: .

i!!jj$!

11:

•

THE

BIOLOGICAL BULLETIN

PUBLISHED BY

THE MARINE BIOLOGICAL LABORATORY

Editorial Board

W. C. ALLEE, University of Florida A. K. PAKPART, Princeton University

L. R. BLINKS, Stanford University BERTA SCHARRER, University of Colorado

K. W. COOPER, University of Rochester ALBERT TYLER, California Institute of Technology

L. V. HEILBRUNN, University of Pennsylvania JOHN H. WELSH, Harvard University

M. E. KRAHL, University of Chicago DOUGLAS WHITAKER, Stanford University

E. T. MOUL, Rutgers University RALPH WlCHTERMAN, Temple University

DONALD P. COSTELLO, University of North Carolina
Managing Editor

VOLUME 106

FEBRUARY TO JUNE, 1954

Printed and Issued by

LANCASTER PRESS, Inc.

PRINCE & LEMON STS.

LANCASTER, PA.

11

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

Subscriptions and similar matter should be addressed to The
Biological Bulletin, Marine Biological Laboratory, Woods Hole,
Massachusetts. Agent for Great Britain : Wheldon and Wesley,
Limited, 2, 3 and 4 Arthur Street, New Oxford Street, London,
W. C. 2. Single numbers $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 15 and September 1, and to Dr.
Donald P. Costello, Department of Zoology, University of North
Carolina, Chapel Hill, North Carolina, during the remainder of
the year.

Entered as second-class matter May 17, 1930, at the post office at Lancaster,
Pa., under the Act of August 24, 1912.

LANCASTER PRESS, INC., LANCASTER, PA.

CONTENTS

No. 1. FEBRUARY, 1954

ARVY, L., AND M. GABE

The intercerebralis-cardiacum-allatum system of some Plecoptera 1

CABLE, R. M.

A new marine cercaria from the Woods Hole region and its bearing on
the interpretation of larval types in the Fellodistomatidae (Trematoda :
Digenea) 15

CHILD, C. M.

Redox indicator patterns in relation to echinoderm exogastrulation. II.
Reduction patterns 21

FRASER, RONALD C.

The utilization of some carbohydrates by in vitro cultured chick blas-
toderms in wound healing 39

HARRISON, JOHN R., AND IRVING KLEIN

Effects of lowered incubation temperature on the growth and differen-
tiation of the chick embryo 48

MATSUMOTO, KUNIO

Neurosecretion in the thoracic ganglion of the crab, Eriocheir japonicus 60

NICOLL, PAUL A.

The anatomy and behavior of the vascular systems in Nereis virens and
Nereis limbata 69

OKAZAKI, KAYO, AND KATSUMA DAN

The metamorphosis of partial larvae of Peronella japonica Mortensen,

a sand dollar 83

RAY, DAVID T., AND P. W. WHITING

An x-ray dose-action curve for eye-color mutations in Mormoniella. . . . 100

SCHILLER, EVERETT L.

Studies on the helminth fauna of Alaska. XVII. Notes on the inter-
mediate stages of some helminth parasites of the sea otter 107

SLIFER, ELEANOR H.

The permeability of the sensory pegs on the antenna of the grasshopper
(Orthoptera : Acrididae) 122

THOMAS, LYELL J ., JR.

The localization of heparin-like blood anticoagulant substances in the
tissues of Spisula solidissima 129

No. 2. APRIL, 1954

CHADWICK, L. E., J. B. LOVELL AND V. E. EGNER

The relationship between pH and the activity of cholinesterase from
flies. 139

in

68990

iv CONTENTS

FRAENKEL, G., AND GLENN E. PRINTY

The amino acid requirements of the confused flour beetle, Tribolium
confusum, Duval 149

HEILBRUNN, L. V., ALFRED B. CHAET, ARNOLD DUNN AND WALTER L.

WILSON
Antimitotic substances from ovaries 158

MONROY, A., L. Tosi, G. GIARDINA AND R. MAGGIO

Further investigations on the interaction between sperm and jelly-coat

in the fertilization of the sea urchin egg 169

MOOREFIELD, HERBERT H., AND G. FRAENKEL

The character and ultimate fate of the larval salivary secretion of
Phormia regina Meig. (Diptera, Calliphoridae) 178

NORTH, WHEELER J.

Size distribution, erosive activities, and gross metabolic efficiency of
the marine intertidal snails, Littorina planaxis and L. scutulata 185

RYTHER, JOHN H.

The ecology of phytoplankton blooms in Moriches Bay and Great South
Bay, Long Island, New York 198

SCHNEIDERMAN, HOWARD A., AND CARROLL M. WILLIAMS

The physiology of insect diapause. VIII. Qualitative changes in the
metabolism of the Cecropia silkworm during diapause and development 210

SCHNEIDERMAN, HOWARD A., AND NED FEDER

A respirometer for metabolic studies at high gaseous pressures 230

SCHNEIDERMAN, HOWARD A., AND CARROLL M. WILLIAMS

The physiology of insect diapause. IX. The cytochrome oxidase system

in relation to the diapause and development of the Cecropia silkworm . 238

WlCHTERMAN, RALPH, AND FRANK H. J. FlGGE

Lethality and the biological effects of x-rays in Paramecium : Radiation
resistance and its variability 253

No. 3. JUNE, 1954

BODINE, JOSEPH HALL, AND WILLIAM LIONEL WEST

v Effect of adenosinetriphosphate (ATP) on the endogenous oxygen up-
take of developing grasshopper embryos 265

BRENT, MORGAN M.

Nutritional studies on the amoebo-flagellate, Tetramitus rostratus 269

BROOKS, SUMNER C., AND EDWARD L. CHAMBERS

The penetration of radioactive phosphate into marine eggs 279

CHAMBERS, EDWARD L., AND WILLIAM E. WHITE

The accumulation of phosphate by fertilized sea urchin eggs 297

BROWN, FRANK A., JR., MILTON FINGERMAN AND MARGARET N. HINES

A study of the mechanism involved in shifting of the phases of the endog-
enous daily rhythm by light stimuli 308

GRANT, PHILIP

The distribution of phosphorus (P31 and P32) in dorsal and ventral
halves of the Rana pipiens gastrula 318

CONTENTS v

LANE, CHARLES E., J. Q. TIERNEY AND R. E. HENNACY

The respiration of normal larvae of Teredo bartschi Clapp 323

LASKER, REUBEN, AND ARTHUR C. GIESE

Nutrition of the sea urchin, Strongylocentrotus purpuratus 328

McSHAN, W. H., SOL KRAMER AND VERA SCHLEGEL

Oxidative enzymes in the thoracic muscles of the woodroach, Leucophaea
maderae 341

RAO, K. P.

Tidal rhythmicity of rate of water propulsion in Mytilus and its modifi-
ability by transplantation 353

VERNBERG, F. JOHN

The respiratory metabolism of tissues of marine teleosts in relation to
activity and body size 360

WEBB, H. MARGUERITE, MIRIAM F. BENNETT AND FRANK A. BROWN, JR.
A persistent diurnal rhythm of chromatophoric response in eyestalkless
Uca pugilator 371

Vol. 106, No. 1 February, 1954

THE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

THE INTERCEREBRALIS-CARDIACUM-ALLATUM SYSTEM

OF SOME PLECOPTERA ]

L. ARVY AND M. GABE

Laborattiirc d' Anatomic ct Histoloyic contparccs, Sorhoinic, Paris, hrancc

In spite of the interesting taxonomic position of the Plecoptera and the biological
peculiarities of these insects their endocrine glands have been studied only by a
small number of investigators. The existence of corpora allata in Pcrla iiia.i-iina
Scop, was first reported by Nabert (1913). In Xcinnm rallicitlaria. \Yu (1923)
described "lateral ganglia" which perhaps correspond to the corpora cardiaca.
Hanstrom (1940) gave the first exact description of the endocrine glands in four
species of Plecoptera (Nemura varieyata Oliv., Chloroperla 1'ircns Zett.. Isoptcry.v
bunncistcri Pictet, and Pcrla ccphalotcs Curt.). He showed that the anatomy of
the endocrine glands of Nemura differs greatly from that of the three other species.
In Nemura the medially located corpora cardiaca are fused with an unpaired lateral
corpus allatum. while the three other species possess symmetrical paired corpora
allata which are connected with the corpora cardiaca by anatomically defined nervi
corporis allati. Histologically, according to Hanstrom, the corpora carcliac-i of
the Plecoptera contain cells whose general appearance is that of neurons but whose
glandular function seems probable on account of the presence of fuchsinophilic
secretory granules. Likewise, the glandular function of the corpora allata is in-
dicated by the occurrence of acidophilic secretory granules. Regarding the in-
nervation of these endocrine glands, Hanstrom found, in the four species of
Plecoptera studied, the two pairs of protocerebral nerves whose existence he had
demonstrated in other Pterygota, and indicated that their cells of origin have the
same location as in other insects. No new data have been added to this description
in the general survey on the subject by Cazal (1948).

