water. The rate of moulting, which was about 8 to
12% casts per day per barnacle during October-November, fell sharply by Novem-
ber—December and fertilised animals underwent a period of anecdysis for about
6 to 8 weeks. They resumed moulting at a slower rate, reaching a maximum by
May with a slight decrease in June-August, rising again towards the breeding
season in November.

46 Di. CRISP AND Bessa Aaliey

2. In this species the first cast after the period of anecdysis always contained
all tissues of the penis, separated by an abscission layer of new cuticle. Neither
food, temperature nor the act of fertilisation were directly responsible for the
loss of the penis. The evidence indicates that the loss of the penis is part of a
physiological cycle in which gonads undergo recession after they have reached
full development. A new penis gradually developed during the period of summer
growth, reaching its maximum length before the onset of the breeding season.

3. Feeding influenced the moulting rhythm, but the effect made itself felt only
after the first 10 to 15 days following collection from the shore. The moulting
rate of specimens maintained without food for a period longer than this (1.e., for
about 30 to 60 days) fell considerably from the initial value, but animals fed on
Artemia larvae continued to moult at about the same rate as when freshly brought
in from the shore.

4. The moulting rate of specimens given food increased linearly with the rise
in temperature from 3—20° C.; on the other hand the moulting rate of starved
specimens only increased from 3-12° C. and fell considerably with further rise.
Temperatures higher than 25° C. were lethal to both groups.

5. Neither the lunar cycle nor the tidal level had any influence on the moulting
rhythm.

6. The parasite Hemioniscus balani, which caused castration of the female
gonads, did not influence the frequency of moulting.

EEE RA TURE Cian

‘CHURCHILL, E. P., Jr., 1917-18. Life history of the blue crab. Bull. U. S. Bur. Fish., 36:
95-128.

Costitow, J. D., Jr., anp C. G. Bookuout, 1953. Moulting and growth in Balanus improvisus.
Biol. Bull., 105: 420-433.

‘Costtow, J. D., Jr, ann C. G. Bookuout, 1956. Molting and shell growth in Balanus
amphitrite niveus. Biol. Bull., 110: 107-116.

Crisp, D. J., 1954. The breeding of Balanus porcatus (da Costa). J. Mar. Biol. Assoc., 33:
473-494.

‘Crisp, D. J., 1959. The rate of development of B. balanoides L. embryos in vitro. J. Anim.
Ecol. 28 119=132:

Crisp, D. J.. anp P. A. Davies, 1955. Observations im vivo on the breeding of Elmunius
modestus grown on glass slides. J. Mar. Biol. Assoc., 34: 357-380.

Crisp, D. J., anp B. S. Pater, 1958. Relation between breeding and ecdysis in Cirripedes.
Nature, 181: 1078-1079.

‘Crisp, D. J.. AnD H. G. Stuppines, 1957. The orientation of barnacles to water currents.
J. Anim. Ecol., 26: 179-196.

Darwin, C., 1851. A monograph on the sub-class Cirripedia. The Lepadidae or pedunculated
Cirripedes. Ray Society, London.

Darwin, C., 1854. A monograph on the sub-class Cirripedia. II. The Balanidae, the Ver-
rucidae. Ray Society, London.

Kinng, O., 1953. Wird die Hautungsfolge der Amphipoden durch die lunare Periodizitat
beeinflusst? Kieler Meeresforsch., 9: 271-279.

KINNE, O., 1959. Ecological data on the amphipod Gammarus duebem. Veroff. Inst. Meeres-
forsch. Bremerhaven., V1: 177-202.

Korrinca, P., 1947. Relation between the moon and periodicity in the breeding of animals.
Ecol. Monogr., 17: 347-381.

Nouvet, H., 1945. Les relations entre la periodicité lunaire, les marées et la mue des Crustacés.
Bull. Inst. Oceanogr., 878: 1-4.

MOULTING OF BALANUS BALANOIDES 47

Nouvet, H., anp L. Novuver, 1939. Observations sur la biologie d’une Mysis: Praunus
flexuosus (Miller 1788). Bull. Inst. Oceanogr., 761: 1-10.

SoutHwarp, A. J., 1955a. On the behaviour of barnacles. I. The relation of cirral and
other activities to temperature. J. Mar. Biol. Assoc., 34: 403-422.

SoutHwarp, A. J., 1955b. On the behaviour of barnacles. II. The influence of tide level and
habitat on cirral activity. J. Mar. Biol. Assoc., 34: 423-433.

SouTHWArRD, A. J., AND D. J. Crisp, 1954. Recent changes in the distribution of the inter-
tidal barnacles Chthamalus stellatus Poli and Balanus balanoides L. in the British
Isles. J. Anim. Ecol., 23: 163-177.

WueEe ter, J. F. G., 1937. Further observations on lunar periodicity. J. Linn. Soc. Zool., 40:
325-345.

EFFECTS OF TURBIDITY-PRODUGING MATE RLAIS “UNG sk
WATER ON EGGS ANDI ARY AE OR. Wire leave
(VENUS (MERCENARIA) MERCENARIA)

HARRY" © SDA VAS
U. S. Fish and Wildhife Service, Milford, Connecticut

The effect of suspended materials on mollusks and on the survival and growth
of their larvae is of interest to biologists and of considerable importance to shellfish
producers. Since several of the most important commercial species are essentially
inhabitants of the shallow waters of bays and estuaries, almost all of the grounds.
used for the cultivation of shellfish are frequently subject to water rendered turbid
by natural phenomena, such as floods and storms, and by human operations, such
as dredging, bridge- and road-building, etc.

An extensive field study by Lunz (1938) during the dredging of the Intra-
coastal Waterway of South Carolina showed that the mortality of adult oysters in
the dredged area, except where oysters were actually buried by the spoil, was no
higher than in areas remote from the dredging operations. Moreover, he found
no evidence that the physiological condition of the oysters, as judged by the yield of
meats per bushel of oysters, was affected by the dredging operations. He also
reported that the intensity of setting of oysters in an area adjacent to dredging
operations did not differ from setting intensity in areas remote from such opera-
tions, and concluded (p. 134) that “dredging operations apparently had no effect
on spawning and setting.”

In careful quantitative experiments Loosanoff and Tommers (1948) have
shown that as little as 0.1 gram per liter of silt reduced the average pumping rate
of adult oysters by 57 per cent. At a concentration of 1.0 g./l. they found the
reduction in average pumping rate was more than 80 per cent and reached 94 per
cent when the concentrations of silt were increased to 3.0 and 4.0 g./l. They re-
ported that results with kaolin and chalk were similar to those obtained with silt,
and that 0.5 g./l. of Fuller’s earth, the only concentration tested, reduced the rate
of pumping by 60 per cent.

Since the development of techniques for rearing bivalve larvae in the laboratory
(Loosanoff and Davis, 1950), these techniques have been used to determine quanti-
tatively the effects of various factors, such as temperature, species of food organ-
isms, crowding, quantity of foods and salinity on the survival and growth of
bivalve larvae (Loosanoff, Miller and Smith, 1951; Loosanoff and Davis, 1953;
Loosanoff, Davis and Chanley, 1953a; 1953b; Davis, 1953; Davis and Guillard,
1958; Davis, 1958). In the present study they have been employed to determine
quantitatively the effect of various concentrations of several materials suspended
in sea water on the development of eggs of the hard clam, Venus (Mercenaria)
mercenaria, and on the survival and growth of their larvae.

The turbidity-producing materials used in these experiments were clay (kaolin
N.F. VII Mallinckrodt), Fuller’s earth (dusting powder, McKesson), chalk (pre-

48

mere eis Or TURBIDITY ON CLAM VARVAE 49

cipitated chalk U.S.P. McKesson and Robb-ns) and silt. The silt was collected
from the bed of the Wepawaug River, just upstream from our laboratory. The
silt was washed through a 325-mesh, stainless steel screen (meshes approximately
50 microns square) to remove larger particles, collected in a Buchner funnel, where
it was washed with distilled water to remove salt, and then dried at 200.0° C.
Finally, the dried cake of silt was ground in a jar mill. In the later experiments
the Fuller’s earth was also ground in the jar mill. Concentrations of suspended
material are, therefore, all expressed as grams of dry powder per liter.

Figure 1. Turbidity apparatus showing the wheel and method of attachment of bottles. Motor
and reduction gear box indistinctly seen in background.

In making up a suspension, a weighed quantity of the powdered material was
thoroughly mixed with sea water and the various experimental concentrations
prepared by serial dilution.

The method for obtaining fertilized clam eggs in midwinter has previously been
described (Loosanoff and Davis, 1950) as have the methods for determining the
percentage of eggs developing to the straight hinge stage, and for determining
the rate of growth of larvae (Davis, 1953, 1958).

In the present series of experiments, wide-mouthed, 32-ounce polyethylene
bottles with screw caps were used as containers for the larval cultures. To keep

50 HARK Y. Gz DANTS

the materials in suspension we used a modification of an apparatus which V. L.
Loosanoff of this laboratory observed in P. R. Walne’s laboratory at the Fisheries
Experiment Station, Conway, England. The bottles were mounted on a vertical
wheel which, as it rotated, turned the bottles end over end once for each revolution
of the wheel. In our apparatus provision is made for 12 bottles on each side of
the wheel for a total of 24 bottles so that 12 sets of duplicate cultures can be run
concurrently (Fig. 1). The wheel is kept rotating by a constant speed motor
working through a speed reducer giving eight revolutions of the wheel per minute.
The wheel and attached bottles are enclosed in a plywood box in which the air

Ficure 2. Turbidity apparatus showing thermostat control, temperature box and
position of stationary control cultures.

temperature is thermostatically controlled at 24.0° C., using two 60-watt light
bulbs as heaters. A wire mesh shelf is installed in the left upper corner of this
box to hold two additional bottles for stationary control cultures (Fig. 2). In
practice, one pair of bottles on the wheel is used for moving control cultures, leaving
11 pairs of bottles for duplicate cultures of larvae at each of 11 different sets of
experimental concentrations of suspended materials.

The number of larvae suspended in the 800 ml. of sea water and turbidity-
producing material in each bottle was initially the same for all cultures in an

EPRECTS OF TURBIDITY/ON ‘CLAM LARVAE 51

experiment, but ranged from 8000 per bottle in some experiments to about 15,000
in others. The larvae were fed daily, each culture receiving an equal quantity
of a mixture of /sochrysis and Monochrysis. The larvae were examined every
second day when the sea water and its suspended material were changed, but
quantitative samples for counts and measurements were taken only when the
larvae were 48 hours old and on the twelfth day, when an experiment was
terminated.

Preliminary experiments indicated that some fertilized clam eggs could develop
into straight hinge larvae in concentrations of either clay (kaolin) or chalk as
high as 4.0 g./l., but that none developed to the straight hinge stage in equivalent
concentrations of either Fuller’s earth or silt. Subsequent experiments, however,

TABLE [|

Percentage of clam eggs developing to the straight hinge larval stage in different concentrations
of suspended materials. The number of eggs developing to the straight hinge stage
in the stationary controls is considered 100 per cent

Percentages
Concentration g./1.
Silt Clay (kaolin) Fuller’s earth Chalk

Stationary control 100 100 100 100
Moving control 91

0.125 95 82 TS

0.188 90

0.250 96 82 61 approx. 45

0.375 93

0.500 99 So 41

0.750 92

1.000 79 oy 57 approx. 30

1.500 65

2.000 39* 49 50 approx. 39

3.000 0

4.000 0 42** ASae* ADPLOKs4a

* Averaged 195.80 y)
** Averaged 203.90 1 | At 12 days when kept in these turbidities only during the first 48
*** Averaged 199.60 u| hours after fertilization and then returned to normal sea water.
**** Averaged 189.10 |

have shown that if Fuller’s earth was first ground in a jar mill, some larvae de-
veloped normally at 4.0 g./l. Similarly, ground silt at concentrations of 3.0 or
4.0 g./l. still completely prevented normal development of clam eggs (Table I).
In silt concentrations of 0.75 g./l. or lower, however, there were no significant
differences in the percentage of clam eggs developing normally, while with clay,
chalk or Fuller’s earth there appeared to be a more or less consistent stepwise
reduction in the percentage of eggs developing normally with each increase in the
quantity of suspended material (Table I).

Although in a silt concentration of 2.0 g./l. only 39 per cent of the eggs de-
veloped to the straight hinge larval stage, almost all of these larvae were capable
of surviving and growing to metamorphosis (mean length 195 p at 12 days) if

52 HARRY (C:.DAVAS

returned to normal sea water. Similarly, larvae developing to straight hinge stage
in concentrations of 4.0 g./l. of clay, chalk or Fuller’s earth have been reared to
setting stage after being returned to normal sea water at the end of 48 hours.
Experiments on the effect of suspended materials on the growth of clam larvae
that have developed to the straight hinge stage in our usual sea water have shown
that such larvae cannot grow in concentrations of clay, Fuller’s earth or chalk
as high as those at which some eggs developed. For example, there was no evi-

110

100

a
s
e 4

GROWTH AS PERCENTAGE
a
(oe)

ao
°o

ie) 1 2 3 4
CONCENTRATIONS IN GRAMS PER LITER

Ficure 3. The 12-day increase in mean length of clam larvae, grown in different concen-
trations of suspended materials, plotted as percentages of the increase in mean length of larvae

in control cultures. Each point represents the average for 50 larvae from each of duplicate
cultures in each of two experiments.

dence that the larvae were taking the flagellates provided as food, and there was
no growth of the larvae in a concentration of chalk as low as 0.250 g./l. With
either clay or Fuller’s earth, 0.500 g./l. was the highest concentration in which
the larvae showed evidence of taking food (color in digestive gland) or any in-
crease in size (Fig. 3). In cultures receiving higher concentrations of clay or
Fuller’s earth, many larvae ingested some of the suspended particles, apparently
in sufficient quantity to block the digestive tract and cause death. Kaolin in a
concentration of 0.500 g./l. caused approximately 50 per cent mortality in 12

BRPECTS OF TURBIDITY ON CEAM LARVAE 53

days and almost complete mortality at all higher concentrations. There was no
appreciable mortality of larvae in 0.500 g./l. of Fuller's earth in 12 days, but at
all higher concentrations mortality exceeded 90 per cent.

When silt was used as the suspended material the results were quite different
from those with clay, Fuller’s earth or chalk. Clam larvae showed evidence of
taking food (color in digestive gland) even at a concentration of 4.0 g./l. of silt,
but growth was negligible, and it is doubtful if any larvae in natural waters could
reach metamorphosis in either 3.0 or 4.0 g./l. of silt. Growth of most larvae was
also considerably retarded by concentrations of 2.0 g./l. or even 1.5 g./l. of silt,
but some larvae under these conditions had reached setting size (maximum length
185 » and 200 pn, respectively) by the twelfth day. Growth of clam larvae was
approximately normal in 0.750 g./l. of silt and, at all lower concentrations, was
somewhat better than that of larvae in the control cultures (Fig. 3). There was
no appreciable mortality of clam larvae, within 12 days, in any of the concentrations
of silt tested.

Soil particles ranging in size from 62 microns to 4 microns are listed as silts,
while particles ranging in size from 4 microns to 0.24 microns are listed as clays.
From our experiments it would appear that it is the larger particles (coarse silt
62 to 31 microns) in the silt and unground Fuller’s earth, or aggregates of particles
in other substances that interfered with development of clam eggs, while the
smaller particles, characteristic of the clays, probably had little effect on egg devel-
opment. The results of experiments on growth of larvae, however, seem to indicate
that the larger particles, characteristic of coarse silts, may have little effect on
growth. It seems to be the smaller particles, as in clays, precipitated chalk, and
finely ground Fuller’s earth, which are about the size of food cells, that interfere
most with growth of larvae. This interference appears to be primarily mechanical
through blockage of the digestive tract, but there is some indication that it may also
have been, at least in part, an indirect result of an adverse effect of the suspended
materials on the food organisms.

The more rapid growth of clam larvae noted in lower concentrations of silt
and at the lowest concentrations of clay and Fuller’s earth is perhaps due, in part,
to chelation of toxic substances, and with silt, in part, to positive growth factors.
An improvement in the rate of growth of algal cultures upon addition of soil
extract has long been known to botanists, and soil extract has been a basic ingredient
in many of the media for algal cultures.

The author wishes to express his appreciation of the suggestions and construc-
tive criticism given by Dr. V. L. Loosanoff, Director of Milford Laboratory.
Thanks are also due the mechanical staff who constructed the wheel and constant
temperature box, to Kenneth Spencer for preparing the figures and to Miss Rita
Riccio for her careful editing of the manuscript.

SUMMARY

1. Some clam eggs developed normally in concentrations of 4.0 g./l. of clay,
precipitated chalk or finely ground Fuller’s earth, although the percentage develop-
ing normally decreased as the concentration of these suspended materials increased.

54 HARRY C.pAVvis

2. In silt concentrations below 0.75 g./l. the percentage of clam eggs devel-
oping normally was not significantly different from that in control cultures but
decreased progressively in successively higher concentrations.

3. None of the clam eggs developed normally in silt concentrations of 3.0 or
4.0 g./L.

4. Larvae resulting from clam eggs developing in high concentrations of each
of the suspended materials were reared to metamorphosis after being returned to
normal sea water at 48 hours.

5. Clam larvae were unable to grow in concentrations of clay, chalk or Fuller’s
earth as high as those at which some eggs developed.

6. The highest concentration of chalk was 0.25 g./l. and 0.5 g./l. was the
highest concentration of clay and Fuller’s earth at which clam larvae showed any
growth and mortality exceeded 90 per cent at all higher concentrations.

7. Ina silt concentration of 0.75 g./l. growth of clam larvae was approximately
normal and at lower concentrations was slightly faster than that of larvae in control
cultures.

8. In silt concentrations of 1.0 to 2.0 g./l. growth of clam larvae was retarded
and at 3.0 and 4.0 g./l. growth was negligible.

9, Even at a silt concentration of 4.0 g./l. there was no appreciable mortality of
clam larvae within 12 days.

LITERATURE, CipED

Davis, H. C., 1953. On food and feeding of larvae of the American oyster, C. virginica.
Biol. Bull., 104: 334-350.

Davis, H. C., 1958. Survival and growth of clam and oyster larvae at different salinities.
Biol. Bull., 114: 296-307.

Davis, H. C., anp R. R. GurLiarp, 1958. Relative value of ten genera of micro-organisms as
foods for oyster and clam larvae. U. S. Fish & Wildlife Service Fish. Bull. 136, 58:
293-304.

LoosANorF, V. L., AND H. C. Davis, 1950. Conditioning V. mercenaria for spawning in
winter and breeding its larvae in the laboratory. Bzol. Bull., 98: 60-65.

LoosAnorF, V. L., AND H. C. Davis, 1953. Utilization of different food organisms by clam
larvae. Anat. Rec., 117: 646.

LoosanorF, V. L., H. C. Davis And P. E. CHANLEy, 1953a. Behavior of clam larvae in
different concentrations of food organisms. Anat. Rec., 117: 586-587.

LoosAnorF, V. L., H. C. Davis ann P. E. CHANLEy, 1953b. Effect of overcrowding on rate
of growth of clam larvae. Anat. Rec., 117: 645-646.

LoosanorF, V. L., W. S. MILLER AND P. B. Situ, 1951. Growth and setting of larvae of
Venus mercenaria in relation to temperature. J. Mar. Res., 10: 59-81.

LoosANorr, V. L., AND FrANcES D. Tommers, 1948. Effect of suspended silt and other
substances on rate of feeding of oysters. Science, 107: 69-70.

Lunz, R. G., 1938. Part I. Oyster culture with reference to dredging operations in South
Carolina, 1-135. Part II. The effects of the flooding of the Santee River in April
1936 on oysters in the Cape Romain area of South Carolina. Report to U. S. Engineer
Office, Charleston, South Carolina, 1-33.

meee REGULATION AND SURVIVAL OF THE RED ALGA,
Pere vin PRERPORATA, IN DILUTED AND
CONGENTRATED SEA. WARE R A

RICHARD IW: EPPLEY AND CHARLES, C..CYRUS

Department of Biology, University of Southern California, Los Angeles 7, California

Scott and Hayward (1953, 1954), Biebl (1956), and one of us (Eppley, 1958a,
1958b) have considered that a significant feature of ion transport in algae may lie
in maintaining a relatively constant intracellular ionic environment, 1.e., cellular
homeostasis. That transport mechanisms are of survival value is attested to by
the great salinity tolerances established for intertidal algae (Biebl, 1952, 1953,
1956, 1958). Although there is increasing interest in homeostatic mechanisms
in marine organisms (e.g., Bullock, 1958), data on ion regulation in algae during
osmotic stress are scant.

Blinks (1951) has reported the accumulation of K by Valonia from slowly
concentrating sea water. Ulva lactuca loses some K in 70 per cent sea water which
is reaccumulated when external Na is returned to normal (Scott and Hayward,
rs).

Ion transport has been implicated in a number of algae in addition to those
mentioned. Bergquist (1958a) has studied Na and K movements after desiccation
in Homosira banksu, MacRobbie and Dainty measured K and Na fluxes in
Rhodymema palmata (MacRobbie and Dainty, 1958a) and in the brackish water
Nitellopsis (MacRobbie and Dainty, 1958b). In each case a high cytoplasmic
K content and an Na content somewhat lower than that of the medium are to
be noted.

To a degree, salinity tolerances of algae correlate with their intertidal zonation
(Biebl, 1952). However, differences are less marked among intertidal and deeper
water forms. Obviously other factors are also involved in intertidal zonation
and Doty (1946) contends that the duration of exposure, 1.e., submergence or
emergence, is of prime importance.

Doty and Archer (1950) further suggest that bright sun, rain, and freezing
weather are, in that order, the most critical factors in intertidal algal survival.
Kanwisher (1957) has recently reported the effects of freezing and drying on
the respiration of some intertidal algae. The present paper reports a study of
cellular cation concentrations in Porphyra perforata, a red intertidal alga, in
different sea water concentrations and the role of ion transport in response to
osmotic stress is discussed.

METHODS

Tissue volumes. Tissue water was taken as the difference between fresh
weight (after blotting twice with tissue paper) and dry weight (24 hours at

1 This work has been supported by a grant from the National Science Foundation (G-5674).

55

56 RICHARD: W. -EPPEEY AND CHARLES". (CYRUS

105° C.), divided by the fresh weight. Apparent free space (AFS) (Briggs and
Robertson, 1957) was obtained by allowing S*°O, to diffuse into the algal tissues
from sea water for two minutes, blotting twice, then placing the tissues in 10 ml.
distilled water for 10 minutes. Aliquots of the distilled water containing the
labelled sulfate were then counted with gas flow apparatus. The results are ex-
pressed as volume per unit fresh weight. Apparent osmotic volume (AOV)
(Briggs and Robertson, 1957), corresponding roughly to the microscopically esti-
mated cytoplasmic volume, was taken as the difference between tissue water and
AFS for each sea water concentration. :

Ion contents. Sodium and potassium contents were determined by flame
photometry. Results are expressed in units milli-equivalents per kilogram fresh
weight (meq./Kg. FW), or milli-equivalents per liter AOV.

Artificial sea waters. One hundred per cent sea water was prepared according
to the following formula: NaCl, 0.55 M@; MgSO,, 0.027 M; MgCl, 0.027 M;
CaCl,, 0.01 M@; KCl, 0.01 M. The concentrations were adjusted proportionately
for diluted and concentrated sea waters. In some cases diluted sea water was
prepared by adding distilled water to natural sea water. Calcium chloride was
omitted for Ca-free sea water.

100

80 poe

Mion sce fo) gee water

°

PER CENT TISSUE VOLUME

O 100 200

PER CENT SEA WATER

Ficure 1. Tissue water and apparent free space (AFS) of Porphyra perforata as functions
of sea water concentration. Apparent osmotic volume is taken as the difference between the
two curves.

CATION REGULATION IN PORPHYRA Si

RESULTS

Tissue volumes at different sea water concentrations. Tissue water and AFS
are shown as functions of sea water concentration in Figure 1. AOV is taken as
the difference between the two curves. Individual points refer to samples taken

800
150
e@
°o
600
K,
100
K.
¥.
a
K
400 0
°
* a ae
wre K, 50
—~e Ks
oO

200 a be

va 5 ;

°o

“ fe)
0 100 200

PER CENT SEA WATER

Ficure 2. Potassium content (K;), expressed on the basis of apparent osmotic volume
(meq./L. AOV), and potassium accumulation ratios (KiK.) for Porphyra perforata after 24
hours in different sea water concentrations.

58 RICHARD ‘W. EPPLEY ‘AND (CHARLES G./CYRUS

after 24-hour exposure. However, samples taken at two minutes give identical
results. Volume equilibration takes place very rapidly in the monostromatic
algal tissues.

Tissue water follows a linear relationship with concentration. Dry weight
provides an increasing component of tissue weight with increasing concentration.
Values for weight changes at 50 and 150 per cent sea water are similar to those
reported for shrinking and swelling of Porphyra tenera blades (Ogata and Takada,
1955).

100
Total Na
°
eas |
80
2 60
z
uJ
> aan
° ae i
oy ae
o °o
2 Bound Na

@) fee) 200

PER CENT SEA WATER

Ficure 3. Total tissue sodium and bound sodium (the latter determined for killed

tissues) in Porphyra perforata after 24 hours at different sea water concentrations. Units
meq./Kg. FW.

Decrease in AFS in diluted sea water suggests that swelling of the cells
squeezes water from the intercellular spaces. At high concentrations the reverse
may hold, the extracellular material shrinking disproportionately from the cells.
That the intercellular material, a galactan sulfate (Eppley, 1957), is not free to
swell indefinitely is indicated by Miwa’s study of Porphyra tenera (Miwa, 1940).
In this species, and probably also in P. perforata, at least three structural carbo-
hydrates exist. “Outer membranes,’ forming a sandwich about the blades, prob-
ably restrict swelling. Miwa found these polysaccharide membranes to be hydro-
lized only with difficulty. The galactan sulfate lies between and immediately

CATION REGULATION IN PORPHYRA 59

surrounds the cells, each of which has its own discrete wall. None of the three
polysaccharides could be identified by Miwa as cellulose.

AOV, taken as an estimate of cytoplasmic volume, shows a relationship to
concentration which is reverse to that followed by AFS. Swelling is noted in
diluted sea water, a fairly constant volume is maintained between 50 to 100 per
cent sea water, and above this value the volume decreases. The cells do not be-
have as perfect osmometers, nor do they show typical plasmolysis. Both these
characteristics are likely due to the extracellular polysaccharides, each of which
seems to have distinct swelling and shrinking characteristics. We have no evidence

120

NO | 30 %
j~
a
G
Wj |
=

100

30:% = Ca
90
dist. water
80
0) 3 fe) 15 20
HOURS

Figure 4. Weight, expressed as per cent of the initial weight, of Porphyra perforata
tissues in 30 per cent sea water with Ca, 30 per cent sea without Ca, and in distilled water.
Time course.

that the volume control between 50 and 100 per cent sea water is due to water
secretion. Nor does there seem to be any necessity for proposing this as an
explanation.

Ion contents in different sea water concentrations. K contents, expressed on
an AOV basis, and K;/K, accumulation ratios are shown in Figure 2. Cell K
follows a linear function of sea water concentration, although a 35-fold, or greater,
accumulation is evident throughout. Considerable retention of K occurs in diluted
sea waters, resulting in high accumulation ratios, although the actual K content

60 RICHARD ‘W. EPPLEY AND CHARLES (IC "CYRUS

is lower here than in 100-200 per cent sea water. Little K is bound by the
extracellular polysaccharides (1-2 meq./Kg. FW), although large amounts of K
may be adsorbed to structural components in other algae, such as Homosira (Berg-
quist, 1958a). Potassium is the principal cellular cation of Porphyra at all con-
centrations studied.

The distribution of Na is less clear, due to appreciable extracellular adsorption.
Amounts bound by killed tissues are independent of concentration of sea water for

100

CONTENT

K

HOURS

Ficure 5. Potassium contents (meq./Kg. FW) of Porphyra perforata tissues in 100 per
cent sea water, 30 per cent sea water with Ca, and 30 per cent sea water without Ca. Time
course.

a given plant (Fig. 3), but considerable variation in bound Na is evident in differ-
ent plants. Therefore Na contents are not expressed on an AOV basis. Through-
out the concentration range studied, however, Na is excluded by the cells. The
degree of exclusion varies with concentration, but is approximately 10-fold at 100
per cent sea water and increases with increasing sea water concentration.

Total K plus Na thus varies in a roughly linear fashion with salinity, while K
is accumulated and Na partially excluded throughout. A similar situation is re-

CATION REGULATION IN PORPHYRA 61

ported for several halophilic bacteria (Christian and Ingram, 1959). In these
bacteria the freezing point depression of cell sap varies directly with sea water
concentration, with K contributing significantly to cell osmotic pressure, as does Na.

Calcium and survival in diluted sea water. Tissues placed in 30 per cent sea
water without calcium, or in distilled water (Fig. 4) lose weight rapidly after an
initial increase. Addition of calcium chloride, 0.01 M, prevents the weight loss.

Calcium-free 30 per cent sea water also results in rapid loss of K (Fig. 5),
while with Ca the K loss is much depressed. Cell Na, data not shown, increases
slightly without Ca (with the concentration gradient), but remains at a lower and
constant value with Ca.

TABLE]

Potassium accumulation and sodium extrusion in the presence and absence of CaCl: or
SrClo (10 mM /L.). Tissues were first agitated in K-free sea water 20 to 30 hours
to render them low in K and high in Na so that subsequent net transport
could be measured. They were then transferred to sea water plus KCl.

K uptake Na extrusion

Expt. Duration of expt. in hours meq./Kg. FW meq./Ke. FW
1 a5
initial 0 0
—Ca 59 26
+Ca 55 38
2 1:5
initial 0 0
—Ca 46 41
+Ca 60 il
5 19.5
initial 0 0
—Ca —7 13
+Ca 96 47
4 16.5
initial 0 0
—Ca 15 13
+Ca 50 3
+Sr Sik 33

Calcium in net transport of K and Na. To investigate the role of Ca in net
uptake of K and extrusion of Na, tissues were rendered low in K and high in Na
by soaking them in K-free sea water (Eppley, 1958b). They were then placed
in sea water with KCI (10 or 20 meq./L.) and the ion contents of the tissues were
followed in time. Initial uptake of K and extrusion of Na (Table I) is independ-
ent of Ca presence. However, after longer periods net active movements are
reduced, indicating loss of K and gain of Na in the absence of Ca. Strontium
(SrCl,, 0.01 MW) may substitute for Ca in preventing loss of K and gain of Na.

Our results are best explained by assuming that Ca-lack gradually brings about
leakage through the cell membranes, resulting in increasing movements of salts
along their concentration gradients. It seems unlikely that Ca plays any direct

62 RICHARD Wi EPPLEEYVANDIGEHARDES CAGYRUS

role in the operation of ion transport mechanism, 7.e., active transport, but it could
well be required to maintain K:Na selective sites on the membrane, or participate
in labile membrane structure. é

DISCUSSION

Tissue volumes. Because of the lack of a central vacuole and because micro-
scopic estimates of AFS correspond roughly to those measured with S*°O,, we
have assumed that AFS in Porphyra includes the extracellular water but no cyto-
plasmic component, as is suggested for some higher plant tissues (Briggs and
Robertson, 1957). With these assumptions any tissue water not corresponding
to AFS must be cellular. AFS has been measured with sucrose for tissues in
100 per cent and diluted sea waters, with results identical to those obtained with
radiosulfate. Whether the experimental values include all the extracellular water
may be open to question, as well as whether a cytoplasmic component is included.
Interestingly, Bergquist (1958b) reported that KCN increases AFS values in
Homosira. Whether this represents an actual increase in AFS or modification
of anionic binding sites was left an open question.

It is with the above reservations that we have taken AOV as the difference
between tissue water and AFS, and have calculated K concentrations on this basis.

We do not necessarily presume a uniform cytoplasmic distribution of K, al-
though we plan to investigate this point further. Some localization may occur in
vacuomes which are variable in number, but which may comprise a small fraction
of cytoplasmic volume. These are visible with neutral red staining. Mitochondria,
the chromatophores, and local membrane vesiculations, if they occur, could be
sites of variation in cytoplasmic ion concentrations.

While Porphyra cells do not behave as perfect osmometers, our results are con-
sistent with the view that active water movements do not take place. Variations
in AOV with salinity seem to be due to different degrees of shrinking and swelling
of the protoplast and the extracellular, structural polysaccharides.

Ion regulation in concentrated sea water. In corroboration of Biebl’s findings
(Biebl, 1953), we find that Porphyra may survive for. some time in 200 per cent
sea water. Over the studied range, external ion concentrations of the medium
increase equally. Selective precipitation of salts, 1.e., CaSO,, occurs only with
concentration above 300 per cent. Our investigations have not extended to such
concentrations.

Accumulation ratios for K are identical for 100 and 200 per cent sea water.
Cell Na shows some increase with concentration, but is lower than that of the
medium; thus cation selectivity is retained.

In these experiments changes in salinity were abrupt. Coenocytic algae such
as Valoma (Blinks, 1951) do not tolerate such rapid changes. Lack of a large
central vacuole in Porphyra cells may be of importance in this regard. There is
no plasmolysis and thus no mechanical injury to the cell membranes with rapid
salinity change. AFS and AOV adjustments are almost immediate, as indicated
by identical AFS values at two minutes and 24 hours of exposure to concentrated
sea water.

Porphyra perforata grows between the 3- and 3.5-foot tide levels, referred to
San Francisco (Doty, 1946). Here that alga is regularly covered and uncovered

CATION REGULATION IN PORPHYRA 63

by the tide twice daily. The maximum period of exposure is 6-8 hours. Concen-
tration of sea water between the algal blades during this period seems unlikely to
exceed that tolerated in our experiments, even on the hottest days of summer. Thus
we feel that osmotic stress under these conditions is insufficient to result in break-
down of ion transport, resulting in death. Survival in concentrated sea water
encountered in the field is not a problem, in our view.

A more serious problem to the emerged algae on hot days is heating, and
possibly photoxidation. In summer the superficial algal blades are bleached.
Underlying blades (the blades overlap one another) may survive.

Transport and survival in diluted sea water. A 6-8-hour exposure to rain
could result in serious injury. or death, as indicated by our experiments. Lack of
Ca results in rapid loss of K and of weight, and these losses are speeded during
osmotic stress. In 100 per cent sea water, free of Ca, Porphyra may survive up
to 30 hours, compared with only 6-8 hours in 30 per cent Ca-free sea water.

Lack of potassium (Eppley, 1958b) also may influence survival because of the
necessity of K for extrusion of Na. Unpublished results indicate that respiration
is inhibited about 60 per cent in the absence of Na. Thus the presence of Ca, K,
and Na (and probably also Mg) is required for normal operation of cellular proc-
esses, one of which is ion transport.

It is of some interest that even in distilled water Porphyra tissues adsorb about
10 mM/Kg. FW of Ca. This amount, probably associated with the extracellular
galactan sulfate, is not apparently available to the critical sites involved in main-
taining membrane selectivity. The implication of this finding may be of some
importance in clarifying the role of trace elements passively adsorbed to extra-
cellular polysaccharides in algae. Thus trace element “accumulation” (Black and
Mitchell, 1952), in which the extent of adsorption was not determined, might be
re-interpreted in the light of this result, with respect to the survival value of such
“accumulation.” The likelihood remains, however, that in the presence of exchange-
able cations, adsorbed trace elements could be made available to the cell surfaces
for real accumulation.

Growth experiments which might clarify this problem have not been possible,
due to our failure to obtain growth of Porphyra in the laboratory. The recent re-
sults of Kanazawa and Kashiwada (1959) on the culture of Porphyra tenera cast
some hope on our aspirations of doing so, however. |

A heavy rain, by washing away the sea water film normally present between
and around the blades even during emergence, would almost certainly abolish
membrane selectivity, decrease respiration, induce loss of cellular cations, and
result in high mortality if the blades were exposed long enough. In southern
California Porphyra perforata largely disappears in the winter, and rain may be
involved in this mortality. Other factors are also involved, however.

At the site studied, approximately two miles north of the Los Angeles-Ventura
County boundary in California, mass movements of sand were observed during
the winter of 1958-1959. The old Porphyra bed was entirely covered and a new
bed appeared about 50 yards south in the following spring. A smaller sand move-
ment partially covered the new bed in July, 1959. While the extracellular poly-
saccharides may act as cushions against moderate wave impact and sand scouring,
the plants may be torn loose or shredded in a heavy surf or be buried by sand

64 RICHARD W. EPPLEY AND’ CHARLES CYRUS

movements. Such physical processes may be important in survival and distribu-
tion as well as physiological tolerances.

SUMMARY

1. Over a wide range of salinity (10 to 200 per cent sea water) cells of
Porphyra perforata accumulate K and partially exclude Na. Apparent osmotic
volume is nearly constant between 50 and 100 per cent sea water. This imperfect
volume control is thought to be due to differential shrinking and swelling of struc-
tural polysaccharides and not to active water secretion.

2. Survival of Porphyra in diluted and concentrated sea water is discussed
with respect to ion transport. ain is potentially a more serious threat to sur-
vival, due in part to breakdown of 1on transport, than is concentration of sea water.

Calcium seems especially important in maintaining membrane selectivity toward
K and Na.

LITERAGURE iCrDE)

Bercguist, P. L., 1958a. Evidence for separate mechanisms of sodium and potassium regu-
lation in Homosira banksui. Physiol. Plant., 11: 760-770.

Bercguist, P. L., 1958b. Effect of potassium cyanide on apparent free space in a brown alga.
Nature, 181: 1270.

Bresi, R., 1952. Ecological and non-environmental constitutional resistance of the protoplasm
of marine algae. J. Mar. Biol. Assoc., 31: 307-15.

BresL, R., 1953. Lichttransmissionsanderungen an Meeresalgen, im besonderen an Porphyra
umbilicalis {. laciniata. Osterreich. Bot. Zeitschr., 100: 179-202. |

Brest, R., 1956. Zellphysiologisch-okologische Untersuchungen an Enteromorpha clathrata
(Roth) Greville. Ber. Bot. Gesell., 69: 75-86.

Brest, R., 1958. Temperatur- und osmotische Resistenz von Meeresalgen der bretonischen
Kuste. Protoplasma, 50: 217-42.

Brack, W. A. P., anp R. L. MitcHeti, 1952. Trace elements in common brown algae and
in sea water. J. Mar. Biol. Assoc., 30: 575-84.

Bunks, L. R., 1951. Physiology and biochemistry of algae. Jn: Manual of Phycology (G.

. M. Smith, ed.), Chronica Botanica Co., Waltham, Mass., pp. 263-91.

Briccs, G. E., Aanp R. N. Rospertson, 1957. Apparent free space. Ann. Rev. Plant Physiol.,
8: 11-30.

Buttock, T. H., 1958. Homeostatic mechanisms in marine organisms. Jn: Perspectives in
Marine Biology (A. A. Buzzati-Traverso, ed.), Univ. Calif. Press, Berkeley, Calit.,
pp. 199-210.

CurisTIAN, J. H. B., anp M. INcRAm, 1959. The freezing points of bacterial cells in relation
to halophilism. J. Gen. Microbiol., 20: 27-31.

Doty, M. S., 1946. Critical tide factors that are correlated with the vertical distribution of
marine algae and other organisms along the Pacific Coast. FEcol., 27: 315-28.

Doty, M. S., ano J. G. ArcHeEr, 1950. An experimental test of the tide factor hypothesis.
Amer. J. Bot., 37: 458-64.

Eppiey, R. W., 1957. Cation binding and exchange by killed red algal tissues. Esp. Cell Res.,
13: 173-74.

Epptey, R. W., 1958a. Sodium exclusion and potassium retention by the red marine alga,
Porphyra perforata. J. Gen. Physiol., 41: 901-11.

Epprey, R. W., 1958b. Potassium-dependent sodium extrusion by cells of Porphyra perforata,
a red marine alga. J. Gen. Physiol., 42: 281-88.

Kanazawa, A., AND K. KasHrtwapA, 1959. Studies on the metabolism of algae. I. The
effects of amino acids and vitamins on the growth of Porphyra tenera. Mem. Fac.
Fish. Kagoshima Univ., 7: 187-91.

KANwISHER, J., 1957. Freezing and drying in intertidal algae. Biol. Bull., 113: 275-85.

CATION REGULATION IN PORPHYRA 65

MacRospir, E. A. C., ann J. Darnty, 1958a. Sodium and potassium distribution and transport
in the seaweed Rhodymenia palmata (L.) Grev. Physiol. Plant., 11: 782-801.
MacRossisz, E. A. C., anp J. Datnty, 1958b. Ion transport in Nitellopsis obtusa. J. Gen.
Physiol., 42: 335-53.

Miwa, T., 1940. Biochemische Studien tuber die Zellmembran von Braun- und Rotalgen.
Japan. J. Bot., 11: 41-127.

OecaTa, E., AND H. Taxapa, 1955. Elongation and shrinkage in the thallus of Porphyra tenera
and Ulva pertusa caused by osmotic changes. Osaka Univ. Polytech. J., Ser. D, 6:
29-40.

Scott, G. T., Anp H. R. Haywarp, 1953. Metabolic factors influencing the sodium and
potassium distribution in Ulva lactuca. J. Gen. Physiol., 36: 659-71.

Scott, G. T., anp H. R. Haywarp, 1954. Evidence for the presence of separate mechanisms
regulating potassium and sodium distribution in Ulva lactuca. J. Gen. Physiol., 37:
601-20.

Scott, G. T., anp H. R. Haywarp, 1955. Sodium and potassium regulation in Ulva lactuca
and Valoma macrophysa. In: Electrolytes in Biological Systems (A. M. Shanes, ed.),

Amer. Physiol. Soc., Washington, D. C., pp. 35-64.

THE EFFECTS OF THYROXIN: AND GROW VE Ein ois
ON LIVER POLY BEGIiiGE

I. I. GESCHWIND, M. ALFERT AND C. SCHOOLEY 2

Hormone Research Laboratory, and Department of Zoology and its Cancer Research Genetics
Laboratory, University of California, Berkeley 4, California

Polyploidy in the cells of the hepatic parenchyma is known to be influenced by
changes in the endocrine state produced by hypophysectomy and growth hormone
administration (Di Stefano and Diermeier, 1956; Leuchtenberger, Helweg-Larsen
and Murmanis, 1954). We have confirmed the effects of both the former
(Geschwind, Alfert and Schooley, 1958) and the latter (see below) on liver poly-
ploidy, but have shown that although the hormonal environment may normally
play a role in the development of polyploidy, it is not indispensable to that devel-
opment (Geschwind, Alfert and Schooley, 1958).

In the course of experiments designed to investigate the hormonal incre nec-
essary to stimulate growth in the genetic dwarf mouse (Cole, Geschwind, and
Bern, unpublished experiments), it was found that highly purified bovine growth
hormone preparations were far less effective in stimulating body growth than were
the crude preparations used for this purpose by Fgnss-Bech (1947). Since con-
tamination of the crude preparations with the thyrotropic hormone was strongly
suspected, it was decided to investigate the effects of thyroxin (more readily avail-
able in pure form than is the thyrotropic hormone), alone and in combination with
growth hormone, on body growth and liver polyploidy. The results of the experi-
ments on liver polyploidy in these animals, together with the results of a similar
series of experiments in the hypophysectomized rat, are reported below.

MATERIALS AND METHODS

Male and female dwarf mice, 40 to 65 days of age, were apportioned into four
groups, one of which served as a control group. The animals in each of the other
groups were injected with either 50 micrograms twice weekly of dl-thyroxin (the
dose suggested by Nielsen, 1953), 25 micrograms daily of a bovine growth hormone
preparation or both hormones at the aforementioned dose levels. Injections were
continued for 21 days.° |

Male rats of the Long-Evans strain were hypophysectomized at 28 days of age.
One week later the animals were apportioned into four groups, one of which served
as a control group. The animals in each of the other groups were injected with
either 2 micrograms daily of |-thyroxin, 25 micrograms daily of the bovine growth
hormone preparation, or both hormones at these respective dose levels. Injections
were continued for 22 days.

‘Supported by University of California Cancer Funds.

2 With technical assistance from Charles Jordan and Norma O. Goldstein.

3 We gratefully acknowledge gifts of materials from the following: Dr. C. H. Li for the
hormones and Beth Cole for the livers of the dwarf mice.

66

HORMONE EFFECTS ON LIVER POLYPLOIDY 67

All animals were sacrificed on the day after the last injection, and the hepatic
left lateral lobes were removed from the anesthetized animals, sliced freehand,
and fixed in acetic-alcohol. The procedure followed for the preparation of sec-
tions for counting and the actual counting procedures have been previously de-
scribed (Alfert and Geschwind, 1958). Approximately 1000 parenchymal cells
were scored in the sections obtained from each animal. The mean percentage
distributions of the various types of hepatic parenchymal cells in each of the
experimental groups in both series, along with the mean initial and terminal body
weights of the animals in each group are recorded in Tables I and II.

RESULTS AND DISCUSSION

In the genetic dwarf mouse, the administration of either growth hormone or
thyroxin results in a decrease in the number of hepatic mononucleate diploid cells
and an increase in tetraploid and binucleate cells (Table I). Of the two hormones,
thyroxin, at the dose administered, appears to be more effective in producing these
changes. By far the greatest effect, however, was observed in those animals re-
ceiving the combined hormone treatment. Such animals also showed the greatest
weight gain during the period of injections.

It was difficult to determine unequivocally the manner by which thyroxin
brought about its observed effect, since at least three different interpretations of
the data could be advanced. One is that thyroxin enhanced polyploidization
directly, either by acting as a specific mitogenic agent or by promoting the need
for accelerated liver function. A second interpretation might be that the changes
in the liver were simply the result of the administration of a large dose of a com-
pound that had to be conjugated and metabolized by the liver, or that was toxic

TABLE [|

The distribution of hepatic parenchymal cell types in control and experimental dwarf mice

Body weight Mean percentage distribution of cell types*
Group
Initial | Terminal 2n 2n bi 4n 4n bi 8n
g. g.
Control** 6.1 v4 81.4 Lit 3.9 0.1 0.1
(78.2-83.1)f) (10.5-15.3) | ( 1.7— 5.2) | (0.0-0.2)| (0.0-0.3)
Thyroxin 6.5 10.0 53.4 18.1 24.3 3.0 0.1
(49.0-60.4) | (12.2-23.7) | (21.9-26.1) | (1.4-4.8)| (0.0-0.2)
Growth hormone tak 10.3 69.0 20.9 8.1 0.9 0.0
(65.4-70.9) | (17.0—23.2) | ( 4.1-10.3) (0.6-1.1)
Thyroxin + 6.2 2A 31.6 8.6 52.8 4.2 1.6
growth hormone (23.4-37.9) | ( 5.6-10.2) | (45.7-61.7) | (3.3-5.7)] (1.0-2.3)

*2n = diploid; 2n bi = diploid binucleate; 4n = tetraploid; 4n bi = tetraploid binucleate ;
8n = octaploid. (In all experiments approximately 1% of the cells could not be classified.)
** Three animals per group.
Tt Range of individual results.

68 I. I) GESCHWIND, M,; ALFERT AND) C. SCHOOLEY

TABLE II

The distribution of hepatic parenchymal cell types in control and experimental
hypophysectomized rats

Body weight Mean percentage distribution of cell types*
Group
Initial | Terminal 2n 2n bi 4n 4n bi 8n
g g
Control** 74 82 70.1 24.0 4.8 0.2 0.0
(67.3-74.5)t| (20.6-26.5) | ( 4.1-5.8) | (0.1-0.4)
Thyroxin (Me 93 Sas) 24.0 22.0 0.7 0.1
(46.6-57.8 | (18.1-26.9) | (19.4-24.6) | (0.1-1.1)| (0.0-0.2)
Growth hormone 74. 103 53.2 24.0 20.6 0.8 0.1
(50.7-58.1) | (19.9-29.6) | (17.3-23.6) | (0.1-1.2)} (0.0-0.4)
Thyroxin + 12 123 Dal 9.5 61.4 3.9 1.9
growth hormone (19.8-26.3) | ( 6.4-14.7) | (56.0-68.1) | (2.4-6.2)} (1.6—2.1)

*, ** | Symbols have the same meaning as in Table I.

to the parenchymal cells in the dosage employed. Finally, it was possible that
thyroxin, in its capacity as a hormone, promoted repair of pituitary acidophiles
which are normally absent in the dwarf mouse (Ortman, 1956), and by so doing
stimulated growth hormone secretion. There is good evidence that thyroxin re-
pairs pituitary acidophile cytology in the thyroidectomized rat (Koneff, Scow,
Simpson, Li and Evans, 1949; Contopoulos, Simpson and Koneff, 1958), and
concomitantly restores growth activity (Eartley and Leblond, 1954; Contopoulos
et al., 1958). Such an effect on the acidophiles could explain the findings of
Carriére (1955) that in the thyroidectomized rat, hepatic nuclei with diameters
of only 6.0 uw are found after a considerable post-operative interval, whereas in
normal controls or in thyroidectomized animals treated with growth hormone,
thyroxin, or both, nuclei with diameters of 6.5 and 8.0 » are seen. The probability
that growth hormone secretion is depressed after thyroidectomy could also explain
our own unpublished findings that in animals which had been thyroidectomized
at 24 days of age, the distribution of hepatic parenchymal cell types found at 114
days was that characteristic of 40- to 45-day-old animals. The change, in the
normal animal, in the distribution of cell types with age has been previously
reported (Alfert and Geschwind, 1958).

In order to eliminate the possibility that thyroxin was acting to enhance poly-
ploidization by repairing pituitary function, experiments were conducted in hypo-
physectomized rats. Such animals have a further advantage of being far more
sensitive to thyroxin than is the dwarf mouse, permitting a much smaller dose of
thyroxin to be employed (2 micrograms daily to a rat weighing 75 grams versus
50 micrograms twice weekly to a mouse weighing 6 grams). Thus, any effects
due to conjugation and metabolism by the liver, or to toxicity, are minimized.

The results of these experiments (Table II) reveal that the effect of thyroxin
on the hypophysectomized rat is comparable, at the dose levels chosen, to that

HORMONE sR ERECTS ON LIVER, POLY PLOIDY 69

obtained with growth hormone. In both cases a 25% decrease in the number of
mononucleate diploid cells has occurred, along with a 5-fold increase in the
number of tetraploid cells. As in the case of the dwarf mouse, the greatest effect,
however, was found in those animals receiving both hormones, for in the livers
of such animals the total number of diploid cells was only one-third that found
in the control animals, while the numbers of both tetraploid and octaploid cells
increased markedly. The synergistic effect of the two hormones administered
together is also reflected in the body weight increments of the treated animals.

These experiments therefore indicate that thyroxin (and presumably thyro-
tropic hormone) as well as growth hormone contributes to the hormonal environ-
ment which affects the development of hepatic polyploidy. It should also be noted
that thyroxin alone can be at least as efficient as growth hormone in promoting the
progression of hepatic polyploidy; thus, as we also have demonstrated by the
results of a previous experiment (cochaid Alfert and Schooley, 1958), the
suggestion of Leuchtenberger et al. (1954), concerning the supposed essential
role of growth hormone in somatic polyploidy, cannot be maintained.

SUMMARY

1. The effect of thyroxin and of growth hormone on liver polyploidy has been
investigated in the dwarf mouse and in the hypophysectomized rat.

2. Either hormone, acting alone, stimulated polyploidization in these experi-
mental animals; the combination of both hormones synergized to produce an
hepatic cell picture dominated by higher polyploid classes.

ERE RATURE Crrep

Arert, M., ann I. I. GescHwinp, 1958. The development of polysomaty in rat liver. Exp.
Cell Res., 15: 230-232.

CARRIERE, R., 1955. Influence of growth hormone and thyroxine on nuclear size in rat liver.
Anat. Rec., 121: 273.

Contoroutos, A. N., M. E. Stmpson anp A. A. Konerr, 1958. Pituitary function in the
thyroidectomized rat. Endocrinology, 63: 642-653.

Di Sterano, H. S., anp H. F. Drermeter, 1956. Effects of hypophysectomy and growth hor-
mone on ploidy distribution and mitotic activity of rat liver. Proc. Soc. Exp. Biol.
Med., 92: 590-594.

Earttey, H., anp C. P. Lesronp, 1954. Identification of the effects of thyroxine mediated by
the ee ae Lory Son ee 54: 249-271.

Fg¢nss-Becnu, P., 1947. <A study on growth hormone of anterior pituitary lobe. Ejnar Munks-
eaard, Copethagen, 1947.

GescHwinp, I. I, M. Atrert anp C. ScuHoorey, 1958. Liver regeneration and hepatic
polyploidy in the hypophysectomized rat. Exp. Cell Res., 15: 232-235.

Konerr, A. A., R. O. Scow, M. E. Simpson, C. H. Li anp H. M. Evans, 1949. Response by
the rat thyro-parathyroidectomized at birth to growth hormone and to thyroxin given
separately or in combination. II. Histological changes in the pituitary. Anat. Rec.,
104: 465-475.

LEUCHTENBERGER, C., H. Fr. Hetwec-LarsEN Aanp L. Murmanis, 1954. Relationship be-
tween hereditary pituitary dwarfism and formation of multiple desoxyribose nucleic
acid (DNA) classes in mice. Lab. Invest., 3: 245-260.

NIELSEN, E. L., 1953. Studies on hereditary dwarfism in mice. XIV. Effect of thyroxin
and growth hormone on growth. Acta Path. Microbiol. Scand., 32: 316.

OrtMAN, R., 1956. A study of some cytochemical reactions and of the hormone content of
the adenohypohysis in normal and in genetic dwarf mice. J. Morph., 99: 417-432.

SURFACE ANTIGEN: DYNAMICS IN THE SEM LE M@Omin
DICTYOSTELIUM, DISCOIDEUM

JAMES H. GREGG

Department of Biology, University of Florida, Gainesville, Florida, and Marine Biological
Laboratory, Woods Hole, Mass.

The mechanisms by which cells in a morphogenetic system effect coordinated
movements and specific associations constitute basic developmental problems that
have been the subject of many investigations. That such mechanisms possibly
involve forces similar to those existing between antigens and antibodies was recog-
nized and elaborated by Tyler (1947, 1957), Weiss (1947), and Spiegel (1954a,
1954b). A recent comprehensive review of the problems involved in such morpho-
genetic processes as cell adhesion and cell migration has been presented by De Haan
(1958) and need not be recapitulated here.

Morphogenetic expression in the slime mold, Dictyostelium discoideum, re-
quires oriented movements of cells, a process which involves changes in shapes of
the cells. The stimulus for the changes in shape may be an external one (via
acrasin) in which large numbers of cells are affected from a common direction
simultaneously (Bonner, 1947, 1950). Following the aggregation process the
cells are organized into a pseudoplasmodium in which the cell surfaces are in
intimate association with each other. Throughout morphogenetic movements the
cells maintain contact with each other. Acquisition of the ability to adhere to one
another implies a change in the nature of the surface of the cells. Consistent with
this is the evidence obtained by Gregg (1956) that new antigens appear at the
onset of aggregation. |

It is the purpose of this study to investigate changes in the surface antigens of
the cells from the beginning aggregation stage to the formation of the mature
spores. Such a study might be expected to reveal not only the pattern of antigens
during morphogenesis but something of the possible role of surface antigens in
maintenance of the integrity of the pseudoplasmodium and in regulation of the
course of cellular movement.

METHODS AND MATERIALS

The slime molds were cultured according to the general methods employed
by Bonner (1947) with the exception that the bacterial associate was Aerobacter
aerogenes rather than Escherichia coli. The cultures were maintained at 22° C.
during the growth phase. During the period of aggregation they were transferred
to a 17° C. incubator, the lower temperature of which extended the migration and
culmination phases.

1 This investigation was supported in part by a research grant E-1452 from the National
Institutes of Health, Public Health Service.

70

SURFACE ANTIGEN DYNAMICS (All

Preparation of injection-antigens

Migrating pseudoplasmodia were collected by cutting discs of nutrient agar
bearing aggregates on the verge of migration. These discs were transferred to
similar sized spaces in plates on non-nutrient agar (Gregg and Bronsweig, 1956).
After the migrating pseudoplasmodia had accumulated on the non-nutrient surface
they were collected individually with a hair loop and transferred into 1.0% NaCl.

The mature spores were collected by sweeping the tops of the mature sorocarps
with a 3-mm. glass rod. The spores adhered to the glass rod and were transferred
into 1.0% NaCl.

An antigen was also prepared by washing mature spores in 0.5% NaCl for
10-15 minutes. The spores were then centrifuged and the supernatant carefully
removed to serve as an antigen.

These collecting procedures enabled migrating pseudoplasmodia, mature spore,
and mature spore surface antigens to be obtained essentially free of Aerobacter.
The antigens were stored in 1.0-ml. aliquots in a deep freezer at minus 15° C.

Each of the three types of antigens contained 0.2 mg. N/ml. Two rabbits
each were injected with the three types of antigens. The injections were given
every other day until a total of nine injections per rabbit had been given (0.5 ml.
antigen /injection ).

Preparation of antisera and normal (pre-injection) sera

The rabbits were bled by heart puncture five days following the final injection.
The blood was allowed to stand for approximately two hours at room temperature
before the clot was ringed, then placed in the refrigerator overnight before cen-
trifugation. Complement was destroyed by heating the antisera at 56° C. for
30 minutes. The antisera were divided into 1.0-ml. aliquots and placed in the
deep freezer at minus 15° C.

The normal (pre-injection) sera were obtained and prepared in the same
manner as described for the antisera.

Preparation of absorbed antisera

The amoebae or spores were harvested from the culture plates by the methods
described under the sections concerning the preparation of injection-antigens and
test-antigens. However, in the absorption procedure all of the supernatant was
removed from the packed cells. In a typical absorption, packed cells were mixed
with the particular antiserum in the ratio of one volume of cells to one volume of
antiserum. The preparation was allowed to absorb at room temperature for two
hours followed by 22-24 hours at 10° C. The absorbed serum was then centri-
fuged in an electric micro-centrifuge (Microchemical Specialties Co.) at 9500 rpm
for 15 minutes. The serum was then carefully removed from the packed cells
and stored at minus 15° C. until desired.

Preparation of test-antigens for agglutination

These antigens were prepared by alternately washing and centrifuging (125 g)
aggregating amoebae or spores twice in 0.5% NaCl. The resulting sedimented

72 JAMES H. GREGG

cells were diluted with about one volume of 0.5% NaCl (approximately 2.5 x 105
cells/ml.) for use in agglutination tests. The cells were placed in agglutination
chambers in 10.0-pl. volumes (Gregg and Trygstad, 1958). The antiserum or
normal (pre-injection) serum was mixed with the cells in 10.0-yl. volumes with

Ficure la. D. discoideum amoebae + spore antiserum (magnification X 36). lb. D. dis-
coideum amoebae + spore normal serum (magnification X 36). lc. D. discoideum spores
+ spore antiserum (magnification <X 100). 1d. D. discoideum spores + spore normal serum
(magnification x 100).

a small glass rod. The presence or absence of antibodies in the serum was de-
tected by agglutination of the amoebae or spores to various degrees (Fig. la, b, c, d).

All antisera and normal (pre-injection) sera were tested on amoebae or spores
independently. However, the results obtained from antisera produced in response

TABLE I[

Antiserum titer obtained by agglutination of cells

Type of antiserum Type of cell agglutinated Highest dilution of antiserum
MA Amoebae 1:80
SA Amoebae 1:80
SSA Amoebae 1220

to the same type antigen or normal (pre-injection) sera have been pooled. The
titers of the various sera were obtained by determining the least concentration of
antiserum diluted with 1.0% NaCl, which would produce detectable agglutination
of amoebae or spores (Table I).

SURFACE ANTIGEN DYNAMICS 73

Determination of the amounts of antigenic material extracted from the surfaces
of the cells

For reasons which will become apparent it was necessary to determine quanti-
tatively the amounts of antigenic material which may be extracted from the sur-
faces of the cells composing migrating pseudoplasmodia and the mature spore
masses. These determinations were made in two ways as follows:

A. 1. Ten to fifty migrating pseudoplasmodia (non-nutrient agar migrants)
or mature spore masses placed in 20 pl. of 0.5% NaCl.
2. Cell surface materials extracted by gentle stirring for 10-15 minutes.
3. Cells centrifuged 5 minutes at 4600 rpm in the micro-centrifuge.
4. Washed cells and surface wash analyzed separately for total N (Bruel

et al., 1946; Gregg, 1950) (Table IV).

B. 1. One hundred to five hundred migrating pseudoplasmodia or mature spore
masses were placed in 100-500 pl. 1.0% NaCl.
2. Cell surface materials extracted by gentle stirring for 10-15 minutes.
3. Cells centrifuged 15 minutes at 9500 rpm in the micro-centrifuge.
4. The surface wash was removed and saved for analysis.
5. The washed cells were analyzed for total N.

6. The volume of each surface wash from each type of cell was adjusted such
pl. surface wash
pg. N of washed cells
tion of surface wash in volumes proportional to the number of cells (total N)

from which it was extracted.

7. The surface washes were then subjected to conventional procedures to
determine optimum proportions with various antisera in precipitin tests.

8. Fifty pl. of each surface wash and the optimum amount of each antiserum,
determined in step 7, were mixed in 3 mm. X 40 mm. test tubes and allowed to
incubate for 1—2 hours at room temperature. After incubation in a refrigerator
overnight the precipitates were centrifuged at 9500 rpm for 15 minutes in the
micro-centrifuge. The precipitates obtained were washed once with 1.0% NaCl
and analyzed for total N (Table V).

that =unity. This dilution procedure resulted in the produc-

RESULTS
Migrating pseudoplasmodia-antiserum (MA)

The MA effected maximal agglutination of both amoebae and spores (Table IT).
It should be noted here, however, that spores do not exhibit the massive type of
agglutination obtained with amoebae. The maximal agglutination in spores
amounts to an easily detectable regular pattern of small clumps of spores, but
never the large masses of drastically clumped cells as observed when agglutinating
amoebae.

Migrating pseudoplasmodia-antiserum absorbed with amoebae (MAa)

The MAa did not agglutinate amoebae or spores (Table III).

74 JAMES H. GREGG

Migrating pseudoplasmodia-antiserum absorbed with spores (MAs)
The MAs did not agglutinate amoebae or spores (Table III).

Migrating pseudoplasmodia normal (pre-injection) serum (MN)

The MN used as controls in all agglutination tests along with MA, MAa, and
MAs, did not effect agglutination of either amoebae or spores (Tables II, III).

Mature spore-antiserum (SA)

The SA agglutinated both amoebae and spores to the degree designated as
maximum tor each) type ot cell (Mable Il) (rice flac):

TABLE II

Degree of agglutination effected by various antisera and normal ( pre-injection)
sera on amoebae and spores
+ = maximum agglutination; - = minimal agglutination; and — = no agglutination.
All sera were diluted to one-half of full strength as a result of the addition of amoebae. The numbers
opposite the degree of agglutination values refer to the number of times a particular pattern was observed.

Amoebae Spores
Degree of agglutination Degree of agglutination*
Type of sera

atc = = al 35 =

Migrating 49 26

Antisera Spore 40 24

Spore surface Dil 13
Migrating 48 24
Normal sera Spore 39 22
Spore surface 19. ilps)

*In this table and Table III, + represents the maximal degree to which spores may be
agglutinated.
Mature spore-antiserum absorbed with amoebae (SAa)

The SAa failed to agglutinate amoebae, but effected maximal Sie
of spores (Table IIT).

Mature spore-antiserum absorbed with spores (SAs)

The SAs failed to agglutinate either amoebae or spores (Table III).

Mature spore normal (pre-injection) serum (SN)

The SN used as controls in all agglutination tests along with SA, SAa, and SAs
did not effect agglutination of either amoebae or spores (Tables II, III) (Fig.
Deady)

Mature spore surface antiserum (SSA)

The SSA effected maximal agglutination of both amoebae and spores (Table IT).

SURFACE ANTIGEN DYNAMICS 75

TABLE III

Degree of agglutination effected by various absorbed antisera and normal (pre-injection)
sera on amoebae and spores
+ = maximum agglutination; + = minimal agglutination; and — = no agglutination.
All sera were diluted to one-half of full strength as a result of the addition of amoebae. The numbers
opposite the degree of agglutination values refer to the number of times a particular pattern was observed.

Amoebae Spores
Degree of agglutination Degree of agglutination*
Type of sera

I sb = ata == =
Antisera abs. Migrating 26 10

with amoebae Spore 24 in

Spore surface a 3)
Antisera abs. Migrating 9 4
with spores Spore 14 4
Spore surface fi | 3
Normal sera Migrating 34 | 12
Spore 38 12
Spore surface 10 4

* See footnote to Table II.
TaBLeE IV

Comparison of total N in surface antigens of migrating pseudoplasmodia amoebae and
mature sorocarp spores

4 Ratio of antigen N
; face antigen N

Cell t Na. oF : pug. whole Lg. surface beg. SUT Spore

a vital fia pees cell N antigen N ug. whole cell N NeRWaE
Spores 6 133.3 31.0 0.23

p

2's area 15
Amoebae 4 53.9 8.0 ens
TABLE V

Comparison of total N precipitated from surface antigen preparations of migrating
pseudoplasmodia amoebae and mature sorocarp spores by antibodies

Surface antigens
obtained from Ratio of precipitate N
No. of exp. Spore
Amoebae
Amoebae Spores
ug. N precipitated by spore surface anti-
serum minus controls (normal serum) 3 6.8 30.0 4.4

ug. N precipitated by migrating pseudo-
plasmodia antiserum minus controls
(normal serum) 3 3.0 8.0 24

76 JAMES H. GREGG

Mature spore surface antiserum absorbed with amoebae (SSAa)

The SSAa failed to agglutinate amoebae but effected maximum agglutination
of spores (Table IIT).

Mature spore surface antiserum absorbed with spores (SSAs)

The SSAs failed to agglutinate either amoebae or spores (Table III).

Mature spore surface normal (pre-injection) serum (SSN)

The SSN did not effect agglutination of either amoebae or spores (Tables
i eas ha

The amounts of surface material removed from cells composing migrating pseudo-
plasmodia and mature spores

It was found that in terms of total N of the surface washes, 1.5 times more
material was washed from the cells composing the spore masses than from the
amoebae composing migrating pseudoplasmodia (Table IV). It was assumed
that the N content of the surface washes would serve as an indicator of the amount
of antigenic or surface material which had been removed from the cells.

When surface washes of the two types of cells were precipitated with antisera
the precipitates obtained from spore surface wash proved to be 2.7-4.4 times greater
than those from amoebae surface wash (Table V). This indicates that more anti-
genic material was removed from the cells composing spore masses than from
those composing migrating pseudoplasmodia.

DISCUSSION

Antisera produced in rabbits to antigens composed of cells and washings of
cells from various stages of the slime mold, D. discoideum, effect agglutination of
amoebae and spores. Analysis of the pattern of agglutination effected by the
various types of sera on amoebae and spores demonstrated that qualitatively identical
surface antigens are present during the transition from the beginning aggregate
to the mature spore. However, absorption of various antisera by amoebae and
spores has disclosed the presence of a surface antigen(s) in the spores which is
(are) not common to early aggregates or migrating pseudoplasmodia. It is pos-
sible that the new antigen is composed of the polysaccharide with which spores
become coated during the culmination process (Raper and Fennell, 1952).

An examination of Figure 1 reveals that the maximal degree to which amoebae
may be agglutinated differs from that degree to which spore cells may be agglu-
tinated. It may be seen that spores agglutinate poorly even with mature spore
antiserum as compared with the agglutination of amoebae by the same antiserum.

Shaffer (1958) has demonstrated that D. discoideum amoebae become strongly
adhesive at the onset of aggregation. Furthermore, it has been observed during
the course of the present work that the amoebae composing a migrating pseudo-
plasmodium do not spread readily upon a glass rod. The failure to spread has
been interpreted as being due to the adherence of the cells to each other. However,
the cells composing a spore mass tend to spread readily when tested in the same
fashion, suggesting that the adhesive properties of the spores are relatively poor.

SURFACE ANTIGEN DYNAMICS 77

In seeking an explanation for the particular serological and adhesive properties
exhibited by the amoebae and spore cells it was found that surface materials ob-
tained by washing cells in 0.5% NaCl were antigenic (ring and precipitin tests)
with MA and SA. Such preparations were subjected to two types of analyses to
compare quantitatively the amounts of material which had been extracted from
the amoebae and spore cell surfaces. These determinations revealed that greater
amounts of antigenic material were found in spore wash as compared to amoebae
wash (Tables IV, V). To account for the difference in the amount of antigenic
material recovered from surface washings of amoebae and spores it is suggested that
an increase occurs in the amount of loosely bound surface antigen on the spores.
Another possibility would include an elimination of the antigens from the pre-stalk
and pre-spore surfaces as a consequence of their role in morphogenetic movements
during the culmination process. Antigens ejected from the cell surfaces would
be contained in the spore extracellular spaces. A mechanism by which surface
antigens are discharged and renewed at the surfaces of the cells would result in
the reversibility of cell adhesiveness. Such reversibility is conceivably necessary
if cells in contact with each other are to retain their motility.

Upon injecting spore washings into rabbits, antiserum was produced which
was capable of agglutinating both amoebae and spores (Table II). Such anti-
serum (SSA) when subjected to absorption procedures with amoebae or spores
reacted with test cells in the same way as the antiserum (SA) produced by injecting
whole spores into rabbits (Tables II, III).

The fact that antiserum obtained from whole spore injection cannot be dis-
tinguished from spore surface wash antiserum suggests that the same antigens
were involved in the production of each antiserum. The ease with which spore
surface wash antigen may be removed implies that it is either loosely bound to the
cell surfaces or was discharged from the cells during morphogenesis. Therefore,
it seems plausible to attribute the poor agglutination of the spores to the inability
of the antigens, with which the antibodies combine, to remain attached to the cell
surfaces. Consequently, a reduction in the number of antigenic sites involved in
effecting agglutination of cells in the presence of antisera would occur. Few anti-
genic sites on the spore surfaces would probably suffice for the production of
antisera in rabbits to those particular antigens, which accounts for the agglutinating
ability of the spore antiserum (SA) on amoebae or spores. Another possibility
exists that the poor agglutination of the spores might be attributed to the relative
inaccessibility of antigenic sites on spore surfaces. Such an interpretation, for
the failure of ox red cells to agglutinate in the presence of certain antisera, was
advanced by Coombs et al. (1951). The existence of this type of cell surface
configuration of the spores could conceivably be tested by Coombs eft al.’s method.

The poor agglutinating capacity of spores is paralleled by weak adhesiveness
between the spores. It is interesting to point out that relatively large amounts
of antigen appear in washings of spore cell surfaces. These facts suggest that the
particular agglutinating and adhesive properties of the spores are related to the
appearance of antigen in spore washings.

The author is indebted to Professor Albert Tyler, California Institute of Tech-
nology, for his critical review of this manuscript, and to Mr. Thomas Moore, Uni-

78 JAMES H. GREGG

versity of Florida College of Medicine, for his technical assistance during the course
of this investigation.

SUMMARY

1. Methods are described whereby antigenic material of the slime mold, D.
discoideum, effective in antibody production in rabbits, is prepared.

2. Antisera produced in response to migrating pseudoplasmodia, mature spores
and mature spore surface antigens effected maximal agglutination of amoebae.
Such antisera effected only minimal agglutination of spores.

3. Antisera of the above types absorbed by amoebae and spores in all com-
binations revealed the presence of a surface antigen(s) on the spores which is (are)
not common to amoebae from early aggregates or migrating pseudoplasmodia.

4. It has been suggested that a relationship exists between the adhesive prop-
erties of the amoebae and spores and their agglutinating properties in the presence
of antisera.

LIPERATURE, CIV ED

Bonner, J. T., 1947. Evidence for the formation of cell aggregates by chemotaxis in the
development of the slime mold Dictyostelium discoideum. J. Exp. Zool., 106: 1-26.

Bonner, J. T., 1950. Observations on polarity in the slime mold Dictyostelium discoideum.
Biol. Bull., 99: 143-151.

BrueE., D., H. Horter, K. LINpERSTRoM-LANG AND K. Rozits, 1946. A micro method for the
determination of total nitrogen (accuracy 0.005 ug. N). C. R. Lab. Carlsberg, Ser.
Chim., 25: 289-323.

Coomss, R. R. A., M. H. GLeeson-WHITE AND J. G. Hatt, 1951. Factors influencing the
agelutinability of red cells. IJ. The agglutination of bovine red cells previously
classified as “inagglutinable” by the building up of an “anti-globulin: globulin lattice”
on the sensitized cells. Brit. J. Exp. Path., 32: 195-202.

Dre Haan, R. L., 1958. Cell migration and morphogenetic movements. Jn: Symposium on
the Chemical Basis of Development (W. D. McElroy and B. Glass, eds.) pp. 339-
374. Johns Hopkins University Press, Baltimore, Md.

Grecc, J. H., 1950. Oxygen utilization in relation to growth and morphogenesis of the slime
mold Dictyostelium discoideum. J. Exp. Zool., 114: 173-196.

Grecc, J. H., 1956. Serological investigations of cell adhesion in the slime molds, Dictyostelium
discoideum, D. purpureum, and Polysphondylium violaceum. J. Gen. Physiol., 39:
813-820.

Grecc, J. H., anp R. D. Bronsweic, 1956. Biochemical events accompanying stalk formation
in the slime mold, Dictyostelium discoideum. J. Cell. Comp. Physiol., 48: 293-300.

Grecc, J. H., anp C. W. Trycstap, 1958. Surface antigen defects contributing to develop-
mental failure in aggregateless variants of the slime mold, Dictyostelium discoideum.
Exp. Cell Res., 15: 358-369.

Raper, K. B., Ann D. I. FENNELL, 1952. Stalk formation in Dictyostelium. Bull. Torrey Bot.
Club, 79: 25-51.

SuarrFer, B. M., 1958. Integration in aggregating cellular slime molds. Quart. J. Micr. Scu.,
99: 103-121.

SPIEGEL, M., 1954a. The role of specific surface antigens in cell adhesion. Part I. The
reagegregation of sponge cells. Biol. Bull., 107: 130-148.

SprEGEL, M., 1954b. The role of specific surface antigens in cell adhesion. Part II. Studies
on embryonic amphibian cells. Biol. Bull., 107: 149-155.

Tyter, A., 1947. An auto-antibody concept of cell structure, growth and differentiation.
Growth, 10 (suppl.) : 7-19.

Tyver, A., 1957. Immunological studies of early development. Jn: A. Tyler, R. C. von
Borstel and C. B. Metz (eds.), The Beginnings of Embryonic Development, Amer.
Assoc. Adv. Sci. Publication 48; pp. 341-382.

Wetss, P., 1947. The problem of specificity in growth and development. Yale J. Biol. Med.,
19: 235-278.

Poorer OF POTASSIUM DEFICIENCY ON THE FREE AMINO
Peer x SERN OF THE MUSCLE TISSUE OF PROTEIN-
Mat te PUNDULUS: HETEROCLITUS +?

DAVID EO HANLON 3

Department of Biochemistry, University of New Hampshire, Durham, New Hampshire

In this investigation the effect of a low potassium environment on the free
amino acid pattern of the skeletal muscle tissue of the killifish, Fundulus hetero-
clitus, has been studied in an attempt to determine if the changes observed in the
potassium-deficient rat (Eckel, Pope and Norris, 1954; Iacobellis, Muntwyler and
Dodgen, 1956) could also be found in other members of the vertebrate subphylum.
Apparently the rat, but not the dog (lacobellis, Griffen and Muntwyler, 1957), is
capable of replacing part of an intracellular loss of potassium ions with organic
cations in the form of basic amino acids, provided a sufficient amount of protein
is included in the diet. Amino acids which could serve to restore an intracellular
cation-anion balance include lysine, arginine, and histidine. All three carry a
net positive charge at body pH. This study, therefore, was made to determine
if an intracellular organic cation replacement occurs in potassium-deficient Fun-
dulus and, if so, which basic amino acids are responsible for the replacement of the
lost potassium.

METHODS

Killifish (Fundulus heteroclitus) were trapped in the brackish water of the
Oyster River near Durham early in August of 1958 and transported to the labora-
tory where they were placed in a large aquarium containing artificially prepared
50% sea water (salinity = 18 parts per thousand). The formula followed was
taken from the Marine Biological Laboratory at Woods Hole manual, “Formulae
and Methods, IV,” 1954. After a three-day period of adjustment, one-half of
the animals was left in 50% sea water to serve as controls, whereas the other
half was placed in an adjacent aquarium also containing 50% artificial sea water
but lacking potassium ions. Control and potassium-deficient animals were main-
tained on a diet of powdered, crude casein.

At the end of a two-week period animals were removed from each aquarium
and their tissues prepared for paper chromatography. In order to obtain 10 grams
of skeletal muscle tissue in each case it was necessary to sacrifice 7 control Fundulus
and 8 potassium-deficient Fundulus. The fish were skinned and filleted. Ten
grams of muscle were macerated in a Waring Blendor, using about 100 ml. of
95% ethanol. After 5 minutes of blending at a moderate speed the liquid was

‘ Published with the permission of the Director of the New Hampshire Agricultural Experi-
ment Station, as Scientific Contribution No. 240.

2 The author wishes to thank Dr. A. E. Teeri and Dr. D. Routley for their helpful sug-
gestions in the preparation of this paper.

3 Present address: Department of Zoology, University of Connecticut, Storrs, Connecticut.

79

80 DAVID P. HANLON

transferred to 50-ml. centrifuge tubes and centrifuged for 5 minutes at 3000 r.p.m.
The centrifugate was removed (the precipitate consists chiefly of protein) and
reduced im vacuo to a volume corresponding to 1 ml. per gm. of tissue extracted.
Slight heat was applied to speed reduction. While awaiting use, the extract
samples were kept in the refrigerator.

Two-dimensional ascending paper chromatography was employed to deter-
mine in a semi-quantitative manner the free amino acids present in the muscle
tissue of the control and potassium-deficient Fundulus. Aliquots of 60 pl. were
spotted on Whatman No. 1 filter paper sheets (9” by 12”) and developed for 16
hours with a phenol-water (4:1) solvent. The chromatograms were dried at
room temperature and developed for 16 hours in the second dimension with a
lutidine-water (2:1) solvent. After drying, the amino acids were located by
spraying the chromatograms with a 0.1% solution of ninhydrin in 95% ethanol.
The Sakaguchi test for arginine and the Pauli test for histidine were also employed
(Block, Durrum and Zweig, 1955). The location of lysine was confirmed on
the ninhydrin-developed chromatograms by exposing them to hydrochloric acid
fumes for one minute. The spot representing lysine turns a bright yellow (Szantai,
1957) which is stable for several days. Valine, leucine and threonine also turn
yellow, but the color soon fades. This, taken in conjunction with the relative
positions of the above-named acids on the two-dimensional chromatogram, makes
any confusion of valine, leucine or threonine for lysine unlikely.

Total amino acid concentration was determined quantitatively as follows. Ten
ul. of the tissue fluid extracts were spotted on Whatman No. 1 filter paper. Also
spotted on the filter paper were 10 pl. quantities of a standard solution containing
6 pgm. of amino acid per wl. Since three amino acids, glycine, alanine and taurine,
were found to be responsible for the bulk of the color observed on the ninhydrin-
sprayed chromatograms, the standard was prepared by dissolving 20 mg. each
of these amino acids in 10 ml. of distilled water. It was felt desirable to use
this type of standard to allow for differences in the absorption characteristics of the
ninhydrin complexes. After application the spotted amino acid solutions were
allowed to dry at room temperature, then were sprayed with a 0.5% solution of
ninhydrin in 75% ethanol plus 0.5% 1 N sodium hydroxide. Kay, Harris and
Entenman (1956) report that the presence of sodium hydroxide is essential for
the development of color in the case of taurine, but otherwise has no effect.
Curiously, the author found that sodium hydroxide had no such specific enhancing
effect in the case of taurine, although a general increased color density was ob-
served. The ninhydrin-treated filter paper was dried at room temperature and
then heated at 60° C. for 25 minutes,’ “The colored) areas) were cut fouteeaae
eluted with 5 ml. of a 75% ethanol solution containing 10 mg. of cupric sulfate
(CuSO,°5H,O) in 200 ml. of solution (Giri, Radhakrishnan and Vaidyanathan,
1952). Blanks were prepared by cutting out portions of filter paper equal in
area to those supporting the colored complexes. Optical density of the colored
solutions was read at 540 mp on a Bausch-Lomb “Spectronic 20.”

Two-dimensional paper chromatography was used to determine the amount of
lysine present in the muscle tissue of potassium-deficient and control Fundulus.
Sixty-pl. aliquots were used in the case of the potassium-deficient fish whereas
400 pl. were required in the case of the control animals, due to the very low con-

POTASSIUM-DEFICIENT FUNDULUS 81

centration of lysine in their tissues. Three to 5 pl. of an aqueous solution contain-
ing 6 mg. of lysine per ml. of solution, serving as standards, were spotted and
chromatographed. Chromatograms were developed using phenol-water (4:1) and
lutidine-water (2:1) as solvents. After drying at room temperature the chromato-
grams were sprayed with ninhydrin as before. Apparently the spraying procedure
is the most crucial point in the quantitative determination. Care should be taken
to spray the chromatograms lightly and evenly on one side, allow them to dry,
then repeat the operation on the other side. The chromatograms were heated,
lysine spots cut out and eluted and optical density read as in the case of the
determination of total amino acid concentration.

RESULTS

Fifteen amino acids were found to be present in the extracts of the control
and potassium-deficient animals. The same amino acids were present in both
cases. Furthermore, it was found that (1) the total amino acid concentration

TABLE [|

The total concentration of amino acids and the concentration of lysine in the skeletal
muscle of control and potassium-deficient Fundulus heteroclitus

Total concentration Lysine concentration
(in mg./gm. of tissue) (in mg./gm. of tissue)
Potassium- Potassium-
deficient Rey iuay ! deficient ponte I
Fundulus i Fundulus ves:
Range 4.9-6.2 2.7-3.3 0.32-0.42 0.030-0.035
Mean AS 3.07 0.387 0.033T
No. of determinations 13 17 4 3

* Difference is significant at 0.1% level.
} Difference is significant at 1.0% level.

in the tissues of the potassium-deficient Fundulus appeared to be somewhat greater
than in the controls (Table 1), and (2) the concentration of lysine in the potassium-
deficient fish was much greater than in the controls, whereas the levels of histidine
and arginine did not seem to be disproportionately large in the former. It was
felt that the apparently greater total amino acid concentration observed in the
potassium-deficient animals could not account for the extreme differences seen in
the concentration of lysine.

Table I shows the values obtained from 17 determinations of the total amino
acid concentration in the tissues of the controls and 13 determinations of the
potassium-deficient fish. A certain amount of variation occurs in both cases.
However, the extent of variation is relatively small when the difficulty of the
method is considered. As was suggested by the semi-quantitative determination,
a significantly greater concentration of free amino acids occurs in the muscle tissues
of the potassium-deficient Fundulus. Although in the controls the concentration

82 DAVID P. HANLON

averages 3.0 mg. per gm. of tissue extracted, it reaches a level of 5.4 mg. per gm.
in the potassium-deficient forms, 7.e., a concentration 1.8 times as great.

Table I indicates the values obtained for lysine levels in control and potassium-
deficient Fundulus. Although only three determinations were made on the control
animals and four on the potassium-deficient animals, the degree of variation was
small in both instances and the differences observed in the lysine concentrations
are regarded as significant. Indeed, the concentration of lysine in the skeletal
muscle of the potassium-deficient forms appears to be more than 10 times as great
as that of the control animals, whereas the total amino acid concentration is only
1.8 times as great.

DISCUSSION

If animals are made deficient in potassium an influx of available extracellular
cations would be expected in order to restore the cation-anion balance. In certain
animals the lost potassium can be partly replaced by a gain in sodium. According
to Cooke and his co-workers (1952) the remainder is replaced by an intracellular
migration of protons, the result being the development of a plasma alkalosis.
However, several investigators (Muntwyler, Griffen and Arends, 1953; Iacobellis,
Muntwyler and Dodgen, 1956; Holliday and Segar, 1957) observed no plasma
alkalosis in the potassium-deficient rat. Since Christensen, Riggs and Palatine
(1952) showed that basic amino acids could act as intracellular cations, Eckel and
his co-workers (1954) investigated the possibility of this being the case in potas-
sium-deficient rats demonstrating no plasma alkalosis. Their results revealed that
8 to 40% of the intracellular metallic cation deficiency is replaced by lysine.

The results of this investigation show that, as in the case of the rat, the
cation-anion imbalance appearing in the potassium-deficient Fundulus is partially
overcome by an increase in the intracellular concentration of lysine. In Table I
it can be seen that the concentration of lysine in the potassium-deficient Fundulus
is approximately 10 times greater than in the controls. In terms of per cent of
total amino acid concentration, the value for lysine is 1.1% in the case of the control
animals and 7.0% in the potassium-deficient animals, an increase greater than 500%.

The present results indicate that histidine and arginine contribute little to
the increase in intracellular organic cation concentration in Fundulus. Further-
more no decrease was observed in aspartic and glutamic acids as Jacobellis, Munt-
wyler and Dodgen (1956) found with Wistar rats.

The total free amino acid concentration in the skeletal muscle tissue of potas-
sium-deficient Fundulus is almost twice that of the controls (Table I). The
observed increase may be the result of a decrease in the rate of nitrogen anabolism
and perhaps the establishment of a negative nitrogen balance. Evidence that such
is the case here is supplied by the work of Muntwyler, Griffen and Arends (1953)
and Eckel, Norris and Pope (1958). Both groups reported decreased growth
and a failure to maintain a proper nitrogen balance in the potassium-deficient rat.
Conceivably in Fundulus as well as the rat an increased amount of free amino
acids would be due to a failure of body protein synthesis and in certain cases
(lysine) could serve to ameliorate a state of cation-anion imbalance by functioning
as intracellular cations.

The fact that an identical device for combatting a loss of cellular potassium
can be seen in animals as widely separated as mammals and fish suggests the

POTASSIUM-DEFICIENT FUNDULUS 83

interesting possibility of the universality of such a mechanism. It may be that a
similar or perhaps identical mechanism accompanying a potassium deficiency can
be found throughout the entire vertebrate group, indeed, perhaps in several phyla
of the animal kingdom.

SUMMARY

1. The effect of a potassium deficiency on the free amino acids of the muscle
tissues of Fundulus heteroclitus was investigated employing the method of quanti-
tative paper chromatography.

2. An over-all increase in the total amino acid concentration was observed in
the potassium-deficient fish amounting to 1.8 times that seen in the controls. A
great increase in the concentration of lysine was also observed in the potassium-
deficient forms. In this case the concentration was about 10 times that of the
controls.

3. It is therefore suggested that the mode of combating a potassium deficiency
in the rat also exists in a fish. The interesting possibility that similar or perhaps
identical mechanisms for lessening the effects of a potassium deficiency may be
found throughout other representative vertebrate groups, as well as the inverte-
brates, offers a fruitful field for further comparative investigations.

LITERATURE CITED

Biock, R., E. Durrum Anp G. Zweic, 1955. Paper Chromatography—Paper Electrophoresis.
Academic Press, New York.

CHRISTENSEN, H., H. Riccs anp I. Paratine, 1952. Intense concentration of a, y diamino-
butyric acid. J. Biol. Chem., 198: 17-22.

Cooke, R., W. Secar, D. CHEEK, F. CovILLE AND D. Darrow, 1952. The extrarenal correc-
tion of alkalosis associated with potassium deficiency. J. Clin. Invest., 31: 798-805.

EckeL, R., C. Pope anp J. Norris, 1954. Lysine as a muscle cation in potassium deficiency.
Arch. Biochem. Biophys., 52: 293-294.

Ecxet, R., J. Norris AND C. Pops, 1958. Basic amino acids as intracellular cations in potas-
sium deficiency. Amer. J. Physiol., 193: 644-652.

Giri, K., A. RADHAKRISHNAN AND C. VAIDYANATHAN, 1952. A simple paper chromatographic
method for the study of transamination reactions. Nature, 170: 1025-1026.

Ho.iiay, M., anp W. Secar, 1957. The effect of low electrolyte feeding on the development
of a potassium deficiency. Amer. J. Phystol., 191: 610-614.

IAcCOBELLIS, M., E. MuUNTWYLER AND C. Doncen, 1956. Free amino acid patterns in certain
tissues of potassium deficient rats. Amer. J. Physiol., 185: 275-278.

IAcoBELLIS, M., G. GrRIFFEN AND E. Muntwy ter, 1957. Free amino acid patterns of certain
tissues of potassium deficient dogs. Proc. Soc. Exp. Biol. Med., 96: 64-66.

Kay, R., D. Harris anp C. ENTENMAN, 1956. Quantification of the ninhydrin color reaction
as applied to paper chromatography. Arch. Biochem. Biophys., 63: 14-25.

Muntwyter, E., G. E. GrirFEN AND R. Arenps, 1953. Muscle electrolyte composition and
balances of nitrogen and potassium in potassium deficient rats. Amer. J. Physiol.,
174: 283-288.

SzanTal, I., 1957. Behavior of amino acids developed with ninhydrin in an atmosphere of
hydrogen chloride. Rev. Chim. (Bucharest), 8: 544. (C. A., 52: 2142a, 1958.)

STRUCTURAL AND FUNCTIONAL CEH ANGI S Ui
GENERATION IN TEPRAEY MENA?

GEORGE, G.), HOLGATE

Department of Zoology, Syracuse University, Syracuse, N. Y.

With the development of techniques for synchronizing the division of members
of a population of microorganisms, it has become possible to analyze many changes
that occur in a cell during a generation (the period from the end of one division
to the end of the next). The ciliate protozoan Tetrahymena pyriformis has been
used in this way by Zeuthen and his coworkers in studies directed at biochemical
changes; nitrogen content (Hamburger and Zeuthen, 1958), nucleoside triphos-
phate concentration (Plesner, 1958a), rate of incorporation of P*? (Hamburger
and Zeuthen, 1958), and sensitivity to anaerobiosis (Rasmussen, 1958), dinitro-
phenol (Hamburger and Zeuthen, 1957), and inhibitors of oxidative metabolism
(Hamburger, 1958). Synchronously-dividing populations are also useful for trac-
ing the time course of structural changes. Tetrahymena is uniquely suited to such
a study since it passes through a period of morphogenesis in each generation that
results in duplication of the structures necessary for the formation of a new indi-
vidual before division (body cilia, mouth, contractile vacuoles, cytopyge) (Chatton
et al., 1931; Furgason, 1940), and it undergoes nuclear changes involving a mitotic
division of the micronucleus and an amitotic division of the macronucleus (Maupas,
1883, 1888).

The following is a description of the time course of the structural changes
(those visible with the light microscope) in a generation, in a synchronized popu-
lation of Tetrahymena pyriformis, mating type I, variety 1, and a comparison of it
with the time course of known chemical changes 1n a generation, observed with other
strains of Tetrahymena.

MATERIALS AND METHODS

Division was synchronized by the application of 5 alternate 14-hour exposures
to temperatures of 42.66 +0.2° C. and 35+0.2° C. Other details of procedure
were as described by Holz, Scherbaum and Williams (1957). Following the final
high temperature period, samples of ciliates were withdrawn at 5-minute intervals
for 180 minutes. To determine when the first and second synchronous divisions
occurred, the division index (cells in division + cells counted) was measured for
each sample. The first synchronous division (70-80% synchrony) occurred 50-60
minutes after the last high temperature period, and the second division (45-55%
synchrony) 65-75 minutes after the first division. Changes in the nuclear ap-
paratus (Fig. 1), infraciliature (Fig. 8), and general appearance of the ciliates
during the interval between divisions were followed by Feulgen-staining (Dippell

1 This work was supported by Grant E-797 of the National Institute of Allergy and In-
fectious Diseases, U. S. Public Health Service, to G. G. Holz, Jr.

84

CHANGES IN A GENERATION IN TETRAHYMENA 85

and Chao’s modification of the DeLamater stain; Sonneborn, 1950), silver impreg-
nation (Chatton-Lwoff technique; Corliss, 1953), and the observation of living
individuals with a phase contrast microscope. The temporal positions of the
nuclear and morphogenetic events in a generation were determined by comparisons
of the times of greatest frequency of appearance of these events (in the samples
taken between the first and second synchronous divisions) with the times when
the first and second synchronous divisions occurred.

To establish a basis for comparison of the times of structural changes in mating
type I, variety 1, with the times of changes in physiological and biochemical param-
eters reported for other strains, all generation times of the various strains used
were compressed to 100 arbitrary units and the time courses of all changes adjusted
to fit this scale. For synchronously-dividing cells, the generation time used in
these calculations was that of the synchronous generation during which observation
of a particular structural or functional change was made. With one exception this
was the second synchronous generation (the period between the first and second
synchronous divisions). Nucleoside triphosphate content was determined at vari-
ous times during the first synchronous generation (the period from the end of
the last high temperature exposure to the first synchronous division) (Plesner,
1958a). Data on DNA synthesis (McDonald, 1958) and sensitivity to extremes
of temperature (Thormar, 1956) were obtained with single normal cells, nitrogen
content (Hamburger and Zeuthen, 1958), sensitivity to dinitrophenol (Hamburger
and Zeuthen, 1957), sensitivity to inhibitors of oxidative metabolism (Hamburger,
1958), nucleoside triphosphate content (Plesner, 1958a), and rate of incorporation
of P** (Hamburger and Zeuthen, 1958) were measured with groups of synchro-
nized cells, and rate of respiration (Zeuthen, 1953; Zeuthen and Scherbaum, 1954)
and sensitivity to anaerobiosis (Rasmussen, 1958) with both types. Such a com-
parison is based on several assumptions: first, that there are no gross differences
in the time scales of structural and functional events in a generation in different
strains of Tetrahymena pyriformis with different generation times and nuclear
constitutions (micronucleate and amicronucleate), 1.e., that each event occurs at
the same relative time in every strain; second, that the time characteristics of
structural and functional changes occurring between synchronous divisions can be
compared with those occurring between divisions of normal cells (cells that have
not been heat-treated ).

The first assumption is made on the basis that the structural and chemical
changes that have been studied are of such a fundamental nature, necessary to the
orderly progression of growth and preparation for division, that they probably
follow the same relative time courses in all strains. This assumption was tested
experimentally, so far as structural changes were concerned, by comparing the
time courses of morphogenetic and macronuclear changes in strain GL and in
mating type I, variety 1, with synchronized populations. Strain GL was syn-
chronized by the method of Scherbaum and Zeuthen (1954). Comparison was
also made of the time course of structural changes of mating type I, variety 1,
with that reported by Browning, Varnedoe and Swinford (1952) for strain T-P.
McDonald (1958) and Prescott and Bors (1958), using different strains, reached
fundamentally the same conclusion concerning the period of DNA synthesis in a
generation.

86 GEORGE G. HOLZ,-JR.

Ficure 1. Resting nuclear configuration in Tetrahymena pyriformis, mating type I, vari-
ety 1. In this and the following illustrations of nuclear states, ciliates were not destained to
a degree necessary to bring out fine nuclear detail. This was done to insure visualization of
the cilia of the mouth, to assist in observing time relations of nuclear and morphogenic states.
The marker indicates 10 uw for Figures 1-7.

Ficure 2. Stage of macronuclear enlargement. The micronucleus is peripherally located.
Cilia of the forming mouth of the opisthe are seen.

Ficure 3. Micronuclear anaphase.

Ficure 4. Micronuclear division complete. Macronuclear division and cytoplasmic cleavage
beginning.

CHANGES IN A GENERATION IN TETRAHYMENA 87

The second assumption is more open to question. Cells prepared for syn-
chronous division by the temperature-cycling method have a shorter generation
time (Zeuthen and Scherbaum, 1954), have a greater volume (Scherbaum, 1956),
greater reduced weight (Scherbaum and Zeuthen, 1954), greater dry weight
(Scherbaum, 1957b), and contain more nitrogen (Zeuthen and Hamburger, 1956),
DNA and RNA (Scherbaum, 1957b), ATP and GTP (Plesner, 1956, 1958b) than
do normal cells. During the synchronous divisions these characteristics return
to normal. To liken the time courses of biochemical changes that have been
studied in these cells to the condition in normal cells, we must assume that the
differences noted above influence only the extent of the changes studied and not
their time characteristics with respect to division. In this connection, it is notable
that rate of respiration (Zeuthen, 1953; Zeuthen and Scherbaum, 1954) and sensi-
tivity to anaerobiosis (Rasmussen, 1958) follow the same time courses between
divisions in heat-treated and normal cells. The chemical composition of the heat-
treated cells might cause the morphogenetic and nuclear changes to follow different
time courses with respect to one another. To test this possibility, synchronized
cells, and normal cells from cultures reproducing at an exponential rate, were
compared for nuclear state in silver impregnations showing stages of stomatogenesis.
In such preparations the micronucleus shows up well except during its final division
stages.

RESULTS

The resting micronucleus was spherical and lay next to the macronucleus,
or within a cup-shaped depression in the macronucleus (Fig. 1). The resting state
was assumed immediately after completion of cytoplasmic cleavage and lasted for
about 75% of a generation. At the end of this period the micronucleus moved
laterally toward the periphery of the organism and elongated (Fig. 2). Chromo-
somes appeared at this time. The most clearly recognizable stages of mitosis were
those associated with separation of the chromosomes (Fig. 3). As the chromo-
somes reached the poles of the division figure the latter greatly elongated and the
micronucleus was divided into halves which became separated respectively into
the anterior and posterior parts of the dividing organism (Fig. 4). These daughter
micronuclei then became spherical again. The division of the micronucleus went
to completion as macronuclear division and cytoplasmic cleavage began.

Key To Ficures 1-16

’

Mouth of “resting” (interphase) ciliate.

Mouth of proter (anterior daughter) during stomatogenesis.
Micronucleus.

Macronucleus.

Mouth of opisthe (posterior daughter).

Macronuclear fragment separated during macronuclear division.
Stomatogenous field, site of formation of mouth of opisthe.
Contractile vacuole pore of proter.

Cleavage furrow.

Stomatogenous meridian.

Formation of membranelles.

Formation of undulating membrane.

Ae ee THO OO oP

88 GEORGE G. HOLZ, JR.

Ficure 5. Macronuclear division and cytoplasmic cleavage continuing.

Figure 6. Macronuclear division complete. Cytoplasmic cleavage continuing.

Figure 7. Product of a recent division, showing re-establishment of resting nuclear
configuration.

Ficure 8. Infraciliature of Tetrahymena pyriformis, mating type I, variety 1. Note normal
mouth of proter with typical tetrahymenal configuration, and stomatogenesis to produce mouth
of opisthe. The marker indicates 10 w for Figures 8-11.

,

ve
;
’

a. wT

tty

Ficure 9. Completion of stomatogenesis in opisthe.

Ficure 10. Appearance of cleavage furrow at beginning of division (dorsal surface of
ciliate). Note contractile vacuole pore of opisthe just anterior to furrow on right side of
animal.

Ficure 11. Cleavage.

Ficure 12. Very early stage of stomatogenesis. Kinetosomes appear to left (animal’s
left) of stomatogenous meridian. The marker indicates 5 « for Figures 12-16.

90 GEORGE G. HOLZ. TR:

Ficure 13. Stomatogenesis; anarchic field.

Ficure 14. Stomatogenesis. Very early stage of membranelle organization at anterior
end of field.

Ficure 15. Stomatogenesis. Three membranelles and undulating membrane formed.
FicureE 16. Stomatogenesis; terminal stage.

The macronucleus was an approximately spherical, centrally-located structure
(Fig. 1). Immediately after cleavage, it assumed a resting state characterized
by a moderately dense and homogeneous granular appearance. At the time the
micronucleus migrated to the periphery of the organism, the macronucleus in-

r
»
1
‘
4

eee eee

CHANGES IN A GENERATION IN TETRAHYMENA 91

creased in size and granularity (Fig. 2). As the micronucleus divided, the macro-
nucleus elongated (Fig. 4), and at the time the cleavage furrow separated the
daughter ciliates, it divided into approximately equal halves (Figs. 5 and 6).
Small portions of the macronucleus occasionally separated from the main macro-
nuclear masses at this time and became isolated in the cytoplasm as spherical,
Feulgen-positive bodies which were later resorbed (Furgason, 1940; Scherbaum

et al., 1958).

GENERATION TIME = 100 ARBITRARY UNITS

fe) fe) 20 30 40 50 60 70 80 90 100
8 8 tS
| MicRONUCLEAR EVENTS AB Ic |
| MACRONUCLEAR EVENTS D | E |
| MORPHOGENETIC EVENTS F | |

SS VW.

7 eee eee ee
ee ew ee ee 8 we ee ee

C5 CISUSL Sim elelie Ze Olo.ee) 08,0 sere ete.

7 eee oe wo we 8

aaa ete See Le aie) weave ere ete ca

o°. 8.8 © © © © © 0 0 0 6 oo 0 8 ce 8 ee eo
SNARES BNO Le) SSR eee ace alee ue ie cm teks raleadaltedenere decorate ne
ROO KS SNOASL OSL aS Ker Rieleelle)elelaievereve tare re

woe eee ee ee wee 8 ew ce

| SENSITIVITY TO ANAEROBIOSIS AND TO ) et
INHIBITORS OF OXIDATIVE |
| METABOLISM! |

7 eee ee ee we we ew ow oe ee we te
Ce eee
Co ee ee
ees e eee eee eee ee

| NUCLEOSIDE
| CONTENT

TRIPHOSPHATE

7 eee eee eee ee
eoecee eee ee eee ee
wes eee eee ee ee owe
eee eee eee eee ee ee ee el

| eee
| =
5
= 5 . . 5
7° . Sr ane * . >
. ese e406 2°, % es . area enc ere

RATE OF INCORPORATION P>

CONTENT

| NITROGEN

DNA

SENSITIVITY TO HIGH AND
LOW TEMPERATURES

Age et se er aere Ooo ete, oa ©

Figure 17. Graphic representation of approximate time relationships of structural and
functional changes occurring in Tetrahymena pyriformis in a generation. (A—migration of
micronucleus to periphery, B—mitosis, C—separation of products of mitosis into daughters,
D—macronuclear swelling, E—macronuclear division, F—stomatogenesis, 1-nitrogen; Ras-
mussen, 1958, dinitrophenol; Hamburger and Zeuthen, 1957, fluoride, fluoracetate, azide, ethyl-
urethane; Hamburger, 1958).

92 GEORGE :GyHOEZ. LR:

Morphogenesis 1n Tetrahymena involves formation of new body cilia, a new
mouth (stomatogenesis) for the posterior daughter (opisthe), and new contractile
vacuole pores and a cytopyge (cell anus) for the anterior daughter (proter). Body
cilia are duplicated at a time not yet precisely determined. Stomatogenesis (mouth
formation) was manifested first by the appearance of kinetosomes (basal bodies of
cilia) to the left of the stomatogenous kinety (row of cilia, or primary ciliary
meridian, passing posteriorly from the right side of the buccal overture) near the
middle of the body (Fig. 8). The precise time when stomatogenesis began was
not determined but early stages were seen before 45% of a generation was com-
pleted. Successive stages in the duplication of the kinetosomes of the cilia of the
new mouth and their organization into membranelles are shown in Figures 12-16.
The anlage of the mouth was in the late anarchic field stage (Fig. 14), or early
stage of membranelle organization, at the time the micronucleus assumed a periph-
eral position. Stomatogenesis was complete, and the membranelles of the mouth
appeared to be functional in living ciliates at the time of division of the micronucleus
(Pigs 0. ees MO).

The contractile vacuole pores of the proter were seen in silver impregnations
(Fig. 10) as early as the time of lateral movement of the micronucleus, and the
contractile vacuole was observed functioning in living organisms at this time.

The cleavage furrow first appeared just anterior to the mouth of the opisthe
at the time of separation of the halves of the dividing micronucleus. A space
appeared between kinetosomes of the primary ciliary meridians, in a line around
the equator of the organism (Fig. 10). The furrow deepened until the proter
and opisthe were separated. Members of the genus Tetrahymena should be excel-
lent organisms for testing theories of the mechanism of cleavage since they have
markers, the kinetosomes, in the cortex. An examination of the pattern of infra-
ciliature in the cortex during cleavage revealed no distortion except in the region
of the cleavage furrow, where the cilia-free space appeared (Fig. 10), and where
a slight torsion of the ends of the kineties next to the furrow sometimes occurred
(Fig. 11). Theories which depend upon polar expansions (active or passive) of
the cortex accompanying furrow formation (Swann and Mitchison, 1958) seem to
be untenable for Tetrahymena.

Except for the absence of the micronucleus in strains GL and T-P, no dif-
ferences in the relationships between nuclear and morphogenetic events were noted
when normal cells of strain GL and mating type I, variety 1, were compared,
when normal and synchronized cells of mating type I, variety 1, were compared,
or when normal cells of strain T-P and mating type I, variety 1, were compared
(data of Browning et al., 1952).

Figure 17 is a graphic representation of the approximate time relationships of
various structural and functional changes occurring in Tetrahymena pyriformis in
a generation.

Discussion

As noted by Zeuthen (1958), there are striking associations between those
biochemical changes related to energy production, and the initiation and completion
of nuclear events associated with division and the process of cytoplasmic cleavage.
Rate of respiration, sensitivity to anaerobiosis and to dinitrophenol and inhibitors
of oxidative metabolism, nucleoside triphosphate content, and rate of P** incorpora-

CHANGES IN A GENERATION IN TETRAHYMENA 03

tion reach a maximum at the time of initiation of nuclear events. During the fol-
lowing period of nuclear changes, before the appearance of the cleavage furrow,
the rate of respiration remains high and constant, sensitivity to anaerobiosis and
to dinitrophenol and inhibitors of oxidative metabolism are at a minimum, and
nucleoside triphosphate content and the rate of incorporation of P** decline. When
the cleavage furrow appears, sensitivity to anaerobiosis and to dinitrophenol and
inhibitors of oxidative metabolism begins to rise, while nucleoside triphosphate
content and the rate of P*? incorporation continue to decline. As cleavage pro-
ceeds, the rate of respiration begins to rise again.

The last 25% of a generation is also a period that shows the greatest fluctua-
tions in sensitivity to extremes of temperature. Sensitivity is at a minimum during
the final 74 of cytoplasmic cleavage. It begins to rise just before the completion
of cleavage and rises gradually until the time of nuclear changes, when it rises
abruptly to reach a maximum just before the onset of visible cleavage. It then
drops sharply to a minimum during the first 44 of cleavage (Thormar, 1956). The
most sensitive time coincides with the time of chromosome movement and the
elongation of the micronucleus accompanying its division, just before the macro-
nucleus begins to elongate, and at the time of final organization of the membranelles
of the mouth of the opisthe. These nuclear and morphogenetic stages are charac-
teristic of ciliates prevented from dividing by the cyclic heat treatments used to
induce synchronous division (Holz et al., 1957). The high sensitivity is not
attributable to the effects of temperature on micronuclear division, since the amicro-
nucleate strain GL seems to be blocked at the same point in time (Williams and
Scherbaum, 1959). It is not possible at present to associate the high sensitivity
of this period with any one of the coincident processes occurring (nuclear changes,
morphogenetic changes, preparation for cytoplasmic cleavage). The cyclic heat
treatments do not prevent the early stages of mitosis, amitosis or stomatogenesis,
but do stop these processes at very characteristic stages and have no differential
effect on them such that one progresses in time beyond the others. In normal
cells from cultures which are reproducing at an exponential rate, nuclear changes
and stomatogenesis also follow parallel courses, with time characteristics such
that the early stages of micronuclear division, just prior to macronuclear elonga-
tion, and the late anarchic field stage and early stages of membranelle organization
coincide. The fact that mitosis, macronuclear division and stomatogenesis are
blocked by high temperature at characteristic stages suggests that the heat-labile
biochemical system (systems) postulated by Scherbaum (1957a) and by Zeuthen
(1958) is (are) necessary to the completion of all these processes, or that each
process is dependent upon the others for its normal progression and goes to com-
pletion only when the others are unimpaired.

McDonald (1958), using a Feulgen microspectrophotometric method on the
macronuclei of single cells, observed that DNA synthesis began after passage of
about 10% of a generation, continued for approximately 45%, and ended shortly
before the initiation of nuclear structural changes. In a recent preliminary note
Prescott and Bors (1958) report a slightly shorter period of synthesis starting
at about the same relative time and ending midway in the generation. They fol-
lowed the course of synthesis by autoradiographic detection of the incorporation
of tritiated thymidine. These findings are not in agreement with the earlier report

94 GEORGE G. HOLZ, JR.

of Walker and Mitchison (1957) that a linear synthesis occurs throughout inter-
phase. Duplication of DNA in the first half of a generation is not characteristic
of all ciliates. Euplotes eurystomus (Fauré-Fremiet et al., 1957), Paramecium
caudatum (Walker and Mitchison, 1957), and Paramecium aurelia (Kimball and
Barka, 1959) synthesize their DNA during the last half of a generation.

The author wishes to acknowledge with thanks the helpful suggestions of Dr.
Verner Wulff and Dr. Eric Zeuthen.

SUMMARY

The time courses of structural events occurring in a synchronous generation
of IT. pyriformis, mating type I, variety 1, were compared with one another and
with the time courses of functional events in a synchronous generation, and/or in
a generation of single, normal ciliates, of other strains of the species.

Notre ADDED IN PROOF

Since this manuscript was accepted for publication two important relevant
papers have appeared. Scherbaum et al., J. Cell. Comp. Physiol., 53: 119-138,
1959, reported that in strain GL the pattern of protein amino acids at stages in
normal growth, and during the first synchronous generation of heat-treated cells,
was stable. The free amino acids present at the same stages varied in their con-
centrations, but not in a manner that could be correlated with any of the cytological
changes that occurred. Scherbaum et al., Exp. Cell Res., 18: 150-166, 1959, re-
ported that in strain GL, during the first synchronous generation (end of heat-
treatment to onset of first synchronous division), the dry weight, volume, acid-
soluble phosphates, acid-insoluble heat-labile phosphates, and DNA content of the
average cell increased, and thymidine incorporation into DNA could be demon-
strated. In addition, circumstantial evidence from microspectrophotometric experi-
ments, indicated a burst of DNA synthesis just prior to and during the first
synchronous division. As the authors suggested, substantiation of this finding
would make it unwise to apply information on DNA synthesis obtained with
synchronized cells to the process in normal cells.

LITERALURESCLEED

Browninc, I., N. B. VarNeporE AND L. R. Swinrorp, 1952. Time of nuclear, cytoplasmic and
cortical division of the ciliated protozoan Tetrahymena geleu. J. Cell. Comp. Physiol.,
39: 371-381.

CuHaTton, E., A. Lworr Anp J. L. Monop, 1931. La formation l’ébauche buccale postérieure
chez les cilies en division et ses relations de continuité topographique et génétique avec
la bouche antérieure. C. R. Soc. Biol., 107: 540-544.

Cortiss, J. O., 1953. Silver impregnation of ciliated protozoa by the Chatton-Lwoff technique.
Stain Technol., 28: 97-100.

Fauré-FREMIET, E., C. RouIrteR AND M. GAucHeEry, 1957. La réorganisation macronucléaire
chez les Euplotes. Etude au microscope électronique. Exp. Cell Res., 12: 135-144.

Furcason, W. H., 1940. The significant cytostomal pattern of the “Glaucoma-Colpidium
group,” and a proposed new genus and species, Tetrahymena geleu. Arch. f. Protistenk.,
94: 224-266. é

Hampurcer, K., 1958. Cited in Zeuthen, E., 1958. Periodicity in living cells. Adv. Buol.
Med. Phys., 6: 37-73.

CHANGES IN A GENERATION IN TETRAHYMENA 95

Hampurcer, K., AND E. ZEUTHEN, 1957. Synchronous divisions in Tetrahymena pyriformis
as studied in an inorganic medium. The effect of 2,4-dinitrophenol. Exp. Cell Res.,
13: 443-453.

HampBurcer, K., AND E. ZEUTHEN, 1958. Cited in Zeuthen, E., 1958. Periodicity in living
cells. Adv. Biol. Med. Phys., 6: 37-73.

Horz, G. G., O. H. ScHERBAUM AND N. WILtiAMs, 1957. The arrest of mitosis and stomato-
genesis during temperature-induction of synchronous division in Tetrahymena pyri-
formis, mating type I, variety 1. Exp. Cell Res., 13: 618-621.

KIMBALL, R. F., AnD T. BArKA, 1959. Quantitative cytochemical studies on Paramecium aurelia.
II. Feulgen microspectrophotometry of the macronucleus during exponential growth.
Exp. Cell Res., 17: 173-182.

MaupaAs, E., 1883. Contributions a l’étude morphologique et anatomique des infusoires ciliés.
Arch. Zool. Exp. Gén. (sér. 2), 1: 427-664.

Maupas, E., 1888. Recherches experimentales sur la multiplication des infusoires ciliés. Arch.
mec Exp. Gén. (sér. 2), 6: 165-277.

McDonatp, B. B., 1958. Quantitative aspects of deoxyribose nucleic acid (DNA) metabolism
in an pleronicieate strain of Tetrahymena. Biol. Bull., 114: 71-94.

Presner, P., 1956. Incorporation of P®” into the purine ribonucleotides of Tetrahymena pyri-
formis in heat-treated cultures. Acta Chem. Scand., 10: 161-162.

Priesner, P., 1958a. The nucleoside triphosphate content of Tetrahymena pyriformis during
the division cycle in synchronously dividing mass cultures. Biochem. Biophys. Acta,
29: 462-463.

PLesner, P., 1958b. Cited in Zeuthen, E., 1958. Periodicity in living cells. Adv. Biol. Med.
Phys., 6: 37-73.

Prescott, D. M., Anp K. Bors, 1958. Synthesis of protein, ribonucleic acid and desoxyribo-
nucleic acid over the cell cycle. Anat. Rec., 132: 489.

RAsMussEN, L., 1958. Cited in Zeuthen, E., 1958. Periodicity in living cells. Adv. Biol. Med.
Phys., 6: 37-73.

ScHeErBAUM, O., 1956. Cell growth in normal and synchronously dividing mass cultures of
Tetrahymena pyriformis. Exp. Cell Res., 11: 464-476.

ScHERBAUM, O., 1957a. Studies on the mechanism of synchronous cell division in Tetrahymena
pyriformis. Exp. Cell Res., 13: 11-23.

ScHERBAUM, O., 1957b. The content and composition of nucleic acids in normal and syn-
chronously dividing mass cultures of Tetrahymena pyriforms. Exp. Cell Res., 13
24-30.

ScHERBAUM, O., AND E. ZEUTHEN, 1954. Induction of synchronous cell division in mass cul-
tures of Tetrahymena pyriformis. Exp. Cell Res., 6: 221-227.

ScHerBAum, O. H., A. L. LoupERBAcK AND T. L. JAHN, 1958. The formation of subnuclear
aggregates in normal and synchronized protozoan cells. Biol. Bull., 115: 269-275.

SoNNEBORN, T. M., 1950. Methods in the general biology and genetics of Paramecium aurelia.
J. Exp. Zool., 113: 87-143.

Swann, M. M., anno J. M. Mitcuison, 1958. The mechanism of cleavage in animal cells.
Biol. Revs., 33: 103-135.

THormMar, H., 1956. Cited in Zeuthen, E., 1958. Periodicity in living cells. Adv. Biol. Med.
Phys., 6: 37-73.

WaAtker, P. M. B., ann J. M. Mircuison, 1957. DNA synthesis in two ciliates. Exp. Cell
Res., 13: 167-170.

WuutiAMms, N., Ano O. H. ScHersAum, 1959. Morphogenetic events in normal and synchro-
nously dividing Tetrahymena pyrifornuis. J. Embryol. Exp. Morphol., in press.

ZEUTHEN, E., 1953. Growth as related to the cell cycle in single cell cultures of Tctrahymena
pyriformis. J. Embryol. Exp. Morphol., 1: 239-249.

ZEUTHEN, E., 1958. Periodicity in living cells. Adv. Biol. Med. Phys., 6: 37-73.

ZEUTHEN, E., Ano K. HaAmpurcer, 1956. Cited by Scherbaum, O., 1956. Cell growth in
normal and synchronously dividing mass cultures of Tetrahymena pyriformis. Exp.
Cell Res., 11: 464-476.

ZEUTHEN, E., AnD O. ScHERBAUM, 1954. Synchronous divisions in mass cultures of the ciliate
protozoan Tetrahymena pyriformis, as induced by temperature changes. Proc. 7th
Symp. Colston Res. Soc., pp. 141-156.

ANTIGENS OF THE: SEA URCHIN: SPERM SURFAGEHE

KURT KOHLER2 AND CHARLES B. METZ

Oceanographic Institute, Florida State University, Tallahassee, Florida, and
Marine Biological Laboratory, Woods Hole, Mass.

Morphological and physiological evidence indicates that the initial steps in
fertilization involve the sperm and the egg surfaces. Furthermore, these steps
would appear to be chemical interactions between surface substances. Beginning
with the pioneer studies of F. R. Lillie (1913, 1914), the chemical relationships
have been visualized as between complex molecules; and antigen-antibody systems
have served as useful models (see Tyler, 1948). Most of the present information
has been obtained from studies on agents extracted from sperm and eggs. Notable
among these agents are the sperm isoagglutinin, fertilizin, obtained from the egg jelly
layer and the egg agglutinating antifertilizin obtained from sperm. Although
these and other agents have been studied extensively, their role in fertilization
has not been clearly defined (see Metz, 1957a, for review). On the other hand,
relatively little effort has been directed toward analysis of the intact cell surface.
Contributions in this direction include Tyler’s (1946) study of the effect of specific
antiserum on the fertilizing capacity of sea urchin sperm. In this study non-
agglutinating, univalent antibody prepared against sperm antifertilizin was found
to reduce the fertilizing capacity of sea urchin sperm. In an analogous investiga-
tion on Paramecium, antiserum was found to block initial steps in the mating
process (Metz, 1954). Extension of this line of study would seem to promise
significant information for an understanding of the interaction of egg and sperm
surfaces at fertilization. As a step in such an investigation it seemed desirable to
map out the antigenic structure of the Arbacia sperm surface. Accordingly, in
the present study an effort was made to determine the number of sperm surface
antigens and their distribution with respect to the morphology of the sperm, and
to examine for sperm surface antigens that do not appear in sea water extracts.

Studies on the antigenic structure of sperm are found in publications extending
over a period of 60 years. However, most of such studies were not directed pri-
marily toward an understanding of the fertilization process. More recently Mudd
and Mudd (1929), Henle (1938), Snell (1944), Smith (1949) and Pernot (1957)
among others have demonstrated species, strain and tissue specificities. Certain
of these studies, notably those of Henle et al. (1938) and Pernot (1957), are also
concerned with the number and distribution of sperm antigens. The former study
showed three agglutinogens on bull sperm, one confined to the tails, one restricted
to heads and the third common to both. Pernot’s (1957) agar diffusion and
immunoelectrophoretic study of guinea pig material revealed eleven antigens in

1 Contribution No. 114 from the Oceanographic Institute, Florida State University. Aided
by grants from the National Science Foundation and The National Institutes of Health.

2 Fulbright Fellow. Present address: Max-Planck-Institut fur Virusforschung, Tubingen,
Germany.

96

ANTIGENS OF SEA URCHIN SPERM SURFACE 97

the seminal plasma and seven in alkaline sperm tail extracts. Six antigens were
common to both materials. Pernot also presents some evidence for an antigen
common to both heads and tails, and an experiment suggesting another antigen
restricted to the guinea pig sperm tail. It remains to be shown which if any of
the sperm antigens in Pernot’s extracts are components of the cell surface. Finally,
Dallam and Thomas (1953) have isolated lipoprotein from sperm heads of several
mammalian species. Antisera prepared against such lipoprotein agglutinated sperm.

These studies concern mammalian material. Among echinoderms extensive
serological investigations of the egg have been made by Tyler and Brookbank
(1956a, 1956b), and the Perlmanns (Perlmann, 1957; Perlmann and Perlmann,
1957a, 1957b; see Runnstrom et al., 1959, for review). Studies of the sperm
include a brief reference by Perlmann (1957) and accounts by Tyler and O’Mel-
veny (1941) and Tyler (1946, 1949). In the last report Tyler (1949) lists cross-
agglutination reactions between antiserum and sperm of several species and demon-
strates a minimum of two antigens on the Strongylocentrotus purpuratus sperm
surface.

MATERIAL AND METHODS

Seven echinoderms were employed in the study. These include Arbacia punc-
tulata, Lytechinus variegatus, Mellita quinquiesperforata and Plagiobrissus grandis
from the vicinity of the Florida State University Marine Laboratory at Alligator
Point, Florida, and Arbacia punctulata, Echinarachnius parma and Asterias forbesi
from Woods Hole, Mass.

Gametes were obtained from Arbacia by electrical stimulation (see Harvey,
1956) or by excising the gonads. Treatment with isotonic KCl was avoided since
this method causes release of the antigen-containing dermal secretion (Metz, 1959).
Gametes were obtained from Lytechinus, Mellita, Echinarachnius and Plagiobrissus
by treating the gonads with isotonic KCl.

Sperm and seminal plasma were separated by low speed centrifugation. The
sperm was re-suspended once in sea water and centrifuged again. The super-
natant was discarded and the settled sperm was made up as a 25% suspension by
addition of three volumes of sea water. This stock suspension was used for
injections and absorptions. For spot plate agglutination tests a 1% sperm suspen-
sion was prepared by diluting the stock suspensions to %;5 with sea water.

Injections. To obtain antisera five rabbits were injected with Arbacia sperm
(25% washed sperm in sea water) or sperm extracts, three rabbits were injected
with Lytechinus, one with Echinarachnus and three with Asterias sperm. The
immunizing antigens were administered through intravenous, intraperitoneal and
subscapular injections. In the last instance the antigen was injected as an emul-
sion in Freunds’ adjuvant (Difco). This yielded the antisera with the highest
sperm agglutinin titers. Most of the rabbits received two or three courses of
injections.

Although anti-Arbacia sera from several rabbits were examined with uniform
results only experiments with antiserum from the hyper-immune rabbit “#33” are
recorded here. This rabbit was injected with antigen in Freunds’ adjuvant. The
animal was bled two weeks subsequent to the injection. The immune serum ob-
tained regularly agglutinated sperm to dilutions of 2° to 27°. No sera were
pooled.

98 KURT KOHLER AND CHARLES B. METZ

Tests, assays and absorptions. The sperm agglutination tests were performed
by mixing one volume (usually one drop) of 1% sperm suspension with two
volumes (2 drops or 0.1 ml.) of serum in depression slides. The salt concen-
tration of sera was adjusted to that of sea water by dialysis or addition of an equal
volume of 1.73 X sea water. Reactions were recorded after the mixtures had
remained for three hours at room temperature. Parallel tests with control (pre-
injection) serum were performed in all experiments to control for nonspecific
reactions.

The agglutination of Arbacia sperm with antiserum gave a regular pattern.
When diluted serially in sea water the antiserum gave titers with sharp, repro-
ducible end points. However, the morphology of the agglutinates varies upon
dilution of the antiserum. Antiserum at dilutions to 2-° yielded net-like aggluti-
nates involving head-to-head, head-to-tail and tail-to-tail agglutination. Agglutina-
tion at higher antiserum dilutions was head-to-head, the sperm was motile and
the agglutinates were “bouquet-like” in structure. Sperm in such dilute antisera
fertilized eggs and became more active on addition of fertilizin. This last effect
was associated with a breakdown of the antiserum agglutinates.

The absorptions were done with varying amounts of material. Absorbing
antigen and antibody were usually allowed to react for three hours at room tem-
perature. Absorption over a longer time did not result in lower titers.

Isolation of sperm tatls and heads. Isolated sperm tails and heads were used
in absorption and agglutination experiments. In order to separate sperm tails
and heads washed, living or formalin-killed sperm (25% suspension) was subjected
to the shaking action of the Mickle disintegrator for 20 minutes at room tempera-
ture. The resulting material was then centrifuged at low speed at 4° C. to sediment
the sperm heads. Suspensions of tails and of heads were washed when necessary
by high and low speed centrifugation, respectively, at 4° C.

RESULTS

In this investigation primary interest centers about the sperm “surface” as
opposed to “subsurface” or interior material with antigenic properties. Although
the sperm surface may be difficult to define or delimit at the molecular level, present
requirements are satisfied by an operational definition. According to this definition,
sperm surface antigens (specific combining sites) are those antigens so situated
that their combination with antibody under appropriate conditions results in sperm
agglutination. In any event information concerning the role of the sperm surface
in fertilization is the ultimate objective here and antigenic groups that are avail-
able for participation in the agglutination reaction should be sufficiently accessible
for interaction with the egg surface in the initial stages of fertilization.

In the experiments described below some attention is given to the antigenic
composition of sperm extracts. It is clear, that the “surface” vs. “subsurface@
origin of soluble antigens in sperm extracts is not readily established. For ex-
ample, a single antigenic group might be readily extracted from the cell “interior”
and at the same time exist in bound, unextractable and insoluble form at the cell
“surface.” However, removal of sperm agglutinating antibodies from antisera
by absorption with sperm extracts serves to identify and separate antibodies
against sperm surface antigens irrespective of the origin of the absorbing antigens.

ANTIGENS OF SEA URCHIN SPERM SURFACE 99

Such absorbed sera in turn can serve to separate and identify sperm surface
antigens. The experiments described below should be considered in this con-
ceptual framework.

A) Identification of sperm surface antigens by interspecific reactions

In order to identify sperm surface antigens, cross-absorption tests were per-
formed in interspecific, agglutinating combinations of sperm and antisera. Six
foreign sperms were tested for cross-reaction against antisera prepared against
sperm of four of the species. With the exception of the combination Mellita sperm

TABLE I

Sperm ageglutinating action of antisera before and after absorption with
sperm of various species

Immune sera

Sera Tested with
Prepared Abs.
against with A* 1b; E M Ie Asterias
sperm sperm
At — ~ + + 0 + 0
A A 0 0
A IY =e 0 0
A E =s aie 0
A E+L | + 0
L — | + = - + (weak) z + (weak)
is itp 0 0
is A 0 HS 0
i E =5 ae 0
rE. E+A 0 aia 0
E a == Se == 45 ie
E Pp 0 0 --
E A 0 a.
E yeaa a | 0 0 ?
Asterias — | 0 0 0 +

*A = Arbacia; E = Echinarachnius; L = Lytechinus; M = Mellita; P = Plagtobrissus.
Control (pre-injection) sera failed to agglutinate sperm in parallel tests to those in the table.

vs. Arbacia antiserum, cross-reactions were found among all echinoid combinations
tested. These relations are summarized in Table I.

Certain of these cross-reacting combinations were used in cross-absorption
experiments. These included the series Arbacia, Lytechinus and Echinarachnius
as seen in Table I. Absorption with the homologous sperm rendered antisera
non-agglutinating for the cross-reacting heterologous species. In every case com-
plete absorption with an heterologous sperm yielded a serum which still agglutinated
the homologous species. Furthermore, in most cases such absorbed sera aggluti-
nated the other heterologous species as well. Thus anti-Arbacia serum absorbed
with Echinarachnius sperm still agglutinated Lytechinus and Arbacia sperm.

100 KURT KOHLER AND CHARLES B. METZ

Evidently, then, Lytechinus and Arbacia sperm possess surface antigens lacking
on Echinarachnius. A further absorption of the Echinarachnius sperm absorbed
anti-Arbacia serum with Lytechinus sperm yielded a serum which failed to aggluti-
nate Lytechinus but still agglutinated Arbacia sperm. Accordingly, it is concluded
that the Arbacia sperm surface possesses at least three antigens or antigenic com-
plexes, one specific for Arbacia, a second shared with Lytechinus and a third
shared with Echinarachnius. It is of some interest that Lytechinus sperm absorbs
all Echinarachnius sperm agglutinins from anti-Arbacia sperm serum. This indi-
cates that all the “Arbacia antigens” present on the Echinarachnius sperm surface
are also present on the Lytechinus sperm. Comparable results were obtained in
absorption experiments using anti-Lytechinus sperm serum (Table I).

B) Sperm surface antigens in sperm extracts

In the previous section the Arbacia sperm surface was shown to possess at
least three antigens or antigenic complexes. In view of the possible role of
insoluble as well as soluble sperm surface substances in fertilization (see Metz,
1957a), it seemed of interest to examine sperm extracts for soluble antigens in
common with sperm surface antigens. This matter is of special interest because
it concerns the question whether the antifertilizin extracts from sperm contain the
full complement of sperm surface antigens. In one series of experiments soluble
antigen extracts were prepared by freeze-thawing Arbacia sperm. This is one
method for preparing sperm antifertilizin (Tyler, 1939). All extracts were subject
to low speed centrifugation (3000 x g for 20 minutes) to remove fragments of the
sperm surface. Certain preparations were also subjected to high speed centrifuga-
tion (26,000 X g) to remove the larger sub-cellular particulates. Constant
amounts of anti-Arbacia sperm serum were then absorbed with increasing amounts
of the sperm extracts. The absorbed sera were then assayed for sperm agglutinat-
ing action. As seen in Figures la, 3a, 4a, and 6a, increasing amounts of sperm
extract lowered the sperm agglutinin titer until a constant value was reached.
Beyond this point increasing amounts of extract fail to affect the sperm agglutinating
action. The extracts failed to remove all of the sperm agglutinins (seven experi-
ments). Accordingly, it appears that the extracts do not contain all of the
sperm surface antigens. Some of these evidently are not extractible in sea water
by freeze-thawing and do not appear as components in extracted “antifertilizin.”
Most of these antigens are heat-labile. If the frozen-thawed extract is heated a
precipitate begins to form at 55° C. and after further heating for one minute at
100° C. the supernatant has reduced antifertilizin activity and lowers the sperm
agglutinating titer of antiserum only one dilution step (compare Figure 1, a and b).

In view of the antibody absorbing action of extracts prepared by freeze-thawing
sperm, it seemed of interest to examine extracts prepared by other methods for
antigens in common with those present on the intact sperm surface. Accordingly,
extracts were prepared by heating (100° C. for one minute) followed by centrifuga-
tion, by disruption in the Mickle disintegrator, and by acid extraction.

As seen in Figure 2a (typical of two experiments), extracts prepared by heat-
ing had little if any effect on the sperm agglutinating action of anti-sperm serum.
Evidently, then, heat extraction yields few if any sperm surface antigens. How-
ever, antigenic material is present in extracts prepared by heating, for such extracts

ANTIGENS OF SEA-URCHIN SPERM SURFACE 101

produce one precipitin band in agar when diffused against antiserum (to be pub-
lished). Possibly this is sub-surface material. If so, it suggests that the anti-
fertilizin (egg jelly precipitating and fertilizin neutralizing) activity of extracts
prepared by heating (Frank, 1939) is not due to sperm surface material. Finally,
as seen above, frozen-thawed extracts contain insignificant amounts of surface anti-
genic material after heating to 100° C. Extracts prepared by action of the Mickle
disintegrator also neutralized the sperm agglutinating action of anti-whole sperm
serum. As seen in Figure 2c (two experiments), such absorption reduced the
agglutinin titer to a constant level, but failed to remove all agglutinins.

TEER

ML ANTIGEN

Ficure 1. Effect of absorption with increasing amounts of antigen on sperm agglutinating
action of anti-Arbacia sperm serum. Curve (a): Antiserum absorbed with frozen-thawed
sperm extract, tested on Arbacia sperm. Curve (b): Antiserum absorbed with heated (100° C.,
one minute) frozen-thawed sperm extract, tested on Arbacia sperm. Ordinate: Titer = — loge
of highest antiserum dilution giving sperm agglutination. Abscissa: ml. antigen added to
0.5 ml. antiserum.

Acid extraction of sperm also yields antifertilizin (Tyler and O’Melveny, 1941).
In one experiment anti-whole sperm serum was absorbed with increasing amounts
of neutralized pH 3.0 sperm extract. Figure 2b again shows reduction of the
agglutination titer to a constant level. In an additional comparative experiment
sperm suspensions (12.5%) were extracted at pH 3.2, 7.2 and 9.0 for six hours.
The extracts were subsequently tested for antibody absorbing action. The pH
7.2 (“aged sperm’’) extract lowered the sperm agglutination titer a maximum of
one-half (one-fold) whereas the acid and alkaline supernatants lowered the titer
of the antiserum a maximum of ¥ (three-fold). Thus, acid and alkaline extracts

102 KURT KOHLER AND CHARLES B. METZ

of sperm, like frozen-thawed and Mickle extracts, contain some but not all of the
antigenic groups present on the sperm surface.

Relation of antigens in sperm extracts to fertilizin. In view of the fact that
the soluble antigen preparations contain antifertilizin, it seemed of interest to ex-
amine the relations of such extracts and of anti-sperm serum to fertilizin. Specifi-
cally, it appeared of interest to test for action of fertilizin on the antiserum neutraliz-
ing action of soluble sperm antigen extracts (antifertilizin). To test for neutralizing

nek

O 0.5 10 1S 20
ML ANTIGEN

Figure 2. Effect of absorption with increasing amounts of antigen on sperm agglutinating
action of anti-Arbacia sperm serum. Curve (a): Absorbing antigen prepared by heating
(100° C., one minute) sperm. Curve (b): Absorbing antigen prepared by extracting sperm in
pH 3.0 sea water. Curve (c): Absorbing antigen prepared by Mickle disintegration of sperm.
Curve (d): Absorbing antigen: isolated sperm heads. Curve (e): Absorbing antigen: whole
intact sperm. Absorbing antigen for curves (a), (b) and (c) centrifuged 3000 X g. Each
curve represents a different experiment. Ordinate and abscissa as in Figure 1.

action, fertilizin and antiserum were mixed in varying proportions and the sperm
agglutinin titers determined. The experiment is complicated by the fact that
fertilizin and antiserum both agglutinate sperm. To avoid confusion from the
action of the two different agglutinins, the sperm was examined for agglutination
after sufficient time (three hours) to permit reversal of any agglutinating action
of fertilizin (see methods section). As seen in Figure 3b the sperm agglutinin
titer of the antiserum was not reduced by fertilizin. It appears, then, that the

ANTIGENS OF SEA URCHIN SPERM SURFACE 103

sperm surface and fertilizin do not have antigenic combining sites in common
which are essential for sperm agglutination.

On the other hand, the combining sites of the soluble antigens in frozen-thawed
extracts evidently are blocked by fertilizin. This is again seen in Figure 3,
curve c. In this experiment decreasing amounts of fertilizin were mixed with
increasing amounts of the sperm surface extract (antifertilizin). The mixtures
were subsequently used to absorb the anti-sperm serum. When the extract-to-
fertilizin ratio equalled 0.143 (0.25 ml. extract) the mixture failed to affect the

TITER

te) a) 1.0 1.5 2.0 ML ANTIGEN (curve a,b)
Oo fo 10 30 co 4 © d QUOTIENT = AF/F (curve c)

Ficure 3. Effect of absorption with increasing amounts of antigen on the sperm agglutinat-
ing action of anti-Arbacia sperm serum. Curve (a): Serum absorbed with sperm extract
(antifertilizin) prepared by freeze-thawing sperm and centrifuging at 3000 X g for 20 minutes.
Curve (b): Antiserum absorbed with fertilizin. Curve (c): Antiserum absorbed with anti-
fertilizin-fertilizin mixtures. Ordinate: Titer =— loge of highest antiserum dilution giving
sperm agglutination. Abscissa for (a) and (b): ml. antigen added to 0.5 ml. antiserum.
For (c): Quotient antifertilizin/fertilizin/0.5 ml. antiserum. Two ml. fertilizin and no anti-
fertilizin at quotient = 0. The same fertilizin preparation was used for curve b and c. Abso-
lute amounts of antifertilizin, antisera and total volumes in curves (a) and (c) are identical
along the abscissa.

sperm agglutinin titer of the anti-sperm serum, whereas the same amount of anti-
fertilizin alone reduced the agglutinin titer six-fold. Increase in the proportion of
antifertilizin resulted in mixtures with antiserum neutralizing action. This result
shows that fertilizin neutralizes the ability of the frozen-thawed extract to absorb
sperm agglutinins from the anti-sperm sera.

Distribution of “soluble” surface antigens. Data in the previous section show
that extraction of sperm yields solutions of soluble sperm antigens and that these
neutralize some of the antibodies (agglutinins) for the sperm surface. On the

104 KURT KOHLER AND CHARLES B. METZ

other hand the extracts fail to neutralize all of the sperm agglutinating antibodies.
Accordingly, the extraction procedures fail to remove some “insoluble” antigenic
material from the sperm. Although the antigens in the preparation may not be
extracted from the sperm surface they are antigenically related to, and therefore
serve to identify, surface antigens. In an attempt to localize these “soluble antigens”
on the sperm surface, antisera were treated with extracts prepared by freeze-thawing
(antifertilizin) and then tested for agglutinating action on suspensions of isolated
sperm heads, tails and intact sperm. As seen in Figure 4 such extracts completely
neutralized the agglutinins for sperm tails. Evidently, then, the sperm tail surface
possesses only soluble antigen. In this connection it is of interest that agar diffusion
precipitin tests show only a single band when frozen-thawed extract of isolated

8

ML ANTIGEN

Ficure 4. Agglutination of whole sperm and isolated sperm tails with anti-Arbacia whole
sperm serum absorbed with frozen-thawed whole sperm extract (antifertilizin). Curve (a):
Agglutination of whole sperm. Curve (b): Agglutination of isolated tails. Ordinate and
abscissa as in Figure 1.

sperm tails is diffused against anti-whole sperm serum. This single band merges
with one of four produced by whole sperm extract (to be published). These ob-
servations show that the insoluble antigens are confined to the sperm head. This
view is confirmed by examination for agglutination of whole sperm and heads by
the absorbed sera. As seen previously (Fig. 3a) the agglutinin titer for whole
sperm is reduced to a constant level by absorption with frozen-thawed extracts. On
the other hand, such extracts had no apparent effect on the ability of anti-sperm
serum to agglutinate isolated sperm heads. Evidently, the surfaces of the isolated
sperm heads and the frozen-thawed extracts do not have antigens in common and
the sperm heads, prepared by treatment in the Mickle disintegrator, lack (have
lost?) soluble surface antigenic material.

ANTIGENS OF SEA URCHIN SPERM SURFACE 105

Regional localization of sperm surface antigens. In a previous section it was
shown that whole sperm extract absorbed the agglutinins for isolated Arbacia sperm
tails but failed to absorb all agglutinins for isolated sperm heads. This result
indicates that the sperm tail antigens are soluble, whereas some or all of the antigens
on the sperm head surface are insoluble. It seemed of interest to extend this study
to a more direct analysis of differences between sperm heads and tails. To achieve
this the method used by Henle et al. (1938) for bull sperm was employed. Arbacia
sperm heads and tails were isolated by treatment in the Mickle disintegrator fol-

10
8
C
6
a
uJ
=
be
4
6 |
SNE
O 0.5 1.0 1.5 20
ML ANTIGEN

Ficure 5. Sperm agglutinating action of anti-Arbacia whole sperm serum after absorption
with sperm tails isolated from formalin-killed sperm. Curve (a): Tested on living Arbacia
sperm. Curve (b): Tested on sperm tails isolated from formalin-killed sperm. Curve (c):
Tested on sperm heads isolated from formalin-killed sperm. Ordinate and abscissa as in
Figure 1.

lowed by differential centrifugation. The suspensions of isolated heads and tails
were then used to absorb anti-whole sperm serum. Preliminary experiments showed
that isolated tails always reduce the sperm agglutination titer in absorption ex-
periments. Isolated sperm heads, however, showed varying results and require
further study. The inconsistencies may be related to the presence of very readily
soluble sperm surface antigens which appear in washings of the sperm.
Unfortunately, treatment of living sperm in the Mickle disintegrator resulted in
considerable fragmentation of the sperm tails. This was especially true in the

106 KURT KOHLER AND CHARLES B. METZ

preparation of large amounts of tails for absorption experiments. To facilitate
isolation of unbroken tails sperm were first fixed in 1% formalin, washed in saline
and the tails shaken from the remainder of the sperm in the Mickle disintegrator.

In using formalin-fixed sperm it is assumed that the formalin treatment does
not alter the antigenic structure of the sperm surface. This assumption is valid
to the extent that both living and formalin-killed whole sperm and sperm fragments
agglutinated to the same titer with anti-sperm serum and that living and formalin-
fixed material showed parallel absorbing action in preliminary experiments. For

8 e
BF ee
Wl 6 Se q
fe
od
4 e
2
@) W4 4 6 8 ML ANTIGEN (curve a,b)
(9) 50 100 I50 200 MG DRY WEIGHT (curve c,d)

FicureE 6. Sperm ageglutinating action of anti-Arbacia whole sperm serum absorbed with
Arbacia antifertilizin and sperm “ghosts.” Curve (a) and (b): Absorbing antigen, frozen-
thawed sperm extract. (a): Tested on Arbacia sperm. (b): Tested on Lytechinus sperm.
Abscissa: ml. antigen added to 0.2 ml. antiserum. Curve (c) and (d): Absorbing antigen,
frozen-dried “ghosts.” (c): Tested on Arbacia sperm. (d): Tested on Lytechinus sperm.
Abscissa: mg. dry weight. Ordinates for (a), (b), (c) and (d): Titer =— loge of highest
antiserum dilution giving sperm agglutination.

a final experiment washed sperm was fixed in formalin (1%), washed and treated
for 30 minutes in the Mickle disintegrator. The tails were then isolated from the
intact sperm and heads by centrifugation. Finally, the isolated tails were used in
varying amounts to absorb the anti-whole sperm serum. This absorbed serum
failed to agglutinate tails (Fig. 5b). However, it did agglutinate isolated sperm
heads (Fig. 5c) and living whole sperm (Fig. 5a), but in both instances the
absorbed serum showed lower titers than the unabsorbed serum. Evidently, then,
the sperm heads possess a head specific antigen or antigens which are not present
on sperm tails. The reduction in titer of sperm head agglutination by absorption

ANTIGENS OF SEA URCHIN SPERM SURFACE 107

with tails indicates that some tail antigenic material is also present of the sperm
heads.

Analysis of insoluble sperm surface antigens. In order to confirm the presence
of water-insoluble sperm surface antigens it seemed of interest to attempt to absorb
sera with sperm from which the soluble sperm antigens had been removed. Sperm
“ghosts” were prepared by freezing 25% sperm, removing the supernatant, re-
freezing the re-suspended residue and finally washing the insoluble material five
times in sea water. As seen in Figure 6a, the initial extract of soluble antigens
(antifertilizin) lowered the sperm agglutinin titer four-fold whereas the extracted
“ghosts” when used in excess lowered the titer seven-fold (Fig. 6c). Evidently,
then, the extracted “ghosts” possess insoluble antigens that are not present in the
soluble fraction.

Furthermore, it is of interest that these Arbacia “ghosts” also lowered the sperm
agglutinin titer for Lytechinus sperm (curve 6d), whereas the absorption with
frozen-thawed extract (Fig. 6b) did not affect the titer for Lytechinus. It remains
for future experiments to determine if all the antigens common to the two
species are insoluble, and if the ones in the extracts are the specific antigens.

Presence of sperm surface components in body fluid and egg extract. Results
from several experiments show that neither body fluid from male or female sea
urchins nor egg extract contains sperm surface components involved in antiserum
agglutination. Absorption of whole sperm antiserum with these extracts did not
lower, or lowered only one step, the titer for sperm agglutination.

DISCUSSION

In view of the striking morphology of the spermatozoan, the multiple antigen
structure of cell surface generally (e.g. erythrocytes) and the presence of at least
three distinct surface antigens on bull sperm (Henle ef al., 1938), and two on
Strongylocentrotus sperm (Tyler, 1949), it is not surprising to find several antigens
on the Arbacia sperm surface. The distinct head and tail antigens found on Arbacia
sperm conform to the pattern found for bull sperm. It seems likely that tests of a
wider range of heterologous species would reveal more cross-agglutination reactions
between foreign sperm (e.g. Tyler, 1949) and the anti-Arbacia sperm serum.
Absorption of the antiserum with such foreign sperm might reveal more antigens
in addition to the three so far demonstrated on the Arbacia sperm surface. Further-
more, a more detailed analysis with such absorbed sera, such as testing for agglutinat-
ing action on isolated sperm heads and tails, should localize the individual antigens
or antigenic complexes.

The number and distribution of the sperm surface antigens is not without interest,
for this information contributes to the knowledge of sperm surface architecture.
Furthermore, such antigens could be used as additional points of attack in studies
on the role of the sperm surface in fertilization. However, the relationship of the
sperm surface antigens to antigens present in sperm extracts appears to be of more
immediate interest.

Extracts of sea urchin sperm prepared by heating (Frank, 1939), freeze-thawing
(Tyler, 1939) and acid extraction (Tyler and O’Melveny, 1941) contain an agent
or agents called antifertilizin, which neutralizes the sperm agglutinating action of
fertilizin, agglutinates eggs and precipitates the egg jelly. This antifertilizin in

108 KURT KOHLER AND CHARLES B. METZ

sperm extracts is assumed to come from the sperm surface and to include the
specific sperm surface combining sites with which fertilizin reacts when it reversibly
agglutinates sperm.

In support of this view Tyler (1949) extracted antifertilizin from the Strongylo-
centrotus sperm by acid treatment. Subsequent examination of the cells with the
electron microscope revealed a partial breakdown of the sperm head surface, sug-
gesting this as the site of origin of the antifertilizin. However, the situation is
probably more complicated than this, for as seen in the present study absorption
of anti-whole sperm serum with sperm extracts (antifertilizin) removes tail ag-
glutinins but has little effect on the head agglutinins. This suggests that the
antifertilizin may be present in large amounts on the sperm tail.

Furthermore, agar gel precipitin tests show that the antifertilizin extracts pre-
pared by freeze-thawing regularly contain three to four antigens (to be published).
The surface vs. sub-surface origin of these antigens is not immediately clear. How-
ever, the fact that this mixture of antigens will remove some sperm agglutinins from
anti-sperm serum suggests that at least one of the antigens is of sperm surface
origin. But even this experiment does not exclude the possibility that the absorbing
antigen(s) possesses a combining group 1n common with the sperm surface but is
actually extracted from the interior of the sperm.

In further attempts to identify these antigens the antigen neutralizing action of
fertilizin was examined. Following treatment with fertilizin, antifertilizin failed
to lower the sperm agglutination titer of the antiserum. Clearly, then, fertilizin can
neutralize all the sperm surface antigens in the frozen-thawed extract. The fertilizin
may be highly specific in action and neutralize only one of the four antigens. If
this is the case, it follows that the extract prepared by freeze-thawing contains only
one antigen in common with the sperm surface. In any event, the antigen or
antigens involved appear to be related mainly to the antigens of the sperm tail
(see above).

Studies on the role of specific egg and sperm substances in metazoan fertilization
have been concerned mainly with soluble agents. It is of some interest, then, to
find that extracts of sperm do not contain all of the surface antigens. Evidently,
some antigens are either destroyed by the extraction procedure or remain in the
insoluble residue. The second alternative seems the more likely in view of the
nature of the extraction procedure. Indeed the absorption experiments with sperm
“ghosts” show that insoluble antigens survive the extraction procedure. The role
in fertilization of such insoluble sperm surface antigens remains to be investigated.
The specific combining groups on the sperm surface that function in the antiserum
agglutination reaction may not be identical with molecular configurations that func-
tion in fertilization. Nevertheless it seems likely that such material may form a
part of, or be spatially related to, molecular surface patterns that do function in the
union of sperm and egg. This apparently is the case in Paramecium (Metz, 1954).

Consideration here has been given only to the sperm head and tail. A more
detailed study should take into account the subdivisions of these structures, such as
the axial tip of the tail, and the midpiece and acrosome of the head. Furthermore,
especial attention should be given to comparison of the antigens of the sperm
before and after acrosomal filament discharge, for at least in some forms (e.g.
starfish) activation of the egg results from interaction of the egg surface with the tip

ANTIGENS OF SEA URCHIN SPERM SURFACE 109

of the acrosomal filament (see Metz, 1957b; Colwin and Colwin, 1957; Collier,
1959; Dan, 1956).

SUMMARY

1. Sperm surface antigens are defined as those antigens which are detected by
sperm agglutinating action of antisera.

2. Interspecific sperm agglutination tests show one or more sperm surface
antigens common to all of the five echinoids tested except for a single combination.

3. The Arbacia sperm surface possesses a minimum of three antigens or antigenic
complexes in common with other species. One of these is shared with Lytechinus
sperm, a second with both Lytechinus and Echinarachnius and a third is not present
on sperm of these species.

4. Arbacia sperm extracts prepared by freeze-thawing, Mickle disintegration or
pH 3 treatment all contain “soluble” antigens in common with the sperm surface.
The Arbacia sperm surface also possesses “insoluble” antigens which do not appear
in the above extracts.

5. The “soluble sperm surface” antigen(s) in extracts is neutralized by fertilizin
from Arbacia eggs.

6. It is present on the sperm tail surface and possibly to some extent on the
sperm head.

7. The insoluble sperm surface antigen(s) is confined to the sperm head.

LITERATURE CITED

Coriier, J. R., 1959. The effect of homologous fertilizin on the sperm of Strongylocentrotus
; purpuratus. Acta Embryol. Morphol. Exp., 2: 163-170.

Corwin, A. L., ann L. H. Corwin, 1957. Morphology of fertilization: Acrosome filament
formation and sperm entry. In: The Beginnings of Embryonic Development. A.
Tyler, R. C. von Borstel, C. B. Metz, editors. Amer. Assoc. Adv. Sci. Washington,
D. C., pp. 135-168.

DatiAM, R. D., Aanp L. E. THomas, 1953. Chemical studies on mammalian sperm. Biochim.
Biophys. Acta, 11: 79-89.

Dan, J. C., 1956. The acrosome reaction. Intern. Rev. Cytol., 5: 365-393.

FRANK, J. A., 1939. Some properties of sperm extracts and their relationship to the fertiliza-
tion reaction in Arbacia punctulata. Biol. Bull., 76: 190-216.

Harvey, E. B., 1956. The American Arbacia and Other Sea Urchins. Princeton Univ. Press,
Princeton, N. J.

HeniLe, W., 1938. The specificity of some mammalian spermatozoa. J. Immunology, 34:
325-336.

HENLE, W., G. HENLE AND L. A. CHAmBERS, 1938. Studies on the antigenic structure of
some mammalian spermatozoa. J. Exp. Med., 68: 335-352.

Liriuiz, F. R., 1913. The mechanism of fertilization. Science, 38: 524-528.

Livi, F. R., 1914. Studies on fertilization. VI. The mechanism of fertilization in Arbacia.
J. Exp. Zool., 16: 523-590.

Metz, C. B., 1954. Mating substances and the physiology of fertilization in ciliates. In: Sex
in Microorganisms. D. H. Wenrich, editor. Amer. Assoc. Adv. Sci. Washington,
D. C., pp. 284-334.

Metz, C. B., 1957a. Specific egg and sperm substances and activation of the egg. Jn: The
Beginnings of Embryonic Development. A. Tyler, R. C. von Borstel, C. B. Metz,
editors. Amer. Assoc. Adv. Sci. Washington, D. C., pp. 23-69.

Metz, C. B., 1957b. Mechanisms in Fertilization. In: Physiological Triggers and Discon-
tinuous Rate Processes. T. H. Bullock, editor. Amer. Assoc. Adv. Sci. Washing-
ton, D. C., pp. 17-45.

110 KURT KOHLER AND CHARLES B. METZ

Metz, C. B., 1959. Inhibition of fertilizin agglutination of sperm by the dermal secretion from
Arbacia. Biol. Bull., 116: 472-483.

Mupp, S., AnD E. B. H. Munp, 1929. The specificity of mammalian spermatozoa with especial
reference to electrophoresis as a means of serological differentiation. J. Immunology,
17: 39-52.

PERLMANN, P., 1957. Analysis of the surface structures of the sea urchin egg by means of
antibodies. I. Comparative study of the effects of various sera. Exp. Cell Res., 13:
365-390.

PERLMANN, P., AND H. PERLMANN, 1957a. Analysis of the surface structures of the sea
urchin egg by means of antibodies. II. The J- and A-antigens. Exp. Cell Res., 13:
454-474.

PERLMANN, P., AND H. PEerrmMANN, 1957b. Analysis of the surface structures of the sea
urchin egg by means of antibodies. III. The C- and F-antigens. Exp. Cell Res., 13:
475-487.

Pernot, E., 1957. Recherches sur les constituants antigeniques des spermatozoides de cobayes:
Bull. Soc. Clin. Biol., 38: 1041-1054.

RuNNstTROM, J., B. E. HAGstROM AND P. PERLMANN, 1959. Fertilization. In: The Cell.
Biochemistry, Physiology, Morphology. Vol. I, edited by J. Brachet and A. E. Mirsky.
Academic Press, New York and London, pp. 327-398.

SmitH, A. U., 1949. Some antigenic properties of mammalian spermatozoa. Proc. Roy. Soc.
London, Ser. B, 136: 46-66.

SNELL, G. D., 1944. Antigenic differences between the sperm of different inbred strains of
mice. Science, 100: 272-273.

Ty ver, A., 1939. Extraction of an egg membrane lysin from sperm of the giant keyhole
limpet (Megathura crenulata). Proc. Nat. Acad. Sci., 25: 317-323.

Ty ter, A., 1946. Loss of fertilizing power of sea urchin and Urechis sperm treated with
“univalent” antibodies vs. antifertilizin. Proc. Soc. Exp. Biol. Med., 62: 197-199.

Tyier, A., 1948. Fertilization and immunity. Physiol. Rev., 28: 180-219.

Ty er, A., 1949. Properties of fertilizin and related saineanees of eggs and sperm of marine
aa Amer. Nat., 83: 195-219.

Ty er, A., AND J. W. BROOKBANK, 1956a. Antisera that block cell division in developing eggs
of sea urchins. Proc. Nat. Acad. Sci., 42: 304-308.

Ty er, A., AND J. W. BrooKBANk, 1956b. Inhibition of division and development of sea urchin
eggs by antisera against fertilizin. Proc. Nat. Acad. Sci., 42: 308-313.

Tyier, A., AND K. O’Metveny, 1941. The role of antifertilizin in the fertilization of sea
urchin eggs. Biol. Bull., 81: 364-374.

eee enor OF EFFERENT NERVE ACTIVITY IN THE ROACH,
PeRiPLANETA AMERICANA, BY EXTRACTS OF THE
CORPUS CARDIACUM*

NANCY MILBURN, ELIZABETH A. WEIANT AND KENNETH D. ROEDER?
Department of Biology, Tufts University, Medford 55, Massachusetts

Several attempts have been made to correlate patterns of endogenous activity
as observed in isolated portions of invertebrate nervous systems with the normal
behavior of the intact animal (von Holst, 1934; Roeder, 1955; and Harker, 1956 and
1958). Roeder, Tozian and Weiant (1960) found that decapitation of roaches
or of mantids markedly increases the level of endogenous activity in the efferent
nerves from the cercal ganglion which innervate the phallic musculature and in the
efferent nerves from the metathoracic ganglion which innervate the thoracic muscu-
lature. As the activity in the phallic nerves increases, rhythmic bursts of nerve
spikes are seen. These motor bursts on the several nerve branches leading to the
phallomeres are not obviously coordinated, but they result in ordered, rhythmic
movements of the genitalia. The behavioral significance of this endogenous nerve
activity is suggested by the observation that in male mantids the copulating re-
flexes appear with great intensity following decapitation. Indeed, under natural
conditions, the male is often beheaded by the female before he mates with her
(Roeder, 1935). Control of this sexual behavior and of the efferent nerve activity
in the nerves to the phallomeres and to the thoracic musculature appears to be
effected through the agency of an inhibiting center in the subesophageal ganglion.

In most of these experiments (Roeder, Tozian and Weiant, 1960) there was a
time lag of five to fifteen minutes between the removal of the head and the ap-
pearance of bursts of spikes in the efferent nerves. This time lag might result from
a decreasing discharge of severed inhibitor neurones, augmented by injury potentials.
On the other hand, it might represent the time necessary for the inactivation or
removal of an inhibiting neurohormone. Such a material, if present, could be
transported by proximo-distal flow down axons within the nerve cord to the site of
its action.

A variety of substances are to be found in roaches which affect their behavior
and some of these may be produced within the central nervous system itself
(Beament, 1958; Colhoun, 1958, 1959; Smyth, 1959; and Sternberg and Kearns,
1952). The work of Ozbas and Hodgson (1958) showed that the injection of
corpus cardiacum extracts produced stereotyped behavior in roaches and that these
extracts had a pronounced inhibitory effect on the activity in isolated roach abdom-
inal nerve cords. Their results led us to attempt to block the inhibitory influences

1 The experimental work reported in this paper was made possible by Graduate Training
Grant 2E32 and Research Grant E497 from the United States Public Health Service.

* The authors wish to thank Drs. L. M. Roth, E. R. Willis and B. Stay, Pioneering Research
Division, Army Quartermaster Research and Engineering Command, Natick, Mass. who assisted
in the procuring of many of the species of roaches used in these experiments.

111

112 N. MILBURN, EAD WEIANT AND KD. ROEDER

impinging on the efferent nerve cells in the cercal and metathoracic ganglia by the
use of similar extracts.

MATERIALS AND METHODS

Most of the observations of activity in efferent nerves to the phallomeres and
in the connectives of the abdominal nerve cord were made using male Periplaneta
americana as test animals. For these experiments the animals’ wings and legs
were removed and the bodies were pinned dorsal side down on a cork. Most of the
abdominal contents were removed. Nerve IX or nerve X from the cercal ganglion
(Fig. 1, E?) was picked up on a small tapered silver wire hook electrode and allowed
partially to dry. In many experiments afferent impulses in the nerve and the muscle
potentials resulting from the efferent activity were eliminated by cutting the nerve
distal to the recording electrode. The indifferent electrode was a chlorided silver
wire touching the inner abdominal wall. Simultaneous recordings of nerve activity
in the connectives of the abdominal cord were made by raising a connective on a
pair of silver hook electrodes (Fig. 1, Et). The input from the phallic nerve
fibers was amplified with a Grass P6 preamplifier and observed on a Dumont 304A
oscilloscope. The amplified signals were also recorded on one channel of a Grass
Polygraph inkwriter using a 5P3 integrator preamplifier. The activity from the
abdominal connectives was amplified by a 5P3 preamplifier and appeared on the
other channel of the inkwriter. The two inputs could be transposed at any time
so that visual observations or photographs could be made of either the nerve im-
pulses from the cord or of those from the phallic efferent nerves. Meanwhile, a
continuous record of the integrated activity from both sources appeared on the
inkwriter.

Endogenous efferent nerve activity in the thoracic region was sampled from
an electrode pair placed under nerve’ 3 (Fig. 1°") "trom the leit cide saneume
metathoracic ganglion. The method has been described previously (Weiant, 1958).
All other nerves arising from the thoracic ganglia were severed, eliminating all
sensory input from these segments. The longitudinal connectives between the
thoracic ganglia were left intact, as were the connections of the thoracic cord with
the head ganglia and abdominal ganglia. Peripheral connections of extra-thoracic
regions of the nervous system were also left intact. Activity of the internuncial
fibers within the thoracic connectives was monitored simultaneously with that of
nerve 3 by placing a pair of electrodes under the thoracic connectives above and
below the prothoracic ganglion (Fig. 1, E*).

Many of the same extract samples were tested for their action both on the
phallic motor nerves and on the thoracic efferent fibers. In experiments with both
of these preparations, injections into the head capsule of 0.01 to 0.02 ml. of extract
were made as close to the subesophageal ganglion as possible. In some tests with
the thoracic preparation a drop of extract was applied directly on the metathoracic
ganglion; in other experiments using the abdominal preparation, several drops of
extract were used to bathe the last three abdominal ganglia with their connectives.
Some tests were conducted on Periplaneta from which the corpora cardiaca had
previously been removed.

Extracts were made from the corpora cardiaca of the following species of
roaches: Blaberus cranitfer, Blaberus giganteus, Periplaneta americana, Leucophaea

NERVE ACTIVITY AND CORPUS CARDIACUM 113

maderae, Byrsotria fumigata, and Diploptera punctata. Extracts were from the
glands of adult animals except in one case in which an extract was made from the
corpora cardiaca of penultimate stage nymphs of Periplaneta.

The paired corpora cardiaca were excised according to the technique of Ozbas
and Hodgson (1958) and placed on the end of a small glass pestle. The glands
were removed as quickly as possible after the donor roaches had been caught in
order to minimize the loss of active material due to excitation of the roach (Hodgson,
personal communication). When the requisite number of corpora cardiaca had been
collected, they were triturated ina small mortar. Either Hoyle’s or Pringle’s insect
saline was added from a hypodermic syringe to dilute the extract to the concentra-
tion desired. Pringle’s proved to be the more satisfactory saline for maintaining the
nerve activity in the preparations.

ec!

Figure 1. Diagram of the central nervous system of Periplaneta americana showing
placement of electrodes.

-

In many of the experiments the mortar containing the extract was heated at
95° to 100° C. for five minutes. A coagulum formed in the bottom of the mortar,
leaving a fairly clear solution which was used for the tests. Longer exposures to
heat seemed to impair the activity of the extract, but a brief period of heating left
the potency unchanged. Unheated extracts lost their activity somewhat after four
to five hours at room temperature or after about ten hours at 12° to 14° C. Heat-
ing approximately doubled the time that the extracts would remain active.

Four experiments were conducted with extracts which were quick-frozen on
blocks of dry ice and kept frozen for periods of twenty-four hours to a week. Two
of these extracts were heated before being frozen. None of the quick-frozen
extracts showed any loss of activity. This seems a promising method for preserving
such material.

114 N. MILBURN, E. A. WEIANT AND K. D. ROEDER

RESULTS

Efferent nerve activity of the last abdominal ganglion. ‘The effects of de-
capitation or of nerve cord transection on efferent fiber activity of the last abdominal
ganglion have been described in detail elsewhere (Roeder, Tozian and Weiant,
1960). Ina roach with the nervous system intact, spike potentials are observed
in only a few efferent fibers of the nerves supplying the phallic musculature. In
those fibers which are active, spikes recur regularly and at a relatively low frequency.
If the animal is decapitated, there is a brief injury discharge followed by a return
to the previous level of activity. After five to ten minutes the number of active
fibers in the phallic nerve begins to increase, and a slowly-rising crescendo of
activity is observed due to a steady increase in the spike frequency in both newly—
and previously— active fibers. Typically this increased efferent discharge gradually
becomes patterned into a sequence of impulse bursts. At first the bursts are isolated,
but they soon become rhythmic and increase steadily in frequency for the next

Ficure 2. A. Oscilloscope record of impulses on the phallic nerve of a normal cockroach,
followed by a polygraph record showing the integrated electrical activity from the same
nerve. B-E. Oscilloscope records from the same nerve. The roach had been decapitated
twenty minutes before the record was made. Note the increased activity. F. Polygraph record
of integrated nerve impulses from the phallic nerve of the same roach taken immediately after
B-E. The film speed for the oscilloscope record was 25 cm./sec.; the polygraph tape moved
2.5 mm./sec.

thirty minutes (Fig. 2). Individual efferent fibers appear to follow their own
burst frequency, giving a rhythmic and often complex pattern to the activity in the
compound nerve. In most experiments after thirty minutes the burst pattern
usually reaches its full development and remains constant for some hours. Oc-
casional silent periods have been observed, however.

When an active extract of corpora cardiaca is applied to the nerve cord or
injected into the head of a roach with an intact nerve cord, the effects of nerve cord
transection or decapitation are mimicked (Fig. 3; compare with Fig. 3C, Roeder,
Tozian and Weiant, 1960). The sequence of changes in the phallic nerve activity
follows the same time course in the gradual development of bursts, and the magnitude
of the response is very similar to that released by cord transection. The initial time
lag when extract is applied is similar to that seen with decapitation. Desheathing
portions of the abdominal ganglia and cord does not decrease this interval. The

NERVE ACTIVITY AND CORPUS CARDIACUM tS

only consistent difference is that activity produced by the corpus cardiacum extract
is temporary, declining after about one hour. The bursting rhythm becomes ir-
regular and the activity slowly dies down to its original level. If extract is reapplied
to the cord every twenty minutes or so, the rhythmic efferent bursts may
be maintained for several hours. These effects were observed in about sixty
preparations.

Mg
nN LOL Sone) InAitioliiin ree sliuiny Gary sgh ue Scant Sito

1 lll A AU ATMA

Ficure 3. Polygraph records showing the integrated phallic nerve activity from the fol-
lowing roaches: A and B. Normal cockroaches. C. Roach with abdominal nerve cord exposed
to corpus cardiacum extract, % pair/0.01 ml., for 30 minutes (concentration below threshold). D.
Roach with abdominal nerve cord exposed for 25 minutes to corpus cardiacum extract, 3
pairs/0.01 ml. from penultimate stage nymphs. E. Roach with abdominal nerve cord exposed to
corpus cardiacum extract, 1 pair/0.01 ml., for 30 minutes. F. Roach with abdominal nerve cord
exposed to corpus cardiacum extract, 2 pairs/0.01 ml. for seven minutes and (G) for 30 minutes.
H. Roach which had had 3 pairs of corpora cardiaca in 0.02 ml. of saline injected into its head
capsule 30 minutes before. JI. Roach which had had 4 pairs of corpora cardiaca in 0.02 ml. of
saline injected into its head capsule 35 minutes before.

The effects of extract concentration. The concentration of corpus cardiacum
material necessary to produce the efferent nerve bursts varies both with the size and
the species of the donor roaches and with the sensitivity of the test preparation.
In general, the larger the adult size of the species, the more active were the extracts
of its glands. In terms of their activity when tested on Periplaneta, extracts of
Blaberus giganteus and Blaberus cranufer were about equally potent, while the
glands of other species ranged down in activity in the following order: Leucophaea
maderae, Periplaneta americana, Byrsotria fumigata, and Diploptera punctata.

116 N. MILBURN, E. A. WEIANT AND K. D. ROEDER

The extract is always diluted by the blood of the roach so that quantitative in-
formation is difficult to obtain. An extract made from three or four pairs of
corpora cardiaca diluted with 0.01 or 0.02 ml. of saline and injected into the head
capsule produces pronounced increases in efferent fiber activity in the phallic nerves
followed by rhythmic bursts of spikes. Extracts containing one or more pairs of
corpora cardiaca per 0.01 ml. of saline when applied to the abdominal cord have
similar effects (Fig. 3).

Control extracts were made from abdominal nerve cord tissue and applied in
the same manner to the head ganglia and abdominal nerve cords of test roaches.
Abdominal cord extract caused occasional transient increases in the activity of
both the abdominal cord and the efferent nerves, but the regular and continuing
efferent nerve spike bursts were never seen.

In experiments employing more concentrated extracts, containing up to three
and a half pairs of corpora cardiaca per 0.01 ml., the initial time lag was unaffected,
but the level of phallic motor nerve activity rose more rapidly than with less con-
centrated extracts and the final frequency attained by the efferent nerve spike
bursts was higher. These rapid bursts, when integrated, appeared on the polygraph
with a lower amplitude and a shorter duration than the slower-frequency bursts
obtained with the less concentrated extracts (Fig. 3, F and G). The burst fre-
quency about thirty-five minutes after the first application of extract provides a
rough measure of the potency of the extract.

In four out of seven experiments in which concentrated extracts containing
three and one-half pairs of glands per 0.01 ml. were applied to the abdominal cord,
the phallic motor nerve activity was blocked after twenty to thirty minutes. This
block was only partially reversible by washing. In fact, all attempts to wash out
the effects of the corpus cardiacum extracts met with little success if the extract had
been on the preparation for fifteen minutes or more. This may have been due in
part to the difficulty of irrigating these preparations.

Activity in the abdominal nerve cord. The experiments of Ozbas and Hodgson
(1958) showed that nerve impulses in isolated roach abdominal nerve cords were
blocked when the cords were immersed in extracts of corpora cardiaca. We have
confirmed these observations and have also noticed a short period of excitation
which seems to precede the block in such isolated cords. It has proved extremely
difficult to block nerve impulses in abdominal cords with these extracts if the cords
remain im situ with the cercal afferents intact and the head and thoracic portions
of the cord still attached. When applied to a preparation in which the central
nervous system is almost undamaged, the extracts usually cause a brief increase
in the level of activity in the abdominal connectives. A tendency for some nerve
cells to fire short trains of impulses becomes apparent after the extract has been on
the cord twenty to thirty minutes. A period of depressed activity sometimes fol-
lows the period of excitation.

Simultaneous recordings made from abdominal connectives and from the phallic
nerve do not reveal reciprocal changes in the levels of activity of the cord con-
nectives and the motor nerves after decapitation or after application of extracts
containing one to three pairs of corpora cardiaca per 0.01 ml. As the efferent
nerve activity increases and bursts of nerve spikes begin to appear on the phallic
nerves, the cord activity increases for a time and then usually returns to its previous

NERVE ACTIVITY AND CORPUS CARDIACUM 117

level. The cord appears depressed only when concentrations of extracts higher
than three pairs per 0.01 ml. were used.

Nerve activity in the thoracic region. The changes induced by corpus car-
diacum extracts in the activity of nerve 3 from the metathoracic ganglion and in
the thoracic connectives were somewhat more complex than those obtained in the
abdominal region. Of twenty-three experiments, fourteen showed increased ac-
tivity followed by bursts similar to those observed in the phallic nerves. The
change occurred twelve to twenty minutes after the extract was applied. In four
cases the activity decreased and in four the efferent impulses were blocked. In
these cases where the extract caused a temporary or partial block in cord and
efferent nerves, twenty minutes were allowed for recovery, but the activity did not
return to normal. However, if these roaches were then decapitated, there was
an increase in the frequency of the active fibers in both the thoracic cord and
nerve 3, and in one case there was also an increase in the number of fibers firing
in nerve 3.

In one experiment there appeared to be no change in activity. A temporary
decline or cessation of efferent activity was also noted at one period in four of
the thirteen preparations which showed an over-all increase in activity and bursts.
These changes induced by corpus cardiacum extract correspond closely with those
following decapitation or transection of the thoracic connectives (Weiant, 1958).

Changes in activity recorded from electrodes placed on the nerve cord followed
very closely the changes in nerve 3 described above. It seems probable that the
increased efferent activity was picked up by the cord electrodes, and that this
would have masked any decreased activity that might have occurred in the inter-
nuncial fibers within the connectives.

In a few cases where the extract reached the exposed thoracic tissues, violent
tremors developed in the flight muscles. Since all motor branches from the
thoracic ganglia were severed, this suggests that the extract has a direct action
on the thoracic muscles.

Abdominal pulsations often coincided with and obscured the thoracic nerve
activity. If the abdomen was cut off at this time, there was some decline in the
activity in nerve 3. However, this operation rarely brought the activity down to
the level recorded previous to the application of extract. If severing the abdomen
was followed by decapitation, a slight increase in the frequency of the activity in
nerve 3 occurred.

Synaptic conduction. Six experiments were performed using the classic cercal
nerve-giant fiber preparation (Roeder, Kennedy and Samson, 1947) to determine
the effects of corpus cardiacum extracts on a sensory synapse. [xtracts containing
one to three pairs of glands per 0.01 ml. were applied or injected and left on the
preparation for thirty to forty minutes. The only change which was observed
was a slight lowering of the electrical threshold of the synapse in every case.

Comparison with DDT-toxin. Sternberg and Kearns (1952) have reported
the presence of an excitor substance liberated from the nervous system of roaches
poisoned with DDT. Four attempts have been made to collect this substance
using a technique employed by Sternberg. The abdominal cord activity increased
greatly after exposure to the saline containing the substance and many impulse
trains were seen. No rhythmic spike bursts were observed on the phallic nerves,

118 Ni MILBURN Eo AY WELANT AN DUK? Ds ROE DIG:

however. It was concluded that the DDT-toxin was not the same as the principle
in the corpus cardiacum extracts which causes phallic efferent nerve bursts.

CONCLUSIONS

The foregoing experiments demonstrate that, except for its reversible action,
the extract of corpora cardiaca taken from several species of roaches mimics, in
the American roach, removal of the inhibitory system as postulated by Roeder,
Tozian and Weiant (1960). The simplest interpretation is that the corpus
cardiacum extract suppresses the action of an inhibitory system, there being no
evidence that it has any action on excitatory synapses.

If this is the case, a decrease in the activity of those fibers within the nerve cord
that mediate the inhibition must occur in the presence of extracts of corpora
cardiaca. This could not be detected by electrodes placed in contact with the
outside of the nerve cord, although in the abdominal preparations the extract-
induced increases in cord activity were transitory and insignificant compared with
those occurring in the efferent fibers. Perhaps this is to be expected since the
fibers comprising the inhibitory system may occupy only a small portion of the
cord substance (Hess, 1958), and from the aspect of an external electrode they
are probably shunted by many other fibers either unaffected by, or even released
by, the blockage of the inhibitory system.

One perplexing observation in the earlier work (Roeder, Tozian and Weiant,
1960) was that the descending inhibitory system could not be blocked by KCl or by
anelectrotonus, the only way of blocking it at that time being by cord transection.
To this may be added the current observation that the inhibitory system may be
inactivated not only by applying corpus cardiacum extract to its center or point of
origin in the subesophageal ganglion, but also anywhere along its path from the
head to the locally-originating neurones. It is difficult to visualize a classical
nerve pathway operating in this manner, although no alternative mechanism can
yet be proposed.

SUMMARY

1. Increased activity and regular bursts of efferent nerve spikes are seen on
nerves IX and X from the cercal ganglion of roaches when extracts of corpora
cardiaca are applied to the exposed abdominal cord or injected into the head
capsule. Such motor nerve activity is usually inhibited in the intact animal by
the presence of some influence arising in the subesophageal ganglion (Roeder,
Tozian and Weiant, 1960). The corpus cardiacum extracts appear to suppress
the action of this inhibitory center. The threshold concentration for this effect is
approximately one pair of corpora cardiaca per 0.01 ml. of saline when the extract
is applied to the abdominal cord, or three pairs injected into the head in about
0.01 to 0.02 ml. of saline. In decapitated male mantids a similar type of nerve
fiber activity is believed to result in behavior associated with mating.

2. The corpus cardiacum extracts seem temporarily to increase the activity
of the abdominal nerve cord in preparations with an intact central nervous system.
When high concentrations of extract are used, the cord activity is occasionally
depressed or blocked. The effects of the extract are only partially reversible by
washing the preparation with saline.

NERVE ACTIVITY AND CORPUS CARDIACUM 119

3. The activity of certain efferent fibers in the thoracic region is similarly af-
fected by corpus cardiacum extract, the action mimicking that produced by de-
capitation or cord transection.

4. Corpus cardiacum extracts have no significant effects on the transmission
of impulses at the synapse between the cercal nerve and the giant fibers in an
electrically stimulated preparation.

5. The DDT-toxin of Sternberg does not have the same effect on the roach
nervous system as the active principle in extracts of corpora cardiaca.

6. The corpus cardiacum extracts may be preserved successfully by quick-
freezing. Their activity is decreased by prolonged heating, but warming for five
minutes at 90° to 100° C. leaves the potency unaffected. Such warming seems to
increase by several hours the length of time that an extract will remain potent.

Pini hURE, GUEnD

BEAMENT, J. W. L., 1958. <A paralysing agent in the blood of cockroaches. J. Ins. Physiol.,
2: 199-214.

CotHoun, E. H., 1958. Acetylcholine in Periplaneta americana L. II. Acetylcholine and
nervous activity. J. Ins. Physiol., 2: 117-127.

CotHoun, E. H., 1959. Acetylcholine in Periplaneta americana L. III. Acetylcholine in
roaches treated with tetraethyl pyrophosphate and 2,2-bis(p-chlorophenyl) -1,1,1-tri-
chloroethane. Canad. J. Biochem. Physiol., 37: 259-272.

Harker, J., 1956. Factors controlling the diurnal rhythm of activity of Periplaneta americana
L. J. Exp. Biol., 33: 224-234.

Harker, J., 1958. Experimental production of midgut tumors in Pertplaneta americana L.
J. Exp. Biol., 35: 251-259.

Hess, A., 1958. Experimental anatomical studies of pathways in the severed central nerve cord
of the cockroach. J. Morph., 103: 479-502.

von Horst, E., 1934. Motorische und tonische Erregung und ihr Bahnenverlauf bei Lepidop-
teranlarven. Zeitschr. f. vergleich. Physiol., 21: 395-414.

Ozpas, S., AND E. S. Hopcson, 1958. Action of insect neurosecretion upon central nervous
system in vitro and upon behavior. Proc. Nat. Acad. Sct., 44: 825-830.

Roeper, K. D., 1935. An experimental analysis of the sexual behavior of the praying mantis
(Mantis religiosa L.). Biol. Bull., 69: 203-220.

Roeper, K. D., 1955. Spontaneous activity and behavior. Sci. Monthly, 80: 362-370.

Roeper, K. D., N. KENNEDY AND E. A. Samson, 1947. Synaptic conduction in the giant fibers
of the cockroach and the action of anticholinesterases. J. Neurophysiol., 10: 1-16.

Roeper, K. D., L. Tozian anp E. A. WEIAnNT, 1960. Endogenous nerve activity and behavior
in the mantis and cockroach. J. Ins. Physiol. (in press).

STERNBERG, J., AND C. W. Kearns, 1952. The presence of toxins other than DDT in the
blood of DDT-poisoned roaches. Science, 116: 144-147.

SmyTH, T., Jr. 1959. Polyphenols and behavior of the American cockroach. Anat. Rec., 134:
641.

WEIAnT, E. A., 1958. Control of spontaneous activity in certain efferent nerve fibers from
the metathoracic ganglion of the cockroach, Periplaneta americana. Proc. 10th Int.
Cong. Ent., 2: 81-82.

ON THE HARDENING,.,OF | THE .CHORION..OF. THE, FS Ea
AFTER > FERTILIZATION.” 11.” THE err AN Ten aie
CHORION* THAR DENINGAIN ORY ZIASSUATEPiES

ElJl ONTSUKA

Biological Laboratory, General Education Department, Kyushu University, Fukuoka, Japan

Shortly after fertilization or parthenogenetic activation, soft chorions of fishes
are transformed into rigid structures mechanically non-elastic, and chemically
resistant. This transformation has attracted much attention particularly in recent
years (Nakano; 1956; °T.'S.'" Yamamoto; "1957; Ohtsuka, 1957 ;\Zommpaets ae
Rothschild, 1958), and some interesting facts on substances participating in the
phenomenon have come to our knowledge. However, the mechanism by which fish
chorions are hardened is still a matter of speculation. In an earlier paper of this
series (Ohtsuka, 1957), the possibility was suggested that in Oryzias egg the
process underlying chorion hardening involves an oxidation process. This hy-
pothesis seems to be provided further support in the present paper in which there
are presented data on the effects of various chemical agents on the chorion
hardening. ,

The author wishes to express his hearty thanks to Prof. I. Kawakami for his
kind direction in the course of this investigation. He is also very grateful to
Prof. N. Yoshii for his interest and encouragement.

MATERIAL AND METHODS

The ripe unfertilized eggs of the fresh-water fish, Oryzias latipes, were used as
material. They were taken from matured females and kept in isotonic Ringer’s
solution. The overripe eggs of which chorions had slightly separated from the egg
surface were discarded. The Ringer’s solution used here had the following constitu-
tion: M/7.5 NaCl 100 parts + M/7.5 KCl 2.0 parts + M/11 CaCl, 2.1 parts whose
pH was adjusted to 7.3 by adding NaHCO, (T. Yamamoto, 1944). Sperm
suspension was also prepared in Ringer’s solution.

Treatments with chemical agents were carried out before or after fertilization at
room temperature (22—27.5° C.). In the pretreatment, after some lengths of ex-
posure, aliquots of the eggs were put back into Ringer’s solution and then in-
seminated. Besides, in a few experiments insemination was done in the presence
of reagents. The effect was examined with regard to chorion hardening. ‘This is
possible to judge indirectly from the degree of chorion elevation; it is generally
accepted that the chorion is low when hardening is produced by the reagent whereas
it elevates very highly under inhibitory condition. On this account the volume
of elevated chorions was calculated 50 minutes after fertilization, according to
T. Yamamoto’s formula (1940). But the above criterion was found to have its

120

HARDENING OF FISH CHORION. III. 12%

exceptions as described in the text. Therefore, in all the cases, chorions were torn
with glass needles for the determination of their stiffness after volume measurement.

RESULTS
I. Oxidizing agents

The possibility that oxidation takes part in chorion hardening makes it of interest
to test whether hardening is caused by means of oxidizing agents. Experiments
were conducted on the action of potassium ferricyanide, sodium tetrathionate,
iodine, hydrogen peroxide, sodium periodate, chromic acid and potassium dichro-
mate. All the agents were dissolved in Ringer’s solution except for periodate. This
was made up in Ca-free Ringer’s solution to avoid precipitation by calcium.

When unfertilized eggs were subjected to these solutions (pH 7.3) for periods
of 15 to 30 minutes, chorion hardening took place in the agents other than chromic
acid. Such eggs with hard chorions exhibited no visible cortical changes during
each treatment. Here it should be added that some of the eggs were partheno-

TABLE, T

Effect of the presence of oxidizing agents on the chorion changes of cadmium-pretreated
eggs at fertilization. Temp. 22.7° C.

Volume of chorion Chorion

Agents cu. mm.) hardening
No treatment 1.44 + 0.018 —
Potassium ferricyanide (2 X 10-3 M) 0.99 + 0.015 ar
Sodium tetrathionate (5 x 107-3 M) 0.99 + 0.020 ah
Iodine in potassium iodide (2.5 & 10-4 M) 0.98 + 0.017 He
Hydrogen peroxide (1.5%) 1.02 + 0.014 +
Sodium periodate (4 & 107! M) 0.98 + 0.012 +
Chromic acid (1 X 10-2 M) 1.40 + 0.016 +
Potassium dichromate (2 < 1074 M) 0.97 + 0.011 oh

genetically activated in both hydrogen peroxide and tetrathionate. In a chromic
acid solution the chorion remained unchanged even after more prolonged exposure.
It was found, however, that if the solution is acidic, chromic acid, though less
effective, is capable of inducing hardening; at pH 6.6 it required a lapse of 60
minutes or more until the chorion became hardened. Then, Ringer’s solution of
chromic acid with this pH value was employed in the following experiments, but
did not give any appreciable effect because of its feeble hardening ability. Treat-
ments with oxidizing agents were carried out soon after fertilization. It was
found that except in the case with chromic acid, all the other agents bring about
the lowering of chorion elevation as a result of the acceleration of the hardening
which follows fertilization.

Table I shows a typical result of experiments in which eggs immersed in a
cadmium chloride-Ringer’s solution (2.5 x 10°? M) for 10 minutes were trans-
ferred to the solutions of oxidizing agents immediately following insemination in
Ringer’s solution. Cadmium treatment is known to abolish chorion hardening
(Ohtsuka, 1957). It will be seen, however, that in these eggs except the chromic
acid-treated ones, the chorions elevated were lower so that there developed a narrow
perivitelline space.

£22 EIJI OHTSUKA

It is of special interest that chromic acid is less effective than any of the other
agents in inducing hardening. The discussion on this point will be presented later.

II. Reducing agents

Experiments were performed using the following reducing agents to determine
whether their action on chorion hardening would be reverse to that of oxidizing
agents: sodium sulfide, potassium cyanide, sodium thiosulfate, sodium sulfate,
sodium thioglycolate, ammonium sulfate and potassium ferrocyanide. These agents
were dissolved in appropriately diluted Ringer’s solution so as to become osmotically
equivalent to M/7.5 NaCl, and prepared in concentrations to contain the effective
amount of calcium, since the deficiency of the latter blocks hardening of the chorion
(Ohtsuka, 1957). The pH of all the solutions was made to 7.3 with HCl or NaOH.

Freshly inseminated eggs were put into each isotonic solution of the reducing
agents. Here in both solutions of sulfide and cyanide the chorions were incapable
of hardening, but such an effect of the remaining agents was observed only in a

TABLE I]
Effect of the presence of reducing agents on the chorion changes at fertilization.
Temp. 24.1° C.
Agents Pretreatment Volume of chorion Chorion

(min.) (cu. mm.) hardening

No treatment a 1.05 + 0.014 +
Sodium sulfide (2 10-2 M) oe 1.07 + 0.017 —

Potassium cyanide (3.3 K 107? M) 1.07 + 0.012 —
Sodium thiosulfate (6.6 xX 10-2 M) 5 1.06 + 0.023 —
Sodium sulfate (6.6 * 107? M) 5 1.07 + 0.018 —
Sodium thioglycolate (9 K 1072 M) 30 1.06 + 0.015 —
Ammonium sulfate (6.6 K 107? M) 5 1.06 + 0.020 —
Potassium ferrocyanide (4 & 10-2 1) 10 1.06 + 0.013 —

few cases (e.g., in 2 of 9 experiments with thiosulfate). If, however, fertilization
was done in the presence of these agents with which the eggs had previously been
treated for some periods (5-30 minutes), hardening always failed to occur. Reduc-
ing agents were found to be effective in high concentrations. One of the repre-
sentative results is tabulated in Table II, where it can be seen that under the
influence of reducing agents the chorions did not markedly elevate in spite of the
inhibition of their hardening. And they were resistant to osmotic pressure of the
perivitelline fluid. The factor inhibiting further elevation of the soft chorion in these
circumstances is not known. On the other hand, it was impossible to prevent chorion
hardening when eggs were inseminated in Ringer’s solution after any length of
pretreatments. Furthermore, the chorions once hardened could not be converted
into a soft condition by treatments with reducing agents.

III. Sulfhydryl compounds

It has not yet been established on the nature of the chorion substrate under-
going hardening in fish eggs. The author (1957) was interested in SH groups,

HARDENING OF FISH CHORION. III. 123

Taste LI!

Effect of 5 minutes’ pretreatments with mercaptide-forming agents on the chorion
changes at fertilization. Temp. 25.4° C.

Volume of chorion Chorion

Agents (cu. mm.) hardening
No pretreatment 1.05 + 0.014 ot
Mercuric chloride (2.5 X 1075 M) 1.47 + 0.022 =
p-Chloromercuribenzoate* 1.48 + 0.016 —

* Saturated in Ringer’s solution.

anticipating that their reactivity might contribute to the chorion hardening. To
elucidate this point, two series of experiments were undertaken, one with a
mercaptide-forming agent and the other an alkylating agent. Preliminary study
by a staining method on the composition of soft chorion seems to demonstrate the
presence of SH groups (presumably protein bound-SH groups) ; no nitroprusside
reaction was given, while it was stained with coloured SH reagent (method of
Mescon and Flesch), even when this staining was applied after fixing the chorion
with 2 per cent trichloroacetic acid.

In the first experiments, unfertilized eggs were placed for 5 minutes in two
mercaptide-forming agents in Ringer’s solution, mercuric chloride and p-chloro-

=
on

wan Onnn nen ene n ane eee O=

VOLUME OF CHORION (cu.mm.)*
5

0 3 4 5 6 10
TIME AFTER FERTILIZATION (min.

Figure 1. Changes in chorion volume in PCMB-Ringer’s solution at different times after
fertilization. Dotted line shows a period during which the chorion was burst by osmotic
pressure of perivitelline fluid.

* Points (5, 10 and 15) of chorion volume in Figure 2 of the previous paper (Ohtsuka, 1957)
should be corrected to 0.5, 1.0 and 1.5, respectively.

124 ELL OBMTSUKA

mercuribenzoate (PCMB). (Lapse of more than 5 minutes in these solutions
caused the migration of cortical alveoli towards the animal pole without activation
of the egg.) Insemination was carried out in Ringer’s solution immediately after
the pretreatment. In both cases, chorions were deprived of hardening and there
occurred a marked elevation followed by bursting (Table III). This inhibition is
due to the blocking of SH groups as shown in the following experiment. Accord-
ing to Anson (1945), the combination of mercury with SH groups can be removed
by cyanide. Then, eggs which had been treated with mercaptide-forming agents
for 5 minutes as above were consecutively transferred to potassium cyanide-Ringer’s
solution: (3.3 X 107 M, pH 7.3) prior to insemination. When these eeesiwere
fertilized in Ringer’s solution after 5 minutes, reversal of the inhibition was ob-
served. The chorion showed a normal degree of elevation, owing apparently to
the occurrence of its hardening.

The failure to harden was also produced by exposing eggs to each mercaptide-
forming agent after insemination, and their inhibitory effect could similarly be
cancelled upon subsequent treatment with potassium cyanide. Figure 1 repre-
sents the results of a typical experiment with PCMB at 25.8° C. If PCMB was
applied within 10 minutes after fertilization, further hardening did not proceed

TABLE IV
Effect of the presence of alkylating agents on the chorion changes at fertilization.
Temp. 26.2° C.
Volume of chorion Chorion
Agents (cu. mm.) hardening
No treatment 1.05 + 0.012 +
Iodoacetic acid (3.3 K 107? M) 1.48 + 0.017 _
Chloroacetophenone* (0.05%) 1.47 + 0.024 _

* Saturated in Ringer’s solution.

and the elevated chorions retained softness. It should be noted, however, that
when this treatment was made about 5 minutes after fertilization or later, such
soft chorions became incapable of elevating highly. Unlike the cases of reducing
agents described above, they were burst by an osmotic pressure of the perivitelline
fluid.

In the next experiments, eggs were subjected to treatment with iodoacetic acid
or chloroacetophenone. These alkylating agents were made up in Ringer’s solution
(pH 7.3). It was found that either short treatment of unfertilized eggs is not
inhibitory to chorion hardening when they are subsequently inseminated in Ringer’s
solution. If the duration of these exposures was long, in the case of chloro-
acetophenone fertilized eggs underwent cytolysis and the iodoacetic acid-treated
eggs lost their fertilizability. When, however, both agents were applied after
insemination, the chorions failed to harden. Eggs exposed to alkylating agents
soon after fertilization formed considerably high chorions (Table IV).

IV. Urea and lithium bromide

The above experiments clearly indicate that SH groups primarily participate
in hardening of the chorion. They might be expected to be available as a possible

HARDENING OF FISH CHORION. III. 125

donor of hydrogen bonds. [If this is really the case, the latter thus given would
be responsible for hardening. According to Nakano (1956), the hydrogen bonds
are concerned in the mechanical properties of the chorion of Oryzias. Attempts
were thus made with urea and lithium bromide, which are known to cleave hydro-
gen bonds. The maximum concentrations at which they were tested were pre-
pared in 3.3 X 10° M both in Ringer’s solution. Eggs were treated before or
after fertilization. But neither reagent prevented chorion hardening, even in those
cases where unfertilized eggs were exposed for 60 minutes and then inseminated
therein. Thus it seems that hydrogen bonds play no role in giving rise to a
chorion hardening.

V. Aldehydes

In a staining study, the soft chorion was further found to be periodate-Schiff-
positive (Lillie’s method), which was blocked by the acetylation technique of
McManus. From this result it is evident that a polysaccharide having a-glycol
groups impregnates the chorion and produces aldehydes on oxidation (cf. Lison,
1953). The question arises whether such aldehydes take part in hardening of

TABLE V

Effect of the presence of aldehydes on the chorion changes of cadmium-pretreated eggs
at fertilization. Temp. 27.3° C.

Volume of chorion

Agents (cu. mm.) Chorion hardening
No treatment 1.48 + 0.017 =
Formaldehyde (0.03%) 0.97 + 0.014 oa
Acetaldehyde (0.5%) 0.98 + 0.018 +
Acrolein (0.05%) 1.01 + 0.012 ae
Benzaldehyde (0.2%) 0.99 + 0.015 oe

the chorion. To clarify this point, the effect of aldehydes, such as formaldehyde,
acetaldehyde, acrolein and benzaldehyde, was investigated. Ringer’s solutions of
these agents were adjusted to pH 7.3 by adding NaOH.

The results obtained were essentially the same as those with oxidizing agents
noted already. The chorions became hardened without egg activation in the
presence of the agents other than acrolein. In the latter solution, almost all eggs
were parthenogenetically activated 2-3 minutes after their immersion and there
took place faster chorion hardening than in the control. Table V illustrates that
aldehydes bring about the formation of low chorion when cadmium-pretreated eggs
are treated after fertilization in the same way as in the cases of oxidizing agents.

DISCUSSION

As shown in the foregoing pages, the chorion fails to harden under the influence
of certain reducing agents. Very high concentrations of the agents are charac-
teristic of this inhibition, suggesting the existence of an antagonistic force respon-
sible for hardening. It has been reported in Oryzias egg that a phospholipid in
the perivitelline space, which is originated from the cortical layer itself following
fertilization or artificial activation, is of prime importance in giving rise to the

126 EWI OHTSUKA

chorion hardening, the action of which is regarded as oxidative in nature (Ohtsuka,
1957). Therefore it is most probable that reducing agents inhibit hardening by
interfering with this oxidative effect. On the other hand, the oxidizing agents,
potassium ferricyanide, sodium tetrathionate, iodine, hydrogen peroxide, sodium
periodate, chromic acid and potassium dichromate, were found to have the harden-
ing capacity. The effectiveness of these agents in hardening the chorion is prob-
ably due to oxidation. We shall return to this aspect later. :

Similar results have been obtained with sea urchin eggs by Motomura and
Hiwatashi (1954), who tested the effects of various chemicals on the hardening
of the fertilization membrane, and found that the action of reducing agents is
opposite to that of oxidizing agents; the former inhibits membrane hardening
whereas the latter accelerates it. They offered an explanation that hardening is
concerned with polymerization to which both agents are sensitive in a reverse
manner. But, as just discussed, the present results can be reasonably interpreted
from an oxidative point of view. Still this does not preclude the possibility of
polymerization in the hardening of the chorion.

In view of these considerations, it may be allowable to conclude that the oxida-
tion process is involved in the process whereby Oryzias chorion is converted into
rigid structure. And it appears that it takes part in the initial stage of the harden-
ing process, as is indicated by the fact, for example, that the chorion became
hardened even when treatment with sodium periodate was carried out for 10
minutes, during which no changes were visible, and then kept in Ringer’s solution.
This conclusion is, however, contradicted by Nakano’s observation (1956) that
colloidal substances, such as gum arabic, gum tragacanth, albumin and gelatin, are
capable of hardening Oryzias chorion. ‘These substances are unfortunately non-
oxidative in chemical nature, their mechanisms remaining obscure.

In a more recent study, Zotin (1958) has found a substance indispensable for
the chorion hardening in the perivitelline space of some salmonid eggs. According
to him, this substance, which was termed the hardening enzyme, is secreted from
the cortical layer, but not from the alveoli present there, upon fertilization or
activation, and plays an active role in hardening the chorion. Thus there is a
considerable resemblance between phospholipid in Oryzias egg and the hardening
enzyme of salmonid eggs, particularly with respect to their source and behaviour.
Furthermore, Zotin described that hardening of the chorion in salmonids, which
is due to polymerization of some substances, is blocked by oxygen deficiency.

It has been shown in the present paper that the SH groups (presumably protein
bound-SH groups) of the chorion are essentially necessary for hardening, while
their role bears no relation to their possible capacity to give hydrogen bonds. In
this connection it should be emphasized that the first three of the effective oxidants
noted above, potassium ferricyanide, sodium tetrathionate, and iodine, all oxidize
SH groups (Anson, 1945). A similar action, though less reliable, is known
about hydrogen peroxide (see Hirai, 1957) capable of hardening as well. Hence
these strongly suggest an oxidative conversion of SH into S-S; the latter thus
formed may be responsible for chorion hardening. This change must proceed
beyond the reversible stage, because a hard chorion is not returned to the soft
condition by reducing agents so far as the present study is concerned. Additionally,
there is evidence that a chorion polysaccharide having a-glycol groups participates

HARDENING OF FISH CHORION. III. 127

in hardening. Sodium periodate, another effective oxidant, does change all the
available a-glycol groups to aldehydes (cf. Lison, 1953). Then, if such a type of
oxidation is involved in the chorion hardening, it is worth pointing out that chromic
acid likewise produces aldehydes in oxidizing the polysaccharide but it seems with-
out action on a-glycol groups (cf. Lison, 1953); this may be the reason why
chromic acid is less effective in inducing hardening in spite of its strong oxidative
capacity. It is therefore likely that in hardening a-glycol groups are subject to
oxidation, resulting in the formation of aldehyde groups. Favoring this idea is
the fact that the chorions harden by treatments with aldehyde, such as formalde-
hyde, acetaldehyde, acrolein and benzaldehyde. Thus it may be said that two
substances of the chorion, a protein containing SH groups and a polysaccharide
which has a-glycol groups, form a complex undergoing hardening.

From the above account it follows that the reactivities of both sulfhydryl and
a-glycol groups contribute to the hardening of the chorion. There must occur
some profound changes in these two groups, especially with regard to their, and
perhaps reciprocal, oxidative reactions. However, an alternative possibility re-
mains, based on the following facts: According to the present data, SH groups
become responsible for chorion hardening about 10 minutes after fertilization.
It was further shown that when PCMB, which attacks SH groups, is applied
about 5 minutes after fertilization, the chorion retaining softness is incapable of
elevating very highly, being burst by osmotic pressure of the perivitelline fluid.
These observations reveal some preceding change, probably of polysaccharide,
and favor the assumption that oxidation concerns first the a-glycol groups only
and then it is followed by that of SH. Significant in this respect is a hardening
effect of certain aldehydes mentioned already, and moreover evidence is available
that aldehydes are able to combine with SH groups (Schubert, 1936), which latter
still remain capable of being oxidized (Anson, 1945).

Thus it may be suggested that hardening is due to the combination of at least
two factors in the chorion, an SH group, and an aldehyde produced by oxidation
of a-glycol group. And the former would undergo oxidation to S-S. There
might be an oxidation polymerization through these disulfide bridges. Very inter-
esting is that the reaction between SH groups and aldehydes is related to a skin
tanning (Hirai, 1957). Furthermore, it should be mentioned that potassium
dichromate capable of hardening is an excellent tanning agent. Considering these,
a tanning reaction appears to take place in the chorion hardening. Rothschild
(1958) proposed that fish chorions are hardened by the tanning effect of substance
which is discharged from the cortical alveoli at fertilization or parthenogenetic
activation. The positive effect of the alveolar substance on chorion hardening has
been suggested by Nakano (1956) in Oryzias and by T. S. Yamamoto (1957) in
Oncorhynchus. But the unpublished data on Oryzias egg show that hardening is
independent of the alveolar substance.

SUMMARY

1. Oxidizing agents, such as potassium ferricyanide, sodium tetrathionate,
iodine, hydrogen peroxide, sodium periodate, chromic acid and potassium di-
chromate, cause hardening of the chorion. Chromic acid is less effective than any
of the other agents.

128 El ORs UA

2. In the presence of reducing agents, such as sodium sulfide, potassium cyanide,
sodium thiosulfate, sodium sulfate, sodium thioglycolate, ammonium sulfate, and
potassium ferrocyanide, the chorions fail to harden.

3. Treatment of unfertilized eggs with mercuric chloride or p-chloromercuri-
benzoate induces the inhibition of chorion hardening when they are subsequently
inseminated in Ringer’s solution. The same result is also obtained in the cases
where both agents are applied after fertilization. Either inhibitory effect is re-
moved if the treatment is followed by that with potassium cyanide.

4. Iodoacetic acid and chloroacetophenone likewise inhibit hardening when
eggs are treated after insemination.

5. Neither urea nor lithium bromide has any effect on the chorion hardening.

6. Aldehydes, such as formaldehyde, acetaldehyde, acrolein and benzaldehyde,
are capable of hardening the chorion.

7. Soft chorion is impregnated with a protein containing SH groups, together
with polysaccharide which has a-glycol groups.

8. It is concluded that the oxidation process is involved in the mechanism of
chorion hardening. It is suggested that hardening is due to combination of SH
groups with the aldehydes produced by oxidation of a-glycol groups.

LITERATURE Clip

Anson, M. L., 1945. Protein denaturation and the properties of protein groups. Adv. in
Protein Chem., 2: 361-386.

Hirat, H., 1957. Sulfurs in protein. In: Seibutsukagaku Saikinno Shinpo, Toyko, III: 1-50
(in Japanese).

Lison, L., 1953. Histochimie et Cytochimie Animales. Second éd. Gauthier-Villars, Paris.

Moromura, I., AnD K. Hiwatasut, 1954. Further note on the inhibition and acceleration of
the toughening of the fertilization membrane in the sea urchin’s egg. Sci. Rep. Tohoku
Univ., 20: 219-225.

Nakano, E., 1956. Changes in the egg membrane of the fish egg during fertilization.
Embryologia, 3: 89-103.

OutsuKA, E., 1957. On the hardening of the chorion of the fish egg after fertilization. I.
Role of the cortical substance in chorion hardening of the egg of Oryzias latipes.
Sieboldia, 2: 19-29.

RoTHSCHILD, Lorn, 1958. Fertilization in fish and lampreys. Biol. Rev., 33: 372-392.

Scuusert, M. T., 1936. Compounds of thiol acids with aldehydes. J. Biol. Chem., 114:
341-350.

Yamamoto, T., 1940. The change in volume of the fish egg at fertilization. Proc. Imp. Acad.
Tokyo, 16: 482-485.

YamamorTo, T., 1944. Physiological studies on fertilization and activation of fish egg. I.
Response of the cortical layer of the egg of Oryzias latipes to insemination and to
artificial stimulation. Annot. Zool. Jap., 22: 109-125.

Yamamoto, T. S., 1957. Some experiments on the chemical changes in the membrane of
salmon eggs occurring at the time of activation. Jap. J. Ichthyol., 6: 54-58.

ZoTIN, A. I., 1958. The mechanism of hardening of the salmonid egg membrane after fertili-
zation or spontaneous activation. J. Embryol. Exp. Morph., 6: 546-568.

move LE VMPERATURE BLOCKAGE OF MOLTING IN UCA PUGNAX

EE PASSANO

Depariment of Zoology, Yale University, New Haven, Connecticut

The fiddler crabs, members of the genus Uca, have been the favorite animals
for the experimental investigation of molting in brachyurans by American com-
parative endocrinologists. The facility in collecting and holding these sand and
mud flat burrowers and the simplicity of inducing proecdysis by bilateral eyestalk
removal have resulted in use of these crabs and this technique by many investi-
gators during the past two decades. Casual note has been made of the variation
in proecdysis duration observed, some of which has been attributed to differences
in ambient temperatures. Presumably because of limited facilities, no systematic
study has been made of the effect of temperature on either proecdysis duration or
proecdysis initiation.

In a preliminary observation incidental to a previously reported study of the
endocrinological basis of molting (Passano, 1953) a group of 20 male Uca pugnax
had both eyestalks removed in early winter and were then kept for 35 days in
individual bowls at room temperatures diurnally fluctuating between 14° and 21° C.
There was a high mortality rate, but of the 8 crabs surviving, none molted. When
put in a 27° constant temperature room, three crabs molted after 3, 7, and 8 days.
Since eyestalkless U. pugnax kept at 23.5° C. showed an average proecdysis period
of 19.3 days (Passano, 1953), it appeared as though a 5° reduction in temperature
might block or markedly delay a forced ecdysis initiated by eyestalk removal
(i.e., by extirpation of the X-organ sinus gland complex).

The experiment reported here has two parts: 1) exposure to a constant test
temperature for 23 days, following elimination of the molt-inhibiting hormone
(MIH) by bilateral eyestalk removal, and 2) a post-treatment period at the
optimum temperature for induced proecdysis, to measure the extent of the previous
temperature-induced proecdysis block.

MATERIALS AND METHODS

Approximately 600 male marsh fiddler crabs, Uca pugnax (S. I. Smith), were
collected from a salt marsh near New Haven, Connecticut, in late April (water
temperature 12° C., air temperature 8°-15° C.) and acclimatized to 15° for one
week. Four hundred animals of uniform size were then selected, discarding any
whose major chela was missing or undersized, or which had autotomized more
than one walking leg, so that the selected crabs had approximately the same blood
volume. One eyestalk was removed by cutting across the arthrodial membrane at
the base of the eyestalk with fine scissors and each animal was forced to autot-
omize one of its walking legs, usually the fourth leg on the side opposite to the
major chela. The following day the other eyestalk was removed in a like manner.
Twelve hours later the surviving eyestalkless animals were divided into ten groups

129

130 L. M.-PASSANO

of 38 crabs each, hereafter called Groups A, B, C... J, with the use of a table
of random numbers. Subsequent measurement showed that these groups were
homogeneous in size distribution with a standard length mean, and standard devia-
tion of the mean, of 18.7+1.2 mm. (Table I). Each animal was put in an indi-
vidual covered glass dish (or “fingerbowl”) with sufficient water (1:1 Long
Island Sound Water:tap water; salinity 14%) to cover the maxillipeds. The
water was changed on the sixth, twentieth and thirty-third days after eyestalk
removal. Each group was placed in a separate temperature-controlled room
(Table I) and kept there for the temperature treatment period of 23 days. Those
rooms closest to ambient temperatures, 20° and 22°, showed the largest fluctua-
tions in air temperature, but these rarely exceeded 1° and were rapid and cyclic.
A control group of 19 crabs was placed at the warmest temperature after forced

TABLE [|

The effect of exposure to different temperatures on ecdysis by eyestalkless Uca

Temperature treatment Post-treatment Total
3 days) (20 days at 29°)
Mpanueet gee 4
t andar
Group eel ilenaele Mean
mortality) mm.) Temp. % % re- % % re- % proecdysis
2G molt maining molt maining molt duration
and S.D.
A 37 18.7 Sit 0 92 46 14 46 37.44 .85
B on 18.6 153 0 86 43 26 43 36.84 .79
C 37 18.5 16.3 0 97 54 19 54 35.7+1.00
D Sil 18.6 18.2 0 97 59 16 59 33.44 .97
E 38 18.7 20.0 0 100 58 KS 58 31.8+1.05.
F 36 18.5 222 DBS 72 Sil 8 56 24.8+1.15
G Si 18.8 24.4 ifs) 16 uh 0 84 19.34 .61
H 5)5) 18.5 26.7 76 17 6 0 81 17.44 .63
I Sif 18.5 29.4 89 5 0 0 89 14.8+ .73
i] Sit 18.8 S22 87 0 0 0 87 14.4+ .49
0 11 0 —

Controls 19 19.0 SA 0 53

387 Wen lise 18/2

autotomy of one walking leg, but as expected, none molted nor showed signs of
entering proecdysis. Since the experimental animals were blinded, no regular
photoperiods were given. The crabs were not fed, inasmuch as it was impossible
to insure equal uptake in the different temperature groups or by individual crabs
at the same temperature.

Mortality during the temperature treatment period of the experiment was light.
Deaths during the first week (3.2%) are considered to be due to operational mor-
tality and are not included in the tabulated results.

Although some animals successfully completed ecdysis following the experi-
mental treatment, the largest number died during ecdysis or were unable to with-
draw all of their appendages. Any crabs dying in Stage E (Drach, 1939) were
scored as molted, while those few dying in Stage D, were scored as having molted
the following day. No animals were kept after ecdysis.

LOW-TEMPERATURE BLOCKAGE OF MOLTING t31

After 23 days, those crabs remaining were examined for evidence of proecdysis
initiation. The regeneration of the previously autotomized fourth walking leg
was used as an indicator (Bliss, 1956; Jyssum and Passano, 1957). The animals
were then slowly equilibrated to 29.4° and kept there for a further 20 days to
determine what effect, if any, the initial temperature treatment had had on pro-
ecdysis initiation. When the experiment was ended, 43 days after removal of
the second eyestalk, 8.5% of the animals remained.

RESULTS

Following elimination of the MIH by bilateral eyestalk removal, a total of 242
of the 368 animals (66%) eventually reached or completed ecdysis. None of the
control crabs molted. Table I and Figures 1 and 2 show the differential effect of

100
GP. A — 13.1°C
B -—- 15.3
80 G --- 16.3
D — 18.2
E a 20.0 N
piss \2892 \
wo 6 —__244 RLSRE
nH 60 cae \
wn H 26.7 \
ra 1 ----294 \
© J —— 32.2 \
7
3° 40 \
\
\
\
\
.
N
20 \
\
\
\
8) .

5 10 15 20 25 30 35 40
TOTAL PROECDYSIS, DAYS

Ficure 1. Cumulative percentages of Uca molting in each experimental group following
eyestalk extirpation.

temperature on average proecdysis duration, the per cent completing proecdysis
and the uniformity of response within each group. The low values of the standard
errors of the mean for the proecdysis duration of each group are noteworthy.

1. Low temperature treatment, Groups A, B, and C

Exposure to constant low temperature in the range from 13.1° to 15.3° C.
blocks proecdysis initiation almost completely, in spite of elimination of the MIH.
With a single exception in the group kept at the lowest temperature, none of the
Group A or Group B animals initiated discernible proecdysis during the tempera-
ture treatment period. Only 14% of these crabs showed basal limb bud (Bliss,
1956) regeneration of their autotomized pereiopod, itself a proecdysis-independent
process (Jyssum and Passano, 1957). Again with the single exception already

£32 LU MeePASSANO

noted, none reached the ecdysis stage, Stage E (Drach, 1939), in less than 9
days after being placed at 29.4° (Fig. 1). The mean proecdysis duration in the
post-treatment warm temperature for Groups A and B combined was 14.1 days
with a standard error of the mean of + 0.48 days. This is not significantly dif-
ferent from the value for Group I crabs which were put directly at 29.4° after
eyestalk removal (Fig. 2, shaded bar), although it is shorter.

POST-TREATMENT, 29°C, DAYS

AVERAGE PROECDYSIS, DAYS (TOTAL)

CONFIDENCE
INTERVAL (RANGE

INITIAL TEMPERATURE DURATION, DAYS

10 1S 20 25 30
INITIAL TEMPERATURE TREATMENT. °C

Ficure 2. The effect of various initial temperature treatments on the proecdysis duration
of eyestalkless Uca. The shaded horizontal bar represents the confidence interval (13.7-15.9
days) for animals put directly into 29.4° C., transposed to the upper, post-treatment, portion
of the graph.

Group A’s exceptional animal molted on the fourth day of post-treatment warm
temperature. It showed no limb regeneration whatsoever. Since it is probable
that this individual had already commenced proecdysis prior to the beginning
of the experiment, the increase in the extremes of proecdysis duration caused by
this aberrant animal is shown by a dashed line in Figure 2. _

Group C, kept at 16.3° for the temperature treatment period, also shows a

LOW-TEMPERATURE BLOCKAGE OF MOLTING £35

post-treatment proecdysis duration, 12.7 + 1.0 days, insignificantly different from
that of Groups A and B. However, half the crabs showed some basal limb bud
regeneration after 23 days at 16.3° and one crab, which molted six days later after
being raised to 29.4°, had a “premolt limb bud” (Jyssum and Passano, 1957)
at that time. The other three animals which molted prior to the tenth post-
treatment day again showed no limb regeneration at all.

2. High temperature treatment, Groups G, H, I and J

There was no discernible proecdysis blockage in the groups of animals kept
at 24.4° or above (Figs. 1,2). The two groups kept at the highest temperatures,
J (32.2°) and I (294°), had average proecdysis durations of 14.4+0.49 days
and 14.8+0.73 days, respectively, the slight difference being without statistical
validity. The somewhat greater standard deviation of Group I is due to a slight
skew of the distribution curve towards the longer proecdysis (Fig. 1). In neither
case did any molting occur prior to 10 days, nor 22 days after, removal of the
second eyestalk. Molting occurred in 88% of Groups I and J, and of these 94%
regenerated the autotomized pereiopod.

Groups H (26.7°) and G (244°) also showed a nearly uniform molting
response but here proecdysis duration was significantly increased (P < 0.01%;
P< 0.04%). As in the groups kept at the highest temperatures, nearly all regen-
erated their autotomized walking leg.

3. Intermediate temperature, Groups D, E, and F

None of the crabs kept at 18.2° (Group D) or 20.0° (Group E) reached
ecdysis during the temperature treatment period, but examination of the regenerat-
ing limb buds showed that at least 50% were in proecdysis. During the post-
treatment period at 29.4°, approximately half of those crabs which eventually did
reach ecdysis had done so before the temperature-blocked animals of Groups A,
B and C began to molt (Fig. 1). Nearly all of the Group F molts (22.2°) had
molted by this time.

The largest standard deviation from mean proecdysis duration occurred within
these intermediate temperature groups, in Group F. Many of those animals in
which proecdysis appeared to be blocked failed to survive the 20-day post-treatment
period at 29.5° (Table I).

DISCUSSION AND CONCLUSIONS

Molting after bilateral eyestalk extirpation is due to elimination of molt-
inhibiting hormone (MIH) of the medulla terminalis ganglionic X-organ (Bliss,
1953; Passano, 1953) and consequent formation and release of molting hormone
(MH) by the Y-organ (Echalier, 1959). MH then causes the sequence of
morphogenic and biochemical events collectively termed proecdysis, ecdysis and
postecdysis.

The results reported here show that proecdysis duration is strongly temperature-
dependent. In Uca pugnax a temperature between 29° and 32° gives the shortest
average proecdysis duration. Temperatures above this may be deleterious to this

134 L. M. PASSANO

crab, since most of the normal control animals failed to survive the 42 days of
this experiment, but it is equally plausible that these animals starved to death.
Temperatures somewhat below 29° still permit molting but the proecdysis period
is progressively lengthened, doubling in duration when the temperature falls
from 30° to 20°.

Temperatures of 15° or below effectively block proecdysis. A comparison
between either Group A or Group B and Group I shows that mean proecdysis
duration at 29° is nearly the same whether or not the eyestalkless animals had
first been kept at 15° for three weeks (Fig. 2). No appreciable proecdysis occurs
during this period in these groups. The low temperature treatment did reduce the
total percentage molting to one-half its original value (Table 1). This is presum-
ably due to the continuous depletion of organic reserves during the three weeks.
of low temperature treatment starvation. If the experimental animals had been
fed, this decrease in the percentage of the group reaching ecdysis would probably
have not occurred.

At some temperature between 15° and 22° proecdysis may either fail to com-
mence, or else may be initiated but proceed slowly, after eyestalk removal. The
lengthened mean proecdysis durations of the intermediate temperature groups,
D (18°) and E (20°), are due in part to proecdysis blockage in some of the crabs.
After three weeks at the treatment temperatures, 42% of the 18° animals and
39% of the 20° animals showed no signs of proecdysis initiation. The remaining
crabs had initiated proecdysis but its duration was lengthened by the “low” tem-
perature. It is thus to be expected that these intermediate temperature groups
are the most heterogeneous in their response of all the temperature groups tested
Clio 2).

It is scarcely surprising to find an inverse correlation between temperature
and proecdysis duration. It is unusual to find that a temperature of 15° or even
higher can block completely an animal’s growth, even though this species experi-
ences lower temperatures through much of its life cycle. Since it can survive
at these lower temperatures for long periods of time, 1t seems probable that some
key reaction, required in proecdysis initiation, is being blocked. The first half
of the proecdysis requires MH, since crabs deprived of their Y-organs are perma-
nently blocked in their intermolt stage (Echalier, 1959). Low temperatures might
thus prevent the Y-organs from responding to the elimination of the MIH following
eyestalk removal. Alternatively, these glands might form and release into the
circulation adequate amounts of MH, but the target tissues might be unable to
respond to the MH rise. Yet neither of these explanations accounts for the nearly
complete correlation of proecdysis blockage and basal limb bud regeneration
blockage found in these experiments. Although the main part of limb regenera-
tion occurs only during proecdysis, so that regenerate size is a good index of
proecdysis stage (Bliss, 1956), the initial or basal outgrowth of the regenerating
limb is independent of Y-organ MH (Jyssum and Passano, 1957) and, unlike
proecdysis initiation, is not photosensitive (Bliss, 1956). Therefore if the tem-
perature-dependent proecdysis blockage found here is blocking Y-organ MH forma-
tion or activity, some other temperature-dependent step must be blocking basal
limb bud formation in these starved animals.

Although a few of the animals which molted in this experiment failed to show

LOW-TEMPERATURE BLOCKAGE OF MOLTING 135

any limb regeneration, 92% did regenerate the autotomized walking leg. Perhaps
temperatures of 15° or below limit some metabolic process common to both proec-
dysis and basal limb bud formation. One of the initial events in proecdysis is the
mobilization of hepatopancreas metabolic reserves (Bliss, 1953; Passano, 1960),
noted after eyestalk removal by a rise in oxygen consumption (Edwards, 1950;
Bliss, 1953) and a fall in R.Q. (Bliss, 1953). Stored fats are utilized for proec-
dysis integument growth and mineral mobilization (Renaud, 1949). The basic
metabolic requirements of the animal, as well as basal bud regeneration, must
require storage depletion as long as animals remain unfed. Perhaps the low tem-
peratures prevent sufficient mobilization of hepatopancreas reserves for basal limb
regeneration and proecdysis initiation, as well as basal metabolism. It is possible
that such mobilization is normally controlled by an eyestalk hormone in the intact
animal at higher temperatures, since eyestalk removal causes a rise in oxygen
consumption whether or not the Y-organs are intact (Passano, unpublished).
It is not known whether the same rise in O, consumption follows bilateral X-organ
extirpation in the Y-organless animal; if so, the MIH may be controlling proecdysis
initiation by limiting the fat depot mobilization necessary for MH synthesis.

An alternative explanation might be developed from the hypothesis that all
crustacean growth processes are controlled in an identical manner (Bliss, 1956).
Thus basal limb growth would be under MH control, but this MH would originate
in some extra-Y-organ site such as the individual limb bud blastema. It would
support this alternative if it were found that Y-organless crabs could be forced
into proecdysis by initiating mass basal limb bud regeneration. Multiple regen-
eration in normal crabs can lead to ecdysis even though environmental conditions
are unfavorable (Bliss, 1956), but such treatment might activate their intact
Y-organs. If such extra-Y-organ sources of MH occur, then low temperatures
could be blocking MH formation or activity, irrespective of the hormone’s origin.

Uca (including U. minax (Le Conte) and U. pugilator (Bosc) in addition
to the species used here) occurs on the East Coast of the United States as far
north as Cape Cod, but does not commonly occur north of the Cape (Rathbun,
1918). The notable discontinuity at Cape Cod in the littoral fauna is primarily
due to the cooler summer water temperatures of the Gulf of Maine, yet the tem-
perature experienced by adult Uca must be primarily determined by ambient air
temperature rather than water temperatures. Since there is no marked air tem-
perature differential between the littoral areas north and south of Cape Cod, it
seems likely that the temperature limit on Uca acts on growth of its pelagic larvae
rather than on post-larval growth. Perhaps, then, the significance of the proecdysis
temperature block found in these experiments is that it reflects an identical limita-
tion of proecdysis initiation occurring in Uca larvae, so that larvae hatching in the
cooler Gulf of Maine waters are unable to grow because they are unable to molt.
Caution must be maintained, however, in attributing species limits to simple tem-
perature effects (Moore, 1958).

SUMMARY

1. A uniform population of male Uca pugnax was used in studying the effect
of 10 temperatures on proecdysis resulting from eyestalk removal.

2. Proecdysis duration is shortest at 29° to 32° C. and lengthens significantly
at lower temperatures.

136 Lb. MauPrASSAN®

3. Proecdysis initiation is markedly temperature-sensitive. At 15° or below
initiation is completely blocked; at 15° to 20° a substantial proportion of the crabs
fail to begin proecdysis.

4. Temperatures which block proecdysis initiation also block basal limb bud
regeneration (a molt-independent growth process). It is hypothesized that a
metabolic event common to both processes is being blocked.

5. The physiological and ecological significance of the proecdysis temperature
block is considered.

LITERATURE CITED

Buss, D. E., 1953. Endocrine control of metabolism in the land crab, Gecarcinus lateralis
(Fréminville). I. Differences in the respiratory metabolism of sinusglandless and
eyestalkless crabs. Biol. Bull., 104: 275-296.

Buss, D. E., 1956. Neurosecretion and the control of growth in a decapod crustacean. In:
Bertil Hanstrom, Zoological Papers in Honour of His Sixtyfifth Birthday, November
20th, 1956 (K. G. Wingstrand, ed.), pp. 56-75. Zoological Institute, Lund.

DracHu, P., 1939. Mue et cycle d’intermue chez les Crustacés Décapodes. Ann. Inst. Océanog.
(Paris) (N.S.), 19: 103-391.

EcHALIER, G., 1959. L’organe Y et le déterminisme de la croissance et de la mue chez
Carcinus maenas (L.), Crustacé decapode. Ann. Sci. Nat. Zool. et Biol. Animale,
[Zee 57;

Epwarps, G. A., 1950. The influence of eyestalk removal on the metabolism of the fiddler
crab. Physiol. Comparata et Oecol., 2: 34-50.

Jyssum, S., Anp L. M. Passano, 1957. Endocrine regulation of preliminary limb regeneration
and molting in the crab Sesarma. Anat. Record 128: 571-572.

Moore, H. B., 1958. Marine Ecology. John Wiley and Sons, Inc., New York.

PassAno, L. M., 1953. Neurosecretory control of molting in crabs by the X-organ sinus gland
complex. Physiol. Comparata et Oecol., 3: 155-189.

Passano, L. M., 1960. Molting and its control. Im: The Physiology of Crustacea (T. H.
Waterman, ed.). Academic Press, Inc., New York. Pp. 473-536.

Ratusun, M. J., 1918. The Grapsoid Crabs of America. U.S. Natl. Museum Bull., 97: 1-461.

ReNAupD, L., 1949. Le cycle des réserves organiques chez les Crustacés Décapodes. Amn. Inst.
Océanog. (Paris) (N.S.), 24: 259-357.

eee >t ITION OF SKELETAL STRUCTURES IN THE CRUS-
eee bE HISTOLOGY OF THE GASTROLITH SKELETAL
Piso COMPLEX AND THE GASTROLITH IN THE CRAYFISH,
ORCONECTES (CAMBARUS) VIRILIS HAGEN—DECAPODA!?

DOROTHY F. TRAVIS 2
Biological Laboratories, Harvard University, Cambridge 38, Mass.

Little is known regarding the basic mechanisms by which certain cells and tissues
participate in synthesis and calcification of organic matrices. The gastrolith discs
and the branchial exoskeleton provide particularly useful experimental material in
the Crustacea for such studies. Attention, in this paper, will be confined to the
histological study of the gastrolith discs of the fresh-water crayfish, Orconectes
virilis Hagen.

The paired gastrolith discs are composed of the cuticular lining of the stomach,
the thickened gastric epidermis, and the underlying sub-epidermal connective tissue.
They are located in the anterior lateral walls of the cardiac stomach of the crayfish.
Their usefulness as sites of activity for the study of cellular processes involved in
the synthesis and calcification of organic matrices is evident when it is realized that
the epidermis of these modified portions of the stomach wall becomes competent to
synthesize and calcify the gastrolith matrix preceding molt. This activity culminates
at the end of the premolt period in the formation of hard calcified disc-shaped
gastroliths which lie in a sac or pouch now formed between the epidermis and
cuticular lining of the stomach. At the same time gastrolith formation occurs, the
epidermis of the exoskeleton is participating in resorption of mineral and organic
constituents. At the molt, the fully formed gastroliths are shed with the old
stomach lining into the stomach. Following molt, these gastroliths are gradually
broken down and resorbed. Some of their mineral constituents are resorbed by
the gastrolith epidermis and hepatopancreatic epithelium (Travis, in preparation).
These mineral constituents are conveyed by the blood to the epidermis underlying
the exoskeleton of other areas and are re-used in synthesis and calcification of
their organic matrices.

While the description of the presence of gastroliths before molt and _ their
gradual disappearance following molt has been given by a number of authors
(Réaumur, 1712; Chantran, 1874a, 1874b; Braun, 1875; Huxley, 1879; Herrick,
1895 ; Irvine and Woodhead, 1889; Husson, 1952; and others), only certain aspects
of the histology of the gastrolith discs have been described by Braun (1875) for the
European crayfish, Astacus fluviatilis, and by Herrick (1895) for the American
lobster, Homarus americanus. The histological changes have neither been described
with reference to stages of the molting cycle nor to the synthesis of the non-calcified

1 This investigation was supported in part by a Special Fellowship (HF-8000) from the
National Heart Institute, United States Public Health Service.
2 Research Fellow—Harvard Biological Laboratories, Cambridge 38, Mass.

137

138 DOROTHY Be TRAVIS

skeletal components of the gastrolith discs, and only with brief reference to the
synthesis and calcification of the gastrolith itself. Accordingly, it is the aim of
the present paper to deal with histological changes which are associated with the
formation of non-calcified skeletal components of the gastrolith discs and those
involved in the synthesis and calcification of the gastrolith itself.

MATERIALS AND METHODS

Animals. Since the duration of each stage of the molting cycle, the length
of the intermolt periods, and frequency of molts for each size group have not been
established for either field or laboratory Orconectes virilis, only males ranging in
size from 40-49 mm. carapace length and molting during July and August were
used in these studies. All animals were previously collected from the Cambridge
Reservoir and maintained in the laboratory in concrete tanks containing rocks and
soapstone slabs placed over the bottom to provide secluded areas for crayfish
retreat and basking. A gently flowing stream of water, ranging in temperature
from 21.8-25.0° C., was maintained at a level of approximately 2.5 inches in the
tanks.

Stages of the molting cycle. Stages of the molting cycle were determined ac-
cording to Drach (1939, 1944). This method is based primarily on morphological
features of the exoskeleton, and clearly delimits four major stages—A, B, C, D—each
of these being divisible into a number of substages. Postmolt, consisting of Stages
A, B, early, and middle C, is characterized by progressive hardening of the pre-
exuvial layers; formation, progressive thickening and hardening of the endocuticle ;
the beginning of feeding; and formation and progressive thickening of the mem-
branous layer. The imtermolt condition (Stage C,) is characterized by the com-
pletion of all components of the exoskeleton—the epicuticle, exocuticle, endocuticle,
and membranous layer. At this stage, one of comparative “rest” or “stability,”
the completed membranous layer with the adhering epidermis can be stripped from
the remaining portion of the exoskeleton. Premolt, consisting of early, middle,
and late Stage D, is characterized by a series of integumentary transformations
which occur preparatory for the ensuing molt. The major portions of the old
exoskeleton, both organic matrix and mineral salts, are resorbed; the new pre-
exuvial layers, consisting of the epicuticle and exocuticle, are deposited under the
old; the animals cease to feed; and the termination of the period is marked by
the molt. Stage D, is a new stage, previously referred to by Drach in connection
with the shrimp molting cycle (Comments at Harvard, 1957). Stage D, in the
crayfish is a stage in the molting cycle which cannot be distinguished from Stage C,
by external features of the exoskeleton. The membranous layer with the adhering
epidermis can be stripped from the exoskeleton in both stages but in Stage Dp,
gastrolith deposition has begun, the gastroliths in this case being thin plate-like
structures.

Histological methods. Three to five animals were killed for each specific stage
of the molting cycle. The gastrolith discs were quickly removed, placed on a glass
slide and straightened in a drop of fixative before they were transferred to bottles
containing larger volumes of fixative. Although the author has used various
fixatives for general histological studies of crustacean material (1951, 1955, 1957),

CRUSTACEAN SKELETAL DEPOSITION 139.

Bouin’s fixative containing 1% calcium chloride (anhydrous) was used in these
studies and has proved to be the best fixative. Fixation was carried out for at
least a 24-hour period, followed by washing in 70% ethyl alcohol for the same
period, dehydration in 80%, 95%, 97% ethyl alcohol for 30 minutes each, two
changes of 100% ethyl alcohol for 15 minutes each and infiltration according to
Peterfi’s Celloidin-paraffin Method (Pantin, 1948). Sections were cut at 6 4,
deparaffinized, coated with 0.5% celloidin for one minute, allowed to dry and
stained with Mallory’s triple stain and Ehrlich’s haematoxylin and eosin for general
histological studies.

OBSERVATIONS AND DISCUSSION
The intermolt condition—Stage C,

Intermolt is marked not only by the completion of the calcified skeletal com-
ponents and the membranous layer of the rigid exoskeleton, but by the completion
of the non-calcified skeletal components of the gastrolith discs. The epicuticle of

CHANGES IN THE GASTROLITH DISC DURING THE MOLTING CYCLE
STOMACH

PYLORIC STOMACH----
GASTROLITH DISC

ESOPHAGUS

I=EPICUTICLE

2=E XOCUTICLE

3=ENDOCUTICLE

4=EPIDERMIS

5=CONNECTIVE TISSUE

6=BLOOD CELL FORMING TISSUE

1Sv,
Me
‘
\)
\

MM VY
~< =a

Figure 1. Diagram showing the histological changes which occur in the gastrolith disc
during significant stages of the molting cycle. Enlarged section of the gastrolith disc is from
an intermolt animal (Stage C:), in which all skeletal components of the disc are completed,
more clearly shown in Figure 2. Note significant changes in the development of cuticular
components, epidermal height, connective tissue thickness, abundance of reserve cells (repre-
sented as black oval bodies). In Stage Do, note gastrolith (G) forming between epidermis
and cuticular components of the disc, and in Stage Ds, note hypertrophy and folding of the
epidermis after it has retracted from the fully formed gastrolith and undergone a spurt of growth.

140 DOROTHY, Fi DRAATS

these latter components, adjacent to the lumen of the stomach, constitutes one
surface of each gastrolith disc, while the basement membrane of the sub-epidermal
connective tissue, adjacent to the hemocoel, constitutes the other surface of each disc
(Figs. 1, 2, 4). Although the gastrolith discs, a few millimeters in diameter, are
modified portions of the stomach wall, they can be distinguished from the remainder
of the stomach wall in Stage C, by distinct differences in the structure and com-
position of their cuticular components and by the nature of their epidermis.

The completed skeletal components of the gastrolith discs of an intermolt
animal show three major differentiated layers in contrast to the four observed in

——— —

LZ”, P=
(01006, " Coe ¥

a) Seca YN 9. FY," o\00\ofo}
On| Tee ARG Ye PARI GKe BES" lf pag Lolo
Be GN at Ayes

VO a ae a ran ent gee ree ee My

ay |

|"

ny

Figure 2. Enlarged diagram of the gastrolith disc of an intermolt animal (Stage Ci).
Note the three differentiated skeletal layers—(1) epicuticle, (2) laminated exocuticle, (3)
endocuticle with characteristic, chemically complex granules which mark the gastrolith disc
histologically from the ordinary stomach lining (S). Note the tissue complex of gastrolith
disc, which consists of a single layer of columnar epidermal cells (E), the sub-epidermal
connective tissue (C), composed of cells of Leydig, and the transient reserve cells (represented
as black oval bodies). Also note the blood cell-forming gland (4) and blood vessels in the
connective tissue.

the more rigid branchial exoskeleton. Beginning with the lining of the stomach
(Figs, 1,2, 4):, thesevare:

1. The epicuticle. This is a very thin non-calcified component, approximately
3—4 » in thickness and composed of a glyco-lipoprotein complex (Travis, in prepara-
tion). It is neither laminated nor crossed by pore canals or tegumental ducts.
Elaboration and hardening of the epicuticle begins before molt (Stage D,) and is
completed by Stage C,. This epicuticle differs from that of the branchial exo-
skeleton in being thinner and not being calcified or crossed by tegumental ducts.
It stains red with Mallory’s.

2. The exocuticle. This is a laminated component of approximately 8-12 p» in
thickness and is also composed of a glyco-lipoprotein complex (Travis, in prepara-

CRUSTACEAN SKELETAL DEPOSITION 141

tion). Deposition does not begin until late Stage A and is completed in Stage B.
In contrast to the exocuticle of the branchial skeleton the gastrolith disc exocuticle
is thinner, non-calcified, lacks pore canals and tegumental ducts, and is formed
during a different stage of the molting cycle. The outer zone stains light blue
with Mallory’s while the inner stains dark blue, with a reddish haze at the junction
of the exo- and endocuticle.

3. The endocuticle. This is the thickest of the three layers. It reaches a
maximum thickness of 15—25 » in Stage C, and stains very light blue with Mallory’s.
This layer differs from the same layer of the branchial exoskeleton in that it 1s
neither laminated nor crossed by pore canals or tegumental ducts. The endocuticle
begins to be deposited in late Stage B and is completed by Stage C,. It appears as
a loose fibrous meshwork and is marked by the presence of round structureless
granules (Figs. 1, 2, 4), which also stain red with Mallory’s. These granules
reach a size of 2.5-3.0 p, the largest being located immediately under the exocuticle.
These are chemically complex granules which contain a glycoprotein complex, a
mixed lipid complex, glycogen at certain stages of the molting cycle, and calcium
(Travis, in preparation). Such granules are not observed in the endocuticle of
the ordinary foregut lining.

The tissue underlying the skeletal components is composed of a single layer of
columnar epidermal cells which lie in immediate contact with the endocuticle (Figs.
1, 2, 4). These columnar epidermal cells are approximately 40 p» in height with
nuclei of about 5 X 13 p. The epidermal layer of the remaining portion of the
stomach is a layer of cuboidal or low columnar cells.

The sub-epidermal connective tissue is very similar to that observed in the
branchial integument. It is predominantly composed of the loose “spongy” type
of connective tissue cells known as “cells of Leydig” (Figs. 1, 2, 4). A number
of transient cells, called “protein cells” by Cuénot (1891, 1893), blood corpuscles
by Hardy (1892), “reserve cells” (Travis, 1951, 1955, 1957), “lipoprotein cells”
(Sewell, 1955), are also evident during the intermolt condition. These transient
or reserve cells contain both acidophilic and basophilic granules when stained with
haematoxylin and eosin and are usually fuchsinophilic with Mallory’s. The granules
vary in size, as observed with both the light and electron microscope (Travis and
Chapman, II), from less than one micron to around four microns, and have the
same histochemical composition as observed in the granules of the endocuticle
(Travis, in preparation). The fine-structure studies indicate that the mature
reserve cells contain large refractile granules which show variable densities and
double membranes, while the immature cells are smaller, may contain clearer non-
refractile cytoplasm and smaller granules of variable densities. These cells, among
other functions, appear to be intimately involved in transport and release of reserves
for synthesis and elaboration of skeletal elements (a more detailed discussion of
these cells will be given in a subsequent paper of this series). In addition to these
cellular elements, the sub-epidermal connective tissue is well penetrated by branches
of the antennary arteries and blood sinuses. Also present in the connective tissue
is a conspicuous blood cell-forming gland (Figs. 1, 2), in which stages of amoebocyte
or reserve cell development may be seen. This blood cell-forming gland has also
been observed by Mataczynska-Suchcitz and Hryniewiecka (1958). Delimiting the
connective tissue from the hemocoel is a basement membrane.

142 DOROTHY FE. fLRAVIS

Figure 3. Two pairs of gastroliths removed from the gastrolith pouches. The exposed
convex surface of the upper pair would lie im situ in immediate contact with the epidermis.
The concave surface of the lower pair would lie in immediate contact with the stomach
lining. 3X.

Ficure 4. Section of the gastrolith disc skeletal-tissue complex (Stage Ci). The epi-
cuticle (1), the exocuticle (2), the endocuticle (3) containing its characteristic chemically
complex granules (arrow), the epidermis (E) and the connective tissue (C). 1800 X.

CRUSTACEAN SKELETAL DEPOSITION 143

~Premolt-Stage D, through D,

During this period of the molting cycle, marked histological changes, associated
with gastrolith deposition, occur in the gastrolith disc tissues. In Stage D,, the
epidermis (with cells of about 75 » in height and bearing nuclei of around 5 X 12 »)
becomes somewhat invaginated. In section, this is evident at the very edge of
_ the gastrolith disc (Fig. 5). These active epidermal cells shortly become branched
and attenuated at their apical ends and many intercellular spaces are then apparent
between the cells (Figs. 1, 6). Their attenuated apices lie in immediate contact
with the forming matrix of the gastrolith. At the ultrastructure level, it is evident
that the attenuated apices observed with the light microscope also bear innumerable
microvillar equivalents (Travis and Chapman, III). From the microvillar equiva-
lents both organic and mineral material are secreted, some microvillar equivalents
secreting granular material into the forming amorphous matrix while others secrete
distinctly beaded-fibrous material. The secreted organic components, not observ-
able with the light microscope, undergo polymerization to form the fibrous lamellae
(Travis and Chapman, II, III) clearly observed in the matrix of the gastrolith
with the light microscope (Figs. 6, 7, 8). Evidence at the ultrastructure level
also reveals that calcium transport and deposition appear to occur by two different
means (Travis and Chapman, III). The granular and beaded-fibrous components
secreted from the microvillar equivalents undergo a change. The former gives
rise almost immediately to small crystals of calcite ranging in size from 440-1060 A,
while the beaded components of the latter eventually give rise to the same size
crystals, presumably of calcite. Crystal formation from the beaded-fibrous com-
ponents occurs simultaneously with polymerization changes associated with the
formation of the fibrous lamellae of the matrix (Travis and Chapman, III). These
small crystals, which form in the matrix of the young gastrolith (Stage D,) and
which are not observed with the light microscope, grow to a size of 7 w» or more
in diameter and are clearly observed in ground sections of the completed gastrolith
(Figs. 12, 13). These ground sections of the completed gastrolith (Stage D,)
also reveal that there is a shift in the mode of crystal deposition as the youngest
and last layers of the fully completed gastrolith are deposited. In this case, crystals
are deposited in such a fashion that rows of crystal prisms or rods (Fig. 10) are
laid down parallel to the longitudinal axis of the epidermal cells and perpendicular
to the parallel lamellae of the gastrolith (Fig. 9). The thickness of the younger
layers of the gastrolith is constituted of these crystal prisms or rods, with one
crystal stacked vertically above another. This thickness is 490-500 p» as compared
to the older portions of the gastrolith (Figs. 9, 11, 12, 13), which constitute a thick-

Ficure 5. Section through the extreme edge of the developing gastrolith (Stage Do).
Note that the developing matrix (M) is filled with granules similar in appearance and chemical
composition to those observed in the completed endocuticle. Also note beginning appearance
of fibrous lamellae (arrow). Epidermis is denoted by (E), connective tissue by (C) and
reserve cells (R). 1800 x.

Ficure 6. Section through a more central portion of the developing gastrolith (Stage Do).
Note the distinct fibrous lamellae of the gastrolith matrix (M) which result from post-secretion
polymerization changes, the attenuated apices of the epidermal cells (E) and the intercellular
spaces between the cells, and the connective tissue (C). Also note the three differentiated skel-
etal layers of the gastrolith disc—(1) epicuticle, (2) exocuticle, (3) endocuticle with character-
istic granules. 1800 x.

144 DORO TEAR. Ati Vals

Ficure 7. A section through the organic matrix of a completed gastrolith (Stage Ds.)
showing its fibrous lamellate nature. Fixed and decalcified in ordinary Bouin’s and stained
with toluidine blue. The more convex surface (arrow) would lie im situ in immediate contact
with the epidermis while the opposite surface would lie in contact with the stomach lining. 16 X.

Ficure 8. A similar section through the organic matrix of the early developing gastrolith
(Stage Do), showing its fine fibrous lamellate nature. Arrow indicates surface (younger
layers) which would lie in situ in contact with the epidermis. 21 x.

CRUSTACEAN SKELETAL: DEPOSITION 145

ness of 2500-3000 » and which show lamellae and crystals oriented at right angles
to the prisms. Preliminary x-ray diffraction studies indicate that the main inorganic
crystals in both the youngest and oldest layers of the gastrolith are calcium carbonate
deposited as calcite.

The sub-epidermal connective tissue becomes greatly compressed or stretched
during this stage of the molting cycle. This is largely due to the size of the devel-
oping gastrolith which is plate-like in nature (Stage D,) and has attained a thickness
at this time of about 300 ». By middle Stage D, the gastrolith has increased to a
thickness of approximately 1 mm. and by the end of D,, a thickness of approxi-
mately 3-4 mm. is achieved in animals of 40-49 mm. carapace length. Reserve
cells remain numerous in the connective tissue and epidermis during the entire pre-
molt period, being heavily concentrated in the apices of the epidermal cells, and
achieve their greatest abundance in late Stage D (D,). During this latter stage
the epidermis has completed the deposition of the gastrolith and retracts from it,
undergoes rapid growth, accompanied by much folding, and achieves its greatest
height (242 ») and nuclear size of 9 X 14 w (Fig. 14). It now begins to elaborate
a new epicuticle which is the only skeletal component of the gastrolith disc deposited
before molt. In areas of the stomach wall on each side of the gastrolith discs both
pre-exuvial layers, epi- and exocuticle, have been deposited. When synthesis of
the gastrolith and new epicuticle are achieved by the gastrolith disc epidermis,
correlated with the many tasks the epidermis performs in other skeletal areas,
molting ensues and the old stomach lining and the completed gastroliths. (Fig. 3)
are released into the stomach.

Postmolt-Stage A, B and C

Stage A. The tissue complex of the gastrolith disc remains greatly hyper-
trophied. The epidermis, in contrast to the folded appearance in Stage D,, is a
flat-surfaced tissue, composed of rows of columnar cells, which maintain a height
of approximately 182 » and an average nuclear size the same as that of a D, animal
(Fig. 15). The reserve cells remain abundant in both the epidermis and connective
tissue, with the greatest number being observed in the apices of the epidermal
cells. The only skeletal component observed in the gastrolith disc early in Stage A

Ficure 9. A ground section of a fully developed gastrolith (Stage Di). The youngest
portion (1) would be in contact with the epidermis, and consists of crystal prisms or rods,
also indicated in Figure 10, which are laid down parallel to the longitudinal axis of the epi-
dermal cells and perpendicular to the parallel lamellae of the gastrolith. Region (2), also repre-
sented in Figure 11, is an older portion of the gastrolith showing the parallel lamellae with
crystals oriented at right angles to the prisms of region (1). Region (3) represents the oldest
layers of the gastrolith, originally corresponding to Stage Do, and would lie in contact with
the skeletal layers of the gastrolith disc which constitute part of the stomach lining, also
represented in Figures 12 and 13. 8.5 x.

Ficure 10. Enlarged portion of region (1) from a ground section of a completed gastro-
lith, depicted in Figure 9. Note crystal prisms or rods of calcite. 400 x.

Ficure 11. Enlarged portion of region (2) from a ground section of a completed gastro-
lith depicted in Figure 9. Note that crystals of calcite in the lamellae (arrows) lie at right
angles to the crystal prisms of region (1). 400 x.

Ficures 12, 13. Enlarged portions of region (2) from a ground section of a completed
gastrolith depicted in Figure 9. Note same arrangement of crystals of calcite—7-70 mw in
diameter—as observed in Figure 11. 400 X.

146 DOROTHY F. TRAVIS

Ficure 14. Section of gastrolith disc showing greatly hypertrophied and folded epidermis
(E), following retraction and growth of this tissue after gastrolith deposition is complete
(Stage Ds). Note beginning of new epicuticle formation (arrow) and numerous dark stain-
ing reserve cells in the apices of the epidermal cells. 900 x.

Ficure 15. Section of gastrolith disc immediately following molt (Stage A). Note the
greatly hypertrophied flat-surfaced epidermal cells (E), the epicuticle (arrow), the only
skeletal component evident in early Stage A, and the numerous dark staining reserve cells
(arrow-R), and connective tissue (C). 900 X.

GRUSTACEAN SKELETAL: DEPOSITION 147

is the epicuticle, partly formed and hardened previous to molt and having a
thickness of only 1.6-2.0 ». Curiously enough, the exocuticle does not apparently
begin to be deposited until late Stage A and is completed in Stage B, achieving a
thickness of approximately 8-12 ». In the areas of the stomach wall on each side
of the gastrolith discs, the exocuticle is completed and endocuticle formation has
already begun.

Stage B. This stage of gastrolith disc formation is marked by the completion
of the synthesis and elaboration of the exocuticle. It is both fibrous and laminated
in nature. This latter aspect is not clearly evident at the microscopic level, but is
distinctly apparent in electron micrographs (Travis and Chapman, II). The
synthesis and elaboration of the endocuticle also begins during the latter part of
this stage but is not completed until late Stage C.

There is a slight decrease in epidermal cell height (112 ») and nuclear size
(5.1 X 11.5 w) and an increase in thickness of the connective tissue when com-
pared to the situation observed in Stage A. During this stage the reserve cells
are still quite abundant.

eager. arly Stage C (C,) and middle Stage €: (C,) are marked by the
reduction in epidermal height (about 80 »), a slight increase in nuclear size
(5 xX 13 »), an increase in thickness of the sub-epidermal connective tissue, a gradual
reduction in the number of reserve cells, and the continued synthesis of the endo-
cuticle, previously described under the intermolt condition. It is during Stage C,
that the interesting granules, observed in the gastrolith disc endocuticle, become
evident at the ultrastructure level, although not becoming distinctly apparent at the
light microscope level until Stage C,. In the tissue complex there is little differ-
ence between the early and middle Stage C condition and that observed in the
intermolt animal. At the end of Stage C when synthesis of the gastrolith disc
skeletal components is complete, the animal has again reached an “intermolt’’
condition (C,) from which the entire cycle repeats itself. Further information
on the development of the non-calcified skeletal components of the gastrolith disc
and of the calcified gastrolith itself will appear in subsequent papers of this series.

SUMMARY

1. The gastrolith discs of the fresh-water crayfish, Orconectes virilis Hagen,
are located in the anterior lateral walls of the cardiac stomach and are themselves
modified portions of the stomach wall.

2. Histologically, the cuticular surface of the completed gastrolith disc (Stage
C,), which forms part of the lining of the stomach, 1s composed of three differen-
tiated layers: a thin non-calcified epicuticle, formed before molt; a thicker non-
calcified exocuticle, laminated in nature, but neither crossed by pore canals, nor
tegumental ducts, formed following molt and completed during Stage B; and the
endocuticle, the thickest of the three layers, characterized by neither being laminated
nor crossed by pore canals and possessing round granules without structure, em-
bedded in a loose fibrous meshwork. The largest of these granules are located in
the outer layers of the endocuticle very close to the exocuticle. The tissue complex
of the completed gastrolith disc is composed of a single layer of columnar epidermal
cells, a loose “spongy” type of sub-epidermal connective tissue, composed of cells of
Leydig and a number of the transient reserve cells which are also observed in the

148 DOROTEHY..FS- TRAVIS

epidermis. The tissue complex is well supplied by branches of the antennary
arteries and has a large blood-cell-forming gland.

3. Marked histological changes are observed during Stage D when premolting
processes, associated with gastrolith deposition, are set into motion. The epidermis
increases greatly in height, invaginates, and begins to elaborate the matrix of the
gastrolith which is filled at its lateral edges with granules like those observed in the
C, endocuticle. At this same time the active epidermal cells take on a branched
attenuated appearance at their apical ends, and many intercellular spaces become
apparent between the cells. From Stage D, through most of Stage D, continued
synthesis, elaboration, and calcification of the gastrolith matrix occur, its thickness
increasing from around 300 » (D,) to 3-4 mm. (D,). Concomitant with these
changes is an increase in the number of reserve cells in both epidermis and con-
nective tissue and a stretching, and thus compression, of the sub-epidermal con-
nective tissues, the latter process being correlated with the increasing size of the
forming gastrolith. Near the end of Stage D (D,), the epidermis has completed
its task of depositing the gastrolith, retracts from it, undergoes rapid growth accom-
panied by much folding and achieves its greatest height. It then begins to elaborate
the epicuticle, the only pre-exuvial skeletal component deposited in this site. When
this task is achieved by the epidermis, correlated with the many tasks it performs
in other skeletal areas, molting ensues and the old stomach lining and the formed
gastroliths are released into the stomach.

4. Postmolt histological changes in the gastrolith disc are associated with con-
tinued synthesis and elaboration of cuticular components. The “intermolt condition”
is achieved in Stage C,, when these synthetic tasks are complete, and from this
stage the entire histological cycle repeats itself.

LIDERATURE GItED

Braun, M., 1875. Uber die histologischen Vorgange bei der Haiitung von Astacus fluviatilis.
Arbeit. aus dem Zool. Instit. Wurzburg, 2: 121-166.

CHANTRAN, S., 1874a. Observations sur la formation des pierres chez les écrevisses. C. R.
Acad. Sct., 78: 655-657.

CHANTRAN, S., 1874b. Sur le méchanisme de la dissolution intra-stomacale des concretions
gastriques des écrevisses. C. R. Acad. Sci., 79: 1230-1231.

Cuknot, L., 1891. Etudes sur le sang et glandes lymphatiques dans la série animale. Arch.
Zoo. Exp. Gen. (Sér. 2), 9: 13-90.

Cuénot, L., 1893. Etudes physiologiques sur les crustacés décapods. Arch. de Biol., 13:

245-303.
DracH, P., 1939. Mue et cycle d’intermue chez les crustacés décapods. Ann. Inst. Oceanogr.,
19: 103-391.

Dracu, P., 1944. Etude préliminaire sur le cycle d’intermue et son conditionnement hormonal
chez Leander serratus (Pennant). Bull. Biol. France Belgique, 78: 39-61.

Harpy, W. B., 1892. The blood-corpuscles of the Crustacea, together with a suggestion as to
the origin of the crustacean fibrin-ferment. J. Physiol., 13: 165-190.

Herrick, F. H., 1895. The American lobster. Bull. U. S. Fish. Comm., 15: 1-252.

Husson, M. R., 1952. <A propos du role des gastroliths chez les Ecrevisses. C. R. Acad. Sci.,
235: 905-907.

Huxtey, T. H., 1879. The Crayfish. First Edition, Kegan Paul, Trench & Co., London.

IRVINE, R., AND G. S. WoopHEAD, 1889. Secretion of carbonate of lime by animals. II. Proc.
Roy. Soc. Edinb., 16: 324-354.

MALACZYNSKA-SuUCHCITz, S., AND S. HryniewrecKaA, 1958. Cytophysiological and cyto-
chemical investigations on the elements of hemolymph in Crustacea. Bull. de l’Acad.
des Polon. des Sciences et des Lettres, Sct. Nat. Ser. B, 14: 319-333.

ee

CRUSTACEAN SKELETAL DEPOSITION 149

Pantin, C. F. A., 1948. Notes on Microscopical Technique for Zoologists. First Edition,
Cambridge University Press, Cambridge, England.

Reaumur, M. D., 1712. Sur les diverses reproductions qui se font dans les Ecrevisses, les
Omars, les Crabes, etc., et entre autres sur celles de leurs jambes et de leurs écailles.
Mém. de l’ Acad. Roy. Sct. Parts, 1712: 295-321.

SEWELL, M. T., 1955. Lipo-protein cells in the blood of Carcinus maenas, and their cycle of
activity correlated with the moult. Quart. J. Micr. Sci., 96: 73-83.

‘Travis, Dorotuy F., 1951. Calcium metabolism in the decapod Crustacea. Thesis for Ph.D.

i degree, Radcliffe College.

Travis, Dororuy F., 1955. The molting cycle of the spiny lobster, Panulirus argus Latreille.
II. Pre-ecdysial histological and histochemical changes in the hepatopancreas and
integumental tissues. Biol. Bull., 108: 88-112.

Travis, Dororuy F., 1957. The molting cycle of the spiny lobster, Panulirus argus Latreille.
IV. Post-ecdysial histological and histochemical changes in the hepatopancreas and
integumental tissues. Biol. Bull., 113: 451-479.

Travis, D. F., anp G. B. CuHapman, 1960. The deposition of skeletal structures in the Crus-
tacea. II. The fine structure of the developing skeletal components of the gastrolith
dises in the crayfish, Orconectes virilis Hagen—Decapoda. Submitted to J. Biophysic.
Biochem. Cytol.

Travis, D. F., ano G. B. CuHapman, 1960. The deposition of skeletal structures in the Crus-
tacea. III. The fine structure of the developing gastrolith in the crayfish, Orconectes
virilis Hagen—Decapoda. Submitted to J. Biophysic, Biochem, Cytol.

EFFECT. OF TEMPERATURE AND SALINITY ON PEAT TORER com
INoT WO: GRAPSOID CRABS, HEMIGRAPSUS MUDUs
AND HEMIGRAPSUS OREGONENSIS?

MAR Y-PLIZABE THs TODD 7;AND]P AULA. DENI

Department of Zoology, University of British Columbia, Vancouver, British Columbia, Canada

Temperature tolerance as an experimental criterion for the demonstration of
physiological change has found many uses and has been reported in various ways
throughout the literature. Most studies on temperature tolerance have been done
on fish, with relatively few experiments on invertebrates. In many cases the lethal
point was determined by the temperature at which death occurred when the animal _
was subjected to slowly increasing temperatures (Huntsman and Sparks, 1924;
Gowanloch and Hayes, 1926). As various authors raised the temperature at dif-
fering rates, the results could not be compared adequately because the time over
which the temperature was increased would be expected to have a marked effect
on the final lethal point. Experiments in which animals were maintained at
constant temperatures for a period of time provided a more accurate determination
of the lethal temperature as well as allowing comparisons among the species. In
this type of experiment, either time to death or per cent survival at intervals was
noted (Fry, Brett and Clawson, 1942; Brett, 1952; Spoor, 1955).

Acclimation to low and high temperatures has been demonstrated in many ani-
mals on the basis of laboratory acclimation, season, microgeography and latitude.
In this discussion, “‘acclimation” includes all types of “demonstrable compensatory
change” (Bullock, 1955). Acclimation has been demonstrated by a higher rate
function at any intermediate temperature when low temperature acclimated animals
are compared with high acclimated ones. Dehnel and Segal (1956) in laboratory
studies on the American cockroach, Periplaneta americana, found a higher rate
of oxygen consumption in equal-weight nymphs and adults after acclimation to
10° C. when compared with 26° C. acclimated animals. Respiration was meas-
ured at 20° C. Previously, the culture had been maintained for at least three gen-
erations at a constant temperature, 27° C. Edwards and Irving (1943) showed
seasonal acclimation in the sand crab, Emerita talpoida. At all experimental tem-
peratures below 20° C., oxygen consumption of animals in summer was less than
that in winter. Ohsawa and Tsukuda (1956) found seasonal acclimation of re-
sponse to temperature (exuding of the body) in the periwinkle, Nodilittorina granu-
laris. Microgeographical acclimation has been demonstrated in the limpet, Acmaea
limatula. Samples taken from lower intertidal levels had a higher rate of heart
beat than those from higher levels (Segal, 1956).

1 These studies were aided by grants from the National Research Council of Canada and
the National Science Foundation of the United States. The authors wish to express their
gratitude to Dr. Earl Segal and Dr. Theodore H. Bullock for their critical reading of the
manuscript.

2 Present address: Department of Zoology, University of Glasgow, Glasgow W. 2, Scotland.

150

TEMPERATURE TOLERANCE IN CRABS fot

Mayer (1914) has shown that at different latitudes there was a similar rate
function within a species at the environmental temperatures of the individuals,
even though these temperatures were different. The rate of pulsations of the bell
of Aurelia aurita was similar in animals off Nova Scotia and off Florida with a
temperature difference of 15° C. Scholander, Flagg, Walters and Irving (1953)
compared many poikilotherms, mostly interspecific but closely related species, from
arctic and tropic regions to find the general consistency of acclimation when rates
of oxygen consumption were investigated. Dehnel (1955) demonstrated a higher
rate of growth in gastropods from high latitudes when compared with low latitude
populations.

Another facet of acclimation is its effect on the temperature tolerance of a species.
A previous temperature history at the upper levels of the physiological temperature
range is known to raise the thermal resistance. Conversely, low temperature accli-
mation will decrease the tolerance to high temperatures. The rates of gain and
loss of heat tolerance appeared to show a consistent trend in diverse groups of
animals. In all cases, gain of heat tolerance was much more rapid than its loss.
Brett (1946) demonstrated that the goldfish, Carassius auratus, required a total of
thirty days to acclimate to 28° C. when brought from 4° C. in 8° C. steps, but
that acclimation occurred at different rates. Twenty days were required from
eemeetort2 » ©. bit only three days from 20° C. to*28° C. Loss of heat tolerance
in the crayfish, Orconectes rusiicus, was not completed by the end of the sixteenth
day when the crayfish were transferred from 22°-23° C. to 4° C. The heat tol-
erance lost after thirteen days at 4° C. was regained in about 24 hours (Spoor,
1955). Ohsawa (1956a) using the periwinkle Nodilittorina granularis, stated
(p. 206), “The acclimatization to higher temperatures in the winter snails is estab-
lished more readily than the acclimatization to lower temperatures in the summer
animals.”

The effects of salinity changes on temperature tolerance have been studied less
extensively. Broekema (1941) demonstrated that the temperature and _ salinity
relations were interdependent in the shrimp Crangon crangon when survival in
various combinations of the two was tested; a low salinity was endured better
when the temperature was high. Wikgren (1953) showed that there was almost
a continuous loss of ions from the crayfish, Potamobius fluviatilis, as the tempera-
ture dropped below 2° C. from about 18° C. when animals were transferred from
tap to distilled water. There was apparently more efficient osmoregulation, normal
loss of ions, at higher temperatures (above 2°-3° C.). The lobster, Homarus
americanus, was shown to have a higher lethal point when both salinity and tem-
perature were high during the period of acclimation. A decrease in salinity with
the acclimation temperature held constant resulted in a lowering of thermal resistance
(McLeese, 1956).

The two species of shore crabs studied in this investigation, Hemigrapsus nudus
(Dana) and H. oregonensis (Dana), are subjected to wide ranges of temperature
and salinity conditions in their natural environment. [Experimental parameters
of temperature and salinity were based on average seasonal conditions, to determine
if these seasonal changes, summer and winter, cause a resulting change in the upper
temperature tolerance of the two species. Also, animals were acclimated to various
combinations of temperature and salinity, to ascertain any resulting changes in

do2 MARY=BLIZABE TH SLODD AND PAUL AL DE EINE

the temperature tolerance within either season. Because of the morphological
similarity and like habitat in this area of these two species, it was hoped that
physiological differences, in this case resistance to high temperatures, could be
demonstrated. This was found to be the case. Additionally, effects of size,
crowding, sex and moulting on the tolerance were noted.

MATERIAL AND METHODS

The experimental animals, H. nudus and H. oregonensis, were collected from
Spanish Bank Beach on the south shore of Burrard Inlet. The Fraser River
flowing into Georgia Strait near the collecting beach is the main cause of the
recorded fluctuations in salinity occurring in this area. The beach has a gradual
slope and changes its contour from sand to rocks and again to sand. The lower
sandy area is exposed only on the lower tides. The crabs are found in the rocky
intermediate area, the main population of H. nudus distributed higher than that
of H. oregonensis. Individuals of both species, however, can be found almost from
highest occurrence to lowest.

The salinity is quite uniform throughout the winter, ranging from about 70
per cent to 80 per cent sea water (based on 100 per cent sea water as 31.88%c).
In the spring, with the greater influx of fresh water from the Fraser River, the
salinity begins to drop and has been recorded as low as 10 per cent in the summer
months. At the same time as the salinity is decreasing, temperature increases, and
the relatively static summer state results where the average salinity is about 35 per
cent. Daily fluctuations in the salinity also occur, as the incoming tide carries a less
saline water along the shore in the collecting region. This muddier Fraser River
water is distinct from the water of higher salinity by a visual demarcation several
hundred yards from shore.

The two species of crabs are found under rocks where a suitable micro-enyiron-
ment is provided by a pool of water or a bed of mussel shells (Mytilus edulis).
Both species are found also in damp sand under rocks, often partially buried.
Both males and females were used in the temperature tolerance experiments when
preliminary experiments showed no difference in the resistance of the sexes. Gravid
females or individuals missing any appendages were discarded, as were any soft-
shell crabs. A weight range was collected, about 0.2-gm. to 6.0-gm. animals,
excluding only very large individuals. Crabs were transported to the laboratory
in canvas buckets with damp sea-weed. The experiments reported here were con-
ducted from 1957 to 1959, inclusive.

In this investigation, holding experiments refer to those in which the tempera-
ture tolerance of the crabs is determined without any previous laboratory acclima-
tion. The animals were tested after at least 24 hours in the laboratory, and during
that period they were held at temperature and salinity conditions which approxi-
mated those in the field. This time period allowed the gut to be cleared partially
so that at the high test tolerance temperatures, deposition of faeces and urine did
not foul the water. In acclimation experiments, on the other hand, animals were
held at previously determined acclimation conditions which differed from environ-
mental circumstances in at least one factor. In these experiments crabs were
acclimated for at least one week in the laboratory. Preliminary experiments indi-

TEMPERATURE TOLERANCE: IN- CRABS 153

cated that acclimation was complete in less than seven days; the tolerance of the
two species, therefore, was tested from seven to a maximum of twenty-one days
in some experiments. Crabs were not fed at any time during the experiments, and
were kept in darkness.

If the acclimation temperature differed greatly from the environmental tem-
perature, the crabs were gradually warmed or cooled until the desired temperature
was reached, usually a period of two hours. Plastic dishes containing slightly
less than 4 liters of water were used for the experimental animals. Holding or
acclimation temperatures were either 5° C. or 20° C., 1° C., based on average
winter and summer temperatures. The salinities were either 75 per cent or 35
per cent sea water, again average seasonal conditions. Hence, four experimental
combinations were used. About thirty-five crabs were placed in each container.
Animals were changed, usually once per 24 hours, to water of appropriate tem-
perature and salinity over the acclimation or holding period.

Temperature tolerance tests were conducted by floating the plastic containers
in water baths in which the experimental temperature did not vary more than
+0.1° C. The appropriate salinity, corresponding to that used during the acclima-
tion or holding period, was employed for each tolerance test. Five animals per
memati per 4 liters of water) were tested at 1° C. intervals, from 28° C. to 35° C.
The five crabs tested at each temperature ranged in weight from small to large.
At each temperature both males and females were tested. Crabs were brought
to the test temperature from acclimation or holding temperatures over a period
of two to four hours at which time the experiment was begun. Preliminary ex-
periments indicated that more time was unnecessary as a more gradual warming
period did not alter the results. The experiment was conducted for 24 hours;
dead animals were removed at 12 hours and live ones were changed to clean water.
All animals were dried with gauze and weighed to the nearest 0.1 gram. Animals
were considered dead only when all movement ceased. Using this criterion as a
death point, at no time did animals revive when returned to lower temperatures.
Immediately prior to death, the sole motion visible was a slight pumping of water
through the gill chambers. Crabs which moulted during the 24 hours of the
experiment were discarded.

The relationship of weight and time to death was determined by checking
at hourly intervals over the 24-hour period and removing and weighing any dead
animals. There was a change of water at 12 hours as noted previously. One
set of experiments was used to test the effect of different numbers of crabs per
dish on the thermal resistance. All experimental conditions were as described
in the foregoing except for the varying densities of animals at each test temperature.

Data were plotted on arithmetic paper as it was believed that this type of plot
gave the most accurate representation of the temperature tolerance curve. Each
point, indicating per cent survival at any given temperature, represented the average
of a number of duplications (each five animals per dish) for those particular con-
ditions. In no instance was this average determined from less than six replica-
tions resulting in a total of at least thirty animals; usually more were used. The
total graph, then, was based on a minimum of one hundred-eighty animals. No
attempt was made to derive a formula for the fitting of each curve; the points
were merely joined. The 50 per cent survival temperature was based on the

154 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

point at which the survival curve crossed the line representing median survival
of the animals. In the graph which represented the relationship between size and
time to death, the regression line is eye-fitted.

An analysis of variance was done for the complete summer data mainly to
determine if a species difference existed. The difference in tolerance between the
species was less during the summer months, so that a significant difference here
definitely indicated one for the complete data, both summer and winter. For indi-
vidual comparisons such as the difference in tolerance between the species at a
given temperature-salinity combination, the “Student’s” f-distribution was used to
compare the 50 per cent survival temperatures. Another method was the com-
parison of the total area under the curve for any two graphs, contrasting the total
cumulative per cent survival. The statistical probabilities resulting from the ¢ test
using this method were approximately the same as when 50 per cent survival
temperatures were compared. Since the latter is a value commonly found in the
literature, this was chosen as a basis for comparison rather than the total cumula-
tive survival value. In all cases where a statistically significant difference has been
demonstrated using the ¢ test, the level of significance is based upon the 0.05
probability point, although the probability in some instances is less than the
0.01 level. 3

RESULTS
SUMMER

Average conditions of salinity and temperature representing the summer months
were 20° C. and 35 per cent sea water. Survival values obtained from summer
animals with no laboratory acclimation, holding experiments, were considered as
the base line curves. These curves were compared with heat tolerances of animals
acclimated to-20°C.,;-75 per cent: sea water; 5°-C., 35 per cent ‘sea "ware eae
5° C., 75 per cent sea water. Temperatures at which 50 per cent survival occurred

for 12 and 24 hours were used as the basis for comparisons.

Holding Experiment: 20° C. and 35 per cent sea water

Results of the base line experiments showed that H. oregonensis was more
resistant than H. nudus at both 12 and 24 hours. The temperature at which 50
per cent survival was found for H. oregonensis was 33.43° C. for 12 hours and
33.16° C. for 24 hours. For H. nudus the values were 32.95° C. for 12 hours and
32.00 C..for 24 hours: (rig. sl).

Acclimation Experiment: 20° C. and 75 per cent sea water

Both species showed a marked increase in tolerance after acclimation to 20° C.
and 75 per cent sea water. This combination of temperature and salinity differed
from the base line conditions only in the salinity, having been changed from 35
per cent to 75 per cent sea water. H. oregonensis again showed more tolerance
than H. nudus with values of 34.50° C. for 12 hours and 34.38° C. for 24 hours,
an increase of 1.07° C. and 1.22° C., respectively, over the base line values. The
50 per cent survival temperatures for H. nudus were 33.62° C. and 33.45° C. for

eS

6 ee ane J ee)

ES aS

TEMPERATURE TOLERANCE IN CRABS 155

12 and 24 hours, an increase of 0.67° C. and 0.95° C. (Fig. 2). The increase found
in both species was significantly different from the base line values, P = .05. It
is important at this point to note that acclimation to the above conditions pro-
vided the most favourable background for withstanding the high test tolerance
temperatures.

Acclimation Experiment: 5° C., 35 per cent sea water

Low temperature and low salinity combined appeared to be slightly less favour-
able for survival during the acclimation period as the number of deaths in both
species was greater than with any of the other combinations, but with fewer deaths
among H. nudus. Results showed the tolerance to be much the same as the base
line values, with no significant difference for 50 per cent survival, 33.28° C. and
33.11° C. for H. oregonensis and 33.38° C. and 32.64° C. for H. nudus for 12
and 24 hours (Fig. 3).