This brief summary indicates that the endocrine glands in the head region of the
Plecoptera are only incompletely known. The study of a larger number of species
seems desirable because of the pronounced anatomical differences in the representa-
tives of this order studied by Hanstrom (1940). Furthermore, none of the papers
quoted above contain any information on two important histophysiological problems
recently brought to light, i.e.. the relationships of the endocrine glands with the

1 Translated from the French by Dr. Berta Scharrer, University of Colorado School of
Medicine, Denver.

1

L. ARVY AND M. GABE

neurosecretory cells, and the changes which these glands undergo in the course of
post-embryonic development.

We have, therefore, undertaken a study of the endocrine glands of the head
region in representatives of seven families of the Plecoptera of the European fauna.2
In the present paper we shall report new data regarding the ''organ system" formed
by the neurosecretory cells of the pars intercerebralis, the corpora cardiaca and the
corpora allata (Scharrer and Scharrer, 1944; B. Scharrer, 1952).

MATERIAL AND METHODS

We were able to examine numerous specimens, in different stages of post-
embryonic development, belonging to the following species (classification and
nomenclature according to Aubert, 1946) :

Perlodidae Perlodes intricata Pict.

Perl-odes mortoni Klap.

Isogenus alpinus Pict.

Isogenus jontium Ris.
Perlidae Per la maxima Scop.

Perla marginata Panz.

Perla cephalotes Curt.

Perla carlukiana Klap.
Chloroperlidae Isoperla grammatica Scop.

Isopcrla rivulorum Pict.
Taeniopterygidae Brachyptera risi Morton

Rhabdiopteryx alpina Kuht.

Capniidae Capnioneura nemuroides Ris.

Leuctridae Lcuctra hippopus Kemp.

Leuctra inermis Kemp.
Nemuridae N cm lira mortoni Ris.

Nemura marginata Ris.

Nemura intricata Ris.

Nemura praccox Morton

Nemura nimborum Ris.

Nemura lateral-is Ris.

The tissues were fixed in Bouin, Duboscq-Brazil, or Carnoy. The material was
embedded in celloidin-paraffin and cut serially at 5 and 7 p. Among stains for
general survey we have used especially hemalum-picroindigocarmine, the triple
stain of Prenant (as modified by Gabe and Prenant, 1949), and azan. The neuro-
secretory cells can be well demonstrated with the latter method, but the study of the
migration of the neurosecretory product along the axons is greatly facilitated by
the use of the chrome hematoxylin-phloxine method of Gomori (1941). Further-
more, we have employed the method of Brachet for the histochemical determination
of ribonucleic acid, the method of Hotchkiss-McManus for the demonstration of
polysaccharides, and Best's carmine stain with the saliva test for glycogen.

2 We are obliged to Dr. J. Aubert, Musee zoologique, Lausanne, Switzerland, and to Dr.
T. T. Macan, Ambleside, Westmorland, England, for supplying us with well preserved material.

NEUROSECRETORY SYSTEM OF PLECOPTERA

RESULTS

The information regarding the intercerebralis-cardiacum-allatum system of the
Plecoptera obtained in the present study concerns (a) the anatomy of the endocrine
glands, (b) the relationships of these glands with the neurosecretory cells, and (c)
the development of these glands of internal secretion in the course of post-
embryonic life.

/. Anatomy of the endocrine glands of the head ret/ion

It is known since the work of Hanstrom (1940) that certain Plecoptera have
paired symmetrical corpora allata, while others have an unpaired lateral corpus
allatum. To these two we can add a third type, characterized by a median un-
paired corpus allatum (Arvy and Gabe, 1953b).

(a) Type: Chloroperla. Under this category Hanstrom classifies animals which
are characterized by the existence of paired corpora allata, i.e. Chloroperla virens,
Isopteryx burmeisteri and Perla cephalotes. The examination of a more extensive
material permits us to state that the classification of Hanstrom is valid for all
Perlodidae, Perlidae, and Chlproperlidae which we were able to examine.

In all representatives of this type the corpora cardiaca are fused in the midline
and surround the dorsal vessel ; they receive two pairs of nerves from the proto-
cerebrum, the nervi corporis cardiaci I and II, whose cells of origin lie in the pars
intercerebralis and next to the corpora pedunculata. The corpora cardiaca give
rise to two nervi corporis allati which are rather short, but thick. The ventral
portion of the corpora cardiaca is fused with the hypocerebral ganglion which re-

C.2.

--JI.O

c.i.

A.

FIGURE 1. Diagrams of the three anatomical types of the retrocerebral glandular complex in
the Plecoptera. A. Type of Chloroperla (according to Hanstrom, 1940), found in Perlodidae,
Perlidae, and Chloroperlidae ; B. Type of Nemura (according to Hanstrom, 1940), found in
Capniidae, Leuctridae, and Nemuridae ; C. Type of Brachyptera (according to Arvy and Gabe,
1953b), found in Taeniopterygidae. a., corpus allatum; c.l., nervus corporis cardiaci I; c.2.,
nervus corporis cardiaci II; n.o., esophageal nerve; n.p., prothoracic nerve; r., recurrent nerve.

L. ARVY AND M. GABE

FIGURES 2-5.

XEUROSECRETORY SYSTEM OF PLECOPTERA

ceives the recurrent nerve and sends off the median unpaired esophageal nerve.
From each corpus allatum a good-sized nerve originates which traverses the
posterior part of the head, receives a branch from the connective which links the
subesophageal and the prothoracic ganglia and branches in the prothorax (Figs.
1A. 10).

(/>) Type: Nemuro. The anatomy of the endocrine glands of the head of
Xcinura varicgata is quite different (Hanstrom, 1940). According to our studies
this type of organization also applies to the Capniidae, the Leuctridae and the
Nemuridae. The corpora cardiaca, fused in the midline, also in this group receive
the same nerves from the protocerebrum as those of the insects of the Chloroperla
type. The unpaired corpus allatum lies asymmetrically at the right side and is
intimately connected with the fused part of the corpora cardiaca and the hypocerebral
ganglion. There are no anatomically defined nervi corporis allati. From the
posterior end of this organ complex arise two esophageal nerves which supply the
stomodaeum (Fig. IB).

(r ) T\pe: Brachyptcra. This type was found only in representatives of the
Taeniopterygidae and resembles that of Nemura with which it has in common the
anatomy of the corpora cardiaca which are fused in the midline and receive the
same innervation from the protocerebrum. The corpus allatum is unpaired but
lies exactly medially (Fig. 2). As in Xemura, it is fused with the corpora cardiaca
and the hypocerebral ganglion. The esophageal nerve arising from the caudal
extremity of this organ complex is unpaired and median (Fig. 1C).

//. Relationships betiveen the endocrine glands and the ncitrosccrctory cells

The morphological peculiarities of the cells of origin of the nervi corporis
cardiaci I of the Plecoptera are known from the work of Hanstrom (1940). This
author emphasizes the acidophilia of the cytoplasm of these cells, compares them
with the elements of the same type which give rise to the nervi corporis cardiaci I
in the Palaeoptera, and homologizes them with the paired frontal organs of the
Apterygota. These cells possess all the morphological attributes of the neuro-
secretory cell as defined by Scharrer (for the bibliography see Scharrer and
Scharrer, 1954). They elaborate an acidophilic product which stains with iron
hematoxylin, with azocarmin and with chrome hematoxylin (method of Gomori).
The secretory product passes along the axons arising from these cells. The course
of the fibers can be followed with particular ease in preparations stained with
chrome hematoxylin-phloxine. Comparable in their major outlines to the course

FIGURE 2. Section through caudal portion of cerebral ganglion, showing also corpora
cardiaca (in center) and nervi corporis cardiaci I, in a larva of Pcrla carlitkiana. Bouin. chrome
hematoxylin-phloxine, X 250. Note neurosecretory cells and accumulation of neurosecretory
product in the corpora cardiaca and their nerves.

FIGURE 3. Neurosecretory cells in the pars intercerebralis of a larva of Perl odes inortom.
Bouin, chrome hematoxylin-phloxine, X 1000. Note abundance of neurosecretory product in the
cells and granules of the same material along the axons.

FIGURE 4. Frontal section through fused portion of corpora cardiaca in a larva of medium
age of Brachyptcra risi. Bouin, chrome hematoxylin-phloxine, X 1000. Accumulation of neuro-
secretory product between the cells.

FIGURE 5. Detail from Figure 2, X 1000. Accumulation of neurosecretory product in
nervus corporis cardiaci I (bottom, left) and in corpus cardiacum.

L. ARVY AND M. GABE

FIGURES 6-9.