Acclimation Experiment: 5° C., 75 per cent sea water

Acclimation to these conditions did not result in much change in the tolerance
when compared with base line conditions. It is possible that the time allowed
for acclimation was not of sufficient duration, since animals collected in the
winter, when these conditions were found in the field, did show a lowering of
the tolerance. However, no change was detected in the temperature at which
50 per cent survival occurred in the summer animals after seven days’ acclimation
compared with a total of thirteen days’ acclimation. The temperatures for 50 per
cent survival for H. oregonensis were 33.20° C. and 32.73° C. and for H. nudus,
33.05° C. and 32.58° C. for 12 and 24 hours (Fig. 4). These values did not have
a significant difference from those found in the base line experiments. When these
animals, however, were compared with animals acclimated to 20° C., 75 per cent
sea water (Fig. 2), a significant difference in tolerance was found, P = .05.
Summer animals, then, when acclimated to both high and low temperatures with
high salinity constant, demonstrated a reduction of thermal tolerance with the
low temperature acclimation.

WINTER

The winter base line represented the heat tolerance of animals collected during
the winter months. These were the winter holding experiments. Also, winter
animals were acclimated to the three remaining combinations: 20° C., 35 per cent
sea water; 20° C., 75 per cent sea water; and 5° C., 35 per cent sea water.

Holding Experiment: 5° C., 75 per cent sea water

The 50 per cent survival temperatures were 32.59° C. and 32.43° C. for H.
oregonensis and 32.00° C. and 31.79° C. for H. nudus for 12 and 24 hours, respec-
tively (Fig. 4). The difference between these winter values and those obtained
for summer animals acclimated to these same conditions (winter) is significant,

P= .05.

HEMIGRAPSUS NUDUS
SUMMER BASE LINE

20°C, 35% SEA WATER

HEMIGRAPSUS OREGONENSIS
SUMMER BASE LINE

20°C, 35% SEA WATER

100
90
80
70
60
50
@|2 HOURS
40 24 HOURS —>
= |2 HOURS
30 24 HOURS —>
20 O
ee |
< lO
Zz
260) ® || 0 @
oc 32.50 32.95 33.16 33.43
a
op)
i HEMIGRAPSUS NUDUS HEMIGRAPSUS OREGONENSIS }
a WINTER WINTER
© 20 °C, 35% SEA WATER 20 °C, 35% SEA WATER
a lOO @— @—©— ©
LJ — — —<
oO. \ re
90
80
70
60
50
40 24 HOURS —>
<< 12 HOURS
30 24 HOURS —>
20 *
10)
Q aes I
32.229 ¥ 32.54 33.04¥ ¥ 33.38
28). 29+ 530... Sli 432.335 34 -S5 28529" DSN SCH SS! CSAS
TEST TEMPERATURE °C,
Figure 1. The influence of a previous history of 20° C., 35 per cent sea water on the

50 per cent survival values at 12 (@) and 24 (CO) hours.
are shown.

Both species in summer and winter
Each point represents the average survival of at least thirty animals.

156

100

ele)

PER CENT SURVIVAL.

HEMIGRAPSUS NUDUS

SUMMER
20 °C, 75 % SEA WATER

HEMIGRAPSUS OREGONENSIS

SUMMER
20 °C, 75 % SEA WATER

—. -- @——-6-—@—08

DX

24 HOURS->

24 HOURS —>

lle e Ihe
33.45 WW 33.62 34.38 W 34.50

HEMIGRAPSUS NUDUS
WINTER

20 °C, 75% SEA WATER

HEMIGRAPSUS OREGONENSIS
WINTER

20 °C, 75% SEA WATER

@—O—©— ©—_ ©— ®

“\

24 HOURS —®>

O ©
33.45 34.32 33.66 34.40

a pea ae oe os 34 35 Ch ed, OD OO oe a ae ~ Oe, OO

Teot LewMtemArume ) C.

Ficure 2. The influence of a previous history of 20° C., 75 per cent sea water on the
50 per cent survival values at 12 (@) and 24 (©) hours. Both species in summer and winter

are shown.

Each point represents the average survival of at least thirty animals.

157

158 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

Acclimation Experiment: 20° C., 35 per cent sea water

Crabs collected during the winter months and acclimated to the above condi-
tions, which corresponded to the summer base line combination, were found to have
survival values quite similar to those found in the summer months. The 12 and
24 hour 50 per cent survival temperatures were 33.38° C. and 33.04° C. for A.
oregonensis, and 32.54° C. and 32.22° C. for H. nudus, the greatest difference
being only 0.41° C. from the summer months (Fig. 1). This meant that in less
than one week the winter animals were able to approach the degree of resistance
found in the summer animals which may have had several months under these
conditions in the field. There was a significant difference in the 50 per cent
survival points between winter animals acclimated to summer base line conditions,
and the winter base line.

Acclimation Experiment: 20° C., 75 per cent sea water

Acclimation to these conditions provided the most suitable environment to with-
stand the test tolerance temperatures; there was a marked increase in the tem-
perature at which 50 per cent survival occurred, from the base line. Again, the
animals acclimated in the winter approximated the tolerance found in summer
animals acclimated to these same conditions, and were significantly different from
the winter base line. In one instance, H. nudus at 12 hours, the tolerance was
even greater. The 50 per cent survival values for H. oregonensis were 34.40° C.
and 33.66° C., and were 34.32° C. and 33.45° €. for H.. nudus for 12 aadyZ

hours (Fig. 2), an average increase of about 2° C. over the base line.

Acclimation Experiment: 5° C., 35 per cent sea water

Animals acclimated to the low salinity, low temperature combination showed
the greatest difference, both from the summer counterpart and the winter base
line. The 50 per cent survival temperatures for H. oregonensis were 30.18° C.
and 30.13° C. tor 12-and 24 hours, and for A: wudus 28:81° ©. and Zacaane
(Fig. 3). This was a drop of approximately 2.5° C. for H. oregonensis from the
winter base line conditions and 3° C. from the survival temperatures obtained
from summer animals acclimated to these same conditions. The drop in the sur-
vival temperatures for H. nudus was even greater, being about 3.0° C. and 4.5° C.,
respectively. When winter animals acclimated to both high and low temperatures
with low salinity constant are compared, the low temperature acclimation results
in a significantly different reduced tolerance (P = .05). During the acclimation
period, the number of deaths was greater in both species than at any other tem-
perature-salinity combination, a similar result to that found with the summer
animals. Contrary to H. nudus being the slightly more resistant during the
acclimation period in the summer animals, that species was less resistant than
H. oregonensis in the experiments involving winter animals.

As seen from the results of the above experiments, the 50 per cent survival
temperatures for both 12 and 24 hours were higher for H. oregonensis than for
H. nudus in all except one case when the value for H. oregonensis was only 0.10° C.
lower. Usually the 50 per cent survival temperature for H. oregonensis was higher
by a much greater amount than this, being in some cases over one degree. Analysis

HEMIGRAPSUS NUDUS HEMIGRAPSUS OREGONENSIS
SUMMER SUMMER
5 °C, 35% SEA WATER 5°C,35% SEA WATER

100 ae Se

3

24 HOURS

i | Wa
32.64 33.38 33.11 W 33.28

HEMIGRAPSUS NUDUS
WINTER
5°C, 35% SEA WATER

HEMIGRAPSUS OREGONENSIS
WINTER
5 °C,35% SEA WATER

PER CENT SURVIVAL

100

12 HOURS
24 HOURS ->

<@ |2 HOURS

30.13 | 30.18
ao aS Oa eS TS SH SS 28 29 30 Sl 32 35 34 35

TEST. TEMPERATURE: °C.

Ficure 3. The influence of a previous history of 5° C., 35 per cent sea water on the
50 per cent survival values at 12 (@) and 24 (©) hours. Both species in summer and winter
are shown. Each point represents the average survival of at least thirty animals.

159

PERCENT? SURVIVAL

100

100

HEMIGRAPSUS NUDUS
SUMMER
5 °C, 75% SEA WATER

@— ©—_6—_ ©——

<< |2 HOURS
24 HOURS —>

O

32.58 33.05

HEMIGRAPSUS NUDUS
WINTER BASE LINE
5 °C, 75% SEA WATER

@—— ©— © —_-®

<= {2 HOURS
24 HOURS —>

3179 {| 32.00

28 2,30 Sl

32. 33 34 35

HEMIGRAPSUS OREGONENSIS
SUMMER
5°C,75% SEA WATER

©— ©— ©— ©

@ [2 HOURS

32.73 33.20

HEMIGRAPSUS OREGONENSIS
WINTER BASE LINE
5°C, 75% SEA WATER

eo—oO—e— ©

\

<< (2 HOURS
24 HOURS —>

O°.
™S
32.43 Ww 32.59

28; 29- SO, Shy S2u 33054, 3a

TEST... EMPERATURE <'C:

Figure 4. The influence of a previous history of 5° C., 75 per cent sea water on the
50 per cent survival values at 12 (@) and 24 (©) hours. Both species in summer and winter
are shown. Each point represents the average survival of at least thirty animals.

160

DEMPERATURE,TOLERANCEVIN CRABS 161

TABLE [|

The change in 50 per cent survival temperature with various temperature-salinity combinations

SUMMER WINTER
A. nudus HA. oregonensts H, nudus H. oregonensis

ar 35% 12 hr. S205" C.* 33.45" 32.54 33.38

sea water 24 32.50 Solo be SWIG 33.04
a0 C., 15% 12 hr. 33.62 34.50 34.32 34.40

sea water 24 33.45 34.38 33.45 33.66

ake 85% 12 hr. 33:98 33.28 28.81 30.18

sea water 24 32.64 Saul! 28.68 30.13

ac. 15% 12 ir. 33205 33.20 32.00* o2.00F

sea water 24 32.58 BOP TIS) Se 7oF oye bey’

* Base line tolerances in the two seasons, summer and winter.

of variance on summer data indicated that H. oregonensis was the more resistant
at lethal temperatures, P = .01. The lower resistance is surprising in H. nudus
as the species is more commonly found at a higher tide level, and it would be
expected, a priori, that a greater tolerance for high temperatures would be of
more advantage to that species. Also, the similarity of the 50 per cent survival
temperatures for 12 and 24 hours within each species was consistent throughout.
At each test temperature the values were very close, and in many cases were
identical. Table I summarizes the results.

Salinity had a pronounced effect on the amount of variability that was obtained
in a series of replications of one of the experimental conditions. Whether the
acclimation temperature was 20° C. or 5° C., but particularly with the low acclima-
tion temperature, the low salinity (35 per cent sea water) resulted in much more
variable survival percentages at the test tolerance temperatures than those found
with the higher salinity. Also, the temperature at which 100 per cent survival
occurred was a much more tenuous point with the lower salinity. This can be
seen readily in any of the low salinity graphs, as there is a gradual slope to the
final breaking-off point, whereas with the 75 per cent salinity, the break is abrupt,
directly from 100 per cent. It should be noted that the difference in temperatures
between the points where 100 per cent or close to 100 per cent survival is found
for 24 hours and complete mortality is very slight.

Effects of size, sex, crowding, and moulting

All weights of crabs did not appear to have the same resistance to the high
test tolerance temperatures. Smaller animals, in both species, seemed slightly
more tolerant than larger ones. There was no direct relationship between weight
and time to death, but results indicated that if any animal lived, it was usually
less than 1 gram. This seemingly greater tolerance in the smaller crabs was

162 MARY-ELIZAGETH TODD ANDI PAUL AL DEH NE

HEMIGRAPSUS NUDUS
WINTER BASE LINE
95°C, 75% SEA WATER

- cal TEST TEMPERATURE 31°C
Ge

==)

O 20
a

z

a

tw 15
>

>

ac

=)

(dp)

on 10
=

=

| 2 3 Is) 6 1 8
WEIGHT IN GRAMS

Figure 5. The effect of weight on the time survived in hours at 31° C. in Hemigrapsus
nudus. The animals had a previous history of 5° C., 75 per cent sea water. Each point
represents one animal. The regression line is eye-fitted.

prevalent when crabs from all temperature-salinity combinations were tested.
One graph was chosen to demonstrate this phenomenon (Fig. 5).

As mentioned previously, both males and females were used in the experiments
when the sex was found to have no influence on the resistance to the high tem-
peratures. To the contrary, crowding, that is more than 5 animals per 4 liters
of water, caused a marked lowering of resistance to high temperatures. A mini-
mum value of 0.333 liters of sea water per l-gram live animal weight was found
to be satisfactory. When less water was allowed, by including more animals per
dish, dying animals apparently affected the water even though they were removed
immediately upon death, because those remaining were more apt to die (Fig. 6).
Thus, with 5 animals per dish, the 50 per cent survival temperature for 12 hours
was 32.95° C., with 10-15 animals per dish, 30.68° C., and with 19-28 animals,
27.86° C. This crowding effect existed only at the high test tolerance temperatures.
As was mentioned in the previous section, when there were at least 35 animals
per dish at the acclimation temperature of 20° C., no mortality due to crowding
was noted.

Since only crabs with a fully hardened carapace were used in the experiments,
the effects of moulting on the temperature tolerance were determined from those
animals which moulted during the 24-hour course of the experiment. These ani-
mals invariably died, even at a temperature where 100 per cent survival normally
occurred. A possible change in permeability of the carapace after moulting may
have rendered the crabs less resistant to high temperatures.

TEMPERATURE TOLERANCE IN CRABS 163

HEMIGRAPSUS NUDUS
SUMMER BASE. LINE
20°C 35% SEA WATER

100

80

SS)
ANIMALS

60

ANIMALS

/ DISH
40

PER CENT SURVIVAL

20

24 26 28 30 32 34
TEST TEMPERATURE C.

Figure 6. The influence in Hemigrapsus nudus of differing densities of crabs per 4 liters
of sea water on the 50 per cent survival temperature at 12 hours with a previous history of
20° C., 35 per cent sea water: 5 animals per dish (©), 10 to 15 animals per dish (@), 19 to
28 animals per dish (A). Each point represents the average survival of at least thirty animals.

fo)

DISCUSSION

Acclimation to a high temperature, with resulting increase in the high tem-
perature tolerance of the species, has been demonstrated many times (Sumner
and Doudoroff, 1938; Brett, 1946; Mellanby, 1954; Spoor, 1955; McLeese, 1956).
The two species of crabs, H. nudus and H. oregonensis, studied here, would seem
to be no exception. Temperature tolerance in conjunction with salinity has been
studied less extensively, and it was of interest to find that the salinity had a very
marked effect on the temperature tolerance of the species. Although results from

164 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

the survival curves occasionally differed in summer animals as opposed to winter
animals, the trend with regard to tolerance in any particular temperature-salinity
combination was the same in both seasons and both species, but there was a
definite species difference.

During the summer months, the environmental temperature is high and the
salinity is low. When both species of winter animals were acclimated to these
conditions, it was found that these acclimated animals had a tolerance at high
temperatures which was approximately as great as that found in the summer
animals. Likewise, these winter animals when acclimated to the high salinity as
well as to the high temperature, demonstrated in both species, again a temperature
tolerance very similar to that found in summer animals acclimated to these same
conditions, by far the most favourable combination to resist heat death. It is evi-
dent that these animals can regain their tolerance to high temperatures very rapidly
after a low temperature history by acclimation in the upper part of the physiolog-
ical temperature range. This rapid gain of heat tolerance has been documented
many times. Mellanby (1954) reported that the heat coma point is shifted with
experimental acclimation at the upper end of the tolerable range in the mealworm,
Tenebrio molitor and mosquito, Aedes aegypti. The gain in tolerance was rapid,
as twenty hours’ acclimation produced as much change as longer acclimation periods.
Spoor (1955) found complete gain of heat tolerance in the crayfish in less than
24 hours with an approximate 18° C. change in temperature (4° C. to 23° C.).
Ohsawa (1956a) determined that gain in heat tolerance by the periwinkle was
complete in less than 48 hours when acclimated to 30° C. from 10° C. One excep-
tion to a fairly rapid gain in heat tolerance is a recorded time of about twenty-two
days for total acclimation in the American lobster when transferred from 14.5° C.
to 23° C. (McLeese, 1956). McLeese suggested that the reason for this unusual
length of time is that the lobster has a long latent period before onset of demon-
strable acclimation, in this case ten days.

A rapid gain of heat tolerance would be of extreme advantage to most inter-
tidal animals. In those animals permanently submerged, a quick augmentation
would appear to be of little direct advantage, as bodies of water only slowly change
in temperature. For intertidal animals, the increase in temperature may be as
much as 10° C. over a period of a few hours in the tidal rhythm. At Spanish
Bank, the collecting area for H. nudus and H. oregonenstis, tide pools have been
recorded at 26° C. to 28.5° C. In this case, a rapid gain in tolerance, particularly
over a period of a few hours, would be most advantageous. Probable increase in
salinity, due to evaporation in tide pools during the low tide period in summer
months, aids the animal in resisting the high temperatures, as experimental re-
sults have indicated that high salinity provides the most favourable environment.

Regarding the low temperature series, the results can not be interpreted as
readily. The seasonal difference is marked when compared with the difference
found in the high temperature series. The change was consistently to a lower
temperature for the 50 per cent survival value in the winter animals. For the
winter base line, 5° C., 75 per cent sea water, a drop of about 1° C., from summer
animals acclimated to these same conditions, was found in both species at 12 and
24 hours. With the low salinity, low temperature combination, loss in heat tol-
erance is much greater. In this low temperature series it is clear that the seasonal

TEMPERA TUREYTOLERANCE IN’ CRABS 165

alteration in the field environment produces a noticeable reduction of tolerance in
the winter animals to high temperatures.

Seasonal change in heat tolerance of a species has been noted by Edwards
and Irving (1943) who found a 10° C. difference in the death point of Emerita
talpoida with summer animals higher than winter ones. Brett (1944) found a
change in upper lethal temperatures in Algonquin Park fishes from spring to fall.
In most cases, the species studied have been acclimated either solely to various
temperatures in the laboratory and the temperature tolerance studied, or the toler-
ance has been tested at different seasons with no laboratory acclimation. Examples
of seasonal comparisons in thermal resistance with the same acclimation temperature
in summer and winter are tolerance studies on several species of fresh water fish
(Hart, 1952), planaria (Schlieper and Blasing, 1953, as reported in Fry, 1958) and
the rainbow trout (Keiz, 1953). These species were more tolerant of high tem-
peratures in the summer months. Ohsawa (1956a) found that summer _ peri-
winkles responded at a higher temperature than winter animals when tested over
the same range. Prior to the test, the summer animals had a 48-hour history at
10° C. which corresponded to winter field conditions.

As in all acclimation experiments in this investigation, the summer animals
of the low temperature series were kept one week before survival was determined
at the test tolerance temperatures. There was no change in the tolerance over the
testing interval, seven to thirteen days. This period of approximately two weeks
was in agreement with that found necessary by other investigators to demonstrate
laboratory acclimation to low temperature. It was evident from experiments
on animals collected in winter months that summer animals acclimated to winter
field conditions of temperature and salinity in no way approached the reduced
tolerance found in winter animals. Likewise, summer animals acclimated to
low temperature plus low salinity had a much greater tolerance than was found
with the corresponding acclimation condition in winter animals. Insufficient
length of time for acclimation is an obvious possibility, but it would be thought
that some change would have shown in the tolerance over the period of two weeks.
A long latent period before the onset of acclimation to the low temperature is
feasible, as a latent period of 10 days was reported in acclimation to a higher tem-
perature in the lobster (McLeese, 1956). Another possibility is that some other
factor or a combination of factors influences the acclimation to low temperatures
but not to the reverse, as winter animals which were acclimated to summer condi-
tions gave approximately the same results as summer animals. Dehnel (1958)
has shown in these two species of crabs, H. nudus and H. oregonensis, that a
simulated seasonal variation of light in the laboratory has a pronounced effect
on metabolism measured by oxygen consumption. It is feasible that summer
animals would never have the same temperature tolerance as that found in the
winter animals solely by acclimating them to winter temperature and _ salinity
conditions. Hoar (1956, p. 365) from experiments on goldfish “. . . concluded
that the seasonal variations in thermal tolerance previously noted in fish main-
tained under constant temperature conditions are photoperiodically controlled... .”
It is hoped that future experiments will determine this relationship more clearly
in these crabs.

Recently, further experiments were carried out on this problem. There was

166 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

an attempt to acclimate summer crabs of both species to the low salinity, low
temperature combination to determine whether results similar to those found in
winter could be obtained. This combination was chosen as acclimation to this
temperature-salinity condition resulted in greatly reduced tolerances in winter.
The summer animals were acclimated for a total of thirty-one days in some experi-
ments. Resistance appeared to be slightly less when compared with tolerances
found after shorter periods of acclimation, but was not as low as those found in
winter. Also, experiments were done where the animals had an eight-hour photo-
period corresponding to light conditions in winter with the above temperature-
salinity relation. Results again indicated a slightly reduced tolerance but did not
appear to differ from those of animals kept in complete darkness. Since two
weeks’ acclimation in the summer animals did not bring about a change over the
testing period, it is important to note that animals cannot be said to acclimate
or not acclimate to a parameter merely because there is no evident change.

In nearly all instances low salinity resulted in a lower value for the 50 per
cent survival temperature than with high salinity, whether the temperature was
5° C. or 20° C. Gross (1957) gives values for H. nudus and H. oregonensis,
calculated from Jones (1941), which indicate the osmotic gradient maintained by
these two species when placed in various concentrations of sea water. The crabs
are isotonic to the external environment at 100 per cent sea water (based on
34.6%). At 75 per cent sea water there is an extremely slight gradient main-
tained, about 2.5 per cent sea water in H. oregonensis and 8.0 per cent in H.
nudus. The gradient maintained at 35 per cent sea water, however, is about
36 per cent in both species. These values cannot be applied directly to the present
experiments, but certainly they can indicate the trend. The maintenance of this
gradient presumably results in more work being done by the animal at the lower
salinity, and in a greater stress placed on it than at 75 per cent sea water where
the gradient is only a few per cent. Additional evidence which suggests more
osmotic work is performed at the lower salinity in these crabs is a 22 per cent
increase in respiratory rate at 35 per cent over that found at 75 per cent sea water
in a 1.0-gram animal measured at 10° C. (Dehnel, 1960). It is probable that
the lower 50 per cent survival temperatures with the lower salinity is due to
the increased metabolic work necessary at that salinity. The lethal or near lethal
temperatures in the test tolerance experiments presumably alone are causing a
marked strain on the metabolic activity of the crabs and the additional strain of
maintaining a large gradient between the blood and the external medium in the
low salinity results in the animal dying at a lower temperature.

In the present investigation, low salinity, combined with low temperature,
particularly in the winter animals, was the most disadvantageous acclimation com-
bination for withstanding the test tolerance temperatures. Wikgren (1953) showed
the greater loss of ions with temperature decrease when transferred from tap to
distilled water. He thought that this greater loss at the lower temperatures could
be assigned to depressed absorption of ions and chloride.

A similar relation between salinity and temperature was found by Broekema
(1941) in the shrimp, Crangon crangon. ‘The salinity optimum for length of life
depended on the temperature. With a rise in temperature, the salinity optimum
was reached by a downward shift in the salinity. High temperatures generally

Pe vPERATURE TORR RANCEIN CRABS 167

were less favourable for survival than low, but low salinity was endured better with
a higher temperature. This was similar to the pattern in the two species of crabs
studied in these experiments ; with low salinity, temperature tolerance was increased
when there was a previous high temperature history rather than a low, particularly
in winter animals. In addition Broekema showed that the difference between
the internal blood concentration and external concentrations in the shrimp was
greater at high than at low temperatures and suggested that at low temperatures
the limits of life are exceeded sooner than when the temperature is high. Jones
(1941) demonstrated a higher osmotic pressure in the blood of H. nudus and
H.. oregonensis with a higher temperature, signifying an increase in the degree
of regulation. Thus, with a difference in temperature acclimation at the low
salinity it is probable that the difference in sustained gradients could carry through
to the temperature tolerance tests, and thereby affect the lethal point. Jones,
however, showed a great variation in the osmotic pressure of the blood when these
crabs were near death from air exposure and felt that this tended to refute a
certain lethal osmotic pressure as the primary cause of death.

Broekema’s findings, demonstrating that shrimp in a high salinity lived longer
at a low temperature, at first appear contrary to the results of this investigation,
where the most favourable history for resisting high temperatures was one of both
high temperature and high salinity. It must be remembered, however, that the
criterion for optimal conditions was quite different, ability to resist lethal tem-
peratures in this investigation, whereas in Broekema’s experiments the length of
life was the factor. The results cannot be compared directly, but it is felt that
they both show the same trend, particularly with respect to low salinities.

Kinne (1958) reported the shifting of heat tolerance in three species of animals,
the polychaete Nereis diversicolor, the amphipod Gammarus duebeni, and the iso-
pod Sphaeroma hookert. The salinity of the pond from which all three species
were collected was about 12% , and in all three species when kept at salinities
below this, heat resistance is lowered. Increased heat resistance results in animals
from the higher salinities, above 12%. Kinne suggested that the alteration of
water and ion balance resulting in increased water content at extremely low
salinities decreases the heat tolerance ; the lowered water content at higher salinities
favourably affects the resistance to high temperatures. The range of test salinities
was beyond the limits of regulation in Gammarus duebem from values given by
Verwey (1957) and at the upper salinities in Nereis diversicolor (Smith, 1955b).
Another explanation than alteration of water content may be necessary when
similar changes in tolerance occur over the physiological regulatory range where
variation is slight. Since the salinity of the pond where the animals were collected
was only 12%, this would result in a fairly large sustained gradient at least in
two of the three species for which data are available. The salinities could cause
increased susceptibility to lethal temperatures. The higher salinities presumably
within limits would approximate more nearly natural conditions in the body fluids
of the animals, to provide a more favourable environment for survival. Another
species studied by Kinne (1956), the hydroid Cordylophora caspia, has slight
osmoregulatory abilities, and again was found to withstand a high temperature
better at a high salinity than at lower ones. Possibly the explanation here could
be that the higher salinity is nearer the concentration of body fluids of the hydroid,

168 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

resulting in a more favourable environment to resist a stress, in this case, high
temperature.

The geographic limitation of the polychaete Nereis is related to the salinity-
temperature balance of its environment. The range of Nereis diversicolor in the
Baltic Sea apparently is limited by the colder parts of the year when it could
not withstand the low salinity expressed as a chlorinity of 4 grams per liter in
which it is able to regulate adequately in the warmer summer months (Smith,
1955a). Studies on Neanthes lighti indicated that it was able to survive in fresh
water in Lake Merced, California, because of its viviparous mode of reproduction
where the young are sufficiently developed at birth that they are able to osmo-
regulate, and that the species is not exposed to severe winter cold. Laboratory
experiments showed that the ability to regulate in water of a chlorinity of 1.0 to
0.09 grams per liter was inhibited at 1.5° C. but possible at 13° C. in worms
tested in vitro (Smith, 1957).

Animals acclimated to temperatures in the upper part of their physiological
temperature range show a remarkable similarity in lethal temperatures. A few
comparisons for estimated 12 hours median tolerance among animals acclimated
to approximately 20° C. and a favourable salinity, if marine forms, are given in
Table Il... For the most part: the lethal tenperature:is at least 30- @.aeieene
species difference found in this investigation, the greater thermal resistance of
Hi. oregonensis is surprising since it occupies a lower level on the beach. Within
the same species, Littorina litorea, the individuals collected at high tide level had
greater resistance to lethal temperatures than those collected a few yards away
at low tide level (Gowanloch and Hayes, 1926). Ohsawa (1956b) determined
species difference for the periwinkles Nodtlittorina granularis and N. vilis. The
latter occurs in the upper zone of the wide range occupied by N. granularis, which
was found to be the more tolerant at lethal temperatures, a similar result to that
found with H. nudus and H. oregonensis.

In most animals, then, and possibly all, a previous high temperature history
increases heat tolerance and a low one decreases it. Along with this is the influ-
ence of salinity where low salinities almost invariably are endured better at a
higher temperature, and high temperature, high salinity combined being most
favourable for withstanding upper lethal temperatures. Low salinity, low tem-
perature generally is least favourable for length of lite and for heat tolerance
(Hies7 )s

TaBLe II
A comparison of tolerances to high temperature in several species

12 hour median tolerance,
eC

Species Source
Hemigrapsus nudus 33.62 —
HZ, oregonensis 34.50 —
Orconectes rusticus 36.4 Spoor, 1955
O. propinquus 35:0 Bovbjerg, 1952
Cambarus fodiens 35.0 Bovbjerg, 1952
Homarus americanus 28.4-30.5 McLeese, 1956
Fyalella azteca 33.0 Bovee, 1949
Ameturus nebulosus 33.4 Brett, 1944
Carassius auratus 34.5 Fry, Brett and

Clawson, 1942

TEMPERATURE TOLERANCE IN CRABS 169

en ©.

50> PER -CENE SURVIVAL: Geen.

SOO 9 °C

Ficure 7. Median tolerance temperatures for Hemigrapsus oregonensts for 12 hours with
previous histories of 20° C. or 5° C. and 35 per cent or 75 per cent sea water. Animals
were collected in winter.

The effect of size and sex on heat tolerance was found to vary in different
animals reported in the literature. In H. nudus and H. oregonensis, smaller ani-
mals seemed to be slightly more resistant to the high test tolerance temperatures
than larger individuals. There was no difference in tolerance between the sexes.
Edwards and Irving (1943) found no difference in tolerance in Emerita talpoida
between males and females, but larger animals seemed slightly more resistant than
small ones. In lethal temperature experiments on young speckled trout, Salvelinus
fontinalis, size did not affect time to death at high temperatures in the same age
group (Fry, Hart and Walker, 1946). Hoar (1955) found an increased tol-
erance at 1°-2° C. in goldfish with an increase in size from 3 to 7.5 cm. In the
crayfish, sex and size did not affect heat tolerance (Spoor, 1955). McLeese
(1956) concluded that in the size range of lobsters studied, from 21 to 28 cm.,

170 MARY-ELIZABETH TODD AND PAUL A. DEHNEL

there is identical response to upper lethal temperatures. Kinne (1958) showed
a decreased tolerance in female Gammarus duebeni to high temperatures, and
increased tolerance in smaller (younger) animals of both sexes. In Sphaeroma
hookeri, smaller males appeared more resistant than larger ones. There was ap-
parently no sex difference in this species. Thus, it is seen that no conclusions can
be drawn regarding the effect of size or sex on heat resistance.

The amount of water available to each animal during the experiments testing
heat tolerance was found to influence markedly that tolerance even though animals
were removed at death. At least 0.333 liter of water per gm. live animal weight
per 12 hours was necessary to prevent death from crowding. The amount of
water provided per gm. varies widely in different investigations. Gowanloch and
Hayes (1926) tested 10 to 15 periwinkles per 250 cc. of water (0.250 liter),
less water than that found necessary in the lethal temperature experiments on
these crabs. Brett (1952) tested 40 fish (average weight about 1 gm.) in a
lethal bath twenty-two inches square by eleven inches deep. This gave about
2.2 liters per gm. Also the exchange of water was equal to the volume of the
tank every 24 hours. Spoor (1955) provided at least 0.1 liter of water for
each animal being tested. The size range was from 17 to 42 mm. McLeese (1956)
tested 10 lobsters at each constant temperature. Taking a value of 400 gm. as
an average weight of the experimental animals, only about 14 cc. (0.0139 liter)
at the most was allowed per gram. This was very much less than would be
expected if the amount for the decapod Crustacea found in this investigation
can be related. A small flow of water into the tank presumably allowed for some
exchange of water, but no values are given. Also, the lethal temperatures are
low both in comparison with other Crustacea and particularly with the values
obtained by Huntsman (1924) for stages IV and V in the lobster. An estimated
lethal temperature of about 34° C. is greater than that obtained by McLeese under
the most favourable acclimation conditions. Too many animals per volume of
water may account in part for results. In many cases in the literature, the volume
of water used in the experiment is not indicated, but presumably tests were con-
ducted to ensure a satisfactory density of animals. It is probable that the volume
needed varies in different species.

Moulting during the 24 hours of the experiment was found to have an adverse
effect on high temperature tolerance in H. nudus and H. oregonensis. Crayfish
moulted successfully in temperatures from 12° C. to 36°\C€.; the stage maeeme
moult cycle had no effect upon heat tolerance (Spoor, 1955). To the contrary,
McLeese (1956) showed that the average survival time for soft-shell lobsters
was less than that for hard-shelled individuals. As results from various workers
differ, nothing definite can be concluded about a relationship between moulting
and death at high lethal temperatures.

SUMMARY

1. The influence on heat tolerance was determined of seasonal change and
laboratory acclimation to various temperature-salinity combinations, for two species
of grapsoid crabs, Hemigrapsus nudus and H. oregonensis.

2. There was a seasonal change in 50 per cent survival in both species when
base lines from summer and winter were compared.

TEMPERATURE TOLERANCE IN CRABS 171

3. A definite species difference in tolerance to high temperatures was found
to exist, but both species reacted similarly to any particular temperature-salinity
combination.

4. Acclimation to a high temperature generally increased the resistance to
lethal temperatures whereas acclimation to low salinity generally decreased it.
High temperature, high salinity was the most favourable combination to with-
stand the high test tolerance temperatures.

5. Gain in heat tolerance whether the salinity was low or high was rapid, less
than one week.

6. Winter tolerances with both low and high salinities in the low temperature
series were not demonstrated in the laboratory with summer animals acclimated to
these same conditions. Various reasons are suggested which might explain this
apparent discrepancy.

7. Moulting during the test tolerance experiments adversely affected the re-
sistance. The number of animals per dish at each test temperature had a pro-
nounced effect on tolerance. The sex of the crabs did not affect the survival, but
smaller animals appeared to be slightly more resistant.

LITERATURE CITED

Bovpyerc, R. V., 1952. Comparative ecology and physiology of the crayfish Orconectes pro-
pinquus and Cambarus fodiens. Physiol. Zo6l., 25: 34-56.

Bover, E. C., 1949. Studies on the thermal death of Hyalella azteca Saussure. Biol. Bull., 96:
123-128.

Brett, J. R., 1944. Some lethal temperatures of Algonquin Park fishes. Pub. Ont. Fish.
Res. Lab., No. 63: 5-49.

Brett, J. R., 1946. Rate of gain of heat tolerance in goldfish (Carassius auratus). Pub. Ont.
Fish. Res. Lab., No. 64: 5-28.

Brett, J. R., 1952. Temperature tolerance in young Pacific salmon, Genus Oncorhynchus.
J. Fash. Res. Bd. Can., 9: 265-323.

BroekeMA, M. M. M., 1941. Seasonal movements and the osmotic behaviour of the shrimp,
Crangon crangon L. Arch. Néerl. Zool., 6: 1-100.

Buttock, T. H., 1955. Compensation for temperature in the metabolism and activity of poikilo-
therms. Biol. Rev., 30: 311-342.

DEHNEL, P. A., 1955. Rates of growth of gastropods as a function of latitude. Physiol. Zodl.,
28: 115-144.

Deunet, P. A., 1958. Effect of photoperiod on the oxygen consumption of two species of
intertidal crabs. Nature, 181: 1415-1417.

DEHNEL, P. A., 1960. Effect of temperature and salinity on the oxygen consumption of two
species of intertidal crabs. Bzol. Bull., 118 (in press).
DeEHNEL, P. A., AND E. SEGAL, 1956. Acclimation of oxygen consumption to temperature in
the American cockroach (Periplaneta americana). Biol. Bull., 111: 53-61.
Epwarps, G. A., ANd L. Irvine, 1943. The influence of temperature and season upon the
oxygen consumption of the sand crab, Emerita talpoida Say. J. Cell. Comp. Physiol.,
21: 169-182.

Fry, F. E. J., 1958. Temperature compensation. Ann. Rev. Physiol., 20: 207-224.

Fry, F. E. J., J. R. Bretr anp G. H. Crawson, 1942. Lethal limits of temperature for young
goldfish. Rev. Can. de Biol., 1: 50-56.

Fry, F. E. J., J. S. Hart anp K. F. Wacker, 1946. Lethal temperature relations for a sample
of young speckled trout, Salvelinus fontinalis. Pub. Ont. Fish. Res. Lab., No. 66: 9-35.

Gowan ocu, J. N., AnD F. R. Hayes, 1926. Contributions to the study of marine gastropods.
I. The physical factors, behaviour and intertidal life of Littorina. Contr. Can. Biol.,
N.S., 3: 135-165.

Pi MARY-ELIZABETH TODD AND PAUL A. DEHNEL

Gross, W. J., 1957. An analysis of response to osmotic stress in selected decapod crustacea.
Biol. Bull., 112: 43-62.

Hart, J. S., 1952. Geographic variations of some physiological and morphological characters in
certain freshwater fish. Pub. Ont. Fish. Res. Lab., No. 72: 1-79.

Hoar, W. S., 1956. Photoperiodism and thermal resistance in goldfish. Nature, 178: 364-365.

HuntsMAN, A. G., 1924. Limiting factors for marine animals. 2. Resistance of larval lobsters
to extremes of temperature. Contr. Can. Biol., N.S., 2: 89-93.

HuntTsMAN, A. G., AnD M. I. Sparks, 1924. Limiting factors for marine animals. 3. Relative
resistance to high temperatures. Contr. Can. Biol., N.S., 2: 95-114.

Jones, L. L., 1941. Osmotic regulation in several crabs of the Pacific Coast of North America.
J. Cell. Comp. Physiol., 18: 79-92.

Kez, G., 1953. Uber die Beziehungen zwischen Temperatur-Akklimatisation und Hitzeresistenz
bei eurythermen und stenothermen Fischarten (Squalius cephalus L. und Trutta iridea
W. Gibb). Naturwiss., 40: 245-250.

Kinng, O., 1956. Uber den Einfluss des Salsgehaltes und der Temperatur auf Wachstum, Form
und Vermehrung bei dem Hydroidpolypen Cordylophora caspia (Pallas), Thecata,
Clavidae. Zool. Jahrb., Allg. Zool. u. Phys., 66: 565-638.

KINNE, O., 1958. Adaptations to salinity variations—some facts and problems. Reprinted from
Physiological Adaptations (Amer. Physiol. Soc., Wash., D. C.), 92-106.

Mayer, A. G., 1914. The effects of temperature upon tropical marine animals. Pap. Tortugas

Lab., 6: 3-24.
Metiansy, K., 1954. Acclimatization and the thermal death points in insects. Nature, 173:
582.

McLeeser, D. W., 1956. Effects of temperature, salinity, and oxygen on the survival of the
American lobster. J. Fish. Res. Bd. Can., 13: 247-272.

Ousawa, W., 1956a. The experimental acclimatization in the temperature response relation
and the heat tolerance of the periwinkle, Nodilittorina granularis (Gray). J. Inst.
Polytechnics, Ser. D, 7: 197-217.

Ousawa, W., 1956b. The species difference in the concentration and temperature response
relations and the heat tolerance of periwinkles. J. Inst. Polytechnics, Ser. D, 7: 219-227.

Ousawa, W., anp H. Tsuxupa, 1956. The seasonal variation in the temperature response
relation and temperature tclerance of the periwinkle, Nodilittorina granularis (Gray).
J. Inst. Polytechnics, Ser. D, 7: 173-188.

SCHOLANDER, P. F., W. Fiacc, V. WaLTERS AND L. Irvine, 1953. Climatic adaption in Arctic
and tropic poikilotherms. Physiol. Zoél., 26: 67-92.

SecaAL, E., 1956. Microgeographic variation as thermal acclimation in an intertidal mollusc.
Biol. Bull., 111: 129-152.

Situ, R. I., 1955a. On the distribution of Nereis diversicolor in relation to salinity in the
vicinity of Tvarminne, Finland, and the Isefjord, Denmark. Biol. Bull., 108: 326-345.

Situ, R. I, 1955b. Comparison of the level of chloride regulation by Nereis diversicolor in
different parts of its geographical range. Biol. Buill., 109: 453-474.

Smitu, R. I., 1957. A note on the tolerance of low salinities by nereid polychaetes and its
relation to temperature and reproductive habit. Ann. Biol., 33: 93-107.

Spoor, W. A., 1955. Loss and gain of heat tolerance by the crayfish. Biol. Bull., 108: 77-87.
Sumner, F. B., anp P. Doupororr, 1938. Some experiments on the temperature acclimatization
and respiratory metabolism in fishes. Biol. Bull., 74: 403-429.
Verwey, J., 1957. A plea for the study of temperature influence on osmotic regulation. Ann.

Biol., 33: 129-149.

WrxKcreN, B., 1953. Osmotic regulation in some aquatic animals with special reference to the

influence of temperature. Acta Zool. Fennica, 71: 1-102.

BIOLOGY MATERIALS

The Supply Department of the Marine Biological Labora-
tory has a complete stock of excellent plain preserved and
injected materials, and would be pleased to quote prices on
school needs.

PRESERVED SPECIMENS

| for
Zoology, Botany, Embryology,
and Comparative Anatomy

LIVING SPECIMENS

for
Zoology and Botany

including Protozoan and
Drosophila Cultures, and
Animals for Experimental and
Laboratory Use.

MICROSCOPE SLIDES

for
Zoology, Botany, Embryology,
Histology, Bacteriology, and
Parasitology.

CATALOGUES SENT ON REQUEST
:, ae
Supply Department

MARINE
BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS | RAG ALENT ca
a ' : /\
: Page |
BAGNARA, JOSEPH 1 . / iy
Tail a ita a caH of Xenopus i in normal development and regenera- |
BOR Vi NT ela Sl bert ah oat WS ae voit AF Staal AM OMT Us AUS. Na a |
BECKMAN, CAROLYN, AND ROBERT MENZIES |
The relationship of reproductive temperature ae the geographical
range of the marine woodborer Limnoria tripunctata. . . Conant ge ees

BOYD, CARL M. | }
The larval stages of Pleuroncodes planipes Stimpson (Crustacea, Ke

Decapoda, Galatheidae).............4. ig Be Ma are las a vole oe MAY Sex
CRISP, D. J., AND B. S. PATEL | ¥ iY :
The moulting cycle in Balanus balanoides De, An shin jaidle sh opace are OBR Sagal

DAVIS, HARRY C.
Effects. of icine ceo libeas? & materials in sea water on eggs arid nN
3 larvae of the clam (Venus ‘(Mercenaria) mercenaria).............. 48
EPPLEY, RICHARD 'W., AND CHARLES C. CYPRUS Vaca
Cation repuldibn and survival of the red alga, Ports perforata, in Vg
diluted and concentrated sea)water........ 0.6.00 0.0.00 ee edo. 65)
GESCHWIND, I. 1., M. ALFERT AND C. SCHOOLEY, aH Reg ke
- The éffects of thyron and prow teil hormone on lives Fate “iat OU
| GREGG, JAMES H. Bo
Surface antigen dynamics in the slime mold, Dictyostelium digs’ A
| COROT, SAT ie NL Ge Rie + Mina caictone ated yal Waa aban feel cele wee \> i i 70-
HANLON, DAVID P. NEN \ bee ay re
The effect of potassium deficiency on ‘the tee} amino atid ciebea of. a’
the muscle tissue of protein-maintained Fundulus heteroclitus. aro 79
HOLZ, GEORGE G., JR. ‘Ai! daa
Structural dua functional changes i ina generation in Tetrahymena. 84 me
KOHLER, KURT, AND CHARLES B. METZ | ‘ SOR al
Antigens of the sea urchin sperm surface........0..4..,4--4++4-. 96 \f
MILBURN, NANCY, ELIZABETH A. WEIANT AND KENNETH D. ROEDER |
The release of efferent nerve activity in the roach, rineatiaiariaven ameri- ,
cana, by extracts of the corpus cardiacum Nee kU) ult “Ao a (Rael vid, MEL
OHTSUKA, EIJI | Sa
On the hardening of the chorion of the fish egg after fertilization. |

III. The mechanism of chorion hardening in Oryzias latipes..... OND
PASSANO, L. M. | a

Low fel baratuve blockage of molting i in Uca pugnax ere | cg ig Be iy,
TRAVIS, DOROTHY F. oe B

The deposition of skeletal structures in the Crustacea. I. The his- ie i

tology of the gastrolith disc skeletal tissue complex and the Barton
in the crayfish, Orconectes (Cambarus) virilis Hagen—Decapoda. . 137

TODD, MARY-ELIZABETH, AND PAUL A. DEHNEL i
Effect of temperature and salinity on heat tolerance in two grapsoid *
crabs, Hemigrapsus nudus and Hemigrapsus oregonensis ........./ 150

}

Vohinans 118 . | | Number 2
514.0573 | :

THE

BIOLOGICAL BULLETIN

PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY |.

Editorial Board
DAVID W. BISHOP, Carnegie Institution of _ V. L. LOOSANOFF, U.S. Fish and Wildlife
Washington Service
HAROLD C. BOLD, University of Texas C. L. PROSSER, University of Illinois
FRANK A. BROWN, JR., Northwestern University BERTA SCHARRER, Albert Einstein College of
Joun B. Buck, National Institutes of Health Medicine

Lissrz H. HyMAN, American Museum of FRANZ SCHRADER, Duke University
Natural History WM. RANDOLPH TAYLOR, University of Michigan

J. LOGAN IRVIN, University of North Carolina CARROLL M. WILLIAMS, Harvard University

DONALD P. COSTELLO, University of North Carolina
Managing Editor

APRIL, 1960

Printed and Issued by
LANCASTER PRESS, Inc.

PRINCE & LEMON STS.
LANCASTER, PA.

INSTRUCTICNS TO AUTHORS

(te a

/ ‘.

The Biological Bulletin accepts papers on a variety of subjects of biological interest. f In
, general, however, review papers (except those written at the specific invitation of the Editorial —
Board), short preliminary notes and papers which describe only a new technique or method —

without presenting substantial quantities of data resulting from the use of the new method cannot - e

be accepted for publication. A paper will usually appear within three months of the date of
its acceptance. — a ; ur x aru ‘ ee A

The Editorial Board requests that manuscripts conform to the requirements set below;

those manuscripts which do not conform will be returned to authors for correction before they.

are refereed by the Board. : |
1. Manuscripts. Manuscripts: must be typed in double spacing (including figure. legends,

~ foot-notes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 83 by 11 inches. They.

should be carefully proof-read before being submitted and all typographical errors corrected

legibly in black ink. Pages should be numbered. A left-hand margin of at least 13 inches —

should be allowed.

2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on Lendaets sheets and

placed in correct-sequence in the.text.. Because of the high-cost of setting such material in type, ©

authors are earnestly requested to limit tabular material as much as possible. Similarly, foot-
notes to tables should be avoided wherever possible. If they are essential, they should be indi-

cated by asterisks, daggers, etc., rather than by numbers. Foot-notes in the body of the text /—

should also be avoided unless they are absolutely necessary, and the material. incorporated into

the text. Text foot-notes should be numbered consecutively and typed double-spaced on a sepa-,
rate sheet. Explanations of figures should be typed double-spaced and placed on separate sheets

at the end of the-paper. :

~ 8 A condensed title or tunning head of no more than 35 letters and spaces should be included. f

4, Literature Cited. The list of papers cited should conform exactly to the style set ina

recent issue of The Biological Bulletin; this list should be headed LITERATURE ‘CITED,

and typed double-spaced on separate pages. | ;

5. Figures. ‘The dimensions of the printed page, 5 by 73 inches, should be kept in mind in
preparing figures for publication. . Illustrations should be large enough so that all details will be
clear after appropriate reduction.. Explanatory matter should be included in legends as far as
possible, not lettered on the illustrations. Figures should be prepared for reproduction as line

cuts or halftones; other methods will be used only at the author’s expense. Figures to be repro- —

duced as line cuts should be drawn in black ink on white paper,’ good quality tracing cloth or ’

blue-lined coordinate paper; those to be reproduced.as halftones-should be mounted on Bristol ~

Board, and any designating numbers or letters should be made directly on the figures. Al] - ‘1
figures should be numbered in consecutive order, with no distinction between text- and plate- |
figures. .The author’s name should appear on the reverse side of all figures, as well as the desired

reduction.

6. Mailing. Manuscripts should be packed flat; large illustrations may be rolled in Spiatteaer vi

tube. All illustrations larger than 83 by 11 inches must be accompanied by photographic

teproductions or tracings that may be folded to page size. SS Gis

Reprints. Authors will be furnished, free of charge, one: hundred reprints without covers.
Additional copies may be obtained at cost; approximate prices will be furnished by the Managing ~

Editor upon request. —

THE BIOLOGICAL BULLETIN —

THE BIOLOGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc:, Prince and —

Lemon Streets, Lancaster, Pennsylvania.

Subscriptions and similar matter should be addressed to ‘The Biological Bulletin, Marine
Biological Laboratory, Woods Hole, Massachusetts. All subscriptions expire with the December
issue and are renewable prior to the next succeeding February issue. Single numbers, $2.50.
Subscription per volume (three issues), $6.00. |

Communications relative to manuscripts should be sent to the Managing Editor, Marine
Biological Laboratory, Woods Hole, Massachusetts, between June 1 and September. 1, and ‘to

Dr. Donald P. Costello, P.O. Box 429, Chapel Hill, North Carolina, during the remainder of | f

the year.

Second-class postage paid at Lancaster, Pa.

uae

i
Ne VA

Te

BACK ISSUES

The Marine Biological Laboratory
needs many numbers of

THE BIOLOGICAL BULLETIN
to replenish its stock, greatly de-
pleted after meeting the needs of
individuals and libraries during the
last 25 years. Persons willing to
contribute any numbers, volumes,
or runs may send them, express col-
lect, to the
Marine Biological Laboratory,

Woods Hole, Massachusetts

to minimize
glass breakage
specify...

FUSIFORM
BORE

THOMAS HR-108 SAFETY STOPPERS are of bloomless, black, resilient, heat resisting
composition, with low sulfur content and long life. Straight bore of the conventional stopper
has been replaced by a fusiform bore with bevelled entrance. Internal diameter at center
approximately twice that of a sealing band at either end. Offer the following advantages:
SAFETY—Glass tubes can be easily inserted and removed, particularly by students, with greater safety than

with straight bore. Insertion is facilitated by the bevelled entrance of the bore, and the tube slides through
only the two sealing bands.

LARGER TUBES—Fusiform bore permits insertion of larger tubes than is possible with straight bore. For
example, tubes 6 to 11 mm can be inserted in sizes 5 to 9 inel.

SUPERIOR COMPOSITION—The stock contains more than 70% of pure virgin gum, possesses great elas-
ticity, is free from surface sulfur and contains less than 0.75% free sulfur; no reclaimed rubber is used.

HEAT RESISTANT—Temperatures up to 110°C do not affect shape, size or structure.
LONG LIFE—Tests indicate useful life approximately five times that of common stoppers.

8807. Safety Stoppers, Thomas HR-108, as above 8808. Safety Stoppers, Thomas HR-108, as above

described, one hole (fusiform bore). described, but two hole.

Size Approx. no. Carton Per Per Size Approx. no. Carton Per Per
Number per Ib. Contains Dozen Carton Number per lb. Contains Dozen Carton
ae arian LI On en ol OSes AR ZAd oe Oven 007 6s ster 69.168. HS Tae
OIE oa JE lta beet eae LOS Es Ae Ar A a en 10.36 Oca TEL Occ toee LOS ace 1 Sit, meee 11.23
iar Nea ad oH eet cte Morden: 8 PisOeccnrs 4 10.02 We hy bon Soe t aa SGunt ee 1.480 ee 10.78
Tet NO ee as eae VAS ois: 8.07 7A Pe A Aeros: his a ate 158s 22 8.65
a Reedy ire DOF. oie OOK. is OP eee ee 7.60 Die er duaween es DO ay om OD vn te TiO eee 8.15
Bh Siox eee Ce eee AS cance. ae AeONi ae. 6.60 AN sate tae AOR see AS ee 1eh a 6.79
Semios tina: Aes Oi toed BO a: 202. ct 5.52 Siete oo reer Bee snk oe Le Megas 2.04 Sn 5.58
Dor, end: Obese: SO ie ae 2M Ouiee) vs 5.94 By ee Doda oe SO vane 20202 oe 6.23
Bie tates Dil edie OW) Ay ce Mes OL eee OMS 65 be 28. 2 BO ee ee

Oo eee 7a en Be SOR Mee ZHAO Rd: 5.68 OU are uses 7A eaest eee OU s taser 2. Ocoee 5.81

Oty ae ae ae UG ee oe oases Sel2 yey cuss. 4.26 oy SA oe Se ie ain ae LO as ee SUG ee 4.31
+ Re ge ase fart 1 bes eae LB ah eee: S164 aoe 4.97 Je eee Rey Sot OF se aera NS, Seca 3-60. ae 4.92

5% discount from carton prices in lots of 6 cartons; 10% discount from carton prices in lots of 12 cartons
20% discount from carton prices in lots of 24 cartons; 30% discount from carton prices in lots of 48 cartons
one size, or assorted in carton units

For more detailed description, see pages 1066 to 1068 of our general catalog

ARTHUR H. THOMAS COMPANY

A. H.T. co.

More and more laboratories rely on ia 4 Laboratory Apparatus and Reagents

Rages sau i a i ia Det me 5.05 (eater iene Dhectastors 183) 5.15
VINE ST. AT BRD e PHILADELPHIA 5, PA;
i

ik

Vol. 118, No. 2 April, 1960
THE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

Meta sOlLisM OF SULFUR AMINO ACIDS IN. MYTILUS EDULIS
AND RANGIA CUNEFATA+

RENN Ee ALEEN 27 -AND J; AWAPARA

Biology Department, Rice Institute, Houston, Texas

Sulfur amino acids are important metabolites in living systems. Their me-
tabolism is well known in mammals—the reactions involved in their breakdown
are known and so are the end products formed. In invertebrates, however, less is
known about the metabolism of sulfur amino acids. It is known that some marine
invertebrates contain in their tissues large amounts of taurine, (Kelley, 1904;
Henze, 1905; Mendel, 1904; Kossel and Edlbacher, 1915; Okuda, 1920; Ackermann
et al., 1924) but not much is known of its origin or its role in the animal. Recently,
several papers have been published in which the probable origin of taurine is
mentioned. In one paper, Shibuya and Shunji (1957) reported that hypotaurine,
a precursor of taurine, was present in Septifer virgatus; Ouchi (1959) claimed
that hypotaurine was present in numerous species of molluscs. Hypotaurine, first
found in the rat by Awapara (1953), is formed from cysteine sulfinic acid and
converted to taurine as demonstrated by Awapara and Wingo (1953). Taurine,
then, could probably be formed in invertebrates by the same sequence of reactions
as in the rat. Cotty et al. (1958) studied the metabolism of sulfur amino acids
in Musca domestica and concluded that methionine serves as a precursor of taurine.

In a recent paper, we reported that taurine is present in all marine invertebrates
studied, but was absent or not detectable in fresh water and terrestrial molluscs
(Simpson, Allen and Awapara, 1959). This comparative study did not tell us
whether sulfur amino acids are metabolized differently by marine molluscs and
fresh-water molluscs. The end products of sulfur metabolism could be different;
we needed to know more about the intermediate steps. We have performed a
number of experiments on Mytilus edulis, which is known to have large amounts
of taurine, and on Rangia cuneata which has none detectable by paper chroma-
tography. The results of these experiments show that both can convert methionine
to cysteine, probably in the same manner as mammals do, and that cysteine can
be oxidized to various products some of which can be decarboxylated to give taurine.

1 This work was supported by grants from the Robert A. Welch Foundation, Houston,
and from the National Institutes of Health, U. S. Public Health Service.
2 The data reported here is part of the Ph.D. thesis of Kenneth Allen.

173

Copyright © 1960, by the Marine Biological Laboratory

gmiiroONt sor ZY 196"

WATITL TION

174 KENNETH ALIEN AN Die Avy Aap AREA

MATERIALS AND METHODS
Animals

M. edulis was obtained from the Marine Biological Laboratory, Woods Hole.
The animals were put in sea water cooled to 5° C. and kept there until they were
used. Fk. cuneata was collected from the San Jacinto River, Harris County, Texas.
They were kept in large aerated aquaria.

Chemicals

S*°-methionine, S*°-cystine and S*-sodium sulfite were obtained from Abbott
Laboratories, Oak Ridge. They were chromatographically pure. S**-taurine was
prepared from S**-sulfite and B-bromoethylamine hydrobromide according to the
method of Cortese (1943). The crude taurine was purified by chromatography
on a column of Dowex 50 in the H* form. The final product was chromatograph-

ically pure. Cystine was reduced to cysteine according to the method of Lucas
and Beveridge (1940).

Administration of compounds

Solutions of S*°-methionine, S*°-cysteine and S*°-taurine were prepared and
administered to both M. edulis and R. cuneata in the following manner: the shell of
M. edulis was opened and kept open with wound spreaders. The foot was located
and 1 ml. of the solution carefully injected into the foot; a 1l-ml. syringe with a
22-gauge needle was used. This method was suggested by Dr. T. W. Potts, of the
University of Birmingham, England. RR. cuneata was injected through a hole cut
carefully in the shell; the needle was inserted through the mantle and gill into the
fleshy part of the viscera. After the injection the animals were returned to aquaria
containing water of the appropriate salinity and temperature.

Extraction and fractionation of amino acids

Extractions were carried out by the method of Awapara (1948). The extracts
were used in some instances for paper chromatography and in others for isolation
of intermediates. Most of the time the extracts were fractioned into various
groups, using 10n exchange resins according to the scheme shown in Table I.
According to this scheme the extracts containing one gram of tissue per cc. were
first put through a column (10 cm. X 1 cm.) of Dowex 50 in the H* form (200-400
mesh). The column was washed with distilled water until the effluent became
neutral. All the anions and non-polar compounds were washed through by this
procedure, along with taurine, cysteic acid and cysteine sulfinic acid. All other
amino acids were eluted from the resin bed with 25 ml. of 2 N NH,OH;; the acid
effluent (A) and the NH,OH eluate (B) were evaporated to dryness; the residues
redissolved in a small amount of water. Fraction A, which contained taurine,
cysteic acid, cysteine sulfinic acid and sulfate, was fractionated using a strong anion
exchange resin—Dowex 2 in the OH~ form. The solution was put through a
small column of the resin (5 cm. X 1 cm.) and washed with 50 ml. of boiled and
cooled distilled water. Elution was carried out in two stages: (1) with 10 ml.
of 2 N acetic acid to remove taurine, and (2) with 10 ml. of 2 N HCl to remove

METABOLISM OF SULFUR AMINO ACIDS 175

cysteic and cysteine sulfinic acids. Fractions A, and A, were evaporated to
dryness ; the residues were dissolved in small amounts of water and the evaporation
repeated to drive off most of the acid. Finally the residues were taken up in
small portions of water and stored for further work.

Identification of products

Intermediates and final products were identified by several methods: (a) The
various fractions were chromatographed on paper; radioautographs were made of
these papers using x-ray film. After a three- to five-week exposure the papers

TABLE |
Fractionation of extracts by means of ion exchange resins
EXTRACT

Water soluble
Substances

Dowex 50 Ht

Effluent |2 N NHzOH

| ata
FRACTION A FRACTION B
Acidic Amino Acids Neutral and Basic
Amino Acids
Taurine
Cysteine sulfinic
Cysteic
Dowex 2 OH

2 N Acetic Acid 2 NALE!

| |
FRACTION A, FRACTION A,

Taurine Cysteine sulfinic
Cysteic

were developed with ninhydrin. The ninhydrin spots were matched against the
blackened areas on the film. Tentative identifications were thus made on the basis
of Ry values. (b) Some of the fractions were chromatographed on paper strips and
the strips analyzed for radioactivity in a continuous strip counter. The position
of the peaks was used as criterion for identification. Also, chromatography on
strips was carried out using the various unknown fractions but mixed with known
substances suspected to be present in the fractions. A perfect overlap between
the peaks and the ninhydrin spots on the strip was used as criterion of identity.
(c) When sufficient evidence was available for the existence of an intermediate,
it was isolated, purified and its radioactivity determined. If the amounts present
were insufficient for isolation, carrier was added, then isolated, purified by re-
crystallization or precipitation and its radioactivity determined. When a compound
was isolated without carrier, other criteria were used such as infrared spectrum.

176 KENNETH ALLEN AND J. AWAPARA

RESULTS
1) Formation of cystathionine

Both M. edulis and R. cuneata were given a solution of S*°-methionine (ap-
proximately 500,000 counts). After 24 hours the organisms were extracted and
the extracts fractionated. Fraction B in each case was chromatographed on paper

Ficure 1. Radioautograph of fraction B from M. edulis after the administration of
S35_methionine. Solvents: phenol (P) and lutidine (L). Methionine (1), methionine sulfone ?
(2), hypotaurine (3) and cystathionine (4).

according to the method used by Simpson, Allen and Awapara (1959), using
phenol-water in one direction and lutidine-water in the second direction. Radio-
autographs of these chromatograms were made, The chromatograms from RK.
cuneata had no cystathionine. We cannot exclude its formation, for it could easily
be a transient intermediate in this animal. It is very likely that it is formed, as we
shall see in discussing formation of cystine. In Figure 1 is shown a radioautogram
from M. edulis. Spot 4 is cystathionine and spot 3 is hypotaurine. Both of these

METABOLISM OF SULFUR AMINO ACIDS liver)

O CA CSA it

Figure 2. Strip chromatogram (lower) and radioactivity distribution (upper) from
the acid fractions of R. cuneata. Solvent: phenol-water. Cysteic acid (CA), cysteine sulfinic
acid (CSA) and taurine (T).

intermediates were recognized by their R¢ values (cystathionine 0.27 in 72%
phenol and 0.20 in 65% lutidine; hypotaurine 0.62 in 72% phenol and 0.42 in
65% lutidine). Spot 1 is methionine and spot 2, probably methionine sulfone.

2) Formation of cystine

Fraction B, from part 1, was used to identify cystine or cysteine. Neither of
these amino acids can be chromatographed satisfactorily. Cystine is very insoluble
and lends itself to isolation. In both instances unlabelled cystine hydrochloride
was added to the fractions. Then hydrogen peroxide was added to oxidize cysteine
to cystine. Cystine was precipitated from the solution by carefully adjusting the
pH to 7. The crystals were filtered, washed, redissolved in dilute HCl and

ee

O CA al

Ficure 3. Strip chromatogram and radioactivity distribution from the acid fraction of M.
edulis. Cysteic acid (CA) and taurine (T).

178 KENNETH ALLEN AND J. AWAPARA

reprecipitated by the same procedure. The washed and dried crystals were used
for determining radioactivity.
M. edulis 68 counts/minute/mg.

R. cuneata 120 counts/minute/mg.

Cystine or cysteine is formed in both organisms. Cysteine oxidizes readily to
cystine but if one assumes that transulfuration takes place in these animals, then
cystathionine and cysteine are the intermediates. Unfortunately cystathionine
was not detected in R. cuneata. If it is not formed we must postulate another
mechanism for the formation of cysteine and no other mechanism is known at this

time.

a S 4 >

Ficure 4. Radioautographs of the acid fraction of R. cuneata after the administration

of methionine. (1) After 1 hour, (2) after 5 hours, (3) after 10 hours, (4) after 15 hours
and (5) after 24 hours. (A) cysteine sulfinic acid, (B) cysteic acid, (C) taurine, (D)

unknown.

METABOLISM OF SULFUR AMINO ACIDS 179

COUNTS/MIN x 10°
Ww iS

MN

DAY

Ficure 5. Disposition of administered S*°-taurine by FR. cuneata.

3) Oxidation products of cysteine

Pooled fractions A, and A, from both organisms were mixed with cysteic acid,
cysteine sulfinic acid and taurine, then chromatographed using 72% phenol. The
strips were analyzed with a continuous strip counter, then developed with ninhydrin.
In Figures 2 and 3 are shown the results obtained for R. cuneata and M. edulis.
The latter had only one radioactive spot: taurine. Cysteic acid was formed in
R. cuneata in relatively high amounts. A second experiment was performed on
R. cuneata to find out the time needed for the formation of these intermediates.
S*°-methionine was injected into several organisms; extracts were made at the
end of 1, 5, 10, 15 and 24 hours. The extracts were fractionated and fraction A
of each one chromatographed on strips. Radioautograms were made and they
are shown in Figure 4. Taurine was formed only after cysteine sulfinic acid and

180 KENNETH ALLEN AND J. AWAPARA

cysteic acid were formed; it took 5 hours to form sufficient taurine to be detected.
At the end of 10 hours two S*°-labelled compounds appeared. Compound D has
an R; value identical to that of taurocyamine (.64 in phenol) but gave a negative
Sakaguchi test on paper chromatograms. At the end of 24 hours most of the
end products have disappeared from the animal. Taurine is formed in Kk. cuneata
but not held. This is better shown in Figure 5. In this figure is shown the
rate at which injected S*°-taurine is excreted by R. cuneata. One could argue that
the conditions were abnormal and that the animal ejects an exogenous substance ;
this is not the case as seen in Figure 5; taurine is endogenously formed and it is
also rapidly excreted.

4) Formation of sulfate and taurine

Both sulfate and taurine are formed by M. edulis and R. cuneata. Sulfate was
detected by precipitation as BaSO, with added carrier. In both instances the
BaSO, was radioactive (35 cts./min. and 85 cts./min.). Although the counts
were low, they were significant. The BaSO, was ignited to insure destruction of
any organic sulfur compounds.

Taurine was detected in R. cuneata as already shown; M. edulis is known to
contain very large amounts of taurine (as much as 4 per cent of wet weight) and
to demonstrate its formation from administered precursors would be difficult.
Nevertheless we succeeded in isolating S*°-labelled taurine from M. edulis after
the administration of S**-methionine. The isolated taurine was recrystallized
several times and its radioactivity determined. The specific activity was 27
counts/min./mg. With such low activity the evidence for taurine formation was
somewhat weak. More supporting evidence was obtained by isotope dilution.
To the isolated taurine we added an equal amount of pure taurine; the mixture

FREQUENCY (CM’)
00 3000 2500 2000 1800 1600 00 65)
Go Gs 0 0 een ae en en es a eee ee | a I
00 Fo
(een seafarers [fa ff
Soeeseemesgasaeo ss ="

ERR SSSSiVeEBvEIe

i i ff (ne net fifa safe ay

oo fe
SeSeartae

SaHinanlit

a] [etm

\ / ip |
Go Cee ee ae
[ eo | RERBURP. SEaea

he 8 9
WAVELENGTH (MICRONS)

Ficure 6. Infrared absorption spectra of (A) known taurine and (B) S*5-taurine isolated
from M. edulis after the administration of S?5-methionine.

METABOLISM OF SULFUR AMINO ACIDS 181

was recrystallized and the radioactivity of the pure crystals determined:

Isolated taurine 6 mg. 26 cts. /min./mg.
Mixture 12 mg. 12 cts. /min./mg.

The infrared absorption spectra * of both pure taurine and the isolated taurine were
determined and shown in Figure 6. It appears that /. edulis forms taurine slowly
but retains it. The high concentration of endogenous taurine in this animal makes
it very difficult to establish the rate at which taurine is formed.

DISCUSSION

The metabolism of sulfur amino acids in the two molluscs studied is not dif-
ferent, at least qualitatively, from the metabolism in mammals. The oxidation of
cysteine in both, mammals and molluscs (those studied), gives rise to cysteine
sulfinic acid and eventually to taurine and sulfate (Awapara, 1956). The most
significant difference observed in the two animals studied is the rate at which
taurine is disposed of. M. edulis keeps it by some unknown mechanism defying
a concentration gradient. F. cuneata can produce taurine but cannot hold it. If
we extrapolate these results, the absence of taurine in all the fresh-water animals
(Simpson, Allen and Awapara, 1959) could be explained on the basis of rapid
excretion, not lack of formation. The method of formation of taurine and sulfate
in both organisms again appears similar to the method of formation in mammals.
Cysteine sulfinic acid is either decarboxylated to hypotaurine or forms sulfate after
transamination, splitting of SO,, and oxidation. The absence of hypotaurine in
R. cuneata and the presence of large amounts of cysteic acid is indicative of a
pathway in which cysteine sulfinic acid is oxidized first to cysteic acid and this
decarboxylated to taurine. In M. edulis hypotaurine was present but cysteic acid
was not detected. If there are any differences in the metabolism of sulfur amino
acids in the molluscs studied, the differences are in the intermediates and in the
disposition of the end products.

The role of taurine in marine molluscs remains unknown. The only suggestion
made is that it serves as an osmoregulator.

SUMMARY

1. The metabolism of sulfur-bearing amino acids was investigated in FR. cuneata
and M. edulis, in vivo.

2. Injected S**-methionine gives rise to cysteine/cystine in both animals.
Cystathionine was detected in M. edulis, but not in R. cuneata though it is probably
formed. A demethylation and transulfuration mechanism is postulated for this
conversion.

3. Both R. cuneata and M. edulis form sulfate from injected methionine. Also,
both form taurine but FR. cuneata does not accumulate it whereas M. edulis ac-
cumulates it in large amounts.

4. Taurine is probably formed by different reactions in these species. In R.
cuneata it is probably formed by decarboxylation of cysteic acid, whereas in M.
edulis it is formed by oxidation of hypotaurine.

3We are indebted to Dr. T. Patton from the M. D. Anderson Hospital for this
determination.

182 KENNETH ALLEN AND J; AWAPARA

LITERATURES Clie

ACKERMANN, D., F. Hottz ann F. KutscuHer, 1924. Extractives of Eledone maschata. Chem.
Abst., 29: 3694.

Awapara, J., 1948. Application of paper chromatography to the estimation of free amino
acids in tissues. Arch. Biochem., 19: 172-173.

AwapaAra, J., 1953. 2-Aminoethanesulfinic acid: An intermediate in the oxidation of cysteine
in vivo. J. Biol. Chem., 203: 183-188.

Awapara, J., 1956. Oxidation de la Cystéine en Sulfate et Taurine. Colloque sur la Biochimie
du Soufre Centre National de la Recherche Scientifique. Pp. 99-105.

Awapara, J., AND W. J. Winco, 1953. On the mechanism of taurine formation from cysteine
in the rat. J. Biol. Chem., 203: 189-194.

CortTEsE, F., 1943. Organic Syntheses, Coll. 2 :564.

Cotry, V. F., S. M. Henry anp J. D. Hitcuey, 1958. The sulfur metabolism of insects.
III. The metabolism of cystine, methionine, taurine and sulfate by the house fly,
Musca domestica L. Contr. Boyce Thompson Inst., 19: 379-392. :

Henze, M., 1905. Beitrage zur Muskelchemie der Octopoden. Hoppe-Seyl. Zeitschr., 43:
477-493.

Kettey, A., 1904. Beobachtungen uber das Vorkomenn von Atherschwefelsauren, von Taurin
und Glycin bei neideren Tieren. Beitr. Chem. Physiol. Path., 5: 377-383.

Kosse_, A., AND S. EpiBAcHER, 1915. Beitrage zur Chemischen Kenntnis der Echinodermen.
Hoppe-Seyl. Zeitschr., 94: 264-283.

Lucas, C. C., anp J. M. R. Breveripce, 1940. The analysis of hair keratin. I. A method for
the quantitative removal of cystine from keratin hydrolysates. Biochem. J., 34: 1356-
1366.

Menoet, L. B., 1904. Uber das Vorkommen von Taurin in den Muskeln von Weichtieren.
Beitr. Zettschr. Chem. Physiol. Path., 5: 582.

Oxupa, Y., 1920. Extractive matters of Palimurus japonicus and Loligo breekeri. Chem.
Abst., 14: 2827.

Oucui, S., 1959. 2-Aminoethanesulfinic acid. J. Isolation from a mollusc; Identification
and distribution. J. Biochem. (Japan), 46: 765-770.

Suipuya, S., AND S. Oucut, 1957. Isolation of 2-aminoethanesulfinic acid from a mollusc.
Nature, 180: 549-550.

Smpson, J. W., K. W. ALLEN anp J. AwaApara, 1959. Free amino acids in some aquatic
invertebrates. Biol. Bull., 117: 371-381.

foes Ch OF SALINITY AND TEMPERATURE ON LARVAL
DEVELOPMENT OF SESARMA CINEREUM. (BOSC) REARED
PE eA BORATORY *

JOHN D. COSTLOW, JR., C. G. BOOKHOUT AND R. MONROE

Duke University Marine Laboratory, Beaufort, N. C., Department of Zoology, Duke University,
Durham, N. C., and Department of Experimental Statistics, North Carolina State College,
Raleigh, N. C.

Although environmental factors have been known to affect the rate of develop-
ment, the number of stages, and the survival of marine invertebrate larvae, rela-
tively few studies have been made to determine the effect of a wide range of
salinities, temperatures, and food on larval development under laboratory condi-
tions. Studies on the relationship between salinity and fertilization (Amemiya,
1926), salinity and larval development (Loosanoff, 1948), and salinity and survival
and growth (Davis, 1958) have been made on some mollusks. A limited number
of similar investigations have also been reported on the effects of temperature
(Clark, 1935; Loosanoff and Davis, 1950; and Loosanoff, Miller and Smith, 1951).
While no identical studies have been carried out on other major phyla some
comparable data on echinoderm development in relation to salinity and temperature
may be found in the publications of Mortenson (1931, 1937 and 1938). In general,
the effects of salinity, temperature, and food on the larval development of the
Crustacea are not known, presumably because of the difficulties which are usually
encountered in rearing larvae in the laboratory.

In spite of the persistent interest in decapod larvae and the numerous studies on
life-histories reconstructed from planktonic material, only 7 accounts are known
to us which deal with the complete larval development of Brachyura in the
laboratory (Schlegel, 1911; Lebour, 1928; Hart, 1935; Sandoz and Hopkins, 1947;
Chamberlain, 1957; Knudsen, 1958; and Costlow, Rees and Bookhout, 1959).
None of these authors consider the effects of salinity or temperature on larval
development. Broekhuysen (1937) studied the effects of temperature and salinity
on egg development of Carcinides maenas but did not rear the larval stages. Sandoz
and Rogers (1944) reported the effects of environmental factors on some of the
early larval stages of Callinectes sapidus, reared in the laboratory, and later they
conducted similar experiments on megalops obtained from planktonic material
(Sandoz and Rogers, 1948). More recently, the effects of salinity and temperature
on the complete larval development of Callinectes sapidus Rathbun have been
described (Costlow and Bookhout, 1959) but mortality in this species is high within
a wide range of environmental conditions and cannot be attributed solely to salinity
or temperature. In the Macrura, Templeman (1936a) described the effects of
temperature, light, salinity, and diet on the larvae of Homarus americanus, and
Broad (1957) studied the relationship between food and the number of larval

1 These studies were supported by a grant (G 4400) from the National Science Foundation

183

184 J: D. COSELOW; JR: C/G BOOKHOUD ANDER VvMONR OE

stages of Palaemonetes vulgarus and P. pugio. In the Anomura, Coffin (1958)
considered the effects of salinity and temperature on the larval development of
Pagurus samuelis.

During the past three years we have followed the larval development of a
variety of Brachyura from the Beaufort area, from hatching to the first crab stage
and beyond. ‘The purpose of the present work has been to determine the effect of
various Salinity-temperature combinations on the number of larval stages, the
molting frequency, the duration of larval life, and the survival of the larval stages
of Sesarma cinereum (Bosc).

Inasmuch as only the first zoeal stage of Sesarma cinereum has been figured
(Hyman, 1924) a description of the four zoeal stages and one megalops stage will
be reported in a separate account.

The authors wish to thank Mrs. W. A. Chipman and Mrs. C. King for their
technical assistance throughout the study. Dr. Fenner Chace, U. S. National
Museum, was most helpful in his identification of the adult females, and specimens
from which the larvae hatched have been deposited with the Museum.

METHODS

Ovigerous Sesarma cinereum (Bosc) were obtained at the drift line in the
vicinity of the Duke University Marine Laboratory, Beaufort, N. C., and brought
into the laboratory. The females were maintained separately at 25° and 30° C.
in fingerbowls containing filtered sea water of the salinity in which the larvae were
to be reared. The water, containing 200,000 units of penicillin G per liter, was
changed daily. When hatching was observed, the female was removed to a
separate bowl and the zoeae transferred to mass cultures of approximately 50—/0
per bowl. Fertilized Arbacia eggs and recently hatched Artemia naupliu were
added immediately. Zoeae were then placed in either Syracuse watch glasses (10
per watch glass) or compartmented plastic boxes (6 per compartment) containing
approximately 30 ml. of water of the same salinity, temperature, and type of food
as in the mass cultures. The containers were marked, covered, and placed in
temperature cabinets regulated to provide a photoperiod of 12 hours of light and
12 hours of darkness. Table I gives the salinity and temperature combinations

TABLE [

Initial number of zoeae, number of megalops, and number of crabs reared under the three
temperatures and four salinities indicated

2OONC? DSO MC. SOY (Cy,

p.p.t

Initial Number Number Initial Number | Number Initial Number | Number

number | megalops crabs number | megalops crabs number | megalops crabs
2.5 108 0 0 102 13 0 108 18 0
20:1 96 34 1 105 53 4 113 52 10
26.5 104 52 4 107 48 26 107 20 5
ol 108 21 0 100 2 0 108 9 0

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 185

PERCENT MOENS

\
\
\
\
\
\
\
\
\
\
\
\
\
\
\

\

4

OY AIEEE ELE ERE aE De EEE AEE BE AE GEE DEAE GEIR EDs

N
\
\
a \e =

NINE
689
DAYS

Ficure 1. Comparison of the time of the first molt for zoeae of Sesarma cinereum
reared in the laboratory at different salinities and temperatures. Black—20° C., white—25° C.,,
diagonal—30° C.

used. Larvae reared at 20° C. were hatched at 25° C. and the temperature was
gradually reduced.

The containers and larvae were examined daily under a dissecting microscope,
the zoeae changed to freshly filtered sea water, and Artemia nauplii added. At
this time the number of exuviae in each series was recorded and the mortality
noted. When the megalops stage was reached the larvae were removed to separate
bowls and fed chopped liver in addition to the Artemia nauplii. The first crab, and
all subsequent stages, were maintained on a diet similar to that of the megalops.

RESULTS

The larvae of S. cinereum hatched in all salinity-temperature combinations
shown in Table I. Approximately 100 per cent of the eggs hatched as first zoea,
left the egg mass almost as one swarm, and swam to the light side of the bowl.

The time after hatching at which the first zoea started to molt was similar at
corresponding temperatures, regardless of the variations in salinity (Fig. 1).

186 J.D: -COSTLOW; JR, CSG BOOKHOUT AND: RoOMONEROE

I =
APO-NMYT OF DAMS

ZOOL LLL

S CIELO

6b

\
\

PERCENT MOLTS

OLLIE LLL LLL LL
LL LLL LLL

(fs ae
LLLLSSALAs

DAYS

FicurE 2. Comparison of the time of the second molt for zoeae of Sesarma cinereum reared
at different salinities and temperatures. Black—20° C., white—25° C., diagonal—30° C.

At 26.7 and 31.1 p.p.t. the majority of the first zoeae molt within a more limited
period of time than at 12.5 and 20.1 p.p.t., at the same temperature. For example,
at 30° C., 31.1 p.p.t. molting was limited to days 3 and 4, with 88 per cent molting
on day 3, while at 30° C., 12.5 p.p.t., the period of ecdysis was spread over four
days with no pronounced peak on any one day (Fig. 1). This becomes more
evident with later molts.

As shown in Figure 2, 93 per cent of the zoeae molted for the second time on
days 5 and 6, at 30° C., 31.1-p.p4.; at/30°'C., 12.5 ppt, the zoede didi motsinesms
the second molt until the seventh day and then required five days before all had

PE RGEN ies MOLES

\
\
\
.
\

\
\
\
A

DAYS

Figure 3. Comparison of the time of the third molt for zoeae of Sesarma cinereum reared at
different salinities and temperatures. Black—20° C., white—25° C., diagonal—30° C.

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 187

PERCENT MOLTS

DAYS

Ficure 4. Comparison of the time of the fourth molt for zoeae of Sesarma cinereum reared at
different salinities and temperatures. Black—20° C., white—25° C., diagonal—30° C.

completed ecdysis. The persistent lag and the tendency for each molt to require
a greater period of time in the lower salinities are continued in the third and
fourth zoeal molts (Figs. 3 and 4). The fourth zoeal molt produces the megalops
which completes metamorphosis by molting into the first crab stage. The majority
of megalops in 26.7 p.p.t. molted to the first crab stage before those in corresponding
temperatures at 20.1 p.p.t. (Fig. 5). Megalops at other salinities, 12.5 and 31.1
p-p.t., did not survive to the first crab stage.

The duration of the four zoeal stages, the megalops stage, and the total time
required for development to the first crab stage are affected by salinity as shown
in Figure 6. There is a gradual reduction in the times required from low to high
salinity. The shortest time for zoeal development was 10 to 12 days, at 30° C.,,
31.1 p.p.t., while at 30° C., 12.5 p.p.t. the same development required 13 to 18
days. Similar differences are shown for the other temperatures used (Fig. 6).
Variations in the time of megalops development are also associated with salinity.
At 30° C., 26.7 p.p.t. the duration of the megalops stage was 7 to 8 days while at
30° C., 20.1 p.p.t. the megalops persisted for 10 to 16 days (Fig. 6). Since the

—
Zw i's
LJ
O at 4G
re O
fi: 5
‘ia
Oo
DAYS
Figure 5. Comparison of the time of the molt from the megalops to the first crab
stage of Sesarma cinereum reared at different salinities and temperatures. Black—20° C.,

white—25° C., diagonal—30° C.

188 J. D. COSTEOW, JR Cres BOOKHOUrr AND ik MONROE

zoeal and megalops stages are prolonged in lower salinities, the time required for
complete development to the crab is delayed (Fig. 6).

As might be expected, the first molt occurred latest at 20° C. and earliest at
30° C. (Fig. 1). The time for completion of the second molt is spreadsovermea
greater number of days at 20° C. and by the time of the third molt (Fig. 3), the

40
9
8
U
6
SD
4
3
2
|
30
9
8
7
6
25
4
3
= 2
We |
i eo
=
(eu |
O I
J I
J |
> |
uJ |
OQ
I
Le
O
op)
SS
<
(a)

ny)
O-—-NWAUBDAIWOO—-NMWHhUMUADINDWO NWOO— NWSWO Y

le ee eel bY)

TEMPERATURE. 7c:

FicurE 6. Comparison of minimum and maximum time of zoeal development (white),
megalops development (stippled), and total development (black) to the first crab stage for
larvae of Sesarma cinereum reared at different salinities and temperatures.

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 189

! =e ey ao! PEs eee

alee
el
L \

ED),

i TN \ \
Bie = \ ‘ \ | sm
ee NNN i
stl \ \ \ c Sill
: V NAA
: V NAA
i \ NA [::
: |

TEMPERATURE War

Ficure 7. Per cent mortality of each stage of Sesarma cinereum reared at different tempera-
tures and salinities. S I-S IV =zoeal stages I-IV; M = megalops.

differences due to temperature appear to be even more distinct. At 30° C., the
majority of second stage zoeae have completed the third molt by the time the zoeae
at 25° C. have begun this molt. Similarly, the zoeae at 25° C. have completed the
third molt before those at 20° C. have begun. At molt IV (Fig. 4) there is
an even greater difference in zoeae reared at different temperatures and at 20° C.
the spread of time for the molt is considerable. This difference is carried over
into the molt to the first crab (Fig. 5). As shown in Figure 6, the duration of

190 J. D. COSTLOW,: JR., C:. GF BOOKHOUIVAND Ri MONROE

the total larval life (2.e., first zoea to first crab stage) varies considerably. The
shortest time (20-22 days) required for complete development to the crab stage
is found at 30° C. and the longest time (35-39 days) was for larvae maintained at
20° C. More larvae completed development at ‘25° Ci 26.7 pip.tthantar 2 ee
20.1 p.p.t. but the former required a greater range of time (24-37 days) than the
latter (30-34 days). A Q,, of approximately 1.8 was observed for the time
required for complete development to the first crab stage at 20° C. and 30° C.

Survival of zoeal and megalops stages varied in relation to salinity and tempera-
ture. Figure 7 indicates that mortality of the first zoeal stage was highest at
all temperatures within a salinity of 12.5 p.p.t. and at 20° C., 20.1 p.p.t. Mortality
of the first stage was low (4-11 per cent) at all temperatures in a salinity of 31.1
p.p.t. The second stage zoea showed a high mortality at 12.5 p.p.t. and 20.1 p.p.t.
at 25° C. In all other temperature-salinity combinations mortality of the second
stage was under 10 per cent. Mortality of the third stage zoea varied in all
salinities and temperatures: from) 1 per cent at/307 C207 p.p.t tosZiepemecr:
at 30° C., 20.1 p.p.t. Mortality of stage four zoea considerably depleted the num-
ber of larvae at all temperatures at 31.1 p.p.t. (Fig. 7). The majority of the fourth
stage zoeae died at this salinity in an attempt to molt to the megalops stage. The
number of fourth stage zoeae which successfully molted to the megalops stage was
largest at all temperatures in 20.1 p.p.t. and 26.7 p.p.t. This varied from 35 to 50
per cent of the original number of zoeae. Those series with the largest number of
megalops were also those which showed some survival to the first crab stage and
beyond. In these same series, however, much of the total mortality was due to
death in the megalops stage. For example, while the lowest per cent mortality
(49.8 per cent) ofall zoeal stages occurred at 20° 1©.,' 207 pip. tie sarees
mortality of the megalops stage was observed at this same temperature and salinity.
The highest per cent survival through the first crab stage was 24.4 per cent, at
29916. 20me pap.t.

DISCUSSION

The few successful rearing experiments with Brachyura larvae usually have had
two major features in common: water which was limited in its temperature range
and salinities which fluctuated only slightly. Experimental studies on the effects
of salinity and temperature variation on decapod larvae are usually confined to a
portion of the larval development and do not include all the stages prior to
metamorphosis to the first post-larval stage. Sandoz and Rogers (1944) studied
the first three zoeal stages of Callinectes sapidus but actually give results for only
the first stage. A later study (Sandoz and Rogers, 1948) centered around the
effects of temperature and salinity on the megalops of C. sapidus obtained from
planktonic material. Coffin (1958) followed the development of Pagurus samuelis
in the laboratory at two different temperatures and a variety of chlorinities. In
the Macrura, several authors have made experimental studies on larval develop-
ment. Sollaud (1919), rearing the larvae of Palaemonetes varians microgenitor
at two different temperatures and salinities, found that at 15° C., 6.5 p.p.t. the
larvae completed development with 6 to 8 molts in 23-33 days while at 18° C.,
6.5 p.p.t. only 5 or 6 molts were required and development was complete after
15-23 days. At 15° C., “sea water,” larval development occurred in from 20 to

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 191

29 days with 5, 6, or 7 molts (Sollaud, 1919). Gompel and Legendre (1927)
found that the first and second stages of Homarus vulgarus could withstand a
salinity of 44 (1.033) but died at 23.3 (1.017). Their work does not include
attempts to rear the larvae through successive molts and no reference is made to
feeding. Templeman (1936a) studied the effects of a variety of environmental
factors on larval development of H. americanus, and Broad (1957) followed the
relationship between diet and number of larval stages of Palaemonetes pugio and
Palaemonetes vulgaris.

In the present study three temperatures were combined with four salinities to
provide twelve distinct environments. Food was always abundant in all environ-
ments, and the photoperiod was constant.

Hatching

Within the temperatures and salinities used there was little difference in the
hatching process. Hatching was 100 per cent in all conditions and the so-called
“pre-zoea” was never observed. Immediately after the zoeae left the egg mass
some were removed, examined under a microscope, and were found to possess a
dorsal spine and exposed maxillary hairs. Many authors have reported the “pre-
zoea’ from a variety of crabs hatched in the laboratory: Birge (1883), Panopeus
sayi; Hyman (1925), Eurypanopeus depressus, Hexapanopeus angustifrons,
Menippe mercenaria, Neopanope texana sayi, and Panopeus herbstw; Lebour
(1928), all British species she described; Churchill (1942), Callinectes sapidus ;
and Chamberlain (1957), Neopanopeus texana sayi. As early as 1903 Williamson
noted in experiments with Carcinus maenas that normal hatching produced only
first zoeae; if the egg mass were washed immediately prior to hatching, the “pre-
zoea’ could be obtained. Sandoz and Rogers (1944) observed that the number
of “pre-zoea”’ of C. sapidus at hatching was associated with salinity. At 25 to
30 p.p.t. few “pre-zoeae” were obtained while at 10.9 p.p.t., 90 to 100 per cent of
the larvae hatched as “pre-zoeae’’ and apparently never molted to the first zoeal
stage.

Sandoz and Rogers (1944) also found that the few zoeae which did hatch at
14 p.p.t. were sluggish and did not show the characteristic positive phototactic re-
sponse. At 18 p.p.t. the zoeae were active and positively phototropic and at 24
p.p.t. the larvae were quite active (Sandoz and Rogers, 1944). Broekhuysen
(1937) found that development of the eggs of Carcinides maenas was normal
and that hatching occurred at 28.3-40 p.p.t., 10° C., and at 18-40 p.p.t., 15.6° C.
He suggests that the range of salinity in which the crab can develop is shifted
toward the more saline waters and further comments that salinity and temperature
together determine the limits for development. In the present study the first zoeae
of S. cinereum were active in most salinities and always positively phototactic.
At 12.5 p.p.t. the zoeae were not as active as in the higher salinities but did seek
the lighted side of the bowl.

To assure acclimation of the eggs to both temperature and salinity, the ovigerous
females of S. cinereum were placed in the salinities to be used several days prior
to hatching. Templeman (1936a) used larvae obtained from one egg mass, which
were hatched at one salinity and temperature, and transferred them to the experi-
mental salinities approximately two hours after hatching. He does not indicate

192 J. D, COSTLOW, JR.,°C. GPBOCOKHOLT ANDaRs tIONROn

whether the change was gradual or abrupt. Sandoz and Rogers (1944) ap-
parently followed a similar approach with some of the first zoeae of C. sapidus
but allowed 24 hours before transferring them to the different salinities. A
salinity range of 23 to 28 p.p.t. was found to be optimum for the hatching of
zoeae of C. sapidus (Sandoz and Rogers, 1944). Coffin (1958) hatched the eggs
of Pagurus samuelis at one salinity and temperature and immediately transferred
the zoeae to the experimental conditions (personal communication). In these three
studies, however, the best results were obtained with zoeae in salinities similar to
the water in which the eggs were hatched. Thus, one questions whether the
salinities described as “optimum” might not be due to the fact that larvae reared
in water of the salinity at which they were hatched were not required to acclimate
to abrupt salinity changes. In the present study separate crabs were used for
each salinity to avoid this disadvantage. This procedure may, however, introduce
a variable which is difficult to recognize or control: differences in viability of
larvae obtained from different females. It is thought that this factor may account
in part for the slight differences in molting frequency and duration of total larval
life in larvae reared in 20.1 p.p.t.

Larval development

In the present study there is definite evidence that the duration of the individual
stages and the over-all time required for complete development were affected by
the salinities used. In the lower salinities, 12.5 and 20.1 p.p.t., there was a
tendency for molting of the individual stages to begin later than at the higher
salinities. While this may be seen for each stage in Figures 1-5, it is best shown
in the total time required for zoeal development (Fig. 6). At comparable tem-
peratures, there is a gradual reduction in the time of development from low salinity
to high salinity. Templeman (1936a) found that the larvae of H. americanus
required approximately the same number of days to develop to stage IV in salinities
of 21.8 to 31.8 p.p.t. He suggests that molting may occur earlier at 21 p.p.t. and
25 p.p.t. but the small number of larvae (12) used at each salinity do not give
conclusive evidence. Templeman (1936a) also found that as the salinity was
reduced, the number of first stage larvae which molted to the second stage also
decreased: at 11.6 p.p.t. none molted to the second stage and in higher salinities,
42.5 p.p.t., the results were similar. The results of Sandoz and Rogers (1944)
with zoeae of C. sapidus indicate that the duration of the first stage is similar in
both 20 and 25 p.p.t., at comparable temperatures. In salinities below 18 p.p.t.
and above 29 p.p.t. the activity of the larvae decreased. A similar analysis of the
duration of stages II and III is not given. In a later study Sandoz and Rogers
(1948) followed the time required for megalops of C. sapidus to molt to the first
crab stage in different salinities and in two rather broad temperature ranges. As
the salinity decreased, from 31 to 10 p.p.t., a greater length of time was required
for completion of the final larval molt (Sandoz and Rogers, 1948). At 11-17° C.,
the increase in time, with the same decrease in salinity, is even more obvious but
as existing light conditions varied, the two experiments should not be compared
directly. Inasmuch as the megalops were obtained from the plankton, and the
age of the individuals was not definitely known, it is difficult to evaluate the
significance of the 17 hours difference in molting. Coffin (1958) shows the dura-

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 193:

tion of each stage of Pagurus samuelis at several chlorinities. With the exception
of the first zoeae, the duration of the stages was less at a chlorinity of 16 (sal. 28.9
p-p.t.) than at 19.3 (sal. 34.9 p.p.t.). This comparison includes the first three zoeal
stages only (Coffin, 1958). Sollaud (1919) notes that a decrease in salinity,
presumably from approximately 35 p.p.t. (“sea water”) to 6.5 p.p.t., increases the
time required for complete larval development of Palaemonetes varians micro-
genitor. In the present study, a decrease in salinity prolonged the length of zoeal
development of S. cinereum. Although the time required for development of
S. cinereum megalops to the first crab stage follows this trend, it is not as definite
as the results for the zoeae and there is more overlapping in salinities of 20.1 and
26.7 p.p.t. (Fig. 6).

Survival of S. cinereum zoeae was definitely affected by salinity. As shown in
Figure 7, the first stage zoea withstood the higher salinities (26.7-31.1 p.p.t.)
better than the lower salinities (12.5 p.p.t—20.1 p.p.t.). Mortality in the second
and third stages was relatively low in all salinities. While exact comparisons in
stage cannot be made, it should be noted that mortality in Homarus americanus
(Templeman, 1936a) was highest in the third and fourth stages in the lower
salinities. Mortality of first zoeae of Pagurus samuelis was lowest at a chlorinity of
19.3 (sal. 34.9 p.p.t.), the chlorinity at which the eggs had been hatched (Coffin,
1958). In the remaining larval stages, the per cent mortality was similar in both
16 and 19.3 (sal. 28.9 p.p.t. and 34.9 p.pt.). Mortality of the fourth stage of
S. cinereum varied from 50-83 per cent of the total mortality in 31.1 ppt. In
the vast majority of cases the fourth stage zoea died in the process of molting into
the megalops. Those larvae which did manage to successfully complete the fourth
molt usually died within a few days. ,

The effect of temperature on the duration of the individual larval stage of S. cine-
reum is shown in Figures 1 through 5. With the exception of the zoeae reared at
20.1 p.p.t., duration of the individual stages at 30° C. was approximately half the time
required for comparable development at 20° C. At 25° C. there was considerable
overlapping in the time of development. In general there is a greater range of
time for all molts at 20° C. than at 30° C. although an insignificant number of late
molts at the latter temperature occasionally spread the over-all time. Templeman
(1936a) did not consider the effect of temperature on duration of the individual
stages, except indirectly and with light as an additional variable. He does give,
however, the time for development to the fourth stage of Homarus americanus
reared at different temperatures. He found that at 24° C., 10.5 days were required
while at 14° C., 26.5 days were required for comparable development. Sandoz and
Rogers (1944), while not giving similar figures for C. sapidus, do note that below
20° C. none of the first zoeae completed the molt into the second stage. Sandoz
and Rogers (1944) also state that when other factors were favorable, active feeding
occurred at temperatures of 19° to 23° C. The effect of temperature on the time
required for completion of the four zoeal stages and the megalops stage of S.
cinereum is shown in Figure 6. There is some overlapping at 25° C. but the
minimum and maximum at 20° C. and 30° C. are quite discrete.

Broad (1957) found that the number of stages of Palaemonetes pugio and
Palaemonetes vulgarus varied according to the diet, and Templeman (1936a,
1936b) found that a reduction in food produced a longer intermolt period and

194 J. D. COSTLOW,. JR., ©. G@ BOCKHOUT ANDi R MONBOE

sometimes produced an “extra” larval stage. In the present study a super-abundance
of food was present in all 12 environments and variations in number of larval stages
were never observed. The four zoeal stages and one megalops were found to be
consistent throughout the study of 1266 larvae. Thus, while variations in diet
may account for additional stages in the laboratory, variations in temperature and
salinity do not aiter the number of stages of S. cinereum.

Ecological significance

S. cinereum, ranging from the Chesapeake Bay to Mexico (Rathbun, 1917),
is found at Beaufort, N. C., in areas of varying salinity. Pearse (1936), during
the summer of 1928, found adult crabs adjacent to waters varying in salinity from
0.5 to 18.9 p.p.t. During the summer of 1958 and 1959 the water in the collecting
areas was 26.5 to 35 p.p.t. Although the limits of the breeding season are not
known at Beaufort, N. C., ovigerous females of S. cimereum are abundant during
the summer months and Sutcliffe (1950) found the greatest numbers of decapod
larvae in the plankton at Beaufort throughout these months. In the laboratory,
hatching was successful in salinities of 12.5 to 31.1 p.p.t. and it may be assumed
that hatching of the larvae takes place naturally in waters where there is a wide
salinity range. The time required for complete larval development of S. cinereum
in the natural environment is not known. Because of differences in species, tem-
perature, and food it is impossible to directly compare our results with the findings
of other rearing studies. However, it may be noted that Sandoz and Hopkins
(1947) found that the four zoeal stages and one megalops stage of Pinnotheres
ostreum required 25 days, at 23° C., for complete development. Chamberlain
(1957), rearing Neopanope texana sayi at 24° C., reported that development of
the same number of stages required 20-25 days. Hart (1935), while commenting
that the length of larval life is dependent on a number of environmental factors,
gives four weeks for the two zoeal stages and one megalops of Pimnotheres taylort,
but does not mention the temperature or salinity used in the laboratory rearing.
At 25° C. Sesarma cinereum required 24-37 days and at 30° C., 20 to 22 days
(Fis 6)k: :

While it is beyond the scope of this study to discuss the theories of plankton
retention or the effects of tidal currents on zooplankton, some of our results may
be compared with the findings of some recent field studies. Bousfield’s (1955)
excellent study on barnacle larvae in the Miramichi Estuary also includes some
references to decapod larvae. He did not find the larvae of R/ithropanopeus
harrissi in salinities higher than 23.5 p.p.t. but did find them in salinities as low
as 4 p.p.t. Connolly (1925) also notes that larvae of this same species are found
in brackish to fresh water but does not give salinities. Bousfield (1955) states
that salinity is probably the chief factor affecting survival of barnacle larvae in
the upriver limits of the estuary. Although not identified, either by species or
stage, Sutcliffe (1950) found the largest number of decapod larvae in waters of
high salinity (31.0-35.4 p.pt.) and in temperatures of 24.9-30.0° C. He did
not, however, investigate a variety of depths. It has been suggested from field
studies that the survival of larvae of estuarine forms is associated with salinity and
temperature and, as indicated by Bousfield (1955), also associated with vertical
distribution and horizontal water movements. In the laboratory (Fig. 7) 100 per

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 195

cent mortality occurred at low salinities (12.5 p.p.t.) and at high salinities
(31.1 p.p.t.), whereas in the intermediate salinities (20.1 and 26.7 p.p.t.) some
zoeae passed through all larval stages and metamorphosed to the first crab stage
at all three temperatures. The constant conditions of temperature and salinity,
used in the laboratory studies, are obviously not found in the natural estuarine
environments. If zoeae were hatched and retained in water of low or high salini-
ties in nature, they might not survive. If, however, the vertical and horizontal
water movements carried them to intermediate salinities, their chances of survival
would be increased. An examination of Figure 7 suggests that optimum con-
ditions of temperature and salinity exist for each larval stage of S. cinereum.

Estimation of optimum conditions

The salinity-temperature ranges for each larval stage of S. cinereum have been
determined from laboratory-reared material. If these same environmental limits

adh &

TEMPERATURE

5 ne) i) 20 fess 30 35 40

SALINITY p.p-t

Ficure 8. Estimation of per cent mortality of first zoeal stage of Scsarma cinereum
based on the fitted response surface to observed mortality under twelve different combinations
of salinity and temperature.

196 J. DV€OSPLOW, JR. CG BOOKHOUT AND Ry MONROE

i

LE
TAO) INS) fa) INS) (Gu) (G5)
(py SS] (00) Ko) (Sy) == IN)

mM Mm
WwW fw
BSS SARS Sate PASE RS Ses Reese
nN
O

TEMPERATURE

FAO) = UNO) NY)
©) = |)

f

/

Gals Loy) = A} (00) “e)
O

hihi Pere eee eee Pee Sa
5) 1O iS 20 (2S) 30 S)S) 40

SALINITY pp.t

Ficure 9. Estimation of per cent mortality of second zoeal stage of Sesarma cinereum
based on the fitted response surface to observed mortality under twelve different combinations
of salinity and temperature.

prevail in nature, one may make an estimation of the distribution and mortality of
larval stages of this species in the natural environment. By using statistical
methods it is possible to postulate the response of the larvae under a greater variety
of environmental conditions than is possible from direct observations in the
laboratory.

The basic experimental design used in this study is usually referred to as a
factorial design. Specifically the plan was a 3 X 4 factorial using three levels of
temperature and 4 levels of salinity, making in all 12 combinations of experimental
conditions. If it is reasonable to assume that the effects of temperature and
salinity are continuous, and further, that a temperature-salinity interaction may
exist, then the 12 experimental combinations may be regarded as a sample in a
continuum of temperature and salinity. Proceeding further one may postulate
the existence of a continuous response (1.e., per cent mortality) as a function of
temperature and salinity, and that there may exist either unique optimum combina-
tions of the two factors or that several combinations of temperature and salinity

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 197

may produce the same response. The estimation of this functional relationship has
come to be called “‘the fitting of a response surface” (Box and Youle, 1955).
For the present study the surface fitted is described by the quadratic form

rs bo + b1x%1 + Doxe + bisx? + Dan%e? + DioxiXe

where y = arcsin Vper cent mortality, x, = temperature in °C., x2 = salinity
in p.p.t. Then 6) = constant, 6; = linear effect of temperature, 5. = linear
effect of salinity, 61: = quadratic effect of temperature, bo: = quadratic effect
of salinity, 61. = interaction effect between temperature and salinity.

The b’s were calculated from the experimental points and the observed mortality
by the method of least squares. A separate equation was obtained for each stage.
The contours of the surface for several selected mortality percentages were then
calculated and plotted in Figures 8, 9, 10, 11, and 12.

re

TEMPERATURE

2 10 Be) 20 eo 30 Go) 40

SALINITY pup.t.

Ficure 10. Estimation of per cent mortality of third zoeal stage of Sesarma cinereum
based on the fitted response surface to observed mortality under twelve different combinations
of salinity and temperature.

198 J..D..COSTLOW, JR.,\C) G4 BOOKHOUT AND ik NONR OE

ZG

TEMPERATURE

SALINITY) Fete

Ficure 11. Estimation of per cent mortality of fourth zoeal stage of Sesarma cinereum
based on the fitted response surface to observed mortality under twelve different combinations
of salinity and temperature.

The contours in Figures 8 through 12 represent the predicted values of
temperature and salinity which would produce the mortality indicated by the
contour line. It should be noted that the usual dangers of extrapolation beyond
the ranges of observed data are as inherent in this method of prediction as in
any other.

Two other uses may be made of the statistics computed in the fitting process:
(1) an estimated optimum combination, if one exists; and (2) tests of significance
of individual effects using the residual error mean square as an estimate of the
experimental error.

The following maximum or minimum conditions were observed :

Hemp; °C; Salinity, p.p.t.
Stage I 2555 21.9 min. mortality = 1.9%
Stage II 250 12.4 max. mortality = 20.4%
Stage III 26.0 24.1 min. mortality = 3.5%
Stage IV No maximum or minimum

Stage V No maximum or minimum

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 199

In making individual tests of significance the significance levels chosen were
5 per cent, 10 per cent, and 20 per cent since the exploratory nature of this study
was to identify possible leads toward important factors or combination of factors.
Effects testing significance at the 5 per cent level were denoted as ‘marked
effects,” those testing at the 10 per cent and 20 per cent level as “some effects.”
T and J? were used to denote, respectively, the linear and quadratic effects of
temperature, S and S* the same for salinity and (T xX S), the interaction. A
summary of these tests follows:

“Over-all

“Marked effects”’ “Some effects”’ correlation”’
Stage Il S) IE TEE 0.872
Stage Il Se 0.742
Stage III Te Ss Sade OSS) 0.829
Stage IV S246) SSS) 5 0.877
Stage V S (Pe 5S) 0.847

It is worth noting that the interaction effects which begin to show up in the
third stage are reflected in the contours in Figures 8 through 12. Stages I and II

re

TEMPERATURE

a
15 20 eo 30 6 he 40

SAN “aa
Ficure 12. Estimation of per cent mortality of megalops stage of Sesarma cinereum
based on the fitted response surface to observed mortality under twelve different combinations
of salinity and temperature.

200 J. D. COSTLOW, JR; CGC: Gs BOORHOUTL ANDER? MONK OE

with no interaction gave approximately concentric circles for contours. Stage III
shows a distortion of contours into pronounced ellipses. Stages IV and V show a
“ridge” pattern manifesting the interaction and the lack of a unique maximum
or minimum.

It is also possible to argue that the basic equations may be simplified by
eliminating those factors which showed no effect and re-fitting the simpler model to
the observed data. In some cases this may be desirable or even the objective of
the analysis. In this study, however, an attempt was made to describe the totality
of the response surface and to infer from the changing patterns with each zoeal
stage a possible explanation of the mortality associated with each molt. We have,
therefore, a sort of predicted mortality pattern as a function of temperature and
salinity from which to postulate about the basic mechanisms involved in the
system. Some such postulations which appear reasonable follow.

At 26-34 p.p.t. and 22-30° C., mortality of 10 per cent or less may be expected
in the first zoeae. With lower or higher temperatures and salinities, the mortality
increases as shown in Figure 8.

In the second zoeal stage the picture is quite different (Fig. 9). It was found
that less than 10 per cent mortality could be expected at all temperature-salinity
combinations other than 10-15 p.p.t., and 24-26° C. Even with these limits no
more than 20 per cent mortality would be expected. Thus, it appears that the
second larval stage is tolerant to almost any temperature-salinity condition which
might exist in an estuary.

As shown in Fig. 10, the tolerances of the third zoeal stage are somewhat simi-
lar to the pattern exhibited for the first zoeal stage. While the range of temperature
is only slightly changed (21-31° C.) the salinity range (23-28 p.pt.) is more
limited for a survival of 90 per cent or more. )

The pattern of survival for the fourth zoeal stage does not correspond to that
shown for any other (Fig. 11). As the temperature of the water increases, up
to 35° C., they can withstand reductions in salinity as low as 3 p.p.t. At lower
temperatures, however, mortality increases with slight changes in salinity.

The megalops stage (Fig. 12) is tolerant to a wide range of temperatures at
high salinities (30-40 p.p.t.) and, interestingly enough, at high temperatures it
can withstand low salinities. As the temperatures decrease, however, the resistance
to low salinity is more restricted.

The successful completion of the life history of S. cinereum appears to depend
largely on the fourth zoeal stage. That is, survival and molting to the megalops
can only occur in estuarine salinities. Those which may be carried out to the
high salinities of the sea, or trapped in tide pools of high salinity, would be killed.
Once the megalops stage is reached, however, it can withstand the salinity of the
ocean and a wide range of temperatures, and it can also survive in lower salinities
at higher temperatures.

CONCLUSIONS

From a study of 1266 zoeae of Sesarma cinereum (Bosc), reared in the labora-
tory under 12 different conditions of salinity and temperature, the following con-
clusions may be made:

1. Hatching of the first zoea of S. cinereum occurred successfully at salinities
of 12.5, 20.1, 26.7, and 31.1 p.p.t. None were observed to hatch as “pre-zoea.”

ENVIRONMENTAL EFFECTS ON CRAB LARVAE 201

2. Optimum salinities exist for the development of each larval stage. At 12.5
p-p.t. there was a delay in molting while at 26.7 and 31.1 p.p.t. molting proceeded
more rapidly.

3. At 12.5 and 31.1 p.p.t. none of the larvae successfully completed development
to the first crab stage. Mortality of zoeae at 12.5 p.p.t. was highest at the first
zoeal stage and spread throughout the remaining larval stages. Mortality was
highest in the fourth zoeal stage at 31.1 p.p.t. Some zoeae completed development
to the crab at 20.1 p.p.t. but the highest per cent survival (24.4) to the first crab
stage occurred at 26.7 p.p.t.

4. Temperature has a more definite effect on length of larval development than
on mortality. At 20° C., 35-39 days were required for complete development to
the first crab. At 30° C., comparable development occurred in 20-22 days.

5. Variations in temperature and salinity did not produce “extra” stages in
S. cinereum. Those that completed development always passed through four zoeal
stages and one megalops.

6. Using statistical methods, the maximum and minimum conditions for survival,
the effects of salinity, temperature, and the combined effects of salinity and tem-
perature have been postulated for each larval stage within a wide range of salinity-
temperature combinations. The statistical analysis, as well as the observed results,
show that salinity is the chief physical factor which confines S. cinereum to an
estuarine environment.

LITPRATURE ClreDp

AmeEmiIvA, I., 1926. Notes on experiments on the early developmental stages of the Portuguese,
American, and English native oysters, with special reference to the effect of varying
salinity. J. Mar. Biol. Assoc., 14: 61-175.

Birce, E. A., 1883. Notes on the development of Panopaeus sayt. Stud. Biol. Lab. Johns
Hopkins University, 2: 411-426.

BousFietp, E. L., 1955. Ecological control of the occurrence of barnacles in the Miramichi
estuary. National Museum of Canada, Bulletin 37.

Box, G. E. P., Anp P. V. Youre, 1955. The exploration and exploitation of response surfaces :
An example of the link between the fitted surface and the basic mechanism of the
system. Biometrics, 11: 287-323.

Broan, A. C., 1957. The relationship between diet and larval development of Palaemonetes.
Biol. Bull., 112: 162-170.

BRoEKHUYSEN, G. J., 1937. On development, growth, and distribution of Carcinides maenas
(L.). Arch. Néerland de Zoologie, 2: 257-399.

CHAMBERLAIN, N. A., 1957. Larval development of the mud crab Neopanope texana sayi
(Santh). Broil. Buil., 113: 338.

CuurcHitt, E. P., 1942. The zoeal stages of the blue crab, Callinectes sapidus Rathbun.
Ches. Biol. Lab. Publ. No. 49, pp. 1-26.

Ciark, A. E., 1935. Effects of temperature and salinity on early development of the oyster.
Prog. Rept. Atl. Biol. St., St. Andrews, N. B., 16: 1-10.

Corrin, H. G., 1958. The laboratory culture of Pagurus samuelis (Stimpson) (Crustacea,
Decapoda). Walla Walla College Publ. No. 22, pp. 1-5.

Conno.ty, C. J., 1925. The larval stages and megalops of Rhithropanopeus harrisi (Gould).
Contr. Canad. Biol., N. S., 2: 329-337.

CostLtow, J. D., Jr. AND C. G. Booxnout, 1959. The larval development of Callinectes
sapidus Rathbun reared in the laboratory. Biol. Bull., 116: 373-396.

CostLtow, J. D., G. Rees anp C. G. Booxuout, 1959. A preliminary note on the complete
larval development of Callinectes sapidus Rathbun under laboratory conditions.
Limnol. Oceanogr., 4: 222-223.

202 J. Di COSTEOW wy CAGeBOOREOU LAN DERY MONROE

Davis, H. C., 1958. Survival and growth of clam and oyster larvae at different salinities.
Biol. Bull., 114: 296-307.

GomPeEL, M., anp R. Lrecenpre, 1927. Effets de la Température, de la Salure et du pH sur
les larves de Homards. C. R. Soc. Btol., 94: 1058-1060.

Hart, J. F. L., 1935. The larval development of British Columbia Brachyura. Canadian J.
Res., 12: 411-432.

Hyman, O. W., 1924. Studies on the larvae of crabs of the family Grapsidae. Proc. U. S.
Nai. Mus., 65: Art. 10, 1-8.

Hyman, O. W., 1925. Studies on the larvae of crabs of the family Xanthidae. Proc. U. S.
Nat. Mus., 67: Art. 3, 1-22.

Knupsen, J. W., 1958. Life cycle studies of the Brachyura of Western North America, I.
General culture methods and life cycle of Lophopanopeus leucomanus leucomanus
(Lockington). Bull, So. Calif. Acad. Sci, 57: 51-59.

Lesour, M. V., 1928. The larval stages of the Plymouth Brachyura.’ Proc’ Zoole soc.
London, 473-560.

LoosANorF, V. L., 1948. Gonad development and spawning of oysters (O. virginica) in low
salinities. Anat. Rec., 101: 55.

LoosAnorr, V. L., anp H. C. Davis, 1950. Conditioning V. mercenaria for spawning in
winter and breeding its larvae in the laboratory. Biol. Bull., 98: 60-65.

LoosANorr, V. L., W. S. MILLER anp P. B. Smiru, 1951. Growth and setting of larvae of
Venus mercenaria in relation to temperature. J. Mar. Res., 10: 59-81.

Mortenson, TH., 1931. Contributions to the study of the development and larval forms of
Echinoderms I-II. Kgl. Danske Vidensk. Selsk. Skrifter, Naturh. and Math. Afd.,
Raekke 9, 4: 1-39, Kobenhavn.

MorTENSON, Tu., 1937. Contributions to the study of the development and larval forms of
Echinoderms III. Kgl. Danske Vidensk. Selsk. Skrifter, Naturh. and Math. Afd.,
Raekke 9, 7: 1-65, Kobenhavn.

Mortenson, Tu., 1938. Contributions to the study of the development and larval forms of
Echinoderms IV. Kgl. Danske Vidensk. Selsk. Skrifter, Naturh. and Math. Afd.,
Raekke 9, 7: 1-59, Kobenhavyn.

PEARSE, A. S., 1936. Estuarine animals at Beaufort, North Carolina. J. Elisha Mitchell
Sct. Soc., 52: 174-222.

RatusBun, M., 1917. The Grapsoid crabs of America. U. S. Nat. Mus., Bull. 97, 461 pp.

Sanpoz, M., ann S. H. Hopxins, 1947. Early life history of the oyster crab, Pinnotheres
ostreum (Say). Biol. Bull., 93: 250-258. :

Sanpoz, M., AND R. Rocers, 1944. The effect of environmental factors on hatching, moulting,
and survival of zoea larvae of the blue crab, Callinectes sapidus Rathbun. Ecology,
25: 216-228.

Sanpoz, M., anp R. Rocers, 1948. The effect of temperature and salinity on moulting and
survival of megalops and post-larval stages of the blue crab, Callinectes sapidus. Va.
Fish. Lab., unpubl. MS, 12 pp.

SCHLEGEL, C., 1911. Sur 1 développement de Maia squinado, Latr. C. R. Acad. Sct., Paris,
153: 480-482.

SoLLAup, E., 1919. Influence des conditions du milieu sur les larves du Palaemonetes varians
microgenttor Boas. C. R. Acad. Sci., Paris, 169: 735-737.

Sutc.iFFE, W. H., 1950. A qualitative and quantitative study of the surface zooplankton at
Beaufort, North Carolina. Ph.D. Thesis, Duke University, unpublished. 137 pp.

TEMPLEMAN, W., 1936a. The influence of temperature, salinity, light, and food conditions on
the survival and growth of the larvae of the lobster (Homarus americanus). J. Biol.
Bd. Canada, 2: 485-497.

TEMPLEMAN, W., 1936b. Fourth stage larvae of Homarus americanus intermediate in form
between normal third and fourth stages. J. Biol. Bd. Canada, 2: 349-354.

Witiiamson, H. C., 1903. On the larval and early young stages, and rate of growth, of
Carcinus maenas (L.). 21 Rep. Fish. Board for Scotland, vol. 3, Sci. Invest., 136-179.

PeeeeOMPLETE LARVAL DEVELOPMENT OF SESARMA
PeveReUM (BOSC) REARED IN THE LABORATORY 4

FOMN DD. COSTLOW, JR: AND C. G. BOOKRHOUT

Duke University Marine Laboratory, Beaufort, North Carolina and the Department of
Zoology, Duke University, Durham, North Carolina

Numerous incomplete descriptions exist of larval development of the Grapsidae
but they are usually based on reconstructions from planktonic material or confined
to the first zoeal stage obtained from hatching in the laboratory. Cano (1891)
described the development of Pachygrapsus marmoratus up to the crab stage but
Hyman (1924) questioned the staging of the three zoeal stages by Cano (1891)
and describes 5 zoeal stages from material collected at Beaufort, N. C. Conn
(1884), comparing portions of larvae of Brachyura and Macrura, includes some
appendages and the telson of Sesarma before and after hatching. Hyman (1924)
gave a description of the first zoeae and appendages of Sesarma reticulata and, in
describing the first stage of S. cinereum, noted that the two species are very similar
morphologically. Aikawa (1929) described the first zoeae of Sesarma sp. briefly
and figured the first stage of S. picta (1937), comparing setation of some append-
ages with the description given by Hyman. Thus only the first zoeal stage of any
species of Sesarma has been described.

The purpose of the present study has been to rear the larvae of Sesarma
cinereum (Bosc) in the laboratory, from hatching to the first crab stage, and
provide a description of all larval stages.

The authors wish to express their appreciation to Mrs. W. A. Chipman and Mrs.
Doris H. King for their assistance throughout the study. We also wish to thank
Dr. Fenner A. Chace for identification of the adult females from which the eggs
were obtained. Adults from which the zoeae were obtained have been deposited
with the U. S. National Museum.

METHODS

Ovigerous Sesarma cinereum females were maintained in finger bowls con-
taining filtered sea water until hatching occurred. The larvae were reared in
groups of 10 zoeae per bowl and fed Artemia nauplii and Arbacia eggs. They
were changed to freshly filtered water in clean bowls and the bowls were examined
daily for exuviae. Larvae were preserved in Bouin’s fixative at known intervals
during development and exuviae were kept in 70 per cent alcohol to check
setation of appendages.

Drawings were made to scale with the aid of a Whipple disc mounted in the
_ocular of a compound microscope. The larval appendages were dissected from

1 These studies were aided by a grant (G 4400) from the National Science Foundation.

203

204 JOHN D. COSTLOW, JEo AND (CG BOCKHOUT

+
eee
| Be

Ficure 1. Side (A) and ventral view (B) of first zoeal stage of Sesarma cinereum with
appendages. C, antennule: D, antenna; E, mandible; F, maxillule; G, maxilla; H, endopodite
of first maxilliped; I, endopodite of second maxilliped.

LARVAL STAGES OF SESARMA’ CINEREUM 205,

each stage and drawn to a different scale from that used for the whole larvae. The
chromatophore pattern was determined from living zoeae of known age.

RESULTS

There are four zoeal stages and one megalops stage in the complete larval
development of Sesarma cinereum (Bosc). The major characteristics of each
larval stage are as follows:

First zoea (Fig. 1)

The cephalothorax has a short, tapering dorsal spine which curves caudally.
The rostral spine is short, approximately equal in length to the antennae and twice
the length of the antennule. The eyes are not stalked. The abdomen consists of
5 segments plus the telson (Fig. 1, A). The second abdominal segment has a
short lateral knob, directed laterally, and a small spine is present on the mid-lateral
line of the third abdominal segment. The posterior margins of abdominai seg-
ments 2—5 terminate in short, slightly rounded lateral spines which overlap the
next segment. There are 6 spines on the inner surface of the telson (Fig. 1, B).
The pattern of chromatophores is consistent for all four zoeal stages of S. cimereum.
The location of the melanophores is as follows: 1, median to the eyes, extending
ventro-laterally on each side of the cephalothorax; 2, a single chromatophore,
median and slightly dorsal to the eyes; 3, ventral to the heart; 4, mandibles and
labrum ; 5, basopodite of first maxilliped ; 6, first abdominal segment, dorsal to gut;
7, posterio-ventral border of remaining abdominal segments.

The short, conical antennule (Fig. 1, C) bears three terminal aesthetes of
equal length and two smaller, non-plumose setae. The antenna (Fig. 1, D) con-
sists of a protopodite which tapers gradually to a point and bears stiff hairs on
the distal portion. The exopodite is a distinct segment, approximately one-third
the length of the protopodite, and terminates in two, non-plumose setae of unequal
length. The mandible (Fig. 1, E) has a large ventral tooth and several smaller
teeth. The two-segmented endopodite of the maxillule (Fig. 1, F) bears four
terminal plumose setae plus two shorter setae and a third, longer seta. The
basal and coxal endites both bear 5 spines. The unsegmented endopodite of the
maxilla (Fig. 1, G) is slightly bifurcated and has three terminal and two sub-
terminal spines. The basal endite, also bifurcated, bears a total of 9 spines and
7 spines project from the coxal endite. Four soft, plumose hairs fringe the distal
border of the scaphognathite and the apical tip bears one hair. The setation of
the 5-segmented endopodite of the first maxilliped (Fig. 1, H) is 1, 1, 1, 2, 4, and
the exopodite bears 4 plumose swimming setae (Fig. 1, B). Setation of the three
segments of the endopodite of the second maxilliped (Fig. 1, I) is 0, 1, 5, and
four plumose swimming setae extend from the exopodite (Fig. 1, B).

Second zoea

The eyes are stalked. A spine projects from the protopodite of the maxillule,
just below the endopodite (Fig. 2, E). Seven spines project from the basal endite,
and the coxal endite has 6 spines (Fig. 2, E). The endopodite of the maxilla
(Fig. 2, F) bears three terminal and two subterminal setae. This arrangement

206 JOHN D. COSTLOW, JRo ANDI CG. (BOORHOUT

G

Figure 2. Side (A) and ventral view (B) of second zoeal stage of Sesarma cinereum
with appendages. C, antennule; D, antenna; E, maxillule; F, maxilla; G, endopodite of first
maxilliped; H, endopodite of second maxilliped.

LARVAL STAGES OF SESARMA CINEREUM 207

remains unchanged in later larval stages. The endites of the protopodite are
bifurcated and a total of 10 spines project from the basal endite while the coxal
endite bears 8 spines. The scaphognathite (Fig. 2, F) has 5 soft hairs on the
distal margin and the apical tip bears three hairs. The setation of the endopodite
of the first maxilliped is 2, 2, 1, 2, 5, and 6 swimming setae are found on the
exopodite (Fig. 2, B and G). On the endopodite of the second maxilliped the
setation is 0, 1, 6, and the exopodite bears 6, swimming setae (Fig. 2, B and H).

Third zoea

The third maxilliped, chela and pereiopods are visible as small buds beneath
the cephalothorax (Fig. 3, A). The abdominal segments have increased to 6.
Segments 2—5 bear small pleopod buds. A pair of small setae fringe the posterior
margins of segments 2—6 on the dorsal surface.

iimenendopodite bud of the antenna (Fig. 3, D) is equal in length to the
exopodite but is not segmented. The basal and coxal endites of the maxillule
bear 8 and 6 spines, respectively (Fig. 3, E). Ten spines are found on the basal
endite of the maxilla (Fig. 3, F) and the coxal endite has 9 spines. The number
of hairs on the distal margin of the scaphognathite has increased to 8 and 4 hairs
project from the apical tip. Each of the exopodites of the first and second maxil-
lipeds now bears 8 swimming setae while setation of the endopodites remains as
in the second stage (Fig. 3, Gand H). The rudiment of the third maxilliped is
present but is unsegmented and has no setation (Fig. 3, I).

Fourth zoea

The rudimentary thoracic appendages are further developed and project below
the border of the cephalothorax (Fig. 4, A). The lateral margins of the cephalo-
thorax are fringed with short setae. Pleopod buds are larger and on segments
2—5 the buds have short setae (Fig. 4, A and B). A pair of small spines is added
on the inner margin of the telson (Fig. 4, B).

The antennule (Fig. 4, C) is inflated in the basal region and terminates in 6
flat aesthetes of varying size and one short seta. The endopodite of the antenna
(Fig. 4, D) is equal in length to the protopodite, terminates in one short, non-
plumose spine, and is partially segmented. The mandible (Fig. 4, E) has numerous
teeth on the cutting surface and a small bare, unsegmented palp is added. Eleven
stiff spines project from the basal endite of the maxillule (Fig. 4, F) and the coxal
endite has a total of 7 spines. The basal endite of the maxilla (Fig. 4, G) bears
12 spines and the coxal endite has 12. The scaphognathite now has a total of 23
soft hairs on the periphery. Setation of the endopodite of the first maxilliped has
increased to 2, 3, 1, 2, 6, and there are 9 swimming setae on the exopodite (Fig. 4,
Band H). The exopodite of the second maxilliped has 10 swimming setae. The
third maxilliped (Fig. 4, J) is partially segmented and bears a few, short setae.

Megalops

Cephalothorax without rostal or dorsal spines and provided with hairs at
lateral edges (Fig. 5, A and B). Center of short rostrum depressed. Abdomen
of six segments and telson which bears 8 short setae on distal margin. One small

208 JOHN -DisCOSTLOW, JEU ANDAG GC, BOOKHOUT

Figure 3. Side (A) and ventral view (B) of third zoeal stage of Sesarma cinereum with
appendages. C, antennule; D, antenna; E, maxillule; F, maxilla, G, endopodite of first
maxilliped; H, endopodite of second maxilliped; I, third maxilliped.

LARVAL STAGES OF SESARMA CINEREUM 209

Figure 4. Side (A) and ventral view (B) of fourth zoeal stage of Sesarma cinereum
with appendages. C, antennule; D, antenna; E, mandible; F, maxillule; G, maxilla: El
endopodite of first maxilliped; I, endopodite of second maxilliped; J, third maxilliped.

,

210 JOHN D. COSTLOW, JR. AND C. G. BOOKHOUT

0.1 mm

Figure 5. Side (A) and dorsal view (B) of megalops of Sesarma cinereum with
C, antennule; D, antenna, E, mandible; F, maxillule; G, maxilla; H, first

appendages.
J, third maxilliped ; K, ventral view of abdominal segments.

maxilliped; I, second maxilliped ;

LARVAL STAGES OF SESARMA CINEREUM 211

dorsal spine on posterior margin of abdominal segments 2-6. The chromatophore
pattern is as follows: 1, two on the dorsal surface of rostrum; 2, anterior and
posterior peripheries of eye stalks; 3, lateral borders of cephalothorax ; 4, posterior
edge of carapace, dorsal to gut; 5, dorso-lateral surfaces of abdominal segments
1-6; 6, mandibles and labrum.

Antennule (Fig. 5, C) composed of enlarged base and peduncle with a flagellum
of three short segments bearing a terminal tier of 5 aesthetes and a subterminal
tier of 6 aesthetes. Antenna (Fig. 5, D) of 9 segments with setae concentrated on
distal three segments. Mandible (Fig. 5, E) with large ventral tooth and a palp
of two segments bearing four short, stiff spines on fringe of second segment.
Maxillule (Fig. 5, F) with four terminal setae on endopodite and two additional
setae on basal segment. Basal endite has 12 spines and coxal endite has 9 spines.
Endopodite of maxilla (Fig. 5, G) with three terminal and two subterminal spines.
Basal endite bears a total of 14 spines, and 13 project from coxal endite. The
margin of the scaphognathite is fringed with approximately 30 soft hairs. The
two-segmented exopodite of the first maxilliped has 5 terminal setae and three
setae project from the first segment at the junction of the segments (Fig. 5, H).
The unsegmented endopodite is reduced and bears four terminal spines. The basal
endite of the protopodite has 8 prominent spines and the coxal endite bears 5.
The epipodite has two setae on the basal portion and terminates in three hairs.
The second maxilliped (Fig. 5, I) has an endopodite of four segments with stiff
spines on the terminal segment. The two-segmented exopodite has 5 terminal
setae. The third maxilliped (Fig. 5, J) has well developed endopodite of 5 seg-
ments, all bearing stiff spines. The two segmented exopodite terminates in 5 setae.
The epipodite is fringed on the basal margin by numerous short spines and 8 soft
non-plumose hairs arise from the distal portion.

The dactylopodite of the fifth pereiopod has three long setae with minute serra-
tions on the inner surface (Fig. 5, B). Setation of the exopodites of the pleopods
of abdominal segments 2-6 is 13, 13, 13, 11, and 8. Endopodites with one short
seta are found on all but the uropods. Two small hooks are located on the inner
surface of all endopodites of the pleopods.

DISCUSSION

A comparison of the results of the present work with previous descriptions of
Sesarma larvae is limited to the work of Hyman (1924) and Aikawa (1937) with
first stage zoeae. Setation of the exopodites of the maxillipeds is identical for
first zoea of S. cinereum, S. reticulata (Hyman, 1924), and S. picta (Aikawa,
1937) but, since four swimming setae are found on the maxillipeds of most first
zoea described to date, the feature is of little diagnostic value. Hyman’s (1924)
figures indicate that some slight differences exist in setation of the maxilla. We found
three terminal and two subterminal setae on the endopodite of the maxilla and
Hyman (1924) figures a 3:1 setation for the first zoea of S. reticulata. In the
present study the coxal endite of the maxilla had 7 spines and Aikawa (1937)
shows 8 spines for the first zoea of S. picta. In previous studies on a variety of
decapod larvae (Hopkins, 1943, 1944; Hart, 1935; Connolly, 1923; Costlow and
Bookhout, 1959) the setation of all appendages has been shown to increase pro-
gressively through larval development. Thus, while the differences in setation of

212 JOHN DU COSTLOW: JRZANDFG, GCG BOOKHOUT

S. cinereum, S. reticulata, and S. picta first zoea are slight, they may become
greater as larval development continues.

The carapace of S. cinereum does not contain lateral spines. Although Hyman
(1924) noted the absence of lateral spines on the carapace of first zoea of S.
reticulata and S. cinereum, Aikawa (1929) groups Sesarma sp. with larvae of
Heterograpsus and Hemigrapsus because the first zoeal stage had rostral, dorsal,
and lateral spines. A later description of the first zoea of Sesarma picta (Aikawa,
1937) notes the absence of lateral spines. In most of these descriptions the first
zoea had been figured from larvae hatched in the laboratory. The presence of
lateral spines on the carapace of Sesarma sp. (Aikawa, 1929) must have been the
result of improper identification of the adult crab.

In the present study all larvae which completed development passed through
four zoeal stages. Hart (1935) described five zoeal stages from the laboratory
rearing of two other Grapsidae (Hemigrapsus nudus and Hemigrapsus oregonensis )
and Hyman (1924) assigned 5 zoeal stages to Pachygrapsus marmoratus by re-
- constructing the sequence of larval development from planktonic material. Hyman
(1924) notes (p. 2), “The metamorphosis of the family seems to follow the usual
brachyuran formula. There are at least three zoeal stages and there are probably
five.’ Aikawa (1929) criticizes Hyman (1924) for his generalizations concerning
the general structure of the grapsid larvae. The presence of four zoeal stages in
the larval development of S. cinerewm further points out the danger in generalizing
on the number of larval stages a particular species of decapod “should” have.
While some other grapsid crabs have been shown to have five zoeal stages, recent
studies on the larval development of a variety of Brachyura larvae reared in the
laboratory (Costlow and Bookhout, 1959; unpublished results) have shown that
zoeal stages may range in number from 2 to 8, depending upon the species.

In the present study only one megalops stage was observed for S. cimereum.
Hyman (1924) figures and described two megalops stages for Pachygrapsus
marmoratus, although he does not indicate whether he actually observed the molt
from the first to the second megalops or based the description on two slightly
different megalops obtained from the plankton. Although only one megalops stage
has been described for most crabs, Aikawa (1937) figures two stages for Plagusia
dentipes and found three distinct modes in megalops body length. From Aikawa’s
(1937) account it is difficult to determine whether or not he actually observed the
molt to the second megalops stage. Lebour (1928) discusses the several megalops
stages attributed to Neopanope texana sayi by Birge (1883) and by Hyman (1925)
and concludes it is probable that only one stage exists. More recent studies on
laboratory-reared larvae (Chamberlain, 1957) have shown only one megalops
stage for N. texana sayi. Although only one megalops stage may exist under
optimum conditions, it is conceivable that dietary deficiencies could produce a second
megalops stage in the same manner in which “extra stages” in the larval develop-
ment of other species have been associated with food (Templeman, 1936; Broad,
1957,);.

Studies on megalops stages of other Grapsidae (Hyman, 1924; Hart, 1935;
Aikawa, 1937) have figured an antennule composed of a swollen base with the
distal segment of the peduncle bearing a segmented and an unsegmented flagellum.
In S. cinereum megalops the segmented flagellum with the sensory aesthetes is

LARVAL STAGES OF SESARMA CINEREUM 23

present (Fig. 5, C) but the unsegmented flagellum is missing and a small spine
remains in its place. The absence of the unsegmented flagellum on the antennule
may possibly be used to differentiate S. cinereum from the megalops of other species
which have been described previously.

Recent publications on rearing larvae in the laboratory should contribute to
our knowledge of larval development and to our ability to identify planktonic forms.
If complete descriptions of appendages and setation are omitted, the value is con-
siderably reduced. Without accurate descriptions of possible diagnostic features,
comparative studies on planktonic material and the positive identification of larval
-decapods will not be possible.

SUMMARY AND CONCLUSIONS

The larval development of Sesarma cinereum (Bosc) has been followed in the
laboratory from hatching to the first crab stage. Twelve hundred zoeae were
maintained on Artemia nauplii and Arbacia eggs under constant conditions of
temperature, salinity, and light. From these studies the following conclusions
‘may be made:

1. Four zoeal stages and one megalops stage are found in the complete larval
development to the first crab. These are described and figured. No variation in
the number of zoeal stages was observed.

2. The setation of all functional appendages in each of the four zoeal stages
has been described and figured and may be used as possible diagnostic characteristics
for larvae of Sesarma cinereum. Setation of the maxillule and maxilla increases
progressively with larval development whereas setation of the endopodites of the
maxillipeds does not change with the second and third stage zoea.