NEUROSECRETORY SYSTEM OF PLECOPTERA

of the nervi corporis cardiaci I of the Palaeoptera, the corresponding nerves in
the Plecoptera show certain peculiarities in their anatomy. The cells of origin
occupy a caudal position (Fig. 2), and the nervi corporis cardiaci I which cross
the midline in the anterior portion of the protocerehrum leave the cerebral ganglia
shortly after the decussation so that they accomplish a relatively long extra-
ganglionic course on the ventral surface of the cerebral ganglion before entering the
corpora cardiaca. The techniques used show in the intraganglionic portion of the
nervi corporis cardiaci I very fine granules along the axons (Fig. 3). The
extraganglionic portion of the nerves is much richer in secretory products, which
accumulate markedly at the point where the nervi corporis cardiaci I change their
course (Figs. 2, 5 and 6). In representatives of the type of Chloroperla the
neurosecretory material appears in form of elongated and knotty masses, of big
granules and droplets. This accumulation of the neurosecretory product where
the fiber bundles arising from neurosecretory cells change their direction represents
a rather frequent occurrence according to E. Scharrer (personal communication).

The neurosecretory product can be demonstrated between the cells of the
corpora cardiaca. The study of preparations stained with Gomori's chrome
hematoxylin method suggests also here an arrangement along the nerve fibers.
We have never found a trace of the substance stainable with chrome hematoxylin
within the cells of the corpora cardiaca. The small secretory granules occurring in
these elements stain intensely with phloxine (Figs. 4-8).

In the species of which Chloroperla represents the type and which possess
anatomically well defined nervi corporis allati, the neurosecretory substance is very
abundant in these nerves. One can trace it without the slightest difficulty to the
corpora allata, and in preparations stained with the Gomori technique its destination
can be observed (Figs. 7, 9, 13, 14, 15). The majority of the fibers of the nervus
corporis allati, which are neatly outlined by the secretory product, ramify under
the connective tissue capsule of each corpus allatum. From these subcapsular
plexus, clearly defined by the accumulation of the neurosecretory material, issue
fibers also charged with neurosecretory material which ramify between the allatum
cells. As in the case of the corpora cardiaca, this product remains extracellular.
The transport of the neurosecretory material does not terminate in the corpora
allata. The nerves which originate from them and run to the prothorax also con-
tain a greater or less amount of the material stainable with chrome hematoxylin.

In the species whose cephalic endocrine glands belong to the types of Nemura
and Brachyptera, the passage of the neurosecretory product into the corpus allatum
appears less pronounced. As a matter of fact, there exists no anatomically defined

FIGURE 6. Corpus cardiacum of an old larva of Pcrla maryinata. Bouin, chrome hema-
toxylin-phloxine, X 1000. Neurosecretory product between the cells.

FIGURE 7. Frontal section through retrocerebral glandular complex of an old larva of
Nemura mortoni. Bouin, chrome hematoxylin-phloxine, X 1000. Esophageal \vall (bottom),
corpora cardiaca (above it), and corpus allatum (to the right). Presence of neurosecretory
product between the cells of the corpora cardiaca and of the corpus allatum.

FIGURE 8. Frontal section through hypocerebral ganglion and corpora cardiaca of a larva
of Lcuctra iiicnnis. Bouin, chrome hematoxylin-phloxine, X 1000. Neurosecretory product be-
tween the cells of the corpora cardiaca.

FIGURE 9. Nervus corporis allati and corpus allatum of a young larva of Pcrla carlukiana.
Bouin, chrome hematoxylin-phloxine, X 1000. Accumulation of neurosecretory product in the
nervus corporis allati (top, left) and under capsule of corpus allatum.

8

L. ARVY AX I) M. GABE

v *•• *-
* • f^,

Mr

// •

74

FIGURES 10-15.

NEUROSECRETORY SYSTEM OF PLECOPTERA

nervus corporis allati. The corpus allatuin is innervated by a certain number of very
thin fibers which are accompanied by a substance staining with Gomori's hema-
toxylin. but which ramify directly among the cells of the corpus allatum without
forming a subcapsular plexus, so that there exists no real accumulation of the
neurosecretory material. Only the study of sections with very powerful magnifica-
tions shows the existence of a phenomenon which, although being less spectacular
than in representatives of the type of Chloroperla, probably possesses the same-
physiological significance.

///. Development of the endocrine glands in the course of post-embryonic life

A comparison of the endocrine glands in various stages of post-embryonic de-
velopment shows the following facts :

(a) The neurosecretory cells of the pars intercerebralis reach their maximum
activity at a stage considerably ahead of the imaginal molt. In the larva whose
wing buds are still far from having reached their maximal development one finds
the most pronounced transport of neurosecretory substance along the nervi corporis
cardiaci and the greatest accumulation of this substance in the corpora cardiaca and
allata. The secretory activity of the pars intercerebralis is less noticeable in older
larvae and in the imago.

(b) The corpora cardiaca reach their maximal size at a larval stage which
corresponds to the maximal secretory activity of the pars intercerebralis. Their
volume remains stationary in the older larva and starts to diminish after the adult
stage is reached.

(c) The corpora allata develop in the same fashion. Their maximal size
falls into the middle of the larval life. During the second half of the post-embryonic
period these organs undergo a pronounced atrophy, which is the more clearly
visible since it coincides in time with the increase in size of all organs in the
head region other than the corpora cardiaca. The size of the corpora allata of
the imago is substantially smaller than that of larvae which are about halfway
through their post-embryonic development.

FIGURE 10. Cross section through caudal portion of cerebral ganglion of larva of medium
age of Isopcrla graininatica. Dubosq, azan, X 250. Fused corpora cardiaca (center) and
corpora allata (on either side of upper third of esophagus).

FIGURE 11. Cross section through head of a larva of medium age of Brachyptera risi.
Carnoy, Prenant's triple stain, X 100. Esophagus in center, above it the unpaired medial corpus
allatum, above the corpus allatum the common, fused portion of corpora cardiaca.

FIGURE 12. Detail of Figure 11, X 1000. Note mitosis in corpus allatum.

FIGURE 13. Nervus corporis allati (at left) and beginning of corpus allatum in an old
larva of Pcrla mqrginata. Bouin, chrome hematoxylin-phloxine, X 1000. Note accumulation of
neurosecretory product in nervus corporis allati and non-neurosecretory ganglion cell at entrance
of nerve into the corpus allatum.

FIGURE 14. Nervus corporis allati and corpus allatum in a larva of medium age of Pcrla
carlukiana. Bouin, chrome hematoxylin-phloxine, X 1000. Accumulation of neurosecretory
product in nerve and subcapsular plexus, as well as in thin fibers penetrating corpus allatum.

FIGURE 15. Nervus corporis allati and corpus allatum of a larva of medium age of Pcrla
ccphalotcs. Technique and magnification as in Figure 14. Note abundance of neurosecretory
product in the nerve and between the cells of corpus allatum.

10 L. ARVY AND M. GABE

DISCUSSION

From the anatomical point of view, the differences between the various
Plecoptera studied are even more marked than was expected from the work of
Hanstrom (1940). One should perhaps mention here that the anatomy of the
endocrine glands is always the same in all representatives of the same family, and
that the grouping of the different families according to the morphology of their
retrocerebral glandular complex results in an arrangement which is in agreement
with present concepts regarding the taxonomy and phylogeny of the Plecoptera.

Concerning the structure of the corpora cardiaca and allata, certain data reported
in this paper correspond to facts well established in other insects. Thus, the sig-
nificance of the connection between the cells of the pars intercerebralis and the
corpora cardiaca has been known since Hanstrom (1940). B. and E. Scharrer
(1944) compared this neurosecretory system with the hypothalamic-hypophyseal
system of the vertebrates. The newer techniques for the demonstration of the
neurosecretory product, in particular the chrome hematoxylin-phloxine method,
greatly facilitate the study of this neurosecretory pathway, and the accumulation of
the neurosecretory product in the corpora cardiaca described by B. Scharrer (1951)
in Leucophaca maderae, was confirmed in all insects studied with sufficiently selec-
tive methods. We should like to point out in this connection that the use of the
chrome hematoxylin-phloxine method permits one to correct an error of interpreta-
tion due to the use of unsuitable techniques, i.e., the description of "pseudopodial
processes" of the "chromophile cells" of the corpora cardiaca by Cazal (1948).
These "processes" are apparently nothing but accumulations of neurosecretory
products along nerve fibers.