3. The setation of all appendages of the megalops has been described and
figured. The absence of unsegmented flagellum on the antennule of the megalops
may serve as a diagnostic character for the differentiation of Sesarma cinereum
megalops in planktonic material.

EIPERATURE CITED

AIKAWA, oe 1929, On larval forms of some Brachyura. Rec. Oceanogr. Wks. Japan, II:
17-55.

a) ee 1937. Further notes on Brachyuran larva. Rec. Oceanogr. Wks. Japan, IX:

-162.

Birce, E. A., 1883. Notes on the development of Panopaeus sayi. Stud. Biol. Lab. Johns
Hopkins Univ., 2: 411-426.

Broan, A. C., 1957. The relationship between diet and larval development of Palaemonetes.
Biol. Bull., 112: 162-170.

Cano, G., 1891. Sviluppo postembrionale dei Dorippidei, Leucosiadi, Corystoidei e Grapsidi.
Mem. Soc. Ital. Sci. Nat., 8: 1-14.

CHAMBERLAIN, N. A., 1957. Larval development of the mud crab Neopanope texana sayi
(Smith). Biol. Bull., 113: 338.

Conn, H. W., 1884. The significance of the larval skin of Decapoda. Stud. Biol. Lab. Johns
Hopkins Univ., 3: 1-27.

Connotiy, C. J., (1923). The larval stages and megalops of Cancer amoenus (Herbst).
Contrib. Canad. Biol., N. S., 1: 337-352.

CostLow, J. D., Jr., AND C. G. Booknout, 1959. The larval development of Callinectes sapidus
Rathbun reared in the laboratory. Biol. Bull., 116: 373-396.

214 JOHN: D?-COSTLOW, JRVAND, G)G.-BOOKHOUT

Hart, J., 1935. The larval development of British Columbia Brachyura. I: Xanthidae, Pin-
notheridae and Grapsidae. Canad. J. Res., 12: 411-432.
Hopkins, S. H., 1943. The external morphology of the first and second zoeal stages of the
blue crab, Callinectes sapidus Rathbun. Trans. Amer. Micro. Soc., 62: 85-90.
Hopkins, S. H., 1944. The external morphology of the third and fourth zoeal stages of the
blue crab, Callinectes sapidus Rathbun. Biol. Bull., 87: 145-152.

Hyman, O. W., 1924. Studies on larvae of crabs of the family Grapsidae. Proc. U. S. Nat.
Museum, 65: Art. 10, 1-8.

Hyman, O. W., 1925. Studies on the larvae of crabs of the family Xanthidae. Proc. U. S.
Nat. Museum, 67: Art. 3, 1-22.

Lesour, M. V., 1928. The larval stages of the Plymouth Brachyura. Proc. Zool. Soc.
London, 473-560.

TEMPLEMAN, W., 1936. Fourth stage larvae of Homarus americanus intermediate in form
between normal third and fourth stages. J. Biol. Bd. Canada, 2: 349-354.

eee? OF TEMPERATURE AND SALINITY ON THE OXYGEN
eo oUME TION OF TWO TNTERTIDAL’ CRABS*

Pepe Ne Oe ENE Te

Department of Zoology, University of British Columbia, Vancouver, British Columbia, Canada

Studies on the effect of temperature change in poikilotherms have demonstrated
physiological adaptations which permit varying degrees of independence of environ-
mental temperatures. Various rate functions proceed within certain temperature
limits, at activity levels which show compensatory responses to temperature.
Mechanisms that result in such adaptations allow for degrees of thermal homeostasis
in physiological processes. Latitudinally, thermal acclimation of rate functions
has been demonstrated for many marine poikilotherms (Rao, 1953; Scholander
fg, 1993; Dehnel, 1955, 1956; Roberts, 1957b; and many others). Seasonal
acclimation to temperature is documented (Edwards and Irving, 1943a; Clark,
1955; Roberts, 1957b), as well as intertidal vertical distribution (Segal, 1956).
A general conclusion, based on intra- and interspecific comparisons, may be drawn
from these data. Rate functions, such as weight-specific oxygen consumption,
growth, heart beat, show that northern latitude populations, at their own natural
temperature, have activity rates comparable to winter-adapted populations (at other
latitudes), and low intertidal populations. Further, these three groups, when
compared with their ecological opposites (southern latitude, summer-adapted, or
high intertidal populations), at the same experimental temperature have generally
higher rates of activity. Bullock (1955) and Prosser (1955) have documented
extensively these considerations for poikilotherms.

Another environmental parameter, which has been given considerable attention
by physiological ecologists is the effect of salinity on animal activity. Some of
these studies have concerned mainly responses of poikilotherms to various osmotic
concentrations, and determinations of gain and loss of water and ions in body fluid
and urine, and accompanying weight changes (Jones, 1941; Robertson, 1949, 1953;
Gross, 1954, 1955, 1957; Prosser, Green and Chow, 1955).

Other studies have concerned the effect of various osmotic conditions on
metabolic activity. Flemister and Flemister (1951) working with Ocypode
albicans determined that oxygen consumption at 26° C. was lowest in sea water
(378 mM. C1/L.), which was isotonic with crab blood, this water being considerably
more hypotonic than field conditions at the time of collection (480 mM. CI/L.).
As the concentration of sea water varied from isotonicity, oxygen consumption
increased, the highest rates being found in hypotonic solutions. Lofts (1956)
compared respiratory rates at various salinities from tap water to 65%o, of two
populations of the prawn Palaemonetes varians, one from a low salinity environ-

1 These studies were aided by grants from the National Research Council of Canada and
the National Science Foundation of the United States. The author wishes to express his
gratitude to Dr. Earl Segal and Dr. Theodore H. Bullock for their critical reading of the
manuscript. —

215

216 PAULA. DEANEL

ment (1.3%0 NaCl) the other from a high salinity one (23.5%. NaCl). In both
groups the rate of oxygen consumption decreased as the salinity increased from tap
water. Minimal respiratory rate for the high salinity population was at a salinity
of 26%o, a condition isotonic with the animals. Minimal rate for the low salinity
population occurred at 6%o. This latter value is somewhat hypertonic to the
environment, and the prawns from this area are correspondingly hypertonic.
Oxygen consumption increased as the experimental salinities increased beyond
those of the natural environment. Gross (1955) has shown oxygen consumption
to decrease as a function of desiccation in Pachygrapsus crassipes. Short term
desiccation (several hours) is believed not to be deleterious. In a later paper,
Gross (1957) measured oxygen consumption in Uca at 16° C. as a function of
external concentration of the medium. His results showed that oxygen con-
sumption did not always increase with osmotic stress. These results disagree with
those of Flemister and Flemister (1951) for Ocypode. Marshall, Nicholls and
Orr (1935) found that oxygen consumption of Calanus measured in 50% sea
water and 15° C. was about 70% of respiration recorded in normal sea water
(34%). No generalization may be given regarding the effect sea water con-
centration has on oxygen consumption.

More recently, the combined effects of temperature and salinity on animal
activity have been investigated. Broekema (1941) used different temperature-
salinity combinations to determine the combined effect of these parameters on
length of life of the shrimp Crangon crangon. She found that the optimal salinity
for two-year-old shrimp was about 33% at a temperature of 4° C., whereas at 20°
to 22° C. the optimal salinity was 28 to 29%. When temperature decreases,
salinity must rise. This condition was noted also for younger shrimp, larvae and
eggs. Isotonicity is partially dependent upon temperature; at 20° C. isotonicity
occurs at about 21.5%, at 4° C., at 23%. Further, with respect to osmotic be-
haviour, as temperature drops blood concentration rises in the hypotonic portion of
the salinity range, and falls in the hypertonic part. These data support the fact
that Crangon tolerates low salinities better when the temperature is high. Smith
(1955a) has suggested that summer salinities in the Baltic Sea are not limiting
factors for Nerets diversicolor, but low spring salinities and temperatures adversely
affect osmoregulatory abilities, thus presenting an ecological limitation. In another
paper Smith (1955b) compared chloride regulation in several geographically
separated populations of Nereis diversicolor. One population was collected from
upper River Tamar, England, and the level of chloride regulation was determined
at a series of temperatures (7°, 14° and 21° C.) and over a range of chlorinities
(0.02 grams chloride/liter to 14.08 grams chloride/liter). For a given chloride
content of the medium the three temperatures produced no significant differences in
level of chloride regulation. Further work by Smith (1957) on chloride regulation,
as affected by temperature, in populations of Nereis lighti from the Salinas River,
California has shown that at chlorinities above 1.0 grams/liter, low temperatures
(0.5° C.) did not affect chloride regulation, when adapted to low salinities prior to
exposure to low temperatures. However, animals adapted to 0.5° C. initially, in
a chlorinity of 2.0 grams/liter, and then transferred to fresh water, failed to show
chloride regulation, but did show volume regulation. At higher adaptation tem-
peratures (12° C.) salinity reduction did not result in lowering of the coelomic
chloride level.

RESPIRATION IN CRABS 21s

In the last few years Kinne (1953, 1956a, 1956b) in a series of papers has
reported the physiological effect of temperature and salinity on several species of
invertebrates, with particular reference to growth in hydroids. Kinne (1957)
states (p. 90): “temperature can change (enlarge, narrow or shift) the salinity
range, and salinity can change the temperature range of a species. The effect
of a given temperature depends on the salinity and vice versa.’ Kinne and
Rotthauwe (1952) have presented data for Rithropanopeus harrisi showing that as
temperature decreases blood concentration rises along the entire length of the
curve, except for salinities above approximately 30%o. This crab is hypertonic and
normally lives in waters of very low salinity (1 to 5%). Rithropanopeus has
been shown to withstand low salinities better when the temperature is low. It
should be pointed out that these conditions are the reverse to those reported by
Broekema (1941) for Crangon. Kinne (1956b) observed growth and reproduc-
tion in the hydroid Cordylophora caspia under different temperature and salinity
combinations. It was determined that these hydroids withstood high temperatures
at high salinities better than at lower salinities.

The combined effect of temperature and salinity has been studied with regard
to its influence on temperature tolerance. The pertinent literature has been
reviewed recently (Todd and Dehnel, 1960). In this laboratory Todd and Dehnel
(1960) investigated the influence of seasonal change and laboratory acclimation
to various temperature-salinity combinations on heat tolerance of Hemigrapsus
oregonensis and H. nudus. They found that there was a seasonal change in both
species when winter and summer data were compared. Further, acclimation to
a high temperature increased resistance to lethal temperatures and acclimation to
low salinities decreased this resistance. For both seasons, winter and summer, a
combination of high temperature and high salinity proved most favorable for
resistance to lethal temperatures.

The present investigation is a study of the combined effect of temperature and
salinity on respiratory metabolism in two species of intertidal crabs, Henugrapsus
oregonensis and H. nudus. The facts that both species, in this geographic locality,
occupy similar ecological niches, occur in abundant numbers, and are maintained
readily under laboratory conditions permit inter- as well as intraspecific com-
parisons. It has been demonstrated that the degree (absolute and relative) to
which these crabs acclimate to a given parameter, either temperature and/or
salinity, depends upon the season of the year at which they were collected and the
experimental temperature-salinity combination imposed at either season. This
work extended from the winter, 1955, to the summer, 1957.

MATERIAL AND METHODS

Abundance and suitability to laboratory conditions of these two eurytopic
species, Hemigrapsus oregonensis (Dana) and H. nudus (Dana) permit studies of
temperature and salinity acclimation. Schmitt (1921) lists distribution of H.
nudus from Sitka, Alaska to the Gulf of California, and H. oregonensis from
Prince William Sound, Alaska to the Gulf of California.

Collections of crabs were obtained from Spanish Bank (Latitude, 49° 17’ N;
longitude, 123° 07’ W), Vancouver, British Columbia (Fig. 1). Specific areas on
the beach were marked and animals were collected only from these regions. H.

218 PAUL A. DEHNEL

oregonensis was collected at approximately the 7.0-foot tide level and H. nudus at
the 9.0-foot tide level (based on Pacific Coast Tide and Current Tables, Canadian
Hydrographic Service, Department of Mines and Technical Surveys). Sea water
temperatures and samples were taken at each time of collection and _ salinity
determinations were made on these samples. Animals were returned to the
laboratory in canvas buckets containing dampened sea weed.

Habitat

The Spanish Bank area, from which populations of H. oregonensis and H. nudus
are studied, borders the south shore of a relatively protected bay, which extends in
an east-west direction. Extension of this beach forms a point (Point Grey), along
the south side of which the Fraser River flows (Fig. 1). Spanish Bank beach is
a rocky mud-sand intertidal area. The habitat for these crabs is the narrow
restricted upper rocky area, ranging approximately from 3.0-foot tide level to
10.0-foot tide level. The linear distance is about 150 feet. The fauna of this area
is relatively poor, major elements being Mytilus edulis, Balanus glandula and the
two grapsoid crabs. These species abound in numbers. Below this, the beach
extends its gradual slope into a mud-sand flat which continues for several hundred
yards, before dropping to form the channel. This mud flat is exposed on low tides,
and a similar paucity of fauna is due to extremely low summer salinities, a condition
to be discussed later.

Latitudinal distribution of H. oregonensis and H. nudus is very similar. Eco-
logically, however, these two species are quite different. Hemigrapsus nudus is
an open coast intertidal species, whereas H. oregonensis is an intertidal bay and
estuarine species. However, in this geographic area, both species are found
abundantly, occupying essentially the same habitat. Vertically, H. oregonensis
is located lower in the intertidal, and ranges from approximately the 3.0-foot tide
level (lower area of rocks) to the 8.0-foot tide level. H. nudus is higher inter-
tidally, ranging from approximately 6.0-foot tide level to 10.0-foot tide level. The
zone of H. oregonensis is defined much more clearly than that for H. nudus. The
zones as given above suggest nearly comparable width, a condition which existed at
the beginning of the work, 1955-1956. For the past two years, there has been a
relative stability of the H. oregonensis zone, but the zone of H. nudus has shifted
progressively to a lower position, intertidally. The zone of overlap is broadening
and the lower level of H. nudus now is approximately at the 4.0-foot tide level.
At present, due to apparent contour changes of the beach, definition of areas is
much less evident. Individuals of both species characteristically are found under
the same rock, particularly in the area of overlap, a condition which changes some-
what seasonally. The usual niche occupied by H. oregonensis is under rocks,
frequently partially or completely buried in the mud, which contains considerable
amounts of decaying organic matter. This species is found also in beds of Mytilus
edulis. H. nudus is found under rocks, infrequently in mud, and also in Mytilus
beds, but generally occupies a much less sedimented microhabitat.

During the winter and early spring the zone of overlap is reduced and the
areas occupied by the two species are defined more clearly. This zonation appears
to be correlated with breeding activity. The breeding season for H. nudus is from
January to May, and for H. oregonensis, from February to June. At the beginning

RESPIRATION IN CRABS 209

BOWEN 8
ISLAND

BURRARD INLET

STRAIT

OF

GEORGIA

RIVER

DRAINAGE

LULU, oo ISk AND

Ficure 1. Map of the Spanish Bank area where this study was conducted. Note
proximity of the Fraser River drainage which is responsible for seasonal fluctuations in
salinity.

of the breeding season the two species separate spatially. After the females lay
eggs, the zone of overlap is increased and the two species intermingle. Evidence is
available (to be published elsewhere) which demonstrates a degree of interbreeding
between these two species. An intergrade series includes crabs ranging from H.
oregonensis with characters of H. nudus to the opposite extreme. In other regions
of their distribution, ecological conditions such as low salinity seem to serve as
barriers and prevent interbreeding. In this area, however, no such barrier ap-
parently exists.

Temperature-salinity conditions in this area are rather severe, and warrant a

220 PAUL) A sDEHNEE

description. These relations undoubtedly serve to limit, ecologically, invasion by
a more varied fauna. Seasonal variations impose strict limitations with regard to
temperature and salinity tolerance, and only a select group appears to have adequate
regulatory mechanisms to compensate for these changes. In adjacent regions where
the Fraser River does not have an effect, the fauna is more plentiful and varied.
Previously, it was stated that this beach area extends westward and the south-
facing region borders the Fraser River. Volume of this river changes seasonally,
flooding in the spring and summer, due to interior British Columbia runoff. This
volume of water flows into the Strait of Georgia and currents carry the low saline,
low density water mass into adjacent areas and around Point Grey into the collecting

SALINITY & TEMPERATURE : SEASONAL CHANGE

@ PER CENT SEA WATER

© WATER TEMPERATURE

aw
TT) -+
= mi.
=
= vU
m
q DD
WW >
” o
(=
= e!]
z m
WJ
Oo ~~
(ta i]
Ww S
a

MONTHS

FicurE 2. Mean monthly intertidal sea water temperatures and salinities for Spanish
Bank, Vancouver, British Columbia (1955-1959). Solid circles (@) represent sea water
salinity (%), open circles (©) represent sea water temperature (° C).

area. This fresh water influences greatly the intertidal salinity of Spanish Bank,
not only seasonally but during a twenty-four-hour high-low tidal cycle.

During the winter, intertidal sea water temperatures range from 1.0° C. to
6° C. (Fig. 2). Isolated pools at low tide (which occur during the late evening
and early morning) have been recorded as low as —0.5° C. Fraser River runoff
is at a minimum and local salinities vary from 70% to 80% sea water. These
conditions exist from approximately the end of November until the end of February.
Winter conditions with regard to these two parameters are relatively stable ones.
Spring conditions are transient, involving a rise in temperature and a drop in
salinity. Temperature in late February and early March begins to rise, prior to an
appreciable salinity change, from approximately 5° to 8-10° C. Following this,

RESPIRATION IN CRABS Jas

salinity begins to drop to about 50% sea water. Spring conditions merge into summer
conditions, which extend from approximately the end of May to the end of August.
Spring temperatures continue to rise and salinities drop until average summer
intertidal temperatures of approximately 20° C. and summer salinities of 25% to
35% sea water are obtained. Again, for several months relatively stable tem-
perature-salinity conditions exist. During the fall, from approximately August to
November, a transient temperature-salinity period again exists. Fall salinities rise
relatively more rapidly and earlier than the seasonal lowering of temperature.
Stability, low temperature, high salinity, is reached during December and the
seasonal temperature-salinity cycle is completed.

In this geographical region the usual two low, two high tides per twenty-four
hours are interrupted frequently and only one low or one high tide results. This
condition is important conceivably during the summer months. Off shore, a line
of demarcation results between low salinity, low density Fraser River water with
higher salinity, higher density sea water. During a high tide, sea water over the
intertidal region is mixed well, and the usual low summer salinity exists. As the
tide drops, salinity remains low until the incoming tide brings in initially lower
salinity water, and mixing occurs. This tidal fluctuation in salinity might be of
the magnitude of 15% to 45% sea water.

Seasonal rate-temperature experiments

Male crabs of both species, ranging from approximately 0.3 grams to 6.0
grams, were collected for these experiments and were placed in plastic containers
(1014” x 13” x 414") with lids, in approximately 3.5 liters of sea water. Each
container held approximately 30 animals of one species. Lids were left ajar, and
a gauze cloth was placed in the container to help separate the animals. Sea water
salinity approximated field salinity at the time of collection. Containers were
placed in temperature-controlled refrigerators (+ 1.0° C.) for approximately
twenty-four hours, a sufficient time to allow partial clearing of the gut. Holding
temperature approximated field conditions at the time of collection. Crabs were
kept in darkness and not fed. Following the twenty-four-hour holding period,
oxygen consumption was measured at a series of temperatures, 2°, (winter), 3.5°,
Seenamaer jy a°, L0°, 15°, 20°; 25°, 27°; and 30°. C.. In some cases oxygen con-
sumption at two different temperatures was measured in one day. All experiments
were commenced at approximately the same time each day. The following day
another group was measured, these animals having been held under constant
conditions of temperature and salinity for a maximum time of forty-eight hours.
New collections were made to complete the series. For each experiment, meas-
urements were recorded for twenty-four crabs. Crabs were discarded after
measurements were completed for one day. Animals were changed daily with
water of the appropriate salinity and temperature.

Respiration studies on individuals respiring in air can be carried out most
conveniently on H. oregonensis and H. nudus because of the intertidal environment
occupied. In all respiration studies measurements were made with the Wennesland
modification (1951) of the Scholander microrespirometer (1949). Crabs were
placed in darkened chambers with sufficient sea water to keep the gills moist, and
then transferred to a constant temperature water bath (+ 0.1° C.). Equilibration

222 i APAULLANDEERNETL

time varied with the change between holding temperature and experimental water
bath temperature. One hour for thermal equilibration was always allowed, with
the respirometers open to the atmosphere. A further forty-five minutes to one
hour was allowed for each additional 5° C. change in temperature. Following
thermal equilibration, the respirometers were closed and successive readings were
made at ten-minute intervals. Each experiment lasted from one and one-half
hours to two hours, based on the time required for respiration levels to become
linear with time. Following the experimental period crabs were removed from
the respirometer chambers, dried with gauze and weighed to the nearest 0.1 gram.
Data on individuals were discarded if the animals showed activity, or if high or low
respiratory rates were obtained, due to approaching death. Respiration rates were
given as cubic millimeters of oxygen consumed per gram per hour (weight-specific
oxygen consumption). Data for each experimental temperature were plotted as
weight-specific oxygen consumption as a function of weight on log-log paper.
Regression lines were fitted by the method of least squares. Respiratory rate for
a given weight animal was plotted against temperature on semi-log paper to demon-
strate the rate-temperature relationship.

Temperature and salimty experiments

Methods of collecting and holding were similar to those described for rate-
temperature experiments. For the temperature acclimation experiments, four tem-
peratures: were’ chosen,:5°,; 10°; 15° and) 20% 6. (1.0° ©:)! ~Anmoals gyeherae—
climated for at least one week, maintained under dark conditions and not fed.
Experiments were arranged in such a fashion that a complete series of measurements
was obtained on one species prior to commencing the other. Experiments were
conducted during the late spring and summer months. In one series of experi-
ments, at four acclimation temperatures, a constant salinity of 75% sea water was
used, which corresponded to winter field conditions. In another series, a constant
salinity of 25% sea water was used, which corresponded to summer field conditions.
At the end of one, two and in some cases three weeks, oxygen consumption meas-
urements were made on groups of crabs acclimated to each of the four temperatures
and a constant salinity. Experimental water bath temperatures used were 10°
and 20° C. (+0.1° C.). Respiratory measurements were determined as men-
tioned previously. After the 10° C. experimental temperature was completed, the
water bath was raised to 20° C. (time required approximately one hour) and the
animals were allowed one and one-half hours thermal equilibration. Following
completion of the experiment and weighing, animals were returned to the experi-
mental temperature and salinity conditions. Results were not different when
animals were measured initially at 20° C. and then lowered to 10° C. Respiration
rates were given as cubic millimeters of oxygen consumed per gram per hour
(weight-specific oxygen consumption). Data for each acclimation temperature-
salinity combination at each experimental temperature were plotted as weight-
specific oxygen consumption as a function of weight on log-log paper. Regression
lines were fitted by the method of least squares.

All field and experimental salinities are expressed as percentage sea water,
based on a standard sea water, 31.88%o salinity, 17.65%o chlorinity at 25° C., as
100% sea water. Sea water concentrations were prepared from local sea water,
away from the influence of Fraser River runoff, which ranged from approximately

RESPIRATION IN CRABS 223

90% sea water in the winter to 65% in the summer. For concentrations below
normal sea water, dechlorinated fresh water was added until the desired dilution
was obtained. For concentrations higher than normal sea water, sea salt was
added, based on 31.88 grams per liter of sodium chloride. Experimental salinities
were determined on a 1000-cycle conductivity bridge calibrated to the standard
sea water noted above.

METHOD oF ANALYSIS

As stated previously rate of oxygen consumption was plotted as a function of
body weight (weight-specific) on a double logarithmic system. Such a plot, over
an adequate weight range, gives a straight line with a negative slope. Regression
of weight-specific oxygen consumption against body weight assumes the form:

or:
log O2. — log W = loga + blog W — log W

where O, is oxygen consumption of the crab in mm.? O,/gram/hour, W is body
weight in grams, a the intercept and b the slope of the line. Negative linear regres-
sion coefficients were calculated by the method of least squares. It was necessary
to determine whether statistically significant differences existed between regression
lines representing metabolic rates of crabs measured at different combinations of
Becumiadton temperatures (5°, 10°, 15° and-20° C.), and. salinities (25% and
75% sea water) at the two experimental temperatures (10° and 20° C.). Each
calculated regression line was based on oxygen consumption measurements of at
least thirty-five crabs and in most cases fifty to sixty. These data are suited to
statistical treatment by analysis of covariance, a standardized method outlined by
Ostle (1954) for a randomized sample. The method was to test by analysis of
covariance the null hypothesis that no true differences existed in the effect of
different combinations of temperature and salinity on oxygen consumption, 1.¢.,
that two regression lines could be represented by a single regression line. In the
analysis, negative logarithm values were eliminated by multiplying oxygen con-
sumption rates by 10 before converting to four place common logarithms.

Zeuthen (1953) has suggested that the terms “respiratory rate,’ “metabolic
rate” and “rate of oxygen uptake” be defined as “oxygen uptake per hour per unit
of body size.” This procedure was observed here. Further, “weight-specific
oxygen consumption” connotes the same meaning. Use of the word “acclimation”
is preferred to “acclimatization” following the suggestion of Bullock (1955; see also
Prosser, 1955). This word refers to intra- and interspecific compensatory changes
whether these changes be phenotypic or genotypic. Other descriptive words such
as “regulation,” “compensation” and “homeostasis” are used with no other im-
plications than stated above.

RESULTS
Seasonal rate-temperature experiments

Results of winter and summer oxygen consumption at various experimental
temperatures, for crabs removed directly from field conditions, are shown in
Figure 3 (H. oregonensis) and Figure 4 (H. nudus). Crabs of both species with
a weight of 2.0 grams have been chosen arbitrarily to depict these data.

224 PAUL, AC DEHNEL

Henugrapsus oregonensis: Respiratory metabolism for H. oregonensis con-
sistently is higher for summer animals than for winter ones. These data demon-
strate the fact that there is no acclimation of oxygen consumption to temperature
by this species under field conditions. Shape of the winter and summer curves is
similar in the physiological temperature range. Wéunter animals show no cold
depression at 2° C. and some heat depression occurs between 25° and 27° C.
Summer crabs are depressed greatly at 3.5° C., O,, value between 3.5° and 5° C.
is 836.0, and heat depression is above 27° C. Depression between 27° and 30° C.
is somewhat less for winter crabs. Physiological temperature range for summer

HEMIGRAPSUS OREGONENSIS
20 gram animals

O eee 9-4
Su
gee ae ‘ eo Summer
O ti
®@
ep Winter

oO

O2 /GRAM/ HOUR

MM?

Za5;0 fe) IS 20 2D2fii 30
TEMPERATURE °C

Ficure 3. Seasonal rate-temperature curves, acutely measured, for 2.0-gram Hemi-
grapsus oregonensis. Respiratory rates for this weight crab were chosen from regression lines
determined for weight-specific oxygen consumption data at each temperature. Lines were
fitted by the method of least squares. Summer animals (©) were kept in 25% sea water and
winter animals (@) in 75% sea water, prior to and during the experimental period.

crabs is about, 5° C., to. 27°. C.,. whereas that range for winter crabs\is 27 5@@i%o
25°-27° C.

Hemigrapsus nudus: In Figure 4 approximately the same conditions as noted
for H. oregonensis are presented for H. nudus. Summer-adapted H. nudus con-
sistently are higher than winter ones, showing no respiratory acclimation to tem-
perature under field conditions. Shape of the two curves is somewhat similar
from 10° C. to 25° C. Consideration of the ends of the curves allows further
discussion. Winter-adapted H. nudus show no cold depression at 2.0° C. whereas
summer-adapted H. nudus show considerable depression below 5° C. The Q,,
value between 3.5° and 5° C. is 166.5. At the higher end of the temperature

RESPIRATION IN CRABS 225

range, winter-adapted H. nudus show some heat depression between 25° and 27° C.
Summer-adapted H. nudus likewise are depressed between 25° and 27° C. De-
pression between 25° and 30° C. is considerably less for winter than for summer
animals. Physiological temperature range for summer crabs is about 5° to 25° C,,
whereas that range for winter crabs is 2° C. to 25°-27° C,

Interspecific comparison: Seasonal comparison shows that summer H. ore-
gonensis have a slightly higher absolute oxygen consumption below 15° C. for
any given weight and temperature, than does summer H. nudus for that same weight
and temperature; 20° C. and above, summer H. nudus have a higher absolute
oxygen consumption. Approximately the same rate is found at 15° C. Winter

I50

HEMIGRAPSUS NUDUS
20 gram animals

“100

70

3

Winter

MM? O2/GRAM/HOUR
8

20

Yaaro 00) lO IS 20 eo cr FSO
TEMPERATURE °C

Ficure 4. Seasonal rate-temperature curves, acutely measured for 2.0-gram Hemi-
grapsus nudus. Respiratory rates for this weight crab were chosen from regression lines
determined for weight-specific oxygen consumption data at each temperature. Lines were
fitted by the method of least squares. Summer animals (©) were kept in 25% sea water
and winter animals (@) in 75% sea water, prior to and during the experimental period.

crabs of both species have about the same rate of metabolism at 15° and 20° C.
for any given weight. At other temperatures, the rates for H. oregonensis are
higher.

Both of these species fail to show a shift on the ordinate, winter versus summer
(an acclimation shift), but neither do they show any appreciable difference at the
high end of the temperature range. It would be expected that summer animals
conceivably would show depression of oxygen consumption at a higher temperature
than winter crabs. Actually, winter H. nudus extend their upper limits to about
the same level as summer ones. Summer H. oregonensis has a slightly higher

226 PAUL AY DEHINED

limit than do winter crabs. If the summer curves for both species are considered
as baselines, then it can be argued that winter animals are showing an acclimation
of the upper physiological limit. They do not show the drop to a lower tempera-
ture associated with cold water living. Correspondingly, winter animals have
extended their curves to the left (lower temperature before depression) and
winter animals have a somewhat wider physiological temperature range of oxygen
consumption (2.0° C. to 25°-27° C., both species) than summer animals (5° C.
to 27° C., H. oregonensis, 5°-25° C., H. nudus). Absolute temperature range
for summer H. nudus is 20° C. and summer H. oregonensis is 22° C., whereas that
same range for both species of winter crabs is about 23°-25° C. Such a condition
is the reverse to that found by Segal (1956): for seasonal variations in heart beat of
Acmaea limatula, and for growth rates of latitudinally displaced populations of
gastropod larvae reported by Dehnel (1955).

Another method which was used for comparison between two curves is Q,),
provided limitations are recognized and accepted. Over physiological ranges of
temperature, Q,, values express variations in temperature sensitivity of oxygen
consumption under different thermal conditions, if intervals between temperatures
at which Q,, values are determined are relatively small. The Q,9 values determined
for cold- or heat-depressed regions of the rate-temperature curves, for instance,
would have no biological meaning unless the depression were shown to be fully
reversible or to occur under normally encountered conditions. Evidence obtained
from temperature tolerance experiments (Todd and Dehnel, 1960) shows that
heat depression is reversible in both species, winter and summer, with respect to
50% mortality levels at much higher temperatures than reported here.

Comparison of Q,, values, winter versus summer, shows little consistency.
Between 10° and 20° C. summer H. nudus have considerably higher Q,, values
than winter ones (summer, 10°-15° C., Oj}, =28; winter, \O,, =2.5; summer
15°-20°° C.; O14 = 213; winter, O,, = 1.6). At’ other parts\oi the temperaeime
range, winter crabs have higher values. Winter H. oregonensis between 10° and
25° C. have higher Q,, values (winter), 10°=15° C.,O,, = 1.5; summer, O75 feo
winter, 15°-20°-C,,.0,6 = 1.5;summer ;©,¢-= LS swinter;20°-252-€) © ae
summer, Q,, = 1.2). At other parts of the temperature ranges summer crabs have
higher values.

Scrutiny of Q,, values for both species, winter and summer, shows many values
to be much lower than the generally accepted value for poikilotherms, 2.0 to 3.0.
Lower values are seen fairly consistently over physiological ranges of temperature,
particularly for H. oregonensis, winter and summer.

Effect of acclimation temperature

Hemigrapsus oregonensis: When the four acclimation temperatures (5°, 10°,
15° and 20° C.) are compared at either low (25% sea water) or high (75% sea
water) salinity at 10° C. experimental temperature, it is seen that as acclimation
temperature increases oxygen consumption decreases over most of the weight
range (Figs. 5 and 6). Analysis of covariance of the total of the four regression
lines at either salinity, with respect to their position on the ordinate, gives a P
value of less than one per cent (Table 1). In the case of H. oregonensis (Fig. 5)
when 1.0-gram crabs are compared at 5° C. and 20° C. and a salinity of 25%

RESPIRATION IN CRABS 227

sea water, there is an 87% increase at the lower acclimation temperature. The
difference between these two regression lines is statistically significant (P = 0.01).
Comparisons of the 5° C. acclimation temperature with the two intermediate tem-
peratures similarly show significant differences at the one per cent level of

TABLE: |

Analysis of covariance of rates of oxygen consumption per gram per hour as a function of weight
in Hemigrapsus oregonensis and H. nudus at 10° C. experimental temperature, at two salinities (25%
and 75% sea water) and four acclimation temperatures (5°, 10°, 15° and 20° C.). Combinations
of acclimation temperatures are compared at each salinity. P indicates the significance of the position
of the regression lines on the ordinate. P» indicates the significance of the change in slope of the
regression lines. 6 1s the regression coefficient. yr indicates coefficient of correlation

Accl. sal. (%) Oe ieee ge P Pp Accl. ED: (G2)
Total 0.01 0.01 5° C. —.678 — .8641
5 and 10 0.01 N.S
5 and 15 0.01 NES: 10 —.685 —.8205
25 5 and 20 0.01 0.01
10 and 15 ak Nest 1S) —0G0 =~ '8522
10 and 20 0.05 0.01
15 and 20 N.S 0.01 20 eee — .8081
HI. oregonensis
Total 0.01 N.S SES 2589 — .9345
5 and 10 0.01 Nes:
5 and 15 0.01 0.01 10 = EN — 6216
fe) 5 and 20 0.01 Nes:
10 and 15 N.S 0.05 15 = Gl =a 3 /3Sn8
10 and 20 N.S NES
15 and 20 N.S N.S: 20 S28) —.5208
Total 0.01 N.S 5) OF ey! — .8296
5 and 10 0.01 INES
5 and 15 0.01 N.S 10 —.517 —.8525
25 5 and 20 0.01 NCS:
10 and 15 NaS: INS: iS —.578 — .8793
10 and 20 0.01 NES
15 and 20 0.01 N.S 20 = brs) ao
HI. nudus
Total 0.01 0.05 5° C. —.500 — .7598
5 and 10 N.S. N.S
5 and 15 0.05 NS. 10 —.441 —.8518
75 5 and 20 0.01 0.05
10 and 15 0.05 N.S. fs) —.459 — .7022
10 and 20 0.01 0.01
15 and 20 0.01 0.05 20 — Oot —.9419
probability. Further, statistical comparisons of acclimation temperatures in various

combinations are given in Table I.

At the lower salinity (Fig. 5) it is seen that the regression lines, with the
exception of the 5° C. acclimation temperature, converge toward the higher end
of the weight range, at approximately 2.0 to 3.0 grams.
shows a Statistically significant difference (P = 0.01) in change of slope.

Analysis of the four lines

If the

228 PAULTAS DEEN

200

HEMIGRAPSUS OREGONENSIS
Expt ‘temp: . {076
Sea water 75%

HEMIGRAPSUS OREGONENSIS
200 5° ® Expt temp 10°C
Sea water 25%,

Oo /GRAM/ HOUR

MM>

30

20

[OL ,
Ol O2 = 03 05, O07 218 ZOKU) 90" Oe

BODY WEIGHT IN GRAMS

Ficure 5. Effect of acclimation temperature on weight-specific oxygen consumption with
increasing size in Hemigrapsus oregonensis, at the two acclimation salinities, 25% sea water
(lower) and 75% sea water (upper). Each point represents an animal. Points for 5° C.
(@) and 20° C. (CO) are included to demonstrate variation. Regression lines were fitted by
the method of least squares. Slope values, coefficients of correlation and other statistical
data are given in Table I.

RESPIRATION IN CRABS 229

four acclimation temperatures are compared, as these temperatures increase the
slopes of the regression lines are nearly parallel, whereas the 20° C. line is signifi-
Gally cirerent (P = 0.01) from the other three (Table 1). If the rate of oxygen
consumption of a 2.5-gram crab is compared at 5° C. and 20° C., the increase is
47% at the lower temperature. This same difference exists for the two inter-
mediate acclimation temperatures when either is compared with the 5° C. regres-
sion line. This increase is approximately one-half that determined for a 1.0-gram
animal at the low and high acclimation temperatures.

At the higher salinity, 75% sea water, H. oregonensis shows a similar trend to
that discussed for the lower salinity (Fig. 5). Comparison of 1.0-gram animals at
low and high acclimation temperatures shows an 83% increase in weight-specific
oxygen consumption at the lower temperature. Further comparison of the 5° C.
line with the two intermediate acclimation temperatures results in levels of prob-
bility of the same magnitude but less percentage differences (Table I). Com-
binations of intermediate acclimation temperatures within themselves or with high
and low temperatures result in the same degrees of significance or insignificance as
noted for the response to low salinity in this species.

Again as with the low salinity, slopes (b values) of the four acclimation tem-
peratures decrease as those temperatures increase. Comparing the change of
slope of the four acclimation temperature regression lines, analysis shows that there
is no significant difference (Table I). Specifically, regression lines of the two
lower temperatures are nearly parallel, as is the case for the two higher acclimation
temperatures, and no significant differences are found to exist within these two
pairs. However, comparison of only the 5° C. line with the 15° C. one gives a
Beaistically significant difference (P = 0.01). There is a tendency for the 10°,
15° and 20° C. acclimation temperature lines to converge at the 2.0- to 3.0-gram
weight. This is the same pattern as found at the lower salinity. Again, if a 2.5-
gram crab is compared at 5° C. and 20° C. acclimation temperatures, there is found
a 48% increase in oxygen consumption at the low temperature. This is ap-
proximately one-half that determined for a 1.0-gram crab.

Data are available but have not been presented for the experimental temperature,
20° C. If any of the above comparisons are made at this temperature, the same
trends are observed. They differ only in the fact that regression lines for the
higher experimental temperature are located at a higher position on the ordinate.
At this higher experimental temperature there is no evidence of heat depression, at
either experimental salinity.

Henugrapsus nudus: Comparison of the four acclimation temperatures at low
salinity for H. nudus demonstrates that as acclimation temperature increases,
oxygen consumption decreases (Fig. 6). There is observed a 105% increase for
1.0-gram crabs at the low temperature (5° C.) when this is compared with the high
one (20° C.). The only combination of two lines which is not statistically signifi-
cant is the comparison of the 10° and 15° C. acclimation temperature regression
lines (Table 1). All other combinations give a P value of less than one per cent.

When the slopes of the four acclimation regression lines are compared it is
seen that there are no statistically significant differences (Table I). The four
lines assume nearly parallel relations.

At a salinity of 75% sea water response of these crabs at the four acclimation
temperatures is similar when compared with the lower salinity. As the acclima-

230 PAUL A. DEHNEL

200)
| HEMIGRAPSUS NUDUS
| Expt. temp. 10°C
100 | Oh Sea water 75%
70]
50}
30 |
or
= 20
O
56
SS
=
et. tO)
oc
2 200 HEMIGRAPSUS NUDUS
S Expt. temp. 10°C
s. Sea water 25%
=
xs |00}

OlyrnynQern Osis: (Oreste ZO WSO 50° FOr
BODY WEIGHT IN GRAMS —

Ficure 6. Effect of acclimation temperature on weight-specific oxygen consumption with
increasing size in Hemigrapsus nudus, at the two acclimation salinities, 25% sea water (lower )
and 75% sea water (upper). Regression lines were fitted by the method of least squares.
Slope values, coefficients of correlation and other statistical data are given in Table LI.

RESPIRATION IN CRABS é ZS

tion temperature increases, oxygen consumption again decreases over most of
the weight range. Analysis of covariance of the total of the four lines shows a
significant difference (P = 0.01). If a 1.0-gram crab is compared at high and
low acclimation temperatures there is a 28% increase in oxygen consumption at
the lower temperature. Further comparison of the 20° C. regression line with
either the 10° C. or 15° C. line gives the same level of significance (Table I). If
the 5°, 10° or 15° C. regression lines are compared within themselves, either
there is no significance or a significance to the 5% level, a value considered to be
statistically significant.

With regard to slope change at the higher salinity, the four acclimation tem-
perature regression lines converge at the small end of the weight range (Fig. 6).
Total line comparison for change in slope gives a significance to the 5 per cent |
level. This is true also for all combinations of the four regression lines with the
Eeeemweno. 10> and 20° Co where P= 0.01’ There is an approximate 67%
increase in weight-specific oxygen consumption for a 2.5-gram crab at 5° C. over
that noted for the same weight at 20° C. This percentage increase is more than
twice that observed for a 1.0-gram animal, when the same two acclimation tem-
peratures are compared.

As in the case of H. oregonensis data are available but not presented for 20° C.
experimental temperature for H. nudus. Again, the trends are the same, only the
position on the ordinate differs.

Interspecific comparison: If the above results as determined by the four tem-
perature acclimation regression lines are compared between species, at either
salinity, it is observed that weight-specific oxygen consumption is similar in some
instances. This is the situation at both experimental temperatures. For instance,
at 25% sea water, 5° C. acclimation temperature a 1.0-gram H. oregonensis has
an oxygen consumption per gram per hour of 87 mm.3, and the same weight
H. nudus with identical conditions has a value of 84 mm.? (3.5% difference).
At 20° C. acclimation temperature, H. oregonensis has a value of 47 mm. and
H. nudus, 41 mm? (15% difference). At 75% sea water, 5° C. acclimation
temperature 1.0-gram H. oregonensis and H. nudus are about the same. At 20°
C. there is a 35% difference, H. nudus having the higher rate (compare Figs.
5 and 6).

Effect of acclimation salinity

Hemigrapsus oregonensis: When weight-specific oxygen consumption is com-
pared at low and high salinities at 10° C. experimental temperature and a given
acclimation temperature, the higher rate is found at the low salinity (25%).
Analysis of covariance of the total of the four acclimation temperature regression
lines in relation to the two salinities shows a statistically significant difference
(P = 0.01) between low and high salinity (Table II). At the 5° C. acclimation
temperature (Fig. 7) a 1.0-gram crab shows a 29% increase at the low salinity
over that for the same weight animal at the high salinity. Similarly, at the 20°
C. acclimation temperature there is a 27% increase at the low salinity. In Figure
7 as well as Figure 8 only low and high acclimation temperatures with both salini-
ties have been given. If either of the two intermediate acclimation temperatures

232 PAUL AYDEHNEL

(10° and 15° C.) are compared at both salinities there is no significant difference
CRable iy:

There is a convergence of regression lines of each acclimation temperature at
both salinities toward the higher end of the weight range. However, statistical
comparison of the total acclimation temperature regression lines at the two salinities
or comparison of any regression line, likewise at the two salinities, shows no
significance with regard to change in slope (Table II). There is a tendency for
the slope to decrease at the higher salinity for either the 5° or 20° C. acclimation
temperature regression line.

TABLE II

Analysis of covariance of rates of oxygen consumption per gram per hour as a function of weight
in Hemigrapsus oregonensis and H. nudus at 10° and 20° C. experimental temperatures. Each
acclimation temperature 1s compared at the two salinities, 25% and 75% sea water. P indicates the

significance of the position of the regression lines on the ordinate. P» indicates the significance of the
change in slope of the regression lines

Expt. temp. (°C.) Accl. temp. (°C.) IP Po
diotall 0.01 N.S
5 0.01 N.S:

10 10 N.S. INES:

15 0.05 N.S.
20 0.05 NES:

H. oregonensis

Total INES: N.S.
20 5 NS. N.S.
10 N.S. INES:
15 NGS: NES!
20 N.S Nes:
otal 0.01 INES:

5 0.01 INES:
10 10 0.01 INES!
15 0.01 N.S.

20 0.01 0.05

H. nudus Total INES: N.S
5 NES: N.S

20 10 INES: N.S

15 N.S NeS

20 N.S 0.05

At the higher experimental temperature (20° C.) when any of the four ac-
climation temperature regression lines are compared at both salinities (5° C. at
25% and 75% sea water) there are no statistically significant differences (Table
II). Similarly, there are no significant slope changes under these experimental
conditions.

Hemigrapsus nudus: Consideration of the data for this species is quite different
from that described for H. oregonensis. Weight-specific oxygen consumption
measured at 10° C. experimental temperature is higher when crabs are acclimated
to the low temperature, low salinity combination when compared with low tem-
perature, high salinity (Fig. 8). Covariant analysis of the total four acclimation

RESPIRATION IN CRABS 233

temperature regression lines relative to the two salinities gives a significant differ-
ence (P =0.01). A 1.0-gram crab at 25% sea water shows a 31% increase over
that same weight crab at 75% sea water. The lines from which this weight crab
was taken are significant at the one per cent level (Table II). At the higher ac-
climation temperatures (Fig. 8 and Table II) crabs acclimated to the higher
salinity have a higher rate of oxygen consumption and the differences at each
acclimation temperature for both salinities are statistically significant (P = 0.01).
Comparison of a 1.0-gram crab acclimated to 20° C. and both salinities shows a
25% difference in rate of oxygen consumption.

HEMIGRAPSUS OREGONENSIS

200 Sea eater Expt. temp. 10°C
3 Sea water
x 75%
O
+ 100
Ss
Sea wat
= £0 25%
(E
Sea water
= 50 75%
N
O
m 30
a
=
20

Ol O2 103 QS Of 10 20) 450) 30), £0) 10
BODY WEIGHT IN GRAMS

Ficure 7. Effect of acclimation salinity on weight-specific oxygen consumption with
increasing size in Hemigrapsus oregonensis at two acclimation temperatures (5° and 20° C.).
Regression lines were fitted by the method of least squares. Slope values are given in Table JI,
significance of slope change and other data are given in Table I]. The two intermediate ac-
climation temperatures (10° and 15° C.) are not included but the analyses are given in the
tables.

Reference to Figure 8 and Table II indicates no significant changes in slope
either for a given acclimation temperature regression line compared at two salinities
or total regression lines.

The response of this species at the 20° C. experimental temperature shows
no statistically significant differences when any acclimation regression line is
compared at the two salinities, or when slopes are compared (Table IT).

Interspecific comparison: When these two species are compared with regard to
their acclimation response to low and high salinity, one basic difference is noted.
When both species are acclimated to low temperature, low salinity and low tem-
perature, high salinity the weight-specific oxygen consumption regression lines

234 PAUL A. DEHNEL

HEMIGRAPSUS NUDUS

200 Expt. temp. 10°C
Seo water

25%

a Seo water

ra : ee
eq water

=z !00 75%

a

> Sea water

= 70 25%

©

C50

w

O

MMs

OP) 2) OS pers Oi, eto 20 "930 gn Te
BODY WEIGHT IN GRAMS

Ficure 8. Effect of acclimation salinity on weight-specific oxygen consumption with
increasing size in Hemigrapsus nudus at two acclimation temperatures (5° and 20° C.).
Regression lines were fitted by the method of least squares. Slope values are given in Table I,
significance of slope change and other data are given in Table II]. The two intermediate
acclimation temperatures (10° and 15° C.) are not included but the analyses are given in the
tables.

are similar in position and slope. Low temperature, low salinity combination gives
the greater rate. If, however, crabs are acclimated to high temperature at both
low and high salinities, H. nudus has the higher rate at the high salinity combination
and H. oregonensis at the low salinity combination. Percentage differences of a
1.0-gram animal for both species are approximately the same, 25%.

Effect of size

Hemigrapsus oregonensis: Reference to Figure 5 and Table I shows that there
is a Statistically significant change (P =0.01) in slope (decrease) as the ac-
climation temperature increases, when crabs are measured at 10° C. experimental
temperature, and maintained at the low acclimation salinity. The fact that the
four regression lines converge at the higher end of the weight range shows that
weight-specific oxygen consumption of small animals, except at the low temperature,
is affected by increase of acclimation temperature to a greater degree than large
ones. With increasing weight there is a smaller change in rate for 20° C. ac-
climated animals than for 5° C. ones. Thus, animals acclimated to high tem-
peratures are less size-dependent. The Q,, values increase as weight increases
over the range from 5° to 10° C. Comparison of a small crab (0.8-gram) with
a large one (3.0-gram) shows a Q,, change from 1.9 to 2.4. The Q,, decreases
with weight over temperatures from 10° to 20° C.

RESPIRATION IN CRABS 235

At the high salinity, analysis of covariance of the four acclimation temperature
regression lines shows no Statistical significance relative to slope change. As a
result no real size effect is demonstrable (Fig. 5). The Q,, values at the 5° to
10° C. temperature range increase as weight increases; Q,, of 0.8-gram crab is
1.6, of a 3.0-gram crab, 1.9. An inverse Q,, relationship with weight or no change
is noted at the two other temperature comparisons.

When the two acclimation salinities are compared at each acclimation tem-
perature, relative to slope change no statistically significant differences are found
(Figs. 7, 9 and Table II). Figure 9 compares different weight animals at the
two salinities and over the range of acclimation temperatures. Small crabs (0.8-
gram) and large ones (3.0-gram) have been chosen from the regression lines in
Figure 5. If the per cent change (increase) in the rate (oxygen consump-
tion/gram/hour) is determined for the two salinities at each acclimation tem-
perature the following values obtain. For small crabs at 5° C. there is a 30%
Saepee eerate; 10° C:) 16%; 15° C., 46% and 20°C, 23%. For large crabs
foe) -eam) the’ per cent change (increase) is less:95° C.,. 17%; 10° C., 7%;
15° C., 4%; 20° C., 20%. It is noted that for each acclimation temperature the

200 HEMIGRAPSUS OREGONENSIS
Expt. temp. 10°C

Sea water e@
25 %

re)
fe)

Sea water ©
75%

~J
O

~ 0.8 gm.

25% O animals

Sea water O
T5%

30 30 gm.

O animals

Sea water @

MM° 0,/GRAM / HOUR
Oo
O

lO
O 3. 10) IS 20

ACCLIMATION TEMPERATURE °C

Figure 9. Effect of size on weight-specific oxygen consumption in Hemigrapsus
oregonensis at the two acclimation salinities (25% and 75% sea water) and over the range
of acclimation temperatures (5°, 10°, 15° and 20° C.). These weights were chosen from
regression lines fitted by the method of least squares for data shown in Figure 5. The
statistics for the size effect are given in Tables I and II.

236. PAUL A. DEHNEL

rate for the low acclimation salinity is higher generally than that recorded for the
high salinity.

Hemigrapsus nudus: For this species, at the low salinity, comparison of the
total four acclimation temperature regression lines shows no significant slope change
(Fig. 6 and Table 1). Comparing Q,, values at the temperature interval of 5° to
10° C. shows no increase in Q,, as weight increases; Q,, of an 0.8-gram animal is
2.2; 3.0-gram crab, 2.3. Over the range from 10° to 15° C., there is only a slight

200 HEMIGRAPSUS NUDUS

Expt. temp. 10°C

a lOO Sea water e
S 25 %
<= Sea water
70
>
=
§ 8 ,
~ 50 0.8 gm
Oo Sea water @ animals
25 %
SS
OS Sea water
WS) Yo
PSO)
2 3.0
: i m.
= fe) 9
20 animals

lO

O 5 lO IS 20
ACCLIMATION TEMPERATURE °C

Ficure 10. Effect of size on weight-specific oxygen consumption in Hemigrapsus
nudus at the two acclimation salinities (25% and 75% sea water) and over the range of
acclimation temperatures (5°, 10°, 15° and 20° C.). These weights were chosen from
regression lines fitted by the method of least squares for data shown in Figure 6. The
statistics for the size effect are given in Tables I and II.

change, 1.1 to 1.3, as there is observed over the range from 15° to 20°°C), amman
small crabs to 1.6 for large ones.

Covariant analysis of total four acclimation temperature regression lines at the
high salinity gives a significance in slope change at the five per cent level (Fig. 6
and Table 1). In this instance, however, the four regression lines converge to the
left (small weights) and this demonstrates that large animals are affected to a
greater extent as acclimation temperature increases. With increasing weight there
is) a greater’ change, in rate for ‘20°C. acelumated animals than ior 9 -/C. ones
Animals acclimated to these higher temperatures are more size-dependent. If

RESPIRATION IN CRABS 237

O,, values are compared over the range from 5° to 15° C. there is no increase with
Weenie) Over/the range from 15° to 20° C., the increase is direct with size, 1.3
(0.8-gram) to 2.2 (3.0-gram).

No statistically significant differences in slope exist when weight-specific oxygen
consumption for total lines is compared for both salinities at each acclimation
temperature (Figs. 8, 10 and Table II). It is noted, however, that a five per cent
level of significance results when rate of oxygen consumption is compared at 20° C.
acclimation temperature. Again as with H. oregonensis small (0.8-gram) and
large (3.0-gram) crabs have been chosen from the regression lines in Figure 6 and
presented in Figure 10 for comparison. The per cent change (increase) in the
weight-specific oxygen consumption results in the following values. For small
Beet Ce tbere isa 31% change in rate; 10° C., 10%; 15° C., 8% and 20° C,,
etveeeoen larce crabs the per cent change is; 5° C, 28%; 10° C., 19%; 15° C,,
Ze 7erand 20° C., 9%.

Interspecific comparison: It can be stated generally that increase in acclimation
temperature affects size differentially at low salinity in H. oregonensis, whereas
this effect is not noted in H. nudus. At the higher salinity H. nudus shows a
differential size effect, whereas H. oregonensis does not. When the two salinities
are compared at any acclimation temperature no size effect is noted for either
species. If O,, values are compared, there is a general increase with weight in-
crease in both species at the low salinity over the range 5° to 10° C. At the two
higher temperature intervals generally there is a decrease in Q,, with size or
essentially no change. At the higher salinity, over the temperature range 5° to
10° C., O,, values for H. oregonensis increase with size, whereas there is no change
for H. nudus. For the two higher temperature ranges Q,, values either decrease
or remain the same for H. oregonensis, and increase or remain the same for H.
nudus. For any weight crab at either salinity, Q,, values tend to decrease as
temperature increases. An exception to this is recorded for H. nudus at high
salinity where Q,, increases as the acclimation temperature increases.

Reference to Figures 9 and 10 shows that at the low acclimation temperature
rate of oxygen consumption for large and small animals of both species is always
higher for ones acclimated to low salinity. As acclimation temperature increases,
low salinity continues to result in a higher rate for H. oregonensis, but for H.
nudus higher salinity has a greater effect.

DISCUSSION
Seasonal rate-temperature experiments

Oxygen consumption measurements made on both species of crabs over a
range of temperatures, summer and winter, and kept at the seasonal salinity, show
that summer crabs, for any weight animal, have a higher respiratory rate than
winter ones over the physiological temperature range. Further, heat depression
occurs for winter animals at approximately the same high temperature as for
summer crabs, and winter animals extend the range of low temperature to a point
below that for summer ones. These responses result in a greater temperature
range for winter animals before temperature depression occurs. The data relate
to a condition described by Precht (1951) as inverse compensation (type 5). The

238 PAUL A. DEHNEL

seasonal rate-temperature curves do not correspond to the frequently reported
relationships, winter rates higher than summer (partial compensation, type 3).

Hemigrapsus oregonensis: Use of the acclimated rate-temperature curve has
been suggested as a method for comparing species ecologically, and for determining
whether acclimation to low temperatures shows proportionally a greater compensa-
tion when compared with high temperatures (Bullock, 1955). An acclimated
rate-temperature curve can be plotted on Figure 3. This curve is determined by
noting weight-specific oxygen consumption for winter animals at 5° C. (30 mm.®
O,/gram/hour) with summer ones at 20° C. (72 mm.*). These two points
represent animals seasonally adapted to their own approximate natural tempera-
tures and measured at those temperatures. This acclimated rate-temperature
curve for H. oregonensis has a higher Q,, than either the summer or winter
acutely measured curves. This means that animals show a greater temperature
dependence when adapted to their field temperatures than when measured acutely
over a series of temperatures, and before any demonstrable acclimation occurs.
Bullock (1955) states that the acutely measured curve is steeper than the acclimated
one (more temperature-sensitive). The curves for this species show this not to be
the case, and if the acclimated rate-temperature curve is significant, then O,, may
have no real meaning.

A further point of comparison is weight-specific oxygen consumption for sum-
mer animals at 5° C. and 30° C., and the same comparison for winter crabs at these
two temperatures. Summer animals have the same rate at the low and high
temperatures, and winter ones are similar. If summer and winter animals are
compared together at either the low or high temperature, again their rates are
nearly the same. Summer and winter animals are depressed about the same at the
high temperature and no depression occurs at 5° C. These data, coupled with the
acclimated rate-temperature curves, are good evidence to show that no acclimation
has resulted.

Hemigrapsus nudus: When an acclimated rate-temperature curve is plotted
for: this ;species)»(Fig.: 4), !the | rate) for winter) animals 2ati57 ©. asZz0iname.
O,/gram/hour, and for summer animals at 20° C., 83 mm.*. This curve is steeper
(higher Q,,) than the winter acutely measured rate-temperature curve, and is
approximately the same as the summer curve. Thus, winter animals show a
greater temperature dependence when adapted to their field temperatures. Tem-
perature dependence for summer animals is about the same for acclimated and
acutely measured rates.

When weight-specific oxygen consumption is compared for summer or winter
animals at 5° and 30° C., both have a higher rate at the high temperature. But,
if summer and winter crabs are compared at 30° C., both have the same rate, the
depression being similar. Again this evidence shows that no seasonal acclima-
tion has been found.

Metabolism and size

Relationship of body size to metabolism is not only recognized but is invariably
the subject of controversy when this dependence is discussed in various animal
groups. The general concept states that weight-specific oxygen consumption is
higher for small animals when compared with large ones, measured at a given tem-

RESPIRATION IN CRABS 239

perature and determined for animals of a given species or for closely related ones.
If the logarithm of the rate is plotted as a function of the logarithm of weight a
linear relation exists, with a negative regression coefficient. If total oxygen
consumption is plotted against body weight, large animals have a higher metabolic
rate. These relationships have been shown to exist for poikilotherms and
homiotherms. Absolute changes occur when intra- and interspecific compari-
sons are made relative to (1) temperature, (2) displacement of populations lati-
tudinally and vertically, (3) season, (4) sex, (5) state of nutrition, (6) age,
(7) other extrinsic and intrinsic factors. Relative changes remain, however, as
mentioned above. Whether one is concerned with aoe specific or total oxygen
consumption the problem involves the dependence of the rate on the weight, namely,
the value and significance of the power function,

Ven or Oy —aw?

The regression coefficient employed frequently is 0.67, a value that corresponds
to the relationship between the surface of an animal and its body weight. This
relationship indeed appears to be undoubtedly a spurious correlation as suggested
by Weymouth, Crismon, Hall, Belding and Field, (1944) ; Brody (1945) ; Zeuthen
(1953) and others. Regardless, it would appear ludicrous to assume the validity
of such a “standard” value, particularly in view of the fact that rate functions
vary with the influence of internal and external parameters. Further, there is no
reason to believe that this influence on rate merely shifts the position of the regres-
sion line on the ordinate. Zeuthen (1953) has shown that the regression coeffi-
cient, b, assumes very different values ranging from 1.0 to negative numbers.
Weymouth et al. (1944) compared respiratory rate to body weight in the kelp
crab Pugettia producta with a series of Crustacea and found an exponent of 0.80.
Scholander e¢ al. (1953) likewise found a similar exponent, 0.80 to 0.85 when
arctic and tropical crustaceans and fish were compared. Roberts (1957a) found
for Pachygrapsus crassipes a power function of 0.664. Bertalanffy (1951) has
correlated three different proportions of metabolism to weight with three growth
types, namely 74, 34 and 1. Categorical definition of these power functions sug-
gests them to be species-specific and unalterable, at least with respect to metabolic
relationships. Bertalanffy and Krywienczyk (1953) demonstrated that oxygen
consumption in brine shrimp, Artemia salina, plotted as a function of body weight
follows the surface rule. These animals were cultured in artificial sea water and
maintained at 25° to 27° C.

Another aspect of weight-specific respiration to body weight is the change in
slope when two or more regression lines are compared, these lines resulting from
measurements made under different seasonal or experimental conditions (altering
one or more parameters, such as a constant temperature and two salinities). This
concerns the significance of parallelism or convergence of regression lines at the
low or high end of the weight range. These conditions can be discussed either
by statistical analysis of b values or by comparison of Q,,. As regards the latter,
Rao and Bullock (1954) have reviewed much of the literature. They summarize
generally, stating (p. 38) that there is “a common increase of temperature coefficient -
with size, on the specified assumption of a weight regression.” Ellenby (1951)

240 PAUL Ay DEHNELE

compared body size in the isopod Ligia oceanica, relative to oxygen consumption
and pleopod beat. He found that total oxygen consumption at 25° C. was propor-
tional to the 0.726 power of body weight, but that this value was not statistically
different from 0.66. Oxygen consumption per unit of length? is constant over the
size range (surface area is proportional to length?). Pleopod beat, on the other
hand, gave a 0.66 power of body length at 15° C. and at 25° C. the value was 0.59,
and these are not significantly different from the 0.5 power. Ina later paper on a
study of predicting oxygen consumption, Ellenby and Evans (1956) believed that
in Ligia greater accuracy could be obtained from body weight than from a function
of body length. Oxygen consumption prediction for prepupae of Drosophila
melanogaster was found to be more accurate if based on surface area. Vernberg
and Gray (1953) studied oxygen consumption of brain brei determined at 30° C.
in a series of marine teleosts, and found two species whose rates of oxygen con-
sumption were independent of weight or length.

Clark (1955) studied the effect of temperature on the oxygen consumption of
the terrestrial amphipod Talitrus sylvaticus. He reported an exponent of 0.836
at 25° C. Further he showed that weight-specific oxygen consumption was greater
in small animals (1.5 mg.) than in large ones (21.0 mg.) at temperatures above
15° C. The Q,, values for winter animals between 20° and 30° C. was 2.33 for
small amphipods and 1.66 for large ones. Roberts (1957a) determined oxygen
consumption in the shore crab Pachygrapsus crassipes that were acclimated to
three temperatures for a period up to seventeen days. Over the weight range
used (1 to 40 grams) he found that Q,, varied directly with weight over the ac-
climation temperature range 6° to 23:52 ©) (Oy, — 2-73 at five etams topo aes
thirty-five grams). No change was noted between 8.5° and 16° C. (Q,, for all
weights was 2.66). At the higher acclimation temperature, slope of the negative
linear regression changed from — 0.336 at the two lower temperatures to — 0.270
at the high temperature, a change shown to be statistically significant.

If O,, values for seasonal rate-temperature data are compared for the two
species of Hemigrapsus and for any weight of crab it is seen that generally Q,,
values are relatively low at low temperatures in winter animals. A rise with in-
creasing temperature to approximately 15° C. occurs, then Q,, values decrease as
the temperature continues to rise. The same pattern exists for summer crabs
except that at the lowest temperatures (3.5° to 5° C.) there is cold depression.
When different weight animals are compared at any temperature interval, no
generalization can be made as no trend is noted for either species, summer or
winter. .

Experimental results for the series of acclimation temperature and salinity
combinations fail to show any definite trend of Q,, values with increase in size or
with increase in temperature for a given size. It should be mentioned that as with
seasonal rate-temperature experiments, there is a tendency for Q,, values to in-
crease at lower acclimation temperatures at both salinities, and then to decrease as
acclimation temperatures increase. It is well to note at this point that Rao and
Bullock (1954) have shown that in many instances Q,, shows no trend with size.
These data support this contention.

Dependence of metabolism on body size for the temperature and salinity ac-
climation studies is somewhat at variance with that reported in the literature.
Weight-specific oxygen consumption b values for both salinities and species range

RESPIRATION INTERABS 241

from — 0.685 to — 0.333, or the positive linear regression coefficients range from
0.315 to 0.667 (Table I). Only a very few instances approached the reported
0.66 or 0.75 exponent. For the most part positive regression coefficients are
relatively low values. If individual slopes or total comparison of the four ac-
climation temperatures at either salinity are noted, it is seen that these slopes
change in some cases to a considerable degree. Statistically significant differences
have been noted for H. oregonensis at low salinity, and H. nudus at high salinity.
For any given acclimation temperature, scatter tends to be greatest for small crabs.
As the acclimation temperature increases scatter increases along the entire weight
range. Reference to Table I gives the correlation coefficients, 7, of each of the
acclimation temperature regression lines for both salinities and both species. In
each case the value for r is significant at the one per cent level as determined from
a table of significance of r (Simpson and Roe, 1939). Sample size (N) is greater
than 25 for all calculated lines. It should be mentioned that regardless of in-
creased scatter at the small end of the weight range, a straight line has been
demonstrated to be statistically the best fit, as a parabolic function described less
well weight-specific oxygen consumption data.

From these data it would be difficult to assign a positive linear regression coeffi-
cient to a species and suggest that such a value might be an inherent or fixed
character of that species. It is evident that the regression coefficient is dependent
upon past environmental histories of the animal as well as experimental variables
(temperature and/or salinity) to which the animal is exposed. Variability of the
response of an animal is noted to be different depending upon the weight range,
temperature, particularly, and salinity combinations.

Temperature and salinity relations

In a recent paper Kinne (1956a) has discussed aspects of temperature and
salinity and their biological effect on marine, brackish and fresh water animals.
Generally, he believes that temperature increase intensifies activity and decreases
resistance, and temperature decrease produces the opposite effect. Further, marine
organisms often resist a low salinity at extremely low temperatures. The resistance
and existence of marine and brackish species is facilitated in a high salinity by a
relatively high temperature, and in a low salinity by a relatively lower one. Near
limits of tolerance, low salinity and high temperature are often lethal, but low
temperature and low salinity are tolerated. Heat resistance in brackish water
species depends to a great degree on salinity. Decrease in environmental salinity
causes a decrease in heat resistance, and with an increase in salinity, heat resistance
increases.

Results from experiments on the effect of temperature and salinity on high
temperature tolerance in the two species of Hemigrapsus do not support completely
these ideas. It. has been determined that acclimation to high temperature resulted
in an appreciable increase in temperature tolerance, both in summer- and winter-
adapted animals, and acclimation to low salintiy decreased resistance. In this
geographical area the two experimental combinations of temperature and _ salinity
that produced the greatest (20° C., 75% sea water) and least (5° C., 35% sea
water) resistance to high test tolerance temperatures never occur seasonally in the
environment (Todd and Dehnel, 1960).

242 PAULA DERNEE

Numerous studies have demonstrated that metabolism increases when organisms.
are removed from their normal salinity medium and placed in a stress medium.
Schlieper (1929) showed that oxygen consumption of the crab Carcinus maenas
increased with decreasing salt concentration. Similarly, results were obtained for
Nereis diversicolor and gill respiration of Mytilus edulis. Schwabe (1933) has.
reported a similar situation for the crayfish, Potamobius fluviatilis, as have
Flemister and Flemister (1951) for the crab Ocypode albicans. These reports
concern measurements made in different salinities and at a given temperature. It
has been suggested that this increased rate of oxygen consumption reflected in-
creased osmotic work. Schlieper (1929) suggested that in higher salt concentra-
tions CO,, a general stimulant for cellular respiration, would be removed more
readily from the animal, whereas in lower salinities there would be a tendency for
CO, to accumulate, thus increasing respiratory rate. Later, Schlieper (1935)
proposed the idea that water content of tissues was related directly to the amount
of oxygen consumed. A higher water content increased volume of the tissues and
the surface, which in turn facilitated absorption and hence the consumption of
oxygen. This interpretation that increased metabolic rate results from increased
osmotic stress has been subject to criticism. Krogh (1939) has discussed
thoroughly this aspect of osmoregulation. Wikgren (1953) in an extensive re-
view of osmoregulation as it is influenced by temperature has stated that the
difference in oxygen requirement in fresh water as opposed to isotonic media was
not attributable to osmotic regulation. He suggests that low salt concentrations
stimulate basal metabolism directly by increased swelling of tissues or by influencing
endocrine balance. These ideas in part are in accord with Schlieper (1935).

Gross (1957) reports that certain crabs, e.g., Pachygrapsus, show violent
attempts to escape from a medium which departs much from normal sea water
in concentration. He suggests that increased oxygen consumption in increased
osmotic stress results from increased muscular activity. Pachygrapsus crassipes
has been shown by Gross (1957) to be a good regulator in both hypo- and
hypertonic media. This species lives in an intertidal situation in sea water of a
concentration of approximately 100%, and probably never is exposed to lower
salinities in the field.

Considering the results obtained in this study, it is evident that when H.
oregonensis and H. nudus are acclimated to a series of temperatures at either low
or high salinity, weight-specific oxygen consumption is greatest after acclimation to
the low temperature (5° C.). As that acclimation temperature increases, oxygen
consumption decreases. When low and high salinities are compared at each ac-
climation temperature, the respiratory rate for H. oregonensis is higher at the low
salinity, over most of the weight range for all temperatures (see Figure 7 for
5° and 20° C.). For H. nudus acclimated to the same conditions, the rate is
higher for the low temperature, low salinity combination, but as the acclimation
temperature increases, oxygen consumption is higher at the high salinity (Fig. 8).
The results obtained for H. oregonensis might be considered to be in accord with
the idea of escaping an unfavorable medium as suggested by Gross (1957), or due
to increased osmotic work at lower salinities as suggested by Schlieper (1935)
and others. On the other hand these ideas would not explain the results obtained
for H. nudus. When seasonal environmental changes in this area are considered,
it cannot be stated that one temperature-salinity combination is normal any more

RESPIRATION IN CRABS 243

than the reverse relations that occur at another season. ‘This is in contradistinction
to that for an open coast intertidal species. It is recalled that during the summer a
low field salinity normally exists, comparable to the experimental one, and is
maintained for several months. During this period the temperature is high. The
animals on which these experiments were conducted were summer-adapted ones.
Further, both species are exposed to the same environmental conditions in the
field. In addition, animals returned to the laboratory and placed directly into low
salinity sea water have never been observed to attempt to escape from that water.
There has been no evidence of this at any of the acclimation temperatures with the
combination of low salinity. This is also the situation for winter-adapted crabs,
collected from a high field salinity and low temperature and placed either directly
or gradually into low salinity water in the laboratory. Under the circumstances
of summer field salinities, it seems highly improbable that the higher rate of oxygen
‘consumption at the low salinity is the result of the attempt by these crabs to escape
this low concentration of sea water, particularly since this low salinity is normal for
several months of the year. A priori, one would expect both species to approximate
the same respiratory rate, even perhaps H. nudus to have a somewhat higher rate.
‘This is based on the fact that the conditions to which H. nudus are exposed in
this locality, are the unusual ones when the ecology of this species is recalled over
the latitudinal distribution. H.nudus, for the most part, is an intertidal species and
occupies a habitat somewhat comparable to Pachygrapsus crassipes.

In connection with the observed high metabolic rate at low salinity, the work
of Schlieper (1953, 1955, 1957) should be noted. Weight-specific oxygen con-
sumption was determined for gills of Mytilus edulis collected from high salinity
water (North Sea 30%c) and for ones from the same area but acclimated to low
salinity (15%) for four weeks. Respiration was higher at the low salinity by
nearly a factor of 2. Mytilus living in low salinity water (Baltic Sea, 15%0) had
a higher oxygen consumption also by a factor of approximately 2, than ones col-
lected from the Baltic but acclimated for four weeks to high salinity water (30%o).
Further, weight-specific oxygen consumption for animals acclimated to low salinity
was approximately the same as that recorded for ones normally living in low
salinity water, about 140 ml. O, per gram per hour. High salinity respiratory rate
for both instances was about 80 ml. O, per gram per hour. This instance serves
to demonstrate the fact that tissues of a sessile animal adapted to low salinity water
have a higher respiratory rate when compared with ones adapted to high salinity
water. Further, laboratory acclimation to low salinity results in a higher rate. On
the other hand, measurements of mechanical activity of gill cilia, frequency of heart
beat and heat resistance are lower in low salinity sea water.

Differential metabolic response of Hemigrapsus to salinity is believed, in part, to
be the result of osmotic stress, 1.e., increased work to maintain an osmotic balance.
At high temperatures metabolic activity increases and low salinity may be a greater
stress than a high one. This low salinity could reduce resistance and cause a
greater osmotic problem; the resulting high oxygen consumption reflects osmotic
work. The problem may be resolved, in part, by a present study of osmotic be-
haviour of these two species to determine the responses to field conditions and to
various laboratory combinations of temperature and salinity. Death at high
temperature and low salinity may be due to osmotic breakdown. But the high
temperature at which osmotic breakdown occurs depends upon the acclimation

244 PAULA, (DEEIN IEE

temperature, salinity and season. Gross (1957) has calculated data from Jones
(1941) to demonstrate change in osmotic gradients between blood and external
medium in H. oregonensis and H. nudus as increased osmotic stresses are applied.
These data show that as sea water concentration decreases below normal sea water,
the internal osmotic concentration remains relatively constant.

It is of significance and interest to be able to compare different populations of
the two species of Hemigrapsus. Gross (1955) has demonstrated the general
ability of terrestrial and semi-terrestrial crabs to regulate osmotically in dilute and
concentrated sea water. He has found that H. oregonensis and H. nudus from
southern California can regulate in 150% sea water for about twenty hours.
However, both species from Vancouver die at that concentration in from two to
six hours. At 125% sea water survival is longer, one to two weeks, but mortality
is relatively high during this time.

Recently Verwey (1957) has reviewed the problem of the influence of tem-
perature on osmoregulation and has discussed means whereby marine and brackish
animals attempt to live in a changing environment. For instance, the shrimp
Crangon crangon migrates seaward toward the North Sea in autumn and returns
in the spring to the region of the Dutch coast. Experimentally, it was found that
salinity must increase as temperature drops for survival of this species. By
migrating seaward the shrimp reach waters of higher salinity and somewhat warmer
waters for winter survival, and return in the spring when coastal waters have a
higher salinity. Other species that show a seasonal migration are those that
migrate offshore in spring and return in autumn, such as the spider crab Hyas
araneus, the shrimp Crangon allmanni and the prawn Pandalus montagm. ‘This.
suggests that low salinities are tolerated better when the temperature is low.
Still other species are those that do not migrate, the crab Riithropanopeus harrist
and the amphipod Gammarus duebem. ‘These forms withstand low salinity better
with a corresponding low temperature. They differ completely from C. crangon
in their tolerance to temperature and salinity, are similar to Hyas araneus in
tolerance, but differ from H. araneus in that they do not migrate. Verwey, in
discussing these data, calculated changes in osmotic pressures expressed in atmos-
pheres as opposed to the usual method of expressing salinity data as parts per
thousand, freezing point depression (delta) or per cent sea water. Expression in
atmospheres relates pressure, salinity and temperature. For instance, C. crangon
kept in a salinity of 33% and given a temperature change from 4° C. to 21° C.,,
gives a value of 19.0 atmospheres for the blood at 4° C., and 19.1 atmospheres at
21° C. Sea water of 33% at 4° C. has an osmotic pressure of 22.99 atmospheres.
and at 21° C. the pressure is 24.40 atmospheres. Osmotic pressure of the blood
is unchanged, sea water osmotic pressure has increased with the rise in temperature,
and the gradient between blood and sea water has increased. A salinity of 33%o
is optimal for adult Crangon at 4° C. At a higher temperature (21° C.) the
optimal salinity is about 28%. The absolute differences expressed in atmospheres.
in osmotic pressure of the medium and blood at the optimal salinities for these two
temperatures are the same (see Verwey, 1957, for further discussion and refer-
ences). Migration of Crangon in the autumn results in the animal moving into
water of a higher salinity and hence higher osmotic pressure, eliminating differ-
ences between blood and external medium caused by the drop in temperature, as.
well as the drop in salinity. It is suggested that Crangon attempts to maintain a

RESPIRATION INFECRABS 245

‘constant value in the differences between osmotic pressures of blood and medium
by this migration. In connection with these aspects Panikkar (1940, 1941) found
‘similar relationships for prawns and points out that minimum blood osmotic pres-
sure can be lowered as temperature increases and thus osmotic work to maintain
hypertonicity is less at high temperatures. He has suggested that colonization by
marine animals of fresh and brackish waters in the tropics may be explained by
the fact that range of tolerance to lowered salinity increases with higher tempera-
tures. Euryhaline species can maintain hypertonicity readily and stenohaline or
slightly euryhaline marine species can cope with brackish waters when tempera-
tures are high.

The ideas suggested by Panikkar (1940, 1941) might be extended to help
explain the distribution locally of Hemuigrapsus, particularly H. nudus. Both
species, as noted previously, breed in late winter and spring. ‘The earliest breeding
occurs during the period of low temperature and high salinity. As breeding
proceeds, temperature-salinity relations change until summer conditions exist.
Water currents and tides are such that zoea and megalops larvae from regions near
the outer coast could be carried through the Strait of Juan de Fuca, into the Strait
of Georgia. Temperature-salinity field conditions are approaching summer ones
as the larvae are released by the female and become planktonic. As larvae enter
the Strait of Georgia, Fraser River water is mixing with the more saline water
of the Strait, and the larvae are in a low salinity environment. However, sea
water temperature is rising, and with increasing temperature, tolerance to low
‘salinity increases. Thus, perhaps larvae when exposed to low salinity, resist its
-effect, survive and establish themselves in an area that seasonally has low salinity
and high temperature conditions. Such conditions as these might explain the
invasion of H. nudus into this geographic area. It would appear that H. oregonensis
occupied this area originally, and that secondarily H. nudus became established.
Such an idea is based on the fact that the usual ecological environment of H. nudus
is the open coast intertidal. If an east-west cline of intertidal distribution is con-
‘sidered, H. nudus is very abundant intertidally, whereas the density of the H.
oregonensis population decreases as open coast areas are reached.

Experiments with the larvae of these two species, to determine resistance, would
‘clarify the responses to high test tolerance temperatures, when exposed to various
temperature-salinity combinations. Preliminary experiments suggest that larval
responses are similar to those of the adults, and this relates favorably to environ-
mental conditions that exist during the planktonic and settling period of Hemu-
grapsus larvae. It would be most significant to determine acclimation of oxygen
consumption and temperature tolerance to different temperature-salinity combina-
tions for populations of both species throughout their latitudinal distribution.

SUMMARY

1. Oxygen consumption measurements have been made on two species of crabs,
Hemigrapsus oregonensis and H. nudus. These crabs were either brought from
field conditions and measured directly, or acclimated to different combinations
of temperature and salinity for two to three weeks prior to experimentation.

2. Acutely measured seasonal rate-temperature curves for summer and winter
animals of both species, kept at their seasonal salinity, have shown that summer

246 PAUL A. DEHNEL

animals at all temperatures over the physiological range have a higher weight--
specific oxygen consumption than winter ones.

3. Winter animals of both species are depressed at a lower temperature than
summer ones, and depression at the high temperature begins at about the same
point for both summer and winter crabs. Comparison of summer and winter
animals of both species at 30° C. shows about the same amount of depression.

4. Both species fail to show an acclimation shift on the ordinate, summer
versus winter, but winter animals show an acclimation of the upper physiological
limit.

5. Acclimated rate-temperature curves for both species generally have a higher
Q,, than either the summer or winter acutely measured curves.

6. When summer crabs of both species are acclimated to a series of temperatures.
(5°, 10°, 15° and 20° C.) at either low salinity (25% sea water) or high salinity
(75% sea water), it is seen that as the acclimation temperature increases, oxygen
consumption decreases over most of the weight range. Statistical analysis shows.
these changes in ordinal position of regression lines to be significant at the one
per cent level of probability.

7. In H. oregonensis, weight-specific oxygen consumption is higher when ac-
climated to 25% sea water, at all acclimation temperatures than when acclimated to.
75% and at the same series of temperatures. In the case of H. nudus the respira-
tory rate is higher when crabs are acclimated to the low temperature, low salinity
combination, when compared with low temperature, high salinity. At higher ac-
climation temperatures, crabs acclimated to 75% sea water have the higher rate.
Statistical analysis of ordinal position of regression lines for both species shows.
them to be significant at the one per cent level.

8. When H. oregonensis is acclimated to low salinity and any of the four ac-
climation temperatures, there is a differential size effect; weight-specific oxygen
consumption of small crabs shows a proportionately greater change as acclimation:
temperature increases. And as acclimation temperature increases, size dependence
decreases. Changes in slopes (0) for the total of the four acclimation temperature
regression lines are significant at the one per cent level. If crabs are acclimated.
to the high salinity, no size effect is demonstrable. Comparison of the two ac-.
climation salinities at each acclimation temperature also shows no size effect.
For H. nudus, animals acclimated to low salinity and the four acclimation tem-
peratures show no size effect, but at 75% sea water there is a demonstrable size
effect. Weight-specific oxygen consumption of large animals shows a greater rate
change at higher acclimation temperatures, and there is greater size dependence at
these higher temperatures. Significance of the regression lines is at the five per
cent level. No statistically significant differences in slope exist when the respira-
tory rate is compared for both salinities at each acclimation temperature.

9. Positive linear regression coefficients for these experimental results range
from 0.315 to 0.667. Only a few approach the reported 0.67 exponent. These
data show that the regression coefficient is not an inherent species character, but
is dependent upon intrinsic and extrinsic factors.

10. Values of Q,, have been used to compare seasonal rate-temperature data,
but show little consistency in relation to acutely measured responses as a function
of increasing experimental temperature, size increase or seasonal difference. Simi-.

RESPIRATION IN CRABS 247

larly, the temperature and salinity acclimation data fail to show a definite trend for
size increase or for increase in temperature at a given size. Generally, Q,, values
increase at lower acclimation temperatures, at both salinities, and then decrease as
the temperatures increase.

11. Salinity affects the metabolic response of these two species of crabs to
temperature. Weight-specific oxygen consumption is highest at the low tem-
perature, low salinity combination. As temperature increases rate of oxygen
consumption remains higher at the low salinity for H. oregonensis, but high salinity
results in a higher rate for H. nudus. This greater response to low salinity is
thought, in part, to be the result of increased work to maintain an osmotic balance.
It does not result from increased muscular activity.

12. Differential responses of Hemigrapsus to temperature and salinity, as
measured by oxygen consumption and temperature tolerance, are suggested as a
means by which H. nudus, in particular, became established in the geographic area
of this study.

ELMER ATURE Cri D

von BeErTALANFFY, L., 1951. Metabolic types and growth types. Amer. Nat., 85: 111-117.

voN BERTALANFFY, L., AND J. KrywiENnczyk, 1953. The surface rule in crustaceans. Amer.
Nat., 87: 107-110.

Bropy, S., 1945. Bioenergetics and Growth. New York: Reinhold Publishing Corp.

BrorekeMa, M. M. M., 1941. Seasonal movements and the osmotic behaviour of the shrimp
Crangon crangon L. Arch. Néerl. Zool., 6: 1-100.

Buttock, T. H., 1955. Compensation for temperature in the metabolism and activity of
poikilotherms. Bzol. Rev., 30: 311-342.

Crark, D. P., 1955. The influence of body weight, temperature and season upon the rate of
oxygen consumption of the terrestrial amphipod, Talitrus sylvaticus (Haswell). Biol.
Bull., 108: 253-257.

DeEuNEL, P. A., 1955. Rates of growth of gastropods as a function of latitude. Physiol. Zodl.,
28: 115-144.

Deunet, P. A., 1956. Growth rates in latitudinally and vertically separated populations of
Mytilus californianus. Biol. Bull., 110: 43-53.

Epwarps, G. A., AND L. Irvinc, 1943a. The influence of temperature and season upon the
oxygen consumption of the sand crab, Emerita talpoida Say. J. Cell. Comp. Physiol.,
21: 169-182.

Epwarps, G. A., anp L. Irvine, 1943b. The influence of season and temperature upon the
oxygen consumption of the beach flea, Talorchestia megalopthalma. J. Cell. Comp.
Physiol., 21: 183-189.

Etrtensy, C., 1951. Body size in relation to oxygen consumption and pleopod beat in Ligia
oceamica L. J. Exp. Biol., 28: 492-507.

ELLenBy, C., AND D. A. Evans, 1956. On the relative importance of body weight and surface
area measurements for the prediction of the level of oxygen consumption of Ligia
oceanica L. and prepupae of Drosophila melanogaster Meig. J. Exp. Biol., 33: 134-141.

FLemister, L. J., ano S. C. Friemister, 1951. Chloride ion regulation and oxygen consump-
tion in the crab Ocypode albicans (Bosq). Biol. Bull., 101: 259-273.

Gross, W. J., 1954. Osmotic responses in the sipunculid Dendrostomum gostericolum. J. Exp.
Biol., 31: 402-423.

Gross, W. J., 1955. Aspects of osmotic regulation in crabs showing the terrestrial habit.
Amer. Nat., 89: 205-222.

Gross, W. J., 1957. A behavioral mechanism for osmotic regulation in a semi-terrestrial
crab. Biol. Bull., 113: 268-274.

Jones, L. L., 1941. Osmotic regulation in several crabs of the Pacific Coast of North America.
J. Cell. Comp. Physiol., 18: 79-92.

Kinne, O., 1953. Zur Biologie and Physiologie von Gammarus duebeni Lillj. II. Uber die
Hautungsfrequenz, ihre Abhangigkeit von Temperatur und Salzgehalt, sowie iber

248 PAULJAS DEHINEL

ihr Verhalten bei isoliert gehaltenen und amputierten Versuchstieren. Zool. Jahrb.
Allg. Zool. u. Physiol., 64: 183-206.

Kinng, O., 1956a. Uber Temperatur und Salzgehalt und ihre physiologisch-biologische Bedeu-
tung. Biol. Zentralbl., 75: 314-327.

Kinng, O., 1956b. Uber den Einfluss des Salzgehaltes und der Temperatur auf Wachstum,
Form und Vermehrung bei dem Hydroidpolypen Cordylophora caspia (Pallas), The-
cata, Clavidae. I. Mitteilung uber den Einfluss des Salzgehaltes auf Wachstum und
Entwicklung mariner, brackischer und limnischer Organismen. Zool. Jahrb. Allg.
Zool. u. Physiol., 66: 565-638.

Kinng, O., 1957. A programmatic study of comparative biology of marine and brackish water
animals. Ann. Biol., 33: 87-92. :

KInNNE, O., AND H. W. RotrHaAuwe, 1952. Biologische Beobachtungen tiber die Blutkon-
zentration an Heteropanope tridentatus (Maitland). Kieler Meeresforsch., 8: 212-217.

Krocu, A., 1939. Osmotic Regulation in Aquatic Animals. Cambridge at the University
Press.

Lorts, B., 1956. The effect of salinity changes on the respiratory rate of the prawn Palae-
monetes varians (Leach). J. Exp. Biol., 33: 730-736.

MarsHat., S. M., A. G. NicHottrs anp A. P. Orr, 1935. On the biology of Calanus fin-
marchicus VI. Oxygen consumption in relation to environmental conditions. J. Mar.
Biol. Assoc., 20; 1-28.

OsTLE, B., 1954. Statistics in Research. Ames, Iowa: The Iowa State College Press.
PANIKKAR, N. K., 1940. Influence of temperature on osmotic behaviour of some crustacea
and its bearing upon problems of animal distribution. Nature, 146: 366-367.
PANIKKAR, N. K., 1941. Osmo-regulation in some palaemonid prawns. J. Mar. Biol. Assoc.,

25: 317-359.

Precut, H., 1951. Der Einfluss der Temperatur auf das Fermentsystem. Verh. dtsch. zool.
Ges., 1950, p. 179.

Prosser, C. L., 1955. Physiological variation in animals. Biol. Rev., 30: 229-262.

Prosser, C. L., J. W. Green anp T. J. Cuow, 1955. Ionic and osmotic concentrations in
blood and urine of Pachygrapsus crassipes acclimated to different salinities. Bzol.
Bull., 109: 99-107.

Rao, K. P., 1953. Rate of water propulsion in Mytilus californianus as a function of latitude.
Biol. Bull., 104: 171-181.

Rao, K. P., ano T. H. Butiocx, 1954. Qi as a function of size and habitat temperature in
poikilotherms. Amer. Nat., 88: 33-44.

Roserts, J. L., 1957a. Thermal acclimation of metabolism in the crab Pachygrapsus crassipes
Randall. I. The influence of body size, starvation, and molting. Physiol. Zool., 30:
232-242.

Roserts, J. L., 1957b. Thermal acclimation of metabolism in the crab Pachygrapsus crassipes
Randall. II. Mechanisms and the influence of season and latitude. Physiol. Zool., 30:
242-255.

RogBertson, J. D., 1949. Ionic regulation in some marine invertebrates. J. Exp. Biol., 26:
182-200.

Ropertson, J. D., 1953. Further studies on ionic regulation in marine invertebrates. J. Exp.
Biol., 30: 277-296.

ScHLIEPER, C., 1929. Ueber die Einwirkung nieder Salzkonzentrationen auf marine Organis-
men. Zeitschr. vergl. Physiol., 9: 478-514.

ScHLIEPER, C., 1935. Neuere Ergebnisse und Probleme aus dem Gebiet der Osmoregulation
wasserlebender Tiere. Biol. Rev., 10: 334-360.

ScHLIEPER, C., 1953. Zur Frage der Beziehungen zwischen osmotischer Resistenz und Grun-
dumsatz bei euryhalinen Meeresvertebraten. Naturwiss., 20: 538-539.

ScHLIEPER, C., 1955. Ueber die physiologischen Wirkungen des Brackwassers. (Nach Ver-
suchen an der Miesmuschel Mytilus edulis). Kieler Meeresforsch., 11: 22-33.
Scuvieper, C., 1957. Comparative study of Asterias rubens and Mytilus edulis from the North
Sea (30 per 1,000 S) and the western Baltic Sea (15 per 1,000 S). Ann. Brol., 33:

117-127.

Scumitt, W. L., 1921. The marine decapod crustacea of California. Univ. Calif. Publ. Zool.,

23: 1-470,

RESPIRATION IN \CRABS 249

SCHOLANDER, P. F., 1949. Volumetric respirometer for aquatic animals. Rev. Sct. Instr., 20:
885-887.

ScHOLANDER, P. F., W. Fiacc, V. Watters AND L. Irvine, 1953. Climatic adaptation in
arctic and tropical poikilotherms. Physiol. Zool., 26: 67-92.

ScuHwase, E., 1933. Ueber die Osmoregulation verschiedener Krebse (malacostracen).
Zettschr. vergl. Physiol., 19: 183-236.

SEGAL, E., 1956. Microgeographic variation as thermal acclimation in an intertidal mollusc.
Biol. Bull., 111: 129-152.

Stimpson, G. G., AND A. Rog, 1939. Quantitative Zoology. New York: McGraw-Hill Book
Co.

SmitH, R. I., 1955a. On the distribution of Nereis diversicolor in relation to salinity in the
vicinity of Tvarminne, Finland, and the Isefjord, Denmark. Biol. Bull., 108: 326-345.

SmitH, R. I., 1955b. Comparison of the level of chloride regulation by Nereis diversicolor
in different parts of its geographical range. Biol. Bull., 109: 453-474.

SmitH, R. I., 1957. A note on the tolerance of low salinities by nereid polychaetes and its
relation to temperatures and reproductive habit. Ann. Biol., 33: 93-107.

Topp, M.-E., anp P. A. DEHNEL, 1960. The influence of temperature and salinity on heat
tolerance in two grapsoid crabs, Hemigrapsus nudus and Hemigrapsus oregonensis.
Biol. Bull., 118: 150-172.

VERNBERG, F. J., AND I. E. Gray, 1953. A comparative study of the respiratory metabolism
of excised brain tissue of marine teleosts. Biol. Bull., 104: 445-449.

Verwey, J., 1957. A plea for the study of temperature influence on osmotic regulation. Ann.
Biol., 33: 129-149.

WENNESLAND, R., 1951. A volumetric microrespirometer for studies of tissue metabolism.
Science, 114: 100-103.

WeEymouTH, F. W., J. M. Crismon, V. E. Hatt, H. S. BELDING anv J. Friern II, 1944.
Total and tissue respiration in relation to body weight: a comparison of the kelp
crab with other crustaceans and with mammals. Physiol. Zo6l., 17: 50-71.

WIKGREN, Bo-JuNGaAR, 1953. Osmotic regulation in some aquatic animals with special refer-
ence to the influence of temperature. Acta Zool. Fennica, 71: 1-102.

ZEUTHEN, E., 1953. Oxygen uptake as related to body size in organisms. Quart. Rev. Biol.,
28: 1-12.

LIMB REGENERATION AND ENDOCRINE ACTIVITY Ik iis
CRAMGELS Eka ar

JAMES B. DURAND

Department of Biology, College of South Jersey, Rutgers—The State University,
Camden 2, New Jersey

With the development of the concept of neurosecretion much new information
has been obtained concerning the endocrine control of growth in crustaceans.
Thus, neurosecretory cells in the x organs within the eyestalk send their
secretory products via their axons to swollen axon terminations which are con-
centrated around a blood sinus and called the sinus gland. The secretory products
are stored here for later release into the blood (Bliss, 1953; Bliss, Durand and
Welsh, 1954; Carlisle, 1953; Passano, 1953). One hormone of the x organ is
molt-inhibiting in crabs (Passano, 1953), and evidence has been presented which
indicates that it is produced in the crayfish by the type 2 neurosecretory cells in
the x organ (Durand, 1956). ‘The specific roles of other neurosecretory cells in
the brain and eyestalks, some of which send their axons to the sinus glands, are
unknown. Work on crustacean neurosecretion has been reviewed recently by
Knowles and Carlisle (1956).

In addition to neurosecretory elements, another endocrine gland, the y organ,
is important in the regulation of crustacean growth. This gland, first described
by Gabe (1953), produces a hormone which has been shown to stimulate growth
and molting in Carcinides (Echalier, 1954, 1955).

Because both of these endocrine glands function in the control of growth, it is
important to know the time relationships between their periods of activity. The
secretory activity of the two organs was studied, therefore, along with the regenera-
tion of autotomized limbs since Bliss (1956) showed that certain stages in the
regeneration of limbs in Gecarcinus reflect growth conditions within the animal
in general.

MATERIALS AND METHODS
Ammals

Young Orconectes limosa (12-15 mm. carapace length) were collected from the
Rahway River, Rahway, N. J. during the summer of 1956. Crayfish were then
moved to Camden, N. J. and kept in half pint paraffin-lined containers, one cray-
fish per container. The containers were covered with translucent plastic and kept
in a water bath (temperature 20-25° C.). The water in the containers was
changed twice weekly during July and August and once a week during September

1 This work was supported by a National Science Foundation grant G-2525, and by the
Rutgers University Research Council.

21 wish to express my thanks to Dr. Dorothy Bliss, Dr. Berta Scharrer and Dr. John
Welsh for their helpful criticism of the manuscript.

250

ENDOCRINES AND CRAYFISH GROWTH Zoi

and October. Freshly collected duckweed and algae, upon which the animals feed,
were added to the containers each time the water was changed.

Regeneration

In one group of animals, either on the day of molt or on the day after molt,
each crayfish was made to autotomize its right rear walking leg. Pinching the
meropodite with fine forceps was usually a sufficient stimulus to bring about
autotomy. Animals were examined for signs of regeneration of the limb about
three times per week during July and August and twice a week thereafter to the
middle of October. A regenerating limb grows out from the autotomy plane as
a long thin flexible structure, its segments linearly arranged within a membranous
sac. This membranous sac and its contents, hereafter referred to as a limb bud,
was measured under 15 X magnification with a pair of dividers, and the length was
taken by comparison with the half-mm. scale of a metal machininst’s rule. The
limb bud was measured from the tip to the plane of autotomy.

In another group, crayfish were kept in the laboratory as described above and
were made to autotomize the right hind walking leg within two days of collection.
A record of the regeneration of the leg was made for each crayfish as described.

In order to compare the rates of limb regeneration of different sized animals,
the size of the regenerating limb is expressed as the per cent of the carapace length:

length of regenerating limb
= 2 x 100.
length of carapace

This formula is comparable to that used for studies of limb regeneration in

Gecarcinus lateralis (Bliss, 1956).

FAiistology

Eyestalks and y organs were fixed in Bouin plus 1% CaCl, at various stages of
limb regeneration and embedded in Tissuemat. FEyestalk serial sections (7 ») and
y organ serial sections (10 ») were stained in most cases with the aldehyde fuchsin
technique (Halmi, 1952) with modifications as suggested by Dawson (1953).

Counts of the type 2 neurosecretory cells which were located in the x organ
and which contained secretory material were made according to the method given
in a previous paper (Durand, 1956).

RESULTS
Limb regeneration

Studies of the regeneration of autotomized limbs were carried out in this in-
vestigation. The rate of limb regeneration was shown earlier (Bliss, 1956) to vary
at certain stages of the intermolt period, and it was hoped that limb regeneration
curves might be used as an indicator of endocrine conditions in the animal in
general. As will be seen, it was found that only at certain times during the inter-
molt period did the rate of limb regeneration reflect changes in endocrine conditions
in the animal.

Typical curves of limb regeneration are shown in Figure 1. These are selected
curves for individual crayfish. The animals were made to autotomize on the day

252 JAMES B. DURAND

after collection, and the rate of limb regeneration was measured to the following
molt. Since the animals were collected at random, it is believed that all stages of
the intermolt period were represented in the population at the beginning of the
experiments. Therefore, Figure 1 illustrates limb regeneration which began at
successive stages of the molt cycle.

It is seen that the stages of limb regeneration are similar to those reported by
Bliss (1956) for Gecarcinus. Thus, following autotomy there is a lag period
before any visible signs of regeneration occur; this period probably represents the
time required for healing processes and the general mobilization of local growth

°

30

R VALUE
= ft
°o (@)
(e}
oO a

o

40
DAYS BEFORE MOLT

Figure 1. Limb regeneration curves for selected crayfish.

processes. It is interesting that the length of the lag period (mean, 6 days) is
essentially constant no matter when autotomy occurs during the intermolt period.

Next, there occurs a period of basal growth, during which time the regeneration
of the limb progresses rapidly. This stage is also essentially the same (the slopes
are the same) no matter when it occurs during the intermolt period. The mean
slope of the basal growth stage was 2.2, and 10-11 days were required for the
completion of basal growth.

Following basal growth, linear growth of the regenerating limb ceases, that is,
it enters a period of plateau which lasts until near the end of the intermolt period.
The limb has usually reached about 22% of the carapace length at this time. Ac-
cording to Hodge (1956) differentiation of limb tissue continues during the period

ENDOCRINES AND CRAYFISH GROWTH 290

of plateau. Therefore, only increase in length ceases while development does
continue. The plateau R value is higher in Orconectes than it is in Gecarcinus
(crayfish 22, crab 10). Bliss (personal communication) has suggested that this
might be a reflection of the fact that the segments of the crayfish limb bud are
straight whereas those of the crab are folded upon one another.

For Gecarcinus, Bliss (1956) reported that shortly before the animal molts
growth occurs again. She termed this period premolt growth. Only in some
instances was it possible to detect this period in the young Orconectes used in the
present study. In a few animals a suggestion of the premolt growth stage was
detected (Fig. 1, animals A and B). Since this period is often short, it is possible
that it occurred between the last measurement and the molt of the animal. With
a measurement interval of 3-5 days, therefore, it would be possible to detect in
most cases only a slight increase, if any, in the rate of limb regeneration at the
end of the period of plateau. In view of this possibility, it is interesting that some
animals which were caused to autotomize passed through a period of basal growth
with the normal slope. “However, these animals did not show a period of plateau
(Fig. 1, animals D and E). Instead, growth continued at a rapid rate, and the
animals molted within a few days. It is believed that these animals had been
collected very near the end of their intermolt period and that basal growth and
premolt growth are continuous when autotomy occurs late in the intermolt period.

In summary, it is seen that the regeneration of autotomized limbs follows the
same pattern in Orconectes as in Gecarcinus. Increase in length is restricted to
the periods of basal growth and premolt growth. Since the slope of the basal
growth curve remains the same at different stages in the molt cycle, it appears that
this stage is not directly under endocrine control. On the other hand, premolt
growth apparently is influenced by endocrine changes in the animal. Evidence for
this is found later in this paper and in the study by Bliss (1956).

Y organ

A non-neurosecretory endocrine gland which functions in the control of growth
in crustaceans is the y organ. Originally described by Gabe in many species of
crustaceans (1953), it is located in the second maxillary segment in Orconectes
limosa. It is always closely associated with the base of one of the large mandibular
muscles and is close to, but not in contact with, the hypodermis. In serial cross-
sections it lies dorso-laterally to the circumoesophageal connectives and posterior
to the oesophagus. The y organ of this animal is irregular in shape; its location
and rambling structure would make it difficult to extirpate.

Histologically the cells of the y organ (Figs. 2, 4, and 6) are fairly uniform in
size, and very fine strands of connective tissue divide the organ into short irregular
cords about 1-2 cells in thickness. Blood spaces appear in the organ but are not
particularly numerous. Frequently, however, blood cells can be seen between
the cords of cells. The cells of the y organ possess a very finely granular, slightly
acidophilic cytoplasm with well defined cell boundaries. The nucleus is sharply
outlined and usually contains a single nucleolus. Scattered in the cytoplasm of the
cells the secretory material appears in the form of irregular clusters of very small
aldehyde fuchsin-positive granules. When the secretory material is present in
very large amounts, the cytoplasm in general is stained with aldehyde fuchsin.

JAMES B. DURAND

254

7

FIGURES 2-

ENDOCRINES AND CRAYFISH GROWTH 255

At these times granules or large droplets of secretory material usually are present
also.

During the greater part of the intermolt period, the cells of the y organ contain
moderate amounts of aldehyde fuchsin-positive material (Fig. 4). Apparently all,
or almost all, of the cells possess similar amounts of secretory material at any
particular time.

At molt, however, the y organ undergoes striking changes in its secretory
content. Animals fixed during the first four or five days after shedding possess y
organs in which it has not been possible to detect secretory material with the
aldehyde fuchsin technique. The cytoplasm of these cells is apparently empty of
secretory material (Fig. 2). The disappearance of the secretory material is
definitely correlated with the molt of the animal although the two may not occur
simultaneously. Some animals, but not all, possess empty y organs before molt.
All animals possess empty y organs immediately following molt. In two cases,
animals fixed while actually in the shedding process had y organs which contained
more secretory material than normally (Fig. 6).

It is interesting to note that, while all animals possess secretory granules during
most of the intermolt period, it is not possible to determine from y organ examina-
tion alone how near to an approaching molt the animals are. Careful study of
serial sections of y organs does suggest, however, that the size of the granules in the
y organ cells increases as the animals approach molt.

Neurosecretory cells

In Table I are recorded the counts of the type 2 neurosecretory cells which
contain secretory material in the x organs of animals fixed at different times during
the intermolt period. It is obvious that a large number of cells possess secretory
material at all times except molt. Thus, from 1-2 days preceding molt to 4-5 days
after molt very few type 2 cells contain secretory droplets or granules (Fig. 3).
Indeed, most of the type 2 cells are devoid of all signs of secretory activity during
this period. Many type 2 cells contain granules or droplets of secretory material
from a few days following molt to just before the next molt (Figs. 5 and 7).

Of further interest is the observation that shortly after molt, when secretory

Ficure 2. Y organ from an animal during the first few days after molt. Secretory
material is absent from the cells. Bouin plus calcium chloride; aldehyde fuchsin; 1140 x.

Ficure 3. Type 2 neurosecretory cells in the x organ of an animal on the day after
molt. Secretory material is absent from the cells. Bouin plus calcium chloride; aldehyde
fuchsin; 1140 x.

Figure 4. Y organ from an animal in the basal growth stage of limb regeneration.
The cells contain only a moderate amount of secretory material. Bouin plus calcium chloride ;
aldehyde fuchsin; 1140 x.

Ficure 5. Type 2 neurosecretory cells in the x organ of an animal in the basal growth
stage of limb regeneration. The secretory material in these cells as clumps of very
small granules. Bouin plus calcium chloride; aldehyde fuchsin; 1140 x.

Figure 6. Y organ from an animal during the molt process. This animal possessed a
high type 2 cell count. Note the large amount of secretory material in the cells. Bouin plus
calcium chloride; aldehyde fuchsin; 1140 x.

Ficure 7. Type 2 neurosecretory cells in the x organ of an animal during the plateau
stage of limb regeneration. Note that the secretory material occurs in the form of droplets.
Bouin plus calcium chloride; aldehyde fuchsin; 1140 x.

256

material is again present in the type 2 cells, the material is in the form of numerous
small granules scattered throughout the cytoplasm of each cell.
intermolt period the secretory material is present in the form of fewer and larger
droplets (Table I and Figs. 5 and 7), similar to those observed in Orconectes

Animal number

Counts of type 2 neurosecretory cells which contained secretory material
in the x organ

Stage of molt cycle

JAMES B. DURAND

TABLE [

Type 2 cell count

Later in the

Form of material

Droplets

3 molting 92 +
39 molting 50 +

1 molt + hrs. ) res

5 molt + hrs. 19 ==

6 molt + hrs. 0 =

2 molt + 1 d. 0 _
58 molt + 1d. 3 -

f molt + 4 d. 25 +

8 molt + 4 d. 34 a
23 molt + 4 d. 46 a
24. molt + 5d. 83 a
37 basal gr. 114 +
10 basal gr. 35) _
11 basal gr. 51 _
25 basal gr. 85 +
AO basal gr. 27 ~
4] basal gr. 80 —
12 early plat. 70 +
14 early plat. 118 _
1S early plat. 126 ~
17 early plat. 98 =
26 early plat. 70 _
29 early plat. 62 +
30 early plat. 47
31 early plat. WS +
32 early plat. 74 +
33 early plat. 23 +
18 late plat. oi +
19 late plat. fii +
20 late plat. 66 =
21 late plat. 64 +
22 late plat. 51 =
42 late plat. 49 BS
46 late plat. 31 +
43 premolt gr. 10 +
44 premolt gr. 20 =
45 premolt gr. 6 +

virilis.

production of the material.

In an earlier study (Durand, 1956) it was suggested that the presence of
secretory material in the form of small granules might represent an early stage in the
If this is so, then the absence of any signs of secretory
material in the x organ type 2 neurosecretory cells in the period immediately follow-

Granules

}t+++i+4+4+441

ENDOCRINES AND CRAYFISH GROWTH a0

ing molt would indicate that the synthetic activities of these cells had ceased. It
is during this time that many active growth processes occur.

DISCUSSION

So far we have seen that major changes occur in the two endocrine organs, x
organ and y organ, at the time the animal molts. The secretory activity of the x
organ neurosecretory cells, as indicated by type 2 cell counts, is compared to growth
of regenerating limbs in Figure 8. The figure represents a generalization of the
data, and an explanation of its construction is given in the next paragraph.

75 [ee ree
oO

20
=
=
=>
5 a
4 =
call

Oo w =< e
VU
ax

N
W

10
Me e
bt

6
ry)
6 10 20 30 36
DAYS AFTER MOLT
LAG BASAL GROWTH PLATEAU Piel t

Figure 8. Type 2 neurosecretory cell counts and R values of crayfish with regenerating limbs
at the indicated times during the period of regeneration.

After animals with regenerating limbs had molted, the R values for each
animal were plotted against the per cent of the growth period, and smooth curves
were drawn through the points. The growth period in this case included the
time from the first appearance of a regenerating limb to the molt of the animal.
From each of the curves, R values were obtained at 10% intervals of the growth
period. If the mean lag period of 6 days is subtracted from the mean intermolt
period of 36 days, each 10% interval in the growth period is equivalent to three
days of the mean intermolt period. In Figure 8 the mean R values are plotted at
each three-day interval. Also plotted are the type 2 cell counts from Table I.
The mean length of time required for the completion of basal growth was 10-11
days, and the type 2 cell count for that stage was placed in the middle of that

258 JAMES B. DURAND

time, on day 11 of the intermolt period. Days 16 to 32 would be the mean length
of the plateau stage of limb regeneration if four days are allowed for the period of
premolt growth. The mean type 2 cell counts for early and late plateau, therefore,
were placed on day 20 and on day 28 of the intermolt period. These last two points
represent, respectively, 25% and 75% of the plateau period. As discussed earlier,
the premolt growth stage probably occurred in the last 3-5 days of the intermolt
period, and so the type 2 cell count for that stage was placed on day 33, three days
before molt.

Figure 8 shows that some animals possessed low type 2 cell counts during the
premolt growth stage of limb regeneration and just after shedding. No high type 2
cell counts were found in animals during the first four days after shedding. In
fact, the counts actually dropped to zero within 24 hours of shedding. About 4-5
days after molt the secretory material could be detected again in the perikaryon of
the cells. Two notable exceptions are animals number 3 and number 39 (Table I).
These animals were fixed while shedding but, nevertheless, possessed high type 2
cell counts. It should be pointed out that the y organs of these two animals, unlike
those of other animals at molt, possessed unusually large amounts of secretory
material. Apparently the initiation of the shedding process was independent of
the y organ and x organ secretory content.

The y organ also underwent its greatest change in secretory content just prior
to the molt of the animal. Thus, secretory granules were present in the y organ
cells during most of the intermolt period except for a few days before and after
molt. In addition the secretory granules disappeared from the y organ cells before
the decrease in the type 2 cell count occurred. Thus, premolt animals were found
to have high type 2 cell counts and full y organs, high type 2 cell counts and empty
y organs, low type 2 cell counts and empty y organs. However, no animals were
found to have low type 2 cell counts and full y organs.

The two endocrine organs consequently exhibited cyclical secretory behavior
associated with molt. When their functions are considered, it is not surprising
that they should do so. What is surprising is that both organs contained secretory
material at approximately the same times during the molt cycle. On the basis of
histological observations alone, it is difficult to interpret these results. For
example, the presence of stainable material within a cell can mean either that the
cell is producing and releasing its secretory products or that it is inactive. In the
succeeding paragraphs certain physiological information concerning the two organs
will be presented which suggests a possible interpretation of the observations re-
ported here.

It is known that neurosecretory cells in the x organ produce a molt-inhibiting
hormone in crabs (Passano, 1953). This hormone is passed via the neurosecretory
cell axons to the sinus glands where it is stored for later release into the blood
(Bliss, 1953; Bliss, Durand and Welsh, 1954; Carlisle, 1953; Passano, 1953).
An earlier study by the author suggested that the type 2 neurosecretory cells in the
x organ are the source of the molt-inhibiting hormone in the crayfish since they are
the only neurosecretory cells that show histologically demonstrable quantitative
changes in secretory material at molt. It was also suggested that the disappearance
of secretory material from the type 2 cells could be accounted for by a rapid transfer
of the material from the x organ to the sinus glands at molt. Another hypothesis is

ENDOCRINES AND CRAYFISH GROWTH 209

that the rate of synthesis of the hormone was reduced below the rate of transfer.
Since the type 2 cells are probably the source of the molt-inhibiting hormone and
since the implantation of eyestalks inhibits the premolt growth stage of limb
regeneration in crabs (Bliss, 1956), it is logical to assume that the presence of
stainable material in the type 2 cells, during the intermolt period when molt-
inhibiting hormone hypothetically is released, indicates that these cells are actively
synthesizing the hormone. The absence of stainable material when molt-inhibiting
hormone supposedly is witheld just before molt could result from a decreased rate
of synthesis or an increased rate of transfer from the cell bodies to the sinus glands.
The reappearance of the secretory material in the form of many small granules 4—5

TV GE 2

Y ORGAN CELLS X ORGAN CELLS

a

LIMB
REGENERATION

BASAL EARLY LAKE PREMOLT

| aie | | GROWTH | | PLATEAU | | PLATEAU | | GROWTH |

Figure 9. Diagrammatic summary of the secretory activity of x organ type 2 neurosecretory
cells and the y organ cells in relation to limb regeneration and the molt cycle.

days after molt would result from a rate of synthesis greater than the rate of transfer
from the perikaryon. Unfortunately, no information its available concerning the
secretory content of the sinus glands during the stages studied in this investigation.
However, Pyle (1943) has shown that the sinus glands of Cambarus virilis (now
Orconectes virilis) lose most of their stainable material soon after molt.

Similar reasoning with respect to the condition in the y organ cells suggests
that these cells may be actively secreting their hormone when they are histologically
empty. Thus, it has been shown that the y organ is the source of a growth-
promoting hormone in Carcinides by Echalier (1954, 1955). Travis (1955, 1957)

260 JAMES B. DURAND

has shown that in Panulirus the resorption of the old integument occurs chiefly
during the last three days before molt and the first two after molt. The production
of all layers of the integument in Panulirus is completed by about eight days after
molt. Travis reported that the intermolt period for the animals she used was
65-70 days. If these processes required a proportionate period of time in
Orconectes, they would have occurred entirely during the period in which the y
organ cells were devoid of their secretory products. For these reasons it seems
logical to assume that the y organ cells are secreting their hormone when they are
histologically empty at molt. However, we cannot conclude this until more is
known about the physiology of the y organ and its interaction with the x organ cells.

In summary, an hypothesis can be constructed (Fig. 9) which agrees with the
known facts but which requires additional experimental evidence. During the
greater part of the intermolt period, secretory granules can be demonstrated in the
type 2 neurosecretory cells and in the y organ cells. Hypothetically, molt-
inhibiting hormone is released from the type 2 neurosecretory cells during this time
while growth-promoting hormone is withheld by the y organ cells or released in
insignificant quantities. The basal growth and plateau stages of limb regeneration
can occur during this period. Just before molt and for a short time after,
secretory material cannot be demonstrated in either the type 2 neurosecretory cells
or the y organ cells. Hypothetically, the molt-inhibiting hormone of the x organ
is not released at this time but the growth-promoting secretion of the y organ
is released. At this time both premolt growth of regenerating limb buds and molt
can occur.

SUMMARY

1. Four periods in the regeneration of autotomized limbs can be identified in the
young of Orconectes limosa. The first, designated lag period, is one in which
no growth occurs and lasts about 6 days. The second, the period of basal growth,
is characterized by rapid growth of the regenerating limb. At the end of this period
the regenerating limb is about 22% of the carapace in length. During the third
period, called plateau, no growth occurs; this period occupies the greater part of
the intermolt period. In the fourth period, premolt growth, rapid growth again
occurs about 3-5 days before molt.

2. The y organ and its secretory behavior are described. Changes in the
activity of the y organ can be demonstrated histologically for only a brief period
which extends from about three days before molt to 4—5 days after molt.

3. The changes in the activity of the type 2 neurosecretory cells in the x organ
are described. The most marked changes occur only just before molt and persist
for a period of 4-5 days after molt.

4. The secretory activity of the x and y organs are discussed in relation to the
molt cycle and the regeneration of autotomized limbs. Arguments are presented
in favor of the hypothesis that the y organ cells produce and release their secretory
products when the type 2 neurosecretory cells in the x organ are inactive.

LITERATURE CITED

Buss, Dorotuy E., 1953. Endocrine control of metabolism in the land crab, Gecarcinus
lateralis (Fréminville). I. Differences in the respiratory metabolism of sinusglandless
and eyestalkless crabs. Biol. Bull., 104: 275-296.

ENDOCRINES AND CRAYFISH GROWTH 261

Briss, Dorotuy E., 1956. Neurosecretion and the control of growth in a decapod crustacean.
In: Bertil Hanstrom, Zoological papers in honour of his sixty-fifth birthday, Novem-
ber 20, 1956, (Wingstrand, K. G., ed.) pp. 56-75, Zoological Institute, Lund, Sweden.

Buiss, DorotHy E., J. B. DurAND AND J. H. Wetsu, 1954. Neurosecretory systems in
decapod Crustacea. Zeitschr. f. Zellf. u. mikr. Anat., 39: 520-536.

Caris_e, D. B., 1953. Studies on Lysmata seticaudata Risso (Crustacea Decapoda). VI.
Notes on the structure of the neurosecretory system of the eyestalk. Pubbl. Stas.
Zool. Napoli, 24: 435-447.

Dawson, A. B., 1953. Evidence for the termination of neurosecretory fibers within the pars
intermedia of the hypophysis of the frog, Rana pipiens. Anat. Rec., 115: 63-69.

Duranp, J. B., 1956. Neurosecretory cell types and their secretory activity in the crayfish.
Biol. Bull., 111: 62-76.

EcHALIER, G., 1954. Recherches expérimentales sur le role de “l’organe Y” dans la mue de
Carcinus moenas (L) Crustacé Décapode. C. R. Acad. Sct. Paris, 238: 523-525.

Ecua ter, G., 1955. Role de l’organe Y dans le determinisme de la mue de Carcinides (Car-
cinus) moenas L. (Crustacés Décapodes) ; Expériences d’implantation. C. R. Acad.
Sct. Paris, 240: 1581-1583.

Gase, M., 1953. Sur l’existence, chez quelques Crustacés Malacostracés, d’un organe com-
parable a la glande de la mue des Insectes. C. R. Acad. Sci. Paris, 237: 1111-1113.

Hatrmy, N. S., 1952. Differentiation of two types of basophils in the adenohypophysis of the
rat and mouse. Stain Tech., 27: 61-64.

Hopce, Mary H., 1956. Autotomy and regeneration in Gecarcinus lateralis. Anat. Rec., 125:
O25.

Know tes, F. G. W., Ann D. B. Cartiste, 1956. Endocrine control in the Crustacea. Biol.
Rev., 31: 396-473.

PassAno, L. M., 1953. Neurosecretory control of molting in crabs by the x organ-sinus gland
complex. Physiologica Comparata et Oecologia, 3: 155-189.

Pye, R. W., 1943. The histogenesis and cyclic phenomena of the sinus gland and x-organ in
Crustacea. Biol. Bull., 85: 87-102.

Travis, DorotHy F., 1955. The molting cycle of the spiny lobster, Panulirus argus Latreille.
II. Pre-ecdysial histological and histochemical changes in the hepatopancreas and
integumental tissues. Biol. Bull., 108: 88-112.

Travis, Dorotuy F., 1957. The molting cycle of the spiny lobster, Panulirus argus Latreille.
IV. Post-ecdysial histological and histochemical changes in the hepatopancreas and
integumental tissues. Biol. Bull., 113: 451-479.

STUDIES ON THE STRUCTURE AND PHYSIOLOGY OF THE eieinen
MUSCLES OF BIRDS: 9. AQUANTITATIVE STUDY On hrm
DISTRIBUTION, PATTERN OF SUCCINIC DEHYDROGENASE

IN THE, PECTORALTS, MAJOR MU SCEE OR TH PiGgnek

J. .G.- GEORGE ANDiGa i aR SARA

Laboratories of Animal Physiology and Histochemistry, Department of Zoology,
M. S. Umversity of Baroda, Baroda, India

George and Naik (1958a, 1958b) have shown that the pectoralis major muscle
of the pigeon consists of two distinct types of fibers, a broad glycogen-loaded white
variety with few or no mitochondria and a narrow fat-loaded red variety having
a large number of mitochondria in them. George and Scaria (1958) im a
histochemical study of dehydrogenases (succinic, malic, lactic, and glycerophosphate
dehydrogenases) in this muscle could not detect the presence of any of these
dehydrogenases in the broad white fibers. They therefore concluded that these
fibers in the pigeon breast muscle are a unique system in which none of the oxidative
processes concerned with the above enzymes takes place and therefore cannot be
considered as analogous to the white fibers of the other vertebrate skeletal muscles
studied. These observations should naturally raise some important questions as well
as doubts regarding the exact role of the broad fibers in this muscle and also the
problem of energetics for their contraction. But, however, the above conclusions
were based solely on histochemical observations and can be considered decisive only
if supported by quantitative studies. Here we report the results of a quantitative
study of the distribution pattern of the succinic dehydrogenase activity in the
pectoralis major muscle of the pigeon.

MATERIAL AND METHODS

It is technically impossible by ordinary methods to estimate directly the activity
of any enzyme in the two types of fibers in the pigeon breast muscle after isolating
the fiber types separately. But the indirect method, viz. of estimating the activity
of the enzyme in different layers of the muscle, where the broad and narrow fibers
vary in number, was adopted in the present study. The fiber architecture of the
pectoralis major muscle of the pigeon has been worked out in detail by George and
Naik (1959). They have found that the least number of broad fibers is found in
the fasciculi situated in the middle of the dorso-ventral axis of the muscle and the
number tends to increase above or below this mark, reaching the maximum in the
most superficial and in the deepest layers of the muscle. They also showed that
per unit area of cross-section of the muscle at any particular depth, the number of
broad fibers is inversely proportional to the number of narrow fibers, and derived
the formula

Ye Or ee Ue
262

SUR ACTIVITY IN PIGEON BECTORALIS 263

(where Y stands for the number of narrow fibers and X for the number of broad
fibers) in order to estimate the number of narrow fibers in any one particular square
mm. area, X being known. Throughout this study we made use of this formula
for finding out the number of narrow fibers in a particular layer of muscle which
was used for the estimation of the succinic dehydrogenase activity.

The method employed in the estimation of succinic dehydrogenase was the one
according to Kun and Abood (1949) which makes use of the principle that colorless
TTC (2,3,5-triphenyl tetrazolium chloride) is reduced to a red insoluble com-

Ficure 1. Illustrating the topography of the muscle pieces used for the study of succinic
dehydrogenase activity in the pigeon breast muscle.

pound, formazan, by the enzymic oxidation of succinate to fumarate. The color
developed was extracted in acetone and the intensity measured in a colorimeter.
The amount of formazan formed was directly read from a standard curve plotted by
reducing completely known quantities of TTC by sodium hydrosulphite. Since
the reduction of TTC by hydrosulphite was found to be reversible, the color was
stabilized by adding acetone within one minute of the addition of hydrosulphite to
the TTC solution, during which time the TTC was completely reduced.

In order to ensure that only uniformly well developed pectoralis major muscle

264 jC. GEORGE ANDS Ce] EAL E SARA

was used in the study, well fed and fully grown, normal laboratory pigeons
(Columba livia) were used throughout.

In each experiment a bird was decapitated and the blood completely drained off.
Pieces of muscle from the regions shown in Figure 1 (A and B) were cut off for
each set of experiments. A portion (1) of this muscle piece A was immediately
mounted on the stage of a freezing microtome with the superficial (ventral) side
facing up, so as to obtain serial horizontal sections of the desired thickness starting
from the superficial side of the muscle. The remaining portion (2) of the same
piece was set apart for counting the broad and narrow fibers. The same procedure
was followed for muscle piece B. In most cases it was found necessary to freeze
the muscle block first with the superficial (ventral) face down and after trimming
the deeper (dorsal) side it was reversed so as to render the surface even and
parallel to the edge of the knife. When the block was frozen hard the epimysium
was removed by a superficial stroke of the microtome knife and serial sections of
0.5 mm. or 1.0 mm. thickness as required, were cut. The sections were immediately
transferred to weighing bottles and the weight determined. ‘They were then chilled
and each of the sections was homogenized in 10 ml. of distilled water. The succinic
dehydrogenase activity of this homogenate was then determined using the method
mentioned above.

The incubation mixture in each case contained 0.5 ml. of 0.1 M phosphate
buffer of pH 7.4, 0.5 ml. of 0.2 M sodium succinate, 1.0 ml. of the muscle homog-
enate and 1.0 ml. of a freshly prepared 0.1% solution of TTC in a total volume of
3 ml. The TTC solution was added last, and after shaking, the tubes were in-
cubated for 15 minutes at 37° C. Seven ml. of acetone were then added)toyeden
tube. The tubes were tightly stoppered, shaken well and centrifuged for about
three minutes at about 3000 rpm. The clear supernatant was drawn off and the
intensity of the color read at 420 mp» on a Klett-Summerson photoelectric colorim-
eter. The readings were corrected for a blank with zero time incubation. In-
hibition experiments with malonate confirmed that the color production was solely
due to succinic dehydrogenase.

Part (2) of the original piece of the muscle was sectioned as shown in Figure 1.
The section was mounted in 50% glycerine and the number of broad fibers per
square mm. counted according to the method described by George and Naik (1959).
From the broad fiber count, the number of narrow fibers in the particular area
was calculated using the formula already referred to.

RESULTS

The dehydrogenase activity is expressed in the number of micrograms of
formazan produced by 100 mg. wet weight of the muscle for 15 minutes at 37° C.
under the conditions of the experiment mentioned above. We do not claim that the
values given are absolute because it is likely that freezing and thawing of the
tissue in the process of cutting sections destroys a certain amount of its enzyme
activity. However, this loss of activity could not be considerable and could be
even neglected in view of the fact that our purpose is to compare the activity at
the different depths and in relation to the number of the fibers per square mm.
area.

The succinic dehydrogenase activity and the number of broad and narrow
fibers per square mm. in the different layers of the regions A and B of the breast

Exp.

SDH ACTIVITY IN PIGEON PECTORALIS

TABLE [|

265

Showing the succinic dehydrogenase activity in the different layers of the regions A and B
(marked in Fig. 1) of the pectoralis major muscle of the pigeon and its relation

to the number of narrow fibers.
of five sets of readings.)

(The figures indicate the average values

Succinic dehydrogenase activity in

No. of broad fibers
per square mm.

No. of narrow fibers

Depths of the muscle in per square mm.

mm. starting from the

superficial side eee
(ventral face) —___ |-- _ | —
Region A | Region B | Region A | Region B Region A Region B

t 0-0.5 132 120 iki 200 15.5 + 2.17 23.0 + 1.18

0.5-1.0 100 98 ons SAT) Sos ea 29S8 40.0 + 1.56

0-1 107 101 278 ola 32.22 + 1.12 | 36.75 + 3.13

1-2 90 59 372 558 42.72 + 1.37 | 64.47 + 1.42

2-3 86 50 402 609 46.25 + 1.44 | 69.37 + 0.82

3-4 75 43 465 646 53.75 £1.15 | 75.87 + 0.54

4-5 59 40 549 662 63.25 + 1.73 | 78.77 + 1.32

5-6 A5 48 638 617 72.50 + 1.05 | 70.00 + 1.08

6-7 43 52 648 595 72.80 + 0.92 | 68.50 + 2.16

7-8 41 58 658 564 74.50 + 1.38 | 65.00 + 1.45

8-9 42 66 652 Sli7, 73.40 + 2.10 | 59.00 + 2.80

9-10 45 76 635 457 70.00 + 1.78 | 53.00 + 2.55

S

MIN.

8

MUSCLE/I5

{100 maWET
as
‘@)

RMAZAN
S

SUCCINIC DEHYDROGENASE ACTIVITY

oF 0
O

0

200
NUMBER OF NARROW FIBERS/Mim

Figure 2. Graph showing the relation between the number of narrow fibers and succinic
dehydrogenase activity in the different layers of the pectoralis major muscle of the pigeon.

400 600

2

pg. of formazan per 100 mg. of
wet muscle per 15 minutes

266 J.C. GEORGE AND CHE TALESARA

‘muscle of the pigeon is shown in Table I. The results indicate that there exists a
relationship between the activity of the enzyme and the number of narrow fibers
per square mm. in the different layers of the muscle. It was also decided to find
out the activity of the enzyme in the 0.5-mm. thickness of the most superficial layers
of A and B where the broad fibers are relatively much more concentrated than in
the one-mm. thick layer (Table I).

From the results obtained the regression line is drawn using the equation

S = 0.3 + 11478 ¥-

where S stands for the succinic dehydrogenase activity and Y for the number of
narrow fibers in one square mm. area. If the value of Y is zero the value of S

IN jug. FORMAZAN

NUMBER OF NARROW SUCC.DEHYDROGENASE
FIB ERS/ mm

@) 2. 4 6 8 10
MUSCLE DEPTH IN mmm.

Ficure 3. Showing the number of narrow fibers and the corresponding succinic dehy-
drogenase activity in the different layers of the two regions A and B of the pectoralis major
muscle of the pigeon.

will be 0.3 wg. of formazan, which is an infinitely small measure of activity. If the
values obtained for the succinic dehydrogenase activity are plotted against the num-
ber of narrow fibers per square mm. a linear graph illustrating this relationship is
obtained using the equation of the regression line (Fig. 2). Similarly plotting the
dehydrogenase activity and the number of narrow fibers in the different layers from
regions A and B almost a similar pattern is obtained (Fig. 3). In the most
superficial region, which consists mostly of broad fibers, the succinic dehydrogenase
activity more or less corresponds to that of the number of narrow fibers.

DISCUSSION

From the data presented above it appears that there exists a relationship between
the number of narrow fibers and succinic dehydrogenase activity in the different
layers of the pigeon breast muscle. This relationship can be traced to the fact that

SDEVACTIVATY TN, PIGEON) PECBRORALIS 267

the activity of the enzyme in the broad fibers is negligible. By substitution of the
values in the formula mentioned above we obtain

S = 0.11478 (890.01 — 5.75 X),

(where S stands for the succinic dehydrogenase activity in pg. of formazan per
100 mg. of wet weight of the muscle taken from any region parallel to the surface
of the muscle, per 15 minutes under the conditions of our experiment mentioned
earlier, and X stands for the average number of broad fibers per square mm. area
of cross-section of the muscle). In other terms

S

(890.01 — 5.75X) a

(where K is a constant, the value of which is 0.11478). This indicates that S$
in any particular layer of the pigeon breast muscle is a function of the number of
narrow fibers. The conclusion that can be drawn from this study is that in the
pigeon breast muscle the main bulk of the succinic dehydrogenase is confined to
the narrow fibers and that the broad white fibers possess only a negligible con-
centration of the enzyme.

George and Scaria (1958) and George, Susheela and Scaria (1958) studied,
using histochemical methods, the activity of certain dehydrogenases in the breast
muscle of the pigeon, fowl and bat and the leg muscle of the fowl and frog, and
observed that the concentration of these oxidative enzymes has a relationship with
the color and the mitochondrial content of the individual muscle fibers in the sense
that the red narrow fibers possess a high concentration of these enzymes and
mitochondria, in sharp contrast to the broad white fibers. Similar observations
were made by Nachmias and Padykula (1958) in the skeletal muscle of rat. It
should be pointed out here that within the limitations of the methods employed in all
the above mentioned muscles, all the fibers in every muscle, except the broad
ones of the pigeon breast muscle, dehydrogenase activity was detectable. Never-
theless the activity was varying in the different muscle fibers according to their
diameter, highest activity being in the smallest fibers. In the pigeon breast muscle,
however, where the same histochemical methods were employed the broad white
fibers did not show the presence of any of the dehydrogenases mentioned earlier
(George and Scaria 1958). The quantitative data presented in this paper show
the distribution pattern of the succinic dehydrogenase activity in the different
layers of the pigeon breast muscle and also that the main bulk of the enzyme is
confined to the narrow red fibers.

SUMMARY

The relative distribution pattern of the narrow red and broad white fibers, and
the succinic dehydrogenase activity in the different layers of the pigeon breast
muscle were studied quantitatively. It was found that the activity of this enzyme
in any particular layer of the muscle is more or less related to the number of the
narrow red fibers present there since the main bulk of the enzyme resides in these

fibers.

268

GEORGE,

GEORGE,

GEORGE,

GEORGE,

GEORGE,

j. C. GEORGE AND Cy Le TALESARA

LITERATURE. CITED

J. C., ann R. M. Narx, 1958a. The relative distribution and the chemical nature of
the fuel store of the two types of fibres in the pectoralis major muscle of the pigeon.
Nature, 181: 709-710.

J. C, ann R. M. Narx, 1958b. Relative distribution of the mitochondria in the two
types of fibres in the pectoralis major muscle of the pigeon. Nature, 181: 782-783.
J. C, anp R. M. Nar, 1959. Studies on the structure and physiology of the flight
muscles of birds. 4. Observations on the fiber architecture of the pectoralis major
muscle of the pigeon. Biol. Bull., 116: 239-247.

J. C., ann K. S. Scarta, 1958. A histochemical study of the dehydrogenase activity
in the pectoralis major muscle of the pigeon and certain other vertebrate skeletal
muscles. Quart. J. Micr. Sci., 99: 469-473.

J. C., A. K. SusHeeLa anv K. S. Scart, 1958. Studies on the structure and physi-
ology of the flight muscles of bats. 3. Alkaline phosphatase and succinic dehydro-
genase activity in the breast muscle—a histochemical study. J. Anim. Morph. Physiol.,
5: 110-112.

Kun, E., ann L. G. Axsoop, 1949. Colorimetric estimation of succinic dehydrogenase by tri-

phenyl tetrazolium chloride. Science, 109: 144.

Nacumias, V. T., AND H. A. Papykuta, 1958. A histochemical study of normal and dener-

vated red and white muscles of the rat. J. Biophys. Biochem. Cytol., 4: 47-54.

fee BPP eCTS OF OXYGEN POISONING ON THE POST-
Pei YONIC DEVELOPMENT AND BEHAVIOR OF A
CHALCID« WASP

MARY HELEN M. GOLDSMITH? AND HOWARD A. SCHNEIDERMAN
Department of Zoology, Cornell University, Ithaca, N. Y.

Aerobic organisms survive only within a narrow range of oxygen tensions.
In very low oxygen, respiration becomes impossible; in excessively high oxygen
tensions, oxygen poisoning occurs. This report describes the effect of high pres-
sures of oxygen on the development and behavior of the chalcid wasp, Mormoniella
vilripennis. Few higher organisms are as suitable subjects for studies of oxygen
poisoning as insects. The design of their respiratory and circulatory systems is
such that insects are less prone to carbon dioxide accumulation and aeroembolism
(“bends”) than vertebrates. The small size of Mormoniella also tends to minimize
such complications. Further, Mormoniella is a particularly attractive experimental
animal because the histological details of its post-embryonic development are well
know (Tiegs, 1922).