The existence, in the corpus cardiacum cells themselves, of a second secretory
product which differs from that of the neurosecretory cells of the protocerebrum
has been debated for a longer time than the accumulation of the glandular product
of the protocerebrum. This secretory activity on the part of the corpora cardiaca
themselves exists, however, in a variety of insects. It was reported in the Thysanura
(Gabe, 1953a), the Ephemeroptera (Arvy and Gabe, 1952a), the Odonata (Arvy
and Gabe, 1952b), in Leucophaea maderae (B. Scharrer, personal communication),
Carausius morosus (Stutinsky, 1952), and Bomby.v mori (Arvy, Bounhiol and
Gabe 1953a) ; the case of the Plecoptera constitutes another example. The par-
ticular timing of this secretory process, whose physiological significance was recently
discussed by Wigglesworth (1954), explains why its product cannot be as easily
demonstrated as the accumulation of the neurosecretory material. In fact, the
corpora cardiaca are actively secreting only at well defined periods in the post-
embryonic development. Studies, as yet unpublished, showed that the secretory
activity of the endocrine glands of the Myrmeleonidae is sharply restricted to
certain periods.

Attention should be called to the fact that the existence of two separate secretory
products in the corpora cardiaca corresponds to that in other endocrine glands of
arthropods. The sinus gland of the crustaceans (Malacostraca) in which the
neurosecretory product furnished by the x-organ of Hanstrom and by other neuro-
secretory cells of the central nervous system is stored, contains in decapods (Gabe,
1952a) and isopods (Gabe 1952b, 1952c) a second product of secretion. This
substance is formed in loco and differs from the neurosecretory product in its

NEUROSECRETORY SYSTEM OF PLECOPTERA

chemical constitution. Similarly, the brain gland of the Chilopoda which receives
the product of the neurosecretory cells of the protocerehrum contains a second
secretory product formed in the cells of the organ themselves and different from
the first in its histological and histochemical characteristics (Gabe, 1952d, 1953b).
This structural analogy of the three principal neurosecretory systems in arthropods,
recently emphasized by one of us (Gabe, 1953c), suggests a comparison with the
hypothalamic-hypophyseal system of the selachians (E. Scharrer, 1952) ; in these
the product of the neurosecretory cells of the preoptic nucleus accumulates in the
terminals of the hypothalamo-hypophyseal tract between the cells of the inter-
mediate lobe which possesses its own secretory activity.

The transport of neurosecretory material to the corpora allata and its accumula-
tion between the cells of this organ deserves special emphasis because it represents
the first example among Heterometabola of a phenomenon recently described in
Bomby.v mori (Arvy, Bounhiol and Gabe, 1953a). The extension of the neuro-
secretory pathway, which begins in the pars intercerebralis, to the corpora allata
speaks in favor of the existence of relationships between the neurosecretory cells
of the protocerebrum and the corpora allata (B. Scharrer, 1952; Thomsen, 1952).

The transport of the neurosecretory product in the nerves leaving the corpora
allata in the Plecoptera of the Chloroperla type seems to show that the product of
the cells of the pars intercerebralis can, in certain cases, reach thoracic organs.
This observation is related to the presence of neurosecretory material in the aortic
nerves arising from the corpora cardiaca of the Thysanura (Gabe, 1953a), and to
the existence of this substance in the esophageal nerve of CallipJwra ery throe ephala
(Thomsen, 1954).

The study of the post-embryonic development shows that the relations of the
corpora cardiaca and allata to the neurosecretory cells do not merely represent an
anatomical peculiarity. The maximal size of the corpora cardiaca and the onset
of their own secretory activity coincide with the maximal abundance of the neuro-
secretory product between the cells and the appearance of numerous vacuoles in
the cells of origin of the nervi corporis cardiaci I. The largest size of the corpora
allata is reached at the same time raid seems to be correlated with the arrival of the
neurosecretory product.

The development of the corpora cardiaca throughout the post-embryonic life
of the Plecoptera is very different from that reported for other insects. The
corpora cardiaca reach their maximal size at the time of the imaginal molt in
Ephemeroptera (Arvy and Gabe, 1950, 1952a), in Odonata (Arvy and Gabe,
1952b), and in Panorpa coinnnniis L. (Schwinck, 1951), during pupation in
Ephestia kilhniella Zell. (Rehm, 1951) and in Bomby.v nwri (Arvy, Bounhiol and
Gabe, 1953a, 1953b). This difference in development whose physiological sig-
nificance cannot be determined except by experimentation seems to correspond to
a difference in timing regarding the neurosecretion in the protocerebrum. We
could show a time relationship between the maximal transport of the neurosecretory
product and the maximal size of the corpora cardiaca in the Ephemeroptera and
Odonata (Arvy and Gabe, 1952a, 1952b, 1953a). The same agreement exists in
Ephestia knhniella as shown by the measurements and descriptions of Rehm
(1951). Finally, in Boinby.r nwri the increase in the volume of the corpora
cardiaca takes place at the same time as a "discharge" of the neurosecretory product
of the cells of the pars intercerebralis (Arvy, Bounhiol and Gabe, 1953a, 1953b).

12 L. ARVY AXL) M. GABE

The development of the corpora allata shows the same feature. The maximal
size of these organs corresponds, in the Plecoptera, to a larval stage still far from
the imaginal molt ; these organs undergo atrophy in the second half of post-
embryonic development. In other insects whose corpora allata are innervated
by the protocerebrum the maximal size is reached in the imago. It coincides with
an intense secretory activity in the cells of the pars intercerebralis. In the
Ephemeroptera, the corpora allata develop as in the Plecoptera. Their maximal
size is also reached towards the middle of the post-embryonic period, and the
atrophy which follows corresponds to a reduction of secretion in the cells of origin
of the nervi corporis allati lying in the subesophageal ganglion in the Ephemeroptera,
and in the pars intercerebralis in the Plecoptera.

In general, the study of the modifications which the endocrine glands of the
head region of the Plecoptera undergo during post-embryonic life illustrates the
parallelism between the state of the endocrine glands and the secretory activity in
the cells of origin of the nerves which innervate these organs. This fact underlines
the important role of the neurosecretory phenomena in the physiology of these
insects.

SUMMARY

The histophysiological study of the intercerebralis-cardiacum-allatum system
in 21 species of Plecoptera resulted in the following observations:

1. From the anatomical point of view, the type of Chloroperla ( Hanstrom, 1940)
characterized by paired symmetric corpora allata, exists in the Perlodidae, Perlidae,
and Chloroperlidae. The type of Nemura, characterized by an unpaired, laterally
located corpus allatum, corresponds to the Nemuridae, Leuctridae, and Capniidae.
A third anatomical type of which an unpaired but definitely medial corpus allatum
is typical ( type of Brachyptera ) exists among the Taeniopterygidae.

2. The cells of origin of the nervi corporis cardiaci I of the Plecoptera possess
all the morphological characteristics of neurosecretory cells. Their secretory
product is stainable with acid dyes, azocarmiu, iron hematoxylin, and chrome
hematoxylin. This secretory product migrates along the axons and accumulates
between the cells of the corpora cardiaca. These elaborate a secretory product of
their own which stains with the phloxine of the method of Gomori.

3. The neurosecretory product migrates along the nervi corporis allati and
occurs between the cells of the corpora allata; one also encounters it in the nerves
which run from the corpora allata to the prothorax in the Plecoptera of the
Chloroperla type.

4. The secretory activity in the cells of the pars intercerebralis is at its peak
towards the middle of the larval period ; the phenomena of neurosecretion are less
pronounced in the later stages of larval life and in the imago.

5. The corpora cardiaca and allata reach their maximal volume in larvae which
are still a considerable period away from the imaginal molt. In the latter stages
of post-embryonic development atrophy of the endocrine glands of the head region
is observed. This mode of development is different from that described in other
insects belonging to the Neoptera; it must be understood in relationship with the
peculiar chronology of the neurosecretory activity of the Plecoptera.

NEUROSECRETORY SYSTEM OF PLECOPTKRA 13

LITERATURE CITED

ARVY, L.. J. J. Borxmm AND M. GAHE, 1953a. Deroulement de la neuro-secretion pro-

tocerebrale chez Boinh\.v niori L. au cours clu developpement post-embryonnaire. (.'. 7\.

Acad. Sci, Paris, 236 : 627-629.

ARVY, L., J. J. BOUNHIOL AND M. GABE, 1953b. Donnees histophysiologiques sur la neuro-
secretion chez Bomby.v inori L. et sur ses rapports avec les glandes endocrines. Bull.

biol. Prance Belyique (in press).
ARVY, L., AND M. GABE, 1950. Donnees histophysiologiques sur lo formations endocrines retro-

cerebrales chez les Ecdyonuridae (Ephemeropteres). Bull. Soc. zool. France, 75:

267-285.
ARVY, L., AND M. GABE, 1952a. Donnees histophysiologiques sur la neuro-secretion chez quelques

Ephemeropteres. La Cellule. 55 : 203-222.
ARVY. L.; AND M. GABE, 1952b. Donnees histophysiologiques sur les formations endocrines

retro-cerebrales de quelques Odonates. Ann. Sci. Nat. Zoo!., (//), 14: 345-374.
ARVY, L., AXD M. GABE, 1953a. Donnees histologiques sur la neuro-secretion chez les Insectes

Paleopteres. Zeitsehr. Zcllj. niikr. Aunt., 38: 591-610.
ARVY, L., AND M. GABE, 1953b. Donnees histophysiologiques sur les glandes endocrines

cephaliques de Brucliyptera risi Morton (Plecoptere). Arch. Zool. r.r/>. <>en. (in

press ) .
AUBERT, J., 1946. Les Plecopteres de la Suisse romande. Doctoral Thesis, Faculty of Science,

Lausanne, 128 pages.
CAZAL, P., 1948. Les glandes endocrines retrocerebrales des Insectes (Etude morphologique).