In preliminary notes (Goldsmith and Schneiderman, 1956, 1958) we reported
that the sensitivity of Mormoniella to oxygen poisoning changed in a systematic
way during post-embryonic life. The present paper describes these changes in
detail and identifies the organ systems which are the main targets of oxygen
poisoning at successive stages in the life history.

MATERIALS AND METHODS
1. Experimental animals

Mormoniella vitripennis Walker is parasitic on pupae of muscoid flies. Tiegs
(1922) has described the developmental anatomy of this wasp; Whiting (1955),
and Schneiderman and Horwitz (1958) have recently reviewed its life-history.
The adult female pierces the puparium and lays her eggs on the developing fly
pupa. A few days later the eggs hatch and the larvae begin feeding on the host.
The cells of a larva grow in size throughout the several instars, but according to
Tiegs, there appears to be no larval cell proliferation. At 25° C. a larva feeds for
about four days and molts several times but is unable to defecate. Finally, it
ceases feeding and enters a resting stage during which the larval tissues begin to
break down; simultaneously the imaginal discs proliferate. After only a few cell
divisions the thin partition between the midgut and the invaginated rectum breaks
down, defecation occurs, and the greyish larva becomes white. During the 24
hours immediately before and after defecation, the so-called prepupal period, the

1 This investigation was supported by a research grant (H-1887) from the National
Heart Institute of the Public Health Service.
2 Present address: Biological Laboratories, Harvard University, Cambridge 38, Mass.

269

270 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

larval cells break down and the cell divisions necessary for the formation of the
adult integument, appendages, nervous system, and certain muscles occur. Except
for the thoracic and abdominal muscles, whose myoblasts continue dividing for
about 15 hours after pupation, adult development is a period of refinement, molding,
and differentiation of tissues and organs into the final imaginal form.

A time-table, descriptive of the development from egg to adult, is given in
Figure 1. Following Snodgrass (1935) and Hinton (1946, 1948, 1958), an
instar is considered ended when the epidermis retracts from the cuticle. In
Mormomella, hours or days may intervene between the time of detachment and
the actual ecdysis, which unveils the new instar. Thus the pupal stage actually
begins just prior to defecation when the cuticle of the last instar larva separates
from the epidermis (evidenced externally by wrinkles). The pupa itself is not

CUTICLE LAST LARVAL PUPAL CUTICLE
SHED CUTICLE SHED SHED.
2(2) DEFECATION | DEATH
LARVAL
INGRABS a nas | 4-20 HRS (2 pavee an NG: ADULT

I
!
| :
| fora
) fas

) | RESTING! \attlg

RED-BR.| BLACK
EYE HEAD

PUPA . PINK

eee RETRACTS

EPIDERMIS RETRACTS FROM PUPAL CUTICLE

FROM CUTICLE OF
PREVIOUS INSTAR

Oo 2-3 5-6 6-7 8 9 10 I 12 13 14 21-30
DAYS

Ficure 1. Time-table for the development of Mormomiella. The approximate times at
which the cuticle detaches from the epidermis (signaling the end of an instar) are indicated by
stippling.

revealed until the cuticle of the last larval instar is shed, an event occurring 24 to
48 hours later. The so-called prepupa is simply the pupa enclosed within the last
larval cuticle. Likewise, the pupal stage ends with the detachment and retraction
of the pupal cuticle, although the adult does not emerge for 4 or 5 days. During
this period the wasp, which is properly called a developing adult, is enclosed in two
cuticles—an inner adult cuticle and an outer detached pupal cuticle, which is shed
at eclosion. The exact time at which epidermis retracts from the pupal cuticle,
marking the end of the pupal period, has not been definitely determined for
Mormoniella. Our observations agree with those of Tiegs in suggesting that
this occurs within 24 hours after shedding the final larval cuticle.

Under certain conditions, just prior to entering the resting stage the larva
ceases development and enters diapause (Schneiderman, 1957; Schneiderman and
Horwitz, 1958). At room temperature diapause persists for a year or more

OXYGEN AND INSECT DEVELOPMENT Zi.

until the animal dies. If the diapausing larva is chilled at 5° C. for 3 months or
longer, diapause ends and development resumes after return to 25° or 30° C.
Chilled diapausing larvae may be stored in the cold for a year or more and
provide a convenient source of experimental animals.

Mormoniella were reared at 30° C. on puparia of the fly Sarcophaga bullata.
Since the wasp’s development is accompanied by changes in epidermal pigmentation
which are readily seen through its transparent pupal cuticle (Fig. 1), it was easy
to select animals at specific stages of development and to observe the progress of
development after experimental treatment. To assure that all insects used in a
single experiment were of the same age and were developing at the same rate,
only developing adults that reached a specified developmental stage within six
hours of each other were used in experiments. Unfortunately larval develop-
ment, unlike adult development, is not marked by gross external changes. Diapause,
however, always occurs at the end of larval life; consequently, all diapausing larvae
are at precisely the same stage of development. In the experiments to be de-
scribed, larvae which had entered diapause were collected and stored at 5° C.
for 4 to 8 months before use. These larvae initiated pupal development within
24 hours after return to 30° C. For the present investigation, animals at five
stages of development were used: (1) chilled diapausing larvae (equilibrated at
30° C. for one hour); (2) early prepupae (chilled diapausing larvae placed at
30° C. for 24 hours prior to the experiment and thus just beginning pupal de-
velopment) ; (3) “pink stage” developing adults (24 hours after ecdysis from the
final larval cuticle); (4) “black stage’ developing adults (12 to 24 hours prior
to adult emergence) ; and (5) adults (males and females). In each experiment
groups of about ten animals were placed in one-dram shell vials loosely plugged
with cotton. Several vials were compressed in each pressure chamber. After
exposure and decompression the vials were removed from the chambers and placed
at 30° C., where the progress of development was observed periodically until the
adults emerged or the insects died.

In every experiment, 25 to 50 insects were kept in air at atmospheric pressure
during the experimental period. These air controls indicated the normal rate
of development and percentage of adult emergence for a given population. The
adult emergence of these control insects varied between 75 and 100 per cent.
Therefore, following an experimental treatment only consistent reductions of
50 per cent or more in adult emergence were considered different from the control
values. To permit a comparison of the effects of different experimental conditions,
a number of groups of insects at the same stage of development were selected
from the same population, and each group exposed simultaneously in separate
chambers to specific experimental conditions. The data from each group were
compared with those of other groups in the same experiment. Each experiment
was repeated one to several times. The data presented in this paper are repre-
sentative of that amassed from almost 10,000 animals studied in a total of about
350 separate compressions (Goldsmith, 1955).

2. Compression and decompression

Commercial cylinders of compressed oxygen, nitrogen, and helium (Airco or
Matheson) were used. All gases contained less than 0.5 per cent impurities

272 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

(Schneiderman et al., 1953). The vials of wasps were sealed in transparent
polymethylmethacrylate (Lucite) chambers fitted with brass end-plates and needle
valves (Schneiderman and Feder, 1954) and compressed with a specific gas. In
all cases the gases were superimposed on the atmosphere of air initially in the
chamber. Throughout this paper all pressures are given as gauge pressure. The
final pressure in each chamber was checked on the same gauge, both after com-
pression and before decompression, and the insects from any chamber showing
greater than + 5 per cent variation in pressure during the experimental period
were discarded. The sensitivity of the gauges at the pressures employed was about
=— Fo Pemcetl

Compression was accomplished in one minute, whereas decompression was
performed stepwise over a period of five minutes. The pressure was reduced to
half in the first minute of decompression and then held constant during the second
minute. In the third minute, the pressure was gradually reduced to one fourth and
then again held constant during the fourth minute. During the final minute, the
pressure was gradually reduced to atmospheric.

The chambers were kept at a specific temperature in a thermoregulated water
bath. Most experiments were conducted at 30+ 1° C. and positive pressures
(gauge) of 5 atmospheres (atms.) oxygen. Under these conditions, adult wasps
survived for 1 or 2 hours. Hlence, the time required for ‘compressions ane
decompression occupied only a small fraction of the total exposure time.

RESULTS
1. The effect of pressure on Mormomiella

In each experiment, wasps at various stages of development were exposed to
positive pressures of nitrogen or helium to determine if elevated pressures of
metabolically inert gases had any toxic effects. As Table I shows, prolonged
exposure for 16 hours to 5 atms. of helium had, at most, a slight effect on survival
and development of any stage of the life cycle. Since 5 atms. of helium did not
impair the development of Mormomniella, we may conclude that this pressure per se
does not adversely affect Mormoniella. By comparison with either helium or
nitrogen ° the toxicity of 5 atms. oxygen is striking. For example, in experiment
No. 104, 78 per cent of the control chilled diapausing larvae ultimately emerged
as adults. After 16 hours of exposure to 5 atms. helium, a somewhat smaller
percentage emerged (/0 per cent), but after the same exposure to oxygen only
13 per cent enrer ved:

2. The effects of compression and decompression on Mormomiella

As Paul Bert (1878) first pointed out, the adverse effects of rapid decompres-
sion on vertebrates result in large measure from nitrogen being released from the
tissues too fast to be dissolved. Since Mormoniella is small in size and possesses
a tracheal system, rapid decompression might affect this insect differently from a
vertebrate. To test this, a total of 1,440 chilled diapausing larvae, early prepupae,

3 The survival and development of insects exposed to 5 atms. nitrogen was somewhat more
variable than that of insects subjected to helium, an effect which can be attributed to nitrogen
narcosis (Frankel and Schneiderman, 1958).

OXYGEN AND INSECT DEVELOPMENT PATS)

TABLE [|

Effect of exposing Mormoniella at specific stages of development to 5 atms. of He or Oz
at 30° C. for various periods

Per cent emerging

ISS3Ties INIG oo Sis ole Eee 104* 109* 114°** 115%**
nee of devel.
Duration of Chilled diapausing
exposure (hrs.) larvae Early prepupae “Pink stage’”’ “Black stage’’
Air (controls at atm.
pressure) 78 90 96 100
He 2 88 86 — —
4 94 85 — —
8 64 94 — ==
16 70 82 ==s 100
24 aad =e 56 88
32 = = 84 100
64 60 78 80 92
O» 1 72 Dil — =
2 73 5s 88 96
+ 70 54 80 94
8 63 65 40 0
16 13 20 0 —
24 aa == 0 —
64 aa 0 = =

* About 50 individuals used in each exposure.
** About 25 individuals used in each exposure.

“pink” and “black stage” developing adults were compressed with 10 atms. nitrogen
and then decompressed. The following procedures were used to study the effects
of rapid compression or decompression :

Group I: Controls—maintained at atmospheric pressure.
Group II: Thirty seconds for compression and thirty seconds for decom-
pression.

Group III: Thirty seconds for compression and five minutes for decompression
following the regimen outlined in the section on Methods.
Group IV: Five minutes for compression and five minutes for decompression.

In each group, two tanks were compressed and ten minutes elapsed between
compression and decompression. Regardless of the regimen, wasps exposed after
the start of the prepupal period invariably emerged as normal adults in about the
same time as air controls. By contrast, chilled diapausing larvae proved exceedingly
sensitive to compression. Though 85 per cent of air controls in this experiment
completed adult development, fewer than 40 per cent of the chilled diapausing
larvae emerged as adults when subjected to any of the above procedures. Be-
cause of the unusual sensitivity of this stage to compression and decompression,
only one successful experiment was conducted with some 750 chilled diapausing

274 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

TABLE II

Effect of exposing Mormoniella prepupae to Oz at 30° C. for various periods

Bxptie NO: Ase ocrers a ees 49* (10 atms.) 115** (5 atms.) 116** (5 atms.)

Seal Per cent emerging as adults

Air (control at atm.

pressure) 85 85 82
O» 3 90 aes ~
3 45 = ds

1 59 <= os

) == 50 44

+ 3D 40 44

8 = 46 40

16 0 + |

a2 = = 0

* Each figure is based on about 20 individuals.
** Fach figure is based on about 25 individuals.

MAXIMUM NON-TOXIC EXPOSURE (HRS)

C.D. E.PRE- "PINK "BLACK
LARVA PUPA STAGE" STAGE"

STAGE EXPOSED (DAYS FROM C.D. LARVA)

Ficure 2. The sensitivity of different developmental stages of Mormomiella to 5 atms.
Oo at 30° C. The maximum exposure that had no effect on adult emergence is plotted as a
function of developmental stage from chilled diapausing (C.D.) larva to “black stage.” Stages
withstanding only a brief exposure are the most sensitive to oxygen poisoning.

OXYGEN AND INSECT DEVELOPMENT DS

larvae (Exp. 104, Table 1). In this experiment, for reasons which are not clear,
the controls exposed to 5 atmospheres helium showed no evidence of an adverse
effect of pressure or decompression and compression.

3. Sensitivity of different developmental stages to oxygen poisoning

We have established that, excepting chilled diapausing larvae, pressure as such,
as well as compression and decompression, has only minor effects on the develop-
ment of Mormoniella, and can now consider oxygen poisoning itself. In the first
series of experiments we examined the effects of five atmospheres of oxygen on
various developmental stages. The toxic effects of oxygen were appraised in
terms of: (a) the rate of development of the wasps, (b) the percentage reaching
maturity and emerging, and (c) the duration of exposure necessary to halt all
development completely. Tables I and II give the effects of exposure to oxygen
at several developmental stages on the ultimate emergence of the adults. Figure 3
and Table II record the final stage of development reached by wasps exposed to
oxygen at a specific developmental stage.

To measure the inhibition of adult emergence, we determined for each de-
velopmental stage the maximal exposure time that had no effect on emergence. It
is clear that this insect’s sensitivity to oxygen changes markedly during development
(Fig. 2). Chilled diapausing larvae and “pink stage’? wasps were the least
sensitive and tolerated more than 8 hours of oxygen. Early prepupae were many
times more sensitive and could barely withstand one hour of oxygen. Thus,
during the 24-hour period in which the chilled diapausing larva transformed into
an early prepupa, sensitivity to oxygen increased many-fold. Likewise, after the
“pink stage,’ sensitivity progressively increased during adult development, reaching
its maximum in the adult wasp which could tolerate only one-sixth the exposure
of a “pink stage’ wasp.

4. Oxygen poisoning in chilled diapausing larvae and early prepupae

The above experiments disclosed that chilled diapausing larvae were far less
sensitive to oxygen than early prepupae. To reduce adult emergence of chilled
diapausing larvae by 50 per cent, an exposure to oxygen for 8 to 16 hours was
required; for early prepupae, however, an exposure of less than one hour at 5
atms. must have been sufficient (Table I). Since at 10 atms. an exposure of
20 minutes had no effect on the subsequent emergence of prepupae (Table II,
Experiment 49), it may be inferred that at 5 atmospheres of oxygen an exposure
between 20 minutes and one hour would be required to reduce emergence by half.

Curiously, exposing early prepupae to 5 atms. oxygen for 2, 4 and even 8 hours
had no greater effect on the subsequent emergence of adults than did one hour of
exposure (Table I). Thus, although the early prepupa was the most sensitive
developmental stage studied, this sensitivity was manifest in only half the popula-
tion. The other half showed a sensitivity comparable to chilled larvae. These
results were confirmed in numerous experiments with early prepupae (for example,
Table II, Exp. 115 and 116). The explanation of this phenomenon is not clear ;
there is no marked difference in sensitivity of males and females, nor does this
result stem from effects of compression or decompression. A likely explanation

276 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

is that the resistant members of the population were slower in initiating pupal
development and at the time of exposure to oxygen were still in a stage comparable
to the insensitive chilled diapausing larva.

Eight hours’ exposure to oxygen caused a retardation in the rate of development
of both chilled diapausing larvae and early prepupae; both lagged behind the con-
trols by 1 or 2 days. Though adult emergence was reduced, prolonged exposure
to oxygen failed to cause an arrest in development comparable to diapause. In
fact chilled diapausing larvae and early prepupae defecated, and many continued to

100

°

80

eoeeeeee

60

(EXPOSED AS PREPUPAE)

40

20

Oo !
DEVELOPMENT BETWEEN DEFECA- BETWEEN "PINK FULLY EMERGED
NONE TION AND LAST STAGE" AND ADULT
OR SLIGHT LARVAL MOLT “BLACK STAGE"

PER CENT CEASING DEVELOPMENT

STAGE AT WHICH DEVELOPMENT CEASED AFTER EXPOSURE TO OXYGEN

Ficure 3. The stage of development attained by early prepuae following exposure to
5 atms. oxygen at 30° C. About 50 early prepupae were subjected to each exposure. The
duration of exposure is indicated above each bar. Wasps which completed the larval-pupal
transformation after exposure to oxygen also successfully completed development and emergence,
and almost none of them ceased to develop at stages intermediate between emergence and the
larval-pupal transformation. A similar result was also obtained in the experiment with
chilled diapausing larvae. For early prepupae, the effect of exposures between 1 and 8 hours
duration was not significantly different from an 8-hour exposure.

develop after exposure which markedly reduced adult emergence. As exposure
times were increased to 16 hours, fewer animals developed beyond defecation; an
exposure somewhere between 16 and 32 hours prevented all further development
(Fig. 3). Examination showed that those animals that failed to emerge as adults
ceased developing before or soon after defecation. Thus treating larvae and
prepupae with oxygen blocked the developmental changes which immediately
precede the transformation from larvae to pupae; animals that successfully pupated
almost invariably completed adult development and emerged (Fig. 3).

OXYGEN AND INSECT DEVELOPMENT 2

5. The effects of oxygen on “pink stage’ developing adults

Approximately 80 per cent of the wasps exposed in the “pink stage” to 5 atms.
oxygen for 8 to 16 hours appeared to complete adult development (1.¢., reached the
“black stage”) (Table III). At these doses then, increasing the length of exposure
to oxygen had no effect on the normal differentiation and pigmentation of the
epidermis and appendages. That the longer exposures in this range were toxic,
however, was revealed by a decrease in the ability of the adults to complete
emergence. Thus after exposures in the “pink stage’ for up to approximately
12 hours, most of the resulting “black stage’ wasps emerged; however, after 14 to
16 hours only a fraction of these “black stage’ wasps escaped completely from their
pupal cuticles (Table III). This is in striking contrast to results obtained with

TABLE III
Development of “pink stage’ wasps after exposure to 5 atms. of Oz at 30° C.

Per cent of wasps exposed*

. Ceasing development at

Exposure
Expt. No. a ;
iss.) Lo eee Completing
4 sf emergence
“Pink Stage” | Tnystmediate stages (etmegn | evelopment
300* 8 iS) 4 81 Ta
300 10 14 0 86 65
300 12 19 0 81 19
a2" i Not determined Not determined 80 60
301* 14 Not determined Not determined 85 23
302 14 Not determined Not determined 82 23
300 16 45 38 ty 0
301 16 Not determined Not determined 73 7
302 16 Not determined Not determined 79 13
114 24 100 0 9) 0

* About 30 individuals used in each exposure in experiment 300; about 25 individuals used
in each exposure in experiment 301 and 302.
** “Black stage,’”’ partially emerged adults, and adults.

chilled diapausing larvae and prepupae where most animals that failed to emerge
also failed to complete adult development (Fig. 3). Exposing “pink stage”
wasps to oxygen for longer than 16 hours commonly prevented further develop-
ment, and the animals remained in the “pink stage” until death (Table III).

The results obtained with the “pink stage’ wasps suggest that although ex-
posures up to about 16 hours of oxygen fail to inhibit epidermal differentiation
during adult development, some internal system necessary for ecdysis is damaged.
Since proper functioning of muscles and nerves is necessary for emergence, and
since the development of the muscular system can be easily observed, the effects of
oxygen on this system were studied. This proved a fortunate choice.

278 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

To study the extent of muscle development, wasps that completed adult de-
velopment were immersed in Zenker’s fixing solution. After 24 hours, the animals
were dissected and their thoracic muscles examined. In normal “black stage”
wasps and in adults, the thoracic muscles are grouped in 5 longitudinal and 5
dorsosternal pairs. These thick white bands almost fill the thorax and are the
most prominent tissue in this part of the animal. Oxygen had a striking effect

100
AIR CONTROL

16 HRS.

80

60

40

"BLACK STAGE" ABLE TO EMERGE

Q@ Oe urs. 0,

20

PER: <CENI = OF

O 20 40 60 80 100
PER CENT OF "BLACK STAGE" WITH NORMAL MUSCLES

Ficure 4. The percentage of “black stage’ adults which emerge as a function of the
percentage which develop normal thoracic muscles. In two experiments (Nos. 301 and 302;
see also Table III), “pink stage” wasps of uniform age were exposed to oxygen (5 atms. at
30° C.) for periods that allowed most of them to differentiate to the “black stage.” The
duration of exposure indicated in the figure varied from a period so brief that almost all
adults emerged, to a period so long that no adults emerged. After exposure, the “pink stage”
insects were divided into two groups; in one group, the emergence of wasps was observed
while in the other group each animal that completed adult development was dissected and its
thoracic musculature examined. About 110 insects were dissected.

on the development of these muscles. In “black stage’? wasps unable to emerge
because of oxygen poisoning in the “pink stage,” the thoracic musculature ranged
from thin strands of differentiated muscle to a yellowish, jelly-like mass with no
distinguishable strands of muscle. The amount of differentiated muscle present
was easily recorded as none, small, moderate, or normal.

In the two experiments recorded in Figure 4, it is clear that as oxygen exposure
increased from 12 to 16 hours, the decrease in number of adults able to emerge was

OXYGEN AND INSECT DEVELOPMENT 279

paralleled by a decrease in the number of adults that had normal muscles. As
previously mentioned, the percentage of animals that completed external adult
development after 12, 14, and 16 hours of oxygen was the same (Table III)
whereas the percentage of animals that emerged decreased steadily between 12
and 16 hours of oxygen exposure (Fig. 4). After 12 hours of exposure, all of the
wasps developed at least a few strands of thoracic muscles. After 16 hours of
exposure, however, only 55 per cent of the wasps had any detectable strands (of
muscle), and the remainder had only a jelly-like mass in place of the thoracic
muscles. Thus, with longer exposures, not only did fewer wasps develop normal
thoracic muscles, but the amount of muscle development occurring in animals that
failed to emerge also decreased. In our opinion, inability of the oxygen-treated
“pink stage” wasps to emerge can be adequately accounted for by the observation
that oxygen impaired the development of adult muscles.

6. The effects of oxygen on “black stage” developing adults

In most experiments, 2 to 4 hours’ exposure to 5 atms. of oxygen at 30° C.
caused a detectable decline in the emergence of insects exposed during the “black
stage.” Several lines of evidence indicate that these wasps were not dead but
paralyzed: (a) 24 to 48 hours after exposure of less than 8 hours duration, at the
time when they normally emerged as adults, many of the animals moved their
appendages weakly; (b) after exposures longer than 8 hours, no movement was
observed, but the molting fluid was partially resorbed; (c) animals exposed as
long as 8 hours retained a healthy, shiny, black appearance for at least five days,
whereas animals killed by ether or alcohol fumes soon became dry, shriveled and
dull in appearance. To show that the immobile wasps were still alive, the respira-
tory rate of several groups of animals was measured before, immediately after, and
at intervals of 24 hours following exposure to five atms. of oxygen at 30° C. for
various periods. All groups were the same age when oxygen uptake was measured.
Since there was no way of knowing how many of the animals in each group were
dead or dying as a result of treatment with oxygen, the data summarized in
Figure 5 refer to the average oxygen uptake of a group and may not accurately
mirror the respiratory behavior of any individual.

The oxygen uptake of the controls was recorded until the adults began to
emerge. As seen in Figure 5, the oxygen consumption of the air controls rose
sharply on the day of emergence. In the group exposed to oxygen for 4 hours,
respiration dropped slightly immediately after treatment, but by the next day had
risen to the level of the control. Although adults began to emerge during the
next 24-hour period, fewer than 10 per cent completed emergence. Thus, al-
though respiration returned to normal levels, the paralytic symptoms of oxygen
poisoning remained. Immediately after 5 hours of exposure, oxygen consumption
was only slightly less than after 4 hours of exposure. The initial effects of 8 hours
of exposure were more dramatic, and respiration almost ceased. However, 24
hours after exposure to oxygen for 5 or 8 hours, respiration had noticeably in-
creased. This increase in oxygen uptake continued until the second day after
exposure. In the case of 5 hours of exposure to oxygen, respiration returned to
the pre-treatment level, and yet none of the insects ever emerged. After the
second day, respiration gradually declined, probably signifying that the insects

280 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

2

100% ADULTS EMERGE

ca. |O°% ADULTS EMERGE

O ADULTS EMERGE

OXYGEN CONSUMPTION (microl./mgm./hr.)

O ADULTS EMERGE

Gh l 2 3 4
BEFORE IMMEDIATELY
EXPOSURE AFTER
EXPOSURE

DAYS AFTER EXPOSURE

Ficure 5. The effect of 5 atms. Oe at 30° C. on the respiration of “black stage” wasps.
Groups of 50 wasps were enclosed in perforated gelatin capsules and placed in 20-ml. Warburg
vessels, the center wells of which contained 2% KOH. Oxygen uptake was measured im-
mediately before and after exposure of wasps to 0, 4, 5, and 8 hours of Os. Following this
the respiration was measured daily until adults began emerging or until the gradual decline
in respiration indicated that poisoned animals were dying.

were dying. Similar results were obtained in each of four separate experiments
involving a total of 1100 “black stage” animals. That animals subjected to a
short treatment regained a respiration comparable to controls but still failed to
emerge, suggests that the system poisoned by threshold oxygen exposure accounts
for only a small fraction of the total respiration of the insect.

7. Oxygen poisoning of adults

About 600 adult wasps of known ages were compressed with oxygen to various
pressures at 30° C. Males and females alike exhibited successive stages of
oxygen poisoning similar to those described by Williams and Beecher (1944) for
Drosophila:

motor activity increased and then gradually declined ;
walking ceased but a standing position was maintained ;
the ability to right and to stand disappeared ;

all spontaneous movements ceased ;

reflex excitability disappeared.

wm Wd

OXYGEN AND INSECT DEVELOPMENT 281

Since it was difficult to determine the difference between “slight”? movement and
“no”? movement, the time required for loss of the righting reflex was chosen as a
convenient measure of oxygen poisoning. Ina group of ten adults, the last insect
to lose its righting reflex usually did so no more than 5 minutes after the first
animal had lost its reflex. Hence it made no appreciable difference in the results
whether we measured the exposure time which caused loss of righting reflex in
50 per cent or 100 per cent of the animals.

Figure 6 plots the time for loss of righting reflex as a function of oxygen
pressure. In 10 atms. oxygen, adult Mormomiella lost their righting reflex after
about one-half hour. Poisoning was 8 times faster at 10 atms. than at 3.3 atms.
Even at 1.6 atms., the lowest oxygen pressure used in any experiment, poisoning
occurred. The relation between pressure and duration of exposure for loss of

5
ae
ea
o
oe
oO
a3
ep)
i.e)
—
o
= C)
ae
®
)
) 2 4 6 8 lo I2

ATMOSPHERES OXYGEN

Ficure 6. Time for loss of righting reflex (R.R.) in 100 per cent of 1- to 2-day-old
adult wasps as a function of oxygen tension at 30° C. Twenty adults were exposed at each
oxygen tension.

righting reflex appears to be logarithmic, similar to findings of Williams and
Beecher (1944) on Drosophila.

Both temperature and age affected the sensitivity of adults to oxygen (Fig. 7).
Thus the time required for poisoning decreased at higher temperatures; at 10 atms.
and 20° C. adults lost their righting reflex in about 60 minutes, whereas at 10
atms. and 30° C. they lost their righting reflex in 20 to 30 minutes. Furthermore,
old animals were more sensitive than young ones. Thus O- to 1-day-old wasps
lost their righting reflex in approximately 60 minutes, 1- to 2-day-old wasps in
about 40 minutes, and 4- to 5-day-old wasps in about 30 minutes (cf. Williams
and Beecher, 1944).

Adults removed from oxygen promptly after the loss of righting reflexes
generally recovered completely within 24 hours. But animals kept in oxygen for

282. MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

TA8

60
a“
a
ti (oO
Ww
S
e Mac
SS
oO
wo
eso
(2)
ine
20
=
=
=
=

O | rz 5) 4 S
AGE OF ADULTS — (DAYS)

Ficure 7. Time for loss of righting reflex (R.R.) at 10 atms. Os and at two temperatures
as a function of age of the adult wasp. Poisoning is more than twice as rapid at 30° C. as
at 20° C. At least 50 adults of each age were used at 20° C.; 30 adults of each age were
used at 30° C.

as little as 10 minutes after the righting reflex was lost usually recovered only
partially and failed to regain normal coordination and motor ability.

DISCUSSION
1. The sites of oxygen poisoning in Mormoniella
A. Developmental processes involved

The present experiments reveal that oxygen can have a variety of toxic effects
on Mormomiella: it can prevent pupation, it can prevent imaginal muscle develop-
ment, it can paralyze adults. These different effects on various stages of the life
cycle can be understood best by considering the processes occurring at each de-
velopmental stage which are likely to be affected by oxygen. We know from
ligature experiments on chilled diapausing larvae that at this stage the neuro-
secretory cells of the insect’s brain have not secreted brain hormone and initiated
the events leading to pupal development (Malamud, 1958). Thus, for the first
few hours after their return to room temperatures, chilled larvae are engaged in

OXYGEN AND INSECT DEVELOPMENT 283

minimal developmental activity, a quiescence which sharply contrasts with the
larval-pupal transformation occurring 24 hours later. During the prepupal period
the cells of the integumentary imaginal discs divide and spread rapidly, and the
discs of the legs, wings and antennae grow out of their sacs. This activity is
followed by defecation, whereupon the appendages evert, and a distinct head,
thorax and abdomen become visible. Shortly thereafter the last larval cuticle is
shed and the antennae, legs and mouth parts promptly attain their full length and
adult aspect. During adult development itself the epidermis produces a new
cuticle, with ridges, grooves and bristles, which gradually becomes pigmented.

It is noteworthy that as far as external features are concerned, the most marked
transformation occurs in the prepupa, not in the developing adult. Furthermore,
the prepupal and very early pupal stages are the periods of greatest development
of the muscular and nervous systems and of the proliferation of the imaginal discs.
Thus, most of the migration and fusion of ganglia in the nerve cord and brain
which occurs in the formation of the adult nervous system is completed after the
last larval molt and before the “pink stage.’ Likewise, during the prepupal and
early pupal stages, myoblast divisions are completed in the thorax, and presumptive
muscle cells begin to fuse. This leads to the appearance, just prior to the “pink
stage,” of five, thin, syncytial dorsosternal columns of undifferentiated muscle.
During the “pink stage” these syncytial columns continue to increase in size, and
the cells involved in their insertions differentiate. According to Tiegs (1922),
the development of the adult abdominal and leg muscles is similar. Obviously,
the normal differentiation of the nervous and muscular systems is necessary if the
adult is to emerge.

B. Effects on larvae and prepupae

In terms of these developmental processes, it appears that the major difference
between the oxygen-sensitive early prepupa and the oxygen-insensitive chilled
diapausing larva and “pink stage” (Fig. 2) is the intense proliferation of imaginal
discs in the prepupa. In the chilled diapausing larva, this process has not yet
begun, and imaginal cells are comparatively inactive; in the “pink stage,” prolifera-
tion has already been accomplished. Animals with actively dividing cells appear
more sensitive to oxygen poisoning than animals in which the division of the
imaginal cells has not begun or has just been completed.

C. Effects on “pink stage’ wasps

The major morphogenetic processes during adult development that determine
the ability of the adult to emerge are the differentiation of the muscular and
nervous systems. Preventing the differentiation of either muscle or nerve or
damaging these systems after differentiation would prevent emergence. The
present experiments reveal that oxygen poisoning in the “pink stage’ interrupts
muscle differentiation and so prevents adult emergence.

The failure of muscle development in “pink stage” wasps exposed to oxygen
could be the result of direct damage to the developing muscles. However, it is
worth recalling that several investigators have reported that adult muscle develop-
ment in Lepidoptera is prevented by removal of the pupal nervous system, al-

284 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

though development into an otherwise normal adult proceeds (Nutesch, 1952;
Williams and Schneiderman, 1952). These observations confirm the original
finding of Kopeé (1923) that deganglionation of thoraces of Lymantria dispar
prepupae prevented the development of thoracic muscles. Kopec also found that
adult development of the integument, as well as the external adult appearance of
wings and legs, was not affected by deganglionation. From these results, one
might anticipate that if oxygen poisoning severely damaged the nervous system of
the “pink stage” Mormoniella, muscle differentiation would cease. Whether the
failure of muscle differentiation after oxygen poisoning is caused by a direct action
on the muscle itself or by an indirect inhibition that disrupts the trophic role of the
nervous system remains to be decided.

D. Paralytic effects on “black stage’ developing adults and adults

During adult development the neuromuscular system appears to become more
sensitive to oxygen poisoning than in the young “pink stage.” Thus, when de-
velopment was allowed to proceed beyond the “pink stage” before oxygen was
applied, oxygen sensitivity progressively increased until the “black stage,’ where
even brief exposures prevented emergence. It is of special interest that although
these “‘black stage’ wasps were unable to emerge, their respiration recovered and
was comparable to the controls. The animals were apparently paralyzed and
unable to escape from their pupal cuticles. The observation that this paralysis
had no effect on the insect’s respiration suggests that either (a) the system first
affected by oxygen is responsible for only a small fraction of the total respiration
of the animal, or, alternatively, that (b) the respiratory activity of the affected
system was not severely damaged. .(For interpretation of similar effects in
vertebrates see Stadie and Haugaard, 1946.) After exposures to oxygen for 8
hours, the respiration of “black stage” wasps immediately declined, never to return
to normal. This permanent decrease in oxygen uptake, evoked by prolonged
exposure to oxygen, suggests that the systems of cellular respiration suffered
irreparable damage. The inhibition by oxygen poisoning of the respiration of
developing insects has also been reported by Clark and his colleagues (Clark and
Papa, 1958; and Clark, Harwitz and Rubin, 1958). These workers examined
the effects of one minute of exposure to two atmospheres of oxygen on the
respiration of the braconid wasp Habrobracon. Not unexpectedly, this brief
oxygen exposure had no effect on “pigmented Habrobracon pupae,’ which cor-
respond to the “black stage” Mormoniella. However, in stages whose subsequent
development was adversely affected by these brief oxygen exposures, respiration
was immediately inhibited. Clark and his colleagues do not report a recovery of
respiration such as we found.

The adult Mormoniella was paralyzed by oxygen in much the same way as
the “black stage’ wasp. The experiments revealed that increasing exposure to
oxygen caused an increasing loss of motor activity and coordination. Here again,
either the nervous or muscular systems might be the site of oxygen damage.

It seems more likely to us that oxygen acts on the nervous system rather than
directly on the muscle. Such a view is consistent with the fact that in oxygen
poisoning of vertebrates, when pulmonary damage is avoided, the nervous system
is the system first affected and most sensitive to oxygen. Further support for this.

OXYGEN AND INSECT DEVELOPMENT 285

opinion stems from the findings, already discussed, that oxygen poisoning of
“pink stage’ wasps has the same effect on muscle development as deganglionation.

2. Differences in the sensitivity of various organ systems during development

From the foregoing analysis it seems likely that the differences in sensitivity
of chilled diapausing larvae, early prepupae, and “pink stage’? developing adults
(Fig. 2) reflect an increased sensitivity of dividing cells to oxygen. It also seems
probable that the gradual increase in sensitivity to oxygen which begins during
adult development and continues uninterrupted during adult life reflects an in-
creasing sensitivity of the nervous system to oxygen poisoning. It is interesting
to speculate that impulse conduction and transmission are the functions disrupted
by threshold oxygen poisoning of “black stage” and adult wasps, while the trophic
influence of nerve on muscle is the function impaired in early stages of adult
development.

It is noteworthy that exposure to oxygen of “pink stage’? wasps may inhibit
muscle development, and yet have no effect on the further differentiation of the
integument and appendages. This indicates that in the “pink stage” the integu-
ment is not as sensitive to oxygen poisoning as is muscle. However, it is im-
portant to emphasize that there are times in development when the epidermal
differentiation is sensitive to oxygen poisoning. Thus, when early prepupae were
exposed to oxygen, many animals failed to develop beyond defecation. This was
also true of the less sensitive, chilled diapausing larvae. The larval cuticle was not
shed, and the appendages, head, thorax, and abdomen of the adult insect failed to
form. We can scarcely attribute the lack of development at this stage to myoblast
or nerve cell damage. Failure to develop in the chilled larva and prepupa ap-
parently reflects a failure to the epidermal imaginal discs to fulfill their role.

Differences in the oxygen sensitivity of the various developmental stages of
insects have also been reported by Clark and his co-workers (Clark and Herr,
1954). In Habrobracon they report that emergence of older prepupae and white
pupae was reduced by exposure to one atmosphere of oxygen for one hour at
30° C. This same exposure had no effect on younger larvae-in-cocoons and little
or no effect on developing adults. Recently, Clark and Papa (1958) have made
the remarkable discovery that exposure to two atmospheres of oxygen for five
seconds arrested development of 100 per cent of white pupae of Habrobracon!
From their results it appears that this stage is exceedingly sensitive to oxygen
poisoning. In our experiments, higher pressures and much longer exposures
(10 atms. for 40 minutes, Table II) were required to inhibit adult development
and emergence of Mormoniella exposed as early prepupae. (In our experience,
the early prepupa is the stage most sensitive to oxygen poisoning.) Although we
have not studied the effects of oxygen on Mormoniella immediately after the last
larval molt, the stage which corresponds to Clark’s white pupae, it seems most
likely that Habrobracon and Mormoniella differ considerably in their absolute
sensitivity to oxygen poisoning.

In their studies of oxygen poisoning during adult development, Clark and his
co-workers did not report a rise in sensitivity during adult development (as
reflected in the failure of adults ultimately to emerge), as was found in the case
of Mormoniella. This is not unexpected, since the exposures they employed were
much lower than the ones used in the present experiments.

286 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

3. A comparison of the sensitivity to oxygen poisoning and x-irradiation

Gerschman et al. (1954) suggest that x-irradiation and oxygen poisoning have
a common mode of action, both agents causing the formation within cells of large
amounts of peroxide and free oxidizing radicals such as OH and OH,. They
support this view with the finding that exposure to high pressures of oxygen after
x-irradiation acts synergistically with x-irradiation in decreasing the survival of
mice, and that reagents which protect organisms from radiation damage also
protect mice from lethal effects of oxygen. The argument gains further support
from the fact that oxygen can mimic many of the effects of x-rays, e.g. chromosome
breakage (Conger and Fairchild, 1952). During the development of Mormoniella,
however, the sensitivity to x-irradiation (Kuten, 1955; Schneiderman e¢ al., 1956,
1957) does not parallel the sensitivity to oxygen. Although the effects of oxygen
poisoning and x-irradiation on Mormoniella are strikingly similar during the late
larval and pupal stages, during adult development sensitivity to oxygen gradually
increases while sensitivity to x-irradiation steadily decreases. In “pink stage”
wasps, 4000 r reduced adult emergence by half; in the “black stage” doses as high
as 64,000 r failed to inhibit emergence at all (Schneiderman et al., 1956). Simi-
larly, Clark and Mitchell (1951) showed that resistance of Habrobracon to
x-irradiation increases during adult development. Apparently during adult de-
velopment systems involved in emergence of the adult wasp become increasingly
resistant to x-irradiation while becoming more sensitive to oxygen poisoning.
Thus, although x-irradiation and oxygen poisoning may have a similar mode of
action on dividing and undifferentiated cells, oxygen has a specific effect on the
adult which is not produced by x-irradiation. If oxygen poisoning, like x-irradia-
tion, leads to the formation of free radicals, it is not clear why oxygen has such a
toxic effect on adults, while x-irradiation does not.

The type of analysis employed in the present study permits us to define the
systems which are the target of oxygen and to recognize that the sensitivity of a
developing system can change during its differentiation. Thus although gaps
exist in our understanding of the biochemistry of oxygen toxicity, oxygen offers
itself as a useful tool to the experimental morphologist (cf. Nelsen, 1950; Malamed,
1958). Moreover, for the student of insect growth, the fact that oxygen poisoning
appears to provoke many of the same developmental consequences as degangliona-
tion suggests new approaches to investigations of the trophic function of nerve.

We wish to thank Professor Carroll M. Williams for his most helpful comments
on the manuscript, and Dr. Timothy H. Goldsmith for his interest in discussing
many aspects of this work and care in reading the manuscript. Dr. Williams
kindly provided equipment and space for the experiments on muscle development
in “pink stage’’ wasps.

SUMMARY

1. Experiments were conducted to determine the effects of high pressures of
oxygen on the development and behavior of an insect, the chalcid wasp, Mor-
moniella vitripennis.

2. More than 10,000 wasps in five different stages of post embryonic develop-
ment were compressed with 5 and 10 atmospheres of oxygen, nitrogen, or helium.
Rapid compression and decompression were detrimental to chilled diapausing
larvae but not to other stages. Although the high pressures of helium or nitrogen

OXYGEN AND INSECT DEVELOPMENT 287

had no, or only slight, adverse effect, high pressures of oxygen were exceedingly
toxic at certain stages of development.

3. The chilled diapausing larva was but slightly sensitive; about 12 hours at
5 atmospheres oxygen was required to reduce adult emergence by 50 per cent.
Sensitivity reached a maximum during the early prepupal stage, when less than
one hour of exposure to 5 atmospheres of oxygen reduced adult emergence by
50 per cent. Sensitivity decreased after pupation, and in the “pink stage,” at the
outset of adult development, about 14 hours of exposure were necessary to reduce
adult emergence by 50 per cent. During the rest of adult development and
throughout imaginal life, the sensitivity to oxygen again increased. The organ
systems attacked by oxygen during these different stages were examined.

4. Exposing “pink stage’ wasps to 5 atmospheres of oxygen for about 16
hours stopped development in the “black stage’? just prior to adult emergence.
Dissection of these “black stage’ wasps revealed that they failed to emerge because
oxygen poisoning had prevented their thoracic muscles from differentiating. The
possibility that this block to muscle development resulted either from direct injury
to the muscle cells or from damage to the trophic function of the nervous system
was discussed, and an action on the nervous system was favored.

5. Exposure of “black stage” wasps just prior to eclosion caused an immediate
and abrupt decline in respiration. Even when respiration returned to control
levels within a day, the wasps remained paralyzed within their pupal cuticles and
never emerged.

6. The first visible sign of oxygen poisoning in adults was loss of motor
activity and coordination. After brief exposures, wasps often recovered com-
pletely, but longer exposures evoked permanent paralysis.

7. It was suggested that in the early stages of adult development from the
“pink stage” onward, the trophic influence of nerve on muscle is disrupted by
oxygen; and that just prior to emergence, and in the adult, impulse conduction and
transmission are prevented by oxygen. The gradual increase in oxygen sensitivity
during adult development and imaginal life appears to reflect an increasing
sensitivity of the nervous system to oxygen poisoning.

8. In contrast to the oxygen-sensitivity of the nervous system, the epidermis
of the developing adult was highly resistant to oxygen poisoning. However,
prior to the initiation of adult development, epidermal differentiation was blocked
by oxygen poisoning. The extreme sensitivity of the early prepupa compared
with chilled diapausing larva or the “pink stage” wasp probably reflects the high
sensitivity of dividing epidermal cells.

LITERATURE, CITED

Bert, P., 1878. La Pression Barometrique, Recherches de Physiologie Experimentale.

Criark, A. M., Anp E. B. Herr, Jr., 1954. The sensitivity of developing Habrobracon to oxy-
gen. -Biol. Bull., 107: 329-334.

CriarkK, A. M., Ann C. J. Mircnett, 1951. Radiosensitivity of haploid and diploid Habrobracon
during pupal development. J. Exp. Zool., 117: 489-498.

Crark, A. M., Anp M. J. Papa, 1958. Some effects of oxygen on the white pupae of Habro-
bracon. Biol. Bull., 114: 180-187.

Criark, A. M., G. A. Harwirz ann M. A. Rusin, 1958. Sensitivity to oxygen during post-
embryonic development of the wasp Habrobracon. Science, 127: 1289-1290.
Concer, A. D., ann L. M. Farrcuitp, 1952. Breakage of chromosomes by oxygen. Proc.

Nat. Acad. Sct., 38: 289-299.

288 MARY HELEN M. GOLDSMITH AND HOWARD A. SCHNEIDERMAN

FRANKEL, J.. AND H. A. SCHNEIDERMAN, 1958. The effects of nitrogen, helium, argon and
sulfur hexafluoride on the development of insects. J. Cell. Comp. Physiol., 52: 431-451.

GERSCHMAN, R. D., L. GrvBert, S. W. Nye, P. Dwyer ann W. O. Fenn, 1954. Oxygen
poisoning and x-radiation: a mechanism in common. Science, 119: 623-625.

GotpsmitH, M. H. M. (née M. H. Martin), 1955. Studies on the mechanism of oxygen
poisoning in insects and the effect of carbon dioxide on oe Thesis, Cornell
University.

GotpsmitH, M. H. M., anp H. A. SCHNEIDERMAN, 1956. Oxygen poisoning in an insect.
Face Rec... 125-7560.

GOLDSMITH, M. i. M., AnD H. A. SCHNEIDERMAN, 1958. Oxygen poisoning in an insect.
eae. 10th intone: Cong. Ent., 2: 337.

Hinton, H. E., 1946. Concealed yresee in the metamorphosis of insects. Nature, 157: 552-
BES

Hinton, H. E., 1948. On the origin and function of the pupal stage. Trans. Roy. Ent. Soc.
London, 99: 395-409.

Hinton, H. E., 1958. Concealed phases in the metamorphosis of insects. Sci. Progress, 46:
260-275.

Koperc¢, S., 1923. The influence of the nervous system on the development and regeneration
of muscles and integument in insects. J. Exp. Zool., 37: 15-23.

Kuten, J., 1955. Effect of x-irradiation upon the metamorphosis of the chalcid wasp,
Mormoniella vitripennis. Unpublished thesis, Cornell University.

Matamep, S., 1958. Gastrular blockage in frogs’ eggs produced by oxygen poisoning. JBvol.
Bull. 114: 226-246.

MALAMUD, JEAN G., 1958. An analysis of the mechanism of diapause termination in Mor-
monella vitripennis (Walker) by ligature and other techniques. Unpublished thesis,
Cornell University.

Netsen, O. E., 1950. Temperature and oxygen pressure inhibition of the early development
of Rana pipiens. Anat. Rec., 105: 599.

Ntescu, H., 1952. Uber die Einfluss der Nerven auf die Muskelentwicklung bie Telea poly-
phemus. Rev. Suisse Zool., 59: 294-301.

SCHNEIDERMAN, H. A., 1957. Onset and termination of insect diapause. Physiological Trig-
gers (Ed. T. H. Bullock). American Physiological Society. Pp. 46-59.

SCHNEIDERMAN, H. A., AND N. FepeEr, 1954. A respirometer for metabolic studies at high
gaseous pressures. Biol. Bull., 106: 220-237.

SCHNEIDERMAN, H. A., anp J. Horwitz, 1958. The induction and termination of facultative
diapause in the chalcid wasps Mormomniella vitripennis (Walker) and Tritneptis klugu
(Ratzeburs). “J. aps Biol 352 520-55:

SCHNEIDERMAN, H. A., M. KetcHet anp C. M. Witiiams, 1953. The physiology of insect
diapause. VI. Effects of temperature, oxygen tension and metabolic inhibitors on
in vitro spermatogenesis in the Cecropia silkworm. Biol. Bull., 105: 188-189.

SCHNEIDERMAN, H. A., J. KuTEN Anp J. Horwitz, 1956. Effects of x-irradiation on the
post-embryonic development of a chalcid wasp. Anat. Rec., 125: 625-626.

SCHNEIDERMAN, H. A., J. WEINSTEIN AND J. Horwitz, 1957. Recovery of diapausing larvae
of a chalcid wasp from x-irradiation. Anat. Rec., 128: 618.

Snoperass, R. E., 1935. The Principles of Insect Morphology. Pp. 64-68. McGraw-Hill
Book Co., New York.

StTapigE, W. C., anp N. HAucaAarp, 1946. Oxygen poisoning. X. The effect of oxygen on
the oxygen consumption of the intact mouse. J. Biol. Chem., 164: 257-263.

Tiecs, O. W., 1922. Researches on the insect metamorphosis. Part I: On the structure and
post-embryonic development of a chalcid wasp, Nasonia. Trans. Roy. Soc. S. Aust.,
46: 319-527.

WitiiaMs, C. M., ann H. K. BrEeEcHer, 1944. Sensitivity of Drosophila to poisoning by
oxygen. Amer. J. Physiol., 140: 566-573.

WitiiAMs, C. M., Aanp H. A. SCHNEIDERMAN, 1952. The necessity of motor innervation for
the development of insect muscles. Anat. Rec., 113: 77.

Wuitine, P. W., 1955. A parasitic wasp and its host for genetics instruction and for biology
courses. Carolina Tips, 18: 13. (Published by Carolina Biological Supply Co.,
Elon College, N. C.)

meee AtTION IN THE HAGFISH, MYXINE GLUTINOSA L.

KJELL JOHANSEN 1
University of Oslo, Institute for Experimental Medical Research, Oslo, Norway

Knowledge of the circulation in fishes and in cyclostomes in particular is very
restricted and inadequate for comparison with the data available on circulation in
higher vertebrates. Mvyxine glutinosa L., which is the species used for this study,
belongs to the class Cyclostomata and the family Myxinidae. These animals
represent phylogenetically the most primitive of the existing vertebrates and thus
occupy a critical position in comparative interpretations. This study seems to be
the first one quantitatively undertaking problems related to the dynamics of
circulation in myxinoids.

Fishes are considered to have one circulatory pump, the systemic heart, for
the propulsion of blood through both the gill and the general body capillaries.
Additional hindrance to circulation exists in the special porta and in some species
the kidney capillary systems. The circulatory system in myxinoids is provided
with several specialized structures. Apart from the systemic heart in Myxine
glutinosa there are three accessory hearts on the venous side of the circulation.
These are the portal heart, the cardinal hearts and the caudal hearts. The portal
heart was first described by A. Retzius in 1822. This heart resembles the atrium
of the systemic heart and differs from the other accessory hearts by having a
relatively large lumen and striated muscle fibers with a consequent greater pumping
capacity. The three orifices of the portal heart, the right anterior cardinal vein,
the portal vein and the main portal vein have valves. The two first vessels
represent the afferent channels and the latter the efferent channel of the portal
heart. The cardinal hearts (Cole, 1926) exist as enlarged anterior extremities of
the deep anterior cardinal veins. The caudal hearts (Retzius, 1822) are paired
expansions on the caudal veins. The efferent vessels of the caudal hearts unite
anteriorly to form the median caudal vein. At this point valves are situated which
cause unidirection of the venous blood stream. The present study draws particular
attention to the role of the gills in the myxinoid circulation and results are presented
which demonstrate an active role of the gills for the propulsion of the arterialized
blood.

The six bilateral pairs of gills are located in the region cephalad to the heart
and caudal to the big tongue muscle. The gills are connected to the oesophagus
through the inner gill ducts which constitute the afferent channels for the respira-
tory water. After passing through the lense or pouch-like gills, the water flows
out through the efferent gill ducts which unite on each side and leave the animal
through the two branchial openings. A cross-section through a gill will show
tissue that projects inwards from the periphery of the gill as a number of radial

1 Present address: Dept. of Physiology and Biophysics, School of Medicine, University
of Washington, Seattle 5, Washington.

289

290 KJELL JOHANSEN

plates, each one clothed by the walls of two adjacent evaginations differentiated
from the epithelial lining of the gill. What may be called gill lamellae are
represented by the contiguous walls of two adjacent evaginations plus the interven-
ing vascular tissue.

The afferent gill arteries come from the ventral aorta and form a circular
vessel around each efferent gill duct (Fig. 1). From this circle, several radial

EFFERENT BRANCHIAL

ARTERIES

My

RADIAL ARTERY

WW

EFFERENT

I? VM)

GEE DuUGear

ANN)

AFFERENT
BLOOD. BRANCHIAL ARTERY
Figure 1. (A) Schematic drawing of the vascular arrangement in a gill-body of

Mysxine glutinosa.

AFFERENT GILL DUCT

!

CONSTRICTOR DUCTUS

BRANCHIALIS

CIRCULAR VESSEL
CONSTR. BRANCH. EXT.

GILL Booy CONSTR. BRANCH. INT.

Figure 1. (B) Detail showing the topography of a circular vessel and the muscular elements
in the gill ducts and gill body.

arteries arise which give off a number of vessels at right angles. These vessels
travel in the core of the gill lamellae and join into another series of radial arteries
which open into a second circular vessel around the afferent gill duct. From this
circular vessel two efferent branchial arteries arise which communicate with the
left and right carotid arteries.

A brief histological examination of the gills shows a respiratory epithelium in

CIRCULATION: IN MYXINE, GEUTINOSA ‘L. 291

mm Hg

a 2
CCCECCECEEereeererer ery Teter Btood pressure
ACCEL Ee LELeELL bike boar

10 tf POR Oe ces ee

5 a fretted Korte | 4 [Ld PH jd 4 iditl | Dorsal aorta

fn RG

gl BREE EEEEEET re :

Ss HF Hn! WO EP

Ficure 2. Blood pressure record from the dorsal aorta in Myxine glutinosa.
Time marks, 4 seconds.

contact with a thin sheet of connective tissue in which the blood vessels are running.
Next there are two muscular layers, an external, circular or concentric (constrictor
branchialis externus) and an internal radial (constrictor branchialis internus), the
axis being the gill ducts. All muscle fibers are distinctly striated. A somewhat
similar structure to that in the gill is also present in the gill ducts. The muscula-
ture, however, consists only of ring muscles (Fig. 1B). It is of special importance
that the gill muscles are not a direct continuation of the gill duct muscles, but that
the circular vessels on both sides of a gill body are lying between the ring muscles
in the duct and the ring muscles in the gill itself (Fig. 1B). The structure of the
myxinoid gills is distinctly different from all other fish-type craniotes, even in-
cluding petromyzonts.

MATERIALS AND METHODS

The specimens used for this study were trapped near the Drobak Biological
Station in the Oslofiord in Norway. The animals were kept in aquaria supplied
with running sea water of 4+-7° C. The largest available specimens were used for
the experiments. Their length ranged from 25-40 cm. The unanesthetized animal
was fastened to a dissection board which was placed in a lucite box filled with
oxygenated sea water of 4-7° C. Two incisions were made; the first, in the
region 5-8 cm. cephalad to the cloacal opening, exposed the dorsal aorta and the
caudal vein. A polyethylene catheter (P.E. 50) was introduced into the dorsal

sa See ig aici Blood pressure
" Dorsal aorta

Ventricle

pees

0 _

Ficure 3. Simultaneous blood pressure records from the dorsal aorta (upper curve) and the
heart ventricle. Time marks, 1 second.

292 KJELL JOHANSEN

HE SEE nn

ae ve Blood pressure

t -

erpasegensscunn

Dorsal aorta

wok WN
F

Figure 4. Blood pressure record from the dorsal aorta showing sustained gill contraction.
Time marks, 2 seconds.

aorta and connected with a Statham or Sanborn pressure transducer. Another
incision in the branchial region exposed the systemic and portal hearts. Sometimes
the gills themselves were dissected free in order to follow their movements closely.
The pressures in the ventricle of the systemic heart and in the portal heart were
recorded via cannulas (gauge 26) to the pressure transducers. Care was taken
not to alter the horizontal position of the animals during the experiments and
temperature changes in the lucite box were avoided. All recordings were made on
a Sanborn 4-channel recorder.

RESULTS

The systemic heart showed a frequency different from that of the portal heart.
Both the systemic and portal hearts increased their frequency as a result of in-
creased venous return brought about by digital manipulation of the great veins.

The blood pressure recorded in the dorsal aorta showed a pattern like that
demonstrated in Figure 2. The smaller oscillations are transmissions of the
systemic heart beat through the gill capillaries. The larger paired pressure pulses
were the result of active contractions of the gill musculature. These contractions
most often came in pairs separated by variable times. The frequency of the gill
contractions varied greatly, with average values of 8-10 per minute. Likewise,
the absolute values of blood pressures in the dorsal aorta, as well as the rate of
the systemic heart, showed great individual variations. The heart rates showed

mm Hg
- Blood pressure

Ay ‘Ki

nN
Savieaeay
NSN:

ol
em |

Portal heart

An
| *
a
ie

Figure 5. Blood pressure record from the portal heart. Time marks, 1 second.

CIRCULATION IN MYXINE GLUTINOSA L. 293

variations between 30-40 beats per minute. The mean blood pressure in the
dorsal aorta most often showed values around 5 mm. Hg, varying between 3-8 mm.
Hg. The pulse pressures resulting from the systemic heart beat showed values
of around 2 mm. Hg or 25 mm. of water. All these pressure values were recorded
in the caudal part of the animal in the dorsal aorta. Figure 3 demonstrates
pressures in the heart ventricle and dorsal aorta recorded simultaneously. The
blood-pressure drop across the gill capillaries calculated from the average pressures
in the heart ventricle and dorsal aorta amounted to 4.0 mm. Hg. The lower
tracing in Figure 3 shows that intra-ventricular systolic pressure is around 8.5 mm.
Hg while the diastolic pressure is zero. During other experiments the same values
varied between 15 mm. Hg and zero. Figure 4 shows another feature of the gill
contractions, namely the ability to sustain the contractile state for some seconds.
The pressure in the portal heart (Fig. 5) was consistently between 3-6 mm. Hg.

DISCUSSION

The circulation in Myxine glutinosa as demonstrated in this study differs
markedly from circulation in all other vertebrates studied. Most surprising is
the active role of the respiratory organs for the propulsion of blood. The unique
morphology of the gills in Myxine, however, with the striated musculature and
special vascular architecture might give reasons for anticipating an active participa-
tion of the gills in the circulation of blood. In fact, this is what M. Hofbauer
(1937) suggests in her excellent description of the histology of the gills in
My-xine glutinosa.

It is apparent from Figures 2 and 3 that the pressure wave resulting from a
gill contraction has a very steep diastolic gradient. This means that the peripheral
resistance is low. In attempting to elucidate the physiological importance of the
gill contractions, one has to bring a disputed problem into consideration. This is
the problem of lacunar circulation in vertebrates, first described by Rathke in 1825.
Cole (1926) pointed out that some of the lacunar spaces (blood sinuses) in Myxine
contain red blood cells and he called them “red lymphatics.” These cannot be
considered true lymphatics but must be considered a part of the blood vascular
system. The circulatory system of Myxine must thus be considered a semi-closed
system where the true blood vessels at some particular sites are interrupted by
lacunar spaces. Blood enters these spaces directly from the arteries through
vascular papillae on the branchial and carotid vessels among others. The com-
munication between the lacunar spaces and the veins are established with valved
apertures (see, also, Marinelli, 1956).

A circulatory system of this type can evidently not work as a high pressure
system. On the basis of this knowledge the gill contractions with the consequent
steep transient pressure rises are well explained. Thus it can be observed through
the vessel wall of the dorsal aorta that a steady, slow forward movement of blood
takes place, interrupted at each gill contraction by a massive forward displacement.
The steady, slow flow component apparently results from the permanent pressure
gradient along the arteries between the pressure head from the systemic heart
beat, and the veins. The described ability of the gills in Myxine to sustain the
contractile state for some time seems to be of physiological importance to the animal
by increasing the volume of blood displaced during a gill contraction. A con-

294 KJELL JOHANSEN

tinuous, repeated blood transport of this type obviously necessitates the mentioned
properly sited valves and accessory “‘hearts” on the venous side of the circulation.

The events in the gill pouch during contraction of the gill musculature are
probably as follows: The two circular vessels on the afferent and efferent branchial
arteries, at the entrance and exit of each gill, will each be pressed together when
the striated musculature in the gills and gill ducts contracts. By this act the
blood already arterialized will be forced out in the arterial trunks through the
efferent branchial arteries, thereby causing the relatively high transient pressure
rise throughout the arterial system. On the afferent or venous side of the gill the
inflow of venous blood will stop during gill contraction. At the same time the
water in the gill pouch has been pressed out, which means that the gill when
contracted is in a state where the respiratory function is temporarily stopped. In
other words the gills at this time are solely engaged in the circulation of blood and
water. Upon relaxation of the gill muscles the inflow of water and venous blood
again takes place and the gills resume their respiratory function.

Considerable interest relates to the events in the ventral aorta during a gill
contraction. Recording of the pressure in this vessel without interfering with the
normal circulation is extremely difficult and has not yet been successful. It may,
however, be suggested that a simultaneous contraction of the gills and the systemic
heart creates a pressure in the ventral aorta which surpasses a threshold necessary
to fill the lacunar spaces through the vascular papillae on the afferent branchial
vessels. It may well be reasoned that the outflow of blood through these papillae
and into the lacunar spaces has a pressure threshold only brought about by a high
pressure in the ventral aorta resulting from a simultaneous contraction of the
gills and the systemic heart.

A comparison of the absolute values of pressure in the vessels and heart of
Myxine with those reported earlier on selachians and teleosts are probably of
limited value, considering that lacunar circulation is not present in the latter.
The pressure drop across the gills, although these too differ markedly in their
anatomy from gills in selachians and teleosts, is likely of some interest. Thus,
Schoenlein and Willem (1894) working with selachians reported a pressure drop
in per cent of values proximal to the gills of from 35 to 65%. They worked with
Torpedo sp. and Scyliorynchus catabus. Lyon (1926), working with Carcharias
sp., reported around 25% pressure drop, while J. Mott (1951) recorded 40 to 50%
pressure drop in the common eel, Anguilla anguilla. The present study gave
values of around 50% pressure loss across the gills measuring pressures in the
ventral aorta and heart ventricle. During these recordings the catheter in the
dorsal aorta was pushed in cephalad direction to avoid the pressure drop along
the dorsal aorta itself. The size of the catheter, however, did not obstruct the
circulation in the posterior direction in this vessel.

Another feature of probable physiological importance is explicitly demonstrated
in the architecture of the myxinoid gill (Fig. 1). This has to do with the
principle of counter currents, recently frequently stressed as being of great im-
portance for efficient exchange of materials and energy as well as for maintenance
of concentration gradients in biological systems (Scholander, 1958).

In conclusion it may be stated that the circulatory system in myxinoids, as
judged from this study, shows a very decentralized mode of functioning. In other

CIRCULATION IN MYXINE GLUTINOSA L. 295

words the components actively taking part in the circulation of blood are distributed
throughout the circulatory pattern. Hence the situation differs greatly from the
circulation in other vertebrates where the systemic heart is considered the pre-
dominant site of the blood-moving power.

SUMMARY

1. Electromanometers have been used to study the pressure relations in the
vascular system of Mysxine glutinosa L.

2. The data demonstrate a low arterial pressure level in this species correlating
with the existence of a lacunar circulation.

3. The average pressure in the dorsal aorta ranged from 3-8 mm. Hg with a
pulse pressure of around 25 mm. H,O. The average pressure drop across the
gills, calculated from the average pressures in the heart ventricle and dorsal aorta,
was found to be around 4 mm. Hg.

4. The gills have been shown to take active part in the forward propulsion of
arterialized blood. This is accomplished by contraction of striated muscular
elements in the gills and gill ducts.

5. It is concluded that the circulation in Myxine differs from circulation in
vertebrates generally by a partial shifting of the blood-moving power from the
systemic heart to the gills and the portal, cardinal and caudal auxiliary hearts.

I am indebted to Professor Dr. Bjorn Foyn and cand. real. Finn Walvig at the
Drobak Biological Station for providing the material used in this study.

LETERA TURE, CIPED

Core, F. J., 1926. A monograph on the general morphology of the Myxinoid fishes, based
on a study of Myxine. Part IV. The morphology of the vascular system. Trans.
Roy. Soc. Edin., 54: 309-342.

HorsaAver, M., 1937. Anatomischer und Histologischer Bau der Kiemensacke von My-xine
glutinosa. Biologia Generalis., 12: 330-348.

Lyon, E. P., 1926. A study of the circulation, blood pressure and respiration of sharks.
J. Gen. Physiol., 8: 279-290.

MarINELLI, W., 1956. Mysxine glutinosa L. In: Marinelli and Strenger, Vergleichende Anato-
mie und Morphologie der Wirbeltiere, 2 Wien 1956 (Deuticke).

Mort, J. C., 1951. Some factors affecting the blood circulation in the common eel (Anguilla
anguilla). J. Physiol. (London), 114: 387-398.

RATHKE, M. H., 1825. Bemerkungen tiber den inneren Bau der Pricke oder des Petromyzon
fluviatilis Linn. Danzig.

Retzius, A. A., 1822. Vascular system of Myxine. K. Vet. Akad. Handlgr. Stockholm.

SCHOENLEIN, K., AND V. WILLEM, 1894. Observations sur la circulation du sang chez quelques
poissons. Bull. Sci. France et Belg., 26: 442-468.

ScHOLANDER, P. F., 1958. Counter current exchange. A principle in Biology. AHvalrddets
Skrifter, No. 44. Det Norske Vitenskaps-akademi, Oslo, Norway.

THE EXCRETION OF INULIN AND GEUGOSE BY ira
CRAYFISH ANTENNAL GLAND?

J. A. RIEGEL? AND L. B. KIRSCHNER
Department of Zoology, State Umversity of Washington, Pullman, Washington

The anatomy of the crayfish® green gland (kidney) is relatively well known,
due principally to the investigations of Marchal (1892), Peters (1935) and Maluf
(1939). It consists, proximal to distal, of a coelomosac (saccule), a sponge-like
labyrinth, an elongate (ca. 3 cm.; Maluf, 1941la) tubule and a bladder. The
bladder empties to the exterior of the crayfish by a short duct.

Several investigations of the physiology of the green gland have been under-
taken, but little in the way of unequivocable evidence has been gained concerning
its functional mechanism. Schlieper and Herrmann (1930) were the first to
discover that crayfish urine is markedly hypotonic to the blood. Peters (1935)
removed urine samples from various parts of the green gland. He found that the
chloride content of the samples removed from the distal portions (distal tubule and
bladder) was much less than the chloride content of urine samples removed from
the proximal portions (coelomosac and labyrinth). The chloride content of urine
samples removed from the coelomosac and labyrinth approximated that of the
blood. Peters concluded from his experiments and observations of the microscopic
anatomy of the crayfish green gland that the primary urine, in the form of a blood
filtrate, was formed in the coelomosac and modified by reabsorption of salts (and
possibly water) in the parts distal to the coelomosac. Picken (1936) made
measurements of the colloid osmotic pressure of the blood and the hydrostatic
pressure within the sternal sinus of the crayfish. Since the former pressure
(average = 15 cm. H,O) was exceeded by the latter pressure (average— Z0veum
H,O), Picken concluded that filtration within the crayfish kidney was possible.
Maluf (1941b) analyzed the manner in which the crayfish (Procambarus clarkt)
green gland excreted inulin, xylose and various dyes. Since such substances as
inulin and xylose are not excreted by the vertebrate aglomerular (secretory)
kidney, Maluf reasoned that the crayfish kidney would or would not excrete those
substances depending upon whether that organ formed the primary urine by
secretion or filtration. Despite the fact that both inulin and xylose were excreted
by his experimental animals, he chose to interpret his data to indicate that the

1 Supported by funds for medical and biological research by the State of Washington
Initiative Measure 171, and by a grant (G4925) from the National Science Foundation.

2 Public Health Service Post-Doctoral Research Fellow.

3 Until a recent revision by Bott (1950), the systematics of European crayfishes has
been greatly confused. The species name, Potamobius fluviatilis used by Marchal (1892),
Peters (1935), Schlieper and Herrmann (1930) and Picken (1936) undoubtedly repre-
sents at least two species. Potamobius fluviatilis used by the northern European workers,
probably refers to the present-day species Astacus astacus (L.), while that name used by
French and British workers probably refers to the present-day species, Austropotamobius
pallipes pallipes (Lereboullet).

296

BXCRETIONIIN THE CRAYFISH 297

crayfish green gland forms urine by active secretion of all components, including
water. Martin (1957) suggested that Maluf’s data could alternatively be in-
terpreted to indicate that filtration occurs in the crayfish green gland.

In the present paper, evidence will be presented which will attempt to explain
Maluf’s inulin and xylose (glucose) excretion data within the framework of a
filtration hypothesis.

METHODS AND MATERIALS

The crayfishes used in this study were specimens of Procambarus clarki
(Girard), Orconectes virilis (Hagen) and Pacifastacus leniusculus (Dana). No
species differences were noted in any of the experiments herein described. The
experimental animals were kept in a water bath at 16° C., and all experiments
were carried out at that temperature.

Blood for inulin and glucose analyses was collected as follows: About 0.06
cubic centimeter of blood was drawn into a one cubic centimeter tuberculin syringe
fitted with a 26 or 27 gauge needle after puncturing the ventral membrane between
the merus and carpus of the cheliped (claw). The blood was deposited onto a
piece of Parafilm, rapidly drawn into a 50-microliter self-leveling pipette, and
transferred to a 12 cubic centimeter centrifuge tube. The protein was precipitated
by Flemister’s (1958) modification of the Somogyi (1930) method. The pre-
cipitate was shaken briefly and centrifuged at 3500 revolutions per minute for five
minutes. The supernatant fluid was poured into a ten cubic centimeter test tube.
The precipitate was then broken up with a glass rod, washed with a one cubic
centimeter aliquot of distilled water, centrifuged, and the solution was added to
the first supernatant. The chemical analysis for inulin was done using Flemister’s
(1958) modification of the anthrone method introduced by Young and Raisz
(1952). Glucose was analyzed by the glucose oxidase (GLUCOSTAT, Worth-
ington Biochemical Crop., Freehold, N. J.) method of Saifer and Gerstenfeld
(1958).

Urine for inulin, glucose and sodium analyses was collected as follows: The
front openings of the crayfish’s gill chambers were plugged with a cotton wad
and the epistomal area blotted with absorbant paper. The crayfish was held
ventral side up under a dissecting microscope (observed at ca. 15 X) and the urine
collected from the nephropore by gently tapping the operculum with a capillary
(see Maluf, 1941a, for details of the anatomy of the crayfish nephropore). The
capillaries had been drawn previously to a blunt tip and fire-polished. This
method of urine collection obviated any danger either of contaminating the urine
with blood (as possibly done by Picken, 1936, and others) or making corrections
for water loss when the urine is collected by suction (as necessitated by the method
of Maluf, 1941b). Urine for chemical determinations of inulin and for glucose
analyses was treated in the same manner as described above for blood. Urine for
sodium analysis was directly diluted to an appropriate volume with distilled water.

Blood and urine samples for inulin-C'* analysis were collected volumetrically
as described above, but the samples were plated directly onto one-inch planchettes,
diluted with two cubic centimeters aliquots of distilled water, and dried. The
samples were counted with a conventional windowless gas-flow counter.

Sodium analyses were made with a Coleman flame photometer. Non-radio-

298 J. AYRIEGED AND, LY Brig SCHNER

active inulin was analyzed with the aid of a Bausch and Lomb, “Spectronic 20”
colorimeter.

The general experimental procedure was as follows: crayfishes were removed
from the water bath, wrapped in a paper towel, and permitted to drain for one-half
hour. They were then weighed and the urinary bladders emptied. Fifty-
microliter urine samples were saved for sodium, inulin, and/or glucose analyses,
and the remainder of the urine discarded. Initial blood samples were taken and
any material to be injected was placed into the sternal sinus with the aid of a
%4-cubic centimeter tuberculin syringe fitted with a 26 or 27 gauge needle. The
inulin, glucose and phlorizin concentrations were adjusted so that volumes no
larger than 0.1 cubic centimeter were injected into the animals. The crayfishes were
marked on the branchiostegum (gill chamber) with a grease pencil as a means of

TABLE I

Changes in sodium content of urine taken from experimental animals injected
with inulin (I) and the control (C)

Animal Time Urine Na Time Urine Na Time Urine Na Time Urine Na

number (hours) (% orig.) (hours) (% orig.) (hours) (% orig.) (hours) (% orig.)
[-1 2 . 9 # 22 283 59 341
[-3 8 470 De 450 34 875 57 960
I-5 2 + i a 32, Sy) (ee
I-6 2, * i e 19 4) 34 411
I-8 3 s 9 * 21 258 a sa
I-9 3 56.0 9 40.2 Dal 80.0 et ie:
I-10 3 133 9 * 21 57.4 a: tS
I-11 3 * 9 75.0 Dit 89.7 aoe ih
I-14 3 WI 7i, 9 141 21 246 30 Died
C-2 D 62.2 9 138 22 214 34 411

* No samples obtainable.
** No sample taken.

identification and returned to the water bath. At definite intervals thereafter, the
animals were removed and urine and blood samples taken.

Controls were run with all experiments. The control crayfishes were those
into which no initial injection of inulin, glucose or phlorizin was made, but blood
and urine samples were taken as described above.

All crayfishes were assigned a reference letter corresponding to the particular
substance which was injected into them (e.g., G for glucose, P for phlorizin). This
reference letter is reflected in the tables. Further, a particular number and refer-
ence letter was assigned to only one animal. Thus, animal G-6 in Table II is
the same as animal G-6 in Table III.

RESULTS
Inulin

Of a total of 18 animals injected with inulin, only six continued to produce
urine throughout a period of six to 57 hours (see Fig. 1). Five of the remaining
12 began to produce urine within 21 to 55 hours after injection. In the latter

299

EXCRETION IN THE: CRAYFISH

4k *k ek 9 LL
* OC 87 x *
OT LS 8S * *
*k cs 61 * OOT
2K 2k * 0 0
0 OT 8v 0 Ol
0 v vs 0 8C
0 Li OS 0 8]
g 61 8S * c9
OT LS 8S # 06
0 0 fC 0 0
esoon{s | essoonys (sinoy) asoonys asoonys
aul) poolg aul aul) poolg
Supelied Py poled

[oe]
st

werner rnroonnnre
ANN

(sinoy)
dee £6

2soonys
uli)

N

i=)
N
N
So StS St St St tt St

esoonyd (sinoy)
poold ou

€ polled

Se SE S]) Sy] as

gsoony|s
oulIy)

68
97
Sv
*
OIC
09S
OOT
col
£6

$9
IT

asoonys
poold

Z poled

Roe st es es i st

SOS Ors Or SO Sore

(sinoy)
oul |

asoonys
oulty)

‘uUdye} adures ON yy
‘ajqeurejqo aduies ON ,

16
Ot
8?
¢

-

0
Ot

0

asoonys
poold

T poled

|

Sore SO iS SiO ra a

(sinoy)
oul |

BONS AAN+HOA
eter aly; tt

—=inaN
' ! !

‘ou
jewtuy

asomnjs fo uoysalur ayy sazfp (J) s]o4quor pun (4) sypmrun jouaurdadxa ut (% “Bu U2) S]ara] aSOIN]S AU1AN PUD Poolg_

I] ATaVL

300 J. A, RIEGEL AND’ L, Bo KIRSCHNER

TABLE III

Changes in sodium content of urine taken from experimental animals injected
with glucose (G) and controls (C)

Animal Time Urine Na Time Urine Na Time Urine Na Time Urine Na
number (hours) (% orig.) (hours) (% orig.) (hours) (% orig.) (hours) (% orig.)
G-1 1 167 + 200 7 400 23 334
G-3 1 i. 4. # 7 is Did, 714
G-6 1 "4 4 120 30 s 50 82.6
G-7 1 45.5 4 45.5 30 75.0 58 45.5
G-9 1 1125 4 5 23 1850 50 364
G-10 1 13.9 + - DD re 50 18.4
G-11 1 100 i 3401 De, 66.7 50 167
G-13 4 637 ih 341 26 291 54 1045
G-14 4 31.0 7 116 26 Sel 54 -
G-15 1 120 4 124 26 112 54 84.5
C-4 1 150 4 100 i. sa 19 *
C-5 4 108 7 95.8 30 120 58 275
C-12 iL 100 4 100 20 408 48 15

* No sample obtainable.

cases, the urine/blood (hereafter designated as U/B) inulin values (in milligrams
per cent) were 160/0, 60/0, 280/12, 800/440 and 790/498. Of the seven animals
that did not produce urine after the injection of inulin, three died. Autopsy of
the three crayfishes revealed collapsed urinary bladders. The remaining four

30

2 =inulin-C!4
25 x =inulin

U/B INULIN RATIO
on

fe)

0.1 1.0 lO 100
BLOOD INULIN (ma.%)

Figure 1. Urine/blood (U/B) inulin ratios in relation to blood inulin levels in crayfishes:
injected with radio-active and non-radio-active inulin.

EXCRETION IN THE CRAYFISH 301

crayfishes may have begun again to produce urine, but no attempt was made to
follow urine production longer than 60 hours.

Eleven animals were injected with inulin-Ct*, and eight of them continued to
produce urine throughout the period of the experiments (six hours; see Fig. 1).

It is interesting to note that whenever only a small amount of urine (less than 10
microliters) could be collected, the inulin U/B ratios were significantly higher
than when amounts greater than ten microliters could be collected. Of a total
of 39 urine collections during these experiments, 27 were greater than ten micro-
liters (average = 41.5 microliters). The average U/B inulin ratio at the same
times was 3.4. The average amount of urine collected during the 12 remaining
periods was 5.8 microliters, during which times the average U/B inulin ratio was
9.0. This difference between average inulin U/B ratios of large and small urine
samples is statistically significant at the 5% level (f= 2.25). Sodium determina-
tions were made on urine collected from nine animals into which inulin had been
injected. Urine sodium increased by factors of 1.3 to 9.6 in six of the nine
crayfishes (Table I). The sodium content of the urine from two of the three
remaining crayfishes dropped after the inulin injection and then rose slightly during
the balance of the experiment, although never attaining the concentration of the
initial urine sample. The ninth animal stopped producing urine after the injection
of inulin and 55 hours later the urine sodium was only 4 that of the initial urine
sample. The average sodium content of seven urine samples of less than ten
microliters (average = 3.5L) was greater by a factor of 5.1 than the average
sodium content of 12 urine samples greater than ten microliters (average = 35.4
pL). This result is statistically significant at the 1% level of probability
(3-51).

Glucose

A total of fourteen crayfishes were injected with varying amounts of glucose.
Of that number, one-half produced urine relatively continuously throughout the
experimental period. The results of analyses of blood and urine glucose from
these animals are shown in Table II, where it can be seen that glucose usually (but
not always) appears in the urine under one of two circumstances: (1) the blood
glucose level exceeds approximately 200 milligrams per cent (animals 11 and 21
at four hours) ; (2) the experiment lasts thirty hours or longer. Of seven animals
that stopped producing urine after the glucose injection, two died, and urine was
unobtainable from two others during a period of 21 to 50 hours; the remaining
three again produced urine 23 to 26 hours after injection, and in all cases, ap-
preciable quantities of glucose were present in the urine. Of special interest is
the fact that analysis of urine and blood of the controls (crayfishes not injected
with glucose; see Table II) produced results similar to those obtained from the
experimental animals. That is, blood glucose rose, urine production stopped, and
in one case, glucose appeared in the urine late in the experimental period.

Analysis of the urine sodium content of ten crayfishes into which glucose had
been injected showed that the sodium content of the urine of eight animals increased
by factors of 1.2 to 34 (Table III). Two animals, unexplicably, showed a reduc-
tion in the sodium content of their urine.

Phlorizin caused the appearance of glucose in the urine in all crayfishes into
which it was injected. However, in order to cause the appearance of quantities of

302 J. AV RIEGED AND OD?) By itiies CHING.

TABLE IV

Blood and urine glucose levels (in mg. %) in experimental animals injected
with phlorizin plus glucose (P) and the control

Blood |Urine Time Blood |Urine mairie Blood | Urine Blood | Urine

Animal | Time glu- | elu

no. ||(hours)|: Bee | goa | (hours) || See | Sona | hours) || Sea | cose! |@PUF9)| Gaze |coce || Panam
Pil (ORS AGI Ys Le ta KO) PSS. Se POE hie 3. |) 12 anew ices
Ee? 0 PHA 4 30"). 65 | OM oo mic eo anaae 6 | 102 mg./Kg.
Epes OFF) SOP NO 4 51°) 107) 10 | 44 89341023) 1.32" |) 460 Say oeeme aie
BS Or CBO 60 A 4360} 4370100 | Foy a8i 23% 4) 27.) 18h) seieemeaen
P-6 Oi 2S NiO An | AT) 106 .)2 104) 2804559) | 26. 25). 27" eee
Pay Oa Sith ra 4 70.) 247 | 10. | 36.) 119 | 26%) 58 |) 93.) ‘Soinmecyiieae
CA OM oe 4 | 44 OO 26 Oo ahi Nees aliaes ae

* No sample obtainable.
** Animal killed accidentally.

glucose sufficent to provide unequivocable results from the glucose analyses, prim-
ing with exogenous glucose was necessary. The data shown in Table IV are
taken from six animals into which doses of 12 to 102 milligrams per kilogram of
phlorizin (plus glucose) were injected. The effect of all tested levels of phlorizin
was immediate and prolonged. Differences in effect between the different
phlorizin dosage levels appeared to be very slight, although no attempt was made
to determine minimal and maximal doses.

DISCUSSION

It would be well at the outset to attempt to clarify the terminology to be used
lest the application here of terms coined for vertebrate function lead to misunder-
standing. By “secretion” is meant the movement of material from the blood into
the lumen of the organ against a gradient of electrochemical potential. In secretion
the cells are expending energy to move solutes and/or water into the lumen. The
term “filtration” usually is used quite specifically to indicate the bulk movement
of a solution through a porous membrane by the application of a head of pressure.
The pore dimensions of the membrane determine what is filtered and what is
restrained. Apart from the fact that the necessary net pressure may exist in the
crayfish (Picken, 1936), there is no evidence that filtration in this sense occurs
among Crustacea. In the following discussion, filtration will be used to mean
the formation somewhere in the organ of a (blood-isotonic) solution containing
molecules below a critical size. The process is indiscriminate in the sense that
the chemical species is not important (1.e., everything below a certain size appears
in the “‘filtrate’). In addition, and critical from the point of view of active water
movement, movement of solution occurs from high to low potential. Thus, the
question that is really under exploration is, does the movement of water into the
lumen of the antennal gland require energy? The formation of a dilute urine by
a secretory mechanism would obviously require active water transport.

Factors affecting the variability in urine production

The results of the experiments presented in this paper are similar in most
respects to those of Maluf (1941b), except that Maluf apparently did not encounter

EXCRETION IN-THE CRAY EISH “308

difficulty in collecting urine after the injection of inulin and glucose (xylose).
A difference in experimental procedure between the present work and that of
Malut’s may clarify this discrepancy. Maluf made no effort to obtain urine from
his experimental animals for eight to fifteen hours after the injection of inulin or
glucose (xylose). At that time, he sacrificed the crayfishes and collected the
urine in the bladders. In the present work, attempts were made to collect serial
samples at much shorter intervals after the injection of inulin and glucose.

The cause of the aforementioned variability in urine production is unknown.
However, the only factor which appears to be related to the decrease in urine
production is the mechanical handling of the crayfishes, especially the withdrawal
of blood samples at frequent intervals. It can be seen from the data in Tables I
and III and the discussion of results that an increase in blood glucose created
effects similar to those seen when inulin was injected into the crayfishes. That is,
urine production was quite variable and urine sodium increased. The blood
glucose data from control animals (Table II) show that there was a spontaneous
rise in blood glucose which could only have been due to handling. Further, the
urine sodium content rose in all control animals. Therefore, the variability in
urine production is closely associated with a rise in blood glucose, which in turn, is
possibly due to excessive handling of the crayfishes necessitated by the experimental
procedures.

Filtration versus secretion

There is no positive evidence that the crayfish green gland is a secretory kidney.
However, there are two principal features of the anatomy of that organ which
make it difficult to assign to it the role of a filtration kidney. (1) There ap-
parently is no glomerulus-like structure (the tenuous syncytium of Maluf) against
which a hydrostatic pressure head could be applied to effect filtration. Paren-
thetically, it should be pointed out here that the statement made by Prosser ef al.
(1950, p. 31), “The histological picture (of the crayfish green gland) is not
compatible with filtration, however. The coelomic sac and tubules lie in haemo-
coelic spaces, so that pressure must be similar on all sides, and it is difficult to see
how unidirectional pressure could exist as it does in the vertebrate kidney
glomerulus,’ appears to be based on a misconception of the anatomy of the cray-
fish green gland. As adequately demonstrated by Marchal (1892), Peters (1935)
and Maluf (1939) and confirmed by personal observations, all parts of the crayfish
green gland are well supplied by branches from the antennal and sternal arteries.
These main branches break up into finer and finer branches until eventually the very
finest branches become sinusoids. As stated by Peters (1935, p. 359), “Ich konnte
beim Flusskrebse die Blutkapillaren bis zu den Zellen verfolgen (Abb. 4, 5 und
6). Fast jede Zelle wird an ihrer Basis vom Blute umspult.” Since the vessels
are patent, it is possible that even in the smallest sinusoids there is still sufficient
arterial pressure to effect filtration. That the sinusoids do respond to pressure
may be shown by a quote from Peters (1935, p. 359) concerning the coelomosac,
“Wurde unter grossem Drucke injiziert, so traten zahlreiche, sonst fast kollabierte
Gefasse in Erscheinung (Abb. 4). Bei normalen Blutdruck scheint ein Teil von
ihnen immer kollabiert zu sein.’ (2) The histological appearance of the green
gland suggests abundant secretory activity. Further, the crayfish kidney possibly
can secrete Indigo Carmine, as suggested by the fact that that dye becomes con-

304 7. ACRIBGELY NDVI? Beker G rian Eo

centrated in the urine ( Maluf, 1941b, and personal observations). Maluf (1941b)
utilized these two factors as the main issues when he chose to interpret the results
of his experiments as indicating that the crayfish green gland is a secretory
kidney. For similar reasons, Burger (1957) was loath to say that filtration oc-
curred in the lobster (Homarus americanus) green gland despite good physio-
logical evidence for it. In regard to the first factor, Martin, Harrison and
Stewart (1953), working with the terrestrial gastropod, Achatina fulica, and
Harrison and Martin (1955), working with the cephalopod, Octopus hongkongensis,
obtained good evidence for filtration in those two molluscs despite the apparent
absence from their kidneys of structures analogous to the vertebrate glomerulus.
The second factor is hardly worthy of concern, since abundant secretory activity
need not exclude the green gland from consideration as a filtration kidney. It is
well known that the vertebrate glomerular kidney, although functioning primarily
by filtration, excretes many substances by secretion, including Indigo Carmine
(Smith; 1951). Further, Cyanol is excreted by the crayfish green gland at low
blood concentrations, whereas that dye is not excreted by the vertebrate secretory
kidney, except at high blood concentrations (Maluf, 1941b). Finally, a “secretory
appearance” does not specify the direction of secretion, 1.e., reabsorption could be
indicated.

The inulin and glucose excretion data supporting Maluf’s (1941b) argument
that the crayfish green gland is a secretory kidney need to be examined critically,
since they render questionable much of the evidence which otherwise would support
a filtration hypothesis. Following is a discussion of Maluf’s (1941b) experiments
and results reviewed in the light of results and calculations obtained from the
present study.

Maluf’s conclusion that inulin is secreted rather than filtered derives primarily
from the nature of the relationship between urine and blood concentrations. Maluf
noted that when the blood inulin levels varied between 100 to 500 milligrams
per cent the concentration of inulin in the urine was considerably higher than that
in the blood. Between 500 to 1800 milligrams per cent, the U/B ratios were
approximately unity. There was an enormous scatter in the experimental points
within the lower range, but quite gratuitously, a line was drawn indicating a non-
linear decrease which approached unity as a lower limit. The decrease in the
inulin U/B ratios with increasing blood inulin concentrations was used to support
the thesis that that substance was secreted rather than filtered. Several serious
objections can be raised against this interpretation. In the first place, there was
too much scatter in the experimental points to justify the drawing of a hyperbola
through them. A straight line for the range of blood concentrations between 100 to
500 milligrams per cent would equally well fit the points. Actually, the experi-
mental eae presented in this paper (see Fig. 1) show that there is much more basis
for this, for when the range of blood concentrations is extended down to less than
one milligram per cent inulin, it can be seen that U/B ratios are nearly constant.
Certainly there is no indication of the steep hyperbola indicated in Malut’s (1941b)
paper.

It should be noted, also, that Maluf’s entire argument for the secretion of
inulin is based on fallacious reasoning. Assuming that the rate of secretion of
inulin is limited because the transport mechanism becomes saturated at high blood

EXCRETION IN THE CRAYFISH 305

levels (as stated by Maluf, 1941b), it can be shown that the quantity (N) of
inulin secreted in unit time is related to the blood level by

If, as Maluf showed, the volume of urine formed per unit time is constant, then

V/
Sees ae
ica (B)
where N’, U’ and K are all constants and (B) is the blood concentration. Super-
ficial examination of this equation shows that U/B will be relatively flat over the
range of blood concentrations in which (B) is much less than K, and that it will
approach zero where (B) is greater than K. An asymptotic approach to a lower
limit of one will be realized only if filtration 1s assumed in addition to secretion.
In fact, a kidney that excretes inulin both by filtration and secretion could produce
the pattern found experimentally; that is, inulin U/B ratios greater than one
when the blood inulin concentrations are low, followed by a decrease to unity at
high blood concentrations. While this possibility must be kept in mind, Maluf’s
experiments at high blood inulin levels are open to criticism of such a nature as to
render the significance of his results suspect. The molecular weight of inulin 1s
about 5000 (Westfall and Landis, 1936). Assuming that it is non-penetrating
(which, of course, is implicit in its frequent use for measurement of extracellular
fluid volume), the use of large quantities will raise the colloid osmotic pressure
SEQEP)-oi the blood. Thus, Picken (1936) reported that the COP of crayfish
blood averages about 15 cm. of H,O. An inulin concentration in the crayfish’s
blood of 500 milligrams per cent would increase the COP of the blood by 23 em.
of H.O. Even assuming that the animal could make compensatory adjustments in
blood pressure and extracellular fluid volume, these adjustments could be effective
only over a limited range. The fact that U/B inulin ratios begin to decrease at
about 500 milligrams per cent, and remain near unity at blood levels of 1000 to
1800 milligrams per cent may represent little more than renal failure of some kind
due to abnormal elevation of blood COP. If a lower molecular weight is used
(Hodgman, 1955), the increase in COP would be even greater. Thus, before
Maluf’s (1941b) data could be used to support any hypothesis, a rigorous investiga-
tion should be undertaken to show that the antennal gland functions normally
under conditions of high blood inulin levels. The best that can be said in favor
of a secretory process is that the organ might be excreting inulin by a combination
of filtration and secretion. However, the introduction of an uncontrolled variable
(colloid osmotic pressure) at high blood inulin levels makes the evidence for
secretion tenuous. Since either secretion or filtration would yield U/B inulin
ratios practically independent of the blood concentration of that substance when
the latter is low, the inulin data in this range provide no support for either
mechanism of urine formation.
There are, however, other data which are better explained on the basis of urine
formation by filtration followed by reabsorption. The data in Table II show that
glucose usually is not excreted in the urine, but if the blood concentration 1s

306 J A RIEGEL AND Geb ERS CHNER

sufficiently high (perhaps about 200 mg. %), it will begin to appear. Thus,
glucose may behave in the crayfish as it does in vertebrates with glomerular
kidneys, 7.e., as a threshold substance. For some reason, Maluf (1941b) used
xylose rather than glucose, and his data seem to indicate either no threshold, or a
very low one. However, when his figure showing blood xylose levels is examined,
it can be seen that the initial blood concentrations were invariably much higher
than those tabulated with the urine concentrations ; in fact, they were always higher
than 200 milligrams per cent, although, as in the glucose experiments presented in
this paper, they fell very rapidly. Thus, the blood concentrations shown in Malut’s
table on xylose excretion are misleading in that they represent, in every case,
values much lower than the level immediately following injection. It is not
possible to compute clearances or reliable thresholds in the face of blood levels
that fall (and rise) as rapidly as seen in Maluf’s and the present investigation.
The glucose (xylose) appearing in the urine may all have been excreted when the
blood concentration was very high and the computation of a clearance on the basis
of “average” blood concentrations may actually obscure this information. This
is certainly the case in the present experiments with glucose. Serial sampling
showed (Table II) that urine glucose is zero prior to the injection of exogenous
glucose, and where glucosuria occurs, it always does so in the period following the
peak in blood glucose and returns to zero when the level of that substance in the
blood falls.

Moreover, the experiments with phlorizin are difficult to rationalize on the
basis of secretory activity. The consistent and sustained glucosuria following the
administration of that drug must indicate that glucose normally appears in a pre-
sumptive urine from which it is subsequently reabsorbed. The highest glucose
U/B ratios following phlorizination were 3 to 4, which brings them into the range
of the inulin U/B ratios, and hence supports the conclusion that the latter can
be used to measure fluid turnover in the crayfish green gland.

Therefore, despite the apparent absence from the crayfish green gland of a well-
defined site for filtration, the facts that inulin can be excreted and that phlorizin
causes glucosuria lead the authors to conclude that filtration (in the limited sense
explained at the beginning of this discussion) does occur in that organ. In ad-
dition, it is quite probable that filtration is followed by the reabsorption of water
and a large quantity of solute.

SUMMARY AND CONCLUSIONS

1. Inulin is excreted by the green gland of the crayfish, and within the range
of blood inulin levels used in the experiments presented here, the inulin urine/blood
ratios were always greater than one.

2. Glucose normally is not excreted by the crayfish green gland. However, if
the blood glucose level exceeds about 200 milligrams per cent, it usually appears
in the urine. Further, after thirty or more hours of handling (see paragraph 3
below), many crayfishes began to excrete glucose.

3. Handling (i.e., injection of material, and, especially, serial withdrawal of
blood samples) often resulted in erratic urine flow, concentration of the urine
(reflected by large increases in inulin and sodium concentrations in small, less
than 10-microliter, urine samples).

EXCRETION IN* THE CRAY EISH 307

4. Phlorizin caused an immediate and prolonged glucosuria at all dosage levels
tested (12 to 102 mg./Kg.).

5. Despite the absence of a well-defined site in the crayfish green gland at which
filtration (in a limited sense, defined in the text) can occur, the authors conclude
that because inulin is excreted and glucose causes glucosuria, filtration does occur
in that organ. Furthermore, the filtration is accompanied by the reabsorption of
salts and possibly water.

Pite RA TURE SCHED

Bott, R., 1950. Die Flusskrebse Europas (Decapoda, Astacidae). Abh. d. Senckenberg.
Naturforsch. Gesell., 483: 1-306.

Burcer, J. W., 1957. The general form of excretion in the lobster, Homarus. Biol. Bull., 113:
207-223.

Fremister, L. J., 1958. Salt and water anatomy, constancy and regulation in related crabs
from marine and terrestrial habitats. Bol. Bull., 115: 180-200.

Harrison, F. M., anp A. W. Martin, 1955. Kidney function in a cephalopod. Fed. Proc., 14:
69-70.

Hopeman, C. D., ed., 1955. Handbook of Chemistry and Physics. Chemical Rubber Pub-
lishing Co., Cleveland, Ohio.

Matur, N. S. R., 1939. On the anatomy of the kidney of the crayfish and on the absorption
of chlorid from freshwater by this animal. Zool. Jahrb., Abt. f. allg. Zool. u. Physiol.
d. Tiere, 59: 515-534.

Matur, N. S. R., 1941a. Micturition in the crayfish and further observations on the anatomy
of the nephron of this animal. Biol. Bull., 81: 134-148.

Mauur, N. S. R., 1941b. Secretion of inulin, xylose and dyes and its bearing on the manner
of urine-formation by the kidney of the crayfish. Bzol. Bull., 81: 235-260.
MarcuHat, P., 1892. Recherches anatomiques et physiologiques sur l’appareil excréteur des

crustacés décapodes. Arch. de Zool. expér. ef gen., 10: 57-275.

Martin, A. W., 1957. Recent advances in knowledge of invertebrate renal function. Recent
Advances in Invertebrate Physiology. University of Oregon Publications, Eugene,
Ore.

Martin, A. W., F. M. Harrison Aanp D. M. StTEwArt, 1953. Urine formation in the giant
African Snail. Abs. Com., 19th Internat. Physiol. Cong., 592.

Peters, H., 1935. Uber den Einfluss des Salzgehaltes im Aussenmedium auf den Bau und die
Funktion der Exkretionsorgane dekapoder Crustaceen. (Nach Untersuchungen an
Potamobius fluviatilus und Homarus vulgaris.). Zeitschr. f. Morph. u. Okol. ad.
Tiere, 30: 355-381.

Picken, L. E. R., 1936. The mechanism of urine formation in invertebrates. I. The excre-
tion mechanism in certain Arthropoda. J. Exp. Biol., 13: 309-328.

feossee ©. L., D. W. BisHor, F. A. Brown, Jr, T. L. JAHN anp V.:.J.. Wutrr, 1950.
Comparative Animal Physiology. W. B. Saunders Co., Philadelphia, p. 31.

SAIFER, A., AND S. GERSTENFELD, 1958. The photometric microdetermination of blood glucose
with glucose oxidase. J. Lab. and Clin. Med., 51: 448-460.

ScHLIEPER, C., AND F. HERRMANN, 1930. Beziehungen zwischen Bau und Funktion bei den
Exkretionsorganen dekapoder Crustaceen. Zool. Jahrb., Abt. f. Anat. u. Ont. Tiere,
52: 624-630.

SmitH, H. W., 1951. The Kidney. Oxford Univ. Press, New York. 1949 pp.

Somocyt, M., 1930. A method for the preparation of blood filtrates for the determination of
sugar. J. Biol. Chem., 86: 655-663.

WESTFALL, B., AND E. LANnpts, 1936. The molecular weight of inulin. J. Biol. Chem., 116: 727.

Younc, M., ann L. Ratsz, 1952. An anthrone procedure for determination of inulin in bio-
logical fluids. Proc. Soc. Exp. Biol. Med., 80: 771-774.

THE EFFECTS OF GLUCOSAMINE, HY DROCHEORMDE (OMe tras
DEVELOPMENT OF DROSOPHIBA MEEANOGAS TEI

ME Me ORD al.
Department of Biology, Reed College, Portland 2, Oregon

An unusual type of abnormality in Drosophila was encountered during a series
of feeding experiments involving glucosamine hydrochloride. These abnormal
pupae resembled the cryptocephal recessive lethal described by Hadorn and Gloor
(1943). Since a high incidence of abnormal individuals was recovered after treat-
ment, and none of the other sugars tested produced morphological abnormalities,
a more detailed examination of the conditions responsible for this effect, as well as
a histological examination of the abnormalities, was undertaken. This report
summarizes the observations of the effects produced by feeding glucosamine
hydrochloride to larvae of Drosophila melanogaster.

MATERIALS AND METHODS

The tu” mutant strain which develops melanotic tumors in the caudal fatbody
(Wilson et al., 1955; Rizki, 1957) and the Ore-R wild type strain have been used
in the present study. The abnormal pupae were first noted in experiments in
which larvae were removed from Cream of Wheat-Fleischmann’s yeast medium
and placed on paper strips moistened with a solution of glucosamine hydrochloride.
This same method was therefore used in subsequent experiments, as well as
merely adding the test substance to the medium on which larvae were feeding.
Each group of larvae was collected during a two- to three-hour interval, and
larval ages are counted from the time of oclusion from the egg. All experimental
material was maintained at 23°—25° C. in an incubator.

In the series of starvation experiments, the timed larvae were removed from
the food dishes and washed in a series of changes of a saturated solution of NaCl,
a 2% solution of NaOCl, followed by at least six rinses of distilled water. The
larvae were then placed on tissue paper strips (Kleenex brand) in sterile petri
dishes with filter paper under the covers to prevent the escape of the larvae from
the dishes. Distilled water was used to moisten the paper strips in the control
dishes, and the test solutions were pipetted onto the paper strips in the experimental
dishes. Conditions were standardized throughout the study by using four layers
of paper, size 2.5 X 2.5 inches, with 2 ml. of test solution for each group of 15-20
larvae. In some experiments sterile cellulose pulp was used as the crawling site
for the larvae in place of paper strips. In other experiments the sugar solutions
were added to Cream of Wheat-Fleischmann’s yeast medium. In these cases the
larvae were washed as above and approximately the same area of the petri dish was
covered with the food as the size of the paper strips used in the starvation ex-

1 This investigation was supported by a research grant RG 5285 from the National In-
stitutes of Health, Public Health Service.

308

EFFECT OF GLUCOSAMINE ON DROSOPHILA 309

periments in order to maintain conditions as similar as possible throughout the
study.

The abnormal pupae were dissected in Waddington solution. The histological
material was prepared by fixing larvae and pupae in Carnoy solution, embedding
paraffin, and staining the sectioned specimens with Azure B (Flax and Himes,

1952),

RESULTS

The typical pattern of the abnormality induced by feeding glucosamine hydro-
chloride involved the disproportionate sizes and relative positions of the structures
of the head and thorax. The most striking feature of the abnormal pupae was
the reduced size of the eyes which nevertheless developed normal pigmentation
(Fig. 1). Examination of pupae after removal of the puparium showed that the
cephalic structures were contained in the thorax. The thoracic structures, in-
cluding the legs and the wings, have developed, their distortion in position being a
result of the presence of the head within the thorax. In some cases the legs
appeared shorter than normal and were not fully stretched (Fig. 2). The larval
mouthparts were still attached to the anterior end of the thorax, and many of the
puparia were unusually elongated. The degree of contraction of the larval cuticle
had been less in these cases. In some pupae the bristles were well formed but in
others the development of these structures, particularly in the posterior region of
the body, had been delayed. None of the abnormal pupae hatched, and sectioned
material from these pupae was subsequently examined.

Stained preparation of the sectioned pupae revealed that eversion of the
imaginal discs of the imaginal complex had occurred but these developing head
structures did not emerge from the thorax. Figures 3 and 4 are frontal sections
passing through the brain in an abnormal pupa and a normal specimen, respectively.
In the former, the brain is located in the anterior region of the abdomen. Com-
parison of these sections indicates the degree of morphogenetic displacement which

occurs in the brain mass. In the normal pupa, as the cerebral ganglion is pushed

forward, the anteriolateral parts of the brain are displaced caudolatrad and the first
optic stalks assume a lateral position to the brain. In the abnormal specimens the
antennae and palpi appear normal with respect to their morphological development
but they are found lodged in the area enclosed by the eyes. The carina is not
properly formed. The gonads of the abnormal pupae are well developed and no
apparent abnormality of the external genitalia would be detected. Eversion of the
wings and legs has been complete. A parasagittal section of an abnormal pupa is
presented in Figure 5 to illustrate the development of the thoracic muscles, tergites,
and bristles. In this photograph, the compound eyes, although failing to emerge
from the thorax, show well formed rhabdomeres and crystalline cones, and the
setae are normal in appearance. Several pupae were found which had dwarf
heads, and improper wing eversion was apparent in a few adults.

The first group of abnormal pupae was recovered during a survey of nutritional
factors which were being tested for their effects in the mutant strain, tu”. For
these experiments larvae 65 hours of age were removed from food and placed on
paper moistened with a 10% solution of each test substance. Larvae remained
in these dishes until after pupation, and scoring was completed as the adults
hatched. Under these conditions, glucosamine hydrochloride was the only chemical

310 M. T. M. RIZK1

Ficures 1-4.

— c

EFFECT OF GLUCOSAMINE ON DROSOPHILA 311

__ 06 &

Ficure 5. Parasagittal section of a glucosamine-induced phenocopy of crc in the Ore-R
strain showing the detailed structures of the eye, E; optic lobe, O; longitudinal thoracic muscles,
T; muscles of the appendages, L. A large portion of the brain is in the abdomen demarcated
by A.

which induced any noticeable morphological changes. The list of substances
tested included N-acetyl glucosamine, thiamine hydrochloride, fructose, glucose,
lactose, starch, mannitol, salicin, casein hydrolysate, nucleic acid hydrolysate,
glycogen, melibiose, trehalose, ribose, sorbose, inulin, sucrose, maltose and galactose.
Drosophila larvae 65 hours of age were removed trom food and placed on paper
moistened with distilled water. These larvae completed their development and
normal adults were obtained. Repetition of these conditions with glucosamine
hydrochloride yielded the same type of abnormalities each time, and this method

Figure 1. An atypical pupa within the puparium obtained after feeding glucosamine
hydrochloride to third instar larvae of the Ore-R strain. The concave discoid pigmented eye,
PE, is located in the thorax; the arrow, T, is pointing to the scutellar bristles of the thorax.

Ficure 2. Atypical imagoes removed from the puparium (ftu”). Note the absence of
the cephalic region while the thorax and abdomen with their associated structures are well
formed. Female genitalia, FG; leg, L; male genitalia, MG; wing, W; the larval mouthparts,
LM, still attached to the anterior region of the thorax. The legs are not fully stretched.

Figures 3 AND 4. Comparison of frontal sections of a glucosamine-induced abnormal
pupa (ftu”) and a normal tu” imago. In the abnormal specimen in Figure 3 the brain is
located in the abdomen and the eye discs are flanking the inner lateral sides of the thorax
forming a cavity. Larval mouthparts, LM; carina, C; antenna, A; eye, E; muscles of
flight in the thorax, T.

312 M. TM. RIZKE

was used to test 1%, 5%, 10%, and 20% solutions of glucosamine hydrochloride
and N-acetyl glucosamine in succeeding experiments. In addition these same
solutions were added to a thin layer of food covering approximately the same
area of the dish as the size of the paper strip.

1.0 &
eer 15(3)
as 16(0) |
0.8
esd 8
-
cS)
lad
0.6
<
a 3.4(15)| hia
= 0.5 4y
or /
<q “e
J 0.46 /
isn. :
2 el
o 0.3 7 /
- v7 e
© / © page,
Ps 0.2 / -
ve 19(2) | : AA
ag
W17 (2) eh
0.1 aot | 21(2)/ oF
ae 7

0.0 B2'(0),20(0)
2 5 10 20
% GLUCOSAMINE—HCL

Ficure 6. The yield of abnormal pupae and larvae in the tu” and Ore-R strains obtained
after feeding varying concentrations of glucosamine hydrochloride. The solid lines represent
starvation conditions in the tu” and Ore-R strains, and the vertical lines on these two curves
indicate standard deviations. The results on the fed larvae are represented by dotted lines.
All of the larvae used in this experiment (#756) came from one collection period. The
fraction of larvae affected includes the total effect of feeding glucosamine hydrochloride, that
is, abnormal pupae as well as larvae which died before pupation. At each point on the graph,
the number preceding the parenthesis is the sample size which is followed by a number in
parentheses giving the count of the pupae at this point with the cryptocephal phenocopy. For
example, the higher per cent of phenocopies in the tu” strain under starvation conditions was
obtained with 5% glucosamine hydrochloride; no phenocopies were obtained in the fw" strain
with 10% and 20% glucosamine hydrochloride as the larval lethality approached 100%.

:
(
7

EFFECT OF GLUCOSAMINE ON DROSOPHILA 313

None of the tu” larvae survived in the 20% solution of glucosamine under
starvation conditions, and very few of the Ore-R larvae pupated. Many of the
larvae crawled off the paper at this concentration, so the high mortality is pre-
sumably the combined effect of the glucosamine as well as desiccation. Behavior
of this type was not noticed in the dishes with the lower concentrations of
glucosamine. Combining the number of dead larvae and abnormal pupae obtained,
the fraction of response to glucosamine hydrochloride can be graphically rep-
resented, thus indicating the difference between the two strains of flies. Figure 6
illustrates the results of one experiment on larvae taken from the same collection
period. Direct comparison between larvae on food and those on paper strips can-
not be made, but the difference most probably represents the dilution factor of the
food since the same amount of solution was added to the food surface as that
added to the paper in the dishes. The type of abnormalities induced was the same
under conditions of feeding and starvation, as well as in the two strains of flies.
The optimum concentration for the production of abnormal pupae was 5% for
the tu” strain and 10% for the Ore-R strain when larvae were removed from food
at 65 hours of age, and these were the concentrations which were used in the
experiments from which material was fixed and sectioned. In the present study,
a total of 276 abnormal pupae was obtained from the tu” strain and 99 from the
Ore-R strain. Under optimum conditions of treatment the per cent of abnormal
pupae in both strains was as high as 60%.

Samples of glucosamine hydrochloride obtained from three sources were used
(Nutritional Biochemicals Corp. ; California Corporation for Biochemical Research ;
Krishell Laboratories) and all produced the same type of abnormalities. The
dosage response decreased with the age of the solution of glucosamine hydro-
chloride, so only freshly prepared solutions were used.

DISCUSSION

The typical pattern of abnormality obtained after feeding glucosamine hydro-
chloride to Drosophila larvae is the same as that reported by Hadorn and Gloor
(1943) for the recessive lethal which they termed cryptocephal, crc. This name
appropriately described the characteristic feature of the glucosamine-induced ab-
normal pupae with the head structures contained in the thorax. Histological
examination of the abnormal specimens has shown that the cephalic complex has
undergone development and eversion, and the principal morphological difference
appears to be the displacement of structures resulting from the lack of emergence
of the head. A similar developmental upset was obtained in two stocks, the wild
type Ore-R and the tu” mutant strain, although a higher percentage of abnor-
malities was induced in the latter strain under the same conditions of treatment.
Comparing pupation time in the two stocks, there is an average delay of at least
twelve hours in the tu” strain. Therefore the difference in response to glucosamine
may reflect a difference in developmental stage since the larvae from both stocks
were of the same age at the time of treatment. A survey of the effects produced
by glucosamine hydrochloride when offered at different ages throughout larval
life in both stocks will have to be undertaken before this question of difference of
response can be answered.

Gloor (1944) reported lack of emergence of the head in some of the pupae of
a wild stock of D. melanogaster by using temperature shocks during the prepupal

314 NER Bal

period shortly before the normal time of emergence of the head. The use of
glucosamine hydrochloride as a phenocopic agent may provide a useful tool in
studying the physiological mechanism of emergence of the head in the developing
pupa. The biological role of the aminosugars has been recently reviewed by Kent
and Whitehouse in their monograph (1955). Glucosamine and N-acetylglucosamine
are components of the polysaccharide polymer, chitin. Complexes of chitin and
protein are important constituents of insect cuticle. The incomplete contraction of
the cuticle, apparent in some Drosophila specimens after feeding glucosamine
hydrochloride, may suggest this system as a point of interference. The reported
inhibition of anaerobic glycolysis and O, uptake by glucosamine in leucocytes
(Alonso, 1959) also suggests further examination for such effects in Drosophila
larvae and pupae. The ease with which abnormal pupae were obtained under
starvation conditions was utilized to obtain as large a number of specimens as
possible for the present survey. However, this restriction of starvation may not
be desirable for an analysis of the mechanisms underlying the formation of the
abnormalities since the histological condition of the abnormal specimens obtained
in starvation experiments was inferior to that in the fed experiments.

SUMMARY

Abnormal pupae were obtained after feeding glucosamine hydrochloride to
larvae from two stocks of Drosophila melanogaster, the Ore-R wild type strain and
the mutant melanotic strain, tu’. The characteristic feature of the abnormality
was the position of the head within the thorax, thus suggesting the use of
glucosamine hydrochloride as a phenocopic agent to produce individuals resembling
the recessive lethal cryptocephal.

PITRE RALURE Cia

Atonso, D., 1959. The effects of glucosamine and galactosamine on lactic acid production and
oxygen uptake by guinea pig leucocytes. Bacteriol. Proc., Soc. Amer. Bacteriologists,
59th Meeting, pp. 75-76.

Frax, M. H., anp M. H. Himes, 1952. Microspectrophotometric analysis of metachromatic
staining of nucleic acids. Physiol. Zool., 25: 297-311.

Guioor, H., 1944. Phanokopie einer Letalmutante (crc) von Drosophila melanogaster. Rev.
Suisse Zool., 51: 394-402.

Haporn, E., anp H. Gtoor, 1943. Cryptocephal, ein spat wirkender Letalfaktor bei Dro-
sophila melanogaster. Rev. Suisse Zool., 50: 256-261.

Kent, P. W., ano M. W. WuitenHouse, 1955. Biochemistry of the Aminosugars. Butter-
worth Scientific Publications, London.

Rizxi, M. T. M., 1957. Tumor formation in relation to metamorphosis in Drosophila melano-
gaster. J. Morph., 100: 459-472.

Witson, L. P., R. C. Kine ann J. L. Lowry, 1955. Studies on the fu” strain of Drosophila
melanogaster. Growth, 19: 215-244.

PreereckLCULAR DIGESTION AND HYDROLYTIC ENZYMES
Py AP AGO Yrs OF; PiEANARIANS *

ROBERT M. ROSENBAUM AND CARMEN I. ROLON
Department of Pathology, Albert Einstein College of Medicine, New York 61, N. Y.

It has been established that, in those species studied, intraluminar digestion
does not occur in triclad flatworms (Willier, Hyman and Rifenburgh, 1925;
Jennings, 1957), but that digestion takes place within food vacuoles in the
phagocytic cells of the gastrodermis. A number of investigators have observed the
disintegration of ingested food particles within these cells (Arnold, 1909; Saint-
Hilaire, 1910; Jacek, 1917; von Levitzow, 1943), but the methods they employed
did not permit elucidation of the mechanisms involved except in a most general
way. Kelley (1931), studying the intracelular digestion of nucleoprotein in D.
dorotocephala by means of the Feulgen reaction with and without acid hydrolysis,
showed that the ingested material was hydrolyzed within the first two hours.

To our knowledge, no previous study has characterized the hydrolytic enzymes
in the phagocytes of planarians. The present report considers the relationship of
three such enzymes to current concepts of intracellular digestion.

MATERIALS AND METHODS

Specimens of Dugesia dorotocephala and D. tigrina were used for this study.
The animals were maintained as stock cultures at 18° C. and fed routinely once
every ten days on minced rat or mouse liver. Experimental animals were isolated
in fingerbowls and starved for at least three weeks. Some of these were fed raw
rat or mouse liver, and others boiled liver to determine if enzyme activities in raw
food were transferred to the gut. No differences in enzyme activities between
animals fed raw or boiled liver were demonstrable by the staining methods
employed. The worms were observed during feeding; post-feeding times were
determined from the moment the animals ceased feeding.

For the demonstration of those enzyme activities surviving chemical fixation,
specimens were initially fixed in warm (37° C.) Baker’s calcium-formalin for 5
minutes to prevent discharge of the gut contents. They were then placed in
calcium-formalin at 4° C. for 24 hours, washed for 5 minutes in distilled water
and cut on a CO,-freezing microtome at 8-10y. Material thus prepared was
incubated for acid phosphatase activity by a slight modification of the method of
Gomori (1952). Sections were incubated for 15-30 minutes in a 0.1 M 2-
glycerophosphate and lead nitrate mixture buffered to pH 5.0 with 0.05 M acetate
buffer.

Aminopeptidase activity was visualized by the method of Burstone and Folk
(1956). Animals were placed on Monel grids and frozen in isopentane chilled by
liquid nitrogen to — 160° C. After quenching, the worms were dehydrated for 24

1 This investigation was supported by a research grant, No. RG-5483, from the United
States Public Health Service.

315

316 ROBERT M. ROSENBAUM AND CARMEN I. ROLON

yy

AV

SRL

Ficures 1-5.

.
4
)
|
:

INTRACELLULAR DIGESTION IN PLANARIANS Sie],

hours in a vacuum (10-* mm. Hg) at — 28° C. and infiltrated for 30 minutes with
54-56° C. paraffin. Paraffin sections cut at 8m were incubated at 37° C. in
substrates consisting of either L-leucyl-8-naphthylamide (LNA) or L-alanyl-,-
naphthylamide (ANA) buffered to pH 7.1 with 0.2 M Tris buffer. Control sec-
tions were incubated in a solution of B-naphthylamine in 0.2 M Tris buffer. Garnet
GBC or Fast Corinth Salt V were employed as simultaneous couplers in the in-
cubation medium.

B-glucuronidase activity was demonstrated by the method of Billett and McGee-
Russell (1955). It was not possible to achieve a precise intracellular localization
of the final ferric-8-hydroxyquinoline crystals with formalin-fixed frozen sections.

_ As an alternative, the entire worm was placed in quinolyl-8-glucuronide at pH

4.2-4.5 and incubated in the presence of a saturated ferric-8-hydroxyquinoline
solution at 37° C. for 24 hours, following which they were washed, first in distilled
water, then in oxalate buffer and finally mounted im toto in glycerine jelly.

RESULTS
Histology of the gastrodermis

The gastrodermis consists of a single layer of epithelial cells resting on a

delicate basement membrane. Two distinct cell types are present. The phagocytic
cell is the larger, the more numerous and contains a basal nucleus and a free end
that protrudes into the gut lumen. The cytoplasm is basophilic and has a varying
number of eosinophilic inclusions that progressively disappear after feeding. These
presumably represent the contents of food vacuoles undergoing digestion (Willier,
Hyman and Rifenburgh, 1925).
_ The second cell type is smaller and less columnar than the phagocyte although
its shape varies considerably, depending on the nutritive state of the animal. It
has been variously described as a storage, gland or spherical cell and contains
intensely staining eosinophilic bodies generally considered to be a protein reserve.
The morphology of these two cell types throughout the digestive process has been
considered in detail by Willier, Hyman and Rifenburgh (1925).

Acid phosphatase

No activity was detected in any of the cells comprising the gastrodermis of
starved animals (Fig. 1). Due to the presence of diffuse brown-black pigment

Ficure 1. D. dorotocephala. Starved (14 days) animal showing three segments of
digestive tract (arrows). There is no activity in the gastrodermis. A number of pigmented
cells are seen, as well as abundant pigment in the dorsal epidermis. Gomori lead sulfide
reaction incubated for 30 minutes at pH 4.5. 100 x.

Ficure 2. D. dorotocephala, gastrodermis 6 hours post-feeding. There is acid phos-
phatase activity in many of the food vacuoles but not all cells are equally reactive. 420 x.

Ficure 3. Acid phosphatase activity in fragments of gastrodermis from D. tigrina fed 5
minutes previously. There is a distinct precipitation of lead sulfide within the vacuoles of
the phagocytic cells which in some cases have become rounded during maceration. 620 x.

Ficure 4. Acid phosphatase activity in the gastrodermis of D. dorotocephala 24 hours
after feeding. Notice the intense staining in the food vacuoles of the phagocytic cells. A
weaker reaction is evident throughout the gastrodermis. 100 x.

Ficure 5. Section of gastrodermis from D. dorotocephala 10 days after feeding. There
is some acid phosphatase activity in the cytoplasm but no food vacuoles are present. Several
pigment cells are evident. 420 x.

318 ROBERT M. ROSENBAUM AND CARMEN I. ROLON

FicurES 6-9.

INTRACELLULAR DIGESTION IN PLANARIANS 319

throughout the animal, controls were necessary to distinguish such pigment from
the similarly colored deposits of lead sulfide in a positive enzyme reaction. Sec-
tions serving as controls were either heated in a water bath at 60° C. for 45
minutes or incubated in buffer without glycerophosphate.

Some enzymatic activity could be detected in sections of animals fed 30 minutes
previously. The gastrodermis of animals killed six to twelve hours after feeding
showed distinct acid phosphatase activity. Activity was localized within vacuoles
formed in the phagocytic cells (Fig. 2), but the intensity of the reaction varied
considerably from vacuole to vacuole. Such differences in enzyme activity are
presumably dependent on the amount of material phagocytized, the length of time
the food has been in the vacuole and, hence, the extent of digestion taking place
in the vacuole at the time of fixation.

Attempts were made to visualize the intracellular distribution of acid phos-
phatase during formation of the food vacuoles. Since frozen sections did not
provide optimal preservation of cytological details, a series of small animals, fed
approximately 30 minutes previously, were squashed on chilled coverslips and im-
mediately fixed in cold calcium-formalin. These preparations were then incubated
in buffered (pH 4.8) glycerophosphate and acid phosphatase activity visualized by
precipitated lead sulphide. As early as five minutes after initiation of feeding,
acid phosphatase could be demonstrated within newly forming food vacuoles
(Fig. 3). During formation of the food vacuole, the sites of enzyme activity are
discrete but of varying size depending on the physiological activity of the vacuole.
Four to six hours after feeding, most vacuoles showed such intense acid phosphatase
activity that the entire vacuolar space was filled by a lead sulphide precipitate. No
cytoplasmic enzyme activity could be detected at this time.

Twenty-four hours after feeding, intense acid phosphatase activity occurred
throughout the gastrodermis (Fig. 4). While staining was still most striking
within the food vacuoles, a reaction also occurred in the cytoplasm of the columnar
cells. In all instances, this was limited to the gastrodermis with no staining
demonstrable in the mesenchyme (Fig. 4). This activity, both in vacuoles and
cytoplasm, persisted up to 48 hours, following which a gradual diminution became
evident.

One week after feeding, a few cells in gastrodermis showed slight staining for
enzymatic activity (Fig. 5).

Ficure 6. D. dorotocephala, starved for two weeks and incubated in L-leucyl-p-
naphthylamide for aminopeptidase activity. There is no staining in the gastrodermis (gast.)
but two regions in the mesenchyme (mes.) show enzyme activity. Numerous pigment cells
are present throughout the section, especially in the dorsal epithelium. Photographed in
combination with Corning narrow band filters, Nos. 3486, 9782 and 5120. 100 x.

Ficure 7. Aminopeptidase activity in the gastrodermis of D. dorotocephala 30 minutes
after feeding. Staining is not uniform but its localization within vacuoles is evident. 210 x.

Ficure 8. Aminopeptidase activity localized within vacuoles in the gastrodermis of
D. dorotocephala 12 hours after feeding. There is no reaction in the cytoplasm of the
phagocytic cells. 460 x.

Figure 9. Aminopeptidase activity in the gastrodermis of D. dorotocephala. The picture
is typical of the reaction in animals 24 to 48 hours after feeding. 160 x.

Figure 10. VD. dorotocephala starved for 10 days. No aminopeptidase activity is present.

160 x.

320 ROBERT M. ROSENBAUM AND CARMEN I. ROLON

Aminopeptidases

Although both leucyl- and alanyl-derived substrates were employed, it was
impossible to distinguish differences in activity between the two substrates. The
pH optima (7.0-7.6) and optimal incubation times for both substrates were also
identical. Only slight activity was obtained with both substrates at a more acid
pH (4.5-5.0).

There was some aminopeptidase activity in the mesenchyme of starved animals
although none could be detected in the gastrodermis (Fig. 6). The presence of
enzyme in the mesenchyme of starved animals was probably related to a lowered
nutritional state since an increase in enzyme activity in the mesenchyme of animals
starved for two, four or six weeks, respectively, was always observed. This
suggests that a mechanism is present outside the gastrodermis for utilization of
the animals’ tissues as a food reserve. The ability of planarians to withstand
long periods of starvation with a corresponding reduction in body size is well known.

Amunopeptidase activity was detected in the gastrodermis 30 minutes after
feeding (Fig. 7). The staining appeared strongest in those cells possessing the
largest food vacuoles and, hence, undergoing most active phagocytosis.

Twelve hours after feeding, nearly all the phagocytic cells of the gastrodermis
contained food vacuoles which stained positively for aminopeptidases (Fig. 8).
Staining was most intense 24 to 48 hours after feeding. (Fig..9), following which
there was a rapid decline in activity throughout the gastrodermis.

Seven days after feeding (Fig. 10), no reaction for aminopeptidase was ob-
tained in the gastrodermis. At this stage, some cells in the mesenchyme did
possess a positive reaction for these enzymes (Fig. 10). This appears to support
the observation of Jennings (1957), that phagocytosed material can pass back
into the mesenchyme for subsequent digestion.

Beta-glucuronidase

The method originally suggested by Billett and McGee-Russell (1955) for the
demonstration of B-glucuronidase was not satisfactory for fresh-frozen sections of
planarians. Even tissue from animals fixed for three hours in cold neutral formalin
did not satisfactorily withstand incubation in quinolyl-8-glucuronide at pH 4.0-4.5.
Since intracellular localization of the enzyme could not be obtained in sections, it
was decided to use the im toto incubation method employed by Billett and Mulherkar
(1958) for the chick embryo. Small, recently fed worms were fixed in cold
neutral formalin for three hours. They were then thoroughly washed in several
changes of cold distilled water and incubated in quinolyl-8-glucuronide for 24 hours
at 37° C. as previously described. Control animals were incubated in substrate
to which 0.0005 M potassium acid saccharate had been added as an inhibitor. The
presence of ferric-8-hydroxyquinoline crystals could be detected in the gut of
animals 24 or 48 hours after feeding. No similar reaction was obtained in the
gut of starved animals. Although the technique employed for #-glucuronidase
was not satisfactory from the standpoint of intracellular localization, it did demon-
strate the presence of B-glucuronidase activity parallel in time to the appearance
of acid phosphatase and aminopeptidase.

—

INTRACELLULAR DIGESTION IN PLANARIANS 321

DISCUSSION

While a considerable amount of investigative work has been undertaken on
intracellular digestion following phagocytosis or pinocytosis in organisms such as
Ameba, little is known about the digestive enzymes present in other invertebrates
(Kitching, 1956). In the present reports, the localization of specific hydrolytic
enzymes in some fresh-water triclads has been demonstrated in relation to the food

vacuoles formed within phagocytic cells in the gastrodermis.

The acid environment within food vacuoles has long been known and it has
been assumed that this is related to the function of the digestive process. The pH
optima of acid phosphatase and £-glucuronidase are approximately the same
(4.24.6). Jennings (1957) has obtained a value of pH 4.6 in the food vacuole
of the triclad, Polycelis, six hours after feeding. Our own data (unpublished)
show similar values for Dugesia.

It has not been sufficiently emphasized that aminopeptidase activity, identified
by the methods employed here, does not ipso facto demonstrate the leucine
aminopeptidase considered as a single enzyme by Smith and Spackman (1955).
Other peptidases are capable of splitting leucyl peptides (Fleisher, 1954) and it is
therefore necessary to assume caution regarding the degree of specificity obtained
by the present methods. The similarities, both in the degree and distribution of
enzyme activity as well as pH optima, obtained with the substrates LNA and ANA
in the present investigation, suggest that we are dealing with such a lack in
specificity. Recently, Waldschmidt-Leitz and Keller (1957) have demonstrated
that “leucine” aminopeptidase is a mixture of peptidases and does not represent a
specific enzyme. The histochemical significance of this problem has been con-
sidered in detail by Glenner et al. (1959).

It is interesting to consider the possible mechanisms responsible for release of
enzymatic activity following feeding. The observation that the three enzymes
studied here appear after the majority of food vacuoles have formed suggests a
secretion of enzymes from the cytoplasm into the vacuole. An alternative mecha-
nism would call for the de novo synthesis of enzymes within the forming vacuole
but then one would expect to find enzymatic activity appearing along with the
earliest formation of the food vacuole. Some evidence for the secretion of enzymes
from cytoplasm into vacuoles has come from observations of the microkinetospheres
of certain cell variants of HeLa cultures (Rose, 1957). The localization of acid
phosphatase activity throughout the gastrodermis 24 to 48 hours after feeding
indicates that the cytoplasm of these cells possesses the capacity to synthesize the
enzyme. There is evidence from the present observations that aminopeptidase is
similarly localized in the cytoplasm of cells. In starved animals, for example,
activity was found in the mesenchyme and the possibility that such sites may
represent localized digestion of the animals’ own tissue—perhaps for nutritive
purposes—has been mentioned.

The three enzymes studied represent a class of six or more soluble acid hy-
drolases collectively associated with a group of cytoplasmic particles termed
“lysosomes” by de Duve (1959). Although the lysosome concept has been in-
vestigated most thoroughly in the liver cell, there is increasing evidence that these
particles may be concerned with localized phenomena related to the acid digestion

ay ae ROBERT M. ROSENBAUM AND CARMEN I. ROLON

of cells in general. Evidence for this is available from observations on Ameba
(Bennett, 1956) ; spleen macrophages (Palade, 1956) and, most conclusively, in
“athrocytosis’” phenomena in kidney cells (Straus, 1957).

SUMMARY

1. The presence of the three hydrolytic enzymes, acid phosphatase, ami-
nopeptidase and 8-glucuronidase have been demonstrated in relation to digestion
phenomena in Dugesia dorotocephala and D. tigrina.

2. None of these enzymes were detected in the gastrodermis of starved animals.
Acid phosphatase activity appears 5 minutes after feeding commences and is
closely identified with formation of the food vacuoles. Maximum phosphatase
activity is present in all vacuoles 24 to 48 hours after feeding; a slight reaction
occurs in the cytoplasm of all gastrodermal cells at this time. Three days after
feeding, acid phosphatase activity declines until one week after feeding, no activity
is evident.

3. Aminopeptidase activity is present in the mesenchyme of starved animals
where it is probably associated with “self-digestion” phenomena related to the
nutritional state of the animal. In general, aminopeptidase activity parallels that
of acid phosphatase. While intracellular localization of B-glucuronidase was not
achieved, its activity in the digestive tract was observed to be similar to that of the
other two enzymes studied.

4. The results obtained on a histochemical level are discussed in terms of
current concepts of the mechanism of intracellular digestion.

VITERALURE, CrreD

ArRNoLD, G., 1909. Intra-cellular and general digestive processes in Planariae. Quart. J.
Micr. Sct., 54: 207-220.

Bennett, H. S., 1956. A suggestion as to the nature of the lysosome granule. J. Biophys.
Biochem. Cytol., 2 Suppl.: 185.

Bittett, F., anp S. M. McGee-Russet1, 1955. The histochemical localization of B-glucuroni-
dase in the digestive gland of the Roman snail (Helix pomatia). Quart. J. Micr. Sct.,
96: 35-48.

Bittett, F., anp L. MuLwHerKar, 1958. The localization of $-glucuronidase in the early
chick embryo. J. Embry. Exp. Morph., 6: 52-56.

Burstone, M. S., Anno J. E. Forx, 1956. Histochemical demonstration of aminopeptidase.
J. Histochem. Cytochem., 4: 217-226.

DE Duve, C., 1959. Lysosomes, a new group of cytoplasmic particles. Jn: “Subcellular Par-
ticles,’ ed. by T. Hayashi. Ronald Press, New York.

FietsHer, G. A., 1954. Peptidase in human blood. II. The hydrolysis of additional peptides
by normal serum. J. Biol. Chem., 206: 637-640.

GLENNER, G. G., M. S. Burstone Anp D. B. Meyer, 1959. A study of aminopeptidase activity
in the stroma of neoplastic tissue, with a comparison of histochemical techniques.
J. Nat. Cancer Inst., 23: 857-873.

Gomort, G., 1952. Microscopic Histochemistry. Univ. of Chicago Press, Chicago.

Jacek, S., 1917. Untersuchungen tber den Stoffwechsel bei rhabdocoelen Turbellarien (Sten-
ostomum). Bull. Int. Acad. Sci. Cracovie, 8(B): 241-261.

JenNnINGS, J. B., 1957. Studies on feeding, digestion and food storage in free-living flatworms.
(Platyhelminthes: Turbellaria). Biol. Bull., 112: 63-68.

Ketiey, E. G., 1931. The intracellular digestion of thymus nucleoprotein in triclad flatworms.
Physiol. Zool., 4: 515-541.

KitcHine, J. A., 1956. Food vacuoles. Protoplasmatologia, vol. III, part D3b. Springer,
Vienna; pp. 1-54.

INTRACELLULAR DIGESTION IN PLANARIANS $Z3

von Levirzow, K. G., 1943. Zur Biologie und Vedauungsphysiologie der polycladen Turbel-
larien. Zool. Anzg., 141: 189-196.

Parape, G. E., 1956. The endoplasmic reticulum. J. Biophys. Biochem. Cytol., 2 Suppl.:
85-98.

Rose, G. G., 1957. Microkinetospheres in VP satellites of pinocytic cells observed in tissue
culture of Geys strain HeLa with phase contrast cinematographic techniques. J.
Biophys. Biochem. Cytol., 3: 697-704.

Sarnt-Himarre, C., 1910. Beobachtungen uber die intracellulare Verdauung in den Darmzellen
der Planarian. Zeitschr. allg. Phystol., 11: 177-248.

SmitH, E. L., Aanp D. H. SpAcuman, 1955. Leucine aminopeptidase. V. Activation, speci-
ficity and mechanism of action. J. Biol. Chem., 212: 271-299.

Srraus, W., 1957. Changes in “droplet” fractions from rat kidney cells after peritoneal in-
jection of egg white. J. Biophys. Biochem. Cytol., 3: 933-947.

WaLpscHMipt-LeEitz, E., Aanp L. KeEtier, 1957. Zur Frage der Existenz einer Leucyl-pep-
tidase. Zeitschr. physiol. Chem., 309: 228-238.

Wittirr, B. H., L. H. Hyman awnp S. A. RIFENBURGH, 1925. A histochemical study of
intracellular digestion in triclad flatworms. J. Morph. Physiol., 40: 299-340.

THE IN VITRO REACTION OF LIMULUS Avie © Gyan
LO BAGS:

M. V. SHIRODKAR, A>] WARWICK AND E23." BANG

Marine Biological Laboratory, Woods Hole, Massachusetts, and Department of Pathobiology,
The Johns Hopkins Umiversity School of Hygiene and Public Health, Baltimore 5, Maryland

The horseshoe crab, Limulus polyphemus, an ancient marine arachnoid, reacts
to infection by extensive intravascular clotting (Bang, 1956). Since this is
primarily a cellular reaction, followed by extracellular gelation (Loeb, 1910), it
was of particular interest to study the interaction of bacteria and the cellular
components im vitro. The present paper reports (a) the maintenance of the
normal granular discoid amebocytes in siliconized and unsiliconized glassware, (b)
the destructive effect of the bacteria and bacterial toxin in first changing these cells
to agranular and filiform extended cells and then destroying them, and (c) the
ability of the granular discoid amebocytic preparations to suppress the development
of nonpathogenic bacterial infections in the cultures.’

MATERIALS AND METHODS

Healthy, adult specimens of Limulus, 10 to 12 inches across the shell, mostly
female, were obtained from the Marine Biological Laboratory Supply Department
and kept in running sea water. No data were available concerning the previous
history of infection among these animals.

Preparations without silicone—granular and agranular cells

Thirty to 50 ml. of blood were obtained by cardiac puncture and were placed
directly into a series of unsiliconized 10-ml. screw-cap roller tissue culture tubes
(2 ml. per tube) with routine sterile precautions. After some manipulations to
obtain a uniformly thin layer of material over most of its inner curvature, the
tube was placed in a rotating drum and kept at room temperature (20-25° C.).
The cells were studied by direct microscopic examiantion of the tube and by
removal of drops of the culture fluid* with its suspended amebocytes. The
latter were followed both by phase and direct transmitted light. A combination
of 100 units each of penicillin and streptomycin was used in a few of the early
experiments, but in most tubes no antibiotics were used. The effect of fluid
changes was studied by replacing the media with various concentrations of adult

1This work was supported by a Grant-in-Aid (E135) from the National Institutes:
of Health.

2 Predoctoral Research Fellow, National Institutes of Health.

3 Some of this material has been reported in abstracts: (Shirodkar, Bang and Warwick,,
1958; Warwick and Bang, 1957).

4 Hereafter referred to as “medium” or as “serum.”

324

LIMULUS AMEBOCYTES AND BACTERIA 325

Limulus serum. Wolbach’s modification of Giemsa’s stain was used in order to
determine the cytological details of the cells.

Cultures with silicone—cultures in siliconized glassware

In these experiments the culture tubes, syringes and needles were coated with
silicone (G.E. SC-87 Dri-Film) and air-dried for 3-4 hours before they were
thoroughly washed with the commercial washing powder “Gold Dust,” rinsed
several times with tap and then with distilled water and dry sterilized at 180° C.
for 146 hours. Large volumes of blood could not be withdrawn and transferred
(2 ml. for each roller tube) without clot formation unless the apparatus was thus
siliconized. Particularly careful manipulation was required immediately after
explantation of the blood. The preparation was considered satisfactory only if a
clear majority of the attached and suspended amebocytes was indistinguishable
from those found im vivo, 1.e. discoid, granular, (Lankester, 1884; Loeb, 1902),
as studied by low power microscopy, for at least 48 hours following explantation.

Description of bacteria used

Limulus pathogen. In the summers of 1953, 1954, 1957 and 1958 organisms
were isolated from the peripheral blood of sick adult animals by means of techniques
described elsewhere, and the 1958 stock cultures were maintained by fortnightly
transfers on ZoBell’s sea water agar, enriched with peptone and ferric salts (Bang,
1956). They were stored between transfers at 4° C.. On the basis of bacterio-
logical tests ° the strain isolated in 1958 was identified as a Vibrio sp. It is gram-
negative, short, thick, polarly monotrichous, motile and does not form spores. It is
comma-shaped, particularly in old cultures. Thin capsules are occasionally pres-
ent. The organism is facultatively anaerobic and has an optimum growth tem-
meertero: 22° C., complete cessation of: growth occurring at 5° and 37° C.. It
produces B-hemolytic zones on horse blood agar; liquefies gelatin rapidly (1 day
at 22° C.); shows presence of peroxidase and catalase; does not produce H,S;
gives a weak positive reaction with indole and reduces nitrate to nitrite. The
bacterium is incapable of fixing nitrogen and shows neither fluorescence nor
luminescence. It produces acid, but not gas, in the following carbohydrate media:
glucose, galactose, maltose, trehalose, mannitol, starch, sorbitol (weak at 15 days),
glycerol (weak) and salicin (weak). No acid is produced with arabinose (late
positive, 15 days), xylose, rhamnose, sucrose, lactose, adonitol, dulcitol, inositol
and ethyl alcohol.

A heat-stable toxin was obtained from this bacterium by the same method as in
the previous work.

Bacterium #5 (Limulus nonpathogen). Pure cultures ® of this gram-negative
peritrichous, motile rod were obtained at the beginning of the summer of 1958 from
oysters.

Bacteriological studies showed no fermentation except in ethyl alcohol, which

5 These, as well as the tests on Bacterium #5, were kindly performed by Dr. H. Lautrop
of the State Serum Institute, Copenhagen, Denmark, to whom we are grateful for the ac-
companying description.

6 These, as well as other cultures used here, were isolated by Mr. Stuart Krassner of our
laboratory and kindly supplied to us.

326 M. V. SHIRODKAR, A. WARWICK AND F. B. BANG ©

suggested that it belongs in the Alkaligenes group. However, since it produces.
H,S, it is tentatively designated Alkaligenes-like. When inoculated, even in large
doses, into healthy, adult animals, these bacteria produced no demonstrable disease.

Other bacteria. Eight different bacteria were tested, seven of which had
been isolated from oysters whilst one, a yellow chromogen (#419), was found as a
laboratory contaminant. No attempt was made to fully identify any of them.
They were distinguished one from the other principally on the basis of colony and
cellular morphology. They were arbitrarily numbered 4, 10, 12, 14, 15, 17, 18, 19.

Antibacterial activity of cultures. The experiments designed to test the extent
of antibacterial activity and to determine in which fraction(s) of blood it resided,
consisted essentially of (a) setting up blood culture tubes, adding known amounts
of bacteria, withdrawing aliquots from the mixtures at successive time intervals and
plating these on agar to determine the drop in numbers of viable bacteria; (b)
following the same procedure with tubes containing different fractions of the blood,
such as serum with few granules, serum with many granules and with fresh,
whole blood homogenates. The measures of antibacterial activity were (1) the
largest initial inoculum following whose introduction into the system either no
bacteria (or reduced numbers) were recoverable and (2) the length of time
needed to achieve such reduction or to maintain inhibition at a constant level.

One ml. of sterile, artificial sea water (without NaHCO,) was added to an
agar slant containing a heavy 24-48 hour (room temperature) growth of the
organism and mixed thoroughly. In the earlier experiments, the titers of such
undiluted suspensions of the various bacteria were determined by making pour
plates of serial log,, dilutions in sea water agar. The resulting values for viable
bacteria were found to be consistently in the range of 10° to 10*° cells per 0.1 ml.
We subsequently assumed these figures for. freshly prepared suspensions. Inocula
of 0.1 ml. of various dilutions were added to the experimental and control culture
tubes ; the tubes were kept at room temperature in the rotating drum, and samples
were taken subsequently for culture by means of a platinum wire loop. Colonies
were counted at 24 hours. In the series of experiments designed to yield informa-
tion regarding the fraction(s) of blood containing the factor(s) responsible for the
bacteria-reducing capacity of the cultures, fresh blood was aseptically explanted
into unsiliconized Petri dishes which were then allowed to stand at room tem-
perature for varying periods of time. A_ gel-like clot formed rapidly after
explantation and was adherent to the bottom of the dish as described by Loeb
(1920). Syneresis occurred and the supernatant serum contained a minimum
number of free amebocyte cytoplasmic granules if drawn off within a few minutes
after placing the blood in the dish. With stirring, this number increased steadily
up to about two hours. Aliquots were removed from the supernate at intervals
and the state of the granules was noted under the low and high dry powers of
the light microscope, using blue light and no filter, with reference to shape, size,
color, refractility and Brownian movement.

Three terms which will have particular connotations as used throughout this
report may need clarification. A normal granule is one which appears like those
seen in the intact cell. It has a uniform size; the shape is spherical to slightly
ovoid; it appears green when seen through the ordinary light microscope, using
blue light and no filter; has high refractility and shows a characteristic amplitude
of Brownian movement when in serum. Polymorphic granules are considered

327

LIMULUS AMEBOCYTES AND BACTERIA

S;

granular cell

id

1SCcO

ote the d

N

in.

x

{ to

10n O

iti

add

tore

te culture bef

cy

Amebo

Ficure 1.

~
Nn
fe)
uv
>
x
=
Y
uv
~
~
1S)
}
ow
vu
v
&
<
=
ie
o}
~
Ge
oS
Ke
o}6
00 °°
rcs
as)
—
ENS
—=
v
Os
ed
a 8
Wigs
wok
e
= 2
mS
Y o&
> . fy
v
a
ee
a
Bs
Sie
Ba
oi
oar
n YO
EY
2 8
eh"
oh
Qo,
we.
eA
N
3
i,
ag
+
=
on
—_
—

integrated.

S

i

nf

e apparent]

tended; many hav

x

granules and e

M.. V.. SHIRODKAR; A. WARWICK AND: FB, SANG

Ficures 3-4.

LIMULUS AMEBOCYTES AND BACTERIA 329

either degenerate or pathological, may be rod- or dumbbell-shaped and may be
larger or smaller than the normal. Tvranslucent granules are those which have lost
their normal high refractility. They are usually a pale yellow-green.

RESULTS
In vitro survival of amebocytes

Variation in serum concentration. In the hope that a lower concentration of
Limulus serum frequently renewed might lead to cellular growth, four different
‘concentrations were tested: 10, 25, 50 and 100 per cent, some with glucose added.
The cells remained viable for as long as 36 days at the 10 and 50 per cent con-
‘centrations, but were not followed as long at the 25 and 100 per cent concentrations.
Although long-term observations were not made with the 100 per cent (undiluted )
serum in these first experiments, it was noted that granular discoid amebocytes
were commonly found in the undiluted serum but rarely or never at the 10 per
cent concentration. In all cases, the number of these cells gradually diminished.
In general, cell appearance varies from the thin, flattened, hyaline cells with no
granules and sharp, extended pseudopod-like processes, to the extended but
granular cells and, finally, the granular discoid amebocyte. Attempts to transfer
the amebocytes by direct explantation, trypsinization, or the use of “Versene”’
failed. In none of the direct microscopic examinations, nor in the stained prepara-
tions of the cultures were mitotic figures seen, nor was there any conversion of the
amebocyte tissue into an organized growth.

Undiluted Limulus serum was therefore used in the tests with the bacterial
toxin, and partially degranulated cells were consistently obtained. It was found
unnecessary to replace the nutrient medium (the serum) at any time even in
experiments of several weeks.

Effect of siliconized glassware. By applying silicone to the apparatus, in the
very first experiment and in subsequent good cultures, the amebocytes were main-
tained morphologically indistinguishable from those in the animal for as long as
30 days in the rotating drum at room temperature without replenishment of the
culture medium; more will be said about this in the section on interaction with
bacteria.

Effects of the Limulus pathogen and its toxin on the blood cultures. Since
the generalized pathology of the disease, an intravascular clotting, can be repro-
duced by a heat-stable toxin, the effects of this toxin on amebocytes in vitro were
studied. As is shown in Table I, the addition of various dilutions of toxin to
cultures of amebocytes in normal clean glassware caused prompt changes in mor-
phology, culminating in the apparent disintegration of the cells, and the formation
of an extracellular gel. The cells lost their granules, extended long processes, and
assumed bizarre shapes. They are shown in Figures 1 and 2.

In order to study these changes under higher power, a loopful of fresh undiluted
bacterial (Vibrio) suspension was mixed with a loopful of intact amebocytes from

Ficure 3. Changes in amebocytes caused by Vibrio. Note blebs and needle-like ex-
tensions of cytoplasm, large vacuoles and nucleus displacement. (Phase, 1400 x )

Ficure 4. Amebocytes without addition of Vibrio. The two central cells are intact.
Scme vacuole formation is present in neighboring cells. (Phase, 1400 x )

330 MV. SHIRODKAR, A. WARWICK AND: FBS BANG

siliconized roller tubes and examined on a slide by phase microscopy at 1400 x.
Controls were (a) cells with the sterile artificial sea water without NaHCO,, and
(b) cells without such dilution.

A remarkable set of changes with the pathogen was observed. Destruction of
the cell occurred within 5 minutes while the control cells, though definitely altered
inasmuch as there was some pseudopod and vacuole formation, still contained

TABLE [

Effect of toxin on amebocytes in vitro

Time following inoculation in hours

Dilution of toxin

|
a 1 5
Undiluted ae agar Seah a
x10" ats anal een rar
Loy 1077 at Sate aie ae
Vo Oe Ar ae Sta
ix LO 0 0 0
Control (sterile artificial sea water without
NaHCOs;) 0 0 0

0 = Majority of amebocytes discoid, granular.
+ = Small fraction of amebocytes extended and agranular (‘‘hyaline’’).
+-+ = Larger fraction of amebocytes extended and agranular; some disintegrated.
+-+-+ = Majority of amebocytes extended and agranular or disintegrated.

normal granules and were not greatly changed. The changes in the cells with
the Vibrio were:

(1) A rapid and uncoordinated protrusion and retraction of blebs of cytoplasm,
reminiscent of pseudopod formation.

(2) The formation of multiple intracytoplasmic vacuoles and rapid loss of
granules from the cell. The few granules which remained in the cell seemed
enlarged and lost their normal refractility; later the extruded granules also en-
larged and became translucent.

(3) Coalescence of the vacuoles into larger clear ones which occupied most
of the cell and which often displaced the nucleus. Formation of sharp, needle-like
extensions of the cytoplasm.

(4) Marked ballooning of the cell and cytolysis.

Figures 3 and 4 depict some of these changes.

The addition of the Alkaligenes-like bacteria produced no comparable changes
in intact amebocytes during the same period of time, although here, too, some
pseudopod and vacuole formation were observed.

An extracellular gel is present both during the normal clotting process and the
pathological intravascular clotting (Bang, 1956). When the pathogen was added
to the cultures, both siliconized and normal, a much heavier gel-like material
invariably appeared in the medium and persisted as long as the cultures were
observed. Microscopic examination showed no living cells, but a ropy network
which enmeshed numbers of agglutinated swollen, polymorphic, translucent, spher-

LIMULUS AMEBOCYTES AND BACTERIA 331

ical or elongated bodies which probably were altered granules. Small discrete
dark bodies were also frequently found with these.

Antibacterial action of the discoid granular cells. When the cultures of cells on
siliconized glass were first set up no sterile precautions were taken, and yet the
cultures remained free of bacteria. It was therefore of immediate interest to test
the ability of these “intact” cells to eliminate bacteria. It was found that large
numbers of the Alkaligenes-like bacterium were rapidly eliminated by these cells,
whereas cultures in which the cells had changed to the extended agranular type
(unsiliconized glassware) showed reduction and inhibition of the bacteria but that
this was temporary. Tables II and III summarize typical experiments.

TABLE [I

Effect of granular discoid amebocytes on Alkaligenes-like bacterium. Number
of colonies recovered per loopful of fluid and tissue

Hours after addition of bacteria

Number of bacteria added*

12 6 12 24 48 72 96 120
2.4 X 107 3 0 0 0 0 0 0 0
12 0 0 0 0 0 0 0
0 0 0 0
2.4 105 fi 0 0 0 0 0 0 0
12 0 0 0 0 0 0 0
0 0 0 0
2A X-108 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0

Sterile artificial sea water
without NaHCO; 0 0 0 0 0 0 0 0
Nothing added 0 0 0 0 0 0 0 0

* Varying dilutions of a fresh suspension of bacteria were added to the cultures. The figures
given were determined by the pour-plate method. Three tubes were used for each dilution but
samples were withdrawn from only two up to 24 hours. This served as a check on a possible
influence of the sampling procedure itself.

The apparent drop in bacterial count at 114 hours might be explained by an
estimated dilution of about 10°° to 10-* in the taking of small samples. However,
at 6 hours tubes of the intact system showed no bacteria and from then on none were
recovered from the granular cells. In cultures of extended cells (Table IIT)
there seemed to be some inhibition but between the 24- to 72-hour sampling the
bacterium reappeared and, by 72 hours, confluent growth was obtained. At the
highest dilution (8.5 x 101 bacteria) no organisms were recovered at any time.

The experimental and control cultures were observed microscopically and
loopfuls of fluid and tissue were examined on slides at 34%, 24 and 120 hours.
There was no persistent gelation in the uninfected system or in those cultures which
overcame the infection. The medium in these remained transparent and _ blue.

Ja2 M. V. SHIRODKAR, A. WARWICK AND F. B. BANG

TABLE III

Effect of agranular extended amebocytes on Alkaligenes-like bacterium. Number
of colonies recovered per loopful of fluid and tissue

Hours after addition of bacteria

Number of bacteria added*

2 6 12 24 48 72
S510 42 4 0 1 80 Confluent
33 9 1 0 Confluent Confluent
75 Confluent —
So 1 OF 10 1 0 0 50 Confluent
5 0) fi 100 Confluent
3 Confluent
8.5 & 10! 0 0 0 0 0 ~¢@
0 0 0 0 0 0
0 0
Sterile artificial sea water
without NaHCO; 0 0 0 0 Oe mer)
Nothing added a 0 0 0 0 0

* See footnote to Table II.

Cell changes were proportional to the number of. bacteria inoculated. Cultures
inoculated with 2.4 x 10° bacteria showed a majority of degranulated and flattened
(but living) cells; those with 2.4 x 10° organisms had approximately equal propor-
tions of degranulated and intact amebocytes, whilst the 2.4 x 10* tubes showed a
great majority of intact cells, as did both controls. Furthermore, all cultures
had apparently normal, free granules in the medium throughout the period of
observation, with the greatest numbers at the highest and fewest at the lowest
bacterial inoculum size and in the controls. Agglutinated clumps of dead bacteria
were found in the experimental tubes.

The granular discoid cell cultures were ineffective in preventing the multiplica-
tion of the pathogenic Vibrio. Cultures inoculated with 10, 100, 1000 and 10,000
viable bacteria (direct colony count) showed destructive progressive infection, all

TABLE IV

Range of antibacterial activity of intact system against
various unidentified marine bacteria

Bacterium stock number Antibacterial activity
4
10
15 (slow grower) Good
19
14 Intermediate

18 (slow grower)

12 None
be]

LIMULUS AMEBOCYTES AND BACTERIA 333

TABLE V

Degenerative changes in amebocyte granules in serum (room temperature)
(unsiliconized glassware)

Time (hours) elapsed subsequent to explantation of blood

i 3 4 173 43 72

i~

Few free in serum; | Several in serum; | Many in serum; | Profuse in serum; | Profuse in serum; | Profuse in serum;

all normal in all normal in all normal in most normal, most polymor- all polymorphic,
shape, size, color | shape, size, color | shape, size, color | some polymor- phic, translucent, | most with red
and Brownian and Brownian and Brownian phic, translucent, | some with red tinge and altered
movement movement movement with altered tinge and altered | Brownian move-
Brownian move- | Brownian move- ment
ment ment

yielding confluent growth of bacteria 24 hours after inoculation. This experiment
was repeated with the same result several times.

The effect of these “intact” siliconized cultures was then studied on several
unidentified marine bacteria (Table IV). The activity was considered good if
1000 or more bacteria were eliminated. Thus the antibacterial activity was not
limited to one strain of bacterium.

Eight different bacteria were used and three dilutions per bacterium, as de-
scribed for the Limulus nonpathogen. The activity was studied at 2, 34 and 144
hours at room temperature.

An attempt was made to determine which of the components within the blood
tissue culture system produced antibacterial activity. First, using unsiliconized
glassware, the degeneration time of the cytoplasmic granules, once they had left
the cell, was determined. Then antibacterial measurements were made with sera
containing minimal and maximal numbers of granules as well as with whole blood
homogenates at room temperature and at 0° C.

TABLE VI

Antibacterial activity of whole blood homogenate on Alkaligenes-like bacterium at 0° C*

Number of bacteria recovered per loopful
Hours after addition of bacteria
2 24 48

Fresh homogenate + 1 X 107 bacteria Confluent 102f 100
Controlt +1 107 bacteria Confluent Confluent Confluent
Fresh homogenate + 1 X 10° bacteria Confluent 16 12,
Control +1 10° bacteria Confluent Confluent Confluent
Fresh homogenate + 1 X 10? bacteria Confluent 9 7
Control +1 10* bacteria Confluent 50 Confluent

* After 48 hours at 0° C. all test and control tubes were returned to room temperature and
sO Maintained for another 24 hours, when loopfuls were plated. Confluent growth was obtained
in all cases.

t Control = 2 ml. of whole blood homogenate heated at 60° C. for 4 hour.

t These figures represent the average of two experimental culture tubes.

334 M. V. SHIRODKAR, A. WARWICK AND F. B. BANG

Table V shows the degeneration time of the cytoplasmic granules outside of
the cells to be between 17% and 43 hours subsequent to explantation. This is
interesting because of possible correlation with the 24- to 48-hour inhibition of
bacterial growth seen in the extended cell system (Table III) and also because the
discoid granular cell culture shows many perfectly normal granules 120 hours
after addition of large numbers of the nonpathogenic Bacterium 5.

In tests for antibacterial activity of sera containing varying numbers of granules
and of whole blood homogenates, equivocal results were obtained until the test for
inactivation of the bacteria by a homogenate was done at 0° C. This has been
found favorable for the measurement of the antibacterial activity of Phascolosoma
gouldu blood (Bang and Krassner, 1958).

Table VI presents the results of this preliminary experiment.

DISCUSSION

The im vitro maintenance of Limulus amebocytes was described by Loeb
(1920), who reported that hanging-drop cultures made from his “experimental
amoebocyte tissue” contained well preserved cells showing ameboid movements for
over a week, but eventually showed ‘degenerative changes.’ Cultures showed
active amebocytic migrations but no multiplication. As early as 1905 and 1906,
the same worker had observed the preserving effects, on the cells, of solutions of
both dilute acid and alkali in isotonic NaCl (Loeb, 1905, 1907, 1910) and had
found (Loeb, 1907, 1910, 1928) that the “irreversible conversion of fdiseoid
granular amebocytes to an extended, agranular, “hyalinized” state upon ex-
posure to glass could be delayed by coating the surface with Vaseline or paraffin.
The later publications of Loeb and associates (Loeb and Blanchard, 1922; Loeb,
Bierman and Gilman, 1924; Loeb, Bierman and Genther, 1925; Loeb and Genther,
1927; Loeb and Genther, 1928) dealt primarily with the stimulating effect of acid
and alkali on the cell migrations (“outgrowths”).

It is believed that the 30-day maintenance period for amebocytes in an intact
state in a siliconized system, as reported in the present paper, is the longest
recorded, as is the 36-day period of viability for amebocytes in an extended hyaline
state. Such prolonged maintenance of cells at 20°-25° C. without change of
medium is in itself interesting. Our results are in agreement with those of Loeb
in that (a) undiluted Limulus serum is the best natural medium for amebocytes
(Loeb, 1920) and (b) that no cell multiplication occurs in such cultures (Loeb,
1920 ;-Loeb, Bierman and Genther, 1925).

The siliconized system seems to be well suited for conducting the various
experiments described. The observations regarding in vitro destructive effects of
the toxin on the cells, progressive infection of the cultures, and cell changes
produced by the pathogenic Vibrio itself in vitro, correlate well with effects
previously noted (Bang, 1956) in the living animal, thus providing some indica-
tion as to the probable course of events in pathogenesis and strongly implicating the
toxin in the production of pathology following inoculation of the organism. It
should be noted here that Métalnikov (1927) demonstrated that injection of
endotoxins of several different bacteria (one, even after boiling at 100° C.) were
lethal to caterpillars of Galleria melonella.

The suggestion has been made (Bang, 1956) that amebocyte reactions to trauma

LIMULUS AMEBOCYTES AND BACTERIA _ 335

and infection are ancient and represent basic protective mechanisms. Gelation
of the blood would serve to immobilize the bacteria. During the course of
several experiments conducted in this investigation it was noted that, while the
pathogenic Vibrio was motile in the siliconized cultures, the nonpathogen was
rapidly agglutinated and killed. Thus, trauma could stimulate interactions between
cells and fluid resulting in a clot preventing further ingress of bacteria, those al-
ready within being killed by defensive factors in the blood. The results reported
in the present paper indicate that defense mechanisms against hemorrhage (gela-
tion, clotting) and against infection (production of antibacterial substance or sub-
stances) are intimately linked in the interaction between cellular and fluid elements
of Limulus blood. Loeb, as noted elsewhere (Bang, 1956), described two stages
in clotting and pointed out (1902) that fibers of clotted blood are partly formed
directly from amebocyte protoplasm. These observations were confirmed during
the course of this work. The formation of an extracellular gel during the normal
clotting process, during pathological intravascular clotting and following addition
of the pathogenic Vibrio or its toxin to siliconized or unsiliconized cultures, offers
further evidence regarding the intimacy between the protective mechanisms. The
im vitro gel persisted after addition of the pathogen and contained pathologically
altered intra- and extracellular granules in large numbers. This suggests that the
granules may be involved in some way, and is supported by the finding that they
(a) appear normal in the media of granular cultures which have eliminated the
Alkaligenes-like bacteria, their numbers increasing with increased bacterial inoculum
size; (b) appear altered not only in cultures infected with the pathogen but also
in blood from infected animals (Bang, 1956); and (c) have a degeneration time,
in serum at 20—-25° C., of 1744-43 hours (Table V) which may correlate well
with the 24-48 hour antibacterial activity seen in an extended cell system (Table
Ill). Finally, pretreatment (parenteral) of Limulus with the Vibrio toxin renders
the animal more susceptible to infection with the Vibrio as well as with certain
nonpathogenic bacteria (unpublished observation), this being presumably due to
a lack produced by the toxin, of normal amebocytes and, possibly, of granules.

It has been shown that the intact system alone can eliminate large numbers
of marine bacteria of more than one strain (Table IV), which lends support to
the thesis that the phenomenon is truly antibacterial and not simply a chance ob-
servation. The extended cell and whole blood homogenate systems merely exhibit
different levels of temporary antibacterial activity. An explanation may be that
the intact system provides a steady, pump-like activity on the part of the healthy
granular amebocytes leading to the release of some thermolabile intracellular
component(s) into the serum long enough to kill all of the bacteria. The fact that
as few as 10-100 pathogenic organisms can multiply and overcome the system may
well be ascribed to the inactivating action of their toxin on the hypothetical
intracellular component(s) and, possibly, on the granules. Its destructive action
on the amebocyte itself has already been noted.

SUMMARY

1. For effective im vitro study of the defense mechanisms against bacterial
infection of Limulus polyphemus an unsiliconized and a siliconized culture system
for blood cell maintenance were developed. The former contained viable but

336 M. V. SHIRODKAR, A. WARWICK AND F. B. BANG

partially degranulated amebocytes for at least 36 days, whilst the latter system kept
the cells in an intact state resembling the in vivo for up to 30 days. Replenishment
of the culture medium (undiluted Limulus serum) and use of antibiotics were found
unnecessary. No cell multiplication was noted.

2. The reactions of the blood elements to certain nonpathogenic and pathogenic
marine bacteria (and to their toxin) were studied. When a pathogenic Vibrio was
added to the siliconized system of granular, discoid cells, persistent gelation of the
medium regularly accompanied bacterial growth, the gel containing patho-
logically altered amebocytes and granules. While gelation is characteristic for
shed normal Limulus blood, intact cultures—both uninfected as well as those which
overcome the nonpathogenic Alkaligenes-like bacterium—often showed either
absence or ephemeral presence of gel in the medium. Such cultures also contained
normal granules and showed large clumps of agglutinated dead bacteria con-
sistently, suggesting involvement of the gelation process in immobilization and,
perhaps, killing of the invading organism. The culture systems were active against
several strains of marine bacteria. Experiments with a nonpathogenic Alkaligenes-
like bacterium showed that only the intact, granular cultures showed rapid (6
hours at 20-25° C.) and permanent elimination of large numbers (inoculum sizes
up to several million), the unsiliconized extended cell and whole blood homogenate
systems merely exhibiting different levels of evanescent activity.

3. The successful infection of the intact system, by very small numbers (10-
100), of pathogenic Vibrio clearly established its in witro pathogenicity. Its
thermostable toxin caused prompt changes in the morphology of the intact
amebocytes in culture, even at a 10°? dilution, and a phase microscopic study of
the action of the Vibrio itself on the cells revealed rapid cellular alterations
culminating in cytolysis.

LITERATURE CITED

Banc, F. B., 1956. A bacterial disease of Limulus polyphemus. Bull. Johns Hopkins Hosp.,
98: 325-351.

Banc, F. B., ann S. M. Krassner, 1958. Antibacterial activity of Phascolosoma gouldu
blood. Biol. Bull., 115: 343.

LANKEsTER, E. R., 1884. On the skeleto-trophic tissues and coxal glands of Limulus, Scorpio
and Mygale. Quart. J. Micr. Sci., 24: 129-162.

Logs, L., 1910. Uber die Blutgerinnung bei Wirbellosen. Biochem. Zeitschr., 24: 478-495.

Logs, L., 1902.. On the blood lymph cells and inflammatory processes of Limulus. J. Med.
Res., 7: 145-158.

Loes, L., 1920. The movements of the amoebocytes and the experimental production of amoe-
bocyte (cell-fibrin) tissue. Washington Univ. Studies, 8: 3-79.

Logs, L., 1905. Studies on cell granula and amoeboid movements by the blood cells of Limulus.
Univ. Pennsylvania Med. Bull., 18: 91-95.

Loges, L., 1907. Untersuchungen iiber die Granula der Amoebozyten. Folia haemat., 4:
313-322.

Logs, L., 1910. tber den Einfluss von chemischen und physicalischen Umgebungsanderungen
auf die Blutzellen von Limulus, und insbesondere auf ihre Granula. Pfliger’s Arch.,
131: 465-508.

Lors, L., 1928. Amoebocyte tissue and amoeboid movement. Protoplasma, 4: 596-625.

Loes, L., AnD K. C. Brancuarp, 1922. The effect of various salts on the outgrowth from
experimental amoebocyte tissue near the isoelectric point and with the addition of acid
and alkali. Amer. J. Physiol., 60: 277-307.

Logs, L., H. BrerMANn AND I. P. Gentuer, 1925. The effect of various ions on the experi-
mental amoebocyte tissue of Limulus and their interaction with other variable factors.
Arch. exper. Zellforschg., 1: 17-288.

LIMULUS AMEBOCYTES AND BACTERIA oon

Logs, L., J. M. BrermMAN AND E. GitmAn, 1924. The effect of acid on the amoebocyte tissue
of Limulus in tissue cultures. Proc. Soc. Exp. Biol. Med., 21: 245-248.

Logs, L., ann I. T. GENTHER, 1927. The effect of alkali on amoebocyte tissue of Limulus.
Arch. exper. Zellforschg., 5: 83-110.

Logs, L., AND I. T. GENTHER, 1928. The effect of acid on amoebocyte tissue of Limulus.
Arch. exper. Zellforschg., 5: 355-378.

Mératnixkoy, S., 1927. L’infection microbienne et l’immunité chez la mite des abeilles Galleria
melonella. Monographies de l'Institut Pasteur, Masson et Cie, Editeurs, Paris.

SHriropKaR, M. V., F. B. BANG ann A. Warwick, 1958. Antibacterial action of Limulus
blood in an in vitro system. Biol. Bull., 115: 341.

Warwick, A., AND F. B. Bane, 1957. Survival of marine invertebrate cells in tissue culture
(Limulus and Ostrea). Biol. Bull., 113: 334.

THE, SELECTIVE. AGGU MULATION. OF; BLOOD, PROTEINS by iam
OOCYTES OF SATURN Mem Eis

W Tac AME (Else BR
Zoological Laboratory, University of Pennsylvania, Philadelphia 4, Penna.

The growing oocyte of an insect accumulates proteins from the blood, rather
than relying exclusively on the synthetic activities of the ovaries. Wigglesworth
(1942) demonstrated in a number of blood-sucking insects that the modified
hemoglobin which occurs in the oocytes is derived from the blood. More recently,
an immunologically detected protein occurring in the blood of female saturnid
moths has been found to leave the blood during the maturation of the ovaries and
to accumulate in the oocytes (Telfer, 1954). That such a process might be entailed
in the growth of oocytes other than those of insects is suggested by the anti-
genic and physical similarities between oocyte and serum proteins of chickens
(Schechtman, 1947; Nace, 1953; Schjeide and Urist, 1956) and of frogs (Cooper,
1948; Flickinger and Rounds, 1956).

A variety of other animal cells can accumulate soluble proteins from their
environment and, of these, one of the most thoroughly analyzed has been the
amoeba (Holter, 1959). The mechanism of protein uptake in the amoeba has been
proposed to include two essential features: adsorption of the protein on the surface
of the cell (Marshall, Schumaker and Brandt, 1959), and a subsequent infolding
of the cell surface to form vacuoles (Mast and Doyle, 1934; Chapman-Andresen
and Prescott, 1956). The latter process was first described by Lewis (1931)
who observed mammalian cells in tissue culture ingesting their culture medium.

An analysis of protein accumulation by the moth oocyte has been undertaken.
Included in this analysis is an attempt to determine whether the mechanism in-
volved is comparable to that proposed for the amoeba. Since, unlike the amoeba,
the moth oocyte cannot be observed microscopically during the process of protein
accumulation, studies of the mechanism must in this case be based on less direct
methods. Evidence is presented here that most proteins in the blood of the moth
enter the oocyte to some extent but that, from a quantitative point of view, the
mechanism is highly selective. The bearing of this and other evidence on the
nature of the protein accumulating mechanism will be considered in later papers.

MATERIALS AND METHODS

The capacities of seven proteins to enter the oocytes from the blood of the
Polyphemus moth were assayed by means of a quantitative application of Oudin’s
antiserum-agar technique. Two of the proteins occur in the blood of the Cecropia
moth and normally accumulate in the oocyte of that species. This has already been
demonstrated for one protein, which will be referred to as the “female protein,”

1 This work was performed in part while the author was a Summer Fellow of the
Lalor Foundation, and in part with the support of Phi Beta Psi Sorority.

338

BLOOD PROTEINS IN OOCYTES 339

and it is demonstrated here for a second protein found in the blood. The five
additional proteins were derived from the blood or oocytes of animals other than
insects.

1. Injection of foreign proteins

The protein preparations were introduced into diapausing female pupae of the
Polyphemus moth which had previously been chilled for several months at 6° C.
in order to activate the endocrine mechanism responsible for stimulating their
transformation into moths (Williams, 1956). The Cecropia proteins were in-
troduced into pupae by the transfusion technique previously described (Telfer,
1954). Proteins of non-insectan origins were injected into the haemocoel from
a tuberculin syringe through a No. 28 hypodermic needle. The volume of the
protein solution, in milliliters, was equal to one-twentieth of the weight of the
pupa in grams, in the case of the non-insectan proteins, and to one-fifth of the
weight of the pupa in the case of Cecropia blood. In order to avoid melanin
formation which can occur in the blood in response to surgical schock, ap-
proximately 0.1 mg. of phenylthiourea was included in the solution introduced
into each animal.

2. Preparation of the blood and oocyte fluid of Polyphemus moths

Oocyte growth occurs in Polyphemus primarily during the final half of the
period of transformation of the pupa into the moth. The injected animals were
bled at the end of this period. The blood was collected in test tubes and stored
under mineral oil at — 20° C. The ovaries were thoroughly washed with 0.15
M NaCl, blotted and placed in a mortar. Most of the oocytes were shaken free
of the ovarioles during the washing procedure, and the chorions of these oocytes
were broken by pressure applied with a pestle. The yolky contents that were thus
squeezed out of the chorions were centrifuged at 20,000 X g for thirty minutes at
0° C. After centrifugation, a clear, yellow, aqueous layer, which will be referred
to as oocyte fluid, lay between a sediment and a floating, particulate layer. The
clarified oocyte fluid occupied approximately three-quarters of the volume of the
centrifuged oocyte contents. It was analyzed, along with the blood, for its content
of the protein in question. Melanin formation was prevented in both blood and
oocyte fluid by the addition of crystalline phenylthiourea.

3. Preparation of the protein solutions for injection into Polyphemus pupae

Commercial preparations of chicken ovalbumin, bovine gamma globulin and
bovine serum albumin (all Pentex) were dissolved in 0.15 M NaCl at neutral pH.
A 2% solution of a protein preparation was used for injection into pupae.

In addition, extracts of chicken’s egg yolk and of lobster oocytes were studied.
The chicken yolk extracts were prepared by washing several yolks free of the
white and then breaking them in several times their volume of 0.15 M NaCl. The
mixture was dialyzed against saline solution and finally centrifuged at 20,000 x g.
The final preparation, which was an opalescent yellow solution, was injected into
pupae without further dilution. Judging from the ionic strength of the extraction
medium, this preparation should have included the protein complexes, lipovitellenin
and livetin (Fevold, 1951).

340 WILLIAM H. TELFER

Lobster oocytes were obtained from ovaries dissected from living lobsters
(Homarus americanus). Developing lobster oocytes, after passing through yellow
and pale green phases, finally attain a dark green color. The oocytes extracted
here were at least 2 mm. in diameter and were dark green. They were loosened
from the ovarian wall by agitation in 0.15 M NaCl. After being washed several
times, the oocytes were disintegrated in a Virtis homogenizer in several times
their volume of saline solution. The extract, after dialysis against saline and
centrifugation at 20,000 x g, was a clear, dark green solution. The green color
of lobster oocytes is due to ovoverdin, the carotenoid-protein complex studied by
Stern and Salomon (1937).

Finally, two proteins of Cecropia blood were studied. These were initially
discovered by means of immunological techniques and were designated as antigens
3 and 7 (Telfer and Williams, 1953). Polyphemus pupae were transfused with
whole blood of Cecropia female pupae, which contains both of these proteins.

Most of the non-insectan protein solutions and extracts studied were toxic to
some of the injected pupae and in these cases death occurred within three days
after the injection. Such pupae never initiated their development into moths.
The surviving animals, on the other hand, transformed into moths in the usual
three-week period of time. Their ovaries produced eggs which appeared to be
normal in size and structure, although sometimes reduced in number. Three
preparations, bovine fibrinogen, phosvitin from hen’s eggs and hemocyanin from
lobster blood, were toxic to all pupae injected, and efforts to study the uptake of
these proteins by the oocytes were therefore abandoned.

4. Immunological procedures

Antisera were obtained from rabbits. The reaction between a protein and its
antibodies was studied by means of the antiserum-agar technique developed by
Oudin (1948) and previously applied to the study of insect blood and oocyte
proteins (Telfer and Williams, 1953; Telfer, 1954). A rabbit antiserum was
solidified with agar in glass tubes (3 mm. i.d.) and was then overlayered with
the blood, oocyte fluid or other sample whose content of the homologous protein
was being tested. The antiserum was diluted to an appropriate level with 0.15
M NaCl buffered at pH 7.0. The rate of advance of the zone of antigen-antibody
precipitation through the antiserum-agar (distance travelled/square root of time)
was used as an index to the concentration of the protein serving as the antigen.
The rate of advance of a zone of precipitation was converted to the relative con-
centration of the corresponding protein in the antigen layer by reference to a
standard curve. Standard curves were constructed for each protein studied by
determining the rate of advance of the zone of precipitation produced by a series
of dilutions of a solution of the protein. _

Precautions were taken which were previously recommended (Preer and Telfer,
1957) for reducing the effects of substances not involved in the reaction on the
rate of advance of the zone of precipitation: in order to eliminate errors due to
viscosity differences, the composition of the antiserum-agar was kept constant in
any set of tests; and all tests were set up so that convection mixing occurred in the
antigen layer (tubes vertical, antigen layer above the antiserum-agar, and the
density of the antigen layer greater than that due to the diffusible components of
the antiserum-agar ).

BLOOD PROTEINS IN OOCYTES 341

In addition, it was demonstrated that samples of oocyte fluid, blood and 0.15
M NaCl, which were known to contain equal concentrations of an antigen,
produced zones of precipitation with the same rates of advance (Table 1). There-
fore, with the exception of possible special cases, the Oudin test as applied here
provided a valid criterion for comparing antigen concentrations in the blood and
oocyte fluid of the moth.

5. Antigenic characteristics of the foreign proteins

All of the preparations of non-insectan proteins, except the lobster oocyte
extract, on reacting in Oudin tubes with a 1:3 dilution of their homologous
antisera produced a single dense zone of precipitation and, in several cases, ad-
ditional weak zones. In order to achieve maximum sensitivity for the detection of

TABLE |

Comparison of the oocyte fluid and blood of Polyphemus moths as solvents for ovalbumin
in antiserum-agar tests. Antigen solutions were layered
over a 1:3 dilution of anti-ovalbumin serum

Rate of advance of the zone of precipitation when ovalbumin was diluted with
Dilution of a 2% ovalbumin
solution in antigen layer

Oocyte fluid Blood 0.15 M NaCl*
X 10-3 cm.-sec.-? xX 1073 cm.-sec.~? < 10-3 cm.-sec.-?
te OF on 3.3
is 2 2.9 Sell —
1:4 Ded 2.8 2.8
1:8 a) 25 —
1:16 Dad Died Dee
HISD 1.9 1.9 —
1:64 1.6 1.6 led
12128 15S 13 —
1:256 pa 1a ail

* Since 0.15 MW NcCl is less dense than the 1:3 dilution of antiserum, the Oudin tubes were
placed horizontally in this case in order to assure convection in the antigen layer.

these antigens in Polyphemus blood and oocyte fluid, the antisera were diluted as
much as was consistent with a clear visualization of the densest zone of precipitation.
As a result, the weaker zones were generally not visible and information was thus
obtained concerning only the antigen producing the densest zone. Two dense
zones were visible in the reaction between the lobster oocyte extract and its
homologous antiserum. Since the densities and rates of advance of these two
zones were extremely similar, they were treated as a single zone and the rate of
advance was determined only for the faster of the pair.

RESULTS

1. Demonstration of an oocyte antigen which 1s indistinguishable from a carotenoid
protein of the blood

Evidence that Cecropia oocytes contain a protein which is antigenically similar
to the female protein of the blood was described earlier (Telfer, 1954). Ob-

342 WILLIAM H. TELFER

servations are described here which indicate that, in addition to the female protein,
one of the several carotenoid proteins of the blood also has an antigenic counterpart
in the oocyte.

Adult antiserum d, a pooled antiserum obtained from three rabbits which had
been immunized with an extract of Cecropia female moths, produced two zones of
dense precipitation when overlayered in Oudin tubes with either the oocyte fluid
or the blood of female moths. One of these zones was not ordinarily produced by
male blood and was thus identifiable as resulting from the precipitation of antigen 7,
the female protein. The second, more slowly moving zone was due to precipitation
of a substance previously designated as antigen 3 and characterized as a carotenoid
protein. This identity was established in mutual dilution tests (Telfer and

TABLE II

The rates of advance of the two major zones of precipitation appearing when antiserum-agar
containing adult anti-serum d (1:3) was overlayered with graded mixtures
of oocyte fluid and the blood of female pupae

Composition of the
antigen layer

% blood Rate of advance of the zone of precipitation due to
( % oocyte fluid ) Female protein Carotenoid protein
= 2-01 5< 1052 cur=sec 1.26 107° Cm=secin
pis 2.04 1.29
12.5
15 Zt 1.29
25
a0 DAG 1.35
50
22 DES 1.41
75
Us 229 1.44
87.5
0
—— 2.28 1.49
100

Williams, 1953) in which aliquots of a sample of pupal blood were layered over
graded mixtures of adult antiserum d and adult antiserum a, the antiserum whose
reaction initially distinguished the carotenoid protein.

That the two zones of dense precipitation produced by oocyte fluid, on the one
hand, and female blood, on the other, were due to related antigens was indicated
by the results of mutual dilution tests. When graded mixtures of pupal female
blood and oocyte fluid were layered over adult antiserum d, the two zones of
dense precipitation behaved as shown in Table II. The rate of advance of each
zone shifted gradually from that in the tube containing only blood to that con-
taining only oocyte fluid. There was no detectable splitting, fusion, or crossing of
the two zones. With regard to the precipitations visible in these tests, one can

BEOODYPROTEINSAN OOCYTES 343

therefore conclude that, as in the case of the female protein zone, the antibodies
which form a single zone of precipitation with a carotenoid protein of the blood
also form a single zone of precipitation with an antigen diffusing from the oocyte
fluid.

' Absorption tests provided additional evidence that the oocyte fluid contains an
antigen similar to the carotenoid protein of the blood. Immunization of four
rabbits with oocyte fluid and of three rabbits with female pupal blood led in every
case to the production of antibodies which, according to mutual dilution tests with
adult antiserum d, precipitated both the female and carotenoid proteins of the blood.
All of these antisera, on absorption with an equal volume of a 1:10 dilution of

TASES TT
Absorption of antisera against Cecropia oocyte fluid with the blood of female pupae

Zones of precipitation produced when oocyte fluid was layered over agar containing

Antiserum ~
(final dilution
1:3 in all cases)

Unabsorbed antiserum Antiserum absorbed with blood of female pupae

Intensity of
precipitation

Intensity of

precipitation Rate of advance

Rate of advance

X 10-3 cm.-sec.# x 10-3 cm.-sec.7?
Oocyte fluid a Weak Daihg Weak 2.20
Strong* 2.04 Weak 1.24
Strong** 1.16 Weak 0.81
Oocyte fluid } Weak 226 Weak 2.19
Strong* 1.59 Weak HSS
Strong** 1.36
Oocyte fluid ¢ Weak vAsligt Weak Desh
Strong* 2.10
Weak 1.36
Strong** 0.99
Oocyte fluid d Weak 2.30 Weak 2.39
Strong* SZ
Strong** 125

* Zone of precipitation of the female protein (antigen 7).
** Zone of precipitation of the carotenoid protein (antigen 3).

oocyte fluid, lost their capacity to precipitate either blood or oocyte antigens in
Oudin tests. Thus the oocyte fluid contains antigenic counterparts of all the sub-
stances, including the carotenoid and female proteins, detected in the blood with
these four antisera.

Absorption of the antisera with female blood led to different results. As would
be anticipated, the absorption of female blood antisera with pupal female blood
removed all of the antibodies visibly precipitating either blood or oocyte antigens in
Oudin tests. However, antisera against oocyte fluid, after absorption with female
blood, retained the capacity to form from one to three zones of weak precipitation
when overlayered by oocyte fluid (Table IIT).

344 WILLIAM H. TELFER

Absorption tests thus indicated that the oocyte contains, in addition to the blood
antigens precipitated by these antisera, at least three antigens which have not been
detected in pupal female blood. The fastest zone produced when oocyte fluid
reacted with the blood-absorbed antisera had, in each case, an intensity and rate of
advance similar to the fastest zone produced by the unabsorbed antiserum. Oocytfe
antisera a and b produced additional zones whose failure to appear in the reactions
with unabsorbed antisera was probably due to masking by the faster and denser
zone of female protein precipitation (Table III). An alternative possibility of
interest to the present problem is that at least one of these zones could have been
due to residual antibodies against the oocyte’s carotenoid protein—antibodies whose
failure to react with the carotenoid protein of the blood would thus indicate an
antigenic difference between the carotenoid proteins of the blood and of the oocyte.
If this were the case, however, the amount of antibody involved was small in
comparison with the total amount of carotenoid protein antibodies present in un-
absorbed antiserum. The weak zones produced by a 1:3 dilution of the unabsorbed
antiserum were comparable in intensity to the carotenoid protein precipitation in
1:30 dilutions of unabsorbed antiserum. Thus, at least 90% and possibly all of
the antibodies precipitated by the carotenoid protein of the oocyte were also
precipitated by the carotenoid protein of the blood.

Mutual dilution and absorption tests therefore led to the conclusion that the
oocyte fluid contains a substance which is antigenically similar to and probably
identical with antigen 3, a carotenoid protein of the blood.

2. The relative concentrations of the carotenoid and female proteins in the oocyte
fluid and the blood of Cecropia moths

The concentrations of the carotenoid and female proteins in both the oocyte
fluid and the blood were estimated from the rates of advance of their respective
zones of precipitation. When the oocyte fluid and blood collected simultaneously
from the same moth were layered over a 1:3 dilution of oocyte antiserum c in
Oudin tubes, the zones of carotenoid protein precipitation advanced through the
antiserum-agar at approximately equal rates. The zone of female protein pre-
cipitation advanced through the agar considerably faster from the oocyte fluid than
from the blood, a result which is in accord with earlier observations on the female
protein zone and which can be attributed to a disparity in concentration of this
protein between the blood and the oocyte fluid (Telfer, 1954). Reference of the
rates of advance of these zones of precipitation to standard curves yielded the
relative antigen concentrations summarized in Table IV. This interpretation of
the rates of advance led to the conclusion that the carotenoid protein is approxi-
mately equally concentrated in the oocyte fluid and blood of the thoths, while the
femael protein is, on the average, twenty-eight times more concentrated in the
oocyte fluid than in the blood.

3. The accumulation of Cecropia blood proteins by Polyphemus oocytes

Evidence that the carotenoid protein of the oocyte has its origin in the blood,
rather than being synthesized in the ovary, was obtained from experiments in
which the blood of Cecropia pupae was injected into female Polyphemus pupae.

BLOOD PROTEINSINI@OCY TES 345

TABLE IV

Relative concentrations of the female and carotenoid proteins in the oocyte fluid
and in the blood of female Cecropia moths

Concentration ratio

ee ee (‘ceevteid)
Animal no. blood
Oocyte fluid Blood Oocyte fluid Blood Fem. prot. Carot. prot.

1 Ta5 0.03 1.00 0.77 38 las)
2 0.93 0.03 0.85 0.81 38 teal
3 1205 0.06 1.45 125 18 2
4 0.90 0.04 0.81 0.81 21 1.0
5 0.76 0.03 — — 29 —
6 i15 0.04 0.66 1.20 Sill 1.8
7 1.05 0.06 eZ £220 19 1.9
8 1.05 0.03 0.71 1235 Sf 0.5

Mean 28 ip

When these animals had transformed into moths, their oocyte fluid and blood
were obtained and were layered over adult antiserum d which had previously been
absorbed with the blood of Polyphemus female pupae. After absorption in this
manner, the antiserum retained its capacity to produce its two major zones of
precipitation when overlayered with female Cecropia pupal blood, although, due to
the removal of some antibodies, the zones were somewhat reduced in intensity.
Thus Polyphemus blood contains substances which are antigenically similar to both
the female and carotenoid proteins of Cecropia but which are distinctive enough
so that they are unable to precipitate all of the antibodies against the Cecropia
proteins.

Studies of the amount of the two Cecropia proteins in Polyphemus blood or
oocyte fluid require that the rate of advance of the Cecropia protein’s zone of

TABLE V

The relative concentrations of Cecropia’s carotenoid and female proteins
in the oocyte fluid and blood of Polyphemus female moths

Concentration ratio

Relative concentration of Relative concentration of oocyte fluid
female protein carotenoid protein ( Bigod )
Animal no. ss
Oocyte fluid Blood Oocyte fluid Blood Fem. prot. Carot. prot.
1 1 | 0.05 1.0 Oy 22, 1.4
2 1.1 0.05 0.5 0.6 De 1.0
3 j Ba 0.05 0.8 0.7 30 al
4 0.7 0.05 0.8 0.4 14 2.0
< 1.0 0.05 0.4 0.8 20 0.5

346 WILLIAM H. TELFER

precipitations be unaffected by the Polyphemus proteins present in the antigen
layer. This requirement was found to be fulfilled: the blood of Polyphemus
female pupae was indistinguishable from 0.15 M NaCl as a diluent for the two
Cecropia antigens reacting with the Polyphemus-absorbed antiserum. Thus, the
Oudin test could be used to indicate not only the presence, but also the concentra-
tions of these two Cecropia proteins in Polyphemus moths.

Both the blood and the oocyte fluid of Polyphemus moths which had been
transfused with Cecropia blood produced the two characteristic zones of dense
precipitation in Oudin tests with the Polyphemus-absorbed adult antiserum d.
The carotenoid and female proteins of Cecropia therefore found their way into
the Polyphemus oocytes from the blood, an observation which has already been
reported for the female protein.

When the rates of advance of the two zones of precipitation were converted
to the relative concentrations of their respective antigens, the results recorded in
Table V were obtained. The distribution of Cecropia’s female and carotenoid
proteins between the blood and oocytes of Polyphemus was approximately the
same as their normal distribution in Cecropia (Table IV). This result suggests
that neither the carotenoid protein nor the female protein of the oocyte fluid is
synthesized in appreciable amounts in the ovary, but that both are derived
primarily, and perhaps exclusively, from the blood.

4. Similarities between other antigens of the oocyte fluid and blood

The demonstration that the carotenoid and female proteins of the oocyte fluid
are derived primarily from the blood raises the question as to whether other blood
proteins participate in oocyte formation. ‘The reactions of Cecropia oocyte fluid
with adult antiserum a and with larval antiserum c suggest that this is the case.
These two antisera have been shown to react with a minimum of six different
Cecropia blood proteins, one of which is the carotenoid protein studied here (Telfer
and Williams, 1953). In Oudin tests, each antiserum produced up to seven zones
of precipitation when overlayered by Cecropia oocyte fluid. After absorption with
equal volumes of a 1:10 dilution of oocyte fluid, none of the antisera was able to
produce zones of precipitation when overlayered with Cecropia pupal blood.
Substances which are antigenically similar to at least five blood proteins, in ad-
dition to the female and cartenoid proteins, are therefore present in the oocyte
fluid. This result is consistent with the possibility that many, or even all, of the
proteins normally present in the moth’s blood enter the oocyte fluid to some
extent, although synthesis in the ovary has not yet been ruled out in all these cases.

5. The accumulation of non-insectan proteins by Polyphemus oocytes

Less equivocal evidence that the oocytes can accumulate blood proteins other
than the female and carotenoid proteins was obtained from experiments in which
five antigens which are completely foreign to the moth were injected into the blood.
Neither the blood nor the oocyte fluid of uninjected moths produced zones of
precipitation in Oudin tests with antisera against the five non-insectan antigens
studied. Therefore, the appearance of the antigenic activity of the injected protein
preparations in the oocyte fluid could be interpreted as due to accumulation from
the blood, rather than to synthesis in the ovary.

BLOOD’ PROTEINS‘ IN. OOCYTES 347

Of the five antigens injected into Polyphemus pupae, four were detectable in
Oudin tests in both the blood and the oocyte fluid of the subsequently emerging
moths. Only the major antigenic component of the chicken’s yolk extract was
undetectable in the oocyte fluid. The conclusion thus seems warranted that most
proteins present in the blood are accumulated to some extent by the oocyte.

The relative concentrations of the foreign antigens in the oocyte fluid and blood
were estimated from the rates of advance of their zones of precipitation (Table VI).
The concentration of Cecropia’s female protein in Polyphemus oocyte fluid, relative
to its concentration in the blood at the conclusion of egg formation, was approxi-
mately 200 times greater than that of the non-insectan proteins. Thus, while

TABLE VI

Relative concentrations of foreign antigens in the oocyte fluid
and blood of Polyphemus moths

Concentration ratio (oocyte fluid/blood)

Antigen Number
of cases
Mean Range

Cecropia female protein 21 14-30 5
Cecropia carotenoid protein ia 0.5-2.0 5
Bovine gamma globulin 0.19 0.08—0.60 6
Bovine serum albumin 0.07 0.06-—0.11 7
Ovalbumin 0.07 0.04—0.13 4
Lobster oocyte antigen (probably ovoverdin) 0.10 0.06-0.16 7
Chicken yolk antigen (probably lipovitellenin) | Undetectable in 0.03-0.20 4

oocyte fluid

most proteins in the blood appear to enter the oocyte to some extent, different
proteins vary greatly in their capacity to do so.

6. Localization of the blood proteins within the oocyte

The oocyte fluid analyzed in the Oudin tests described here consisted primarily
of a solution derived from the breakdown of the yolk bodies. When whole oocytes
were centrifuged at 20,000 X g, the strata which were formed within the chorion
appeared different from the strata produced by the centrifugation of crushed
oocytes. While a centripetal layer presumably consisting of fatty materials was
formed in both cases, a non-particulate intermediate layer comparable to the
clarified yellow fluid of crushed oocytes was not produced in whole oocytes. More
than 90% of the volume of centrifuged whole oocytes was occupied by a sediment
which, on dissection, was seen to consist of yellow spheres whose diameter ranged
up to greater than 20 microns. Since more than three-quarters of the volume of
crushed oocytes forms a clear, yellow liquid on centrifugation, it appears that the
yellow spheres of undamaged oocytes are liquified when the eggs are crushed with
a mortar and pestle. The nature of these particles, which we presume to be the
yolk spheres of the cytologists (Wilson, 1925), will be considered in more detail
in a later paper.

While the oocyte fluid presumably contained the non-particulate ooplasm and

348 WILLIAM H. TELFER

fractions of any less voluminous particles which may have been damaged when the
yolk was being expressed from the chorion, its primary constituent must have been
lysed or liquified yolk spheres. The possibility thus appears that the proteins
derived from the blood may be localized in the yolk spheres of the oocyte. This
proposal is confirmed by the fact that the yolk spheres are colored bright yellow,
for at least one of the proteins found here to be accumulated by the oocyte is a
carotenoid protein which contributes to the yellow color of the blood.

DISCUSSION

The oocytes of saturniid moths have now been demonstrated to accumulate
the female protein and a carotenoid protein from the blood. Antigenic similarities
between the oocyte fluid and the blood suggest that other proteins which normally
occur in the insect’s blood may also be accumulated by the oocyte, although in
these cases synthesis within the ovary has not been excluded. In addition, four
proteins of non-insectan origin, after injection into the blood, have been found in
the oocyte. Thus far, the only protein which has not been detected in the oocyte
after injection into the blood is one derived from chicken yolk extract. Whether
some feature of the uptake mechanism prevents the accumulation of this molecule
or whether, in the course of accumulation, its antigenic properties are destroyed
has not been determined. In experiments similar to some of those described
here, Knight and Schechtman (1954) demonstrated that foreign proteins pass
from the serum to the ovum of chickens.

There are large differences in the quantitative distribution of the various
proteins between the oocyte fluid and blood. The data (Table VI) suggest that
the mechanism of accumulation is uniquely adapted to the removal of particular
proteins from the blood, and that others, such as the non-insectan proteins injected
into the pupa may, in effect, be carried into the oocyte as contaminants. This
interpretation presumes that the oocyte fluid: blood concentration ratios measured
here are determined primarily by the capacity of the protein in question to penetrate
the various cellular layers of the ovary and the surface of the oocyte. However,
other events occurring within the injected insect could also affect this ratio and
these must be considered in any attempt to apply the ratio of concentrations to an
analysis of the protein accumulating mechanism.

Structural alterations which occurred in a protein molecule during or after
its entry into the oocyte could, by affecting the rate of advance of the corresponding
zone of precipitation, lead to gross errors in determinations of its distribution ratio.
Such alterations could act in two ways: by altering the protein’s antigenic properties
and thus its capacity to combine with antibodies, or by altering its diffusion
coefficient. A significant alteration in antigenic properties during the accumula-
tion of the carotenoid and female proteins can be ruled out, since as is shown
here, the blood and oocyte representatives of these proteins are extremely similar,
and are possibly identical in their capacity to precipitate antibodies. It is also
unlikely that the antigenic activities of the non-insectan proteins studied were
grossly modified during their accumulation by the oocytes. In these cases, the
zones of precipitation produced by the blood and oocyte fluid of the injected animals
appeared to be equally dense and thus to entail the precipitation of comparable
amounts of antibody. Antigenic changes which might not have been detectable by

BLOOD PROTEINS IN OOCYTES 349

this criterion could not account for the 200-fold differences observed in the oocyte
fluid: blood concentration ratios of the proteins studied here. In a later paper,
data will be presented which indicate that a change in diffusion coefficient does not
occur during the transfer of either the female or carotenoid protein from the blood
to the oocyte. It is therefore improbable that the distribution ratios measured for
the seven proteins under consideration were subject to errors resulting from
molecular alterations associated with the process of accumulation. Knight and
Schechtman (1954) found that, in the chicken also, foreign proteins injected into
the serum were not detectably altered by transmission to the ovum.

A protein’s distribution between oocyte fluid and blood could be affected by
secondary processes which are unrelated to the transfer mechanism: processes such
as adsorption on an insoluble structure in the oocyte, or a selective destruction in
the blood or other tissues after accumulation by the oocyte has been completed.
While such processes may occur to some extent, the following observations indicate
that the difference between the oocyte fluid: blood concentration ratios of the fe-
male and carotenoid proteins does in fact reflect a difference in their capacity to
penetrate the oocyte. During the period of maximal oocyte growth, the con-
‘centration of female protein in the blood decreases 80%, while that of the carotenoid
protein decreases only 20-25% (Telfer and Rutberg, 1960)—a difference which
is readily accounted for by the female protein’s being removed from the blood
at a faster rate than the carotenoid protein. In addition, selective accumulation of
the female protein can explain the effects of ovariectomy on the blood of female
moths (Telfer, 1954). The electrophoretic pattern of the blood of moths which
had been ovariectomized as pupae was similar to that of the blood of normal moths
except for the fact that the moths which had not been allowed to produce eggs
contained excessive amounts of a component possessing the electrophoretic mobility
of the female protein. Comparisons of the female and carotenoid proteins thus
established that the moth oocyte does not accumulate all proteins in proportion to
their concentrations in the blood, but that the transport mechanism is characterized
by a high degree of selectivity.

Selectivity of accumulation suggests that the protein transport mechanism may
be more complex than could be accounted for alone by free diffusion from the
blood to the oocyte. In two other cases of protein transport, it has been necessary
to conclude that a protein combines temporarily with some element of the transport
mechanism. Such a step has been postulated in order to explain the selectivity
and competition demonstrated in the transmission of maternal antibodies to the
foetus and the new-born of a number of species of mammals (Brambell, Halliday
and Morris, 1958). In addition, adsorption of protein on the surface of an amoeba
undergoing pinocytosis is convincingly suggested by kinetic and cytological evidence
(Schumaker, 1958; Brandt, 1958).

Several mechanisms could be proposed to account for the selectivity of protein
accumulation by the oocyte of the moth. Included among these is a temporary
adsorption on the cell surface, as has been proposed for the amoeba. Adsorption
on an intracellular structure such as the yolk spheres is not ruled out, however,
nor is selective permeability of one of the various cellular envelopes which separate
the oocyte from the blood. Distinguishing which of these possibilities may be
correct must await further characterization of the uptake mechanism.

350 WILETAM’ Hy, TEHELEER

A final point which the protein accumulating mechanism of the moth oocyte
may have in common with that of the amoeba is the intracellular disposition of
the acquired protein. In both cases, the protein is associated to a major extent
with cytoplasmic particles: pinocytic vacuoles in the amoeba and yolk spheres in
the oocyte. Proteins with the antigenic properties of serum proteins are also
associated with the yolk platelets of the amphibian oocyte (Cooper, 1948; Flickinger
and Rounds, 1956). It may thus prove possible to establish a homology between
these cytoplasmic structures with regard to their function in protein accumulation
from the environment.

SUMMARY

1. At least seven proteins detectable by immunological techniques in the blood of
the Cecropia moth have antigenic counter-parts in the oocytes produced by the
female moth. While several oocyte antigens are undetectable in the blood of the
female pupa, all of the proteins thus far observed in the blood are also present in
the oocyte. Four out of five proteins of non-insectan origins, after being injected
into the blood of the Polyphemus moth were detectable in the oocyte. Thus, the
oocyte appears to accumulate almost all of the proteins present in the blood.

2. When the blood of Cecropia is injected into females of the Polyphemus
moth, antigen 7, the “female protein,’ and antigen 3, a carotenoid protein, become
distributed between the blood and the oocytes of the host in a manner quantitatively
similar to their normal distribution in Cecropia. These two oocyte proteins may
therefore be derived exclusively from the blood, rather than being synthesized to
an appreciable extent in the ovary. The capacity of these proteins to combine
with homologous antibodies was not detectably altered during their transmission
from the blood to the oocyte.

3. Measurements of the concentrations of several proteins in the oocyte, relative
to their concentrations in the blood, indicate that the mechanism of protein ac-
cumulation is selective. Of the proteins studied, the female protein was the most
avidly accumulated, while the non-insectan proteins were detectable in the oocyte
in relatively small amounts.

4. The proteins derived from the blood are probably localized primarily in the
yolk spheres of the oocyte. Protein accumulation in this case therefore entails both
the penetration of the oocyte surface and association with a cytoplasmic particle.

LITERATURE Crip

BrAMBELL, F. W. R., R. Harimay anv I. G. Morris, 1958. Interference by human and
bovine serum and serum protein fractions with the absorption of antibodies by suckling
rats and mice. Proc. Roy. Soc. London, Series B, 149: 1-12.

Branot, P. W., 1958. <A study of the mechanism of pinocytosis. Exp. Cell Res., 15: 300-313.

CHAPMAN-ANDRESEN, C., AND D. Prescott, 1956. Pinocytosis in the amoeba, Chaos chaos
and Amoeba proteus. Compt. Rend. Trav. Lab. Carlsberg. Ser. Chim., 30: 57-78.

Cooper, R. S., 1948. A study of frog egg antigens with serum-like reactive groups. J. Exp.
Zool., 107: 397-437.

Fevotp, H. L., 1951. Egg proteins. Advances in Protein Chemistry, 6: 188-252.

FLICKINGER, R. A., AND D. E. Rounns, 1956. The maternal synthesis of egg yolk proteins
as demonstrated by isotopic and serological means. Biochim. et Biophys. Acta, 22:
38-42.

Hotter, H., 1959. Problems of pinocytosis, with special regard to Amoebae. Amn. N. Y.
Acad. Sci., 78: 524-537.

BLOOD” PROTEINS IN-OOCYTES oot

Knicut, P. T., anp A. M. ScHEcCHTMAN, 1954. The passage of heterologous serum proteins
from the circulation into the ovum of the fowl. J. Exper. Zool., 127: 271-304.

Lewis, W. H., 1931. Pinocytosis. Johns Hopkins Hosp. Bull., 49: 17-27.

MarsHat, J. M., V. N. SCHUMAKER AND P. W. Branot, 1959. Pinocytosis in amoebae.
Ann. N. Y. Acad. Sct., 78 (Art. 2): 515-523.

Mast, S. O., anp W. L. Doyte, 1934. Ingestion of fluid by amoeba. Protoplasma, 20: 555-
560.

Nace, G. W., 1953. Serological studies of the blood of the developing chick embryo. J. Exp.
Zool., 122: 423-448.

Ouprn, J., 1948. L’analyse immunochimique qualitative; methods par diffusion des antigenes
au sein de l’immunserum precipitant gelose. Ann. Anat. Pasteur, 75: 30-52; 109-130.

Preer, J. R., Anp W. H. TeEtrer, 1957. Some effects of non-reacting substances in the quanti-
tative application of gel diffusion techniques. J. Jmmunol., 79: 288-293.

SCHECHTMANN, A. M., 1947. Antigens of early developmental stages of the chick. J. Exp.
Zool., 105: 329-348.

Scuyeme, O. A., AnD M. R. Urist, 1956. Proteins and calcium in serums of estrogen-treated
roosters. Science, 124: 1242-1244.

ScHuMAKER, V. N., 1958. Uptake of protein from solution by Amoeba proteus. Exp. Cell
Res., 15: 314-331.

Stern, K. G., ann K. Satomon, 1937. Ovoverdin, a pigment chemically related to visual
purple. Science, 86: 310-311.

TELFER, W. H., 1954. Immunological studies of insect metamorphosis. II. The role of a
sex-limited blood protein in egg formation by the Cecropia silkworm. J. Gen. Physiol.,
37: 539-558.

TeLFeR, W. H., anp L. D. Rurperc, 1960. The effects of blood protein depletion on the
growth of the oocytes in the Cecropia moth. Biol. Bull., 118: 352-366.

TELFER, W. H., anv C. M. Wr tAms, 1953. Immunological studies of insect metamorphosis.
I. Qualitative and quantitative changes in the blood proteins of the Cecropia silk-
worm. J. Gen. Physiol., 36: 389-413.

WicGLesworTtH, V. B., 1942. The fate of haemoglobin in Rhodnus prohxus (Hemiptera)
and other blood-sucking arthropods. Proc. Roy. Soc. London, Series B, 131: 313-339.

WixuiAMs, C. M., 1956. Physiology of insect diapause. X. An endocrine mechanism for
the influence of temperature on the diapausing pupa of the Cecropia silkworm. JBuzol.
Bull., 110: 201-218.

Witson, FE. B., 1925. The Cell in Development and Heredity. The Macmillan Company,
New York.

THE EFFECTS OF BLOOD: PROTEIN DEPLETION] ON ieas
GROWTH. OF THE OOCYTES IN, THE CE CRORIA mi@agel

WILETANM He TELFER AND Le DAVID RURBERG

Zoological Laboratory, Umversity of Pennsylvama, Philadelphia 4, Penna.

During the final two weeks in the metamorphosis of the Cecropia moth the
oocytes enter a period of yolk formation, in the course of which their volume
increases 400-fold. The growth of the oocyte during this period is not dependent
exclusively on the synthetic activities of the ovary, but also entails the transfer
of proteins from the blood to the yolk (Telfer, 1960).- An analysis of the
mechanism of protein transfer and its role in the formation of yolk has been un-
dertaken. We describe here some effects of depleting the blood of two prospective
yolk proteins on the growth of the oocyte and on the composition of its yolk.
Advantage was taken of the decrease which normally occurs in the concentration
of the two proteins during the period of yolk formation.

The growth rate of the oocyte is interpreted here as an index to its rate of
yolk formation. The Cecropia oocyte belongs to that class of eggs in which the
yolk inclusions constitute the primary mass of the cytoplasm: centrifugation at
20,000 x g yields a stratum of protein yolk spheres which occupies more than 90%
of the volume of the oocyte. In such a case, gross changes in the rate of yolk
formation should be conspicuously reflected in the growth rate of the oocyte.

Determinations of the growth rate of an oocyte, as with other internal structures
of an organism, are handicapped by difficulties in measurements of the time axis.
In the present case, it would be desirable to measure the individual oocyte at
successive times in its period of yolk formation. However, the oocyte must be
dissected from the moth in order to be measured and its capacity for further
development is thereby destroyed. A second approach would be to measure the
size of the oocytes in a large number of ovaries which varied only in the chrono-
logical stage of their development, and this procedure proved feasible for the
Cecropia oocyte. The correlation between ovarian development and the stage of
the moth’s metamorphosis was found to be sufficiently precise so that we were
encouraged to estimate the time of yolk formation from the directly observable
time of metamorphosis. |

MeErTHops

1. Determination of the time of development

The sequence of developmental changes used in determining the time of the
moth’s metamorphosis is described in Table I. The earliest developmental stage
recognized with precision was the darkening of the tarsal claws which can be seen
through the pupal cuticle with the aid of a dissecting microscope. In Schneiderman
and Williams’ (1954) timetable for the development of Cecropia males, tarsal claw

1 This work was supported by a grant from the Phi Beta Psi Sorority.

352

OOCYTE GROWTH alae

darkening is listed as occurring 13 days after the initiation of the pupa’s develop-
ment into the moth. We arbitrarily designated this stage as day 13 for the female
also, and subsequent stages of development were timed accordingly. Our timetable
is based on the rate of development at 25° C., the temperature at which all animals
were maintained.

2. Dissection of the ovaries and determination of oocyte volume

The ovaries were dissected from 70 females in the last half of their transforma-
tion from pupae into moths. The weights of the animals dissected were within
the range of 4.5 to 5.5 grams, and it should be emphasized that the timing and
magnitude of the changes described are probably valid only for animals within
this size range. The final stages of the dissection and measurements of the size
of the oocytes were completed with the ovaries immersed in 0.15 M NaCl. While
the measurements were generally completed within an hour after the dissection had

TABLE |
Characters used in determining the time of development of the female pupa into a moth

Days of development
°

at Zz5~.C. Diagnostic characters

13 Tarsal claws darken

13-14 Antennal rami darken

15-16 Spines on genitalia turn from white to reddish brown

16-17 Developing ovarian tracheae become lined by a cuticular thread demon-

strable by teasing the tracheae apart

18 Ovarian tracheae become air-filled and thus appear silvery

18-19 Initiation of fore-wing pigmentation: lateral spots appear
19 Advanced wing pigmentation; lateral abdominal pigmentation visible
21 Cuticle softened over facial region
22 Cuticle crisp
aS Eclosion

been begun, neither swelling nor shrinking of the oocytes was visible in ovaries
which had remained for as long as 4 hours in the saline solution.

The paired ovaries ccntain a total of 8 tubular ovarioles, one of which was
selected for study from each animal. Since there was little obvious variation
among the ovarioles with regard to the number or size of the oocytes they con-
tained, the ovariole to be studied was selected at random. The only exceptions to
this procedure were the omission of occasional ovarioles which obviously differed
from their seven partners.

A single Cecropia ovariole generally contains from 30 to 50 oocytes in varying
stages of development (Fig. 1). Their volumes were estimated, beginning with
the most mature oocyte, which is located at the posterior end of the ovariole, and
ending with the smallest oocyte which contained visible yolk. The presence of
yolk was recognized through a dissecting microscope by the fact that it renders the
cytoplasm opaque and colors it yellow. The termination of growth coincides with
the secretion of the chorion. This event was detected by two criteria: since the
chorion is opaque and colorless, it causes the oocyte to appear progressively paler
as it is laid down over the yellow yolk; in addition, the chorion forms a rigid crust
on the surface of the oocyte which thus loses its elastic response to probing.

354 WILLIAM H. TELFER AND’ L. DAVID RUTBERG

The volume of an oocyte was calculated from measurements of its linear
dimensions. The shape of a growing oocyte approximates an ellipsoid with two
equal axes. The disparate axis is the longitudinal one, or that which coincides
with the axis of the tubular ovariole (Figs. 1 and 2). Although the nurse cells
accompanying each oocyte tend to flatten its anterior end (Fig. 2) and thus cause
some deviation from the shape of an ellipsoid, this effect is progressively reduced
as the oocyte enlarges and is undetectable during the last three-quarters of the
volume increase. Therefore, the volume of a growing oocyte could be estimated
by measuring its longitudinal and transverse diameters with a filar micrometer and
by solving the appropriate equation for the volume of an ellipsoid.

Figure 1. An ovariole dissected from a female Cecropia on the seventeenth day of its
transformation from a pupa into a moth. The individual oocytes lie in sequence with the
largest at the posterior end of the ovariole. Small fragments of fat body adhere to the outer
wall of the ovariole.

Figure 2. Several oocytes which were removed from the ovariole shown in Figure 1.
The five nurse cells form a transparent cap at the anterior end of the smaller oocytes. At
the lower left is an oocyte which is nearing maturity. Magnification, 20 X.

The most convenient index to the size of the mature oocyte was its weight,
as determined with a semi-micro balance. Weighing was possible in this case
because the rigid and comparatively impervious chorion allows the egg to be
blotted and weighed without the loss of significant quantities of water from the
oocyte itself.

As in the case of the growing oocyte, the volume of a mature oocyte could
also be calculated from measurements of its diameters. The shape of the mature
oocyte differs from that of the growing oocyte in that its two transverse diameters

OOCYTE GROW TH S08

are unequal. By means which are not apparent, one of the transverse diameters
increases at the expense of the other during or shortly before the process of chorion
formation. The longitudinal diameter appears to be unaffected by this process.
Thus, the volume of a mature oocyte was calculated with the formula for the
volume of an ellipsoid with three differing diameters.

3. Determinations of blood protein concentrations

The concentrations of two antigenically defined proteins in the blood and in
oocyte extracts were determined with the antiserum-agar technique of Oudin. The
two proteins have been described previously (Telfer and Williams, 1953). They
included antigen 3, one of several carotenoid proteins in the blood, and antigen 7,
a sex-limited blood protein which occurs primarily in female pupae and moths.
The validity of the Oudin technique for the determination of the relative con-
centrations of these proteins has been discussed elsewhere, and in the case of the
female protein, has been confirmed by independent methods (Telfer, 1954).

4. Ovarian transplantations to males

Comparisons of the effects of male and female blood on the size of the oocytes
produced were made by transplanting ovaries to males which were then transfused
with samples of either male or female blood. These experiments were performed
on pupae of the Polyphemus moth. The method of CO, anaesthesia and other
surgical procedures used are described in detail by Williams (1959). The tip
of the abdomen was cut off from each male pupa and up to 0.5 of blood was re-
moved through the resulting orifice by compressing the anterior end of the pupa.
An ovary with a small amount of associated fat-body was dissected from a female
pupa and inserted through the abdominal orifice in the male. The haemocoel of
the male was then filled from a pipette with blood which had been obtained either
from female pupae or from other male pupae. Finally, a piece of plastic cover-
slip was sealed over the opening with paraffin. The implanted ovary was dis-
sected from the host at the conclusion of metamorphosis.

RESULTS
1. The volumes of the oocytes in individual ovarioles

The stage of development of an ovariole and its oocytes can be described by
plotting the volume of the successive oocytes as a function of their position in the
ovariole. Three examples of the relationship in the developing Cecropia ovariole
are shown in Figure 3. In all three cases oocyte volume decreased progressively
along the sequence of oocytes. Exceptional in this respect were the volumes of
the first 12 oocytes in the moth which had been developing for 20 days. The fact
that these oocytes were nearly equal in volume can be attributed to their having
completed growth, as was indicated by their all having undergone some degree of
chorion formation. The three examples typify the gradient in developmental stage
seen in an ovariole during the period of yolk formation. Profiles such as these for
animals in varying stages of metamorphosis contain the information required for
constructing growth curves for the cocyte in any particular numerical position.

356 WIELLTANM Hy TELEER AND DAW LD RU rR ERG

+1.0

LOG
VOLUME 20 DAYS
(MM)
0 18 DAYS
16 DAYS
— 1.0

20 25 30

2 10 NUMERICAL
POSITION OF OOCYTE
Ficure 3. The logarithm of the volume of successive oocytes in three ovarioles. ‘The

stage of development of the female from which the ovariole was dissected, is indicated next
to each curve and is expressed in days at 25° C. since the initiation of the pupal-adult molt.

2. Oocyte growth curves

The volume of the first, or most mature oocyte in each ovariole studied was
plotted as a function of the morphologically determined age of the insect from
which it had been dissected. The resulting data indicated that the logarithm of
oocyte volume increases approximately linearly with time (Fig. 4) and then
ceases abruptly at the time of chorion formation. Thus the growth of the first
oocyte is described here as follows:

Vos ig torr NO (4 == Oe ohare Wir: (13)

in which V is the volume of the oocyte in mm.*, ¢ is the stage of metamorphosis
expressed in days at 25° C., and a and b are constants. Analysis of the data by
the methods described in Table II yielded the information that, during the one-
week period for which data were obtained, the percentage relative growth rate
of the first oocyte was 65 = 6% per day while its percentage increment in volume
was 91% per day. Thus, the first oocyte nearly doubles its volume on each of at
least 7 successive days of yolk formation.

3. Comparisons of successive oocytes in the ovariole

Growth curves constructed for the tenth, twentieth and thirtieth oocytes were
similar in form to that of the first oocyte. However, the growth constants differed
in such a manner as to suggest that yolk formation is less vigorous in the more
anteriorly located oocytes.

Each oocyte completes its growth at a later time than its more posterior
neighbor. From the data on the time at which yolk formation is completed

QOCYTE GROWEH Soil

a10

LOG
VOLUME
(MM®)

12 14 16 18 20 22
DAYS OF DEVELOPMENT

Ficure 4. Growth curve of the first or most posterior oocyte in the ovariole of Cecropia.
The logarithm of the volume of the oocyte is plotted against the time of development of the
moth. Chorionated oocytes, © ; oocytes without chorions, 3

(Table II), it can be calculated that in the ovariole as a whole, each oocyte com-
pletes yolk formation an average of 5.2 hours after its predecessor. However, there
is a tendency for the first oocytes produced to follow each other in more rapid
succession than the later oocytes. The average lag between successive oocytes
was 4.2 hours for the first 20 oocytes, and was estimated to be 7.2 hours for the
twentieth to thirtieth oocytes.

Also varying with numerical order in the ovariole is the volume of the oocyte
at maturity. The volume of the first oocyte was 130% of the volume of the tenth

TABLE II

Growth characteristics of oocytes during yolk formation

Numerical position in ovariole

Growth characteristic and its method of calculation

1 10 20 30
Mean volume (+ s.e.) of oocytes possessing chorions (mm.,3) | 4.75 +0.25 | 3.64+0.13 | 3.35 +0.46 3.0
b (equation 1); calculated by the method of least squares 0.28 0.23 0.21 0.18
for the non-chorionated oocytes
Percentage daily increment in volume 91% 70% 62% 51%
[(Antilog 6) —1] 100%
Percentage relative growth rate (% per day) 65% 53% 48% 41%
(2.3 b) 100%
Time when growth terminates: the value of ¢ (equation 18.6 20.1 21.9 24.9
1) when V is the mean volume of the oocytes possess-
ing chorions (days)

358 WILLIAM H. TELFER AND L. DAVID: RUTBERG

5.0

4.0 4.0

3.0

5.0
WEIGHT
(MG)
4.0 4.0

3.0
5.0

4.0 4.0

3.0

| 10 20 30
NUMERICAL POSITION OF OOCYTE

Figure 5. The weights of mature oocytes dissected from moths which had completed
metamorphosis. The weights of successive oocytes from six different moths are plotted as a
function of their numerical position in the ovariole. The reference lines are arbitrarily placed
at 4.0 mg.

oocyte and 142% of that of the twentieth oocyte. The number of chorionated
thirtieth oocytes was too small to permit the comparison to be extended to this
case. The relationship between size at maturity and position in the ovariole was
seen 1n a more exact manner by weighing the successive oocytes in mature
ovarioles dissected from fully developed moths. In all six cases illustrated (Fig. 5),
the later oocytes tended to be smaller than the first oocytes produced.

TABLE III
Comparison of the b-value for the first oocyte with those for numbers 10, 20 and 30

N ical positi + Signifi f diff
a Senet bse . nie Ce ae
1 0.28+0.025 33 os
10 0.23+0.025 41 0.11
20 0.21+0.023 43 0.04
30 0.18+0.026 28 0.008

Chi square test of the significance of the over-all decline; p = 0.003.

OOCYTE GROWTH 359

The value of b, and consequently the rate of growth, also decreases with in-
creasing numerical position (Table II). The difference in b between the first and
the thirtieth oocytes represents a decline in the percentage relative growth rate
from 65% to 41% per day. Statistical comparisons indicated that the decline
observed in b was significant (Table IIT).

Thus, all of the growth characteristics considered here change in such a
manner as to suggest that yolk formation occurs with reduced vigor in the later
oocytes produced in an ovariole. The question then arises as to whether this is
a direct consequence of a depletion in the supply of prospective yolk proteins
available in the blood.

DAYS OF DEVELOPMENT

Figure 6. Changes in the relative concentrations of two proteins in the blood during
the last ten days of the pupal-adult transformation. Upper curve is for antigen 3, a carotenoid
protein. Lower curve is for antigen 7, the female protein. A relative concentration of 1.0
is arbitrarily set as the maximum concentration of each protein in the blood.

4. Changes in the concentrations of two blood proteins during yolk formation

Most, and possibly all, of the moth’s blood proteins are deposited in the yolk
to some extent, and of those studied quantitatively, the female protein is the most
avidly accumulated (Telfer, 1960). It would seem reasonable, therefore, that
depletion of the available blood proteins, and of the female protein in particular,
could affect the growth characteristics of the oocytes. That the concentration of
the female protein in the blood decreases during the last half of the transformation

360 WILLIAM H. TELFER AND L. DAVID RUTBERG

of the pupa into a moth has already been demonstrated (Telfer, 1954). A more
precise determination of the timing of the decrease is described here, as well as
a comparison with the less readily accumulated carotenoid protein.

The results indicate (Fig. 6) that the concentration of female protein de-
creases approximately 80% during the period of yolk formation. The concentra-
tion of the carotenoid protein in the blood also drops during yolk formation, but
in this case the decrease is not nearly as great. The difference between the two
proteins in this regard is consistent with the greater extent to which the oocyte
accumulates the female protein. |

The timing of the decrease in female protein with relation to the growth of the
four oocytes studied is summarized in Figure 7. While the first oocyte undergoes
the final three-quarters of its growth (left hand cross-hatched area in Figure 7)

1.0
REL. LOG
CONG, VOLUME

0

eo)

SOY

19 2i
DAYS OF DEVELOPMENT

23

Ficure 7. Changes in concentration of the female protein (left-hand scale) super-
imposed on the growth curves of the first, tenth, twentieth and thirtieth oocytes (right-hand
scale). Cross-hatched areas indicate the time during which the first and twentieth oocytes
undergo the final three-fourths of their volume increase.

the average female protein concentration approximates one half or more of its
maximum level in the blood. The twentieth oocyte, on the other hand, undergoes
the final three-quarters of its volume increase (right hand cross-hatched area)
when the blood contains only one-fifth of its maximum concentration of female
protein. It thus appears that the first oocytes are produced in an environment in
which the concentration of female protein is greater than that of the later oocytes,
and we are left with the possibility that this accounts for the reduced vigor with
which the later oocytes produce yolk.

5. The effects of a protein’s concentration in the blood on its rate of accumulation
by the oocyte

The most likely mechanism by which a reduction in blood protein concentration
could retard the process of yolk formation would be through a decrease in the rate
of protein accumulation by the oocyte. We therefore sought to determine whether

OOCYTE GROWTH 361

the rates of female and carotenoid protein accumulation are affected by the decrease
which normally occurs in the concentrations of these proteins in the blood. The
information required here, in addition to the growth rates of the successive oocytes,
was the amount of each protein accumulated per mg. of oocyte.

Successive pairs of chorionated oocytes dissected from fully matured ovarioles
were weighed and then extracted with 0.2 ml. of 0.15 M NaCl buffered at
neutrality. Either the oocytes or the extracts were frozen and thawed once, a

1.0

REL. GONC,
OOCYTE WT.

0.6 0.6

sa CAROTENOID

PROTEIN

0.6

FEMALE
a PROTEIN

| 5 10 IS 20 25
NUMERICAL POSITION OF OOCYTE

Ficure 8. Relative concentrations of the carotenoid and female proteins in successive
pairs of oocytes dissected from mature moths. The points represent the mean values for
five different animals. The bars indicate 95% confidence intervals. The relative concentra-
tion of the protein in an extract of standard volume was divided by the fresh weight in
milligrams of the oocytes before extraction. Reference lines are arbitrarily placed at 0.6.

procedure which has been found to lyse the yolk spheres and thus to release their
stored proteins. The concentrations of female and carotenoid proteins in the
extracts were determined with antiserum-agar tests.

From the results summarized in Figure 8, it is apparent that the amount of
female protein extracted per mg. of mature oocyte tended to remain constant along
the length of the ovariole. While there may have been some decline, this was of
questionable significance and it did not approximate the 80% decrease in the

362 WiIELTAM EH. TEEPER AN DVL, DAVID RURBERG

concentration of female protein which occurs in the blood. A similar relationship
was found for the carotenoid protein. Despite the approximate 20% decrease in
the level of carotenoid protein in the blood, the concentration of this protein in
the oocyte tended to remain constant, and perhaps even rose slightly, along the
length of the ovariole (Fig. 8).

When considered independently, these results suggest that the rates of ac-
cumulation of the two proteins are not affected by the decrease in their con-
centrations in the blood. However, a comparison of growth rates indicates that
the time available for protein accumulation is longer in the later oocytes than in the
first oocyte. For instance, the twentieth oocyte required 12 days to increase in
volume from 0.01 mm.’, the volume of the smallest oocytes observed to contain yolk,
to 3.35 mm.*, its volume at maturity. Due to its higher growth rate the first
oocyte requires only 9 days for the same increase in volume. Thus, while the
concentration attained by the proteins may be the same in the two oocytes, the
time required to achieve this concentration is 33% longer in the twentieth oocyte.

It therefore seems likely that naturally occurring reductions in the concentra-
tions of the carotenoid and female proteins in the blood result in their being ac-
cumulated by the oocyte at slower rates. This, in turn, could account for the
slower growth rate and smaller size of the oocytes which produce yolk at the end
of metamorphosis. However, alternative explanations are possible; for instance,
the results summarized in Figure 8 would also be expected if the rate of protein
accumulation were governed primarily by the rate at which space is provided for
the protein in the oocyte. In such a case, the rate of protein accumulation would
be determined by the growth rate of the oocyte, which is the inverse of the relation-
ship proposed above. It therefore proved desirable to seek an independent test
of the extent to which deficiencies of a blood protein can impair the growth of
the oocyte.

6. Oocytes produced by ovaries transplanted to males

Experimental evidence that the level of female protein in the blood may affect
the growth of the oocyte was derived from ovarian transplantation experiments.
Kopeé (1911) observed in a number of Lepidoptera that the oocytes produced by
ovaries which had been transplanted to males are considerably smaller than those
produced by ovaries in females. A similar effect has been observed in Cecropia
(Telfer, 1954) and is demonstrated here for the Polyphemus moth. The follow-
ing experiments indicate that this effect is due to the male’s deficiency in a factor
which is present in the blood of female pupae.

Single ovaries dissected from diapausing female pupae of the Polyphemus
moth were transplanted to the haemocoeles of diapausing male pupae. Half of
the male pupae were then injected with the blood of male pupae and the other half
with the blood of female pupae. The volume of the blood injected was such that
the recipient’s blood contained a 14 to Yq dilution of the injected blood. Fifteen
of the males survived the surgical procedures and completed their development into
moths. In 13 of these the implanted ovary, which was dissected from the moth,
contained oocytes that had completed their development, as was indicated by the
presence of a chorion.

The presence of blood from female pupae enhanced the size attained by the
oocytes developing in males. In the males transfused with male blood the average

OOCYTE GROWTH 363

weight of the mature oocyte was 2.5 mg., the values ranging from 1.9 to 2.8 mg.
In those transfused with female blood, the average weight was 3.6 mg. with a range
of 2.2 to 4.8 mg. The variance in the latter case was significantly greater
(p = 0.04) than that for the male-transfused animals—a fact which suggests that
there were inconsistencies in the effective level of female blood received by the
host. In any case, 6 of the 8 males transfused with female blood produced oocytes
which weighed 3.8 mg. or greater and which were thus within the size range of
oocytes produced by normal ovaries in Polyphemus females.

Blood taken from female pupae—a stage, incidentally, during which yolk
formation does not occur—thus contains something which is lacking in the male
and which enhances the formation of yolk. That the female protein is the
responsible agent is suggested by the fact that it is approximately one thousand
times more concentrated in the blood of female pupae than of males (Telfer, 1954),
and that it is accumulated by the growing oocytes in substantial quantities.

DIscUSSION
1. Comparison with yolk formation in other animals

The growth rate of an oocyte might be expected to serve as a valid index to
the rate of yolk formation only in oocytes whose cytoplasm becomes nearly close-
packed with yolk bodies—as in the case of moths and birds. Even in moderately
yolky oocytes, however, yolk formation, which is limited to the final stages in the
total differentiation of the oocyte (Wilson, 1925), may be the primary cause of the
volume increase occurring at the time.

Despite the hazards in this interpretation, comparisons of the Cecropia oocyte
with several other oocytes whose growth has been described indicate clearly that
the rate of yolk formation varies widely from species to species. Although, to
our knowledge, the growth of the amphibian oocyte has not been the subject of an
analysis in which the parameter of time was measured, one can presume that its
growth during yolk formation is relatively slow from the fact that it spends several
months in the process (cf., Grant, 1953). By contrast the process is rapid in the
chicken and in the fruitfly. In chickens the absolute growth rate of the oocyte
during yolk formation reaches an astonishing 5 grams per day, while the maximum
percentage increment in volume has been measured at 284% per day (Romanoff
and Romanoff, 1949). The relatively minute oocyte of Drosophila melanogaster
completes yolk formation in an estimated 4.7 hours and during this time its
volume doubles approximately once every 0.7 hours (King, 1957). The Cecropia
oocyte with its daily increment in volume of 50-100% per day appears to be
intermediate between these extremes.

Variations such as these are presumably adaptations to various factors in the
over-all economy of the organism. Whatever their explanation, it is apparent that
yolk formation can be a rapid process, resulting in both absolute and relative growth
rates which may be among the fastest for metazoan cells.

2. Conclusions concerning the mechanism of yolk formation in insects

Data concerning the growth of the oocyte in Drosophila melanogaster (King,
1957) are sufficiently extensive to allow a detailed comparison with that of Cecropia.
The growth curves of these two oocytes during their period of yolk formation have

364 WILLIAM H. TELFER AND L. DAVID. RUTBERG

two characteristics in common, one of which is the abruptness with which growth
terminates. There is no period of senility detectable in the yolk-producing mecha-
nism of insects. The correlation between the termination of growth and the first
appearance of the chorion in Cecropia suggests that the two events may be
causally related to each other.

A second similarity is the apparent exponential character of oocyte growth
during yolk formation. A constant relative growth rate, which is implied by this
result, 1s frequently found in biological systems and can in many cases be interpreted
as indicating that the products of growth in turn reproduce. Such an interpreta-
tion is not plausible in the case of yolk formation; it would require that yolk be
produced throughout the substance of the existing yolk mass, whereas evidence
from a number of organisms with relatively yolky oocytes indicates that new ma-
terials are deposited primarily at the periphery of the cell.

Lipophilic dyes injected into the chicken (literature reviewed by Romanoff and
Romanoff, 1949) and radioactive amino acids injected into frogs (Kemp, 1955)
both localized at the periphery of the oocyte rather than throughout its substance.
Evidence that the yolk spheres of the insect oocyte grow primarily at the periphery
of the cytoplasm is suggested by the cytochemical localization of a number of
constituents of the yolk (Bonhag, 1955, 1956; Telfer, 1958). These observations
make it likely that the exponential increase in the yolk mass of the insect oocyte
results, not from self-reproduction of the mature yolk bodies but from an ex-
ponential increase in a peripherally-located mechanism for the formation of yolk.
Whether this mechanism is, simultaneously, a self-reproducing mechanism re-
mains, however, a matter of conjecture.

If equation (1) is in fact an accurate description of the growth of the insect
oocyte, an additional characteristic pertinent to investigations of the mechanism
of yolk formation can be proposed. The volume of the Cecropia oocyte is ap-
proximately 400 times greater at the conclusion of yolk formation than at its
inception. If follows that the absolute rate of yolk formation also increases by a
factor of 400. If one assumes that the oocyte has an approximately constant shape
during yolk formation, its surface area increases by a factor of 400%, or only 50.
Thus a unit of surface area, or perhaps of the cytoplasm underlying the surface,
must produce yolk 8 times faster just before the conclusion of yolk formation than
just after its inception. Either the density or the rate of activity of the yolk-
producing elements at the surface of the oocyte must also increase by a factor of
8, a fact which could prove useful in the cytological identification of the yolk-
producing mechanism. Finally, a similar increase must occur in the rate of
penetration per unit of oocyte surface area by blood proteins and by all other raw
materials required in the formation of yolk.

3. The role of blood protein accumulation in yolk formation

As will be demonstrated more fully in a later paper, the blood proteins ac-
cumulated by the Cecropia oocyte are confined to the protein yolk spheres which
comprise the bulk of the mass of the oocyte. The process of yolk formation in
this insect thus entails the transfer of proteins across the cell surface and their
incorporation into a cytoplasmic particle. It is understandable, therefore, that the
growth of the yolk mass might be sensitive to the availability of blood protein.

OOCYTE GROWTH 365

Of two normally occurring blood proteins and five artificially introduced proteins
thus far studied (Telfer, 1960), the female protein is the most avidly accumulated,
attaining a concentration ratio (oocyte: blood) of 20:1. The evidence presented
here suggests that deficiencies in this protein can impair the growth of the oocyte.
Oocytes reaching maturity in the low female protein environment which prevails
at the termination of metamorphosis grow at a slower rate and are smaller at
maturity than oocytes which grow during an earlier and more bountiful period.
In addition, oocytes developing in males which normally contain only traces of
female protein in their blood attain a 40-50% larger size if the host has been
injected with the blood of female pupae. There is thus a correlation between the
availability of female protein in the blood and the size attained by the oocyte.
Whether the correlation results from the influence of the. female protein alone on
yolk formation is a question whose answer must await the isolation of this protein
in experimentally useful amounts. It is probable however that we are dealing
here with the effects of a blood protein on the growth of the oocyte and the
formation of yolk.

SUMMARY

1. The earliest oocyte to differentiate in an ovariole of the Cecropia moth grows
during the final week of its development with a daily increment in volume which
was calculated to be 91% per day. Growth during this period is due primarily
to the accretion of yolk spheres. It terminates abruptly when the chorion appears
at the surface of the oocyte.

2. Successive oocytes in an ovariole lag behind each other in their development
by an average of approximately 5 hours; the first oocyte terminates growth more
than 6 days earlier than the thirtieth oocyte. As a consequence, the first oocyte
produces yolk at a time when at least two prospective yolk proteins are present in
the blood at higher levels than those prevailing during the growth of the later oocyte.

3. Correlated with a decline in the level of the two blood proteins are reductions
in the growth rates of successive oocytes in the ovariole and in the final size which
they attain. Experimental results suggest that depletion of the female protein in
the blood may be one of the primary causes of the reduced vigor with which the
later oocytes grow.

4. The final concentrations of the two proteins were the same in successive
oocytes in the ovariole. Thus, while the normally occurring depletion in the
blood’s supply of prospective yolk proteins may lead to a reduction in the rate
and amount of yolk formation in an oocyte, it does not affect grossly the quality
of the yolk produced.

5. From a consideration of the oocyte growth curve and of cytological observa-
tions reported for other yolk oocytes, it is proposed that yolk is formed in the
Cecropia oocyte by an exponentially increasing mechanism which is probably
located at the periphery of the cell.

LITERATURE CITED

Bonuwac, P. F., 1955. Histochemical studies of the ovarian nurse tissue and oocytes of the
milkweed bug, Oncopeltus fasciatus (Dallas). I. Cytology, nucleic acids and carbo-
hydrates. J. Morph., 96: 381-440.

366 WILLIAM H. TELFER AND L, DAVID RUTBERG

Bonuac, P. F., 1956. The origin and distribution of periodic acid-Schiff-positive substances
in the oocyte of the earwig, Anisolabis maritima (Gene). J. Morph., 99: 433-464.

Grant, P., 1953. Phosphate metabolism during oogenesis in Rana temporaria. J. Exp. Zool.,
124: 513-543.

Kemp, N. E., 1955. Incorporation of radioactive glycine into proteins of frog oocytes. Science,
121: 471-472.

Kine, R. C., 1957. Oogenesis in adult Drosophila melanogaster. II. Stage distribution as
a function of age. Growth, 21: 95-102.

Kopr¢é, S., 1911. Untersuchungen tber Kastration und Transplantation bei Schmetterlingen.
Arch. f. Entw., 33: 1-116.

Romanorr, A. L., AND A. J. Romanorr, 1949. The Avian Egg. John Wiley and Sons, Inc.,
New York.

SCHNEIDERMAN, H. A., anp C. M. Wit tas, 1954. Physiology of insect diapause. IX. The
cytochrome oxidase system in relation to the diapause and development of the
Cecropia silkworm. Biol. Bull., 106: 238-252.

TeLrer, W. H., 1954. Immunological studies of insect metamorphosis. II. The role of a
sex-limited female protein in egg formation by the Cecropia silkworm. J. Gen.
Physiol., 37: 539-558.

TeLrer, W. H., 1958. Fluorescent antibody studies of blood protein uptake by moth oocytes.
Anat. Rec., 131: 603-604.

Tetrer, W. H., 1960. The selective accumulation of blood proteins by the growing oocytes
of saturniid moths. Biol. Bull., 118: 338-351.

TeLFrer, W. H., ano C. M. WritiAms, 1953. Immunological studies of insect metamorphosis.
I. Quantitative and qualitative changes in the blood proteins of the Cecropia silkworm.
J. Gen. Physiol., 36: 389-413.

WiiiAMs, C. M., 1959. The juvenile hormone. I. Endocrine activity of the corpora allata
of the adult Cecropia silkworm. Biol. Bull., 116: 323-338.

Witson, E. B., 1925. The Cell in Development and Heredity. The Macmillan Co., New York.

BIOLOGY MATERIALS

The Supply Department of the Marine Biological Labora-
tory has a complete stock of excellent plain preserved and
injected materials, and would be pleased to quote prices on
school needs.

=*

PRESERVED SPECIMENS

for
Zoology, Botany, Embryology,
and Comparative Anatomy

LIVING SPECIMENS

for
Zoology and Botany

including Protozoan and
Drosophila Cultures, and
Animals for Experimental and
Laboratory Use.

MICROSCOPE SLIDES

for
Zoology, Botany, Embryology,
Histology, Bacteriology, and
Parasitology.

CATALOGUES SENT ON REQUEST
oor
Supply Department

MARINE
BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS |
ALLEN, KENNETH, AND J, AWAPARA | |
Metabolism of sulfur amino acids in Mytilus edulis and Rangia

})

Page >

i COREA aT Ci Pel tr LEN SAO ME eal Ate PERT Sih) iE «8%

‘COSTLOW, JOHN D., JR., C. G. nuRmoirr AND R. ‘MONROE | VA
The effect of salinity and temperature on larval, Neyelenene of

Sesarma cinereum (Bosc) reared in the laboratory. dhe UNA old on (BB) A

CosTLow, JOHN D., JR., AND C.'G. BOOKHOUT "Ee
The complete tapyal development of Pesarn cinereum (Bosc) reared _

in the laboratory......... hi ae arial eA Le es Lye clade D Mes Nid TN a ag

DEHNEL, PAUL A. , ‘
_ Effect of temperature nd salinity on the oxygen consumption of two

intertidal)crabs./).1 0. oe diyel ee. BES ION AY ca PAT a isk ok ace pf ener ame

|
DURAND, JAMES B. | | Oe en,
Limb regeneration ‘and endocrine activity in the cay ee eee

GEORGE, J. C., AND C. L. TALESARA /
Studies on the structure and physiology of the flight muscles of: birdsy
0% A quantitative study of the distribution | pattern of succinic dehy-

drogenase in the pectoralis major muscle of the pigeon..........., 262

GOLDSMITH, MARY HELEN M., AND HOWARD A. SCHNEIDERMAN
The effects of oxygen poisoning on the post-embryonic development _

_ and behavior of a chalcid WASP... sees. ey AL Mpsuaheed lt Sal NI 269 |
| JOHANSEN, KJELL | ON) iy 7p }
tJ Circulation in the hagfish, Myzine glutinosa Dae eg Pt hafaad tlhe eae
RIEGEL, J. A., AND L. B. KIRSCHNER | Uae TN
The exeretion of inulin and BiGCose, by, the Erayee antennal gland. 296
RIZKI, M. T. M. © | PR
_ The effects of glucosamine hydrochloride on ‘the ‘development of |
Drosophila melanogaster........ Lad b asClefe ale Ia Kelly itt ty ey WOE la 308

ROSENBAUM, ROBERT M., AND CARMEN ROLON |
‘Intracellular digestion and eed enzymes in the phagocytts of
PRAM ABIAMS CONG UN DNS he WL SCAN! a TR rol Valle poe maree coal. 315

SHIRODKAR, M.-V., A. WARWICK AND F. B. BANG

The in vitro reaction of Limulus amebocytes to bacteria........ es (OI,

TELFER, WILLIAM H. |
The selective accumulation of blood proteins by the oocytes of Bate
BED MOEDS 2s ye Seals ee Wik sce sey Pha eae aad T Gd PP MaZaGl SMEYdh y 338

TELFER, WILLIAM H., AND L. DAVID RUTBERG .
The effects of ihacd protein depletion on the growth of the oocytes in
the Cecropia sia BEER iva Mak eb telonacs MA eA OME TANK DOES che Gl 352

peviakenie 118 | ; Number 3_

P THE
- BIOLOGICAL BULLETIN

a PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY

| Editorial Board
DAVID W. BISHOP, Carnegie Institution of V. L. Loosanorf, U.S. Fish and Wildlife
Washington Service
HAROLD C. BOLD, University of Texas C. L. PROSSER, University of Illinois
FRANK A. BROWN, JR., Northwestern University BERTA SCHARRER, Albert Einstein College of |
JouN B. Buck, National Institutes of Health yo ee eae
LIBBIE H. HYMAN, American Museum of FRANZ SCHRADER, Duke University
3 Natural History WM. RANDOLPH TAYLOR, University of Michigan
J. LoGan Irvin, University of North Carolina CARROLL M. WILLIAMS, Harvard University
. DONALD P. COSTELLO, University of North Carolina
Managing Editor
JUNE, 1960

Printed and Issued by
LANCASTER PRESS, Inc.

PRINCE & LEMON STS.
LANCASTER, PA.

“ ' % \ x \7# jiu {ik
\ ' a. }

INSTRUCTIONS $ TO AUTHORS -

The Biological Bulletin accepts papers on.a variety of subjects of biological iitarcut. In

/
|

general, however, review papers (except those written at the specific invitation of the Editorial Wy

Board), short preliminary notes and papers which describe only a new technique or method.
without presenting substantial paper w of data resulting from the use of the new, method cannot
be accepted for: publication. paper will usually appear within three months of hg date, of
its acceptance.

The Editorial Board requests dhael jnanuscripts ae to the requirements set elon:
those manuscripts which do not conform will be returned to authors for correction dain they,
are refereed by the Board. |

1. Manuscripts. ‘Manuscripts must be typed in double spacing (including figure ‘eoenids, .

foot-notes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 83 by 11 inches. | They
should be carefully proof-read before being submitted and all typographical errors corrected
legibly in black ink. Pages should be numbered. A left-hand margin of at least 1} nee
should be allowed.

2. Tables, Foot-Notes, Figure Legend; etc. Tables sul be typed on laparade difaots ne
_ placed in correct sequence in 'the text. Because of the high cost of setting such material in type,
authors are earnestly requested to limit tabular material as much as possible. _ Similarly, foot-
notes to tables should be avoided wherever possible. | If they are essential, , they should be indi-
cated by asterisks, daggers, etc., rather than by numbers.. Foot-notes in the body of the text
should also be avoided unless they are absolutely necessary, and the material incorporated into
the text. Text foot-notes should be mumbered consecutively and typed double-spaced on a sepa-
rate sheet. Explanations of figures should be typed double-spaced and aA on separate sheets
at the end of the paper. Eh |

|

3. A condensed title or tunning head of no more than 35 letters and spaces should be included. Wh

4, Literature Cited. The list of papers cited should conform exactly to the style ae in a4)

recent issue of The Biological Bulletin; this list should be headey LITERATURE CITED,
and typed double-spaced on separate pages. |

_ 5.. Figures, The dimensions of the printed page, 5 by 73 tucked, should be hegt: in hina in
preparing figures for publication. Illustrations should be large enough so that all details will be
clear after appropriate reduction. Explanatory matter should be included in legends as far as
‘possible, not lettered on the illustrations. Figures should be prepared for reproduction as line,
cuts or halftones; other methods will be used only at the author’s expense. Figures to be repro-
duced as line cuts should be drawn in black ink,on white paper, good quality tracing cloth or
blue-lined coordinate paper; those to be reproduced as halftones, should be mounted on Bristol
Board, and any designating numbers or letters should be made directly on the figures. ‘All
figures should be numbered in consecutive order, with no distinction between text- and plate-
figures. The author's name should appear on the reverse side of all figures, as well as the desired
reduction. | ly

6. Mailing. Manuscripts should be packed flat; large illustrations may be rolled ina wate

A
a

tube. All illustrations larger than 8} by 11 inches must be accompanied Fy haath 9 ef

reproductions or tracings that may be folded to page size. — wee,

Reprints. Authors will be furnished, free of charge, one hundred rednnats leith Covers.

Additional copies may be obtained at cost; approximate prices will be furnished ied the Manayins |

Editor upon request.

|

THE BIOLOGICAL BULLETIN

THE BIroLoGICAL BULLETIN is issued six times a year at the Lancaster Press, Inc., Prince and
Lemon Streets, Lancaster, Pennsylvania.

|

Subscriptions and similar matter should be addressed to The Biological Bulletin, Marine
Biological Laboratory, Woods Hole, Massachusetts, All substriptions expire with/the December
issue and are renewable prior to the next, succeeding February issue. Single numbers, $2.50.
Subscription per volume (three issues), $6,00.

Communications relative to manuscripts should be sent to the Managing Editor, Marine

Biological Laboratory, Woods Hole, Massachusetts, between June 1 and September rf and to |

Dr. Donald P. Costello, P. O. Box 429, Chapel Hill, (North Carolina, during the remainder of
the year.

Second-class postage paid at Lancaster, Pa.

BACK ISSUES

The Marine Biological Laboratory
needs many numbers of

THE BIOLOGICAL BULLETIN
to replenish its stock, greatly de-
pleted after meeting the needs of
individuals and libraries during the
last 25 years. Persons willing to
contribute any numbers, volumes,
or runs may send them, express col-
lect, to the
Marine Biological Laboratory,

Woods Hole, Massachusetts

to minimize
glass breakage
specify...

FUSIFORM
BORE

THOMAS HR-108 SAFETY STOPPERS are of bloomless, black, resilient, heat resisting
composition, with low sulfur content and long life. Straight bore of the conventional stopper
has been replaced by a fusiform bore with bevelled entrance. Internal diameter at center
approximately twice that of a sealing band at either end. Offer the following advantages:
SAFET Y—Glass tubes can be easily inserted and removed, particularly by students, with greater safety than

with straight bore. Insertion is facilitated by the bev elled entrance of the bore, and the tube slides through
only the two sealing bands.

LARGER TUBES—Fusiform bore permits insertion of larger tubes than is possible with straight bore. For
example, tubes 6 to 11 mm can be inserted in sizes 5 to 9 incl.

SUPERIOR COMPOSITION—The stock contains more than 70% of pure virgin gum, possesses great elas-
ticity, is free from surface sulfur and contains less than 0.75% free sulfur; no reclaimed rubber is used.

HEAT RESISTANT—Temperatures up to 110°C do not affect shape, size or structure.
LONG LIFE—Tests indicate useful life approximately five times that of common stoppers.

$807. Safety Stoppers, Thomas HR-108, as above 8808. Safety Stoppers, Thomas HR-108, as above

described, one hole (fusiform bore). described, but two hole.

Size Approx. no. Carton Per Per Size Approx. no. Carton Per Per
Number per Ib. Contains Dozen Carton Number per Ib. Contains Dozen Carton
Ls ee i 6 aa 1S see cE eee 15.81 O02 eee 160s = oe 168. 2.2.5 1.34. 22,55 17.04
Oi re i § |S Bee ae 198. 25 sf (Pe 10.36 Ungerer Steen MO Cae [| ho eae 1.385 2-8 11.23
Beret. Suns ere = et 5G. sti TSO a cn 10.02 fs ters Sa ee Oe ee 1.482778 10.78
ee eer ok if eee (pee Se (ELS ape ee 8.07 Ng Bi ph tee (eee ee T: 1.582cceee 8.65
ee See 6025S Ue lees 7.60 5 Ee Seek ace = Gate 1.19 8.15
BS eee = Me a ae ve ge St 6.60 BM ore Beat AG. Soe 2 BS aes 1 .SF Xe 6.79
epee heer Sy ee BG 2 e282 "| opt Ss 5.52 AS Seas seein 2 ee 2.04... ee 5.58
DES wie as Rls <7 Wee tao SO See rage |: 5.94 Ds sees = i Se oe. Fae 2.28. Sao 6.23
Ste Se Lo. 2cs s OO. Aa eee ee ee 6....22... 28..2... 30272 242 eee
G36 cee ee ee Soh cree ae gt eee 5.68 G16 5k eet ae cen 30 2-56... Seaee 5.81
RE eS ae Gann a A Oe ee ang & Ege ee 5.05 0 ae eee | | a ge: See 2.83. 75 5.15
Be et Se | i ae bi eyo he SY oe 4.26 Be ee Omen ate jaf page ee Se Lis he at S416 4.31
“SES Se | Sate Howes 18: oe = pee 4.97 ~ eR see poe 145..-22% Be eae 3:60: 22s 4.92

7, discount from carton prices in lots of 6 cartons; 10% discount from carton prices in lots of 12 cartons
ae discount from carton prices in lots of 24 cartons; 30% discount from carton prices in lots of 48 cartons
one size, Or assorted in carton units

For more detailed description, see pages 1066 to 1068 of our general catalog

ARTHUR H. THOMAS COMPANY

A. H.T. co.

More and more laboratories rely on Thomas Laboratory Apparatus and Reagents
VINE ST. AT 3RD ¢ PHILADELPHIA 5, PA:

Vol. 118, No. 3 June, 1960
THE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

MAGNETIC RESPONSE OF AN ORGANISM AND
ffs SOLAR RELATIONSHIPS *

F. A. BROWN, JR., W. J. BRETT, M. F. BENNETT AND F. H. BARNWELL 2

Departments of Biological Sciences, Northwestern University, Indiana State Teachers College,
and Sweet Briar College, and The Marine Biological Laboratory, Woods Hole, Mass.

It has become increasingly evident in recent years that the living organism is
sensitive to fluctuations in still unidentified geophysical factors, in addition to those
factors with which physiologists and ecologists have customarily concerned them-
selves in the past (Brown, 1959a). Such obvious environmental factors as tem-
perature, light, humidity, gravity, pressure, and sound, are clearly ones toward
which it is usually highly important that organisms exhibit immediate adaptive
responses. Correspondingly, it is well-known that living things, in general, have
their behavior and basic physiology dominated in terms of response to them.

On the other hand, organisms in environments held constant with respect to all
these obvious factors still exhibit physiological and behavioral fluctuations corre-
lated with fluctuations in other such geophysical parameters as atmospheric pressure,
atmospheric temperature, primary cosmic radiation, and general background
radiation, including both their regular atmospheric tidal changes, and their large,
weather-related, essentially aperiodic, changes. The correlations are of such char-
acter as to indicate clearly that other and yet unidentified factors are also effective
for organisms. These less obvious environmental factors appear able to induce or
trigger relatively large biological alterations, not uncommonly into the range of
30% or more in their deviation from longer-term mean values. In general, it does
not appear adaptive that organisms should permit physiological processes to be
regulated by these subtle factors instead of by the more obvious environmental ones,
and not unexpectedly, therefore, homeostatic mechanisms seem normally to be con-
tinuously operative in reaction to the alterations induced by them.

In two particular associations, however, such subtle environmental factors and
their fluctuations may possess clear adaptive value, and there are good reasons to
postulate that organisms have evolved physiological means for utilizing this value.

1 This research was aided by a contract between the Office of Naval Research, Department
of the Navy, and Northwestern University, #1228—03.

2 The authors wish to acknowledge their indebtedness to Professor L. I. Bockstahler of
the Department of Physics for his generous assistance with problems related to the magnetic
field.

367
Copyright © 1960, by the Marine Biological Laboratory

SMITHSONIAN JUN1-3 1960

368 BROWN, JR., BRETT, BENNETT AND BARNWELL

One of these is in temporal orientation, or the timing of the well-known temperature-
independent solar-daily, lunar-daily, synodic monthly, and annual periodicities that
commonly persist in complete constancy of all factors to which the organisms are
usually deemed sensitive. Indeed, it has been shown possible to account for all the
many and peculiar described properties of these long-cycle persistent rhythms in
terms of periodicities effected by the subtle factors (Webb and Brown, 1959;
Brown, 1959). The second association relates to spatial orientation, particularly
as it pertains to the periodic migrations of organisms for feeding and breeding,
and especially to the navigational aspects of the phenomenon.

Suggesting a role of subtle environmental factors in animal navigation is the
demonstration in recent years of a very close interrelationship between the phe-
nomena of temporal and spatial orientation of animals (Hoffman, 1954; Pardi and
Grassi, 1955; Renner, 1959) quite as our own spatial orientation may depend on
relationships to sun, moon or the constellations as a function of time. In both of
these associations, temporal and spatial orientations, control by the obvious factors,
when available, may be the dominating consideration. However, a number of
observed characteristics of these phenomena appear inadequately accounted for in
these terms alone. Such problems might be greatly simplified were it learned that
organisms had available additional information.

On the one hand, it is obvious from available evidence that at least some subtle
geophysical factors are perceived by organisms. And yet, on the other hand,
cursory examinations for the biological effectiveness of changes in various ones of
these within essentially their natural range have in the past yielded generally nega-
tive results. Therefore, it seems obvious that the question must be more critically
re-examined. The work about to be described is the consequence of a brief, in-
tensive re-examination for possible response to the magnetic field. In this study,
reported earlier briefly (Brown, Brett and Webb, 1959; Brown, Bennett and
Brett, 1959; Brown, Webb, Bennett and Barnwell, 1959), not only is it apparent
that an organism responds significantly to the changes in the magnetic field through
alterations in its spatial orientation, but that the response is intimately associated
with the mechanism of temporal orientation of organisms quite as might be expected
were the magnetic orientation to possess usefulness to the organism.

METHODS AND MATERIALS

The common mud-snail, Nassarius obsoleta, was used in this study. Collections
were made at Chappaquoit Beach, West Falmouth, Massachusetts, at approximately
weekly intervals, and the stock collections were maintained on a table in running
sea-water in the ordinary daily illumination changes of the laboratory.

Simple equipment (Fig. 1) was constructed to determine the degree of right
and left turning of snails as they emerged from a narrow corridor into a constant,
symmetrically illuminated field. The apparatus consisted of an aluminum, funnel-
shaped corral fastened in an 8-inch crystallizing dish set upon a polar-coordinate grid
with the opening of the funnel over the center of the grid. A circle of 3 cm. radius
was drawn. The grid was ruled into 22.5° sectors, and the long axis of the corridor
was accurately aligned with the center of the sector labelled zero. Since the sizes of
the snails varied from nearly the width of the corridor to half of that value, even
snails following one or the other corridor wall and continuing on a straight path

MAGNETIC RESPONSE IN SNAILS 369

would, after 3 cm., lie in the zero sector. Successive sectors to the right were given
numbers + 1 through + 4, and to the left, — 1 through — 4. This apparatus was
placed on the bottom of a 17 X 12 X 10 inch box, as shown in Figure 2, in such
a manner that a 41-inch round window illuminated by an enclosed 60-watt in-
candescent lamp and covered with a white diffusing transmitter lay above and
slightly ahead of the corridor exit. A hooded window for observing the snail
movements, and for re-corraling the snails between the successive runs of ten, was

Ne

Ficure 1. Orientation apparatus. View from above of the corral (A) in an 8-inch glass
crystallizing dish placed over a polar grid. All dimensions may be scaled from the 3-cm.
radius of the innermost circle.

located above and to the rear of the corral. Beneath the box was an adjustable
platform with a horizontal turntable upon which an 18-cm. Alnico bar magnet could
be easily placed, centered under the corridor exit and orientated in any compass
direction, or from which the magnet could be removed quickly. The magnet at
14 cm. below the orientation chamber gave a horizontal intensity (1.5 gausses) only
about nine times that of the earth’s field in Massachusetts (.17 gauss). Such a
weak experimental magnetic field was purposely selected to increase the probability
that any response which might be discovered was a naturally occurring phenomenon.

370 BROWN, JR., BRETT, BENNETT AND BARNWELL

3 J

Frcure 2. Lateral view of the diagrammatically-sectioned wooden orientation apparatus, drawn
to scale. A is the corral, M is the bar-magnet.

MAGNETIC RESPONSE IN SNAILS “Vall

Such a field strength would be expected to lie within the range of sensitivity of any
normally operative responder system which might be present for magnetism. Four
identical orientation chambers were available during the study.

The experimental procedures were kept uniform, simple and minimal. The
apparatus was always oriented carefully so that the corridor exit was pointed toward
the magnetic south. Sea-water to a depth of about 2 cm. was added to the crystal-
lizing dish. Eleven to fifteen snails were dropped into the corral. Each completed
experimental series consisted of 60 snail exits, two groups of ten exits without
magnet, two groups of ten with the artificially increased magnetic field oriented
as the earth’s field (with north-seeking pole directed south), and two groups of
ten with the north-seeking pole directed west. The first three groups of ten usually
consisted of one of each of the three experimental conditions, as did also the last
three. The order within the three groups of ten, however, changed continuously.
Throughout most of the period of experimentation, the observer was uninformed
as to the presence or absence, or orientation, of the magnet. Each snail emergence
was accorded one number, that of the sector in which the head, or greater fraction
of it, was located at the instant the snail reached the 3 cm. arc.

It is important to note that the method does not involve the determination of any
final pathway chosen by the snails; rather, it simply yields a measure of any
tendency toward change in direction from the initial path, and the relative degree
of this tendency. The use of this kind of measurement appeared to reduce to a
minimum obviously erratic, exploratory movements.

It was learned from brief exploratory experiments that the mean path of turning
for any given time was not conspicuously influenced by changing the compass direc-
tion of orientation of the experimental chamber.

During the two-month period, June 28 to August 29, 1959, inclusive, 564 such
series of 60 snail passages were obtained. Series were obtained for all hours of
the day from 5 AM through 9 PM, Eastern Standard Time. No single one of
these hours of the solar-day was represented by fewer than 9 or by more than 50
series of 60 snail exits.

RESULTS

It was apparent early in the experimentation that the snails, whether controls
without experimentally imposed magnetic field or experimentals with such imposed
fields, varied greatly from one time to another in both their spontaneity of activity
and their rate of locomotion. This resulted in differences in time for completion
of a single series of 60 runs, ranging from about 12 to 60 minutes. Secondly, the
pattern of emergences of the snails clearly differed significantly from one time to
another under the same experimental conditions. Thirdly, the results of com-
parisons within single series of 60 runs indicated an influence of the experimental
magnetic fields.

Sample patterns of emergence of the snails are illustrated for nine series in
Figure 3. These were selected, without reference to time, for the purpose of
illustrating the wide variety of results. It is quite evident from this figure that the
experimental magnetic fields, even if effective, could not be producing a simple uni-
form response. As seen from the sample patterns, the introduction of the N-S
oriented magnet (B in figure) could, as the average of the two runs of ten snails,
be associated with a mean path to left, (e.g., #1, 2, 7, 9) or to the right (e.g., #3,

BYP BROWN, JR., BRETT, BENNETT AND BARNWELL

=e. On. ae

=a "Oe

Figure 3. The frequency distributions of snail paths found in nine sample series of 60
runs. A, 20 runs in E~W magnet field. B, 20 runs in N-S magnet field. C, 20 control runs
only in the earth’s (N-S) field. See text for discussion.

MAGNETIC RESPONSE ‘IN-SNAIES 373

4, 8) of the controls. The N-S oriented magnet could appear to effect more dis-
persion from zero than the controls within the same series (e.g., #3, 7, 9) or less
(e.g., #2, 4, 8) ; it could appear to effect mean paths to right (e.g., #4, 5) or the
left (e.g., #1) of the E—W oriented magnetic field (A in figure) or to produce
more (é.g., #3, 6) or less (e.g., #2, 8) dispersion than that observed within the
same series for the E—W oriented magnet. Comparably, the E—W oriented magnet
was associated with patterns to left (e.g., #4, 5, 7, 9) or right (e.g., #3, 8) or
with more (e.g., #2, 7, 8, 9) or less (e.g., ##3, 6) dispersion than the controls.
Ordinary tests for the significance of differences between means, or between
standard deviations in the frequency distributions among the three groups of 20
even within these single series, yielded, not uncommonly, probabilities ranging from
the 5% to well less than 0.1% level. There was no suggestion that these differences
could be explained on the basis of any continuous trend in behavior during the
experimental period of a series. In fact, this last possibility was greatly reduced
by having each sample of twenty composed of two groups of ten separated by inter-
vening runs under other conditions.

It seemed quite apparent that if a bona fide response occurred to either the
earth’s magnetic field or the experimentally imposed ones, the response could not
be simple and invariable. Consequently, analyses of the data involved chiefly three
types of values: (a) mean direction and degree of turning; (b) total dispersion
as a measure of right and left turning about the zero axis; and (c) standard devia-
tion as a measure of dispersion relative to whatever was the mean path of the
given series. The forms of frequency distributions were also always inspected.

(4) Mean path of snails: Using all data, it became quickly evident that there
was a daily rhythm in the average degree of turning of the emerging snails, the
animals moving nearly straight ahead at 5 AM and turning increasingly to the left
until noon. Thereafter, they turned progressively less, to a second minimum of
turning about 7 PM. The mean paths for the snails exposed to the N-S and E-W
experimental magnetic fields, and for the controls are plotted as a function of hour
of the day in Figure 4A. Indicated also are standard errors of the means for the
combined three groups for selected hours. In Figure 4B are illustrated the dif-
ferences, hour by hour, between the path of the controls and the snails in each of the
two experimental magnetic fields. Using all 1128 available differences between
experimental and control animals, a mean of — 0.0341 + 0.0104 was obtained.
This indicated that the presence of a magnet results in increased left-turning over
controls (P < .005).

There was no statistically significant difference between the effect of the N-S
field (— 0.0347) and E-W one (—0.0336). It is, however, suggestive that the
3.2% greater effect of the N-—S field is correlated with the actual 12% greater
horizontal intensity of the experimental N—S over the experimental E—W field, due
to the vectorially additive effect of the earth’s natural field in the former expert-
mental condition.

In view of the evident daily rhythm in the mean path observed, together with
the suggestion from Figure 4B that increase in strength of magnetic field in the
early morning hours yields greater right turning, an examination of the effect of
the magnetic fields for the hours 7 AM through 9 PM was made. This gave even
more decisive indication of increased left-turning in response to the experimental

374 BROWN, JR., BRETT, BENNETT AND BARNWELL

oo
~ el
Oo
a
O
a0
ke
<6
oO
Zz -.|
<6
uJ
=
-.2
-
Le
uJ
=)
ay
- 4
+,20
| oa
O +.10
uJ
ire
iL
LJ
O
lo
uw
Fz
oO
ne
=. 6)
—.20

6AM 12 NOON 6PM
SOLAR DAY

Ficure 4. A. Mean paths of the snails in the E-W magnet field (dashed lines), in the
N-S magnet field (solid line), and control snails (dotted lines) as a function of hour of the
solar day. Standard errors of means for all snails are shown for three times of day. B.
The difference of mean paths of snails in the E-W magnet field (dashed line) and N-S magnet
field (solid line) from the paths of the control snails in the same series.

MAGNETIC RESPONSE IN SNAILS 375

FROM ZERO

DEV.

AVERAGE PATH

EPrecc T

MAGNET

6 AM 12 NOON 6PM
SOLAR DAY

Ficure 5. A. Mean dispersion of experimental and control snail paths from the zero
sector as a function of hour of day. Standard errors of means from all snails are shown for
four times of day. B. Difference between experimental snails and controls as a function of
time of day. See Figure 4 legend for key.

376 BROWN, JR., BRETT, BENNETT AND BARNWELL

magnetic fields (P < .001). Also, for the 3 PM hour alone the effect of the
increased magnetic field strength was highly significantly different from zero (t =
3.64; N = 84). These probabilities, which leave no reasonable doubt as to a re-
sponse to the magnet, are found despite highly significant fluctuations of magnetic
response of lunar-day frequency which contribute greatly to the variability of the
response (Brown, Webb and Brett, 1960).

(B) Mean dispersion of paths from zero: Inspection of the frequency distribu-
tions of the mean snail paths as a function of hours of the day, indicated an in-
creasing bimodality of the distribution of the paths, through a maximum degree of
bimodality near noon. A consequence of this is clearly apparent in the form of a
daily fluctuation in average dispersion (Fig. 5A). Standard errors for selected
hours are shown. A broad maximum is centered over noon, and a second maximum
occurs at night. There is also evident a daily rhythm in the difference between
mean dispersions of control and experimental animals (Fig. 5B). For a 9-hour
period centered at noon, the mean dispersion of the animals as the average for the
two magnetic fields was 7.55 + 1.96% (N = 364) greater than that of the controls.
These results suggest, further, that the effect of both magnet positions in early
morning and early evening is to reduce dispersion even below that of the controls
within the same series.

(C) Standard deviation of paths: The next analysis was of the fluctuations in
standard deviation of paths. It was clear that there was a daily fluctuation in this
parameter. This is illustrated in Figure 6A. The significance of this cycle can
be seen from the two standard errors included. In this parameter, as in the others,
the frequency distributions suggested increasing bimodality for the data obtained
over the middle of the day. However, at those times of day, 8-12 AM and 8 and
9 PM, when standard deviations are relatively lower than might be expected in
terms of total dispersion (compare Figs. 5A and 6A), the population of values
more strongly favors one of the two centers in the bimodal frequency distribution.
Figure 6B illustrates the hour-by-hour difference between the control snails and
the snails in the experimentally augmented magnetic fields. The daily pattern
of effect of the imposed magnetic fields upon altering the standard deviation of
paths selected appears to possess at least three maxima. With all the data from
the 17 hours of the day, the standard deviation appeared increased by the magnets,
though not highly significantly (P < .05). For the period 6 AM through 4 PM
alone, the statistical significance increased substantially (P < .01). However, that
an influence of the magnetic fields is being reflected in this parameter is suggested
even more strongly from the similarity of the mean daily pattern of effect for the
two magnetic orientations, N-S and E-W (r=0.72) (N = 17), though the sig-
nificance of this correlation is obviously tempered somewhat by the fact that there
was a common control group for the two experimental groups.

(D) Frequency distribution of magnet responses, relative to hours of day:
In view of the clear indication from the preceding analyses that the imposed magnetic
fields effect to various extents either of two types of response, right or left turning,
it seems reasonable to presume that (1) the small, but statistically highly sig-
nificant, predominance of left turning especially between the hours 7 AM and 9
PM, and (2) the clear suggestion of right turning at 5 and 6 AM are simply
residuals, a consequence of failure of one response to be cancelled exactly by the

MAGNETIC RESPONSE IN SNAILS 377

1.0
”)
a3
loo
<
a8
>
uJ

@) 9
O
x
<
(a)
=
<
ke
Y)

8

+.05
-
O
uJ
iw
LL

ad 0
=
uJ
A
<

= 7% 05

-.10

6AM 12 NOON 6 PM
SOLAR DAY

Ficure 6. A. Standard deviation of experimental and control snail paths, as a function
of hour of day. B. Difference between experimental snails and controls as a function of time
of day. See Figure 4 legend for key.

378 BROWN, JR., BRETT, BENNETT AND BARNWELL

other. To cast further light upon this aspect of the problem, frequency distribu-
tions of the differences obtained between experimental animals and the controls in
the same series are shown as a function of hour of the solar day in Figure 7. From
the general dissimilarity of nearly all these to normal distributions, and the strong
tendency towards bimodal distributions (see especially 9 AM through 4 PM), these
suggest that the organismic response to the imposed magnetic fields is not only real,
but substantially larger than was apparent from the simple study of the pooled
hourly data. The actual response appears normally to amount to magnetic-induced
turning through an average of 0.3 to 0.4, or more, of a sector unit, or a mean
turning of the order of 8-10° during the snail passage in a few seconds over the
3-cm. course.

Further inspection of the distributions suggests that a response to the imposed
magnet occurs also as a deviation from a mean degree of turning characteristic of
the particular hour of the solar day. This is most evident in the 3 PM and 9 PM
distributions. It is also suggested for the 5 and 6 AM hours, when the mean

MAGNET MINUS CONTROL

Figure 7. Forms of frequency distributions of the differences between all experimental snail
paths and control snail paths, as a function of hour of the solar day.

turning of controls is near zero and the response to the magnet is predominantly
that of turning to the right. It appears also evident that the mean effectiveness of
the two experimental magnetic fields passes through minima at 8 AM, 2 PM, and
7 to 8 PM, as indicated by the reduction in the bimodality at these times. This
changing pattern of frequency distributions through the day suggests the presence
of two major daily patterns of oscillation in the extent of the response to the ex-
perimentally increased magnetic fields, with one pattern essentially the mirror-
image of the other, and with cross-over points at these times of minimal response.

DISCUSSION

It is apparent from this study that in the constant symmetrical experimental
field the amount of turning of snails is a function of the time of day, and that a
daily fluctuation is found in both the mean amount of turning, ignoring sign, and

MAGNETIC RESPONSE IN SNAILS 379

in the mean path of the snails taking sign into account. This is apparent for all
snails, both those in the earth’s natural magnetic field and in the imposed fields
where the field strength is increased about 10-fold with the direction of magnetic
lines of force being left either in the same direction as the earth’s, or at right angles
to this. The three groups considered as three samples showed remarkable mean
similarities as functions of hours of the day, though commonly differing very
widely from one another in any individual series (see Fig. 3). This similarity
indicated that the amount of right or left turning, or dispersion of paths from zero,
was not determined primarily by the direction of the field or by its fluctuations
in strength.

Hence, it must be concluded that the response is predominantly a klinokinesis,
or a turning relative to any initial compass direction, either clockwise or counter-
clockwise, as a function of time of day. Of the various possibilities of the factors
concerned in the response, three are immediately apparent. (1) The response
involves a rate of turning relative to the diffusing light source above and slightly
ahead of the path of the snail. This might be interpreted as a move on the part
of the snail to assume a particular angular relationship between the long-axis of its
body (light-compass relationship) and the artificial gigantic “sun” as a function
of the time of solar day; (2) the turning of snails involves a torsional orientation
in response to the magnetic field itself as a function of time of day; (3) the ori-
entational response may not be due to any external factor directly, but simply to
differential bilateral activity of bodily orientation mechanisms. Or, possibly any
two or all might be involved and contribute jointly to the results. However, it is
evident from this study that just as would be expected were the response normally
one of klinokinesis in response to the magnetic-field, so the klinokinesis can be
highly significantly modified through experimentally increased magnetic flux as a
function of time of solar day. There also remains an additional possibility, namely,
that any light-compass orientation of the snails is, in turn, regulated through an
orientation to magnetic field.