Bull. biol. France Beli/uine, suppl.. 32: 1-227.
GABE, M., 1952a. Particularites histochimiques de 1'organe de Hanstrom (organe X) et de la

glande du sinus chez quelques Crustaces decapodes. C. R. A cud. Sci., Paris, 235 : 90-92.
GABE, M., 1952b. Sur 1'existence d'un cycle secretaire dans la glande du sinus (organe

pseudofrontal) chez Oniscus asellns L. C. R. Acad. Sci. Pans, 235: 900-902.
GABE, M., 1952c. Particularites histologiques de la glande du sinus et de 1'organe X (organe

de Bellonci) chez Sphaero/na serratuin Fabr. C. R. Acad. Sci., Paris, 235: 973-975.
GABE, M., 1952d. Sur 1'emplacement et les connexions des cellules neuro-secretrices dans les

ganglions cerebroides de quelques Chilopodes. C. R. Acad. Sci.. Paris. 325: 1430-1432.
GABE, M., 1953a. Donnees histologiques sur les glandes endocrines cephaliques de quelques

Thysanoures. Bull. Soc. zool. France (in press).
GABE, M., 1953b. Particularites histologiques de la glande cerebrale de Scutigera coleoptrata

Link. Bull. Soc. cool. France (in press).
GABE, M., 1953c. Quelques acquisitions recentes sur les glandes endocrines des Arthropodes.

Experientla, IX 19: 352-356.
GABE, M., AND M. PRENANT, 1949. Contribution a 1'etude cytologique et histochimique du

tube digestif des Polyplacophores. Arch. Biologic. 60: 39-77.
GOMORI, G., 1941. Observations with differential stains on human islets of Langerhans.

Amcr. J. Path., 17: 395-406.
HANSTROM, B., 1940. Inkretorische Organe, Sinnesorgane und Nervensystem des Kopfes

einiger niederer Insektenordnungen. Kuiigl. Srciiskn retcnskupsakud. Hdl., (J)

18: 1-265.

NABERT, A., 1913. Die Corpora allata der Insekten. Zeitsehr. i^iss. Zool., 104: 181-358.
REHM, M., 1951. Die zeitliche Folge der Tatigkeitsrhythmen inkretorischer Organe von

Ephestia kiihniclla \vahrend der Metamorphose und des Imaginallebens. Arch. f.

Entn.:, 145: 205-248.
SCHARRER, B., 1951. The storage of neurosecretory material in the corpus cardiacum. Anat.

Rec.. Ill : 554-555.
SCHARRER, B., 1951. tiber neuroendokrine Vorgange bei Insekten. Pflii</ers Arch.. 255:

154-163.
SCHARRER, B., AND E. SCHARRER, 1944. Neurosecretion \"I. A comparison between the

intercerebralis-cardiacum-allatum system of the insects and the hypothalamo-hypophy-

seal system of the vertebrates. Biol. Bull.. 87 : 242-251.
SCHARRER, E., 1952. Das Hypophysen-Zwischenhirnsystem von Scvlliuin stcllarc. Zeitsehr.

Zellforsch. niikr. Anat.. 37 : 196-204.

14 L. ARVY AND M. GABE

SCHARRER, E., AND B. ScHARRER, 1954. Neurosekretion. v. Mollcndorffs Hdbcli. mikr. Anat.,

VI/5 (in press).
SCHWIXCK, I., 1951. Veranderungen der Epidermis, der Perikardialzellen und der Corpora

allata in der Larvenentwicklung von Panorpa coiniuniiis unter normalen und experi-

mentellen Bedingungen. Arch. f. Entzv., 145: 62-108.
STUTINSKY, F., 1952. Etude du complexe retro-cerebral de quelques Insectes avec 1'hema-

toxyline chromique. Bull. Soc. zool. France, 77 : 61-67.
THOMSEN, E., 1952. Functional significance of the neurosecretory brain cells and the corpus

cardiacum in the female blow-fly, Calliphora crythroccpliala Meig. /. E.vp. Biol.,

29: 137-172.
THOMSEN, E., 1954. Experimental evidence for the transport of secretory material in the

axons of the neurosecretory cells of Callipliora crythroccpliala Meig. Pubbl. Stas.

zool. Napoli, 25 (3) : (in press).
WIGGLESWORTH, V. B., 1954. Neurosecretion and the corpus cardiacum in insects. Pubbl.

Sfaz. zool. Napoli, 25 (3) : (in press).
\\'i:. C. F., 1923. Morphology, anatomy and ethology of Nemoura. Bit]!. Llo\d Libr., entoinol.

scr., n°3 : 1-51.

A NEW MARINE CERCARIA FROM THE WOODS HOLE REGION

AND ITS BEARING ON THE INTERPRETATION OF LARVAL

TYPES IN THE FELLODISTOMATIDAE

(TREMATODA: DIGENEA)

R. M. CABLE

Department of Biological Sciences, Purdue University, Lafayette, Indiana, and
The Marine Biological Laboratory, Woods Hole, Mass.

The concept of two great groups of digenetic trematodes as proposed by La
Rue (1926) is based in part on the type of cercaria which is furcocercous in the
order Strigeatoidea and non-furcocercous in the order Prosostomata. However,
the tails of cercariae are subject to such varied and extreme modifications in many
instances as to give little indication of the basic type. Unfortunately, the pioneer-
ing life history studies in some families concerned species having just such ex-
tremely modified larvae and hence did not reveal fundamental relationships, a
knowledge of which has had to await investigations dealing with more "typical"
representatives. Excellent examples of groups in which this situation has proved
to be the case are the families Brachylaemidae and Fellodistomatidae.

In the Fellodistomatidae, the first life history to be reported was that of
Bac tiger bac tiger which according to Palombi (1934) has a trichocercous cercaria
with a non-bifid tail. Subsequent studies have shown that at least some fello-
distomatids have furcocercous larvae as discussed in a recent paper (Cable, 1953)
which should have taken into account also the life cycle of FcUodistoinum jellis as
described by Chubrick (1952). In reporting for that species a furcocercous larva
developing in marine lamellibranchs, her observations lend further confirmation of
the view that the Fellodistomatidae properly belong in the order Strigeatoidea.
This view poses no difficulty in the interpretation of caudal structure of many fello-
distomatid cercariae in which the tail is reduced or even absent. Such instances
obviously are examples of caudal reduction associated with the abbreviation of
free-living activity now recognized to occur in various distantly related families.
It is this modification that has led to classifying larval trematodes into such un-
natural categories as Microcercous Cercariae and Cercariaca.

The interpretation of the cercaria of Bacciger baetigcr and similar larvae in
respect to the furcocercous type is quite another matter. In such larvae, the tail
is well developed without possessing furcae and has paired lateral setaceous tufts, the
elements of which may be joined by delicate webbing to form finlets. Such a
larva was among the fellodistomatid cercariae found by the writer in Puerto Rico
and figured in a recent paper (Cable, 1953). That cercaria was found in but
one clam, before the other larvae included in that paper were seen and hence
before the significance of determining the precise relationship of the excretory
system to the tail was appreciated. Fortunately, the writer was aware that a very
similar cercaria occurs in the Woods Hole region and a brief visit to that area was
made during the summer of 1953 to study the species. It was quickly found,
thanks to Prof. P. S. Crowell, Jr. and students of the Invertebrate Zoology class

15

Cere aria laevicardii sp. nov. (Figs. 1-4)

EXPLANATION OF PLATE I

(All figures concern Cercaria laevicardii)

16

A NEW MARINE CERCARIA 17

through whose assistance over 200 specimens of the clam serving as the host were
made available. As the larva has not yet been reported, it is here described and
named after the host.