Support for the concept that the normal turning of the snail is importantly an
orientation in response to magnetic field itself may be seen first in Figure 5. When
the dispersion is least in the earth’s magnetic field, the effect of increasing magnetic
flux is to reduce it still more, and when dispersion is large in the earth’s field, the
magnet increases it still more. Comparably, as Figure 4 indicates, when left-
turning is high in the earth’s natural field the magnet increases it, and when it is
low, appears to effect turning in the opposite direction, to the right. In a general
manner, but slightly more complexly, the relationships of Figure 6 also give a
similar kind of support for the view of an important role of the normal earth’s
field in snail orientation. The effect of the experimentally augmented field is to
bring about either further increased, or decreased, size of standard deviation,
depending upon which is normally the predominant response characteristic of the
time in the daily cycle.

It is very interesting, also, to compare the two aspects of magnetic orientation,
left-turning and bimodality of distribution about a mean path, to the two signs
of oxidative metabolic fluctuations of Nassarius about a mean daily non-inverting
fluctuation. Both the non-inverting and inverting components in these metabolic
fluctuations appear clearly to be responses to unidentified barometric pressure

380 BROWN, JR., BRETT, BENNETT AND BARNWELL

correlates (Brown, Webb and Brett, 1959). This suggests an intimate relationship
between the mechanisms of temporal and spatial orientation in the snails and points
to the possibility that the experimentally augmented magnetic field has in part
simply increased the strength of whatever orientational physiological mechanism
chanced to be dominant at the time, through a generalized influence upon the
mechanism of cellular oxidative energy transformations. For this effect, the mag-
netic field would need to possess no spatial orientational feature, per se. But while
this effect might account for the induced increases in standard deviation or increased
turning, it could scarcely account for those times of improved precision of orientation
by reduction of turning below that of the controls. The last appears more rationally
accountable in terms of the concept that the sharp and stronger the magnetic field,
the more decisive whatever organismic responses are characteristic of that time of
day to it.
SUMMARY

1. The orientation of snails in a constant, symmetrical field was studied over a
two-month period, June 28 through August 29, 1959, at various hours of the day
between 5 AM and 9 PM.

2. The orientation of snails in the earth’s natural magnetic field was compared
throughout the study with the orientation of snails subjected to a 9- to 10-fold
increase in field strength, with fields both parallel and at right angles to the earth’s
natural field.

3. A daily rhythm in the direction and average amount of turning was found in
the snails; the mean paths of those in the two (N-S; E-W) experimentally aug-
mented magnetic fields were statistically significantly to the left of the controls,
particularly between the hours 7 AM through 9 PM.

4. The mean amount of turning, whether clockwise or counterclockwise
(klinokinesis), in the experimental magnetic field was also increased significantly
over that of controls in solely the earth’s field, and similarly exhibited a daily
rhythm.

5. Certain similarities between the orientational responses to the magnetic fields
and earlier described exogenous metabolic fluctuations in constant conditions, sug-
gest a relationship between them.

6. Evidence is advanced supporting the hypothesis that the orientation of snails
normally includes a true response to the earth’s magnetic field.

LITERATURE, CLLED

Brown, F. A., Jr., 1959a. The rhythmic nature of animals and plants. Amer. Sct., 47: 147-
168. (Reprinted from Northwestern Tri-Quarterly, Nov., 1958.)

Brown, F. A., Jr., 1959b. Living clocks. Science, 130: 1535-1544.

Brown, F. A., Jr., M. F. BENNETT anp W. J. Brett, 1959. Effects of imposed magnetic
fields in modifying snail orientation. Biol. Bull., 117: 406.

Brown, F. A. Jr., W. J. Bretr anp H. M. Wess, 1959. Fluctuations in the orientation of
the mud snail, I/yanassa obsoleta, in constant conditions. Biol. Bull., 117: 406-407.

Brown, F. A. Jr., H. M. Wess ann W. J. Brett, 1959. Exogenous timing of solar and lunar
periodisms in the mud snail, J/yanassa (Nassarius) obsoleta, in laboratory constant
conditions. Gunma J. Med. Sciences, 8: 233-242.

Brown, F. A. Jr., H. M. Wess Ann W. J. Brett, 1960. Magnetic orientation in an organism
and its lunar relationships. Biol. Bull., 118: 382-392.

MAGNETIC RESPONSE IN SNAILS 381

Brown, F. A., Jr. H. M. Wess, M. F. Bennetr anp F. H. Barnwe tt, 1959. A diurnal
rhythm in response of the snail, J/yanassa, to imposed magnetic fields. Biol. Bull.,
117: 405-406.

HorrMan, K., 1954. Versuche zu der im Richtungsfinden der Vogel enthaltenen Zeitschatzung.
Zeitschr. f. Tierpsychol., 11: 453-475.

Parpi, L., anp M. Grassi, 1955. Experimental modification of direction-finding in Talitrus
saltator (Montagu) and Talorchestia deshayelsi (aud.) (Crustacea-Amphipoda).
Experientia, 11: 202-210.

Renner, M., 1959. Uber ein weiteres Versetzungsexperiment zur Analyse des Zeitsinnes und
der Sonnenorientierung der Honigbiene. Zettschr. vergl. Physiol., 42: 449-483.

Wess, H. M., anp F. A. Brown, Jr., 1959. Timing long-cycle physiological rhythms. Physiol.
Rev., 39: 127-161.

MAGNETIC RESPONSE OF AN*ORGANISM AND ITS
LUNAR, RELATIONSHIPS;*

PF, A. BROWN, JR., H. M. WEBB ANDIW. j:c5RETSE

Departments of Biological Sciences, Northwestern University, Goucher College and
Indiana State Teachers College,.and the Marine Biological Laboratory,
Woods Hole, Mass.

It has been demonstrated (Brown, Brett, Bennett and Barnwell, 1960) that
mud-snails, Nassarius obsoleta, initially directed magnetic southward in a symmetri-
cal field constant for all factors normally considered able to influence their orienta-
tion, exhibit a daily rhythm in the direction of their mean paths. The amount of
dispersion of paths of a population of snails about their mean path also displays a
daily rhythm. Both of these characteristics of spatial orientation were shown to be
quantitatively alterable, in a manner highly significant statistically, by experi-
mentally changing the strength of the ambient magnetic field. Although much of
the observed variation, both in the orientation of control animals in the earth’s field
and in the modified orientation in response to experimentally increased strengths of
the magnetic field, was accounted for in terms of the daily rhythm of response,
much variation still remained unaccounted for. The following study was made to
determine whether lunar periodisms in responsiveness to the magnetic field were
also present.

METHODS AND MATERIALS

The apparatus, and methods of obtaining the data, utilized in this study have
already been reported in detail elsewhere (Brown, Brett, Bennett and Barnwell,
1960). In essence, measurements were made of the average amount of clockwise
or counter-clockwise turning of snails as they traversed a 3-cm. course immediately
following their emergence from a straight, narrow corridor directed toward the
magnetic south into an illuminated, symmetrical, constant field. Each of 564
experimental series, obtained during the period June 28 through August 29, 1959,
included two groups of exits of ten snails each in the earth’s natural magnetic field,
two groups of ten in a magnetic field increased 10-fold and oriented as the natural
one (N-S), and two groups of ten in a nearly equally strengthened field rotated
clockwise through 90° so that the north-seeking pole of the bar magnet was directed
west (E-W).

Data were obtained for hours of the day from 5 AM through 9 PM, E.S.T.,
with no hour of this period represented by fewer than nine series, nor more than
50 series, of 60 snail exits.

RESULTS
Lunar-day rhythm of mean path

All the data on orientation of the snails for each calendar day were rearranged
to assume their proper relationship as approximate hours of lunar days. When

1 This research was aided by a contract between the Office of Naval Research, Department
of the Navy, and Northwestern University, #1228-03.

382

RHYTHMS IN MAGNETIC RESPONSE 383

more than one series was recorded for any single hour of a given day, these were
combined to yield a single average value. The average direction and amount of
turning of the snails as a function of hour of the lunar day is indicated in Figure
1 for the control snails, those in the artificially augmented N—S magnetic field and
those in the E—W field. Each of the three differently treated groups is dealt with
here as a separate sample. Not only is there a clear mean lunar-day cycle of

PATH

MEAN

Z+7

ZENITH
LUNAR DAY

FicurE 1. The mean paths of snails in the E-W magnetic field (broken line), N-S
magnetic field (solid line), and control snails solely in the earth’s field (dotted line) as a
function of hours of the lunar day. Standard errors of means at two selected times are
depicted.

turning, but the three samples exhibit a considerable similarity in the form of the
mean cycle despite relatively wide differences between values for the three samples
within individual series of 60 (See Brown, Brett, Bennett and Barnwell, 1960).
Not only is the mean lunar-day cycle of turning of essentially the same ampli-
tude as the comparable, previously described solar-day one, but displays a general
similarity to it in gross form, though appearing to mirror-image the daily one in

384 F. A. BROWN, JR., BH: M: WEBB AND W. J: BRETT

secondary superimposed fluctuations over the period the moon is above the horizon.
As with the solar-day cycle, minimum left-turning or maximum right-turning oc-
curs about the fourth hour of the lunar day or near the time of moon-rise, and
maximum left-turning occurs at lunar nadir. Standard errors of the means for
the fourth (zenith minus 8) and the nineteenth hours (zenith plus 7), calculated
from all data for these hours, are indicated on the figure. These values indicate
the high degree of statistical significance of this lunar-day fluctuation of mean path
of orientation.

EREE Ca

MAGNET

NADIR ZENITH NADIR
LUNAR DAY

Figure 2. The difference of mean paths between snails in the E-W magnetic field
(broken line) and N-S magnetic field (solid line) from control snails as a function of hours
of the lunar day. The values are calculated from all series studied during overlapping three-
hour periods.

Lunar rhythm in effect of experimentally augmented magnetic fields

The average effect of the imposed magnetic fields for overlapping groups of
three lunar-day hours were expressed as differences from the comparable three-hour
control samples and the mean lunar-day fluctuation in this magnetic effect is plotted
in Figure 2. Such three-hour grouping of the data was employed (1) to reduce
as far as possible the apparent significance of individual points based upon relatively
small numbers of series, and (2) to reduce irregularities in the lunar-day cycle
reflecting incomplete randomization of the solar-day cycle as a consequence of the
occurrence of occasional solar days lacking in data. As indicated in Figure 2, a
maximum right-turning response to the experimental magnetic fields occurs near

RHYTHMS IN MAGNETIC RESPONSE 385

the time of moon-rise, and a maximum left-turning response is seen at lunar nadir.
This is true whether one deals with either the N-S or E—-W experimental fields.
The lunar-day pattern suggests three progressively decreasing maxima as one moves
through the lunar day. The overwhelming preponderance of negative values, in-
dicating the induction of left-turning by the experimental procedure, clearly cor-
roborates the results of the earlier solar-day study in demonstrating this response to
the experimentally imposed weak fields.

Z+6
2)
x
| od
<
a
>
WJ
ra)
Z-7T
fa)
a
<
fa)
Zz
<
| od
Y
NADIR ZENITH NADIR

LUNAR DAY

Ficure 3. The standard deviation of the paths of experimental and control snails as a
function of hour of lunar day. Sample standard errors are illustrated. See legend of Figure
1 for key.

Standard deviation of paths

In view of the solar-day rhythm of dispersion of paths established previously
and indicating both right and left turning responses to the magnetic field, the
standard deviations of all the samples of 20 controls, 20 animals in the N—-S experi-
mental field and 20 in the E—-W field were next rearranged to ascertain whether
a lunar-day fluctuation occurred also in this parameter. In Figure 3 is seen the
mean lunar-day fluctuation of standard deviations for the month-period, July 6—
August 4, for each of the two experimental conditions and the controls, separately.
Standard errors of the means for the seventh hour before zenith and the sixth hour

386 F. A. BROWN, JR., H. M. WEBB AND W. J. BRETT

MAGNET EFFECT

MAGNET EFFECT

=0'5

o° =e 10° -1809 -90? o°
ELONGATION OF THE MOON

Ficure 4. A-B.

RHYTHMS IN MAGNETIC RESPONSE 387

after zenith indicate the highly significant character of this lunar-day cycle. When,
instead, all data for the two-month period were used, this cycle was very similar,
but had slightly increased amplitude and statistical significance. However, the
particular monthly period was selected since the most complete series of daily
studies were conducted during this period, and, hence, more complete randomization
of the solar-daily cycle was assured.

Semi-monthly cycle in the mean daily sign and amount of turning

A semi-monthly rhythm of response to the experimental magnetic fields was
noted. This was characterized by the experimental snails turning to the right of
controls just before each new and full moon, and turning maximally to the left of
controls near the times of the moon’s quarters. Differences between the paths of
experimental snails in the N-S or E—W fields and controls were obtained for a

TABLE |

Mean daily signs of magnetic field responses obtained as a function of day of a natural
semi-monthly period

Z,
=
g
hy
=<
a
te
ae
|

+
co
+++++1+e4++C+41
I

apse ioe eal eesiees|
|
|
|
|

total of 64 days during the two-month period of study. The sign of the average
response obtained for each of the days on which any series were obtained, as related
to the days of new or full moon, is indicated in Table I. In view of the paucity, or
absence, of data on some days and the relative abundance on other days it was felt
that a more accurate determination of the mean form of this semi-monthly cycle
could be obtained by calculating averages of all available series for overlapping
periods of three consecutive days each. The results of such a procedure are plotted
in Figure 4A.

The values were obtained by considering each day on a semi-monthly frame of

Ficure 4. A. Semi-monthly cycle in mean daily difference between paths of experimental
and control snails. The mean values from overlapping three-day periods of study comprise
the points. B. The solid line indicates the fluctuation through the synodic month of the re-
sponse to the experimental magnetic field based upon 3- to 5-day grouped data, selected to
provide mean values at about 45° intervals of elongation of the moon relative to the sun. The
broken line shows the relationship after averaging the original data for the two semi-monthly
periods, considering both the times of new and full moon as 0°, and using the single range 0°
to 90°.

388 F. A. BROWN, JR, H.'M. WEBB AND W. J:; BRETT

reference of the scales new moon through new moon minus 14 days and full moon
through full moon minus 14 days. This provided an arbitrary 15-day semi-
monthly period, an adequately close approximation, for the purposes at hand, to
the mean, natural 14.8-day one.

In order to assay the statistical significance of this semi-monthly cycle the
correlation of the effect of the magnets with the elongation of the moon relative to
the sun was assayed. The daily angle of elongation is 12.2°. However, inspection
of Figure 4A suggested that the maxima and minima preceded the days of the new
and full moon, or its quarters, by about half a day, or by about — 6°. ‘This value,
consequently, was treated as the corrected zero one. The data were next combined
into eight groups of days in the synodic month: (1) New moon, — 1, and —2;
(2) —3, —4, —5, —6; (3) —7, —8, —9; (4) — 10, — 11, —12, —13, —14;
(5) Full moon —1, — 2; (6) — 3, — 4, — 5, —6; (7) —7, — 8, — 9; and (8)— 10,
—11, —12, —13, —14. These times were selected to give values centered on
times close to those in which the earth-moon and sun-moon axes were parallel
with one another (0°), at right angles to one another (90°) and midway between
these two (45°). The eight values are plotted in Figure 4B, now corrected for
the — 6°. When these were treated as linearly correlated with the angular elonga-
tion of the moon corrected to zero at — 6° and one now remained with the 0 to
— 90° range of each quadrant of the monthly cycle of lunar elongation, a. coefficient
of correlation of 0.8486 (P < .01) was found. When the same data were further
reduced by combining the original data for those pairs of values related to one
another by 180°, r became extremely large, 0.9994 (broken line in Figure 4B).
This last is obviously a very highly significant value. The substantially lower
value of r with N = 8 than with N = 4, in this instance appeared to reflect a quite
symmetrical difference between the general forms of the pre-new-moon and pre-
full-moon semi-monthly cycles.

The response to experimental magnetic fields appears definitely correlated with
the elongation of the moon, and in a rather regular manner. However, the re-
strictions in the number of values used in the reduced data clearly do not permit
one to conclude with any certainty that the correlation is a truly linear one.

Synodic monthly fluctuation in mean daily standard deviation of paths

In Figure 5A are plotted together the mean daily standard deviations of paths
of the controls, and the experimental snails in the N-S and E-W magnetic fields.
Each is treated separately. During the first forty days of the study there appears
to be a synodic monthly fluctuation in standard deviation, with the maximum devia-
tion about two days prior to new moon and minimum deviation about the time
of full moon. The last 13 days of data are not in any fundamental manner
incompatible with the view of a continuing monthly periodism of the same gen-
eral character, but do contain, over a short period, some tremendous fluctuations
in the standard deviations which clearly can not themselves be part of a simple
monthly cycle of the relative regularity of the first. This suggests significantly
differing responses of the snails of an aperiodic character, in the constant con-
ditions.

Figure 5B is a plot of the mean daily barometric pressure values for the corre-
sponding period of time. The barometric pressure changes appear related more

RHYTHMS IN MAGNETIC RESPONSE 389

than coincidentally to the fluctuations in standard deviations of the snail orienta-
tion. Not only are the snail orientation and the barometric pressure fluctuations
the mirror-images of one another in gross general trends, but when the barometric
pressure fluctuation is treated as a two-day lead correlation on snail orientation,

1.40

z
120
<
a
a
Q 1.00
(a)
a
<
Z
Z .80
Y
.60
30.2
WJ
a
2
Y
Y
W 30.0
x
a
F
< 29.8

ve)
oO
1959 =

JUL |

Ficure 5. A. The mean daily standard deviations of all snail paths for the whole period
of study. Open circles, snails in E-W magnetic field; closed circles, in N-S magnetic field;
half open circles, controls. B. Mean daily barometric pressure for the period corresponding

to (A).

even a number of the major irregularities in the fluctuations of snail orientations,
and particularly those of the last thirteen days, become strikingly accounted for in
these terms. The calculated two-day lead correlation was r= 0.33 + 0.08 (N =
141), clearly highly significantly different from zero.

390 F. A. BROWN, JR., H. M. WEBB AND W. J. BRETT

DISCUSSION

There appears little reason to doubt that spatial orientation of snails, expressed
as an amount of turning, or a klinokinesis, possesses a lunar rhythmicality. This
is displayed first as lunar-daily, or approximately 24.8-hour, periodicities in both
mean orientational paths with signs taken into account and total turning, both
clockwise and counterclockwise. The latter is indicated here in terms of standard
deviation of pathways. It is interesting that the form of the lunar-day cycle, in
the range of hours corresponding roughly with those available for the solar-day
study, namely 5 AM through 9 PM, is strikingly similar to that of the solar-day
one in both gross features and phase relationships. At both the times of sun-rise
and moon-rise the left turning is minimal and it generally increases towards the
early solar and lunar “evenings,’—1.e., times of setting of sun or moon. Also, in
both the solar and lunar days standard deviation of pathways is minimal about the
time of sun-rise and moon-rise, and increases systematically while the sun and
moon are above the horizon, decreasing again as these heavenly bodies set. Fur-
thermore, the amplitudes of the solar and lunar cycles are of the same general
magnitude. The presence of these two similarly conspicuous periodisms would
be expected to give rise to a synodic monthly cycle, and such a cycle appears to
be present. The occurrence of a semi-monthly cycle of influence of the experimental
magnetic fields suggests that either the lunar-day or solar-day cycle of mean path
is bimodal. Though the former is suggestive of bimodality, the complete form
of the solar-daily cycle has not been determined. A semi-monthly rhythm would
conform to that of the recurrence of the tides at a particular hour of the day in
the natural habitat.

It is evident that since the snails were always subjected, whether in the field
or laboratory, to daily illumination changes, a solar-daily cycle could be imparted
to the snails by light cycles. The lunar-day period, on the other hand, could be
gained by the snails in the field from the tides, but during the periods of as long
as 8 to 10 days when the snails were retained in the laboratory any exogenous
lunar period in terms of obvious environmental factors was not available. In
view of the persistence of the lunar-day rhythm with conspicuousness equal to
that of the solar-daily period, it would appear probable that both the lunar-day
cycle and solar-day cycles are being derived continuously in response to some
unidentified subtle environmental factors in a manner comparable to that demon-
strated for the solar and lunar periodisms in Uca pugnax (Webb and Brown,
1958; Brown, 1959) and of the metabolism of Nassarius itself (Brown, Webb
and Brett, 1959).

The similarities between barometric pressure changes and certain parameters
of spatial orientation of snails suggest strongly that orientation in the snails is in
some manner dependent upon subtle environmental factors other than magnetic
fields and that these other factors may also induce considerable alteration in snail
behavior. This is indicated, since even the experimentally augmented magnetic
fields may be overriden by the factor correlated with the barometric pressure.
One possible factor which has been suggested is electrostatic field (Webb, Brown
and Brett, 1959). The correlation with barometric pressure recalls the correlation
observed for metabolic fluctuations in the snails in constant conditions, including
pressure, during the summer of 1958 (Brown, Webb and Brett, 1959), except

RHYTHMS IN MAGNETIC RESPONSE 391

' for its being a mirror-image. Since there is reason to believe (Brown, Brett,
Bennett and Barnwell, 1960) that the spatial orientation described in this study
is at least in some measure a klinokinetic response to the natural and experimental
magnetic fields, it is possible that fluctuations in other subtle geophysical factors
can influence the sensitivity of the magnetic responder-system. However, the
possibility can not be excluded that a responder system for another factor is being
altered by the magnetic field changes. The apparent effectiveness of certain
barometric pressure correlates in modifying the orientational responses, and the
well-established role of these correlates in influencing the rate of cellular oxida-
tions, suggest that the sensitivity to the magnetic field is in some manner related
to rate of oxidative metabolic changes.

These striking relationships between strength and character of magnetic orienta-
tion and phases of the lunar-day and synodic monthly periods contribute, together
with the earlier demonstrated solar-day relationships, a substantial degree of pre-
dictability to the responses of the snail to magnetic forces. Collectively they in-
crease still more not only the probability that this magnetic response is real, but
that it plays a significant role in both the lunar-dominated tidal and the solar-day
periodic behavior of these animals.

It is interesting, further, to note that in the lunar-day fluctuation in orientation,
with sign taken into account, those times of lunar day when the mean path of the
snails is essentially straight or even slightly to the right, the experimentally
augmented magnetic field effects turning to the right, and as the snails normally
orient during the lunar day more and more to the left as one approaches lunar
nadir, the effect of the increased magnetic field is to produce progressively stronger
left-turning. This relationship is quite comparable to that found for the solar
day and provides further support for the view that response to the earth’s natural
field normally occurs.

There has been noted a remarkable similarity in both the gross form and the
detailed trimodal character of progressively decreasing maxima of the lunar-day,
N-S magnetic field responses of snails and the mirror-image of a highly significant
mean lunar-day cycle of spontaneous motor activity found in white mice over
the 5-month, partially overlapping period, March through July, 1959 (Terracini
and Brown, unpublished) similarly treated as a three-hour moving mean. The
maxima and minima were essentially synchronous when the two cycles, studied
in two places, Massachusetts and Illinois, were adjusted to simultaneity (universal
time). The coefficient of correlation for the two trimodal cycles was — 0.85.
Similarly high simultaneous correlations have been reported earlier between fluctu-
ations of the nucleonic component of cosmic radiation in Illinois and fiddler crab
and sea-weed metabolism in constant conditions in Massachusetts (Brown, Webb
and Bennett, 1958). This similarity between the cycles of mice activity and re-
sponse of snails to the magnetic field supports the hypothesis that in this magnetic
response we are dealing with a widely, probably universally, occurring biological
phenomenon.

The proof of a remarkable sensitivity of an organism to one subtle environmental
factor (magnetic field) operating at an intensity only slightly above the earth’s
natural one, and the evidence for its modification in these constant conditions by
fluctuations in a second natural factor, poses even the very fundamental problem

392 F, A. BROWN, JR., H. M. WEBB AND W. J. BRETT

of the dispensability, or indispensability, of the earth’s milieu of subtle factors and
its natural fluctuations, as a normal ecological consideration in organismic survival
and species propagation.

SUM MARY

1. The direction, and mean amount, of turning in snails initially directed
southward into a constant symmetrical, illuminated field displays a lunar-day
rhythm with minimum turning about the time of moon-rise, and maximum turning
at lunar nadir. There is also a lunar-day cycle of standard deviation of snail
pathways, with a minimum about moon-rise and a maximum near moon-set.

2. The response of snails to an experimentally augmented magnetic field also
exhibits a lunar-day rhythm with maximum turning to the left at lunar nadir.

3. The specific character of the lunar-day rhythm of the response to the ex-
perimental magnetic fields gives further support for the view that magnetic field
is normally involved in snail orientation.

4. The mean daily response of snails to experimental magnetic fields, expressed
as differences from the response of controls in the earth’s natural field, displays a
semi-monthly rhythm. Maximum right-turning in response to a magnetic in-
crease of 10-fold over that of the earth occurs one to two days before new and
full moon, and maximum left turning just before the times of the first and third
quarters of the moon.

5. There is a synodic monthly fluctuation in mean daily standard deviations of
- snail paths with maximum deviation about two days before new moon and minimum
deviation about the time of full moon.

6. Some suggestive correlations are demonstrated between barometric pressure
and the spatial orientation of snails in an environment constant with respect to all
generally accepted orienting factors.

7. It is pointed out that similarities in influence of some unidentified barometric
pressure correlates on (a) magnetic orientation of snails, (b) general cellular
oxidations, and (c) spontaneous activity cycles present reasons for postulating
that the latter two phenomena are in some manner related to the magnetic field
response and suggest that response to magnetic field is a widely distributed bio-
logical phenomenon.

8. Evidence is presented that suggests there is a biological influence of a uni-
versal-time-related, rhythmic, environmental factor.

LITERATURE CITED

Brown, F. A., Jr., 1959. Living clocks. Science, 130: 1535-1544.

Brown, F. A., Jr., W. J. Brett, M. F. BENNETT AND F. H. BARNWELL, 1960. Magnetic re-
sponse of an organism and its solar relationships. Biol. Bull., 118: 367-381.

Brown, F. A., Jr., H. M. Wess anp M. F. Bennett, 1958. Comparisons of some fluctuations
in cosmic radiation and in organismic activity during 1954, 1955 and 1956. Amer. J.
Phystol., 195: 237-243.

Brown, F. A., Jr., H. M. Wess ann W. J. Brett, 1959. Exogenous timing of solar and lunar
periodisms in metabolism of the mud-snail, [/lyanassa (Nassarius) obsoleta, in labora-
tory constant conditions. Gunma J. Med. Sct., 8: 233-242.

Wess, H. M., ano F. A. Brown, Jr., 1958. The repetition of pattern in the respiration of
Uca pugnax. Biol. Bull., 115: 303-318.

Wess, H. M., F. A. Brown, Jr. AND W. J. Brett, 1959. Effects of imposed electrostatic fields
on rate of locomotion in Ilyanassa. Biol. Bull., 117: 430-431.

INVESTIGATION OF THE HORMONES CONTROLLING THE DISTAL
RETINAL PIGMENT OF THE PRAWN PALAEMONETES +

MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

Department of Zoology, Newcomb College, Tulane University, New Orleans 18, Louisiana,
and Marine Biological Laboratory, Woods Hole, Massachusetts

The distal retinal pigment of the prawns Palaemon and Palaemonetes has been
observed after appropriate stimulation in the fully light-adapted position and the
fully dark-adapted one as well as in an intermediate state (Kleinholz and Knowles,
1938; Sandeen and Brown, 1952). The position of the distal pigment was a
function of the brightness of the visual field.

Migration of the distal pigment is mediated by blood-borne substances. Klein-
holz (1936) found that extracts of eyestalks from Palaemonetes caused this pig-
ment to migrate toward the fully light-adapted position. Brown, Hines and
Fingerman (1952) obtained the same response with extracts of supraesophageal
ganglia. The latter investigators also provided indirect evidence for a hormone
that would move the distal retinal pigment toward the fully dark-adapted position.
Direct evidence for a dark-adapting hormone in Palaemonetes was supplied by
Fingerman, Lowe and Sundararaj (1959). When extracts of eyestalks were
injected into prawns whose distal pigment was in a position approximately mid-
way between the fully light-adapted and dark-adapted states, a light-adaptational
response occurred and lasted about two hours. A dark-adaptational response
that lasted approximately five hours followed the light-adaptational one. The
supraesophageal ganglia appeared to contain light-adapting hormone alone. No
dark-adaptional response was obtained with extracts of eyestalks that was not
preceded by a light-adaptational one.

The major objectives of this investigation carried out on Palaemonetes were
two-fold. The first aim was to learn if the amount of light-adapting hormone in
the supraesophageal ganglia could be altered by maintaining specimens in light and
in darkness. This aspect was part of a continuing investigation in our laboratory
of the effects of long-term adaptation upon endocrine sources in crustaceans. The
second objective was to effect at least a partial separation of the distal retinal
pigment light-adapting and dark-adapting hormones from each other.

MATERIALS AND METHODS

Two species of prawns, Palaemonetes vulgaris and P. pugio, collected in the
region of Woods Hole, Massachusetts, were used in this investigation.

The method used to determine the position of the distal retinal pigment was a
slight modification of that devised by Sandeen and Brown (1952). The prawns
were placed one at a time, ventral surface down, on the stage of the stereoscopic

1 This investigation was supported in part by Grant No. B-838 from the National Insti-
tutes of Health.

393

394 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

dissecting microscope. With the aid of an ocular micrometer and transmitted
light the following measurements were made: (1) the width of the translucent
portion of the compound eye in a plane parallel to the long axis of the eyestalk,
and (2) the length of the eye from the corneal surface to the distal edge of the
dorsal pigment spot at the base of the eye proper. Sandeen and Brown (1952)
had measured to the proximal edge of this pigment spot. To render the distal
clear portion of the eye more translucent and the proximal edge of this area more
definite, the prawns were submerged in a dish of sea water on the stage of the
microscope. The ratio of width of clear area (measurement 1) to total length
(measurement 2) will be referred to as the distal retinal pigment index. Use
of this ratio minimized the effect on the data of size differences. In a fully dark-
adapted eye the distal retinal pigment abutted against the cornea, the mean distal
retinal pigment index was 0.00. In a fully light-adapted eye the distal pigment
index was about 0.25.

For all experiments wherein prawns were injected with tissue extracts the
assay animals were placed into black enameled pans containing sea water about
one inch deep. The pans were then placed at least one hour before the animals
were injected under an illumination of such intensity that the distal retinal pigment
was approximately midway between the fully light-adapted and dark-adapted
positions. The light intensity used in each experiment will be stated below.
Each extract was assayed on 10 prawns and the dose was 0.02 ml. per specimen.
The assay animals had had one eyestalk removed at least 12 hours before use in
an experiment. Because the eyestalk contained a large quantity of light-adapting
hormone, one-eyed prawns would not be as readily able to antagonize injected
dark-adapting hormone as would intact specimens.

Student’s ¢ test was used for determination of the level of significance between
means. The 5% level was considered the maximum for a significant difference.
Standard deviations and standard errors for the differences between means were
also calculated. The results of the statistical analyses are summarized in Table I.

A Model E-800-2 Filter Paper Electrophoresis Apparatus manufactured by
the Research Equipment Corporation, Oakland, California, was housed in a con-
stant temperature room maintained at 43° F. on the days the electrophoretic
analyses reported below were performed. For each experiment 40 eyestalks were
extracted in 0.2 ml. distilled water. The extract was then centrifuged and ap-
plied with the aid of a hot air blower across a one-half-inch-wide Whatman No.
1 filter paper strip. The region of application was never more than 4 inch long.
This strip was then moistened with 0.1 M sodium hydroxide-boric acid buffer
of pH 9.0 and placed into the migration chamber. After electrophoresis had
proceeded for two hours at 500 V. and 0.1 mA. the strip was removed from the
chamber. The region of application of the extract and one three-inch section from
each side of the origin were placed into individual covered dishes containing 0.3
ml. sea water. These dishes were kept in the cold room for 30 minutes to allow
materials to wash from the filter paper. As a control in the electrophoretic
analyses whenever a strip was removed from the migration chamber a strip of
filter paper three inches long was moistened with buffer and placed into 0.3 ml. sea
water for 30 minutes in the cold room. The fluid from the containers was then
collected in syringes for injection into assay animals.

HORMONES AND RETINAL PIGMENT 395

TABLE [

Summary of the statistical analyses of the data presented herein. N, number of distal
pigment indices used in the analysis; S.D., standard deviation; S.E., standard
error of the difference between the means; t, Student's t; p, probability
value. See text for complete explanation.

iV 43 0.110 0.05—0.22 0.0356 0.00739 0.91 | 0.40
r 39 0.116 0.05-0.18 0.0326
2V 20 +0.034 (—0.02)—( +-0.11) 0.0319 0.01044 0.73.1 0:50
P 20 +0.041 ( —0.04)—(+0.11) 0.0340
3E 30 0.221 0.14-0.30 0.0468 0.0114 9.66 | 0.001
C 30 0.136 0.09-0.23 0.0395
4E 29 0.159 0.10-0.28 0.0363 0.00835 2.33 | 0.05
Cc 30 0.136 0.09-0.23 0.0268
5% 95 0.142 0.05—0.28 0.0376 0.00714 5.52 | 0.001
D 88 0.188 0.09-0.30 0.0519
6S 28 0.162 0.03—-0.28 0.0754 0.0176 3.53 | 0.01
F 9 0.253 0.18-0.29 0.0309
7E 48 0.249 0.11-0.33 0.0578 0.01022 6.10 | 0.001
C 54 0.180 0.10-0.25 0.0432
8E 112 0.087 0.03-0.23 0.0311 0.00436 14.7 | 0.001
Cc 95 0.135 0.07-0.21 0.0299
9E 55 0.224 0.11-0.38 0.0474 0.00827 7.24 | 0.001
Cc 48 0.131 0.03—-0.23 0.0362
10 E 75 0.079 0.03-0.21 0.0311 0.00514 6.02 | 0.001
c 87 0.109 0.05-0.21 0.0342
11 W 55 0.224 0.11—-0.38 0.0474 0.00953 2.81 | 0.01
/ 55 0.185 0.10—0.29 0.0524
i2E 62 0.086 0.03-0.17 0.0359 0.00622 2.49 | 0.02
e 68 0.104 0.05-0.17 0.0348

Analysis 1, Background adaptation of P. vulgaris (V) and P. pugio (P).

Analysis 2, Light-adapting potency of supraesophageal ganglia of P. vulgaris (V) and P. pugio (P).

Analysis 3, Supraesophageal ganglia of specimens two weeks in darkness (E) versus control (C).

Analysis 4, Supraesophageal ganglia of specimens two weeks in light (E) versus control (C).

Analysis 5, Supraesophageal ganglia of prawns in light (L) versus supraesophageal ganglia of
prawns in darkness (D).

Analysis 6, Distal pigment index at start (S) and finish (F) of experiment.

Analysis 7, Eyestalk light-adapting hormone (E) versus control (C).

Analysis 8, Eyestalk dark-adapting hormone (E) versus control (C).

Analysis 9, Eyestalk light-adapting hormone (E) versus control (C).

Analysis 10, Eyestalk dark-adapting hormone (E) versus control (C).

Analysis 11, Comparison of light-adaptation produced by extracts of eyestalks without trypsin
(W) and with trypsin (T).

Analysis 12, Dark-adaptation after electrophoresis of eyestalks (E) versus control (C).

396 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

EXPERIMENTS AND RESULTS
The distal retinal pigment of P. pugio and P. vulgaris

The specimens of Palaemonetes in the Woods Hole area had always been
considered to be P. vulgaris (cf. Allee, 1923). However, Holthuis (1952)
separated the prawns of the Woods Hole area into three species, P. vulgaris, P.
pugio and P. intermedius. No investigator has reported specific differences of
retinal pigment cells in Palaemonetes.

Examination of 451 specimens used in the present investigation revealed the
presence of specimens of two of the species only, P. vulgaris and P. pugio. An
overwhelming percentage (92.0%) was P. vulgaris. The present investigators
used Palaemonetes without regard to species. This procedure was justifiable
because the following two experiments revealed that the distal retinal pigments
of P. vulgaris and P. pugio were physiologically indistinguishable.

(1) This experiment was designed to answer the following question. Are the
mean distal retinal pigment indices of specimens of P. vulgaris and P. pugio ex-
posed to the same intensity of illumination substantially the same? Forty-three
specimens of P. vulgaris and 39 of P. pugio were placed into black enameled con-
tainers under an illumination of 31 ft. c. After one hour the distal retinal pig-
ment index of each specimen was determined. The mean distal pigment indices
were 0.110 for P. vulgaris and 0.116 for P. pugio. Statistical analysis of these
data revealed no significant difference between the means (Table I, Analysis 1).

(2) The next experiment answered two questions. Do differences exist be-
tween the light-adapting potencies of extracts of supraesophageal ganglia from
the two species? Are the responses of specimens of both species to the same
extract different? Extracts with a concentration of 10 organs per ml. of sea
water were prepared from supraesophageal ganglia of P. vulgaris and P. pugio
and were assayed on 10 specimens of each species. The assay animals were in
black enameled pans under an illumination of 31 ft. c. As a measure of the light-
adapting potency of each extract, activity (potency) values were calculated in the
following manner. The sum of the mean distal pigment indices, determined 30
and 60 minutes after the prawns were injected with the extracts, was subtracted
from the sum of the means of the indices determined at the same time intervals
with control specimens of the same species as the prawns injected with the extract
of supraesophageal ganglia. The controls were injected with sea water. This
difference between the sums was a measure of the potency of the extract. Thirty
and 60 minutes were chosen because maximal light-adaptation in response to ex-
tracts of eyestalks and supraesophageal ganglia always occurred 30-60 minutes
after injection. The activity values were as follows: “vulgaris” supraesophageal
ganglia into “pugio,’ 0.068; “vulgaris” into ‘vulgaris,’ 0.068; “pugio” into
“vulgaris,” 0.076; “pugio” into “pugio,” 0.082. An extract made from the supra-
esophageal ganglia of animals randomly selected from the stock aquaria and
assayed on randomly selected assay animals was 0.071. The “vulgaris” into
“pugio” and “pugio” into “pugio” potency values were analyzed statistically to de-
termine if the difference between them was significant. These values were chosen
because they were at the extremes. For the statistical analysis the mean distal
pigment index of the control specimens was subtracted from each index obtained

HORMONES AND RETINAL PIGMENT 397

from prawns injected with the extracts. The algebraic sign was noted. As antici-
pated, the two sets of data were not significantly different from each other (Table I,
Analysis 2).

Influence of light and darkness upon the amount of light-adapting hormone in the
supraesophageal ganglia

One group of prawns was placed in a darkroom, a similar group was placed
into white enameled pans and exposed to a constant illumination of 31 ft. c.
After 14 days the supraesophageal ganglia of specimens of both groups were ex-

0.22

INDEX
O
©

0.18

0.16

0.14

DISTAL PIGMENT

0 | 2
HOURS

Ficure 1. Responses of the distal retinal pigment to extracts of supraesophageal ganglia
of prawns maintained on a white background under an illumination of 31 ft. c. (circles) and
in darkness (dots) for 14 days. Control, half-filled circles.

tracted in sea water (10 organs/ml.) and assayed on one-eyed prawns in black
enameled pans under an illumination of 31 ft. c. Control specimens received
injections of sea water.

Extracts of the supraesophageal ganglia of both groups produced a light-
adaptational response. However, the responses of the. assay animals to the
organs of specimens that had been in darkness was’ much greater than that pro-
duced by the extracts of supraesophageal ganglia of specimens that had been

398 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

illuminated. The experiment was done two more times. The results were
qualitatively the same as those obtained the first time the experiment was done.
Data of the three experiments were consequently averaged and the means were
used in the preparation of Figure 1. The light-adaptational responses of both
groups of assay prawns were statistically significantly different from the controls

0.25

0.16

DISTAL PIGMENT

O 30 60 90 120 150 180
MINUTES

Ficure 2. Response of the distal retinal pigment to several concentrations of extracts of
supraesophageal ganglia of animals collected less than 24 hours before use. Circles, one organ
per 0.02 ml. sea water; dots, % organ per 0.02 ml.; circles half-filled on left, %4 organ per 0.02
ml.; circles half-filled on right, 4% organ per 0.02 ml.; circles half-filled on bottom, 46 organ per
0.02 ml.; triangles, control of sea water.

(Table I, Analyses 3, 4). The distal pigment indices determined 30 minutes
after the extracts had been administered were used in the statistical analyses.
' ‘The difference between the responses of assay animals to the extracts of the
supraesophageal ganglia of the specimens that had been in light and in darkness
was also highly significant (Table I, Analysis 5). Analysis 5 was based on the

HORMONES AND RETINAL PIGMENT 399

distal pigment indices determined 30, 60 and 90 minutes after the extracts had
been injected.

The next experiment was designed to learn how the amount of light-adapting
hormone in the supraesophageal ganglia of freshly collected specimens compared
with the amounts of this hormone in the supraesophageal ganglia of prawns that
had been in darkness or in light for 14 days. Supraesophageal ganglia were re-
moved from prawns that had been collected less than 24 hours prior to use in
the experiment and triturated with sufficient sea water to yield extracts that con-
tained Y5, 1%, %4, % and 1 organ per 0.02 ml. sea water. Each extract was
assayed on 10 one-eyed prawns in black pans under an illumination of 31 ft. c.
The responses were observed for three hours. This experiment was performed
three times. The data of the three experiments were then averaged and the
means were used in the preparation of Figure 2.

Two facts were apparent from inspection of the figure: (1) the higher the
concentration of supraesophageal ganglia in the extracts, the greater the distal
pigment index 30 minutes after the extracts were injected, and (2) the rate
of return of the index to the control level was faster with the higher concentra-
tions than with the lower ones. Furthermore, no indication of a dark-adapting
response was apparent even at the highest concentration.

As a measure of the light-adapting potency of the extracts the differences
between the means of the distal pigment indices of the controls and of prawns
receiving extracts of supraesophageal ganglia were summed. The sums (potency
values) were used to compile Figure 3. In like manner, potency values were
calculated for the extracts of the supraesophageal ganglia of the specimens that
had been in light (0.034) and in darkness (0.204) for 14 days. The coordinates
for the supraesophageal ganglia of specimens that had been in light 14 days (0.2
organ/0.02 ml. and a potency value of 0.034) almost fell on the curve of Figure
3. As the curve was drawn, the potency value expected from an extract contain-
ing two-tenths of the supraesophageal ganglia of one prawn per 0.02 ml. was
0.045. Illumination, therefore, had no appreciable effect on the amount of light-
adapting hormone. Further inspection of the figure revealed that a potency value
of 0.204 would be caused by an extract of supraesophageal ganglia of freshly
collected prawns with a concentration of eight-tenths of an organ per 0.02 ml.
Presumably, therefore, the amount of light-adapting hormone in the supraesophageal
ganglia quadrupled during the two weeks the specimens had been in the dark-
room.

Effect of time in darkness on the ability of the distal retinal pigment to respond
to a dark-to-light change

The object of this experiment was to determine if the period of time specimens
were kept in darkness influenced their subsequent rate of light-adaptation. Prawns
were collected, put into white enameled pans and placed in darkness. Two hours
later the prawns were exposed to an illumination of 31 ft. c. for 30 minutes.
The distal pigment indices of 10 specimens were determined and the prawns were
again put in darkness. They were subsequently exposed to illumination after
having been in darkness 3, 5, 7, 10, 12 and 14 days. The mean distal pigment
indices determined after 30 minutes of illumination gradually increased.

400 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

The experiment was repeated and the results were substantially the same.
The data of the two experiments were averaged and the means were used in the
preparation of Figure 4. The number of prawns gradually decreased while each
experiment was in progress. Totals for both experiments were: 20 on the first

03

Oo
N

POTENCY

0.!

0.0

-25 -1.00 -O75 -0O50 -0.25 000
LOG CONCENTRATION

Figure 3. Relationship between light-adapting potency and concentration of extracts of
supraesophageal ganglia of specimens collected less than 24 hours before use. The concen-
tration is the logarithm of the number of supraesophageal ganglia each assay animal received
triturated in 0.02 ml. sea water.

and third days, 16 on the fifth day, 15 on seventh and tenth days, 12 on the twelfth
day and nine on the last day of the experiment.

The indices recorded on the first and last days of both experiments were
analyzed statistically (Table I, Analysis 6) ; the difference between the means was
highly significant. The longer the prawns had been in darkness, the faster the
distal pigment approached the fully light-adapted position.

HORMONES AND RETINAL PIGMENT 401

Digestion of the light-adapting and dark-adapting hormones by trypsin

The object of the first experiment of this series was to demonstrate that a
boiled solution of trypsin would not inactivate the retinal pigment hormones.
After boiling, the enzyme would have been inactivated but some contaminant
might have possibly destroyed the hormones. This possibility had to be ruled
out. For the first experiment of this group a 2% solution of trypsin in sea water
was prepared; at the same time a freshly prepared extract of eight eyestalks
in 0.2 ml. sea water was boiled for 30 seconds and centrifuged. Equal volumes
of the boiled trypsin solution and the eyestalk extract were mixed together. The

INDE X

DISTAL PIGMENT
oO
©

oO
o

6) 2 a 6 8 te) lo 4
DAYS

Ficure 4. The distal retinal pigment index 30 minutes after prawns had been taken from
darkness and placed on a white background under an illumination of 31 ft. c., versus the
number of days the specimens had been in darkness.

mixture was then injected into one-eyed prawns in black pans under an illumina-
tion of 22 ft.c. Boiled trypsin solution diluted with an equal volume of sea water
was injected into the controls.

The results were those that were anticipated. A light-adaptational response
occurred and lasted two hours. A dark-adaptational response followed. The
extent of the latter was not followed to its completion. Fingerman, Lowe and
Sundararaj (1959) reported that the dark-adaptational response lasted at least
five hours. The experiment was performed two more times and the results were

402 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

O | 2 3 4 5
HOURS

Ficure 5. Responses of the distal retinal pigment to extracts of eyestalks. Top, the
extract was prepared in boiled, centrifuged trypsin solution, circles. Control of boiled,
centrifuged trypsin in sea water, dots. Middle, extract prepared in sea water and kept for
one hour, circles. Sea water control, dots. Bottom, extract in trypsin solution for one hour,
circles. Control, dots.

HORMONES AND RETINAL PIGMENT 403

the same. The data of the three experiments were averaged and the means were
used in the preparation of the top portion of Figure 5.

These data were analyzed statistically (Table I, Analyses 7, 8) and the
amounts of light-adaptation and dark-adaptation were significant. The distal pig-
ment indices obtained 30 and 60 minutes after the extracts were injected were
used for Analysis 7; the indices determined 3.0, 3.5, 4.0 and 4.5 hours after the
extracts were injected were used in Analysis 8.

For the next experiment an extract was prepared of 18 eyestalks in 0.45
ml. sea water. Four-tenths ml. of the centrifuged extract was divided into two
equal portions. To one fraction was added 0.2 ml. sea water and to the other
0.2 ml. of 2% trypsin solution in sea water. Both tubes were kept for one hour
at room temperature, 24° C., at which time the material in the tubes was boiled for
30 seconds and centrifuged. The supernatant from both tubes was injected into
assay animals. Boiled, centrifuged trypsin solution was injected into one-eyed
prawns as a control for the extract of eyestalks treated with trypsin. As a con-
trol for the extract not treated with trypsin, sea water was injected into one-eyed
prawns.

The extract that did not contain trypsin caused light-adaptation and subse-
quent dark adaptation. The trypsin-treated extract caused a small amount of
light-adaptation and no dark-adaptation. The experiment was performed two
more times and each time the results were essentially the same. The data of the
three experiments were averaged and the means were used in the preparation of
the middle (pure extract) and bottom (trypsin-treated extract) portions of Fig-
ure 5.

Statistical analyses (Table I, Analyses 9, 10) of the results depicted in the
middle portion of Figure 5 showed that the light-adapting and dark-adapting
effects were significant. The distal pigment indices determined 30 and 60 minutes
after the extracts were administered were used for Analysis 9, the indices ob-
tained after 4.0, 4.5 and 5.0 hours for Analysis 10.

To determine if the difference between the degrees of light-adaptation produced
by the pure and trypsin-treated extracts was significant, the distal pigment indices
determined 30 and 60 minutes after injection of the extracts were analyzed
statistically (Table I, Analysis 11) and found to be significant at the 1% level.
The amount of light-adapting hormone in the extract had been reduced by the
trypsin.

Electrophoretic analysis of the retinal pigment hormones

The object of the last set of experiments was to separate the dark-adapting
hormone in the eyestalk from the light-adapting one. Separation was accom-
plished by filter paper electrophoresis performed in the manner described above.
The experiment was performed three times.

The results of the three experiments were qualitatively alike and were averaged
(Fig. 6). The region of application of the extract to the strip and the three-inch
section of the strip on the cathodal side of the origin contained light-adapting
and dark-adapting substances. Fluid obtained from the three-inch section on the
anodal side of the origin produced dark-adaptation alone. Distal pigment indices
determined 90, 120 and 150 minutes after the extracts from the anodal portion

404 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

0.18

0.16

INDE X

0.14

O.l2

0.10

DISTAL PIGMENT

0.08

0.06

O | 2 S
HOURS

Ficure 6. Responses of the distal retinal pigment to extracts of eyestalks. These ex-
tracts had been subjected to filter paper electrophoresis. Circles, fraction that migrated toward
the cathode; dots, fraction that migrated toward the anode; half-filled circles, fraction that
remained at the origin; triangles, control.

of the strips and the control strips had been injected were compared statistically
(Table I, Analysis 12). The amount of dark-adaptation was highly significant.

DISCUSSION

The amount of light-adapting hormone in the supraesophageal ganglia in-
creased in specimens of Palaemonetes kept in darkness. This hormone in the
supraesophageal ganglia, therefore, must be normally involved in regulation of
the distal retinal pigment. The site of release of the light-adapting hormone from
the supraesophageal ganglia is unknown. It could be transported via axons to
a storage and release center in the eyestalks (e.g. sinus gland) or could be re-
leased directly from neurosecretory cells in the supraesophageal ganglia into the
blood.

HORMONES AND RETINAL PIGMENT 405

The light-adaptational response produced by the strongest extract of supra-
esophageal ganglia (Fig. 2) was of the same magnitude as that produced by the
extracts of eyestalks (Fig. 5). However, the extracts of the supraesophageal
ganglia did not cause a dark-adaptational response. The dark-adaptational re-
sponse produced by the eyestalk extracts might have been considered overcompensa-
tion by the prawns in removing the injected light-adapting hormone from the
blood. This interpretation, however, can not be correct in view of the results
obtained with the extracts of the supraesophageal ganglia (Fig. 2).

The data shown in Figure 4 reveal that the prawns adapted faster to light
the longer they had been in darkness. The observation that the supraesophageal
ganglia of prawns kept in a darkroom for 14 days contained more light-adapting
hormone than the organs of illuminated specimens can explain the results shown
in Figure 4. The animals light-adapted at a faster rate with increased time in
darkness because they had a greater store of light-adapting hormone to use.

Some physico-chemical properties of the retinal pigment hormones of Palae-
monetes can be adduced from these experiments. Both hormones are heat-stable.
With regard to the chemical nature of the retinal pigment hormones, there is in-
conclusive evidence that the neurosecretory products of crustaceans are polypep-
tides. The sensitivity of these substances to trypsin and chymotrypsin suggests
the presence of peptide bonds. Pérez-Gonzalez (1957) found that the hormone
that dispersed the black pigment in the fiddler crab, Uca pugilator, was inactivated
by chymotrypsin. Kleinholz (personal communication) found that the light-
adapting substance in the eyestalk of Palaemon serratus at Naples, Italy, was
sensitive to trypsin and chymotrypsin. Evidence was presented in Figure 5 for
the trypsin sensitivity of the retinal pigment dark-adapting and light-adapting
hormones in Palaemonetes. Trypsin and chymotrypsin, however, also have weak
esterase activity and inactivation by these enzymes, alone, is not sufficient proof
of the peptide nature of these neurosecretory substances.

Both retinal pigment hormones migrated in an electric field (Fig. 6). At pH
9.0 the dark-adapting hormone separated from the light-adapting hormone so that
light-adaptation did not precede the dark-adaptational response. In contrast,
whenever extracts of eyestalks were injected before electrophoresis a light-adapta-
tional response occurred before the dark-adaptational one (top portion of Figure
5). Because the dark-adapting hormone was found at the origin and on both
sides of it, pH 9.0 must be close to its isoelectric point. The isoelectric point of the
light-adapting hormone must be higher than pH 9.0 because no light-adapting
hormone was found on the anodal side of the origin.

SUMMARY AND CONCLUSIONS

1. Some physico-chemical properties of the hormones activating the distal
retinal pigment of the prawn Palaemonetes were determined.

2. The light-adapting and dark-adapting hormones were heat-stable, inacti-
vated by trypsin and migrated in an electric field. These findings suggest that
both hormones are polypeptides.

3. The amount of light-adapting hormone increased four-fold in the supra-
esophageal ganglia of prawns kept in darkness 14 days. The quantity of this
hormone in the supraesophageal ganglia of specimens illuminated for 14 days did

406 MILTON FINGERMAN AND WILLIAM C. MOBBERLY, JR.

not change detectably. These results demonstrated that the light-adapting hor-
mone from the supraesophageal ganglia was used by the prawns in regulating the
distal retinal pigment.

4. The longer prawns were kept in darkness, the faster was the rate of migra-
tion of the distal pigment toward the light-adapted position when prawns were
illuminated. This increase was explained by the increase in the light-adapting
hormone in the supraesophageal ganglia of specimens kept in darkness.

5. The dark-adapting hormone was separated from the light-adapting one by
filter paper electrophoresis at pH 9.0.

LITERATURE CITED

ALLEE, W. C., 1923. The distribution of common littoral invertebrates of the Woods Hole
Region. Studies in Marine Ecology I. Biol. Bull., 44: 167-191.

Brown, F. A., Jr. M. N. Hines anp M. FINGERMAN, 1952. Hormonal regulation of the
distal retinal pigment of Palaemonetes. Biol. Bull., 102: 212-225.

FINGERMAN, M., M. E. Lowe aAnp B. I. SunpARARAJ, 1959. Dark-adapting and light-adapting
hormones controlling the distal retinal pigment of the prawn Palaemonetes vulgaris.
Biol. Bull., 116: 30-36.

Ho.ttuHuts, L. B., 1952. <A general revision of the Palaemonidae (Crustacea Decapoda Natantia)
of the Americas. II. The subfamily Palaemoninae. Occasional Papers; Allan Han-
cock Foundation Publ. Number 12, pp. 1-396.

KLEINHOLz, L. H., 1936. Crustacean eye-stalk hormone and retinal pigment migration. Biol.
Bull., 70: 159-184.

KLEINHOLz, L. H., anp F. G. W. Know tes, 1938. Studies in the pigmentary system of
Crustacea. III. Light-intensity and the position of the distal retinal pigment in
Leander adspersus. Biol. Bull., 75: 266-273.

PEREZ-GONZALEZz, M. D., 1957. Evidence for hormone-containing granules in the sinus glands
of the fiddler crab, Uca pugilator. Biol. Bull., 113: 426-441.

SANDEEN, M. I., anp F. A. Brown, Jr., 1952. Responses of the distal retinal pigment of
Palaemonetes to illumination. Physiol. Zool., 25: 223-230.

SEX-LINKED INHERITANCE IN ULVA

BJORN FOYN
Biological Station of the University of Oslo, Drobak, Norway

In an earlier report (Foyn, 1959) mention was made of two autosomal mutants,
Slender (S17) and long (1), both of which spontaneously arose in cultures of
Ulva mutabilis and both of which exhibit sex-controlled manifestation. Found
among the young gametophytes from a crossing between the two mutants was a small
dwarfed individual entirely different from the types hitherto known. The individual
was isolated and proved to belong to the — -sex. In order to determine whether it
was a matter of a new hereditary type or only a phenotypical abnormality, the
gametes were put to parthenogenesis. The new germ plants were all of the same
kind as the mother plant, and thus probably represented a new mutant. As de-
scribed earlier (Foyn, 1958) parthenogenesis in this species is normally connected
with doubling of the chromosome number. The parthenogenetic plants therefore
became diploid and produced (haploid) zoospores. Figure 1 shows a few stages in
their development towards gametophytes.

Ficure 1. . Ulva mutabilis. Young dw S1 individuals from zoospores.
a, 10 days old; b, 17 days old.

The gametophytes were fertile after a good month and they were then crossed
with the wild-type. The zygotes germinated immediately to germ plants of the
sporophyte generation. As Figure 2 shows, they resemble the sporophyte germ
plants after crossing between the Slender-type and the wild-type but diverge from
them by the tapered top and the weakly developed rhizoids which do not grow
out from the entire basal part, but only from a single point medially below this.
With some germ plants the basal part can also grow one or a few short branches
such as in the individual on the right of the figure, but this is rare.

The hybrid sporophytes when about 2% months old formed zoospores. The
plants had then grown to a length of 25-30 cm., flat, and with even width (ca. l
cm.) band. Zoodspores from one of the plants were transferred to a petri dish

407

408 BJORN FOYN

which was thereupon placed in darkness until the following day. The zodspores
had by then become distributed evenly over the bottom of the dish and had
affixed themselves to the glass. Inspection 10 days later found the bottom
over-grown by various types of germ plants. A precise count was taken within
a circle-shaped area of 2.3 cm. diameter, and the various types were at the same
time separated, each plant being loosened from the under-bedding with the aid
of a glass needle and transferred to a new petri dish. Altogether, within the
limited area 728 germ plants were found. Of these, 189 belonged to the wild-

1mm

Figure 2. Ulva mutabilis. Twelve-day-old sporophytes from a cross between a dw SI
gametophyte and a gametophyte of the wild-type.

type and 138 to the Slender-type. The remaining 401 consisted for the greater
part of germ plants of the same appearance as in Figure la and thus resembled
the pristine mutant plant completely. A smaller number were more simple,
built more clumpily, and were only weakly ramified (Fig. 3a). Every transition
between the two types was found, however, and an assortment into two distinct
groups had, therefore, to be abandoned. All of them were reckoned as belonging
to the mutant type. Thirty-six germ plants could not be identified; they were,
partly, of different shape and dissimilar to the types hitherto known. The plants

SEX-LINKED INHERITANCE IN ULVA 409

were accordingly carried forward to sex-maturity and the gametes were put to
parthenogenetic development. A survey of the parthenogamete plants 14 days
later showed that not one of them represented new mutants. Four belonged to
the wild-type, 21 to the Slender-type and 11 to the mutant-type. The final result
of the count was therefore:

wild-type: 193 individuals (26.51%)
mutant-type: 376 individuals (51.65%)
Slender-type: 159 individuals (21.84%)

It is therefore a matter of a crossing with two independently segregating fac-
tors of which the one is SZ. The other has been designated dw. Half of the
376 germ plants of the mutant type must have both the chromosomes with S7
and the chromosome with dw the other half, only the chromosome with dw.

Three dishes with altogether 220 mutant germ plants were placed under the
source of light and, gradually, as the plants grew, the largest were isolated,
carried forward to sex-maturity, and crossed with the wild-type. It soon became

V/s mm 14mm

Ficure 3. Ulva mutabilis. Young dw individuals from zoospores. a, 12 days old;
b, 19 days old; c, 30 days old.

evident that dw SJ individuals grow much more rapidly and, therefore, become
fertile earlier than dw-individuals. The first crossings all resulted in sporophytes
of the same kind as the mother sporophyte (Fig. 2). It was only after 84 plants
had been tested that the crossings between the three successive individuals and
the wild-type gave another result, namely, sporophytes of the wild-type. Upon
the formation of zoospores eleven weeks later, they segregated into the mutant
and the wild-type in the ratio 1:1. A precise count was not taken as already at
the first glance into the microscope it was evident that the two types were repre-
sented equally strongly.

For each crossing undertaken, the gametes of the mutant partner were put
to parthenogenesis. In the first 84 cases mentioned the gametes developed, as
expected, mainly to germ plants of the ramified type, but also a few developed to
individuals of the simple type. After the three crossings following, however, the
ratio was reversed: most of the parthenogamete plants became simple whereas
only a few became considerably ramified. Even if weak or no ramification
accordingly must be said to be a typical character of the pure dw germ plant,

410 BJORN FOYN

it is no certain distinguishing feature. One of the three above-mentioned dishes
having mutant germ plants contained only individuals of the simple type, altogether
20; 13 of them later yielded gametes, 10 proved to be pure dw plants and the
remaining three, dw SJ individuals. The two other dishes each contained 100
germ plants, chosen casually, without regard to type or size. Of these, altogether
197 have been tested in crossing with the wild-type with the following result: 118
were dw S71 individuals, 79, dw individuals.

It is apparent from Figures 3c and 4 that the dw plants also later become
considerably ramified, frequently much more so than dw S1 plants. The fertile dw
individual looks like a rosette of young wild-type plants grown together. The
thallus is dark green, often folded, and gives a stunted impression. The two
layers of cells are either grown together throughout or are separated from each
other along the border of the thallus. The thickness of the layers is greater than

Figure 4. Ulva mutabilis. A full grown and fertile dz
sporophyte. Three months old.

in the wild-type. The rhizoids are weakly developed or are not at all distinguish-
able. It is because of this dwarf-like appearance that the mutant factor has been
termed dw.

The fertile dw S71 plants much resemble the dw plants. Nor do they become
larger than these but the two layers of cells are often free throughout and, partially,
distend strongly from each other. Most frequently, they have the usual green
colour.

More interesting than these results, however, was that all the 210 mutant
individuals (both dw and dw S1 individuals) which had been carried forward to
sex maturity proved to belong to the — -sex. Investigation was therefore made
as to how the 193 individuals of the wild-type and the 159 Slender individuals
behaved in this respect. Where the latter are concerned, it had already been

SEX-LINKED INHERITANCE IN ULVA 411

observed during the counting that most of them were ramified which, with this
mutant, is a quite certain sign that they are + -individuals (see below). Only
61 were unramified and these were therefore isolated and carried forward to sex
maturity. All of them revealed themselves as +-plants. Of the wild-type,
100 individuals were tested with regard to sex, all with the same result, namely,
that they belonged to the +-sex. Also in different later crossings which, in
another connection, have been performed with dw-plants, individuals with the dw
factor have proved to be of the —-sex. The factor dw must therefore be lo-
cated in the same chromosome as the factor(s) for the — -sex.

It has not been possible to establish cases of crossing-over between dw and
the sex-factor(s). The crossing best suitable for the discovery of such cases is
that between dw SI and dw* S1 individuals. The crossing can naturally only
result in the same types of gametophytes as the parents, and, apart from eventual
cross-overs, the dw S1 plants will belong to the — -sex and the dw* S17 plants,
the + -sex. As described earlier (Foyn, 1959) the germ plants of the — -sex
of the Slender race are always long, slim and unramified, whereas the germ plants
of the opposite sex can either be like the — -individuals or be small and con-
siderably ramified. This variation rests upon hitherto unknown environmental
conditions, as + -zoOspores from one and the same sporophyte after one swarm-
ing can develop to germ plants, all of which are small and ramified, and, after
another, to both ramified and unramified, or to unramified germ plants only. In
the cases where the first possibility is realized from the + -zoospores of dw dut
S1 Si-sporophyte, naturally a — -individual of the Slender-type will immediately
become evident. Of the 14 petri dishes with their bottom wholly covered by
haploid germ plants from two dw dw* S1 S1-sporophytes, three contained ex-
clusively small and ramified germ plants. In the one dish the germ plants originated
from the one, and, in the other two, from the second sporophyte. In two of the
dishes there were more than 4000 individuals, in the third, more than 5000 indi-
viduals ; thus, from among more than 6500 dw* S1-individuals, not a single cross-
over was observed.

The author would like to express his thanks to Miss Ingeborg Gjoen for the
execution of the drawings, and to Mr. John C. Aird for the translation of the
manuscript.

SUMMARY

A recessive sex-linked mutant is described that arose spontaneously in cul-
tures of Ulva mutabilis. It has not been possible to establish cases of crossing-
over between the sex chromosomes.

LITERATURE CITED

Foyn, B., 1958. Uber die Sexualitat und den Generationswechsel von Ulva mutabilis (N.S.)
Arch. f. Protistenkund., 102: 473-480.

Foyn, B., 1959. Geschlechtskontrollierte Vererbung bei der marinen Griinlage Ulva mutabilis.
Arch. f. Protistenkund., 104: 236-253.

INFLUENCE OF CARBONIC ANHYDRASE INHIBITORS ON SHELL
GROWTH OF A FRESH-WATER SNAIL, PHYSA
HETEROSTROPEA *

JOHN A. FREEMAN
Department of Biology, Winthrop College, Rock Hill, S. C.

The enzyme carbonic anhydrase catalyses reversibly the reaction
COs + ‘EEO —— H.COs

(Berliner and Orloff, 1956) and has been shown to accelerate both deposition
and dissolution of calcium carbonate in vitro (Meldrum and Roughton, 1933).
In view of its possible implication in shell formation, the mantle tissues of molluscs
have been examined and the enzyme demonstrated in several of them (Florkin
and de Marchin, 1941; Maetz, 1946; Freeman and Wilbur, 1948; Wilbur and
Anderson, 1950; Stolkowski, 1951; Tsujii and Machida, 1953; Ikinago, 1954;
Kawai, 1954a, 1954b; Larraneta and Ponz, 1954; Clark, 1957). However,
among 20 species examined, Freeman and Wilbur (1948) found little or no
carbonic anhydrase activity in mantle tissue in two, a pelecypod and a gastropod,
both shell-formers, and Stolkowski (1951) found none in the mantles of Ostrea
edulis and several other shell-forming molluscs. Such results suggest that at least
in some species the enzyme is not essential for shell formation.

Stolkowski (1951) apparently first used a carbonic anhydrase inhibitor,
benzenesulfonamide, in studies of the role of the enzyme in molluscs, observing
decreased shell regeneration in Helix aspersa in the presence of the drug. He
concluded that the enzyme is important in determining the crystalline form in
which calcium carbonate appears in the shell. Wilbur and Jodrey (1955) found
that two carbonic anhydrase inhibitors, 2-benzothiazolesulfonamide and Diamox
(2-acetylamino-1,3,4-thiadiazole-5-sulfonamide), markedly decrease the rate of de-
posit of Ca*® in the shell of Crassostrea virginica. Abolins-Krogis (1958) in a
study of shell regeneration in Helix pomatia concludes on the basis of as yet un-
published data on carbonic anhydrase inhibition that (p. 36) the enzyme “is
necessary for the formation of calcium carbonate crystals in the regenerating
shell.” These studies suggest a role of carbonic anhydrase in shell formation
but are inconclusive in establishing it. The two inhibitors used by Wilbur and
Jodrey, 2-benzothiazolesulfonamide and Diamox, exhibited quite different extents
of inhibition of Ca*® deposition in the intact animal, only Diamox affecting
deposition in a manner consistent with the hypothesis that the results were due
to carbonic anhydrase inhibition. Even though in the mantle-shell preparation
used in part of the studies the two drugs had effects consistent with the idea that
the enzyme is important in determining the rate of calcium deposit, the side effects
of one of the drugs suggest caution in interpretation of the data. Further, the

1 This study has been supported in part by a grant from the Southern Fellowships Fund.
412

CARBON ANHYDRASE AND SNAIL GROWTH 413

data are consistent with another interpretation, that carbonic anhydrase influences
the exchange of calcium between the shell and the extrapallial fluid. This is an
exchange which may not include the reactions which serve as a pathway for shell
formation.

The study of Abolins-Krogis (1958) indicates that calcium carbonate crystals
did not develop in the presence of a carbonic anhydrase inhibitor. Since the
reaction catalyzed by the enzyme occurs at a very appreciable rate, proceeding 90
per cent to equilibrium in 200 seconds (Berliner and Orloff, 1956), it would be
expected that during inhibition of the enzyme deposition of calcium carbonate
would occur at a decreased rate but would not cease.

One method of resolving the dilemma would be to study the growth of shell
in the presence of a number of carbonic anhydrase inhibitors, for if a series of
side effects were operative among the drugs in decreasing shell development
these effects should show different extents or at least statistically modify in dif-
ferent ways the pattern for the growth of the populations treated with different
drugs. On the other hand, if the effect on shell development is statistically
similar in the case of several known carbonic anhydrase inhibitors this would
strengthen the evidence for a direct role of the enzyme in shell development. In
the present study the hypothesis that carbonic anhydrase is important in molluscan
shell development has been tested by a study of the influence of five carbonic
anhydrase inhibitors on the growth of the shell of the pond snail, Physa hetero-
stropha.?

METHODS

Adult snails of a local population were maintained as stock cultures on lettuce
with oyster shell chips in the bottom of aquaria. Young snails not over a month
old were removed from the stock aquaria as required and placed in 4-inch finger-
bowls containing 100 ml. water which had been equilibrated with precipitated
calcium carbonate for at least a week. Three or four snails from about 2 to 5 mm.
in length, which were individually distinguished in the course of measurements,
were placed in each fingerbowl along with lettuce. Twenty-four hours later each
snail was measured, using a micrometer calibrated to 0.01 mm., and the water
was replaced. At this time appropriate volumes of solutions of the drugs (made
up in the CaCOs;-saturated water) were substituted for a part of the water in
dishes containing the experimental animals. Subsequently the water or inhibitor
solution was replaced daily and lettuce was added as needed. (For information
on feeding, see Table I.) The snails were measured on alternate days for ten
days. In the few cases in which snails were lost, died or were accidentally killed,
the measurements were included if data were available for six or more days of
growth. (Such losses were: 8 snails died during the major series of observations,
6 were killed accidentally during handling, and there were 3 losses which were
unaccounted for among 672 snails.)

The experiment was carried out in the room in which the snails were reared,
at a temperature which in most experiments approximated 26° C. (In a pre-
liminary study the temperature approximated 21°, as indicated in appropriate
places in the tables. )

2 The author appreciates the identification of the snail by Dr. H. Van der Schalie of the
University of Michigan.

414 JOHN A. FREEMAN
TABLE [|
Growth of Physa heterostropha fed in three ways. Data for snails having median

lengths during growth periods of 3.00-5.99 mm.

Growth (mm./day)

Feeding No. snails
Median Mean Range
I 12 0.11 0.10 + 0.015 0.05-0.13
II 55 0.25 0.27 + 0.010 0.07-0.38
III 30 0.40 0.41 + 0.014 0.28-0.49

Observations were at ca. 26°. Fresh lettuce was available to all snails continuously, the
amount of decaying lettuce varying. In feeding method I, each morning any piece of lettuce
showing signs of decay was removed and replaced with fresh lettuce. In method II, each piece
of lettuce was left until all parts showed signs of decay. Then it was removed and replaced with
fresh lettuce. In method III, decaying lettuce was available at all times and decaying pieces
were removed when only plant fibers remained.

The inhibitors used were Diamox (2A),? 2-benzothiazolesulfanilamide (2B )*
(Miller, Dessert and Roblin, 1950), benzenesulfonamide (BS), p-toluenesulfonamide
(TS) (Krebs, 1948) and sulfanilamide (SA) (Mann and Keilin, 1940).

RESULTS

Examination of preliminary data in graphs indicated approximately uniform
growth rates for snails having median lengths ranging from 2 to 6 mm. during
growth periods and lower rates for those having lengths outside this range. More
critical examination of the data and subsequently of data for all control groups,
using the Kruskal-Wallis test (Kruskal and Wallis, 1952), indicates the range
of lengths having statistically uniform growth rates is more restricted, from 3.00
to 5.99 mm. when data are examined in classes having 0.50 mm. ranges. All data
for controls have been examined and, with the exception of a single class among
those subjected to feeding method I, show no statistically significant variations
in growth rates among classes having median lengths of 3.00 to 5.99 mm. within
any method of feeding at a given temperature. As determined by the Kruskal-
Wallis test, probability (p) for controls ranges from 0.3 to 0.8, with the exception
indicated. Statistically significant variations among classes within experiments
were obtained in each case in which a broader range of median lengths was used.
Inclusion of all data in calculations would have made no difference in the over-all
picture, for among the control experiments the mean growth rates of snails sub-
jected to different feeding regimes would have been 0.10, 0.24 and 0.37 instead
of 0.10, 0.25 and 0.41, and similar differences were found for experimental popula-
tions. The difference is in part determined by the proportion of measurement
made on snails outside the range 3.00 to 5.99 mm.

Preliminary experiments indicated that the manner of feeding greatly in-
fluenced the rate of growth. Although fresh lettuce supported slow growth, the
most rapid growth was obtained only if decaying lettuce was available to the

3 Diamox and 2-benzothiazolesulfanilamide were supplied by Dr. Emmanuel Waletzky of
American Cyanamid Co.

CARBON ANHYDRASE AND SNAIL GROWTH 415

TABLE II

Growth of Physa heterostropha in 2-benzothiazolesulfonamide. (Data for periods
in which median lengths were 3.00-5.99 mm.)

Growth (mm./day)

Feeding 6G} : roe “ay No. snails
Median Mean Range
I Zi 0.0 12 0.11 0.10 + 0.015 0.05-0.13
0.1 12, 0.12 0.11 + 0.020 0.08-0.19
0.5 1 0.10 0.12 + 0.018 0.01-0.28
Ops 12 0.01 0.02 + 0.005 0.00-—0.08
Il 26 0.0 “85 0.25 0.27 + 0.010 0.07-0.38
0.1 17 0.21 0.22 + 0.013 0.12—0.30
0.2 15 0.17 0.18 + 0.013 0.06-0.26
0.5 49 0.05 0.07 + 0.005 0.00-—0.14
1.0 14 0.00 0.00 + 0.001 0.00-0.02

snails at all times. The rate of growth of snails used in control experiments at 26°
has been compared for three feeding methods with the results indicated in Table I.

In all experiments subsequently presented in this report feeding was by method
III which gave most rapid growth, unless otherwise indicated, and data are for
growth periods during which the median lengths of the snails were from 3.00 to
5.99 mm.

The effect of 2B on growth was quite different from that of the other four
drugs and so is presented separately in Table II. Additional information on the
effects of 2B were obtained in preliminary experiments conducted at ca. 21°. In
these experiments, of 27 snails exposed to concentrations of 2B from 2.0 to 10.0

TABLE III

Comparison of growth rates of Physa heterostropha treated with concentrations of 2A, T'S,
BS and SA showing maximum inhibition of growth with untreated controls. Data
for periods in which median lengths of snails were within the range

3.00-5.99 mm.
Feeding | THPR* | Drug | Prageane |o.snais| Mean gipwh | Feresnt | pee

III ZT none 0.0 16 0.23 + 0.014

2A 0.5-15.0 25 0.15 + 0.009 35 <0.00003
II 26 none 0.0 55 0.27 + 0.010

2A 2.0-5.0 65 0.15 + 0.007 44 0.0013
III 26 none 0.0 30 0.41 + 0.014

BS 50 15 0.26 + 0.014 37

A 50 16 0.26 + 0.016 Bi <0.00003

SA 10.0—50 30 0.28 + 0.015 32

* Temperatures varied within approximately two degree ranges.
** Values for p have been determined using the Mann-Whitney U test for comparing the data
for control and experimental groups (Mann and Whitney, 1947).

416 JOHN A. FREEMAN

mg. per liter, 24 died within two days and the remaining three within four days.
At concentrations of 0.5 and 1.0 mg. per liter, ca. 50% mortality occurred within
four days. One of six died between the tenth and twelfth day in 0.25 PPM 2B and
none of 15 maintained as controls or in 0.05 PPM 2B died within 12 days. Mor-
tality was lower at 26°, 4 of 87 snails dying within 10 days in 2B solutions of
concentrations 0.5 and 1.0 mg. per liter. Among 55 controls and 32 snails in 0.1
or 0.2 PPM 2B, none died within 10 days.

The four remaining drugs when used with snails growing at the more rapid
rates (feeding methods II and III) showed maximum inhibition of growth to
comparable extents. The results of these experiments are presented in Table III.
The data for concentrations showing maximum inhibition have been examined in
classes (as indicated above for control experiments) using the Kruskal-Wallis
test and found to be statistically homogeneous (p ranges from 0.1 to 0.5). Data
for BS, TS and SA obtained in simultaneous observations have been similarly
compared as a group with their controls. The value of p thus obtained was
< 0.00003. :

A series of experiments using Diamox was run to compare inhibition of growth
of slowly growing snails with that of rapidly growing ones. Mean growth rates

TABLE IV

Growth rates of snails subjected to various concentrations of Diamox and fed for slow
growth ( Feeding I) at 26° C.

Conc. Diamox 0.0 0.1 0.2 0.3 0.4 0.5 1.0 125
Mean growth rate (mm./day) 0.10 0.12 0.10 0.07 0.10 0.08 0.10 0.10
S.E. of mean 0.015 | 0.019 | 0.016 | 0.013 | 0.016 | 0.011 | 0.014 | 0.016

observed are shown in Table IV. When these control and experimental groups
are compared, no statistically significant differences between the growth rates of
the groups of snails are found (p = 0.3, using the Kruskal-Wallis test).

DISCUSSION

The effect of 2B on the snail Physa heterostropha is quite different from that
of the other drugs studied and is of a nature which indicates its action on the
snail is not simply one of inhibition of carbonic anhydrase. This result agrees
well with the conclusion of Wilbur and Jodrey (1955) that this drug has an
effect on calcium deposition in the oyster that is independent of its inhibition of
carbonic anhydrase and in their experiments was shown to be independent of the
mantle.

The results of experiments using Diamox, BS, TS, and SA with snails fed for
rapid growth are consistent with the hypothesis that rapid shell growth in this
mollusc depends upon the activity of carbonic anhydrase and it is unlikely that
such congruence of effects would occur if the four drugs were acting on different
enzyme systems. At low growth rates induced by feeding method I, Diamox did
not further decrease growth rates, indicating that under at least certain circum-
stances of slow growth carbonic anhydrase activity is insignificant in shell growth

CARBON ANHYDRASE AND SNAIL GROWTH 417

of the snail. Such a result does not, however, preclude the possibility that the
enzyme may have increased importance in determining the rate of growth under
other circumstances limiting shell development, as low temperatures.

The results here reported are entirely consistent with the hypothesis that under
certain circumstances carbonic anhydrase is essential for rapid shell development
in the snail used but that in some conditions the enzyme is not necessary for shell
growth.

SUMMARY

1. The rate of growth of Physa heterostropha may be controlled in the labora-
tory by using different modes of feeding.

2. Under the conditions of the experiments, growth rates for each feeding
method, with the possible exception of that giving slowest growth, are constant over
the range 3.00 to 5.99 mm. median shell lengths during two-day growth periods
but are slower at both higher and lower lengths.

3. The drug 2-benzothiazolesulfonamide completely inhibits growth of the snail
at concentrations about 1 mg. per liter. At 21° C. it is lethal at concentrations
above this but appears to be somewhat less toxic at 26° C. Its effect is not due
to the action of the drug as a carbonic anhydrase inhibitor but to an unknown side
effect.

4. Diamox, benzenesulfanilamide, p-toluenesulfonamide and sulfanilamide, at
the concentrations exhibiting maximum effect on growth, exhibit comparable de-
grees of inhibition of shell growth in snails fed for rapid growth.

5. Diamox at concentrations up to 1.5 mg. per liter does not decrease the
growth rate of snails induced to grow slowly by the feeding method employed
though this concentration inhibited growth to a significant degree in rapidly grow-
ing snails.

6. The results are consistent with the hypothesis that carbonic anhydrase is
essential in rapid shell development of some molluscs but may be insignificant in
slow growth induced by poor feeding conditions, and afford additional evidence
that the enzyme is significant in mollusc shell formation.

LITERATURE CITED

Axo.tns-Krocis, A., 1958. The morphological and chemical characteristics of organic crystals
in the regenerating shell of Helix pomatia L. Acta Zool., 39: 19-38.

Berttner, R. W., anv J. Ortorr, 1956. Carbonic anhydrase inhibitors. Pharmacol. Rev.,
8: 137-174.

CrarKk, A. M., 1957. The distribution of carbonic anhydrase in the earthworm and _ snail.
Australian J. Sci., 19: 205-206.

Frorkin, M., ann P. pe Marcuin, 1941. Distribution de l’anhydrase dans les tissus de
Yanodante. Arch. Internat. Physiol., 51: 130-132.

FREEMAN, J. A., AnD K. M. Wrirzur, 1948. Carbonic anhydrase in molluscs. Biol. Bull.,
94: 55-59.

Ixtnaco, K., 1954. The metabolism of the pearl oyster, Pinctada martensu—I. On the
carbonic anhydrase. Bull. Japanese Soc. Fisheries, 19: 925-928. (In Japanese with
English summary.

Kawal, K., 1954a. Carbonic anhydrase in pearl oyster. I. Distribution and some properties
of the enzyme. Mem. Coll. Sci., Univ. Kyoto., Ser. B, 21: 39-44.

Kawal, K., 1954b. Carbonic anhydrase in the fresh-water mussel (Hyriopsis schlegeli)
and physiological significance of the enzyme. Japanese J. Mal. 18: 39-46. (In
Japanese with English summary.)

418 JOHN A. FREEMAN

Kress, H. A., 1948. Inhibition of carbonic anhydrase by sulphonamides. Biochem. J., 42:
525-528.

KruSKAL, W. H., ano W. A. WAL is, 1952. Use of ranks in one-criterion variance analysis.
J. Amer. Statist. Assc., 47: 583-621.

LarrANETA, M. G., AND F. Ponz, 1954. Alkaline phosphatase and carbonic anhydrase in some
gastropods. Rev. Espan.: Fisiol., 10: 225-234..

Maetz, J., 1946. L’Activité anhydrasique de quelques tissus d’Invertébrés. Bull. Inst.
Oceanogr., 43: 1-20.

Mann, H. B., anp D. R. Wuitney, 1947. On a test of whether one of two random variables
is stochastically larger than the other. Ann. Math. Statist., 18: 50-60.

Mann, T., AND D. Ketmtn, 1940. Sulphanilamide as a specific inhibitor of carbonic anhydrase.
Nature, 146: 164-165.

Metprum, N. U., ano F. J. W. Roucuton, 1933. Carbonic anhydrase. Its preparation and
properties. J. Physiol., 80: 113-170.

Mitter, W. H., A. A. DeEsserRT AND R. O. Rosin, Jr., 1950. Heterocyclic siifonamtgee as
carbonic anhydrase inhibitors. J. Amer. Chen: Soc., 72: 4894-4896.

SToLKowSsKI, J., 1951. Essai sur le déterminisme des fares minérologiques du calcaire ches
les étres vivants (calcaires coquilliers). Ann. Inst. Oceanogr., 26: 1-113.

Tsujyu, T., AnD A. Macuipa, 1953. Carbonic anhydrase and the formation of pearls. Kagaku,
23: 148.

Wirpur, K. M., ann N. G. AnpErson, 1950. Carbonic anhydrase and growth in oyster and
Busycon. Biol. Bull., 98: 19-24.

Witpur, K. M., anno L. H. Joprey, 1955. Studies on shell formation. V. The inhibition of
shell formation by carbonic anhydrase inhibitors. Bzol. Bull., 108: 359-365.

Peet yolOLOGY OF SKELETON FORMATION IN CORALS. III.
CALCIFICATION RATE AS A FUNCTION OF COLONY WEIGHT
AND TOTAL NITROGEN CONTENT IN THE REEF CORAL
MANICINA AREOLATA (LINNAEUS)

THOMAS F. GOREAU! AND NORA I. GOREAU

Physiology Department, University College of the West Indies and
The New York Zoological Society

This paper deals with an experimental study of the relationship between
colony size, total nitrogen content, surface area and rate of calcification in the
reef-building coral Manicina areolata (Linnaeus).

Tamura and Hada (1932) stated that coral growth ceases after certain colony
sizes are attained. The Stephensons (1933) reported that the percentage annual
increase in surface area and diameter was much greater in small colonies than in
large ones. No attempt was made by these authors to determine whether the
faster growth rate of younger corals was due to some inherent physiological factor,
or whether it was due to the increasing disproportion of skeletal mineral matter,
e.g. volume, in relation to the surface area occupied by the polypal mass, as corals
get larger and older. Abe (1937) found that freshly settled larvae of Fungia
actiniformis grew about three times faster between the first and fifth than from
the sixth to the sixteenth days. Motoda (1940) believed that the decrease in
the growth rate of the reef-building coral Goniastrea aspera with increase in
colony diameter was the result of physiological senescence, and the progressive
restriction of surface area.

To investigate the relationship of colony size on total nitrogen content and
on the calcification rate in a reef coral, we chose Manicina areolata as the experi-
mental animal. This species is plentiful on nearby reefs. Colonies of all sizes
were obtained in numbers and even the largest were small enough to fit con-
veniently into our experimental vessels. Manicina areolata is found on the coral
sand, shingle and grass bottom environments which are associated with the reef
tract south of the town of Port Royal, Jamaica. This species is unique among
West Indian reef-building Scleractinia in that it spends only a relatively short
initial part of its life cycle attached by a stalk to small bits of shingle, shells or
other corals. On reaching weights of approximately ten grams, the colonies
usually break free, probably by accident, and fall unattached to the bottom where
they continue to grow and proliferate.

PROCEDURE

Colonies of Manicina areolata, weighing between 0.047 and 148.5 grams wet,
were collected on 11 May 1959 from the lagoon of Drunkenman’s Cay, without

1 Mailing address: University College of the West Indies, Mona, Jamaica, W.I.
419

420 THOMAS F. GOREAU AND NORA I. GOREAU

exposing them to contact with air. These were taken to the laboratory and kept
for two days in large, well illuminated tanks provided with fresh circulating
sea water to acclimatise them to their new environment. Colonies showing obvi-
ous damage, infestations of Pyrgoma, worms or other boring animals were re-
jected. To avoid injury, those of the smaller colonies which were still growing on
pieces of rock or shell were not removed from their attachments.

After acclimatisation, the colonies were transferred, still under water, into
large polyethylene vessels which were calibrated to contain three and a half liters.
To this volume of unfiltered sea water was added an amount of buffered Ca*®CL,?
sufficient to give a specific activity of about 10* counts per minute per milligram
of calcium in the aquatic medium. Illumination was provided by a twin bank of
40-watt daylight fluorescent tubes located about 12 inches above the water level.
The water was stirred and aerated by a fine stream of wet air forced through
a sintered glass filter stick which served as airstone. The pH of the medium
increased from 8.20 to 8.34 during the experimental runs, and the water tem-
perature fluctuated between 29.3° C. and 27.5° C. The experiments were allowed
to run for 50 hours, water samples being taken after one, 24 and 48 hours for the
measurement of the specific activity.

The runs were terminated by transferring the corals from the Ca*®-enriched
water into fresh “cold” running sea water for two hours to allow washing out of
all activity not deposited into the skeleton. The specimens were killed by putting
them into a solution of slightly alkaline 4% formaldehyde. Later, the corals were
placed on filter paper to remove the adherent fixative. Algae and other en-
crusting organisms were carefully scraped away from the non-living base of each
colony, and the length, width and weight were recorded for individual specimens.
The colonies were then dissolved in separate flasks containing concentrated HCl,
the volume of which was adjusted to the weight of the coral. After solution
of the mineral matter the tissue was homogenised with a Potter grinder, and the
final volume of the suspension carefully determined. The total nitrogen contained
in each colony was calculated from replicate 3-ml. aliquots treated according to
the Kjeldahl method.

The weight of calcium laid down during the experiment by each specimen
was determined from the amount of calcium*® incorporated and the mean specific
activity of the water in which the corals were run, according to the method de-
scribed in a previous paper of this series (Goreau, 1959a). The final results
were expressed in terms of weight of calcium deposited per gram of wet weight,
e.g. mg. Ca gm.-1, or as the amount of calcium deposited per milligram of nitrogen,
e.g. mg. Ca mg. N-.. The latter will be called the “specific calcification rate,”
and the use of the term “growth” in this paper is taken to mean increase in weight
due to calcium carbonate deposition.

Besides using normal colonies of M. areolata containing large numbers of
zooxanthellae, experiments were run concurrently, and under identical conditions,
on a series of five colonies which had previously been grown in darkness for three
months to cause complete loss of the zooxanthellae and boring algae. The coenosare
of these specimens was colourless and quite translucent in the expanded condition.

2 Supplied as Ca-45-P-2 by the Oak Ridge National Laboratory.

SIZE AND GROWTH. IN A REEF CORAL 421

100

o.

LEGEND

@ COLONIES WITH ZOOXANTHELLAE
© COLONIES WITHOUT ZOOXANTHELLAE

TOTAL NITROGEN PER COLONY, IN MILLIGRAMMES (T)

oO!
Ol ! ' 10 100

TOTAL WET WEIGHT OF COLONIES, IN GRAMS (y)

Ficure 1. The relation between wet weight and total nitrogen content in colonies of the
reef coral Manicina areolata. The solid circles represent colonies with zooxanthellae, the
open circles are from colonies lacking zooxanthellae. The regression line was calculated by
the method of least squares using the data from normal colonies only. The slope of the line
is very close to unity, thus indicating that the volume, as measured by the weight, and the
surface area, as measured by the total nitrogen, increase at the same rate. The significance
of this is explained in the text. The position of the open circles near the regression line
indicates that in comparison with normal colonies of the same weight, the nitrogen content
of the bleached colonies is not significantly reduced by the absence of both zooxanthellae and
boring algae.

In one colony a faint green discolouration was seen in the skeleton, indicating
that some boring algae were still present although zooxanthellae were absent.

Dead colonies of M. areolata of different sizes, from which all tissues had
been removed by maceration, were run in each series to test the amount of
equilibrium exchange of calcium between the bare skeleton and the aquatic medium.
No corrections were necessary for exchange since this occurred at an extremely
low level in comparison with active deposition of calcium by living colonies.
There was also no correlation between equilibrium exchange rates and the weight
of the colonies used.

422 THOMAS F. GOREAU AND NORA I. GOREAU

RESULTS AND OBSERVATIONS

We found that the total nitrogen content of specimens of Manicina areolata
was directly related to their wet weight over the range of sizes investigated.
This is shown in Figure 1 where the wet weight of the corals is plotted against
their total nitrogen content. Logarithmic scales are used because of the large
numerical range of the values. The nitrogen content of the smallest colonies tested

LEGEND

—@®— COLONIES WITH ZOOXANTHELLAE
---Q--- COLONIES WITHOUT ZOOXANTHELLAE

MILLIGRAMS CALCIUM DEPOSITED PER MILLIGRAM NITROGEN (C)

TOTAL NITROGEN PER COLONY, IN MILLIGRAMS (N)

Figure 2. Graph showing the relation between the specific calcification rate and the total
nitrogen in colonies of Manicina areolata. Corals with zooxanthellae are represented by
solid circles, the bleached corals by open circles. The slope of the solid regression line was
calculated by the method of least squares whereas the lower broken line was fitted by eye.
These curves show that calcium is deposited at a progressively slower rate as the size of the
colonies increases. The specific calcification rate of corals with zooxanthellae was about 30
times higher than that of bleached colonies of similar size.

was too low to determine by our Kjeldahl method; hence data from these were
omitted from Figures 1 and 2.

The position of the points in Figure 1 indicates a good linear correlation be-
tween the wet weight V and the total nitrogen T. The regression line was fitted
by the method of least squares and yields the equation,

V=0.51 T1088 (1)

SIZE AND GROWTH IN A REEF CORAL 423

for the relation between Y and 7. The power of T is very nearly unity, which
indicates that the ratio of wet weight to the total nitrogen is more or less constant
over a large range of colony sizes. This means that, in individuals of M. areolata
weighing between 0.1 and 148.5 grams, the amount of living tissue remains ap-
proximately the same in proportion to the total mass of skeletal calcium carbonate.

In colonies lacking the zooxanthellae, the relation between the weight of the
individual specimens and their total nitrogen content does not appear to differ
significantly from that of normal colonies of similar weight, as shown by the data
plotted in Figure 1. Our results for M. areolata conflict with the assertion by
Odum and Odum (1955) that the average living reef coral colony contains three
times as much plant matter as animal matter. Bleached Manicina areolata, which
lacked both zooxanthellae and boring algae, did not show the expected decrease
in the protein nitrogen to weight ratio as compared with normal colonies of the
same size. If these corals were to contain as much plant matter as stated by the
Odums, then the nitrogen content of the bleached and zooxanthella-less corals
should have been much less, size for size, than that of normal colonies. The fact
that it was nearly the same would imply that the aggregate plant biomass associ-
ated with WM. areolata is very much less than that found by the Odums for other
species of corals.

With increasing size, we observed a striking fall in the calcification rate. The
smallest colony tested, weighing 0.047 gram wet, deposited 92 mg. Ca gm. coral
hr.’ whereas the largest, weighing 148.5 grams wet, deposited only 1.28 mg. Ca gm.
coral hr.-t. This is a 76-fold difference. The above calcification rates are given
in terms of weight rather than nitrogen because the smallest colonies used con-
tained too little nitrogen to be reliably measured by our Kjeldahl method.

On the whole, the relationship between the specific calcification rate C to the
total colony nitrogen content N, as plotted in Figure 2, is given by the equation,

N =93.4 er € (2)

whereas the relation between calcification rate per gram of coral W to the total
wet weight V, not shown in the figure, is given by the equation,

Boog es (3)

These formulae show that the growth rate, as expressed in terms of the rate of
calcium deposition, decays as an exponential function with increase in size. Similar
but less marked trends were observed in colonies lacking zooxanthellae, but in
these the calcification rate, size for size, was 30 times less than that of normal
colonies of M. areolata.

The intercept of the regression line with the x-axis in Figure 2 implies that
colonies having more than about 102 mg. total nitrogen do not grow any more.
There is no evidence for this, and the decremental relationship cited above may
hold only for colonies below a certain size range. Our results show that much
larger colonies still grow slowly, but we do not yet have enough data to determine
what kind of relationship, if any, may be applied to describe growth versus size in
such cases.

For corresponding colony sizes, the calcification rates observed in these experi-
ments on M. areolata with zooxanthellae were similar to those previously re-

424 THOMAS F. GOREAU AND NORA I. GOREAU

ported for this species under field conditions (Goreau and Goreau, 1959), but
growth rates of the zooxanthella-less specimens were lower by a factor of about
1.5. A possible reason for the discrepancy is that the bleached M. areolata used
in the present experimental series had been grown in laboratory tanks in darkness
for three months whereas those used in our previous study were found growing
in the decolourised state under a coral head in the natural environment of the
reef. Under these conditions, the growth rate of the naturally bleached colonies
may have been faster because of their better nutritional status.

DISCUSSION

In the coral Manicina areolata there is an approximate one to one relationship
between the wet weight of whole colonies and their protein nitrogen content, and
the specific calcification rate is inversely related to the weight, 1.e., size, of the
individual corals.

The linear relationship between colony mass and the amount of protein nitrogen
in specimens, with and without zooxanthellae, is extremely interesting. Except
for the boring algae, all living tissue in scleractinian corals occurs outside the sur-
face of the skeleton. Therefore, the amount of nitrogen per colony will be a
function of the total surface area covered by the living polyps. In massive hemi-
spheroidal corals, an increase in the colony size causes the weight, 7.e., mass of
skeletal calcium carbonate, to increase in proportion to the cube of the radius,
whereas the tissue mass, 1.e., the surface area, increases only in proportion to the
square of the radius. The relationship, which has been demonstrated by Kawaguti
(1941) for the massive coral Gomiastrea aspera, does not hold for M. aerolata
where the volume and surface area increase in direct proportion to each other.
Inspection of M. areolata colonies of various sizes shows why this is so. Very
small young individuals are shaped like simple cups. These grow in size and
proliferate by incomplete intratentacular budding. This results in a lengthening
of the tentacular ring and oral disk, along with an increase in the number of the
stomodea. At first, this growth produces a simple linear elongation of the colony
until the length is about twice the width. As growth continues with the budding
of new stomodea, the valley becomes sinuous and the edge of the theca folds
so that the whole colony presents a convoluted appearance which becomes in-
creasingly complex as the coral grows larger. A typical growth series is shown
in Figure 3. In this manner, the area, as measured by the protein nitrogen, and
the volume, as measured by the weight of calcium carbonate, retain a one to one
relationship to each other, as demonstrated by the slope of the regression line
in Figure 1.

Such a large increase in the surface area relative to the weight is almost cer-
tainly adaptive response to the ecological conditions in which M. areolata lives.
This coral is very common in shallow water; the larger colonies are usually found
lying loose on sand or shingle bottom, frequently in association with the marine
grass Thalassia (Goreau, 1959b). Our reef surveys have now shown that it 1s
also found in large numbers on the sand and rubble bottom of canyons in the
reef buttress zone at depths down to at least sixty feet where it is one of the few
corals which successfully and consistently colonise this particular habitat (Goreau,
unpublished field observations). It is therefore clear from its ecology that M.

SIZE AND GROWTH IN A “REEF CORAL 425

areolata is adapted for life on unstable substrates where it is constantly exposed
to the danger of burial by shifting sediments, or to being overturned by wave
turbulence and currents. The species has a very efficient mechanism for ridding
itself of sediment by means of flagellar currents and moving sheets of mucus.
It is also one of the few corals possessing a well developed righting reaction. We
have observed colonies weighing up to about 65 grams successfully right themselves
after having been turned upside down. This was accomplished by a great swelling

eo

WE
Ze
SS

Ficure 3. Relation between size and development of surface convolutions in a growth
series of the coral Manicina areolata. The dry weights of these specimens are 0.09 gm., 0.74
gm., 6.2 gm., 10.3 gm., 21.4 gm., 58.7 gm., and 117 gm., respectively. The smallest individual
shown here has a simple cup shape and only one stomodeum. Proliferation of additional
stomodea by intratentacular budding at first produces linear elongation of the theca; later this
becomes gradually more complex by formation of folds and branchings. Eventually, the
typical convoluted brain-like surface pattern seen in the largest specimen is achieved. The
continuous valleys of the specimens shown indicate that they are all clones descended from
single planulae. Colonies of WW. areolata which are formed through fusion of more than one
planula invariably have a corresponding number of separate and distinct valley systems.

of the coenosarc through imbibition of water which was then vigorously ejected
from the stomodea in the form of jets directed against the sand, thus undermining
the colony on one side and rocking it until it was gradually righted in the normal
position (Goreau, unpublished observations). It is almost certain that the
ability of this coral to cleanse itself of sediment and to right itself if overturned
must depend on its having a large surface area with respect to mass.

426 THOMAS F. GOREAU AND NORA I. GOREAU

An interesting parallel is offered by the fungiid corals of the Indo-Pacific,
which occupy an ecological niche in reefs very similar to that inhabited by M.
areolata in the Western Atlantic, and which show striking analogies in their
adaptive specialisations even though these corals belong to a different suborder
of the Scleractinia. The young stages of both are attached to the substrate by
stalks which subsequently break so that the older colonies are usually found free.
In the fungiids, the stalk continuously buds off new colonies, in Manicina it does
not. Both corals are commonly expanded in the day time; both increase their
surface areas in more or less direct proportion to their mass, the fungiids doing
this by assuming a flattened disk-like shape, Manicina by becoming more and
more convoluted as its size increases. Both corals can inhabit environments
where sedimentation and substrate instability prevent colonisation by most other
species, largely through their innate ability to remain “afloat” on top of sand and
shingle. In both, the highly sensitive youthful stages remain attached until they
are heavy enough to resist displacement and turning over by anything but the
strongest waves and currents.

Unlike the surface area, the specific calcification rate of M. areolata decreases
with an increase in colony size. Motoda (1940, quoted by Wells, 1957) studied
growth in Gomiastrea aspera and concluded that progressive restriction of the
surface area causes crowding of polyps in the center of the colony, thus slowing
the growth rate as the spherical form is approached. We do not believe that these
factors operate in the case of M. areolata since the surface area of this coral in-
creases in direct proportion to the size. All colonies used in our experiments
were descendants of single planulae as shown by their continuous valley patterns.
There was no evidence that the stomodea in the center of the colonies were more
crowded than those of the periphery.

Of all other variables which could influence the calcification rate with increasing
age, the most obvious are the zooxanthellae. These dinophycean algae occur
in the gastroderm of all hermatypic corals as intracellular commensals, and are
known to affect the metabolism and growth rates of their hosts in a number of
important ways. The presence of photosynthesising zooxanthellae constitutes a
most effective stimulus to the mechanism of calcification, causing a very large in-
crease in its rate in all species of reef corals we have tested so far (Goreau and
Goreau, 1959). This effect is probably mediated by a decrease in the HCO;
concentration at the site of calcification when the zooxanthellae are photosynthesis-
ing, 1.e., removing CO, from the system (Goreau, 1959a). As is shown by the
data given above, the calcification rate of M. areolata without zooxanthellae falls
to extremely low levels in comparison to colonies of the same size which contain
the zooxanthellae; nevertheless the growth rate of the smallest of our bleached
colonies was nearly the same as that of the largest of the normal colonies that
were tested. There is no evidence that the low calcification rate we found in
the large specimens was due to a systematic reduction in the number of zooxanthellae
with increasing age or size. On the contrary, large colonies of M. areolata are
usually a darker brownish green than the young colonies which are mostly pale
greenish brown in colour. Histological cross-sections of the oral disk show that
there are about half as many zooxanthellae in the average thickness of the gastro-
dermis of small colonies about one centimeter long than in larger colonies about
ten centimeters long. This means that the younger smaller colonies seem to have

SIZE AND GROWTH IN A REEF CORAL 427

fewer zooxanthellae per unit of weight than older larger colonies. The precise
nature of this relationship is now being investigated in our laboratory. The de-
crease in calcification rate with increasing size is perhaps not due primarily to any
variation in the number of zooxanthellae, but it occurs independently, as shown
by our data on corals without zooxanthellae, in which this effect was also ob-
served though it was much reduced in amplitude.

The mechanism by which age affects growth and other metabolic processes (cf.
Brody, 1945) is not well understood. Although the possibility is suggested by
our experimental results on M. areolata, it is still not clear whether other corals
undergo a true process of senescence if all environmental factors remain optimal.
There are probably very great individual variations in this respect among the
different species of reef-building corals, as is indicated by the enormous variations
in the size of some of them. Single colonies weighing several hundred tons are
not uncommon. Many coral colonies are clones which are descended from single
planulae. In these a progressive reduction in the speed of growth processes would
be expected with increasing age of the coral. Similar phenomena have long been
recognised, for example, in protozoa such as Paramecium and Uroleptus (Calkins,
1926) in which clonal ageing results in gradual failure of cell division.

There are almost certainly exceptions to this among the corals. For example,
our observations on the branching West Indian staghorn coral, Acropora cervi-
corms, suggest the possibility that there is no reduction in the over-all growth
rate with increasing size and/or age. In this species most of the growth takes
place in the initial few centimeters of the branch tips occupied by large pale
terminal polyps which have a far higher calcification rate than the much smaller
lateral polyps immediately behind (Goreau and Goreau, 1959). These fast grow-
ing apical polyps are often seen to differentiate apparently in a spontaneous way,
giving rise to new branches in parts of colonies which are already comparatively
old and where the calcification rate is very low. The ability of these corals to
grow new branches whenever and wherever needed must confer upon them an
enormous adaptive advantage, since this enables them to respond rapidly to damage
by waves and sediment. By using some of the lower branches as stilts to hold the
main part of the colony off the bottom, it becomes possible for A. palmata to
colonise sandy unstable habitats which would normally be inaccessible to most
other corals having a more determinate pattern of growth.

This work was supported by grant G-4017 of the National Science Founda-
tion to the New York Zoological Society, and by institutional funds from the
University College of fhe West Indies. We wish to thank Professor D. M.
Steven for putting at our disposal the facilities of the University’s Marine Station
at Port Royal for some of the experiments described above. We also record
our grateful acknowledgment to Dr. R. C. Read and Mr. D. M. Hum of the Uni-
versity College Department of Mathematics for their advice on the statistical
treatment of some of the data.

SUMMARY AND CONCLUSIONS

1. A study was made of the relation of colony weight to the total nitrogen
content and calcification rate in the reef coral Manicina areolata (Linnaeus).
Colonies weighing between 0.047 and 148.5 grams were used in our experiments.

428 THOMAS F. GOREAU AND NORA I. GOREAU

2. In Manicina areolata, the ratio of the surface area, as measured by the
total nitrogen, and the volume, as determined by the weight of the coral, remained
nearly constant over the range of colony sizes tested. This is due to the fact
that the surface of the coral becomes more and more folded as the colony grows
older and larger. In those massive hemispheroidal corals which lack convoluted
surfaces the increase in area would occur as a function of the square of the radius
whereas the volume would increase as a function of the cube of the radius. It
is suggested that the disproportionately large increase in the surface area of M.
areolata with respect to growth in size, or volume, is an adaptive specialisation
which-aids this coral to colonise sandy or otherwise unstable bottom substrates,
to clear itself more rapidly of sediment by means of flagellar currents, and to right
itself 1f overturned by waves or currents.

3. The specific calcification rate of the smallest colonies was about 76 times
greater than that observed in the biggest specimens of our series. The slopes of
the plotted curves indicate that the growth rate decays as an exponential function
of the size, 1.e. the age of the coral, which can be described by a relationship of
the type X = Ae where X = total colony size, weight or nitrogen content, and
Y =the amount of calcium deposited per unit weight or nitrogen. The growth
rate of the largest colonies was somewhat greater than expected, and there is
reason to believe that the relationship given above may hold only for colonies
weighing less than about 100 grams.

4. Similar but much less marked decreases of the specific calcification rate
with size were observed in M. areolata colonies which lacked zooxanthellae. It
is suggested that, although the zooxanthellae are powerful stimulants to the
calcification reaction, they are not as such responsible for the decay of the growth
rate with increasing size and/or age.

5. The causes for the reduction in specific calcification rates with age are not
understood, but may be due to an effect of senescence, similar in principle to the
reduction in cell division which has been found to occur in old protozoan clones.
Evidence is presented that some of the branching acroporid corals may not undergo
a slowing of growth with age since these are able to sprout new fast growing
branches even from the older parts of the colony. The survival value of this is
discussed.

LITERATURE, CITED

Ase, N., 1937. Post larval development of the coral Fungia actiniformis var. polawensis
Doederlein. Palao Trop. Biol. Sta. Stud., 1: 73-93.

Bropy, S., 1945. Bioenergetics and Growth. New York, Reinhold Publishing Company, pp.
1023. .

CaLkIns, G. N., 1926. The Biology of the Protozoa. Philadelphia, Lea and Febiger, pp. 502.

GoreaAu, T. F., 1959a. The physiology of skeleton formation in corals. I. A method for
measuring the rate of calcium deposition under different conditions. Biol. Bull., 116:
59-75.

GoreAu, T. F., 1959b. The ecology of Jamaican coral reefs. I. Species composition and
zonation. Ecology, 40: 67-90.

GorEAU, T. F., and N. I. Goreavu, 1959. The physiology of skeleton formation in corals. II.
Calcium deposition by hermatypic corals under various conditions in the reef. Bizol.
Bull., 117: 239-247.

Kawacutl, S., 1941. On the physiology of reef corals. IV. The growth of Gomniastrea aspera
measured from numerical and areal increase of calyces. Palao Trop. Biol. Sta. Stud.,

2: 309-317.

SIZE AND GROWTH.IN A REEF CORAL 429

Moropa, S., 1940. A study of growth rate in the massive reef coral Gontastrea aspera Verrill.
Palao Trop. Biol. Sta. Stud.,. 2: 1-6.

Opum, H. T., anp E. P. Onum, 1955. Trophic structure and productivity of a windward
coral reef community on Eniwetok Atoll. Ecol. Monogr., 25: 291-320.

STEPHENSON, T. A., AND A. STEPHENSON, 1933. Growth and asexual reproduction in corals.
Gt. Barrier Reef Exped. Sci. Rep., 3: 167-217.

TAmuRA, T., AND Y. Hana, 1932. The growth rate of reef building corals inhabiting the
South Sea Island. Tohoku Imp. Univ. Sci. Rep., 7 (4): 433-455.

Wetts, J. W., 1957. Annotated bibliography—Corals. Geol. Soc. American Memoir, 67
(1) :1087-1104.

THE ACTION OF VARIOUS GOITROGENS IN INHIBITING
LOCALIZATION OF RADIOIODINE IN’ TEE aHvVee@ge
AND THYMUS’ GLANDS OF LARVAL TREE TOADS

W. GARDNER LYNN? AND JAMES NORMAN DENT?

Biology Division, Oak Ridge National Laboratory,? Oak Ridge, Tenn., Department of Biology,
The Catholic University of America, Washington 17, D. C., and Department
of Biology, The University of Virginia, Charlottesville, Virginia

Because of the differential affinity of the thyroid gland for iodine, properly
adjusted doses of radioactive iodine (I'**) can produce enough radiation to destroy
the thyroid completely without seriously damaging surrounding tissues. Immer-
sion of tadpoles of Hyla versicolor at midlarval stages for 24 hours in a solution
containing 20 uc./ml. of I*** results in rapid radiation damage to the thyroid so
that no functional remnants are present after 10 days (Dent and Hunt, 1952).
Administration of certain chemical substances known to inhibit the thyroid’s ability
to concentrate or bind iodine might be expected to prevent the thyroid from re-
ceiving enough radiation to cause permanent damage. Rugh (1953) found treat-
ment with thiouracil effective in preventing or decreasing radiation damage in the
thyroids of Triturus pyrrhogaster given a single injection of 1134 (25 yc.).

Dent and Hunt (1952) observed that, in the tadpole, iodine is bound in other
organs, notably the thymus, to a lesser degree than in the thyroid. It was demon-
strated by Lynn and Dent (1957) that phenylthiourea inhibits iodine binding in
both the thyroid and the thymus.

There are several different types of thyroid inhibitors with diverse modes of
action. In the present study we have shown that some of them differ in degree
and duration of effect by testing their capacities for preventing or diminishing
radiation damage to the thyroid and thymus during administration of a large dose
ones,

MATERIALS AND METHODS

Tadpoles of Hyla versicolor collected near Oak Ridge, Tennessee, were main-
tained in the laboratory at 21° =1° C. until they reached late limb bud stages
(equivalent to stages III to V of the Taylor and Kollros (1946) series). At this
time they were divided into control and various experimental groups. Drugs to
be tested were dissolved in the spring water in which the experimental animals
were raised. The tadpoles were fed crumbled pellets of Purina Rat Chow, an
excess of food being available for ~ 8 hours each day. The culture media were

changed daily.

1 Present address: Department of Biology, The Catholic University of America, Wash-
meton 1/7, D.C:

2 Present address: Department of Biology, University of Virginia, Charlottesville, Va.
Work supported in part by A. E. C. Contract AT- (40-1) -2000.

3 Operated by Union Carbide Corporation for the U. S. Atomic Energy Commission.

430

GOITROGENS AND [131 LOCALIZATION 431

Figures 1 through 5 are photomicrographs of sections of larval specimens of Hyla versicolor.

Figure 1. Thyroid gland from control specimen with histological evidence of moderate
secretory activity.

Figure 2. Arrow points to remnant of a thyroid follicle from an animal that received
a thyroid-destroying dose of I1*1 and no goitrogen. Thyroid glands from most of the animals
treated in this fashion were completely destroyed.

Figure 3. Thymus gland from control specimen. Note cords of cells and small sinus.

Figures 4 anp 5. Thymus glands showing severe radiation damage. The one in Figure
4 has a single cyst, the one in Figure 5 a large cyst and two smaller ones.

432 W. GARDNER LYNN AND JAMES NORMAN DENT

Figures 6 through 11 are photomicrographs of contact autoradiograms made from sections
of larval specimens of Hyla versicolor treated with [131. .

Ficure 6. Thyroid region of control tadpole after a tracer dose of 1131. The spots
produced by the two lobes of the thyroid gland are fused into a single spot the size of which is
many times that of the gland (compare size of images of eyes).

Ficure 7. Thyroid protected from the histological damage of the thyroid-destroying
dose of [181 by goitrogen and then given a tracer dose of I1*1. The size of the spots indicates
that the gland is functioning at a lesser rate than glands of control animals.

Ficure 8. Thyroid protected from histological damage. No tracer dose was given.
The iodine present is residual from the thyroid-destroying dose.

Ficure 9. Thymus region of control animal given only a tracer dose of I131. The lobes
of the thymus have produced the spots on either side of the photograph. They are slightly
larger than the sections of the gland that produced them. The double spot on the left repre-
sents concentrations of [1*1 in two sinuses. The diffuse spot in the lower center represents
the heart.

GOITROGENS AND [331 LOCALIZATION 433

After a 15-day period * of goitrogen administration one series of tadpoles was
given a large (thyroid-destroying) dose of radioiodine by immersion for 24 hours
in a solution containing 20 pe./ml. of I'**. They were then passed through two
baths of spring water and maintained in spring water, with the usual daily feeding
schedule, for 10 days. These animals and some not previously exposed to radio-
iodine were given a tracer dose of radioiodine by immersion for 24 hours in a
solution of 1 pc./ml. of I’**. Ina second series of animals, the same procedures
were followed except that the tracer dose was omitted. All tadpoles were kept
in spring water for an additional day and were then fixed in Bouin’s fluid, em-
bedded, and sectioned. Contact autoradiograms were prepared after the method
of Dent and Hunt (1952) by apposing slides bearing sections of the larvae to
Eastman medium-contrast lantern slide plates for an exposure time of 8 days. All
sections were subsequently stained with Harris’s hematoxylin and Ponceau de
xylidine-orange II (Gray, 1952). Figures 1-5 are photomicrographs of sections
of larval specimens. Figures 6-11 are photomicrographs of contact autoradio-
grams made from sections of larvae treated with [**?.

RESULTS
1. Thyroid gland

(a) Effects of radioiodine treatment on normal thyroid. Tadpoles kept con-
tinuously in spring water and given only the tracer dose of I**? provided the basic
controls in this experiment. Histological observations of thyroid glands from
eight such animals indicated moderate secretory activity (Fig. 1). Large follicles
were lined with cuboidal epithelium and contained much acidophilic colloid.
Chromophobe droplets were present at the periphery of the colloid mass in most
follicles. The marked ability of these thyroids to concentrate iodine is evidenced
by an intense black spot on the autoradiographic plates over each lobe that fused
with the spot representing its companion lobe (Fig. 6).

Administration of the large dose of I*** to nine tadpoles resulted in complete
destruction of the thyroid in six of them. In the other three, tiny spots running
through two or three sections on the autoradiograms corresponded to minute thy-
roid remnants on the slides. Two of these remnants consisted of isolated colloid
masses with no surrounding cells, the other of several follicles with severely dam-
aged epithelia (Fig. 2). These tadpoles, having received no drugs, constitute
the experimental controls of the series receiving the thyroid-destroying dose of
I?**. Table I summarizes the histological findings for all the experimental animals
that received the large dose of [**?.

4In preliminary experiments, the period of treatment with phenylthiourea varied from 25

to 9 days. Within these limits the protective effect was not correlated with the period of
goitrogen administration. The standard period used in subsequent experiments was 15 days.

Ficure 10. Thymus region of animal given both thyroid-destroying and tracer doses of
1131, The large, dark spot on the right represents the right thymus; the left is not present
in this section.

Ficure 11. Thymus region of animal given only thyroid-destroying dose. Only a small
part of the right thymus is present. The left thymus has produced a large spot, indicating
considerable retention of [**?.

434 W. GARDNER LYNN AND JAMES NORMAN DENT

(b) Effects of treatment with phenylthiourea. In experiments performed to
test the degree and duration of effect of phenylthiourea three variations in pro-
cedure were followed: (1) treatment with 0.01% phenylthiourea was continued
during the 24-hour period of I'* administration, the I'*' solution being made up
in phenylthiourea solution; (2) treatment with phenylthiourea was stopped at the
time of ['*' administration, the I‘ solution being made up in spring water; (3)
treatment was stopped two days before I*** administration,

The thyroids of eight tadpoles given phenylthiourea before and during I‘!
treatment were indistinguishable histologically from those of the basic control tad-
poles, although their physiological activity was reduced in that they produced
autoradiograms about half the size of those produced by control thyroids (Fig. 7).

When phenylthiourea treatment was discontinued at the time I'*! administration
was begun, the protective value of the treatment was greatly decreased. Thyroids
of five animals handled in this manner showed extensive radiation damage with
destruction of cells, pyknosis of nuclei, and, in some regions, complete disruption

TABLE I

Degree of radiation damage in thyroid glands of tadpoles after immersion in solutions of goitrogens
and subsequent immersion in solutions containing 20 yc./ml. of I! (thyroids of controls
immersed in iodine without goitrogen treatment were destroyed)

Drugs in which animals were immersed

Time of removal from goitrogen solution
Phenylthiourea KC104 KSCN KCI1O3 KIO;
(0.01%) (0.01%) | (0.005%) | (0.005%) | (0.005%)
At end of I! administration 0 0 0 0 0
At beginning of I'*! administration +. + (0) ++ NG IoD
Two days before |**! administration + + -+ (0) +++ NT NT

Legend: 0 = No histological evidence of radiation damage.
+--+ = Evidence of extensive radiation damage.
+-+-+ = Evidence of complete or essentially complete destruction.
NT = Not tested.

of follicles. Three of these thyroids produced no autoradiographic images; small
isolated spots corresponding to less seriously damaged follicles were present in
the others.

Discontinuance of phenylthiourea treatment two days before I'** administration
resulted in complete destruction of the thyroid.

(c) Effects of treatment with perchlorate. Three sets of experiments corre-
sponding with those just described for phenylthiourea were carried out with
0.01% KCIQ,.

In contrast with phenylthiourea, KClO, completely protected the thyroid not
only when given during I'*' administration, but even when discontinued two days
earlier. No histological abnormalities were seen in thyroid sections of this series,
and autoradiograms showed dense spots representing the thyroids. Like those
from thyroids protected by phenylthiourea, the spots were somewhat smaller than
in untreated controls but indicated, nevertheless, a relatively high I"! content.

GOITROGENS AND [131 LOCALIZATION 435

Autoradiograms prepared from sections of animals that did not receive the tracer
dose of radioiodine still showed small but definite spots for the thyroid (Fig. 8),
indicating that a part of the heavy dose of radioiodine was retained in the protected
glands.

Treatment with a lower concentration of KCIO, (0.005% ) gave similar results,
indicating that these two concentrations are equally effective.

(d) Effects of treatment with thiocyanate. Although animals were treated
with both 0.01% and 0.005% KSCN, the higher concentration was so toxic that
only a few animals survived the experiment, and they were discarded. The lower
concentration prevented morphological damage to the thyroid when treatment was
continued during I'*' administration. Cessation of treatment at the time I** was
given, however, resulted in extensive damage, only a few badly disrupted remnants
of the thyroid being discernible. These remnants were sometimes represented by
tiny spots on the autoradiograms. Cessation of pretreatment two days before I**?
administration resulted in complete destruction of the thyroid.

(e) Effects of treatment with chlorate and iodate. A single series of experi-
ments was run to test the effects of treatment with 0.005% KCIO, and 0.005%
KIO, that was continued during I**t administration. Under these conditions these
substances prevented any visible histological damage to the thyroid. Autoradio-
grams showed large intense spots representing the glands.

2. Thymus gland

(a) Effects of radioiodine on the thymus of control animals. At the stage
studied, the thymus gland is made up of loosely defined cords of cells surrounding
small sinuses filled with lymphocytes and erythrocytes (Fig. 3) so that the gland
presents a relatively compact appearance in section (Speidel, 1926; Dent and
Hunt, 1952).

Administration of the large dose of I**! to previously untreated animals resulted
in extensive damage to the thymus. In most of them the thymus was represented
only by a single large vesicle with a few intact cells around the periphery (Fig. 4).
Such glands produced much more intense spots on autoradiograms than did the
normal thymus (Figs. 9 and 10). In the relatively few instances in which a
number of separate vesicles were present (Fig. 5), rather than a single large one,
the autoradiographic spots always corresponded to the vesicles. The material
in these vesicles thus had a higher I**! content than the cells of the thymus.

(b) Effects of treatment with goitrogens. For all the preceding treatments
discussed, histological and autoradiographic study was made of the thymus as well
as the thyroid. Only one treatment proved completely effective in preventing
thymus damage under the conditions of these experiments. Treatment with 0.01%
KCIO, continued during I**! administration prevented cystic change in the thymus
in all nine animals treated. Judging from the extent of damage, similar treatment
with 0.005% KCIO, or 0.005% KSCN gave partial protection, but 0.005% KCIO,,
0.005% KIO, and 0.01% phenylthiourea gave none. No evidence of protection
was seen in any tadpoles in which goitrogen treatment was discontinued before or
at the time of I*** administration.

Any large vesicles present in the damaged thymus were represented by intense
spots on autoradiograms. Because these spots were present whether or not a

436 W. GARDNER LYNN AND JAMES NORMAN DENT

tracer dose of I*** was given (Figs. 10 and 11), it is evident that they resulted
primarily from residual iodine of the original dose.

DISCUSSION

Work with mammals indicates that, with the exception of phenylthiourea, the
goitrogens used in this study interfere with the thyroid’s function by inhibiting its
ability to concentrate iodide (Wyngaarden et al., 1952). Phenylthiourea, however,
permits iodide concentration but prevents its binding to protein (Pitt-Rivers, 1950;
Roche and Michel, 1955). Under the experimental conditions of this investiga-
tion all these substances protected the thyroid against retaining sufficient I*** to
cause histological damage to the gland, provided that the goitrogen treatment was
continued during the administration of I***, whereas only KClO, was effective
when pretreatment was discontinued either two days before or immediately before
I**+ administration.

Longer periods between the administration of the large I**t dose and fixation
of the specimens might produce progressive damage since even those thyroids that
were considered to have been completely protected retained a large amount of I***
10 days after the thyroid-destroying dose was given. In mice, damage resulting
from a low dose (25 pc.) of I*** is not apparent until many months after the
iodine has been injected (Dent, Gadsden and Furth, 1955). Our experiments,
however, were directed toward testing the relative effectiveness and duration of
effect of the various goitrogenic drugs and were not concerned with devising a
method for protecting the thyroid from radiation damage. Continuation of the
goitrogen treatment for some time after exposure to I*** would presumably pro-
vide complete protection.

It is also noteworthy that, in a number of the experiments in which the thyroid
seemed completely undamaged histologically, autoradiograms nevertheless indi-
cated that the gland’s ability to take up I*** from the tracer dose was considerably
decreased. This may be interpreted as evidence of physiological damage in the
absence of histological changes, or it may indicate a residual inhibitory effect of
the drug as seen in the rat after the administration of propyl thiouracil (Durbin
Eb Obs 1957 5,

The only similar experiments with an amphibian are those of Rugh (1953),
in which adult specimens of Triturus pyrrhogaster were given a series of four intra-
peritoneal injections of a 1% suspension of thiouracil over a period of 3 weeks.
These animals and untreated controls then received an intraperitoneal injection of
25 pc. of I***. The first indications of damage to the thyroid were seen in the
controls after one month, and almost complete destruction had occurred by two
months. The specimens given thiouracil showed no damage at one month, minor
damage at two months, and clear signs of protection even at 7 months. Even
more complete protection might have been afforded, Rugh pointed out, by con-
tinuing administration of the goitrogen after I**t treatment. Although our ex-
periments were undertaken with a different aim and technique, our results are in
accord with those of Rugh.

Our observations also confirm the earlier observation of Dent and Hunt (1952)
that the amphibian thymus has a differential affinity for I**t and that cystic changes

GOITROGENS AND [231 LOCALIZATION 437

occur in this gland when sufficient I'** is administered. The amount of I**! taken
up by the thymus is apparently quite low compared with that concentrated by the
thyroid. [*** concentration in cysts of damaged thymus tissue is apparently much
higher than in normal cells of the thymus.

The protective action of goitrogenic drugs in the thymus differed from that in
the thyroid. Only one treatment (0.01% KCIO, continued during I**! administra-
tion) completely prevented cystic changes in the thymus. None of the drugs pro-
vided any protection for the thymus if their use was discontinued before ['** ad-
ministration. These goitrogens are all believed to interfere with the thyroid’s
ability to take up iodine from the circulating blood. Since KCIO, was the most
effective inhibitor of I**t uptake by both thyroid and thymus, the same mechanism
may be operative in both cases. The lesser degree of effectiveness of the drugs
for the thymus compared with the thyroid may indicate that the thymus is dam-
aged by lower radiation levels than is the thyroid.

Although protective to the thyroid, phenylthiourea proved to be completely
without protective value in the thymus. Although phenylthiourea caused a de-
crease in I*** uptake by the thymus (Lynn and Dent, 1957), the present results
indicated that this reduction is not sufficient to protect the thymus from damage
by a thyroid-destroying dose of I**?.

SUMMARY

1. Larvae of Hyla versicolor were immersed for 15 days in spring water con-
taining phenylthiourea, KCIO,, KSCN, KCIO,, or KIO, followed by a 24-hour
immersion in 20 pe./ml. of I**1, a thyroid-destroying dose.

2. When treatment with any one of these drugs was continued during the
period of I*** administration, radiation damage to the thyroid was prevented.

3. The inhibitory effect of KCIO, on the thyroid persisted even when treatment
was discontinued two days before immersion in the I™* solution.

4. The effects of phenylthiourea and KSCN were dissipated more quickly since
the thyroid showed extensive damage when treatment was stopped immediately
before I*** administration and complete destruction when it was stopped two days
before.

5. In control animals, the thyroid-destroying dose of I'** extensively damaged
the thymus gland as well.

6. Of the drugs tested, only KCIO, sufficiently inhibited the I**' uptake by the
thymus to prevent damage, and only when it was present during the period of
exposure to [**?.

LITERATURE CITED

Dent, J. N., ann E. L. Hunt, 1952. An autoradiographic study of iodine distribution in
larvae and metamorphosing specimens of anura. J. Exp. Zool., 121: 79-98.

Dent, J. N., E. L. Gapspen anp J. Furtu, 1955. On the relation between thyroid depression
and pituitary tumor induction in mice. Cancer Res., 15: 70-75.

Dursin, P. W., C. W. Astinc, M. E. Jonnston anv J. G. Hamitton, 1957. The effect of
propyl thiouracil on the recovery of rat thyroid gland from irradiation with astatine?!!.
Anat. Rec., 129: 17-37.

Gray, P., 1952. Handbook of Basic Microtechnique. Blakiston Company, Philadelphia.

Lynn, W. G., anno J. N. Dent, 1957. Phenylthiourea treatment and binding of radioactive
iodine in the tadpole. Biol. Bull., 113: 160-169.

438 W. GARDNER LYNN AND JAMES NORMAN DENT

Pirt-Rivers, R., 1950. Mode of action of antithyroid compounds.
RocHeE, J., AND R. MicHeEt, 1955.

Physiol. Rev., 35: 583-610.
RucH, R., 1953. Thiouracil modification of effect of radioiodine (1181) on the thyroid
Radiology, 61: 391-400.

SPEIDEL, C. C., 1926. Studies of hyperthyroidism. II. The significance of changes in the
thymus glands of thyroid-treated frog tadpoles. Amer. J. Anat., 37: 141-157.
Taytor, A. C., AND J. J. Korzros, 1946. Stages in the normal development of Rana pipiens

larvae. Anat. Rec., 94: 7-23.
WYNGAARDEN, J. B., B. M. WricHt AnD P. Ways, 1952. The effect of certain anions upon

the accumulation and retention of iodide by the thyroid gland. Endocrinology, 50:
537-539.

Physiol. Rev., 30: 194-205.
Nature, biosynthesis and metabolism of thyroid hormones.

PiveotiGATION OF THE FERTILIZATION INHIBITING ACTION
OF ARBACIA DERMAL SECRETION?

CHARLES B. METZ

Oceanographic Institute, Florida State University, Tallahassee, Florida
and Marine Biological Laboratory, Woods Hole, Mass.

Several investigations (Oshima, 1921; Pequegnat, 1948; Harvey, 1956; Metz,
1959a) have shown that Arbacia releases a fertilization inhibiting agent or agents
when subjected to conditions of stress. The agent is present in a yellowish fluid
called “dermal secretion” (Oshima, 1921) and is probably released from certain
pigment- containing cells of the integument (Pequegnat, 1948). Indeed this dermal
secretion is probably responsible for the fertilization es action ascribed
to Arbacia blood by Lillie (1914).

Unfortunately, little information is as yet available concerning the mechanism
by which the dermal secretion inhibits fertilization. In a previous study (Metz,
1959a) it was found that the dermal secretion inhibits the agglutination of Arbacia
sperm by fertilizin, the specific sperm isoagglutinin obtained from the jelly layer
surrounding eggs. This agglutination inhibiting action was found to result from
an inactivation of the specific combining sites of the fertilizin that react with the
antifertilizin of the sperm surface.

The parallel inhibiting action of dermal secretion on fertilizin agglutination
of sperm and on fertilization suggests that the latter action may also result from
inactivation of fertilizin combining sites. The present study was undertaken to
investigate the mechanism of fertilization inhibition by the dermal secretion and
particularly to determine if this action resulted from an inactivation of fertilizin.
The experiments have been reported briefly elsewhere (Metz, 1959b). They
indicate that fertilizin is not the site of inhibitor action.

MATERIALS AND METHODS

Two species of sea urchins were used in the experiments, namely Lytechinus
variegatus and Arbacia punctulata. Animals of the former species were obtained
from the large population in the vicinity of the Florida State University Marine
Laboratory, Alligator Point, Florida. The latter (Arbacia) were also obtained
from this area, and from Panama City, Florida. This species was also obtained
from the Woods Hole, Massachusetts, area and used at the Marine Biological
Laboratory.

Gametes were obtained from Lytechinus by inducing the animals to shed by
treatment with 0.5 M KCl. Gametes were obtained from Arbacia by electrical
stimulation.

1 Aided by grants from the National Institutes of Health and the National Science Founda-
tion. Contribution #108 from the Oceanographic Institute, Florida State University.

439

440 CHARERS 2. METZ

Solutions of dermal secretion were obtained by immersing Arbacia in tap or
distilled water for 1-3 minutes, rinsing them in sea water and collecting the yellow-
ish fluid that drained from the animals (Metz, 19592). The pH of such dermal
secretion is about 7.5. The pH was adjusted to that of sea water (8.0-8.2) be-
fore use.

RESULTS

In the presence of Arbacia dermal secretion both Arbacia and Lytechinus eggs
fail to fertilize, even though these eggs are bombarded by highly motile sperma-
tozoa. Such eggs fail to elevate fertilization membranes or show any indication
of nuclear activity. Clearly, then, the dermal secretion blocks fertilization at its
initial stages. The first step in analyzing the mechanism of this action of the

TABLE |

Percentage of eggs cleaving after insemination with Arbacia sperm washed from dermal secretion

Dermal secretion—treated sperm S.W.—treated sperm
Dilution of treated
sperm : ;
Washed in S.w, | Re-suspended in D-S. | Washed in S.W. Reo

1 100% (200+) 60% (103) 100% (200+) 100% (200*)
1/5 100% (200+) 100% (200+) 100% (200*) 100% (200*)
1/25 90% (128) 100% (200*) 99% (200+) 100% (200*)
1/125 62% (98) 100% (97) 97% (119) 100% (200*)
1/625 20% (109) 61% (116) 64% (108) 93% (123)

Figures in parenthesis are the number of eggs counted. The four mixtures were prepared in
the following proportions: 0.5 ml. approximately 20% sperm suspension; 1.5 ml. D.S. or S.W.
These mixtures were centrifuged and the sperm suspended in 1.5 ml. S.W. or in the supernatant.
This procedure was repeated and all four sperm suspensions were finally diluted with 25 ml. S.W.
(sperm “dilution 1’’ = approximately 0.4% undiluted semen). The suspensions were diluted
serially as indicated. Two drops of unfertilized eggs were added to 4 ml. of each sperm dilution.
In a control experiment eggs failed to fertilize when inseminated in the D.S. supernatant from the
centrifuged sperm suspension.

dermal secretion was to determine if the agent acted upon the sperm, the egg, or
both gametes.

Action on the sperm. In the previous studies (Pequegnat, 1948; Metz, 1959a)
it was found that the dermal secretion from Arbacia markedly enhanced both
motility and respiration of Arbacia sperm and that these effects were long lasting.
Evidently, then, any fertilization inhibiting action on the sperm must result from
some subtile effect such as “premature discharge” of the acrosomal filament or
blocking of an essential receptor site on the sperm surface.

To test for such action Arbacia sperm were treated with dermal secretion, sub-
sequently washed free of the solution by centrifugation and finally tested for
fertilizing capacity. As seen in Table I, Arbacia sperm washed from dermal se-
cretion are as effective in fertilizing eggs as sperm washed from sea water. Fur-
thermore, dilution of dermal secretion-treated sperm with sea water restores
fertilizing capacity. Thus (Table 1) the sperm in dermal secretion gave only 60%
fertilization at “dilution 1” but yielded 100% fertilization at higher dilutions.

FERTILIZATION INHIBITOR ACTION 441

This inhibition at “dilution 1” is attributed to an action of dermal secretion on the
eggs. At the higher dilutions the dermal secretion is evidently diluted beyond
the minimum inhibitory concentration. Finally, the data suggest that washing,
as opposed to re-suspension in the supernatant, reduces the fertilizing capacity of
control sperm. This is in accord with Hayashi’s (1945) observations on the
beneficial effect of seminal plasma on sperm. Results similar to those in Table I
were obtained in a second experiment using Arbacia sperm and eggs. Comparable
results were also obtained in two experiments with Lytechinus sperm and eggs
using Arbacia dermal secretion. However, the results were not as striking in
these experiments because the more delicate Lytechinus sperm does not survive
centrifugation and re-suspension as well as Arbacia sperm.