Specific diagnosis: distome, non-ocellate, trichocercous cercaria developing in
simple sporocysts in the visceral mass of a marine lamellibranch. Body of cercaria
yellowish in life, tail with 28 pairs of slender, lateral finlets, each composed of setae,
usually 10 in number, united by a delicate web; finlets shorter at each end of tail
and closer together at its proximal end. Associated with bases of finlets on each
side is a uniform row of similar nuclei which are prominent in stained specimens.
Dorsal and ventral caudal fins are absent; tail widest at about mid-length and
rather blunt posteriorly. Entire body and tail finely spinose, cuticle of body thick
and with fine striae. Suckers about equal with the ventral sucker embedded in
a prominent protrusion somewhat anterior to mid-level of body and overhanging a
ventral depression of the fore body. Oral sucker not at extreme anterior end of
body, with the mouth opening ventrally. Prepharynx extremely short, pharynx
well developed, esophagus about as long as pharynx and receiving just anterior
to ventral sucker the ceca which reach about halfway between ventral sucker and
posterior end of body; ceca thick-walled and with somewhat inflated blind ends.
Cephalic and cystogenous glands not evident. Primordia of reproductive system
well developed ; testes dorsal, symmetrical, and just posterolateral to ventral
sucker ; ovary more ventral, median, and just posterior to level of testes ; other
primordia are represented by strands of nuclei and a prominent mass posterodorsal
to ventral sucker. Excretory vesicle U- or almost V-shaped, with wide arms
extending anterior to ventral sucker and with large, refractile concretions. From
each arm of the excretory vesicle, a ciliated recurrent tubule extends posteriorly
to about mid-level of body and receives an anterior and a posterior collecting
tubule, each of which is joined by two groups of flame cells, three cells per group.
The excretory formula is accordingly 2 [(3 + 3) + (3 + 3)]:= 24 flame cells.
A distinct bladder sphincter is present and from it the prominent caudal excretory
tubule extends the length of the tail, bifurcating just before reaching the pair of
embryonic excretory pores at the posterior tip of the tail. Measurements in milli-
meters of cercariae killed in hot sea water, mounted without pressure, and selected
for specimens showing a minimum of body flexure are as follows : body length
0.185-0.243, maximum body width 0.10-0.13; tail 0.426-0.517 long and 0.050-
0.054 in maximum width exclusive of appendages. Caudal finlets range in length
from 0.095 near base of tail to a maximum of about 0.30 elsewhere. Oral sucker
0.035-0.040 in diameter, usually a little wider than long; ventral sucker 0.038-
0.040 and pharynx 0.022-0.025 in diameter. Sporocysts elongate, young ones
with pointed ends which are very motile ; older sporocysts up to 3.0 in length,
rounded or truncate posteriorly and with a pointed anterior end bearing a terminal
birth pore.

FIGURE 1. Entire cercaria in ventral view, drawn to scale from a heat-killed specimen
with internal structures added from observations on living and stained larvae.
FIGURE 2. Detail of caudal finlet.

FIGURE 3. Sketch of cercaria to show resting attitude near surface of water.
FIGURE 4. Embryology of the excretory system.

18 R. M. CABLE

Host : Lacvicardiutn nwrtoni Conrad.

Locality: Lagoon Pond, Martha's Vineyard, Massachusetts.

Incidence of infection: 25-33 % of clams collected in August, selected for large
size, and opened for examination.

Although over 200 clams were isolated in bowls of sea water for 48 hours, no
cercariae emerged spontaneously. In some of the clams then opened, the infection
was immature but from the visceral mass of others, large numbers of evidently
fully developed cercariae escaped and remained alive over 24 hours. They swam
energetically, tail-first with the body bent ventrally on the base of the tail. The
larvae made rapid progress, often swimming somewhat erratically in one direction
and then spinning around before coming to rest near the surface of the water with
the body downward and the tail contracted into a coiled mass (Fig. 3). During
rest periods, which were frequent, the body would contract and expand and on
several occasions cercariae were observed creeping upside down in an inch-worm
fashion with the suckers attached to the surface film. No photactic behavior
was observed.

Several species of cercariae resembling C. laevicardii have been described, mostly
by earlier workers whose accounts are so inadequate that a critical evaluation
of them is impossible. Dollfus (1925) gave a summary of trichocercous larvae
known at that time, dividing marine species into two groups, one in which eye-
spots are present, and one in which they are absent. Subsequent studies have
revealed that such a distinction may be an artificial one, for instances are known
in which one cercaria may be ocellate whereas another larva in the same family
lacks eye-spots. On the basis of known life histories, it is certain that the tri-
chocercous cercariae listed by Dollfus have adults belonging to at least three
distinct families, the Lepocreadiidae, the Monorchiidae, and the Fellodistomatidae.
Furthermore, the last two groups and perhaps all three have some larvae that
are not trichocercous. Thus in distinguishing the larvae of these families, the
morphology of the body and type of molluscan host are more dependable than is
the structure of the tail which can be positively misleading. Of the non-ocellate
cercariae listed by Dollfus, C. setifcra Miiller ncc Monticelli (the larva of Bacciger
bacciger according to Palombi, 1934), C. villoti, C. pclsenceri, C. chiltoni, and
C. pectinata Huet nee Chilton may be assigned to the Fellodistomatidae. C.
laevicardii evidently differs from all of these in at least one of the following respects :
size of body, tail and suckers ; proportionate length of body and tail ; and number
of setaceous tufts. Among the ocellate cercariae listed by Dollfus, it seems highly
probable that C. elcgans Miiller also is a larva of the Fellodistomatidae although
the molluscan host is unknown. In a personal communication, Prof. G. R. La Rue
has informed the writer of what evidently is an ocellate fellodistome larva taken
in plankton from Lake Pontchartrain, Louisiana.

The development of the excretory system provides an interpretation of Cercaria
laevicardii and similar larvae in respect to the furcocercous type with which they
obviously are closely related. In young embryos (Fig. 4), each definitive excretory
tubule terminates with a ciliated largement which is joined by capillaries from two
flame cells. With further development, four flame cells are seen on each side and
this pattern persists until the embryo is well advanced. As the tail develops, the
definitive tubules extend its full length to open posteriorly and at about the time

A NEW MARINE CERCARIA 19

the caudal finlets first appear as small krrobs, the tip of the tail shows a tiny but
distinct notch separating two short, terminal papillae with the excretory pores at
their tips. A faint suggestion of this condition is evident in the fully developed
cercaria. After the caudal tubules fuse, the resulting excretory canal is large and
conspicuous, just as in furcocercous larvae of the Fellodistomatidae.

From these observations and other studies, it seems evident that larvae of the
Fellodistomatidae are basically furcocercous and show at least three types of
modification : ( 1 ) symmetrical reduction of the entire tail until in some forms it
has become a mere knob of cells or lost altogether; (2) disappearance of the
furcae without a corresponding reduction of the tail stem; and (3) reduction of the
stem only, with the furcae becoming greatly elongate. The first type of modifica-
tion has a counterpart in other trematode families such as the Brachylaemidae
which is closely related to the Fellodistomatidae and the Microphallidae and
Monorchidae which are not. In all such cases, caudal modification is associated
with reduction or suppression of free-swimming activity of the cercarial stage.
However, such an adaptation cannot explain the second type of modification in
which the development of finlets from paired lateral setaceous tufts makes the tail
an exceedingly effective natatory organ but no more so than in trichofurcocercous
species. The third type of modification is exemplified by the unnamed cercaria
described by Jones and Rothschild (1932). That larva develops in a marine
bivalve, Nucula nucleus, and has a tail with an extremely short stem and long,
slender furcae which are very extensile. From the structure and host relationship
of that cercaria as interpreted in the light of recent studies, there seems no doubt
that its adult is a member of the Fellodistomatidae.

Caudal modifications shown by cercariae of the Fellodistomatidae may have
considerable phylogenetic significance. Jones and Rothschild (1932) observed
that the cercaria from N . nucleus superficially resembled bucephalid larvae. That
this resemblance may be significant is indicated by the papers of Allison (1943)
and Cable (1953) who presented evidence that the Brachylaemidae, Fello-
distomatidae and Bucephalidae may form a related group within the Strigeatoidea.

Life history studies of the last two decades have supported and greatly extended
La Rue's (1926) concept of two orders of digenetic trematodes, the Strigeatoidea
and the Prosostomata. Without exception, it has been found that the definitive
excretory pores of cercariae in the Strigeatoidea are posterior in location, at or
near the tips of the furcae when present, although secondary openings nearer the
body may develop later as in the Bucephalidae. This position of the excretory
pores is characteristic also of the rediae and sporocysts in at least some families of
the Strigeatoidea. The most striking example of this situation is provided by the
family Bivesiculidae in which Le Zotte (1954) has found that the cercariae are
furcocystocercous and produced in rediae in which not only is the posterior end
cleft but also the excretory pores are at the tips of the resulting lobes just as they
are at the tips of the furcae in the cercarial stage. Furthermore, the excretory
patterns of the redia and cercaria differ only in respect to the number of flame
cells in certain groups, the number of flame cell groups being the same in both stages.

On the other hand, the location of the definitive excretory pores in the cercariae
of the Prosostomata is variable. In some families, the pores are at the junction
of the body and tail and in others on the sides of the tail at varying distances from

20 R. M. CABLE

the body. In a few groups, the pores are well toward the posterior end of the tail
and thus approach the situation described above for C. lacvicardii. In the
Strigeatoidea, the excretory vesicle of the cercaria is always thin-walled, whereas
in the Prosostomata it may be either thin- or thick-walled, the latter condition being
evidently a secondary one. These facts and many others beyond the scope of a
brief discussion suggest that as a group, the Strigeatoidea is more primitive than
the Prosostomata. Yet the two orders have so much in common that their having
arisen independently from turbellarian stock is inconceivable ; their point of
divergence must have been considerably removed from that source. It may be
doubtful whether the Fellodistomatidae is the extant group closest to such a point
of divergence but at least present knowledge of that family affords a plausible
explanation of how the larvae of one order could have arisen from those of the
other. In this connection, it is of interest to note that in the Fellodistomatidae
the genital pore of the adult is anterior to the ventral sucker whereas in other
families of the Strigeatoidea it is either at the extreme posterior end of the body or
closer to that end than to the anterior extremity except in some blood flukes.
Thus in respect to the location of this opening, the fellodistomatids are more like
the Prosostomata than strigeatoids.

SUMMARY

Cercaria laevicardii sp. nov. is described from the marine lamellibranch,
Lacvicardhtin mortoni. The cercaria develops in simple sporocysts and is similar
to the larva of Bacciger bacciger. The excretory formula is 2 [(3 + 3) + (3 + 3)]
and the structure and development of the excretory system is such that the. larva
is interpreted as having been derived from the furcocercous type. Various caudal
modifications in fellodistomatid cercariae are discussed and the possible phylogenetic
significance of some types is mentioned.

LITERATURE CITED

ALLISON, L. N., 1943. Leucochloridiomorpha constantiae (Mueller) (Brachylaemidae), its

life history and taxonomic relationships among digenetic trematodes. Trans. Amer.

Micros. Soc., 62 : 127-168.
CABLE, R. M., 1953. The life cycle of Parvafrcma borinqueiiae gen. et sp. nov. (Trematoda:

Digenea) and the systematic position of the subfamily Gymnophallinae. /. Parasitol.,

39: 408-421.
CHUBRICK, G. K., 1952. (The immature stages of the trematode Fellodistomnm fellis Nicoll,

1909, from invertebrates of the Barents Sea.) (In Russian.) Zoo!. Zhur., 31:

653-658.
DOLLFUS, R. P., 1925. Liste critique des cercaires marines a queue setigere signalees jusqu'a

present. Trav. Stat. Zool. Wimercux, 9 : 43-65.
JONES, E. L, AND M. ROTHSCHILD, 1932. On the sporocyst and cercaria of a marine distomid

trematode from Nucula. Parasitol., 24 : 260-264.
LA RUE, G. R., 1926. Studies on the trematode family Strigeidae (Holostomidae) No. III.

Relationships. Trans. Amcr. Micros. Soc., 45: 265-281.
LE ZOTTE, L. A., JR., 1954. Studies on marine digenetic trematodes of Puerto Rico : the family

Bivesiculidae, its biology and affinities. /. Parasitol., 40 (in press).
PALOMBI, A., 1934. Bacciger bacciger (Rud.), trematode digenetico : fam. Steringophoridae

Odhner. Anatomia, sistematica e biologia. Pub. Stas. Zool. Napoli, 13: 438-478.

REDOX INDICATOR PATTERNS IN RELATION TO ECHINODERM
EXOGASTRULATION. II. REDUCTION PATTERNS

C. M. CHILD

Hopkins Marine Station and School of Biological Sciences,
Stanford University, Stanford, Calif.

Continued use of redox indicators on echinoderm material during the last six
years has brought to light certain characteristics of the patterns of intracellular
oxidation and reduction of the indicators, particularly in their relation to exogastrula-
tion, and has made it desirable to call attention again to certain features of these
patterns. Some of these were not known to be present, and the physiological
significance of certain others was still uncertain at the time of early studies of in-
dicator patterns. Intracellular oxidation patterns have already been considered
in a preceding paper (Child, 1953c), with suggestions concerning their significance
in exogastrulation. The present paper is concerned with reduction patterns.

The first studies of redox indicator patterns in echinoderms also described
only reduction patterns ; these became visible after staining by certain redox dyes
only when external oxygen was decreased to a certain critical level. The dyes
became hydrogen acceptors and in the case of methylene blue and various other
dyes merely became colorless, or with diazine green (Janus green) reduction to
the red diethyl safranine occurred first and might be followed by further reduction
to colorless, with return to red after oxygen increase (Child, 1936a, 1936b). In
these papers presence of distinct regional differentials in rapidity of reduction was
demonstrated. These constituted gradient patterns of reduction obviously cor-
related in some way with the physiological axes and with the course of morphogenesis.
At that time nothing was known concerning intracellular oxidation patterns of
the indicators; consequently the physiological significance of certain features of
the reduction patterns was not clearly recognized.

As a background for the recent studies of reduction pattern, it seems necessary to
call attention briefly to some of the more important results of the earlier papers.
The material consisted of Strongylocentrotus purpuratus, S. franciscanus, Dendraster
excentricus, all echinoids, and the asteroid starfish, Patiria miniata. In normal
development (i.e., the course or courses of development under as nearly as possible
natural conditions and without experimental modification) the egg, cleavage
stages, and the earlier blastulae showed a reduction gradient decreasing basipetally
from the apical region without any visible change in this pattern at the time of
formation of the micromeres in the echinoids. In Patiria there are no micromeres ;
also the basipetal differential in rate of reduction seemed to be somewhat greater
in the starfish than in the echinoids. It was further noted that when the cells of
the cell wall of blastulae and early gastrulae were stained throughout, reduction
progressed from the blastocoelar surface outward, and during and after immigra-
tion of primary mesenchyme cells in echinoids they reduced more rapidly than any
other cells. The decrease in rate of reduction from the blastocoel outward was
regarded as resulting from lower oxygen content in the blastocoel than in the
external fluid and was believed to be rather an incident of development than of any

21

C. M. CHILD

real significance. The arrows used to indicate directions of decrease in rate
of reduction were intended to indicate primarily differentials in the polar reduction
gradients and other regional gradients of later development. In most figures the
arrows indicate reduction beginning on the blastocoelar surface of the cell- wall,
but in their further course they were often drawn entirely outside the body. With
approach of gastrulation a second reduction gradient appeared in the basal region
with decrease acropetally, and variation in extent toward the apical region with
different experimental conditions.1 In somewhat later studies a similar reduction
pattern of intracellular indophenol was observed in Dcndraster and Patiria (Child,
1941b, 1944).

In all these earlier papers attention was repeatedly called to the possible sig-
nificance of this change in the intracellular reduction pattern with approach of
gastrulation, as suggesting a change of some sort in the oxidation-reduction
mechanism, apparently associated with activation of primary mesenchyme and
prospective entoderm in echinoids and of prospective entoderm in the starfish,
perhaps a step in differentiation of the basal region. It is accompanied or almost
immediately followed by immigration of the mesenchyme in echinoids and by
entodermal invagination in both echinoids and starfish.2 In the early papers
on reduction patterns it was further noted that in the entogastrula the reduction
differential in the entodermal cell wall underwent a reversal in direction, at least
in the apical entoderm, with reduction no longer progressing from the blastocoelar
surface but from the archenteric cavity.3 This change was regarded as probably
resulting from lower oxygen content in the cavity than elsewhere. The apical
archenteric region attains high developmental activity and high susceptibility to
inhibiting agents during these stages. Adequate supply of oxygen through the
blastopore (anus) appears improbable and fluid in the cavity apparently moves
toward, rather than from the anus. Oxygen diffusing inward from the exterior
must now pass, not only through the ectoderm, but through the entodermal cell-wall
to reach the archenteric cavity. With external oxygen decrease by a reducing
agent, and with the oxygen uptake of the ectodermal and entodermal cell-walls, it
appears highly probable that oxygen content in the blastocoel will become lower
than elsewhere, as the course of reduction indicates.

In the evaginated entoderm of the exogastrula no such reversal in direction
in the entodermal cell-wall was found (Child, 1936b, p. 484). If the preceding
interpretation of the reversal is correct, no reversal is to be expected. The internal
entodermal cavity in the exogastrula is still a part of the blastocoel and diffusion of
oxygen is merely through the entodermal cell-wall. With external oxygen de-
crease, the oxygen uptakes of the entoderm cells and of mesenchyme in the echinoids
and dissociated entoderm cells in the starfish, which are evidently not dead in
most cases, all contribute to decrease oxygen content in the blastocoel below that
elsewhere and so to determine entodermal reduction in the exogastrula decreasing
from the blastocoel outward. If the exogastrular ectoderm is thick enough to show

1 The basipetal reduction gradient is indicated in Child, 1936a, Figs. 8-21, the later ap-
pearance of the acropetal reduction pattern in Figs. 22-26 and 29-31.