It is evident from these experiments that dermal secretion does not inhibit
fertilization by irreversible action on the sperm. Accordingly, attention was di-
rected toward possible effects on the egg.

Action of dermal secretion on the egg. According to Pequegnat (1948) wash-
ing in sea water at least partially restores fertilizability to dermal secretion-treated
Arbacia eggs. Harvey (1956, page 57), on the other hand, states that “it has
been my experience that sometimes the eggs are not fertilizable, even after re-
peated washings with sea water.” Observations in the present study indicated
that reversibility of inhibition by washing varies from preparation to preparation.
Eggs washed from some dermal secretion preparations fertilize readily whereas
those from others show varying degrees of reduced fertilizability. This is evidenced
in two ways. As seen in Table III, experiment 1, eggs washed from a dermal
secretion preparation may approach 100 per cent fertilizability. However, sub-
stantially higher sperm concentrations are required to fertilize such eggs than are
required for the controls. A second expression of reduced fertilizability is de-
layed cleavage. In some experiments Arbacia eggs washed from dermal secretion
and subsequently inseminated fail to cleave synchronously with the controls. In
these, first cleavages are asynchronous and delayed one or two divisions as com-
pared to the controls.

These results indicate that the probability of a “successful sperm-egg contact”
(Rothschild, 1956) is lower for the treated eggs than for the control eggs under
comparable conditions. The reduced probability of a successful sperm-egg contact
in turn could result from a correspondingly reduced number of receptor sites on
the egg surface. In the case of cleavage delay the treated eggs may not fertilize
at the moment of insemination because time is required for a successful sperm-egg
contact to be made. The assumption is then that the egg develops at the normal
rate and cleaves on schedule following a successful sperm-egg contact. This view
is supported at least to the extent that fertilized eggs show no cleavage delay when
placed in the dermal secretion as early as one to two minutes following insemination.
However, eggs placed in dermal secretion a few seconds after insemination fre-
quently show arrested fertilization membrane elevation and sperm entry (especially
clear in Lytechinus eggs). These observations suggest an inhibition of the cortical
reaction. In any event these experiments show that washing does not readily
restore dermal secretion-treated eggs to the control level of fertilizability. Ac-
cordingly, it appears likely that the dermal secretion inhibits fertilization by an
effect upon fertilizability of the eggs, as opposed to sperm. This view is supported
by studies with enzymes as recorded in a later section.

442 CHARLES B. METZ

The effect of dermal secretion on the formation of blebs by inseminated oocytes.
The observations described above suggest that the dermal secretion inhibits, blocks
or otherwise prevents an early step in fertilization. In a further test of this as-
sumption dermal secretion was examined for its effect on the formation of blebs
or papillae on. oocytes when these cells are inseminated. As is well known (e.g.,
Rothschild and Swann, 1949) sea urchin oocytes are not fertilizable. However,
these cells, when exposed to sperm, form numerous blebs on their surfaces.
Formation of such blebs has been attributed to an interaction between the sperm and
oocyte surface. To test for the effect of dermal secretion on oocyte bleb formation,
suspensions of eggs containing oocytes were placed in dermal secretion solutions,
sperm was then added and the mixtures examined for bleb formation and com-
pared with appropriate controls. Many such observations using both Arbacia and
Lytechinus eggs show that in the presence of dermal secretion bleb formation does
not result to a significant degree, whereas numerous blebs formed in the control
oocytes. Some inhibitory reaction evidently results which prevents interaction of
sperm and oocyte or blocks the bleb-forming response mechanism of the oocyte.
In any event these observations support the view that the dermal secretion blocks
an early step in fertilization.

In a further attempt to localize the site of action of the dermal secretion, the
egg surface was subjected to mild dissection and then tested for sensitivity to the
inhibiting action of the dermal secretion.

The effect of dermal secretion on fertilizability of acid-treated eggs. The first
agent selected for such dissection was acid sea water. As is well known (Tyler,
1948) acid sea water dissolves the jelly surrounding the egg and thereby removes
the fertilizin from the egg except possibly for a layer bound to the egg surface.
In two experiments Lytechinus eggs were exposed to pH 4-5 sea water and in four
experiments Arbacia eggs were exposed to pH 5 sea water to remove the jelly
layer. These eggs were subsequently washed in sea water, treated with dermal
secretion and in some cases washed again and tested for fertilizability over a wide
range of sperm concentrations. In all experiments the acid-treated eggs showed
the same sensitivity to the fertilization inhibiting action of dermal secretion as did
control eggs. Table II gives two experiments, one using Arbacia eggs and the
other Lyiechinus eggs. In both experiments it is clear that acid-treated and sea
water-treated eggs are both sensitive to the fertilization inhibiting action of the
dermal secretion. The well known reduction in fertilizing capacity upon removal
of jellies from eggs (Tyler, 1948) is not pronounced at the sperm concentrations
employed but indications of this effect are seen in the experiment with Lytechinus
eggs.

Since the sensitivity of eggs to the fertilization inhibiting action of the dermal
secretion is not affected by acid treatment, it follows that the site of action of the
dermal secretion is not the egg jelly. It is also presumably not the egg fertilizin
since fertilizin constitutes the jelly. However, it is possible that a layer of
fertilizin remains bound at the egg surface after the acid treatment and that this
is an essential layer blocked by the dermal secretion (e.g. Tyler, 1948).

The effect of dermal secretion on fertilizability of eggs treated with proteolytic
enzymes. Proteolytic enzymes were used to achieve a further mild dissection of
the egg surface. As is well known (e.g. Tyler and Metz, 1955; Tyler, Monroy
and Metz, 1956), eggs treated with proteolytic enzymes lack jellies, they fail to

FERTILIZATION INHIBITOR ACTION 443

form normal fertilization membranes, they become polyspermic readily and
hybridize more readily than control eggs when inseminated with sperm of foreign
species. The following proteolytic enzyme preparations were used: “crystalline
protease” (Nutritional Biochemical Corporation) in 0.1% sea water solution;
“crystalline trypsin” (Nutritional Biochemical Corporation) in 0.01% sea water
solution. Enzymatic action of the trypsin was stopped prior to addition of dermal
secretion by washing the eggs in an excess of trypsin inhibitor (ovomucoid, Nu-
tritional Biochemical Corporation, 0.1% solution in sea water; or soybean trypsin
inhibitor, Mann Research Corporation, 0.1% in sea water). In all, ten experi-
ments were performed. These included five with Arbacia eggs, one using protease
and four employing the combination of trypsin and trypsin inhibitor. Experiments

TABLE I

Effect of Arbacia dermal secretion on fertilizability of acid-treated eggs as indicated by
per cent cleavage

Acid-treated eggs S.W.-treated eggs
Sperm dilution

D.S.-treated S.W.-treated D.S.-treated S.W.-treated
1 23% (124) 91% (141) eA Taney) 90% (200*)
Arbacia 1/5 Bo e158) 97% (107) 27 ass) 90% (2007)
eggs 1/25 1% (200+) | 90% (152) 1% (200+) | 90% (200+)
1/425 0% (200°) 81% (128) 0% (200°) 97% (105)

1 14.5% (124) 88% (148) 19% (114) 96% (119)

Lytechinus 1/6 6% (114) 97% (147) 4.5% (150) 98% (134)
eegs 1/36 0%: 42008) 46% (140) LZ C200") 35% (130)
1/216 ou (200) 14% (180) 0% (200°) 74% (182)

Arbacia eggs: Jellies were removed by acidification to pH 5.4, eggs were then washed twice
in 50-ml. samples of S.W., 0.5 ml. eggs was added to 2.4 ml. D.S. or S.W. and finally three-drop
samples of these eggs were added to 5-ml. samples of Arbacia sperm. Sperm dilution 1 = 0.1%.

Lytechinus eggs: Jellies were removed by acidification of pH 3.9. Eggs were washed twice in
S.W., 0.75 ml. eggs was then added to 2 ml. D.S. or S.W. They were subsequently washed twice
in S.W. Four drops eggs were added to 5 ml. of Lytechinus sperm. Sperm dilution 1 = 0.66%.
D.S.-treated eggs at sperm dilution 1 showed high incidence of polyspermy. Many of the un-
cleaved eggs contained multipolar spindles.

using Lytechinus eggs included four applying protease treatment and one using
the trypsin-trypsin inhibitor treatment. In all ten of these experiments enzyme-
treated eggs were subsequently treated with dermal secretion (or sea water),
washed in sea water and inseminated. With this procedure the enzyme-treated
eggs fertilized equally readily whether they received the dermal secretion or sea
water treatment. Proteolytic enzyme treatment alone, like acid treatment, lowers
the fertilizing capacity of the eggs (e.g. Tyler and Metz, 1955); In Labler lit
the results of two of the ten experiments, one using Arbacia and the other Lytechinus
eggs, are presented. In both experiments the fertilizability of sea water-treated
eggs was markedly reduced by treatment with dermal secretion. Thus in experi-
ment 1 to even approach comparable fertilization (100% of the eggs), over 125
times the control sperm concentration was required to fertilize the dermal secretion-

444 CHARLES B. METZ

treated eggs. In experiment 2 the difference in sperm concentration (to achieve
1% fertilization) was greater than 1000-fold. On the other hand no significant
difference is seen between the fertilizability of dermal secretion and sea water-
treated eggs that had been pretreated with enzyme. Indeed, comparison of the
enzyme-treated and sea water-treated eggs shows that on an absolute basis enzyme

TABLE III

Effect of proteolytic enzymes on sensitivity of eggs to the fertilization inhibiting action
of Arbacia dermal secretion

Enzyme-treated eggs S.W.-treated eggs
Sperm
dilution
D.S.-treated S.W.-treated D.S.-treated S.W.-treated
Experiment 1 1 99% (149) 97% (154) 89% (158) 100% (200*)
trypsin-trypsin 1/5 98% (151) 68% (187) 91%t (162) 100% (200+)
inhibitor-treated 1/25 42% (295) 21% (186) 37 (156) 100% (200°)
Arbacia eggs i725 8% (199) 5% (187) <1 = (200%) 100% (200°)
Experiment 2 1 94% (107) 93% (97) <1F"'(200*) 93% (88)
Protease-treated (82%)* (51%)* (6%)*
Lytechinus eggs
1/6 87% (118) 83% (87) 0% (200°) 98% (96)
(32170) (47%)* (2%)*
1/36 81% (131) 12% (149) 0% (200°) 65% (113)
(13%)* (0%)* Cao
1/216 37% (169) 25% (128) 0% (200°) 35% (108)
(1%) nya)” (0%)*
1/1296 14% (133) Ly (2008) 0% (2007) <3% (200%)

% = Per cent cleavage.

Experiment 1. Two-ml. samples of Arbacia eggs were added to 4-ml. samples of S.W. and
0.05% trypsin (in S.W.). After 45 minutes the eggs were washed twice in S.W., 1.5 ml. of the
eggs were transferred to 10 ml. 0.1% ovomucoid in S.W. Subsequently the eggs were washed
again, samples were treated with D.S. and S.W. and washed. Four drops eggs were added to
5 ml. Arbacia sperm dilution series. A duplicate control experiment omitting the ovomucoid
treatment gave similar results. Sperm dilution 1 = 0.1% semen. Degree of polyspermy was not
recorded in this experiment. Figures in parenthesis are the total number of eggs counted.

t Cleavage was delayed two divisions as compared to controls.

Experiment 2. Five-tenths-ml. samples of unfertilized Lytechinus eggs were placed in 5 ml.
0.1% protease solution or S.W. After 89 minutes 1-ml. samples of protease- and S.W.-treated
eggs were each added to 1-ml. samples of D.S. and S.W. Twenty minutes later the eggs were
washed in S.W. and 4-drop samples were added to 5 ml. Lytechinus sperm at the dilutions indi-
cated. Sperm dilution 1 = 16% of undiluted semen, approximately 2 X 108 sperm/ml. in this
case.

* Percentage of eggs counted showing the abnormal cleavage characteristic of polyspermy.

pretreatment results in considerable improvement in fertilizability following ex-
posure to dermal secretion. This improvement is evidently limited at the level of
the fertilizability of ¢nzyme-treated control eggs.

In the experiments just described the enzyme-treated eggs were washed free
of dermal secretion and inseminated in sea water. If the enzyme-treated eggs are

FERTILIZATION INHIBITOR ACTION 445

inseminated directly in the dermal secretion they fail to show the marked improve-
ment in fertilizability. In the discussions given below this reservation concerning
the loss of dermal secretion sensitivity following trypsin treatment is assumed.

Two possible explanations for the improving action of proteolytic enzyme
pretreatment are evident. The enzyme could destroy the activity of the dermal
secretion and thereby protect the eggs from this inhibitor, or the enzyme could
act directly upon the egg. Even after several washings sufficient enzyme might be
carried over or remain adsorbed to the egg to prevent the inhibition by destroying
the dermal secretion. This explanation might suffice for the experiments with
protease-treated eggs. However, prior to exposure to dermal secretion the trypsin-
treated eggs were exposed to an amount of trypsin inhibitor (ovomucoid, four
experiments with Arbacia eggs; soybean inhibitor, one experiment with Lytechinus
eggs) sufficient to inhibit all trypsin activity. Furthermore, in one experiment the
proteolytic enzyme treatment was applied after exposure of the egg to dermal
secretion. This post-treatment with enzyme was as effective as pretreatment in
rendering dermal secretion-treated eggs fertilizable. Furthermore, in preliminary
experiments the fertilization inhibiting action of dermal secretion was not ap-
preciably affected by long exposure to trypsin. Clearly, then, trypsin does not
render eggs insensitive to dermal secretion by inactivating this inhibitor. The
alternative, second possibility is that the enzyme destroys or removes some dermal
secretion sensitive site or substance of the egg which is essential in normal fertiliza-
tion. Such action in turn evidently exposes some alternative pathway for fertiliza-
tion. This alternative pathway is reversibly inhibited by dermal secretion because
enzyme-treated eggs fail to fertilize when inseminated in dermal secretion.

Relation of dermal secretion action to the block to polyspermy. The response
of enzyme-treated eggs following dermal secretion treatment paralleled the enzyme-
treated sea water controls not only quantitatively but also qualitatively. As is
well known, eggs treated with proteolytic enzymes undergo polyspermic develop-
ment at sperm concentrations considerably lower than that required to obtain
polyspermy in normal eggs. This susceptibility to polyspermy was not confined
to enzyme-treated control eggs. Thus, in the enzyme-pretreated group of experi-
ment 2, Table III, dermal secretion-treated eggs yielded at least as high an incidence
of abnormal (polyspermic) cleavage as the enzyme-treated sea water controls.
This was true at all sperm dilutions.

These observations suggest that the site of action of dermal secretion may be
associated in some way with the block to polyspermy. Accordingly it seemed of
interest to test eggs pretreated with another polyspermy inducing agent for sensi-
tivity to the fertilization inhibiting action of dermal secretion. Nicotine has long
been known to be a very effective polyspermy inducing agent (see Rothschild,
1953; Hagstroém and Allen, 1956, for literature) and was therefore selected as the
polyspermy inducing agent. Nicotine diluted to 2 x 10°* in sea water proved to
be optimal for inducing polyspermy in Arbacia eggs. However, in two experi-
ments such treatment did not lower the sensitivity of Arbacia eggs to the fertiliza-
tion inhibiting action of dermal secretion. When tested quantitatively over a wide
range of sperm concentrations, the nicotine-pretreated eggs showed the same
degree of fertilization inhibition as control eggs after exposure to dermal secretion.
Apparently, then, the sensitivity of eggs to the dermal secretion is not invariably
associated with the block to polyspermy.

446 CHARLES B. METZ

Relation of the fertilization inhibiting action of dermal secretion to fertilizin.
As suggested above, the fertilization inhibiting action of dermal secretion and its
reversal by proteolytic enzymes is most readily explained by assuming a combina-
tion of the inhibitor with an essential egg substance. Such combination blocks the
essential substance and prevents it from entering into reactions involved in the
initial stages of fertilization. The substance that immediately suggests itself is
fertilizin, and especially so since dermal secretion prevents combination of fertilizin
with sperm by inactivating the sperm receptor sites of the fertilizin molecule (Metz,
1959a). If the dermal secretion inhibits fertilization by an essentially irreversible
combination of the inhibitor with fertilizin at or near the egg surface, then the
fertilization inhibiting action of the dermal secretion should be destroyed by ad-
dition of an excess of fertilizin.

TABLE IV

Effect of fertilizin on fertilization inhibiting action of dermal secretion

D.S. diluted in fertilizin D.S. diluted in S.W. S.W. + fertilizin S.W. alone
DS.
dilution Sperm Sperm Sperm Sperm
Cleavage aggluti- Cleavage aggluti- Cleavage aggluti- Cleavage aggluti-
nation nation |. nation nation
1/2 0% (200+) a 0% (200*+)| — [99% (136) |++++ | 100% (200+)| +

1/10 | 0% (200+) |+++ | 0% (200*)| +
++4++

1/50 | 13% (159) |++++| 6% (140) | +

1/250 |97% (119) |++++ |86% (131) | +

Aliquots of a dermal secretion preparation were diluted serially in sea water and in fertilizin
as indicated, two drops of unfertilized Arbacia eggs were then added to 0.8-ml. samples of each
D.S. dilution, and finally 5 ml. 0.0003% Arbacia sperm were added to each sample. % = per cent
cleavage; values in parenthesis are the number of eggs counted. To test for fertilizin excess in
the mixtures 2-drop samples of the supernatant from each mixture were withdrawn after insemina-
tion of the eggs. These were mixed with 2 drops 1-2% sperm and the agglutinating action re-
corded ona — to ++-+4 scale in the table. The sperm agglutination titer of the original fer-
tilizin solution was >729.

To test for such inactivation of the fertilization inhibiting action of the dermal
secretion by fertilizin, three experiments using Arbacia eggs were performed. In
these experiments dermal secretion was diluted serially in fertilizin, constant
amounts of eggs and sperm were added to the mixtures and the percentage of
cleavage determined and compared with appropriate controls. The mixtures were
also tested for presence of excess fertilizin by examining for sperm agglutinating
action. One experiment, typical of the three, is given in Table IV. As seen in
the controls the concentration of sperm used was sufficient to fertilize 99% of the
eggs. However, it was necessary to dilute the dermal secretion to 1/250 to ap-
proach this value. Furthermore, the dermal secretion inhibited fertilization to the
same dilutions (1/50 to 1/250) whether diluted in fertilizin or sea water. In-
deed, at dilution 1/50, five times the dilution required to achieve fertilizin excess,
only 13% of the eggs fertilized. This parallel loss of fertilization inhibiting action

FERTILIZATION INHIBITOR ACTION 447

when diluted in fertilizin and sea water, combined with the demonstration of a
fertilizin excess at 1/10 dilution, clearly shows that the fertilization inhibiting
action of dermal secretion is not impaired by fertilizin. Evidently, then, fertilizin
does not neutralize the fertilization inhibiting action of dermal secretion. There-
fore, it seems unlikely that dermal secretion inhibits fertilization by combination
with fertilizin.

It should be noted that fertilizin treatment alone did not have a marked effect
on fertilization in these experiments. This is in keeping with previous observations
on Arbacia (Tyler and Metz, 1955). However, fertilizin treatment does have
a marked effect in some other species (Tyler, 1941).

DISCUSSION

The experiments described here show that the dermal secretion from Arbacta
inhibits fertilization by action on the egg, not the sperm. The agent presumably
blocks an essential reaction or reactions in the initial stages of fertilization. One
such response of the egg that is evidently blocked by the dermal secretion is the
formation of the fertilization membrane. In fact, treatment with dermal secretion
a few seconds after fertilization arrests membrane formation. Such eggs have
membrane elevated over only a part of the egg surface. It will be of interest to
determine if other cortical phenomena of fertilization, such as cortical granule
discharge (Motomura, 1941; Runnstrom, 1949; Endo, 1952) and the changes
in optical properties observed in dark field illumination, are also arrested by the
dermal secretion. These observations should give some indication of whether the
agent inhibits the propagation of the cortical responses of the egg or interferes only
with the process of membrane elevation (see Metz, 1957, for interrelations of
cortical phenomena).

Unfortunately, the relationship of the blebs formed by inseminated oocytes to
stages of fertilization of normal eggs is somewhat obscure. Therefore, no detailed
interpretation of such inhibition in terms of fertilization is warranted. However,
the blebs suggest that some of the initial stages of fertilization have taken place.
Accordingly, inhibition of bleb formation by the dermal secretion indicates that
this agent can inhibit initial steps in fertilization. It is not unlikely that the dermal
secretion inhibits essential steps in the interaction of the sperm and the eggs, as well
as arresting manifestations of the propagated responses of the egg.

The fertilization inhibiting action of the dermal secretion presumably results
from interaction of the inhibiting agent with some essential substance or substances
of the egg. Such interaction evidently results in destruction or blockage of the
essential substance. This in turn reduces the number of receptor sites on the egg
and the probability of a successful sperm-egg contact. This view is consistent with
the observation that inhibitor-treated eggs can sometimes be fertilized if insemi-
nated with high sperm concentrations.

Cases of cleavage delay may also be explained in formal fashion as a reduction
of the probability of a successful sperm-egg collision. Thus eggs with a reduced
number of receptor sites should have a lower “fertilization rate” (e.g. Rothschild,
1956; Hagstr6m and Hagstrom, 1954) than control eggs because of the increased
time required for the less probable event. Another possibility is that cleavage
delay is a manifestation of a reversal of inhibition such that washed dermal se-

448 CHARLES B. METZ

cretion-treated eggs undergo a slow reversal of inhibition. It should be noted,
however, that other factors such as delay in the propagated response of the egg
have not been eliminated as explanations for delayed cleavage of dermal secretion-
treated eggs.

The fertilization inhibiting interaction between the dermal secretion and the
essential egg substance could result in enzymatic destruction of the latter. It
appears more likely, however, that the inhibitor acts by relatively undissociable
combination with an egg substance. This view is supported by the observation
that washing sometimes results in at least partial restoration of fertilizability to
dermal secretion-treated eggs. It is also consistent with the observed reversal of
inhibition by proteolytic enzymes.

The identity of the egg substance or substances with which the dermal secretion
combines to inhibit fertilization remains obscure. This material is evidently not
the sperm isoagglutinin, fertilizin. Several lines of evidence support this view.
Fertilizin constitutes the jelly layer surrounding the egg. Eggs from which this
jelly has been removed by acid sea water treatment are dermal secretion sensitive.
Evidently, then, fertilization inhibition is not mediated by action of the dermal
secretion on the jelly. However, this experiment does not rule out the possibility
of action on an acid-resistant layer of fertilizin bound to the egg surface (e.g.
Tyler, 1941). Further “dissection” of the egg surface was achieved by treating
eggs with proteolytic enzymes. Following such treatment the eggs were rela-
tively insensitive to the fertilization inhibiting action of dermal secretion. Evi-
dently the enzyme digests or otherwise eliminates the primary egg substance with
which the dermal secretion combines. However, there is some question if even
the relatively drastic action of proteolytic enzymes removes all the fertilizin from
the egg surface (Tyler and Metz, 1955) in a reasonable time. As a final test for
a role of fertilizin in the inhibition of fertilization by dermal secretion, fertilizin
and dermal secretion were mixed and the mixtures were assayed for fertilization
inhibiting action. The fertilizin had no effect on the inhibiting action of dermal
secretion. This was true even when the fertilizin was present in large excess as
measured by sperm agglutinating action.

In view of these results it appears that the dermal secretion does not inhibit
fertilization by an action on fertilizin. This is of particular importance because
previous studies (Metz, 1959a) have shown that dermal secretion of Arbacia
destroys the sperm combining sites of the fertilizin molecule.

Apart from consideration of the nature of the egg substance involved in the
inhibition, the reversal of inhibition by either “pre- or post-” treatment of eggs
with proteolytic enzymes has interesting implications for other aspects of fertiliza-
tion. According to the interpretation given above dermal secretion inhibits
fertilization by combination with a substance that is essential for fertilization of
the normal egg. Pretreatment of the egg with proteolytic enzymes removes this
essential egg substance and renders the egg relatively insensitive to the inhibitor.
Evidently coincident with the removal of the essential egg substance by enzyme,
an alternative pathway(s) to fertilization is exposed. Whether this pathway
resides in the sperm-egg attachment mechanism, the activation initiating mechanism
or the propagative and cortical response mechanism remains to be established.
However, it should be noted that the “alternative pathway” exposed by proteolytic

FERTILIZATION INHIBITOR ACTION 449

enzyme treatment is associated with a loss of the block to polyspermy and a re-
duction in fertilization specificity (Bohus Jensen, 1953; Hultin, 1948a and 1948b;
Tyler and Metz, 1955) and is itself reversibly inhibited by dermal secretion. This
last conclusion follows from the fact that enzyme-treated eggs must be washed
free of dermal secretion before they will fertilize. Whatever the site of action of
the proteolytic enzymes the action does not result in drastic visible effects on the
unfertilized egg, for sections of trypsin-treated and control eggs are indistinguish-
able when seen with electron optics (Parpart, personal communication ).

This investigation suggests the use of fertilization inhibiting agents as markers
or labels for particular sites or substances involved in fertilization. Extension
of the investigation to include detailed examination of other fertilization inhibitors
should prove interesting for these may act at other sites. Indeed, the studies of
Branham and Metz (1959) indicate that the inhibition of fertilizin agglutination
of sperm by extracts from Fucus and by dermal secretion results from two quite
different mechanisms. It will not be surprising if these two agents are found to
inhibit fertilization by action at different sites in the complex of reactions that re-
sult in fertilization.

SUMMARY

1. Sperm washed from Arbacia dermal secretion fertilize eggs as readily as
sperm washed from sea water.

2. Eggs washed from Arbacia dermal secretion do not fertilize as readily as
controls. At best such eggs require high sperm concentrations to achieve fertiliza-
tion. In some experiments the treated eggs show a marked cleavage delay. It
is concluded that the dermal secretion inhibits fertilization by action on the egg,
not on the sperm:

3. Addition of dermal secretion to eggs several minutes after insemination has
no effect on the development of the eggs. Similar treatment a few seconds after
insemination results in arrest of fertilization membrane elevation and sperm penetra-
tion. The dermal secretion also inhibits the formation of blebs in inseminated
oocytes. It is concluded that dermal secretion inhibits by blocking an initial
stage(s) in fertilization.

4. Jellyless (acid-treated) eggs fail to fertilize after treatment with dermal
secretion. They have undiminished sensitivity to the inhibiting action of the
dermal secretion.

5. Proteolytic enzyme-treated eggs are relatively insensitive to the inhibitor.
They fertilize as readily as controls after treatment with dermal secretion, provided
they are washed free of this inhibitor. They fail to fertilize if inseminated in
dermal secretion.

6. The dermal secretion has undiminished fertilization inhibiting action in the
presence of an excess of fertilizin.

7. The experiments suggest that the dermal secretion inhibits fertilization by
combining with some egg substance that is essential for fertilization of the normal
egg. The egg substance is apparently not the sperm isoagglutinin, fertilizin.
Treatment of the egg with proteolytic enzymes eliminates the essential egg sub-
stance and simultaneously exposes an alternative pathway to fertilization. The
latter is reversibly inhibited by dermal secretion.

450 CHARLES B: METZ

LITERATURE.CIyED

Bonus JENSEN A., 1953. The effect of trypsin on the cross fertilizability of sea urchin eggs.
Exp. Cell Res., 5: 325-328.

BRANHAM, J. M., anp C. B. Mertz, 1959. Inhibition of fertilization and agglutination in
Arbacia by extracts from Fucus. Biol. Bull., 117: 392-393.

Enpo, Y., 1952. The role of the cortical granules in the formation of the fertilization mem-
brane in eggs from Japanese sea-urchins. Exp. Cell Res., 3: 406-418.

Hacstrom, B. E., anp R. D. ALLEN, 1956. The mechanism of nicotine-induced polyspermy.
Exp. Cell Res., 10: 14-23.

Hacstrom, B., anp B. Hacstrom, 1954. A method of determining the fertilization rate in
sea-urchins. Exp. Cell Res., 6: 479-484.

Harvey, E. B., 1956. The American Arbacia and Other Sea Urchins. Princeton University
Press, Princeton, New Jersey.

Hayasul, T., 1945. Dilution medium and survival of the spermatozoa of Arbacia punctulata.
I. Effect of the medium on fertilizing power. Biol. Bull., 89: 162-179.
Huttin, T., 1948a. Species specificity in fertilization reaction. I. The role of the vitelline
membrane of sea-urchin eggs in species specificity. Arkiv Zool., 40A: 1-9.
Huttin, T., 1948b. Species specificity in fertilization reaction. II. Influence of certain factors
on the cross-fertilization capacity of Arbacia lixula (L.). Arkiv Zool., 40A: 1-8.

Lituiz, F. R., 1914. Studies of fertilization. VI. The mechanism of fertilization in Arbacia.
J. Exp. Zool., 16: 523-588.

Metz, C. B., 1957. Mechanisms in fertilization. In: Physiological Triggers, T. H. Bullock,
editor. American Physiological Society, Washington, D. C.

Metz, C. B., 1959a. Inhibition of fertilizin agglutination of sperm by the dermal secretion
from Arbacia. Biol. Bull., 116: 472-483.

Metz, C. B., 1959b. Studies on the fertilization inhibiting action of Arbacia dermal secretion.
Biol. Bull., 117: 398.

Motomura, I., 1941. Materials of the fertilization in the eggs of Echinoderms. Sct. Rep.
Tohoku Imp. Univ., Ser. 4, 16: 345-363.

Osuima, H., 1921. Inhibitory effect of dermal secretion of the sea urchin upon the fertilizability
of the egg. Science, 54: 578-580.

PEQUEGNAT, W., 1948. Inhibition of fertilization in Arbacia by blood extract. Biol. Bull.,
95: 69-82.

ROTHSCHILD, Lorp, 1953. The fertilization reaction in the sea urchin. The induction of
polyspermy by nicotine. J. Exp. Biol., 30: 57-67.

RoTHScHILD, Lorn, 1956. Fertilization. John Wiley and Sons, N. Y.

RoTHSscHILD, Lorp, AND M. M. Swann, 1949. The fertilization reaction in the sea-urchin egg.
A propagated response to sperm attachment. J. Exp. Biol., 26: 164-176.

Runwnstrom, J., 1949. The mechanism of fertilization in metazoa. Advances in Enzymol.,
9: 241-327.

Tyrer, A., 1941. The role of fertilizin in the fertilization of eggs of the sea urchin and other
animals. Biol. Bull., 81: 190-204.

Ty er, A., 1948. Fertilization and immunity. Physiol. Rev., 28: 180-219.

Tyrer, A., AnD C. B. Metz, 1955. Effects of fertilizin-treatment of sperm and trypsin-treat-
ment of eggs of homologous and cross-fertilization in sea urchins. Pubbl. Staz. Zool.
Napoli, 27: 128-145.

TyLer, A., A. Monroy anp C. B. Metz, 1956. Fertilization of fertilized sea urchin eggs.
Biol. Bull., 110: 184-195.

PROTEIN CHANGES IN DEVELOPMENT ??

MELVIN SPIEGEL
Department of Zoology, Dartmouth College, Hanover, New Hampshire

Biologists have long been in agreement with the thesis that morphological dif-
ferentiation is either preceded by, or accompanied simultaneously by, an underlying
chemical differentiation. In the testing of this hypothesis, considerable attention
has been focused on the qualitative and quantitative changes occurring in the
proteins of developing embryos.

The usual biochemical techniques have been employed to describe and to
quantitate these changes. One should mention the solubility studies of sea urchin
embryo proteins by Mirsky (1936), the electrophoresis of sea urchin embryo pro-
teins by Monroy (1950), and the electrophoresis of amphibian embryo proteins by
Flickinger and Nace (1952). Recently the serological technique has been used
quite extensively; the precipitin test by Cooper (1948), Ebert (1951), Harding
et al. (1955) ; the Oudin technique by Spar (1953) and by Cooper (1948).

In addition to these in vitro studies a number of im vivo studies have been
carried out. Ebert (1953, 1955) and his co-workers have studied the effects of
specific antisera on living chick embryos, and Tyler and Brookbank (1956) have
studied the cytotoxic effects of specific antisera on sea urchin eggs. We have
carried out similar studies on dissociated sponge cells (Spiegel, 1955). The results
of these investigations have been extensively summarized by Nace (1955) and
by Tyler (1957).

Briefly, the results obtained through the use of the serological techniques on
amphibians and sea urchins are as follows. In the sea urchin, it has been demon-
strated by Perlmann and Gustafson (1948) and Perlmann (1953) that new anti-
gens appear during development. These new antigens were first detected in the
gastrula stage. Harding et al. (1954, 1955), working with lethal hybrid sea
urchin embryos, were able to detect paternal antigens at the blastula stage.

In the frog, it seems fairly well established from the work of Cooper (1946,
1948, 1950) and of Spar (1953) that new antigens are found in the blastula and
gastrula stages. Ten Cate and Van Doorenmaalen (1950) and Flickinger e¢ al.
(1955) have shown that there is good correlation between the time of appearance
of lens antigen and of the lens itself. Flickinger and Nace (1952) have detected
new antigens in the tail-bud stage.

Clayton (1953) has further demonstrated that between the blastula and
gastrula stage, before neurulation, and between neurulation and the formation of
the tail-bud, a synthesis of antigenic material occurs. Ectoderm and archenteron
roof contain fractions specific to themselves as well as having common antigens.

1 Supported by grant E-1365 from the U. S. Public Health Service.
* Part of this work was carried out in the Department of Biology, Colby College, Water-
ville, Maine.

451

452 MELVIN SPIEGEL

A word of caution seems warranted when considering these results. It
should be mentioned that Tyler (1957) has pointed out that in investigations of
this kind (in which antisera produced against saline extracts—after suitable ab-
sorptions—have been used to detect changes in antigens) there is the uncertainty
as to whether or not a particular antigenic structure remains associated with a
saline-soluble constituent and as to whether or not it is available for reaction in
specific absorption procedures. The appearance of new antigens or the loss of
old antigens may represent a change in solubility, rather than the synthesis or de-
struction of an antigen. The above evidence, therefore, which has been repre-
sented as a synthesis of new material may rather reflect changes in solubility of
pre-existing substances. This point will be more fully considered in the discussion
of the results presented in this paper.

The study to be reported here involved the use of the technique of zone electro-
phoresis to follow the changes occurring in the proteins of developing embryos
and to compare these changes with adult organ proteins.

MATERIALS AND METHODS
I. Preparation of extracts

A. Developmental stages. Embryos of the frog, Rana pipiens, from the un-
fertilized egg through stage 21 (Shumway, 1940) were used. Eggs were ob-
tained by pituitary injection of large adult females and fertilized by the usual method
of stripping directly into a suspension of macerated testes in 10% Holtfreter’s
solution, pH 7.8. The embryos were then washed three times with sterile 10%
Holtfreter’s solution, pH 7.8, and incubated at 19-20° C. until the desired stage
was attained. Extracts were prepared from embryo populations in which at least
95% of the embryos were of the same stage and developing normally. Regardless
of the stage at which the extract was made, 100-200 embryos of each population
were always carried through stage 22 and if subsequent development of these con-
trols showed more than 5% of the embryos to be abnormal, the extract was dis-_
carded. For the unfertilized egg extracts, 100-200 eggs were fertilized and incu-
bated as described above. If more than 5% of the embryos were abnormal, the
extract was discarded.

The jelly of individuals of the desired stage was then removed by the papain-
thioglycolate method of Spiegel (1951). After washing five times with sterile 10%
Holtfreter’s solution, pH 7.8, 500-1000 jelly-free embryos or eggs were allowed
to settle out, under gravity, in a 50-ml. cellulose nitrate centrifuge tube. Excess
supernatant fluid was removed by aspiration, leaving only the small amount of
solution that was trapped in the interstices between individuals. The loosely
packed embryos were frozen-thawed, homogenized in a glass homogenizer, and
centrifuged at 23,000 g for 10 minutes. The clear supernatant fluid minus the
fat layer was removed by aspiration and stored at — 20° C. until used. All de-
velopmental stages were treated in identical fashion. The temperature throughout
the extraction procedure was 4° C.

B. Adult organs. The following organs were dissected from adult frogs of
both sexes: brain, gastrocnemius muscle, and small intestine. These organs were
then extracted by the identical method used for developmental stages. Adult blood

PROTEIN CHANGES IN DEVELOPMENT 453

was obtained by cardiac puncture from adult male and female frogs, diluted with
an equal amount of 10% Holtfreter’s solution, pH 7.8, centrifuged at 23,000 g
for 10 minutes, the supernatant fluid, strongly colored by hemoglobin, removed
by aspiration, and stored at — 20° C. until used.

II. Electrophoresis

Before electrophoresis, the extracts were thawed and dialyzed in the cold vs.
three changes of the electrophoretic buffer, Veronal, pH 8.6, »=0.05, for 24
hours. The dialysates were centrifuged at 23,000 g for 10 minutes and 0.02 ml.
of supernatant fluid was applied to filter paper strips (Spinco Part No. 300-028).
The 8-strip Spinco Durrum-type electrophoresis cell and the Spinco Model R
regulated power supply were used. Electrophoresis was carried out at room
temperature using 25 ma constant current for 6 hours. Under these conditions
rather remarkable reproducibility was obtained. Each stage or adult organ extract
was prepared a minimum of five times and each preparation subjected to at least
three separate runs, making a total of at least 15 electrophoretic runs per type of
extract.

III. Staining and scanning of paper strips

At the completion of the run, the papers were dried at 115° C. for 30
minutes and stained by the rapid brom phenol blue procedure for proteins described
by Durrum (1958). The strips were rinsed in methanol for 6 minutes, stained
for 30 minutes in 0.1% brom phenol blue in methanol, followed by three rinses
in 5% acetic acid of 6 minutes duration per rinse. They were then blotted, dried
at 115° C. for 15 minutes, exposed to concentrated ammonium hydroxide vapor
for 15 minutes, and scanned with the Photovolt Densitometer Model 525 (using
a 505-millimicron broad band filter) coupled to the Varicord Variable-Response
Recorder. For glycoprotein identification, the periodic acid-fuchsin sulfite method
of Koiw and Gronwall (1952), as described by Durrum (1958), was used. The
strips were scanned as described above.

RESULTS

A. Developmental stage extracts

The brom phenol blue protein pattern obtained with unfertilized egg extracts
is shown in Figure 1. Seven cation-bands and two anion-bands were noted. We
chose to use the term bands rather than proteins since the absence of additional
data such as electrophoresis at pH’s other than 8.6, salt fractionation, etc., did
not allow us to conclude that each band represented a single protein. Indeed,
careful examination of the densitometer records often revealed that a major band
was composed of two or more small peaks. For example, the cation-band labelled
B in Figure 1 was actually composed of three smaller peaks, readily reproducible
from run to run. In addition to these stained bands there was an additional band,
labelled fl, which was not stained by brom phenol blue but which charred as a
result of drying at 115° C. for 30 minutes and appeared as a yellow-brown band
in the unstained strip. It was the only band which fluoresced under ultraviolet
illumination. It did not correspond to any of the brom phenol blue-stained bands

454 MELVIN SPIEGEL

feet BSecU SEs Ssa53sseeassadeece O# $5SES00 EEDESUIGG0 550350063 100EIERU01 ITESIEE

SHEE EEE G Gf asp eft He ut
E585 Egss: anes 75ss25=55= ze EAN isa as UNF. ATE Ti
=S55 SSESE55: Sn a i Hh
Zbeg Spas S25! S55:'252555 fog super ldeacEsr
Sead goss eae! S0se \aESeGd eee , 1E sai ten tne

BES [F5H! SLEEEGED: PaENC BH ioe etescdsbeess cen lUbssle) 2

= ESE: a EL EES== face pete laos giaee MAREE )

HH Bees payee PRE woe ii =

Has NN
Seeseeene.’
ssstdes WiaERA‘E

MTU TT
ay

|

a ill
val
it
i

pppey bea!
init
ba | 43gttt
Titi |
tit ;
|

SHEER EEREES ie
epee es
a
bane me NEE owe A es Te oe

ee Pao Wis see

sls I i ae ES SE
tat ey ert Heres ee Le

a Ee
ye eae
ee ee
ree iad

ae UEEZEY alone Senss sons TEC: me 1 Lad
SH ees ek
See! S oes \eoee 25g WUS550 2552 Pare
A ~—
= SAU ares oe FAAS EL
SE= aun Ent SESrere Sg i: Fe ah
Se eae
PREGA zd a
TREE Si mas ee,
jonas eae ee
aS ae (gel ie
a2 ES Bete GESEGE a Hane

: Sos S/S eae
oe

RE eee PES
RV Ach A HPT Pat

Ficure 1. Densitometer tracings of brom phenol blue-stained paper strips of develop-
mental stage extracts after electrophoresis. Ordinate labelled 0 represents point of application.
Values 0-100 on the 0 ordinate represent the relative density of brom phenol blue stain. To
the right of the 0 ordinate are the anions; to the left are the cations. Abscissa units represent
relative distance away from the point of application.

PROTEIN CHANGES IN DEVELOPMENT 455

Fee O Vv SEREESDEDEGUEGEEEEEEOEZEEECESEERRERECOETI UB
tt BOOBS Robes ESSE 5EGSS S220 RR UEN TESESE? Rit sae a he PELL

> / ee
NE 2 S05 G5e52 FSS85

. uae

aiai casa 127104 eset ETERS
SAT

BETTY
Mitt
st

\

| rat

—|== Farag ores ETEEd a Vas57 soca" cad LE

mE BEE slit EF A. SBSHE/ CESEESREEERTEGS TEgEY NLS

pH FBS \ee54 /RERESEREEI SPSTATGET

bED ESEGN (2530829 veered ETE) Ea! #LEGE

eed EREZ4 1 HANG 0A40025 a9 FEES VON) 080050 1500000000 2050068- 0580003 (2HC0
pEED SS TS5EPSJBGE3 20°23 PESEREGESASEELSENG!

2) Ea zi=Ee
au GES) 3280 eens 0 ETE 05d LO

eo HS SERESERE ESE oa be
SEESES SPREE ROSE RYEEa tN nal Pe ueterite
B55028 ee

‘
sesfsees

ry

ie it |
Nae etl
hi @ Agi Baas ffs
Dea Aid & i eeae ae Pare ae Ee |
SE ia A Ss? oe
‘i A ee) 1
SOE) Oe 2 PP A

Cia a ao ; a Seas ma

2 Epes Soa a eee

:
connel Hi

aa var nthe t
_
ithe : | | =:

Ficure 2. Densitometer tracings of brom ee blue-stained paper strips of
developmental stage extracts after electrophoresis. Symbols as in Figure 1.

456 MELVIN SPIEGEL

SSSSS5
= =
5 =
SSSESSSE52
seesecces
Se 2SSssesg=5
Se 52:23 6222
aie =
Sceseesoees
Pepe ee
SSSSssesam0
3 SSS5eSaaes
BER er ers =
= =
EERE ESE =

100 = Sa Soca os os os os os os oo os

soe so
=ss=5

S555: 25522 S5s== 555
jsreee F
Seaitaais

measeee
ifs

He

saa ao- mae ove os ees om

BE SEEEEE EESEEEEELE Bt gec2222=2222s222 z eeeetity |b HEE saaeea0sseeceses0e==36 aii
BESSSS=SSSS==5== or Pea seaseas=s sesssses=
giitetsteteatetes yet BRAIN oe
SSi==== SSSSSS=es=5=== SESSSSee=T=eees fae =2
seefeesssunssscettfeeesrsse ce surat
si uss En asta
S55 sassssss==S====s== gegascaeaesaseese
PEEELP SEES
SES5=25 s55s2 sees
23
== SScSe
zene Se
ZEEE =! 1525/55: E555 s5a= 052s! 258" SNEED ee
F ==! rsE5! [eres ara! sora2iars!=or57 594 eee REPEEEE
asee z gas snssssoss2aese= feeee sees: essa! Seem Sey Eosaz2=. S220

sees sess CgeaE Eieea giz: ses: 75222558: 295505227 ree | Be 225q2552
=a50 f27500E=as SSggEEEE: =! :5u570510/500 7820/55 4#252 fesGd 12554 1eUcfsacaezéfocesc== "=! S55

Ficure 3. Upper: Densitometer tracings of ee acid-fuchsin sulfite-stained paper
strips of developmental stage extracts after electrophoresis. Middle and bottom: Densitometer
tracings of brom phenol blue-stained paper strips of adult organ extracts after electrophoresis.
Symbols as in Figure 1.

-4-449
1! 1
: Le

PROTEIN CHANGES IN DEVELOPMENT 457

and its limits were demarcated in the figure by the two lines on either side of the
label ff in the figures.

The patterns obtained with fertilized egg extracts and stage 10 embryo extracts
are shown in Figure 1. It was readily noted that following fertilization there was
a decrease in the concentration of the H band which was perhaps comparable to the
“Mirsky protein” found in the sea urchin egg. There was also a quite noticeable
progressive increase during development in the concentration of the cation-band
C. All other bands remained essentially constant in concentration through stage
15 (Figs. 1 and 2).

By stage 21, bands A, F and G were no longer detected (Fig. 2). The presence
of band J was questionable and at best was in very low concentration and migrated
as a broad band rather than as a sharply defined entity. The over-all protein
concentration, under the conditions of extraction and electrophoresis employed,
was markedly reduced. Band H, however, increased in relative concentration and
by stage 21 approximated the relative concentration found in the unfertilized egg.

2S 0 SS es ae

See feet SKEL. MUSCLE teeee
zj/J]04/g1 21112121 lgzed i] :224211 21242112552 HEH RIEHIHIIH
23227 HF255F2H5:2355552 25 (SSS SSS SS

Sod SepEs Ieoa ::|53] HEELLEUS] Hust! HEEHIEESEHEE 2TH
EH Eepey ges 12522 ruses poset ested LQ@lA fesse 232 i2=ss Eeegpree
= FZ ===

i
i

tH
|

{hit

\
\

BAH HTH

mnt

we
J

——
2
==
7 =
<= #4 .
ss - pers eee
335-5 —

Ficure 4. Densitometer tracings of brom phenol blue-stained paper strips of adult organ
extracts after electrophoresis. Symbols as in Figure 1.

458 MELVIN SPIEGEL

Staining with the periodic acid-fuchsin sulfite method for glycoproteins revealed
the presence of 5 stained bands, all migrating as cations (Fig. 3) in the fertilized
egg extract (two hours after fertilization). One of these bands, K’, did not cor-
respond to any of the brom phenol blue-stained bands. The remaining bands,
B’, C’, D’ and E,’ corresponded in position to the B, C, D and E bands, respectively,
of the brom phenol blue-stained strips.

The periodic acid-fuchsin sulfite pattern of the unfertilized egg extract was
identical with the above results. Through stage 15, however, there was a progres-
sive decrease in the stainability or concentration of these bands and by stage 21, the
presence of carbohydrate was no longer detected. Reasons for labelling these
bands with primes are considered in the discussion section.

B. Adult organ extracts

Patterns obtained with extracts of adult organs are demonstrated in Figures
3 and 4. These patterns revealed that relatively few of the ion-bands present in

TABLE I

Brom phenol blue-stained proteins of adult organ extracts present in unfertilized
egg extracts

Band
Adult organ extract
ft B A H di
Intestine = =e os = Se,
Brain wa = oy ae ae
Skeletal muscle + = = ae eu
Blood me au ae ae re

+ Indicates presence of band.
— Indicates absence of band.

developmental stages were detected in adult organ extracts. The majority of the
cation-bands were not detected; in intestine and brain extracts, none were de-
tected; in skeletal muscle only the ff band was detected; in the blood extract only
the A and B bands were present. All of the anion-bands were detected, although
not all were present in each adult organ extract.

Bands labelled with a question mark in the figures do not necessarily represent
new or different proteins but rather represent pre-existing proteins which were un-
masked by the changes in concentration of substances having almost identical
mobilities. In addition, the anion-band labelled with a question mark present
in the blood extract probably represents the serum proteins masked by hemoglobin.
In the unstained strip the band was red in color.

The results obtained with adult organ extracts are summarized in Table I.

DISCUSSION

It is apparent from the results summarized in Table I that although the majority
of the proteins extracted from adult organs are present in the unfertilized egg, the

PROTEIN CHANGES IN DEVELOPMENT 459

protein patterns of adult organ extracts do exhibit differences. There are some
components present in more than a single organ but others are apparently confined
to a single organ. For example, bands H and J are present in both skeletal
muscle and in brain extracts; the fl band, on the other hand, is present only in the
skeletal muscle extract.

These patterns suggest that all, or most, of the proteins found in adult organs
are first found in the unfertilized egg. As development proceeds some of the cells
lose their ability to manufacture one or more particular proteins. Correlated with
this loss of ability to synthesize a particular protein is a corresponding morphologi-
cal differentiation. Differentiation, therefore, could involve a loss of the ability
to synthesize a protein rather than the synthesis of a new protein. The work
of Ebert (1953, 1955) on the distribution of cardiac myosin in the chick embryo
is in support of this hypothesis. Ebert demonstrated that early in development
cardiac myosin is rather uniformly distributed throughout the blastoderm. As
development proceeds, the ability to synthesize cardiac myosin is eventually re-
stricted to those cells located in the two heart-forming regions.

An alternative to the above explanation is equally probable. The differences
observed in the protein patterns of adult organs and those of the embryo may
reflect changes in the solubility of the proteins of the unfertilized egg or of adult
organs rather than the loss or gain of any synthetic capacity. As a result of this
line of reasoning, one could assume that the same proteins are present in the adult
brain as in the unfertilized egg, but some of these proteins are insoluble, perhaps
under the conditions of extraction, in the adult organ. At present, the more
probable hypothesis cannot be determined. It is likely that both processes are
involved in differentiation. The solubility and location of the proteins detected in
this investigation are unknown. Whether we are dealing with the soluble proteins
of the cytoplasm or those proteins soluble in the small amount of 10% Holtfreter’s
solution (and ultimately in the electrophoretic buffer) used for extraction is not
readily apparent. Further studies are necessary to determine location, solubilities,
etc. of these proteins. Why the present work does not show the appearance of
“new” proteins (proteins insoluble in the unfertilized egg but soluble at later
stages)—if the latter hypothesis is correct—cannot be answered at the present
time.

It must be emphasized that the identification of proteins in this study has been
made solely on the basis of mobility in an electric field. It is possible that new and
different proteins from those of the unfertilized egg are being synthesized through-
out development. These new proteins could have the same or closely matching
mobilities as those of the early embryo, would therefore not be detected as new
proteins, and could perhaps give the impression of a static population of protein
types which does not exist. Further characterization of these proteins by other
biochemical techniques will resolve this question. Of interest is the observation
that, under these conditions, the majority of the embryonic proteins are cations;
those of the adult are primarily anions. It is possible that those cation-bands
which are lost during development represent the yolk proteins and are utilized
as an energy source.

If the over-all hypothesis be true, that differentiation can be brought about by,
or is associated with, the loss of the ability to synthesize a pre-existing protein

460 MELVIN SPIEGEL

and/or changes in solubilities of pre-existing proteins, one must consider the
ever-increasing body of literature apparently demonstrating the presence of new
antigens in development. All of these papers, to our knowledge, have dealt
solely with saline-soluble antigens. The appearance of a new antigen in develop-
ment may rather represent a change in solubility rather than a synthesis of new
antigenic material. For example, it is entirely possible that an antigen present
in the blastula stage is saline-insoluble but in the gastrula stage is soluble in saline.
Such an antigen, after the usual absorptions, etc. have been carried out, would
then be described as a new antigen arising during gastrulation rather than as a
change in solubility of a pre-existing substance. The demonstration by Markert
and Mgller (1959) of multiple forms of enzymes—the isozymes—which are tissue,
ontogenetic, and species-specific can also be explained, rather simply, on the basis
of changes in solubility.

The recent paper by Solomon (1959) offers a striking demonstration of
changes in location, and therefore of solubility, of a particular enzyme during de-
velopment. It was shown that the mitochondrial glutamic dehydrogenase of the
chick embryo liver increased in activity after the twelfth day of incubation and this
rise was followed by a sudden drop of glutamic dehydrogenase activity in the
supernatant fluid (0.25 M sucrose) after 15 days of incubation. Supernatant fluid
glutamic dehydrogenase activity is four times that of mitochondria at 7 days’
incubation ; 6 times at 12 days’ incubation; 14 times at 20 days’ incubation. The
thesis that changes in the intra-cellular distribution of enzymes may occur during
differentiation has been put forth by Krahl (1950), and Gustafson (1954).

The work of Schechtman (1955) and Nace (1953) has further demonstrated
that the serum proteins of the chick are present in the unincubated egg and are
probably transferred from the maternal circulation to the egg.

The results obtained using the periodic acid-fuchsin sulfite staining technique
for glycoproteins remain to be considered (Fig. 3). The band K’ does not corre-
spond to any of the brom phenol blue-stained bands and is probably a polysaccha-
ride. The fact that the remaining four cation-bands, B’, C’, D’ and E’, corre-
spond in position to the B, C, D and E bands, respectively of the brom phenol
blue-stained strips, coupled with the observation that the periodic acid-fuchsin
sulfite-stained bands were detected in early stages but not in stage 21 extracts, has
led to the following interpretation.

It is possible that these bands represent a polysaccharide migrating at the same
rate as proteins composing the band, or they may represent a glycoprotein. By
stage 21, synthesis of these polysaccharides or glycoproteins has ceased or their
solubility has changed. It would therefore be justifiable to label these bands as
distinct from other proteins present in the band. It is also possible that the pe-
riodic acid-fuchsin sulfite technique is less sensitive than the brom phenol blue
technique. Due to the decrease in protein concentration of the stage 21 extracts
(Fig. 2) a positive brom phenol blue test but negative polysaccharide test results.
Which of these interpretations is correct remains to be determined. The labelling
of these bands with prime letters indicates the possibility of their being discrete
entities.

It is realized that the present paper by no means offers conclusive proof, but
merely suggests that differentiation may, in part, involve the loss of ability to
synthesize a particular protein and/or changes in the solubility of pre-existing

PROTEIN CHANGES IN DEVELOPMENT 461

proteins. The intra-cellular locations, concentrations, tissue distributions, and
further biochemical characterizations of these proteins are being carried out in
this laboratory at the present time.

SUMMARY

1. Extracts of developmental stages and of four adult organs of the frog,
Rana pipiens, were examined by the technique of paper electrophoresis.

2. Seven protein cation-bands and two protein anion-bands stained with brom
phenol blue were detected in extracts of developmental stages from the unfertilized
egg through stage 15. By stage 21, two of the cation-bands were no longer de-
tected.

3. Examination of the adult organ extracts revealed organ- specie proteins
in addition to proteins common to all organs tested.

4. The results were interpreted in terms of the hypothesis that differentiation
involves, in part, the loss of ability to synthesize a particular protein(s) and/or
changes in the solubility of pre-existing proteins.

PTE RAO RE Chie

CLAyToN, R. M., 1953. Distribution of antigens in the developing newt embryo. J. Embryol.
Exp. Morphol., 1: 25-42.

Cooper, R. S., 1946. Adult antigens (or specific combining groups) in the egg, embryo, and
larva of the frog. J. Exp. Zool., 101: 143-172.

Cooper, R. S., 1948. A study of frog egg antigens with serum-like reactive groups. J. Exp.
Zool., 107: 397-437.

Coorrr, R. S., 1950. Antigens of frog embryos and of adult frog serum studied by diffusion
of antigens into agar columns containing antisera. J. Exp. Zool., 114: 403-420.

Durrum, E. L., 1958. Paper electrophoresis. Jn: A Manual of Paper Chromatography and
Paper Electrophoresis, by R. J. Block, E. L. Durrum and G. Zweig. Academic
Press Inc., Publishers, New York, N. Y. Pages 489-667.

Epert, J. D., 1951. Ontogenetic change in the antigenic specificity of the chick spleen.
Physiol. Zéol., 24: 20-41.

Expert, J. D., 1953. An analysis of the synthesis and distribution of the contractile protein,
myosin, in the development of the heart. Proc. Nat. Acad. Sci., 39: 333-344.

Esert, J. D., 1955. Some aspects of protein biosynthesis in development. Jn: Aspects of
Synthesis and Order in Growth, D. Rudnick, editor. Princeton University Press,
Princeton, N. J. Pages 69-112.

FLICKINGER, R. A., anp G. W. Nace, 1952. An investigation of proteins during the develop-
ment of the amphibian embryo. Exp. Cell Res., 3: 393-405.

FLIcKINGER, R. A., E. Levi anp A. E. Situ, 1955. Some serological experiments relating
to the embryonic development of the lens. Physiol. Zool., 28: 79-85.

GUSTAFSON, ai 1954. Enzymatic aspects of embryonic differentiation. Intern. Rev. Cytol.,
3: 277-327.

Harpinc, C. V., D. Harpinc anp P. PertmMann, 1954. Antigens in sea urchin hybrid embryos.
Exp. Cell Res., 6: 202-210.

Harpinc, C. V., D. Harpinc anp JI. W. Bampercer, 1955. On cross fertilization and the
eae of hybrid embryonic development in echinoderms. Exp. Cell Res., suppl.,
3: 181-187.

Koiw, E., anp A. Gronwa tt, 1952. Staining of protein-bound carbohydrates after electro-
phoresis of serum on filter paper. Scand. J. Clin. and Lab. Invest., 4: 244-246.

Kraut, M. E., 1950. Metabolic activities and cleavage of eggs of the sea urchin, Arbacia
punctulata. Biol. Bull., 98: 175-217.

MarkKert, C. L., ann F. M@tier, 1959. Multiple forms of enzymes: tissue, ontogenetic, and
species specific patterns. Proc. Nat. Acad. Sci., 45: 753-763.

462 MELVIN SPIEGEL

Mirsky, A. E., 1936. Protein coagulation as a result of fertilization. Science, 84: 333-334.

Monroy, A., 1950. A preliminary electrophoretic analysis of proteins and protein fractions
in sea urchin eggs and their changes on fertilization. Exp. Cell Res., 1: 92-104.

Nace, G. W., 1953. Serological studies of the blood of the developing chick embryo. J. Exp.
Zool., 122: 423-448.

Nace, G. W., 1955. Development in the presence of antibodies. Annals N. Y. Acad. Sci.,
60: 1038-1055.

PERLMANN, P., AnD T. Gustarson, 1948. Antigens in the egg and early developmental stages
of the sea urchin. Experientia, 4: 481-482.

PERLMANN, P., 1953. Soluble antigens in sea urchin gametes and developmental stages.
Exp. Cell Res., 5: 394-399.

ScHECHTMAN, A. M., 1955. Ontogeny of the blood and related antigens and their significance
for the theory of differentiation. Jn: Biological Specificity and Growth, E. G. Butler,
editor. Princeton University Press, Princeton, N. J. Pages 3-31.

SHuMWway, W., 1940. Stages in the normal development of Rana pipiens. I. External form.
Anat. Rec., 78: 139-147.

SoLtomon, J. B., 1959. Changes in the distribution of glutamic, lactic and malic dehydrogenases
in liver cell fractions during development of the chick embryo. Dev. Biology, 1:
182-198.

Spar, I., 1953. Antigenic differences among early developmental stages of Rana pipiens. J.
Exp. Zool., 123: 467-497.

SpreceL, M., 1951. A method for the removal of the jelly and vitelline membrane of the egg
of Rana pipiens. Anat. Rec., 111: 544.

SpreceL, M., 1955. The reaggregation of dissociated sponge cells. Annals N. Y. Acad. Sci.,
60: 1056-1078.

Ten Cate, G., anpD W. S. Van DoorENMAALEN, 1950. Analysis of the development of the eye-
lens in chicken and frog embryos by means of the precipitin reaction. Konink. Med.
Akad. Wetenschap., 53: 894-909.

Tyier, A., 1957. Immunological studies of early development. Jn: The Beginnings of Em-
bryonic Development. Amer. Assoc. Adv. of Sci., Washington, D. C. Pages 341-382.

TyLer, A., AND J. W. BrookBaNnk, 1956. Antisera that block cell division in developing eggs
of sea urchins. Proc. Nat. Acad. Sct., 42: 304-308.

meow SPECIFICITY IN EMBRYONIC AND’ ADULT
CYMATOGASTER AGGREGATA STUDIED BY
SCALE. TRANSPLANTATION

EDWARD L. TRIPLETT AND SUSANNE BARRYMORE

Department of Biological Sciences, Umversity of California, Santa Barbara,
Goleta, California

The method of scale transplantation first used by Mori (1931) has proved to
be of great value in studies on tissue specificity in fishes. Goodrich and Nichols
(1933) and Hildemann (1956, 1957a, 1957b) have shown that, whereas autografts
are always successful, homografts are invariably rejected by the host. These
studies have established that the immune responses to homografts so profusely
studied in mammals and birds are also observable in the goldfish (Carassius
auratus ).

The purpose of this investigation was to extend our knowledge of homograft
reactions in fishes. Answers were sought to the following questions: (1) To
what extent are adults of Cymatogaster aggregata (Gibbons) immunologically re-
active toward scale homografts? (2) Are adults of this species capable of being
sensitized to scale homografts (1.e., will secondary, tertiary, etc., grafts be more
rapidly rejected than the primary graft)? (3) Does the reactivity of embryonic
and immature C. aggregata toward scale homografts differ from that of the adults?
(4) Assuming affirmative answers to the above questions, can a sensitized, pregnant
female transfer this homograft sensitivity to intra-ovarian embryos?

MATERIALS AND METHODS
Experimental animals

C. aggregata is a marine embiotocid fish which is distributed over the west
coast of North America from southern Alaska through Baja California (Tarp,
1952). Like all embiotocids this species is viviparous, the young spending about
five months in the hollow ovary and increasing in size from 0.25 mm. (fertilized
egg) to about 42 mm. (newborn) in standard length (Eigenmann, 1889; Triplett,
unpublished data). Animals used in this investigation were collected by angling
from Goleta pier, Goleta, California.

Methods

The method of grafting was essentially similar to the one devised by Hildemann
(1957b). Adult fish were anesthetized in tricaine methane sulfonate (MS 222),
1/5000 dissolved in one per cent sodium chloride (MS 222 forms a white pre-
cipitate in sea water). Survival was greatest when the anesthetic was maintained
at about 10° C. Grafting was done simply by removing a scale from the ventral
surface of the host and replacing it with a scale from the dorsal surface of the

463

464 EDWARD L. TRIPLET? AND SUSANNE BARRY MORE

same fish or another fish. The transplant is easily distinguished since the scales
of the dorsal surface of the body are heavily pigmented in contrast to those of the
ventral surface which contain few or no melanophores. Grafted adults were
maintained in aquaria provided with running sea water. They were fed ad
libitum on pieces of mussel (Mytilus califormianus).

Embryos were obtained by caesarian section using aseptic technique. They
were then transferred to fingerbowls containing a modification (Triplett, unpub-
lished data) of sterile White’s medium (1954) and kept there until time was
available for grafting. The embryos were anesthetized in 1/10,000 MS 222
dissolved in the aforementioned solution. An incision was then made in the
flank skin with iridectomy scissors. Into this was placed either a homograft
consisting of a sterile, adult, trimmed scale, or an autograft consisting of a piece of
pigmented skin. The scales of embryos, when present, are too small and lightly
pigmented to be used as autografts. Adult scales were sterilized by immersion
for one minute in 0.004 M mercuric chloride dissolved in modified White’s medium
and washed with sterile medium before grafting. The scale remained alive and
apparently healthy after such treatment. Embryos were grafted and maintained at
17° C. in a sterile culture chamber.

All host fishes, embryonic and adult, received a homograft and most fishes
received an autograft. After it was established that autografts would survive and
become healthy, some of the newborn in subsequent experiments were not given
autografts in order to avoid this additional injury. In some cases reciprocal
grafting of pairs of fishes was done. In others a single donor contributed scales
to between three and eleven hosts.

The principle criterion for assaying the grafted scales was the condition of the
scale melanophores. The melanophore is a single pigment cell of a radially
dendritic pattern. The melanophores are usually somewhat separated from one
another and readily visible under a dissecting microscope. These pigment cells
contain many granules of melanin which respond to physiological changes to spread
out into the cell’s extremities, or to aggregate in the center of the cell. Ina
typical situation the melanophores of autografted scales, after an initial healing
period, remained healthy and completely pigmented for an indefinite period of
time. After a given time homograft melanophores began to undergo degenerative
changes. The pigment aggregated as a black dot in the center of the melanophore
and over a variable interval of time proceeded to fragment. The melanin granules
were eventually phagocytized. In a few cases an inflammatory reaction similar
to that described by Hildemann (1957b) and Goodrich and Nichols (1933) was ob-
served. The time required for fragmentation of all the melanophores of the trans-
planted scale is here termed the reaction time.

RESULTS
Adults

The adults of this species, like many pelagic fishes, are difficult to maintain
in a laboratory without running sea water. Therefore the temperature, of neces-
sity, was that of the ocean at that particular time of the year. In order to measure
the effect of temperature on the reactivity of animals toward homografts one
series was performed between March and July (average temperature = 17° C.),

PISSUE* SPECIFICITY IN; PERCH 465

and another series was done between August and October (average temperature
esi 20? -C.).

In these series a single donor contributed scales to either three, four or five
hosts. In all, 37 hosts and eight donors survived long enough to yield data. The
hosts received as many as six successive grafts from the same donor, new grafts
being placed-as soon as fragmentation of the melanophores of the previous graft
was complete. As mentioned above all the autografts placed on hosts remained
alive and healthy for the duration of the experiment (in some cases more than
three months).

Both autografts and homografts underwent an initial healing phase during
which time the melanophores that had been injured during the operation frag-
mented and were phagocytized. After the third post-operative day degenerating
melanophores could no longer be found in the autografts, but progressive loss of
melanophores continued in the homografts until all had disintegrated (reaction
time). The reaction time for homografts is shown in Table I. Here it can be
seen that the average reaction time for grafts at 17° C. is 7.0 days. The reaction
time of subsequent grafts decreased progressively until at the fourth grafting it
averaged 3.5 days. Any further grafting was not effective in increasing homograft
sensitivity.

It can also be seen in Table I, in confirmation of Hildemann’s (1957b) findings,
that increased temperature of the external medium has the effect of decreasing
the reaction time. It was also observed that at the higher temperature the primary
homograft has a more pronounced effect in producing sensitivity to subsequent
grafts. These observations are paralleled by those of Cushing (1942) who found
that the ability of fishes to produce antibody against injected erythrocytes or
spermatozoa is influenced by temperature in such a way that circulating antibody
appears sooner at higher than at lower temperatures.

Some evidence has been gathered which indicates (as might be expected)
that fishes of this species, comprising a natural population of unknown interrelation-
ships, bear some but not all scale tissue factors (antigens?) in common. Six fish
of the series discussed above, which had received between four and six successive
homografts, were each given a graft from a donor that had not previously been
used. The average reaction time for this series was 5.6 days. The value for a
primary graft would have been 7.0 days (average), and the value for a fifth, sixth
or seventh set reaction would have been 3.5 days (average). Since the realized
value is in between these two, the statement that each fish bears distinctive as well
as common homograft factors seems justified. Medawar (1946) has made similar
observations on rabbits.

Significance of reaction time

A small series of operations was performed in an effort to determine the
relationship of the reaction time (melanophore disintegration time) to the com-
plete destruction of the living tissue of the homograft. The method, first used
by Hildemann (1957b), consisted of returning homografts to the donor after
variable lengths of time. Failure of the graft to re-establish itself in the original
donor indicated irreparable damage or death.

In this series five pairs of adult fish were used. Each donor contributed 10

466 EDWARD L. TRIPLETT AND SUSANNE BARRYMORE

scales to its partner, and the grafted scales were returned to the donor at daily
intervals. It was found in all five cases that scales returned to the host on or before
the third day after the original grafting became re-established in the donor. Those
returned on or after the fourth day did not survive. These data lack statistical

TABLE |
Reaction time in days for scale grafts to adult fish

Operations performed between March and July—Average water temperature 17° C.

Specimen no. ist graft 2nd graft 3rd graft 4th graft 5th graft 6th graft

1-1 11
2 | 7
3 9 9 6 3 3 3
4 ? 8 3 3 3 4
5 — 4

2-1 9 11
2 11 iW
3 11 11
4 ti 9
5 fe 10 4 4 3

3-1 7 9
2 6
3 6 6 4 3
4 6 5 6 5
5 9 4

4-1 11
2 6 5)
3 Le 6 3 3 4
4 iM) 5 6 3 4 3
5 af 5 6 4 4

5-1 6 7 4
2 7 m
3 6 7 3
4 6 8 4
5 6 5 5

Average 7.0 y fe 4.5 3.5 320 38

Operations performed between August and October—Average water temperature 20° C.

1B-1 2
Z 5
a 5 3
+ 7

2B-1 5 3
2 r ( 3
3 5
4 5 a
5 ‘4

4B-1 7 3
2 7 3
3 5 3

Average 5.8 4.0

EIssUE; SPECIFICITY IN} PERCH 467

force but suggest that homografts are no longer viable in a suitable environment
several days before the reaction time (7.0 days at this temperature). It is felt,
however, that the reaction time as used in the various experiments reported here
is valid since all other data indicate that it is a function of host homograft sensi-
tivity.

Embryos

Two series of experiments were done on intra-ovarian embryos. The hosts
of the first series consisted of six siblings averaging 29 mm. standard length, and
the hosts of the second series consisted of 11 siblings averaging 19 mm. standard
length. The hosts in each series received homografts from a single adult, the
operation and postoperative care being done in the manner described above.

Since the grafts had to be examined every day, and since the medium was rich
in organic materials some difficulty was experienced in maintaining asepsis. In
the first series, four animals lived 16 days, one lived 20 days and one lived 24 days.
In the second series eight animals lived 20 days, and three lived 30 days.

An initial healing period lasted between three and five days. During this time
damaged melanophores of both autografts and homografts were phagocytized.
After this time both autografts and homografts remained alive and in excellent
condition as long as the host lived. Since some of the embryos lived as long as
30 days there is a strong indication that intra-ovarian animals are incapable of
eliciting the homograft reaction.

Immature fishes

The animals used in this series were divided into two groups on the basis of
their parturition age. There survived in the first group 11 animals grafted one to
three days after birth, and in the second group seven fishes grafted 15 days after
birth. Some, but not all animals received an autograft consisting of a 0.25-mm.
piece of skin, and all received a scale homograft from an adult fish. Each donor
contributed to between three and seven hosts. All animals in this series were
cultured at approximately 17° C. in running sea water.

The autografts remained alive for the duration of the experiment. The
homografts were without exception broken down in a manner similar to that de-
scribed for the adult host homografts. It can be seen in the first two columns
of Table II that the reaction time for homografts is approximately in inverse pro-
portion to the age of the fish. The average reaction time for one- to three-day
fishes is 13.2 days, whereas that for 15-day fishes is 9.7 days (that for adults is 7.0
days at the same temperature).

Embryos from sensitized parents

These experiments were performed in an effort to discover whether or not
induced homograft sensitivity in parent fishes could be transferred to intra-ovarian
embryos. A total of 11 pregnant females was sensitized. Ten were given a single
scale homograft and one was given three successive homografts. The young
were born from 10 to 30 days after the parent had received the last graft. The
newborn animals each received a scale homograft one to three days after birth

468 EDWARD L. TRIPLETT AND SUSANNE BARRYMORE

TABLE II]
Host aged 1-3 days Host aged 15 days Host aged 1-3 days Host aged 1-3 days
Post-partum. Post-partum. Post-partum. Parent given Post-partum. Parent given
Parent not sensitized. | Parent not sensitized. 1 or 2 successive homografts. 3 successive homografts.
Specimen | Reaction Specimen Reaction Specimen Reaction Specimen Reaction
no. time no. time no. time no. time
S-6-59-la 16 S-6-59-7a 8 S-8-59-1a fe S-8-59-19a il
1b 16 7b 10 1b if 19b fl
te 16 UG 10 fe fl 19c ii
S-6-59-2 5 7d 10 id 9 19d 7
3 15 Te 10 le 9 19e 7
4 12. The 10 S-8-59-2 £3 19f 9
Gayl 7g | 10 S-8-59-7 6 19g 9
6b 1 S-8-59-9a 3 19h 9
6c 12 9b 9 19i 11
6d 12 S-8-59-10a 9
6e Lyf 10b M4
S-8-59-13a 8
13b 8
ise 10
S-8-59-14a 8
14b ‘yt
14c ct
14d 16)
14e 1
S-8-59-18a 8
18b 8
18c 10
18d 10
18e 10
S-8-59-21a 8
21b 8
21c 8
21d 11
S-8-59-22a 4
22b 6
Average A322 | 8.6 8.1
reaction
time
Totakbnon 714 7 30 9
specimens

from the same donor that had been used for the parent. The results are given in
the third and fourth columns of Table II. The results for the progeny of singly
grafted parents are tabulated separately from those for the progeny of the triply
grafted parent. By comparing column one with columns three and four it can be
seen that the reaction time for one- to three-day young from sensitized parents
(average reaction time = 8.6 and 8.1 days) is significantly shorter than that for
young of comparable age from unsensitized parents (average reaction time = 13.2

SESSUE: SPECIBICITY IN’ PERCH 469

days). It can also be seen by comparing columns three and four that the parent
with three successive grafts produced young with slightly greater sensitivity
(average reaction time = 8.1 days) than did the parents with a single graft (aver-
age reaction time = 8.6 days). The evidence leaves little grounds for doubt that
homograft sensitized parents can transfer this sensitivity to intra-ovarian young.

DISCUSSION

These experiments indicate a basic similarity between adult C. aggregata and
Carassius auratus (Hildemann, 1957b) concerning the homograft response. Both
species reject homografts, and both can be sensitized by successive grafting. In
both species the time for homograft rejection is shortened by increase in tempera-
ture. Barrymore (unpublished data) has made similar observations on adult
swordtails (Xiphophorus heller1) as have Triplett and Barrymore (unpublished
data) on Amphistichus argenteus and Hyperprosopon argenteum. Fin transplanta-
tions on xiphophorin fishes yield similar results (Kallman and Gordon, 1957).
There is good reason to suspect that these phenomena could be observed in teleosts
in general.

To the authors’ knowledge only one publication to date has dealt with tissue
transplantation in older embryonic or immature fishes. Humm et al. (1957) have
transplanted melanomas from adult platyfish-swordtail hybrids to embryos of both
species and their hybrids. These workers have found, in contrast to the present
authors, that homografts on embryonic fishes will not survive. This difference
in results could be explained in one or both of two ways. First, it should be noted
that the tissues grafted by Humm and his colleagues were not normal and should
not be expected to behave as such. The inability of the transplanted melanomas
to survive could be interpreted as a physiological deficiency other than immunologi-
cal incompatibility.

A second and more likely possible explanation must take into account the dif-
ferent developmental rates of platyfishes and swordtails as compared with C.
aggregata. The young of the former two species become free-living in about three
weeks (Hooper, 1943), whereas the latter spends about five months in the ovarian
cavity of the parent. A close correlation has been noted between the time of
parturition in mammals (e.g., Freund, 1930), hatching in birds (e.g., Murphy,
1914; Waterman, 1936) or metamorphosis in amphibians (Schwind, 1933, 1937;
Triplett, 1958) and the development of the ability to react immunologically to im-
planted or injected foreign materials. If our interpretation of Humm’s work is
correct those animals that were not killed by mechanical blockage, resulting from
a rapidly proliferating graft, retained the transplant until well after the normal
time of birth. The present experiments show that in C. aggregata homograft
incompatibility is expressed soon after birth. This is quite possibly also the case
in xiphophorin fishes used by Humm et al.

There has been considerable speculation as to whether or not circulating anti-
body is responsible for tissue incompatibility (reviewed by Snell, 1957). The
first clear proof of an immune reaction to homografts was published by Gorer
(1937) who found that tumor homografts could elicit the formation of circulating
antibodies capable of agglutinating donor red cells. More recent work (see Snell,
1957) has established that humoral antibodies are formed in a variety of species

470 EDWARD L. TRIPLETT AND SUSANNE BARRYMORE

against a variety of transplanted tissues. Santos et al. (1959) have shown that
the serum of graft-sensitized rodents, upon injection into other animals, can pro-
duce a passive heterograft sensitivity.

The present experiments on transfer of homograft sensitivity from parent to
intra-ovarian embryos can be explained best by assuming that circulating antibody
is produced against scale homografts. It is probable that antibody accumulates
in the ovarian fluid and is ingested by the embryos (no placental connection exists).
This hypothesized method of transfer is made more plausible by the fact that
Hemmings and Morris (1959) have been able to observe the absorption of antibody
from the gut in young mice. In addition, Brambel et al. (1959) have observed
that homologous antitoxin is absorbed from the uterine cavity to the fetal circulation
in the rabbit.

Hosoda et al. (1955) have reported experiments Of interest in this context.
They found that hens injected with horse erythrocytes produce antibodies and
transfer these antibodies to offspring.

SUMMARY

1. Embryonic and adult Cymatogaster aggregata have been subjected to scale
homotransplantations.

2. Homografts are rejected after a short period of time. The time required
for graft rejection is temperature-dependent. |

3. Increased sensitivity to homografts results from successive transplantations.

4. Intra-ovarian embryos used in these experiments were not capable of re-
jecting scale homografts.

5. Immature free-living fish rejected homografts, but more slowly than adults.

6. Pregnant females can transfer homograft sensitivity to intra-ovarian embryos.
It is hypothesized that circulating antibody is passed to the embryo via the ovarian
fluid and the embryonic hindgut.

LITERATURE’ CITED

BrAMBEL, F. W. R., W. A. Hemmincs ann C. L. Oaxktey, 1959. The relative transmission
of natural and pepsin-refined homologous antitoxin from the uterine cavity to the
fetal circulation in the rabbit. Proc. Roy. Soc. London, Ser. B, 150: 312-317.

CusuHin«, J. E., 1942. An effect of temperature upon antibody-production in fish. J. Immun.,

45: 123-126.
EIGENMANN, Cart H., 1889. On the development of California food fishes. Amer. Nat.,
23: 107-110.

FreuND, J., 1930. Influence of age upon antibody production. J. Immun., 18: 315-324.

GoopricH, H. B., AND Rowena Nicnots, 1933. Scale transplantation in the goldfish, Carassius
auratus. Biol. Bull., 65: 253-265.

Gorer, P. A., 1937. The genetic and antigenic basis of tumor transplantation. J. Pathol.
Bacteriol., 44: 691-697.

Hemmincs, W. A., AND J. G. Morris, 1959. An attempt to affect the selective absorption
of antibodies from the gut in young mice. Proc. Roy. Soc. London, Ser. B, 150:
403-409.

HitpeEMANN, W. H., 1956. Scale homotransplantation in goldfish (Carassius auratus).
Transpl. Bull., 3: 67-68.

HILDEMANN, W. H., 1957a. Early onset of the homograft reaction. Transpl. Bull., 3: 144-145.

HiLtpEMANN, W. H., 1957b. Scale homotransplantation in goldfish (Carassius auratus). Ann.
New York Acad. Sci., 64: 775-791. ;

PISSUE( SPECIFICITY IN“ PERCH 471

Hooper, A. F., 1943. The early embryology of Platypoecilus maculatus. Copeia, 4: 218-224.

Hosopa, T., T. KANEKo, K. Moct anp T. Ase, 1955. Transfer of antibodies in the hen to
her eggs and to the offspring. Bull. Natl. Inst. Agric. Sci. (Japan) Ser. G. Animal
Husbandry, 10: 125-129.

Hum, D. G., E. E. Crarx anp J. H. Hum, 1957. Transplantation of melanophores from
platy-swordtail hybrids into embryos of swordtails and platyfish and their hybrids.
J. Exp. Biol., 34: 518-528.

KALLMAN, Kiaus D., AND Myron Gorpon, 1957. Transplantation of fins in Xiphophorin
fishes. Ann. New York Acad. Sct., 71: 305-320.

Mepawar, P. B., 1946. Immunity to homologous grafted skin. I. The suppression of cell
division in grafts transplanted to immunized animals. Brit. J. Exp. Path., 27: 9-14.

Mori, YAsuMASA, 1931. On the transformation of ordinary scales into lateral scales in the
goldfish. J. Fac. Sci. Imp. Univ. Tokyo, 2: 185-194.

Murpny, J. B., 1914. Studies on tissue specificity. II. The ultimate fate of mammalian
tissues implanted in the chick embryo. J. Exp. Med., 19: 181-186.

Santos, GEorGE W., LEONARD J. CoLE AND RicHArD M. Garver, 1959. Antigenic stimuli for
transplantation immunity to rat bone marrow heterografts in lethally x-irradiated
mice. J. Immunol., 83: 66-73.

ScHWIND, JosEPH L., 1933. Tissue specificity at the time of metamorphosis in frog larvae.
J. Exp. Zool., 66: 1-14.

ScHWIND, JosEPH L., 1937. Tissue reactions after homoplastic and heteroplastic transplanta-
tion of eyes in the anuran amphibia. J. Exp. Zool., 77: 87-107.

SNELL, C. D., 1957. The homograft reaction. Ann. Rev. Microbiol., 11: 439-458.

Tarp, Frep H., 1952. A revision of the family Embiotocidae (the surfperches). Fish
Bulletin No. 88, State of Calif. Dept. of Fish and Game, Bureau of Marine Fisheries.
1-99.

TRIPLETT, Epwarp L., 1958. The development of the sympathetic ganglia, sheath cells, and
meninges in amphibians. J. Exp. Zool., 138: 283-312.

WaterRMAN, A. J., 1936. Heteroplastic transplantation of embryonic tissues of rabbit and rat.
Amer. J. Anat., 60: 1-25.

Waite, Pump R., 1954. The Cultivation of Animal and Plant Cells. The Ronald Press
Co., New York, p. 90.

SOME ASPECTS OF OSMOREGULATION IN EMBRYONIG VANS
ADU TOY MATOGAS DER AGG TG Ae aN) eae
EMBIOTOCID FISHES

EDWARD L. TRIPLETT AND SUSANNE D. BARRYMORE

Department of Biological Sciences, University of California, Santa Barbara,
Goleta, California

The Embiotocidae is a family of marine (with one exception) teleosts which
is somewhat unique in that all species possess a truly viviparous type of develop-
ment. The young remain in the hollow ovary of the parent for a period of about
five months, during which time they increase in size from 0.24 mm. (fertilized
egg) to about 42 mm. (newborn) (Eigenmann, 1890, 1894). It was learned while
devising a technique for culturing the embryos in vitro (Triplett, unpublished) that,
unlike the adults, they are capable of tolerating only very limited variation in the
osmotic concentration of the external medium.

The object of this study was to learn more precisely how well these animals
can regulate to variations in the concentration of the external medium both as
adults and during development. In addition, preliminary experiments were per-
formed in an effort to find out why the adults are better able to regulate than the
embryos.

MATERIALS AND METHODS
Experimental animals

Cymatogaster aggregata (Gibbons) was the most easily obtained of the embi-
otocids and was therefore the principal organism used in the investigation. This
species is widely distributed over the west coast of North America, ranging from
southern Alaska through Baja California (Tarp, 1952). It has been observed in
estuaries where the salinity is quite variable (Hubbs, 1917) as well as in bays and
along the open coast. Animals used in these experiments were collected by
angling from Goleta Pier, Goleta, California.

Less extensively investigated were Amphistichus argenteus (Agassiz) and
Micrometrus minimus (Gibbons). These species have about the same distribution
as C. aggregata, but so far as the authors are aware, they do not enter brackish
water. For these experiments they were collected with a beach seine on the
campus beach at Goleta, California. Only adults of these two species were used.

Methods

The general approach was to determine the total osmotic pressure of the serum
and the change in body weight after exposure to experimental salinities for periods
ranging from six hours to nine days.

1 These studies were aided by a contract between the Office of Naval Research, Department
of the Navy, and the Regents of the University of California, NR 104-466, and by a Grant
from the Research Committee of the University of California, Santa Barbara, Goleta, Cali-
fornia.

472

OSMOREGULATION IN EMBIOTOCIDS 473

The experimental animals were divided into four groups on the basis of their
stages of development. Group 1 comprised intra-ovarian embryos ranging in
standard length from 18 mm. to 25 mm. At this stage the animals were com-
pletely transparent except for a thin band of melanophores along the lateral line and
a few scattered melanophores in the head region. The fins were hypertrophied
and possessed highly vascularized dermal flaps between the rays. The hindgut
was also hypertrophied and protruded in the cloacal area. The fins and hindgut
probably serve as a pseudo-placenta (Eigenmann, 1890, 1894). The trailing edge
of the caudal fin was convex or flat. Group 2 consisted of intra-ovarian embryos
measuring between 27 mm. and 31 mm. in standard length. These animals still
possessed hypertrophied fins and a hypertrophied protruding hindgut. Melanin
deposition had increased considerably, and guanophores appeared on the dorsal
half of the animal, rendering it semi-opaque. Scales had differentiated visibly at
this time. The trailing edge of the caudal fin was becoming concave. Group 3
comprised intra-ovarian animals nearing parturition age and ranging in standard
length from 32 mm. to 38 mm. The body was completely opaque and the adult
pigment pattern was acquired. The fins, though not yet of adult proportion, had
begun to atrophy. The hindgut had also atrophied and protruded only slightly.
The trailing edge of the caudal fin was deeply concave and V-shaped as in the
adult. Group 4 comprised newborn animals and adults measuring between 39 mm.
and 115 mm. standard length. Preliminary experiments on group 4 animals of
different lengths (ages) established that their osmotic responses were identical.

A minimum of four animals from each group was placed directly from holding
tanks (running sea water) or directly from the ovarian cavity into each of the
following solutions of sea water diluted with distilled water: 100% (A = 1.86),
75%, 50%, 25%, 20%, 15%. For each series of measurements four adult animals
were kept in an aerated battery jar containing 3000 ml. of fluid and maintained at
16° C., or four embryos were kept aseptically in a fingerbowl containing 1000 ml.
of fluid and maintained at the same temperature. Serum freezing point determina-
nations and weight measurements were made on four different animals of each
group at each dilution after six hours and one, two, four and eight or nine days.
Freezing points are here given in sodium chloride equivalents.

Weight determinations were made on blotted animals with a triple beam
balance to an accuracy of 0.01 gram. All animals in group 4 were weighed before
and after exposure to the experimental medium. Weight measurements were not
possible with groups 1, 2, and 3 because these animals will not live after such
handling.

Blood for freezing point determinations was taken by cardiac puncture or by
removal of the tail at the level of the caudal peduncle. In the adults the particulate
matter was removed before placing the serum in capillaries. This was not possible
with the embryos since too little blood was available. Instead the freezing point
capillary was inserted into the heart, and whole blood was drawn into the tube.
The red cells and the clot were then packed by centrifuging, thus rendering the
serum clear enough for freezing point determinations. Freezing points were
determined with an apparatus modified after one used by Gross (1954), and since
the modification offers some advantages over the original it is briefly described
here. It consists of a Fiberglas-lined, insulated box containing four meters of one-
half inch copper tubing. Two glass windows allow light to be passed through

474 EDWARD L. TRIPLETT AND SUSANNE D. BARRYMORE

the box. The box is filled with brine, and the brine is cooled by circulating freon-
14 through the copper tubing. When the desired temperature is obtained the
freon-14 is turned off, and the previously frozen capillaries containing the fluid
to be measured are inserted in the brine between the observation windows. As the
box warms very slowly the fluid in the capillaries melts. The melting point can
be observed with great accuracy (+ 0.02° C.) by using crossed polaroids.

In an effort to learn something about differences in ionic exchange between
different age groups preliminary experiments have been performed using chlorine-
36 as a tracer. The technique consisted of placing the experimental animal, after
peritoneal injection of radio-chloride, in a divided chamber so that the gills and
half the body were on one side. Aliquots of the fluid in the two compartments
were measured at intervals over a three- to eleven-hour period for radioactivity.
The fluid in both compartments was a modified White’s (1954) medium (.220
M/L NaCl equiv.) slightly hypertonic to the serum.

RESULTS AND DISCUSSION
Changes in serum freezing point and body weight in experimental media

Cymatogaster aggregata: The results of freezing point determinations and
weight measurements on adult (group 4) C. aggregata are summarized in Figure
1, each point on the graph representing the average of measurements made of at
least four fishes. It can be seen that the viable range of salinities for this species
is quite large, ranging in these experiments from 20 per cent to 100 per cent sea
water. Adults remained alive no longer than three hours in 15 per cent sea water,
during which time they increased 3.83 per cent in weight.

At all salinities below 100 per cent sea water there was an initial drop in serum
tonicity which reached a minimal value between one and two days after exposure
to the experimental media. Accompanying this was a corresponding increase in
wet weight. In general, the minimal serum osmotic value decreased as the salinity
of the external medium was decreased. Less convincingly, Figure 1 indicates that
as the salinity of the external medium is decreased, the time required for return
to normal weight and serum tonicity increases. At all salinities tested above 20
per cent sea water, regulation is complete or nearly complete within nine days.

It will be noted that there was a small temporary increase in serum tonicity
and decrease in wet weight of fishes maintained in 100 per cent sea water. The
authors attribute this to an increased permeability of the integument resulting from
handling. It is presumed that a similar increase in integumentary permeability was
also experienced by fishes kept in other sea water dilutions, and this must be kept
in mind when interpreting the results.

The results of freezing point determinations on the various embryonic stages
are summarized in Table I, and corresponding figures for the adults are included
for comparison. It can be seen that the viable range of salinities for embryos in-
creases with increase in age. Group 1 and group 2 were able to survive for the
complete duration of the experiment only in media between 25 and 36 per cent sea
water (36 per cent sea water is equivalent osmotically to the ovarian fluid in which
the animals normally reside), whereas group 3 was capable of regulation at all
salinities between 25 and 75 per cent sea water.

However, it was noted that, in general, it required a greater alteration of the

OSMOREGULATION IN EMBIOTOCIDS A475

serum osmotic potential of embryos than of adults to kill the animals. In these
experiments group 1 and group 2 animals died when this value was above .232 M/L
or below .114 M/L. Group 3 animals died at values above .223 M/L or below
156 M/L. Possibly for this reason the embryos live longer at 15 per cent sea
water than do the adults. This may be a reflection of the fact that less differ-
entiated systems are generally more tolerant of their physiological environment.

% \NCREASE WET WEIGHT

220
Lap, Ae
—- ie KO ee es ee
> vd EN TS Oi as
E
ad
SE eS a =
a Oe RS en ee a gee
ee er ee ae a Oe he ee ES SS Se ee
<
=
=
al90
WJ
uJ
o
c
TI TM hs Suk eit ea
3 " ”
Oo " rT)
u "
5
«I70
uJ
”

160

I 2 3 4 5 6 Li 8 9
TIME (DAY)

Figure 1. Upper graph: the wet weight of C. aggregata adults plotted as a function
of time media of different salinities. Lower graph: tonicity (sodium chloride equivalents) of
C. aggregata adult serum plotted as a function of time in media of different salinities.

476 EDWARD L. TRIPLETT AND SUSANNE D. BARRYMORE

TABLE I

Serum tonicity of C. aggregata of different developmental stages after exposure to media
of variable salt concentrations for variable intervals of time

External medium

% Sea water 100 75 50 | 25 | 20 | 15
NaCl equiv. (M/L) .568 .326 284 142 114 .085
Age group Time exposed

1. Transparent 6 hours death death .230 .160 BS s5) 20
1 day death .164 131 ints,
2 days .185 death death
4 days .192
6 days .190
8 days A195

2. Semi-opaque 6 hours 32 .210 205 .160 =135 125
1 day death death 245 .156 .138 114
2 days death 181 156 BIE
4 days .185 death death
6 days .190
8 days .195

3. Opaque 6 hours vat .198 .196 .192 .198 179
1 day 233 202 .209 ah 197 .179
2 days death 236 3221. ? ? death
4 days DE | Ais) 195 .156
6 days ? ? ? death
8 days 211 .209 .206

4. Adult 6 hours ? ? ? ? ? death
1 day .206 .199 .193 .193 .181
2 days .206 watt Bait .169 siieil
4 days 216 BAIT .209 .199 181
6 days ? ? f ? ?
9 days .209 .206 ZG .196 .209

Micrometrus minimus and Amphistichus argenteus: Similar though less ex-
tensive experiments were performed with the adults of these two species. It was
expected that since they do not enter brackish water, the ability of M. minimus and
A. argenteus to regulate to differentiate salinities would be less than that of C.
aggregata, but this was not the case. Both were quite capable of remaining alive
in media between 20 and 100 per cent sea water over the nine-day interval. Fur-
thermore, serum freezing points of animals living in any of these dilutions at the
end of this time were the same as those of animals living in the normal environment

(serum = .209 M/L).

Ton exchange studies using Cl** tracers

The decreased ability of embryos to regulate to different tonicities of the ex-
ternal medium as compared to adults could be explained in several ways. One
possible explanation is that these animals may not yet have differentiated the

OSMOREGULATION IN EMBIOTOCIDS 477

branchial salt secretory mechanism that has been demonstrated in various adult
teleost fishes (Smith, 1930). Both adults and embryos of C. aggregata were
examined for this mechanism by utilizing the divided box technique first used by
Smith (1930). The animals were injected intraperitoneally with 0.25 ml. of an
isotonic salt solution with an activity of 0.125 microcuries per milliliter. At two-
hour intervals or less, 250-lambda samples were withdrawn from the anterior and
from the posterior chambers and measured for radioactivity with an accuracy of
1.0 per cent.

Three adults ranging in standard length from 105 mm. to 115 mm. and three
sibling embryos (group 3) measuring 33 mm. were used. The results for one
animal of each group are shown in Figure 2. The general results were about the
same for any animal in a given age group. It can be seen that in the adults the
activity of the anterior chamber is consistently higher than that of the posterior
for the duration of the experiment. This would indicate that a branchial salt
secretory mechanism is operative. However, in the embryos, the activity of the
posterior was consistently higher than that of the anterior chamber. This is tenta-
tively interpreted to mean that the ‘‘salt pump” in the gills is not yet in operation
or is operating at a much reduced rate. Though the embryos were placed in the
chamber so that half the body was on each side, it was roughly estimated that 74
per cent of the body surface area was contained by the posterior chamber. The
hypertrophied fins account for the greater part of this area. This could account
for the higher Cl** value in the rear chamber by assuming that chloride loss is about
equal over the whole body surface (including the gills).

These tracer studies are quite obviously not sufficient to draw definite con-
clusions. The sample is small, and desirable control experiments have not been
performed. The experiments do suggest, however, that the “salt pump” operative
in the adult is absent or operating less efficiently in the embryo. Further efforts
in this direction are planned.

8

a os

2 =)

o7 iS

ra

o a e
x6 P \ o 6

/ \ aq

a : a
a” 22 hie 0 5

=

= z
24 =4

= =

= =
=" a: 3

o rs)
2 2

ee ae ee AD C5) 18. 18 20 25 53.0
TIME (HOURS) TIME (HOURS)

Ficure 2. Radioactivity in front and back chambers of a divided box so arranged that
the anterior half of a chloride*®-injected fish is contained in the front chamber and the posterior
half of the fish is contained in the back chamber. Dotted lines = front chamber; solid lines =
back chamber. Left graph=measurement for adult fish; right graph = measurements for
embryonic fish.

478 EDWARD L. TRIPLETT AND SUSANNE D. BARRYMORE

Another clue concerning the lack of osmoregulatory ability in embryos was
gained by observing the volume of the hindgut lumen in media of different osmotic
concentrations. It was noted that embryos placed in any of the hypertonic solu-
tions (> 36% sea water) experience a swelling of the hindgut. The hindguts
of animals placed in hypotonic solutions shrink. It is well known that marine
teleosts maintain osmotic equilibrium by drinking sea water and excreting the
excess Salt, principally through the gills. It is hypothesized that these embryos also
drink the external medium. If they should be placed in a hypertonic medium
there would be created an osmotic gradient between body fluids and hindgut
lumen resulting in body dehydration and swelling of the hindgut lumen. The
opposite gradient would result if a hypotonic medium were drunk by the embryo.
The question remains as to why this does not happen in the adult.

SUMMARY

1. The osmoregulatory abilities of adult and embryonic Cymatogaster aggregata
were tested by determining serum freezing points and weight changes after ex-
posure to dilutions of sea water ranging between 15 and 100 per cent.

2. Adults were able to regulate in external media ranging between 20 and 100
per cent sea water. The greatest deviations of weight and serum freezing point
from the normal occur after one to two days in the experimental media. Regula-
tion in all media is complete at or before nine days of exposure.

3. The ability of embryos to regulate is proportional to their stage of develop-
ment. The youngest animals tested regulate only in media with values between
25 and 36 per cent sea water.

4. Similar though less extensive experiments were performed with adult
Amphistichus argenteus and Micrometrus minimus. The osmoregulatory ability
of these species is identical to that of adult C. aggregata.

5. Preliminary experiments using Cl°* as a tracer indicate that the branchial
salt secretory mechanism present in the adults is absent or operating less efficiently
in the embryos of C. aggregata.

6. Observations on the volume of the gut in different external media indicate
that the embryos drink the medium, thus creating an osmotic gradient between the
body fluids and the lumen of the gut. The result is dehydration or “edema,”
depending upon the tonicity of the external medium.

LITERATURE CITED

EIGENMANN, Cart H., 1890. The development of Micrometrus aggregatus, one of the vivip-
arous serf perches. Amer. Natur., 23: 923-927.

EIGENMANN, Cart H., 1894. Cymatogaster aggregatus (Gibbons): A contribution to the
ontogeny of viviparous fishes. Bull. U. S. Fish Comm., 12: 401-478.

Gross, WARREN J., 1954. Osmotic responses in the sipunculid worm Dendrostomum sosteri-
colum. J. Exp. Biol., 31: 402.

Hupsss, Cart L., 1917. The breeding habits of the viviparous perch, Cymatogaster. Copeia,
47: 72-74.

SmitH, H. W., 1930. Absorption and excretion of water and salts by marine teleosts. Amer.
J. Physiol., 93: 480-505.

Tarp, Frep H., 1952. A revision of the family Embiotocidae (surf perches). Fish Bull. No.
88, California Div. Fish and Game, 99 p.

Wuirte, Pure R., 1954. The Cultivation of Plant and Animal Cells. The Ronald Press
Co., New York, 239 p.

INDEX

ACCLIMATION, by crab, 150, 215.

Accumulation of blood proteins by moth
oocytes, 338.

Acid phosphatase in planarians, 315.

Action of goitrogens on tree toad larvae, 430.

Aldehydes, action of on fish egg chorion, 120.

Adult Cymatogaster, osmoregulation in, 472.

Adult Cymatogaster, tissue specificity in, 463.

AuFerT, M. See I. I. GEscHWIN»D, 66.

Alga, red, cation regulation in, 55.

Alga, sex-linked inheritance in, 407.

ALLEN, K., AND J. AWAPARA. Metabolism of
sulfur amino acids in Mytilus and Rangia,
173.

Amebocytes, Limulus, in vitro reactions of,
324.

Amino acid pattern of K-deficient Fundulus,
79.

Amino acids, sulfur, metabolism of, 173.

Aminopeptidases in planarians, 315.

Amoebae, surface antigen dynamics of, 70.

Amphibian, protein changes during develop-
ment of, 451.

Amphibian larvae, treatment of with radio-
iodine, 430.

Amphibian melanophores, 1.

Amphistichus, osmoregulation in, 472.

Anatomy of circulatory system in hagfish,
289.

Anatomy of crayfish gastroliths, 137.

Anatomy of Pleuroncodes larvae, 17.

Antennal gland of crayfish, excretion by, 296.

Antigen dynamics of slime mold, 70.

Antigenic studies of moth oocytes, 338.

Antigens of sea urchin sperm surface, 96.

Appendages of Pleuroncodes larvae, 17.

Arachnid, in vitro reactions of amebocytes of,
324.

Arbacia dermal secretion,
hibiting action of, 439.

Arbacia sperm antigens, 96.

Asterias sperm antigens, 96.

Awapara, J. See K. ALLEN, 173.

fertilization in-

BACTERIA, in vitro reactions of Limulus
amebocytes to, 324.

BaGnara, J. T. Tail melanophores of Xeno-
pus in normal development and regenera-
tion, 1.

Balanus, molting cycle in, 31.

Bane, F. B. See M. V. SurropKar, 324.

Barnacle, molting cycle in, 31.

BARNWELL, F. H. See F. A. Brown, Jr,
367.

BARRYMORE, S. See E. L. Triprett, 463, 472.

BECKMAN, C., AND R. Menzies. The re-
lationship of reproductive temperature
and the geographic range of the marine
woodborer Limnoria, 9.

Behavior of chalcid wasp, effects of oxygen
poisoning on, 269.

BENNETT, M. F. See F. A. Brown, Jr., 367.

Birds, flight muscles of, 262.

Blockage of molting in Uca, 129.

Blood inulin levels in crayfish, 296.

Blood protein depletion, effects of on growth
of Cecropia oocytes, 352.

Blood proteins, selective accumulation of by
moth oocytes, 338.

Blood vascular system of Myxine, 289.

BookHout, C. G. See J. D. Costiow, Jr,
133-..203:

Boyp, C. M. Larval stages of Pleuroncodes,
Wve

Breeding of Balanus, 31.

Breeding experiments with Ulva, 407.

BRE. Weide SceaHaAl DRONCN +. pRisc GO%,
382.

Brown, F. A., Jr, H. M. Webb anv W. J.
Brett. Magnetic response of an organ-
ism and its lunar relationships, 382.

Brown, F. A., Jr., W. J. Brett, M. F. Ben-
NETT AND F. H. BARNWELL. Magnetic
response of an organism and its solar
relationships, 367.

CALCIFICATION rate in corals, 419.

Cambarus gastrolith deposition, 137.

Carbonic anhydrase inhibitors, effect of on
snail shell growth, 412.

Carotenoid protein, effect of on growth of
Cecropia oocytes, 352.

Cation regulation of Porphyra, 55.

Cecropia oocytes, growth of, 352.

Cell division in Tetrahymena, 84.

Chalcid wasp, effects of oxygen poisoning on,
269.

479

480

Chalk, effects of on Venus eggs and larvae,
48.

Changes, protein, in development, 451.

Chorion of fish egg, hardening of, 120.

Chromatography of Fundulus muscle, 79.

Chromatography of mollusc amino acids, 173.

Chromatophores of Xenopus, 1.

Chromosomes of Tetrahymena, 84.

Circulation in hagfish, 289.

Cirripede, molting in, 31.

Clam eggs and larvae, effects of turbidity-
producing materials on, 48.

Clay, effects of on Venus eggs and larvae, 48.

Cold, action of on Uca molting, 129.

Cold, effect of on heat tolerance of crab,
150.

Cold, effect of on oxygen consumption of
crabs, 215.

Colony weight in relation to calcification
rate, in corals, 419.

Columba, succinic dehydrogenase
muscles of, 262.

Compression, effects of on wasp, 269.

Control of distal retinal pigment migration
in Palaemonetes, 393.

Corals, skeleton formation in, 419.

Corpus cardiacum extract, action of on roach
efferent nerve activity, 111.

CostLtow, J. D., Jr, anp C. G. BookuHovut.
The complete larval development of
Sesarma reared in the laboratory, 203.

Costtow, J. D., Jr. C. G. BookHouT AND
R. Monroe. The effect of salinity and
temperature on larval development of
Sesarma reared in the laboratory, 183.

Crab, fiddler, blockage of molting in, 129.

Crab, horseshoe, in vitro reactions of amebo-
cytes of, 324.

Crab, larval stages of, 17, 183, 203.

Crabs, heat tolerance in, 150.

Crabs, intertidal, oxygen consumption of, 215.

Crayfish, limb regeneration and endocrine ac-
tivity in, 250.

Crayfish antennal gland, excretion of inulin
and glucose by, 296.

Crayfish gastrolith, deposition of, 137.

Crisp, D. J., anp B. S. Pater. The moulting
cycle in Balanus, 31.

Crustacean, blockage of molting in, 129.

Crustacean, deposition of skeletal structures
fly air

Crustacean, distal retinal pigment of, 393.

Crustacean, heat tolerance of, 150.

Crustacean, larval stages of, 17, 183, 203.

Crustacean, limb regeneration and endocrine
activity in, 250.

Crustacean, oxygen consumption of, 215.

in flight

INDEX

Crustacean antennal gland, excretion of
inulin and glucose by, 296.

Cycle, molting, in Balanus, 31.

Cycle, molting, in crayfish, 137.

Cyclostome, circulation in, 289.

Cymatogaster, osmoregulation in, 472.

Cymatogaster, tissue specificity in, 463.

Cyprus, 'G..C.. See R. W.. Eperey, 35)

Cystathionine, formation of in molluscs, 173.

ARK, role of in migration of prawn
distal retinal pigment, 393.

Davis, H. C. Effects of turbidity-producing
materials in sea water on eggs and larvae
of the clam, Venus, 48.

Decapod crustacean, gastrolith deposition in,
137;

Decapod crustacean, larval stages of, 17.

DEHNEL, P. A. Effect of temperature and
salinity on the oxygen consumption of
two intertidal crabs, 215.

DEHNEL, P. A. See M.-E. Topp, 150.

Dent, J. N. See W. G. Lynn, 430.

Depletion of blood protein, effect of on growth
of Cecropia oocytes, 352.

Deposition of skeletal structures in Crustacea,
1a%

Dermal secretion, fertilization inhibiting ac-
tion of, 439.

Developing Cymatogaster, tissue specificity in,
463.

Developing Xenopus, melanophores in, 1.

Development of Cecropia oocytes, 352.

Development of clam eggs in turbid sea wa-
ter, 48.

Development of dermal
echinoderm eggs, 439.

Development of Drosophila, effects of gluco-
samine on, 308.

Development of Sesarma, 183, 203.

Developmental stages of Cymatogaster, osmo-
regulation in, 472.

Developmental stages of Pleuroncodes, 17.

Developmental stages of wasp, effects of oxy-
gen poisoning on, 269.

Diapausing pupae of Mormoniella, sensitivity
of to oxygen poisoning, 269.

Dictyostelium, surface antigen dynamics in,
70.

Digestion in planarians, 315.

Distal retinal pigment of Palaemonetes, 393.

Distribution of Limnoria, 9.

Distribution pattern of succinic dehydrogenase
in bird flight muscles, 262.

Drosophila development, effects of glucosa-
mine on, 308.

Dugesia, intracellular digestion in, 315.

secretion-treated

INDEX

Duranp, J. B. Limb regeneration and endo-
crine activity in the crayfish, 250.

Dwarf mice, liver polyploidy of, 66.

Dynamics of surface antigens of slime mold,
70.

EK CDYSIS of barnacles, 31.

Ecdysis of crabs, in relation to temperature
and salinity, 150.

Ecdysis of crayfish, in relation to gastrolith
deposition, 137.

Ecdysis in Uca, 129.

Echinarachnius sperm antigens, 96.

Echinoderm dermal secretion, fertilization in-
hibiting action of, 439.

Echinoderm sperm antigens, 96.

Ecological significance of Uca molting, 129.

Ecology of corals, 419.

Ecology of Hemigrapsus, in relation to oxy-
gen consumption, 215.

Ecology of Porphyra, 55.

Ecology of Sesarma, 183.

Effect of K deficiency on Fundulus amino
acids, 79.

Effect of salinity and temperature on larval
development of Sesarma, 183.

Effect of temperature and salinity on crab
oxygen consumption, 215.

Effect of temperature and salinity on heat
tolerance of crabs, 150.

Effect of thyroxin and growth hormone on
liver polyploidy, 66.

Effects of blood protein depletion on growth
of Cecropia oocytes, 352.

Effects of glucosamine on Drosophila devel-
opment, 308.

Effects of turbidity-producing materials on
eggs and larvae of Venus, 48.

Efferent nerve activity of roach, 111.

Egg chorion, fish, hardening of, 120.

Eggs, Venus, effects of turbidity-producing
materials on, 48.

Electric stimulation of roach efferent nerve,
111.

Electrophoretic studies of prawn hormones,
393.

Electrophoretic studies of Rana proteins, 451.

Elminius, molting of, 31.

Embiotocid fishes, osmoregulation in, 472.

Embryonic Cymatogaster, osmoregulation in,
472.

Embryonic Cymatogaster, tissue specificity in,
463.

Embryonic stages of wasp, effects of oxygen
poisoning on, 269.

Embryos, protein changes during development
of, 451.

481

Emersion, role of in molting of Balanus, 31.
Endocrine activity in crayfish, 250.
Endocrine studies on tree toad, 430.
Endocrine studies on prawn, 393.

Enzyme distribution in bird flight muscles,
262.

Enzymes, hydrolytic, in phagocytes of plana-
£1a01S, “85,

Epptey, R. W., ann C. C. Cyprus. Cation
regulation and survival of the red alga
Porphyra in diluted and concentrated sea
water, 55.

Excretion of inulin and glucose by crayfish
antennal gland, 296.

Extracts of corpus cardiacum, action of on
roach efferent nerve activity, 111.

Exuviae of barnacles, 31.

FEEDING, role of in shell growth of Physa,
412.

Feeding of Limnoria, in relation to tempera-
ture, 9.

Feeding of planarians, 315.

Feeding in relation to molting of Balanus, 31.

Female protein, effect of on growth of

Cecropia oocytes, 352.

Fertilization, hardening of fish egg chorion
after, 120.

Fertilization inhibiting action of Arbacia der-
mal secretion, 439.

Fertilizin in relation to sea urchin sperm
antigens, 96.

Fiddler crab, blockage of molting in, 129.

Filtration in crayfish antennal gland, 296.

FINGERMAN, M., AnD W. C. Mosperty, Jr.
Investigation of the hormones controlling
the distal retinal pigment of the prawn
Palaemonetes, 393.

Fish, osmoregulation in, 472.

Fish, scale transplantation in, 463.

Fish egg chorion, hardening of, 120.

Flatworm, digestion in, 315.

Flight muscles of birds, 262.

Formation of skeleton in corals, 419.

Foyn, B. Sex-linked inheritance in Ulva,
407.

FREEMAN, J. A. Influence of carbonic an-
hydrase inhibitors on shell growth of a
fresh-water snail, Physa, 412.

Frog, protein changes during development of,
451.

Fruit fly development, effects of glucosamine
on, 308.

Fullers earth, effects of on Venus eggs and
larvae, 48.

Function of bird flight muscles, 262.

Functional changes in Tetrahymena, 84.

Fundulus, amino acid pattern in, 79.

482

(,ALATHEID crab, larval stages of, 17.

Gastrodermis of planarians, histology of, 315.
Gastrolith histology, in crayfish, 137.
Genetics of Ulva, 407.

Geographical range of Limnoria, 9.

GerorcE, J. C., anp C. L. TALESARA. Studies
on the structure and physiology of the
flight muscles of birds. 9., 262.

GescHwinp, Ir Ty Mo Arrerr Anp “C:
ScHooLey. The effects of thyroxin and
growth hormone on liver polyploidy, 66.

Gills of Myxine, 289.

Glucosamine, effects of on Drosophila de-
velopment, 308.

Glucose, excretion of by crayfish antennal
gland, 296.

Glucuronidase in planarians, 315.

Goitrogens, action of on tree toad larvae, 430.

GotpsmiTH, M. H. M., anp H. A. SCHNEIDER-
MAN. The effects of oxygen poisoning
on the post-embryonic development and
behavior of a chalcid wasp, 269.

GorEau, TF. Any Ne I Gorrauy = fie
physiology of skeleton formation in corals.
J has n HS),

Gradient of melanophore differentiation, 1.

Grafts of scales in fish, 463.

Grapsoid crabs, heat tolerance in, 150.

Green gland of crayfish, excretion of inulin
and glucose by, 296.

Grecc, J. H. Surface antigen dynamics in
the slime mold, Dictyostelium, 70.

Growth of Cecropia oocytes, 352.

of clam larvae in turbid sea water,

Growth of coral colonies, 419.

Growth of regenerating crayfish limb, 250.

Growth of shell in Physa, 412.

Growth hormone, effect of on liver polyploidy,
66.

HAGFISH, circulation in, 289.

Hanton, D. P. The effect of potassium de-
ficiency on the free amino acid pattern
of the muscle tissue of protein-maintained
Fundulus, 79.

Hardening of fish egg chorion, 120.

Hatching of Sesarma, 183, 203.

Heat, action of on Uca molting process, 129.

Heat, effect of on oxygen consumption of
crabs 215:

Heat tolerance of crabs, 150.

Heat-treatment of Tetrahymena, 84.

Hemigrapsus, heat tolerance in, 150.

Hemigrapsus, oxygen consumption of, 215.

Hemioniscus, role of in molting of Balanus,
Si:

INDEX

High temperature, effect of on oxygen con-

sumption of crabs, 215.

Histochemistry of planarians, 315.

Histology of amphibian thyroid and thymus
treated with radioiodine, 430.

Histology of crayfish gastrolith, 137.

Histology of Drospohila larvae, 308.

Histology of limb regeneration in crayfish,
250.

Histology of Myxine circulatory system, 289.

Hoiz, G. G., Jr. Structural and functional
changes in a generation in Tetrahymena,
84.

Hormone, growth, effect of on liver poly-
ploidy, 66.

Hormone of roach, action of on efferent
nerve activity, 111.

Hormones, molting, of Uca, 129.

Hormones of Palaemonetes, 393.

Horseshoe crab, in vitro reaction of amebo-
cytes of, 324.

Hydrolytic enzymes in
planarians, 315.

Hyla, treatment of with radioiodine, 430.

phagocytes of

[LYANASSA, magnetic response of, 367,
382.

Immunologic studies of sea urchin sperm, 96.

Immunological studies on slime mold, 70.

Immunology of blood protein accumulation by
moth oocytes, 338.

Infection of Limulus with bacteria, 324.

Influence of carbonic anhydrase inhibitors on
snail shell growth, 412.

Inheritance in Ulva, 407.

Inhibiting action of Arbacia dermal secretion
on fertilization, 439.
Inhibition of radioiodine

goitrogens, 430.
Inhibitors, carbonic anhydrase, effects of on
snail shell growth, 412.
Intertidal alga, cation regulation in, 55.
Intertidal crabs, respiration of, 215.
Intracellular digestion in planarians, 315.

localization by

Inulin, excretion of by crayfish antennal
gland, 296.
Invertebrates, metabolism of sulfur amino

acids spyanliZo!
Investigation of hormones controlling distal
retinal pigment of Palaemonetes, 393.
Ion transport in Porphyra, 55.
Isopod, marine, distribution of, 9.
Isopod, parasitic, of Balanus, 31.

Circulation in the hag-

OHANSEN, K.
fish, Myxine, 289.

INDEX

AOLIN, effects of on Venus eggs and
larvae, 48.
KirscHner, L. B. Sce J. A. RIEGEL, 296.
KOHLER, K., AnD C. B. Metz. Antigens of
the sea urchin sperm surface, 96.

| ABORATORY culture of Sesarma, 183,
203. ;

Larvae, Venus, effects of turbidity-producing
materials on, 48.

Larval development of Sesarma, 183, 203.

Larval stages of Pleuroncodes, 17.

Larval tree toads, treatment of with radio-
iodine, 430.

Larval Xenopus, melanophores in, 1.

Light, role of in migration of prawn distal
retinal pigment, 393.

Light, role of in molting of Balanus, 31.

Light-sensitive melanophores of Xenopus, 1.

Limb regeneration and endocrine activity in
crayfish, 250.

Limb regeneration in Uca, 129.

Limnoria, distribution of, 9.

Limulus amebocytes, in vitro reactions of, 324.
Liver polyploidy, effect of thyroxin and
growth hormone on, 66.
Localization of radioiodine in

larvae, 430.

Low temperature, effect of on heat tolerance
of crab, 150.

Low temperature, effect of on oxygen con-
sumption of crabs, 215.

Low temperature blockage of Uca molting,
129.

Lunar cycle, in relation to molting of Balanus,
31.

Lunar relationships of magnetic response of
Nassarius, 382.

Lynn, W. G., anp J. N. Dent. The action
of various goitrogens in inhibiting locali-
zation of radioiodine in the thyroid and
thymus glands of larval tree toads, 430.

Lytechinus sperm antigens, 96.

tree toad

MAGNETIC response of an organism, 367,
382.

Manicina, calcification rate in, 419.

Marine woodborer, ecology of, 9.

Mechanism of chorion hardening in fish egg,
120.

Melanophores of Xenopus, 1.

Mellita sperm antigens, 96.

Menzies, R. See C. BECKMAN, 9.

Mercenaria eggs and larvae, effects of tur-
bidity-producing materials on, 48.

Metabolism of intertidal crabs, 215.

Metabolism of sulfur amino acids by molluscs,
173.

483

Metabolism of wasp larvae, 269.

Metz, C. B. Investigation of fertilization in-
hibiting action of Arbacia dermal secre-
tion, 439.

Metz, C. B. See K. KOu Er, 96.

Micrometrus, osmoregulation in, 472.

Migration of Nassarius, in relation to mag-
netic force, 367, 382.

Mitsurn, N., E. A. WerAnt anv K. D.
Roeper. The release of efferent nerve
activity in the roach, Periplaneta, by
extracts of the corpus cardiacum, 111.

Mitosis in Tetrahymena, 84.

Mosserty, W. C., Jr. See M. FINGERMAN,
393.

Mollusc, magnetic response of, 367, 382.

Mollusc eggs and larvae, effects of turbidity-
producing materials on, 48.

Mollusc shell growth, 412.

Molluscs, metabolism of sulfur amino acids
by, 173.

Molting, blockage of, in Uca, 129.

Molting of crabs, in relation to temperature
and salinity, 150.

Molting cycle in Balanus, 31.

Molting processes of crayfish, in relation to
gastrolith deposition, 137.

Molting process, in relation to neurosecretory
activity of crayfish, 250.

Molting of Sesarma, 183, 203.

MonroE, R. See J. D. Costitow, Jr., 183.

Mormoniella, effects of oxygen poisoning on,
269.

Morphogenesis of slime molds, 70.

Morphogenesis in Tetrahymena, 84.

Morphology of Pleuroncodes larvae, 17.

Morphology of Sesarma larvae, 203.

Moth oocytes, accumulation of blood proteins
by, 338.

Moth oocytes, growth of, 352.

Motor activity of oxygen-poisoned wasps,
269.

Mouse liver polyploidy, effect of thyroxin
and growth hormone on, 66.

Mud-snails, magnetic response of, 367, 382.

Muscle of Fundulus, amino acids in, 79.

Muscles, flight, of birds, 262.

Musculature of oxygen-poisoned wasp larvae,
269.

Mussel, metabolism of sulfur amino acids by,
73:

Mutant Drosophila, effects of glucosamine on
development of, 308.

Mutants of Ulva, 407.

Mytilus, metabolism of sulfur amino acids by,
173.

Myxine, circulation in, 289.

484

NASSARIUS, magnetic response of, 367,
382.

Nerve activity of roach, 111.

Nervous system of wasps, effects of oxygen
poisoning on, 269.

Neurosecretory activity in crayfish, 250.

Neurosecretory hormones of roach, action of
on efferent nerve activity, 111.

Nitrogen content of corals, 419.

HTSUKA, E. On the hardening of the
chorion of the fish egg after fertilization.
LE 120:

Oocytes, Cecropia, growth of, 352.

Oocytes, moth, selective accumulation of blood
proteins by, 338.

Orconectes, limb regeneration and endocrine
activity in, 250.

Orconectes gastrolith deposition, 137.

Orientation of snails, in response to magnetic
force, 367, 382.

Oryzias, chorion hardening in, 120.

Osmoregulation in embiotocid fishes, 472.

Osmotic relations of Porphyra in dilute and
concentrated sea water, 55.

Ovarian differentiation in Cecropia moth, 352.

Oxidizing agents, action of on fish egg
chorion, 120.

Oxygen consumption of intertidal crabs, 215.

Oxygen poisoning of chalcid wasps, 269.

PALAEMONETES, distal retinal pigment
of, 393.

Parasite, role of in molting of Balanus, 31.

Parasitic wasp, oxygen poisoning of, 269.

Passano, L. M. Low temperature blockage
of molting in Uca, 129.

Para? BOS. See Di J. Ceres sik

Pectoralis major muscle of pigeon, succinic
dehydrogenase in, 262.

Penis loss in Balanus, 31.

Perch, osmoregulation in, 472.

Perch, tissue specificity in, 463.

Perchlorate, action of on radioiodine-treated
amphibian larvae, 430.

Periodicity, tidal, role of in molting of Balanus,
S10

Periplaneta, efferent nerve activity in, 111.

Phagocytes, hydrolytic enzymes in, 315.

Phenylthiourea, action of on _ radioiodine-
treated amphibian larvae, 430.

Phosphatases, acid, in planarians, 315.

Physa, shell growth of, 412.

Physiology of skeleton formation in corals,
419.

Pigeon flight muscles, succinic dehydrogenase
in, 262.

Pigment, distal retinal, of Palaemonetes, 393.

INDEX

Pigment cells of Xenopus, 1.

Plagiobrissus sperm antigens, 96.

Planarians, intracellular digestion in, 315.

Pleuroncodes, larval stages of, 17.

Poisoning, oxygen, of chalcid wasps, 269.

Polyphemus oocytes, selective accumulation
of blood proteins by, 338.

Polyploidy, -mouse liver, effect of thyroxin
and growth hormone on, 66.

Population studies with Limnoria, 9.

Porphyra, cation regulation in, 55.

Post-embryonic chalcid wasp, effects of oxy-
gen poisoning on, 269.

Potassium-deficient Fundulus, amino acids of,
79.

Prawn distal retinal pigment, 393.

Protein changes in development, 451.

Protein depletion, effect of on growth of
Cecropia oocytes, 352.

Protein-maintained Fundulus, amino acid pat-
tern of, 79.

Proteins, blood, selective accumulation of by
moth oocytes, 338.

Protozoan, structural and functional changes
in, 84.

RADIATION damage to thyroid and thy-
mus of amphibian larvae, 430.

Radiocalcium, use of in study of calcification
rate in corals, 419.

Radiochlorine, use of in osmotic studies on
fish, 472.

Radioiodine uptake by tree toad larvae, 430.

Radiosulfur, uptake of by molluscs, 173.

Rana, protein changes during development of,
451.

Range, geographical, of Limnoria, 9.

Rangia, metabolism of sulfur amino acids by,
173:

Rate of molting of barnacles, 31.

Red alga, cation regulation in, 55.

Reducing agents, action of on fish egg chorion,
120.

Reef coral, calcification rate in, 419.

Regeneration, limb, of crayfish, 250.

Regeneration of limbs in Uca, 129.

Regeneration of Xenopus tails, 1.

Regulation of cations in Porphyra, 55.

Relationship of reproductive temperature to
distribution of Limnoria, 9.

Release of efferent nerve activity in roach,
i Ge

Reproductive temperature and distribution of
Limnoria, 9.

Respiration of intertidal crabs, 215.

Respiration of wasp larvae, 269.

Respiratory organs of Myxine, as blood-pro-
pellors, 289.

INDEX

Response, magnetic, of an organism, 367, 382.
Retinal pigment of Palaemonetes, 393.
Rhythmic occurrence of gravidity in Limnoria,

Rhythms of response in Nassarius, 367, 382.

RIEGEL, J. A., AND L. B. KirscHNER. The ex-
cretion of inulin and glucose by the cray-
fish antennal gland, 296.

Rizxi, M. T. M. The effects of glucosamine
hydrochloride on the development of
Drosophila, 308.

Roach efferent nerve activity, 111.

Roeper, K. D. See N. Mirzsurn, 111.

Roton, C. See R. M. Rosensavum, 315.

RosENBAUM, R. M., ANpd C. Roton. Intra-
cellular digestion and hydrolytic enzymes
in the phagocytes of planarians, 315.

Rutserc, L. D. See W. H. Te Fer, 352.

ALINITY, effect of on crab oxygen con-
sumption, 215.

Salinity, effect of on heat tolerance of crabs,
150.

Salinity, effect of on larval development of
Sesarma, 183.

Salinity relations of Porphyra, 55.

Salt balance in fishes, 472.

Saturniid moths, selective accumulation of
blood proteins by, 338.

Scale transplantation in Cymatogaster, 463.

SCHNEIDERMAN, H. A. See M. H. M. Gotp-
SMITH, 269.

ScHootey, C. See I. I. Gescuwinp, 66.

Sea urchin dermal secretion, fertilization in-
hibiting action of, 439.

Sea urchin sperm antigens, 96.

Sea water, turbid, effects of on Venus eggs
and larvae, 48.

Seasonal differences
crabs, 150.

Seasonal differences in oxygen consumption
of crabs, 215.

Seasonal variation in molting of Balanus, 31.

Secretion, dermal, of Arbacia, fertilization
inhibiting action of, 439.

Selective accumulation of blood proteins by
moth oocytes, 338.

Serology of Dictyostelium, 70.

Serology of moths, 338, 352.

Serology of sea urchin sperm, 96.

Sesarma, larval development of, 183, 203.

Sex-linked inheritance in Ulva, 407.

Shell growth in Physa, 412.

SuiropKar, M. V., A. Warwick AND F. B.
Banc. The in vitro reaction of Limulus
amebocytes to bacteria, 324.

Silicone, use of in in vitro culture of Limulus
amebocytes, 324.

in heat tolerance of

485

Silt, effects of on Venus eggs and larvae, 48.

Size of colony in relation to calcification rate,
in corals, 419,

Size of Limnoria, in relation to temperature,
9.

Size in relation to oxygen consumption of
Crabssi2ls;

Skeletal structures in Crustacea, deposition
Onl Sz

Skeleton formation in corals, 419.

Slime mold, surface antigen dynamics in, 70.

Snails, magnetic response of, 367, 382.

Snail shell growth, 412.

Solar relationships of magnetic response of
Nassarius, 367.

Specificity, tissue, in Cymatogaster, 463.

Sperm antigens, sea urchin, 96.

SPIEGEL, M. Protein changes in development,

. 451,

Stages, developmental, of Pleuroncodes, 17.

Statistical study of Sesarma development in
relation to temperature and salinity, 183.

Structure of bird flight muscles, 262.

Structural and functional changes in Tetra-
hymena, 84.

Succinic dehydrogenase distribution in pigeon
muscle, 262.

Sulfhydryl compounds, action of on fish egg
chorion, 120.

Sulfonamides, effects of on snail growth, 412.

Sulfur amino acids, metabolism of, 173.

Surface antigen dynamics of slime mold, 70.

Survival of crabs at high and low tempera-
tures, 150.

Survival of Porphyra in diluted and concen-
trated sea water, 55.

Synthesis of protein during development, 451.

‘T AIL melanophores of Xenopus, 1.

TALESARA, C. L. See J. C. GeorGE, 262.
Taurine, formation of in molluscs, 173.
Teleost, osmoregulation in, 472.

Teleost, scale transplantation in, 463.

Teleost egg chorion, hardening of, 120.

TeL_rer, W. H. The selective accumulation
of blood proteins by the oocytes of
saturniid moths, 338.

TELFER, W. H., anp L. D. Rutserc. The
effects of blood protein depletion on the
growth of the oocytes in the Cecropia
moth, 352.

Temperature, effect of on crab oxygen con-
sumption, 215.

Temperature, effect of on heat tolerance of
crab, 150.

Temperature, effect of on larval development
of Sesarma, 183.

Temperature, effect of on Uca molting, 129.

486

Temperature, reproductive, and distribution
of Limnoria, 9.

Temperature, role of in molting of Balanus,
Sl.

Tetrahymena, _ structural
changes in, 84.

Thiocyanate, action of on radioiodine-treated
amphibian larvae, 430.

Thymus localization of radioiodine by tree
toad larvae, 430.

Thyroid localization of radioiodine by tree
toad larvae, 430.

Thyroxin, effect of on liver polyploidy, 66.

Tidal periodicity, role of in molting of
Balanus,, 31.

Tissue specificity in Cymatogaster, 463.

Tissue volumes of Porphyra at different sea
water concentrations, 55.

Toads, tree, treatment of with radioiodine,
430.

Topp, M.-E., anp P. A. DEHNEL. Effect of
temperature and salinity on heat toler-
ance in two grapsoid crabs, 150.

Transplantation of scales in Cymatogaster,
463.

Travis, D. F. The deposition of skeletal
structures in the Crustacea. I., 137.
TRIPLETT, E. L., AND S. BARRYMORE. Some
aspects of osmoregulation in embryonic
and adult Cymatogaster and other embio-

tocid fishes, 472.

and __ functional

TRIPLETT, E. L., anp S. BARRYMORE. Tissue
specificity in embryonic and _ adult
Cymatogaster, studied by scale trans-

plantation, 463.

INDEX

Trypsin, effects of on echinoderm eggs, 439.

Trypsin, effects of on prawn hormones, 393.

Turbidity-producing materials, effects of on
Venus eggs and larvae, 48.

CA, low temperature blockage of molting
in, 129. :
Ulva, sex-linked inheritance in, 407.

VENUS eggs and larvae, effect of turbidity-
producing materials on, 48.

Vibrio, effect of on Limulus amebocytes in
vitro, 324.

Volume increment in Cecropia oocyte, 352.

WARWICK, A. See M. V. SHtIRopKar,
324.

Wasp, effects of oxygen poisoning on, 269.

Wess, H. M. See F. A. Brown, Jr., 382.

WEIANT, E. A. See N. Mirsurn, 111.

Weight in relation to calcification rate in
corals, 419.

Weight-specific oxygen consumption of crabs,
VANS

Woodborer, marine, distribution of, 9.

X ENOPUS melanophores, 1.

Y OLK production by Cecropia oocytes, 352.

7, OOXANTHELLAE, role of in calcifica-
tion of coral colonies, 419.

BIOLOGY MATERIALS

The Supply Department of the Marine Biological Labora-
tory has a complete stock of excellent plain preserved and
injected materials, and would be pleased to quote prices on
school needs.

—

PRESERVED SPECIMENS

for
Zoology, Botany, Embryology,
and Comparative Anatomy

LIVING SPECIMENS

for
Zoology and Botany

including Protozoan and |
Drosophila Cultures, and
Animals for Experimental and

. Laboratory Use.

MICROSCOPE SLIDES

for
Zoology, Botany, Embryology,
Histology, Bacteriology, and
Parasitology.

CATALOGUES SENT ON REQUEST
Th
Supply Department

MARINE
BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS 70/00 4. aaa

Bs: , / tt

BROWN, F. A., JR., W. J. BRETT, M. F. BENNETT AND F.H. BARNWELL
Magnetic response of an organism and its solar relationships... . sca! 367

BROWN, F. A., JR., H. M. WEBB AND W. J. BRETT | me SA ae
s, Magnetic response of an organism and its lunar relationships.......' 382 ie

FINGERMAN, MILTON, AND WILLIAM C, MOBBERLY, JR.

Investigation of the hormones controlling the distal retinal pigment a
of the prawn Palaemonetes. a ake Bol Caeph\« Mie rldncta KANety Bo silat gee alae 393 9
FOYN, BJORN ) 5 Ns seat yy A
Shs Sidmed inheritance ih Ulva... Fey als o'n GALEN Merge Bieta 5h aor
FREEMAN, JOHN A. | ay
Influence of eaabonic anhydrase inhibitors on shell growth of, a oe
fresh-water snail, Physa heterostropha...... Bee a Be ES Gh 412: cag
GOREAU, THOMAS F., AND oral. GorEau SNCS: Soe a
The physiology of skeleton formation in corals. Ill. Calcification ==
rate as a function of colony weight and total nitrogen eontent in the a eo
reef coral Manicina areolata (Linnaeus).,..-.......... ts Save eee be
LYNN, W. GARDNER, AND JAMES NORMAN DENT a
The action of various goitrogens in inhibiting localization of rdiv= wa
‘todine in the thyroid and thymus glands of larval tree a Ne } +} (BO pe
“ Ee ee
METZ, CHARLES B. i p | Ces
Investigation of the fertilization inhibiting action of Arbacia Yeas ea: oa
Serrefion.. <6 TER A A Gr eG Seve tahi ss he
SPIEGEL, MELVIN | j oe hg
Protein changes in development: ..0....00.....6 00.0 e nee J. 451
TRIPLETT, EDWARD L., AND SUSANNE BARRYMORE ;
Tissue specificity in embryonic and adult Cymatogaster aperegata ae
studied by scale transplantation....... wait, Met oa Tete SS 620 asics ye 463. Rs;

TRIPLETT, EDWARD L., AND SUSANNE D. BARRYMORE

Some aspects of osmoregulation in embryonic and adult Cymatogaster 7
ageregata and other embiotocid fishes.............. Le ENR: 412

Ris
1 a
“a R 1 }
7 hear
.H f
de
= ’
hy > &
é
fs ve
' aie
j 7 We 7 e “ae Y
, wy }
' 3 q
- £ - f
™ ;
LA f
cAU NW
q §
.
ba |
“4
\
:
i
nl
| G
te,
mY
~
i
i
y
:

ie i
, f
A . fad |
7
j
if
“yl
5 . 5
;
N ae
A 4
\
—— i
. rT.
I Vil Pepe wy
fi ¥
i,
‘
f
,
fj
2
7
iy
ff :
i P

775%

( ‘ a WpeRtins un iia Vie
| Hh EMRE ak Seay

6. Oe ae Ral

So.

/
“Fay wy wee ee

"WOW

3 9088 01228 2919