2 Child, 1936a, p. 450; 1941a, p. 129; 1941b, p. 525, bottom; 1944, pp. 450-51. The reduction
pattern in Dendraster was described only briefly without figures, 1941b. The new acropetal re-
duction of Patiria was indicated in Child, 1944, Figs. 11 and 13-20.

3 Child, 1936a, Figs. 33 and 36-38.

REDOX PATTERNS AND EXOGASTRULATION

a cell-wall gradient, reduction progresses from the blastocoel outward there,
as in the entoderm.

The frequently repeated studies of redox indicator patterns of echinoderms
in recent years confirm in general the earlier observations on the reduction pat-
terns of normal development, and perhaps contribute something to the physio-
logical analysis of these patterns. As regards exogastrulae, however, the earlier
observations are less complete, and conditions determining the entodermal reduc-
tion pattern observed in almost all exogastrulae were not analyzed. In almost
every one of thousands of exogastrulae reduction was found to begin at the ect-
entodermal junction or in the entoderm near it. A very few elongated exo-
gastrulae of Dcndrastcr were found with entodermal reduction decreasing from
the tip (Child, 1936b, Figs. 21 and 22) and later (Child, 1941b, p. 527) it was
noted that in some elongated Dcndrastcr exogastrulae entodermal reduction pro-
gressed from the tip, but little attempt to account either for the usual course of
entodermal reduction from the ect-entodermal junction or the few cases of
reduction from the entodermal tip was made. The rapid development, and often
the great elongation of the entoderm in exogastrulae, suggest that the evaginating,
like the invaginating, entoderm develops with activity decreasing from the tip.
It was suggested in the early paper that this gradient might have been reversed
in direction by the much greater susceptibility of the tip to the exogastrulating
agent, but this suggestion was not fully discussed. It appears from recent studies
that a reversal of this sort may occur in some of the more extreme degrees of
exogastrulation, though by no means in all exogastrulae.

When the early studies were made it was not clearly recognized that the condi-
tions under which the reduction pattern becomes visible represented, or might
represent, differentially inhibiting factors in addition to the agent used for
exogastrulation. If the tip of the evaginated entoderm is the most active and most
susceptible region, the external oxygen decrease necessary for intracellular reduction
may also be a highly important factor in decreasing or obliterating, or perhaps even
reversing, the direction of the oxidation gradient. Lack of oxygen has been
found to be a differential inhibitor with the same relation to oxidation gradient
pattern as other differentially inhibiting agents.

Staining by oxidized dye, particularly by the relatively toxic diazine green, used
very largely in the earlier study, is also a differential inhibitor with the same
relation to gradient pattern as others. If staining is too long continued, reduction
in the more susceptible region or regions is retarded. If the entodermal tip is
the most susceptible region of the evaginated entoderm it will be most retarded
in reduction. In the earlier study of reduction the oxygen decrease necessary for
reduction was brought about gradually by the oxygen uptake of a number of
animals sealed in a small volume of water or sometimes in dilute dye solution.
Consequently oxygen decrease occurred gradually and the critical level for reduc-
tion was attained after a variable length of time. During this time the animals
remained stained by oxidized dye.

And finally, it is possible that the physiological age of the exogastrula may
sometimes be a factor concerned to some degree in determining the reduction
pattern. In normal plutei and starfish larvae the oxidase gradient patterns
gradually decrease with the progress of starvation of the larvae and may almost
completely disappear while the larvae are still motile. The exogastrula is not a

24

C. M. CHILD

stage in the progress of development. It represents the end of that form of develop-
ment when it has attained its final stages. Exogastrulae may live for days after
growth and elongation of the evaginated entoderm have ceased, but it seems
probable that they usually, if not always, die from other conditions than starvation.
The entodermal oxidase gradient evidently persists in exogastrulae for a considerable
time (Child, 1953c) ; it sometimes seems to become less evident in older exo-
gastrulae, but comparison of the degree of differential in this gradient or in the
reduction gradient in different individuals is of no real value.

With continued exposure to moderate degrees of differential inhibition by the
exogastrulating agent, there is the possibility of development of differential tolerance
to the agent. With return of material to water, even after relatively extreme
degrees of inhibition there is often a very considerable degree of differential re-
covery in definite relation to the gradient pattern, and a continuation of exogastrular

FIGURES 1-3. Slightly modified from figures of the early study of reduction pattern for
greater clarity as regards gradient pattern: Figure 1, Dcndrastcr, six days in LiCl M/50; Figure
2, Strongylocentrotus, extreme crowding in water; Figure 3, Patiria, 60 hours in LiCl M/30.
Further data in text.

modification. Differences in susceptibility in eggs of the same or of different lots
and the difficulty of exposing different individuals, even in the same container, to
conditions that are really similar, emphasize the desirability of covering similar
experimental ground repeatedly with different lots of material.

At the time of the first study of reduction pattern nothing was known con-
cerning the indicator oxidation or oxidase pattern. The reduction pattern must
now be considered in relation to what has been learned concerning the oxidase
pattern. The oxidase gradient of the. cell-wall decreasing from the blastocoelar
surface in blastulae and early gastrulae of normal development (Child, 1953c)
suggests that under natural conditions oxygen content in the blastocoel may not
differ greatly from that outside but may become much less than outside after
external oxygen decrease, in consequence of oxygen uptake by cells of the wall
and mesenchyme and dissociated cells in the blastocoel. The purpose of this
discussion is merely to call attention to various factors which are or may be con-

REDOX PATTERNS AND EXOGASTRULATION

earned in determining reduction patterns; few suggestions concerning the roles
of particular factors in individuals are possible.

Figures 1-3 are exogastrulae of Dendraster, S. purpuratus and Patiria from
the early paper, but with positions of the arrows indicating directions of decrease
in rates of reduction altered in order to show more exactly the patterns described
in the text of that paper. In Figure 1 (6 days in LiCl M/50 from 2-cell stage)
the ectoderm still shows a very slight polar gradient. In Figure 2 (extreme
crowding in water) there is evidently entodermization of the apical ectoderm with
elongation of the entodermized region outward, as in basal exogastrulation (Child,
1948). In both the basal and the apical exogastrulation reduction progresses from
the ect-entodermal junction and in the ectoderm the polar gradient is still present,
except basally. In Figure 3 (60 hours in LiCl M/30 from 2-cell stage) reduction
progresses from a region of the entoderm, perhaps entodermized ectoderm, nearer
the ect-entodermal junction than the entodermal tip. In all cases reduction in the
cell-wall progresses from the blastocoel outward. These were the reduction
patterns observed in thousands of exogastrulae of all four echinoderms.

MATERIAL AND METHODS

The three echinoderms chiefly used in the early study of reduction and in the
preceding paper on oxidation pattern, Dendraster e.rcentricus, S. purpuratus and
Patiria miniata, were again used in this further study of reduction patterns.4 The
dyes used were chiefly diazine green and in some cases methylene blue. Diazine
green, with the two steps in reduction and with color change, first from the blue-
green of the oxidized dye to the red diethyl safranine, and second, the further
reduction to colorless, has been more useful than other dyes which merely lose
color on reduction. Intracellular reoxidation of colorless reduced diazine green
to the red diethyl safranine is possible, but further reoxidation to the blue green,
fully oxidized dye does not usually occur in echinoderm material, though it has
been observed in some other organisms. Diazine green is more toxic than many
other "vital" dyes and has been used in concentrations of 1/100,000 and 1/50,000,
and usually only with staining periods from 5-15 minutes.

Patterns of intracellular indophenol reduction with loss of color are similar
to the dye reduction patterns. As repeatedly described in earlier papers, intra-
cellular indophenol reaction (the Nadi reaction) results from oxidation of the
reagents, para-aminodimethyl aniline (dimethylparaphenylene diamine) and
cc-naphthol, catalyzed by an oxidase, often regarded as cytochrome oxidase. Both
of the indophenol reagents are toxic but with use in very low concentrations both
the intracellular reaction with deep blue color and reduction to colorless are
possible in apparently uninjured embryos and larvae and in motile stages still mov-
ing. Also alkali is not required to dissolve the naphthol. Use of this indicator
is more fully described in earlier papers (e.g., Child, 1944, 1953c and various
other papers).

4 The kindness of the Director and staff of the Hopkins Marine Station in providing ma-
terial, facilities for work, and in some cases for transportation of material to Palo Alto, and of
Dr. Olin Rulon for sharing Dendraster material and transporting it to Palo Alto is again grate-
fully acknowledged.

26 C. M. CHILD

Although certain organisms or certain regions or organs can reduce intra-
cellular methylene blue and low intracellular concentrations of diazine green and
indophenol without external oxygen decrease, intracellular reduction of these and
various other redox indicators is not generally characteristic of development or
life under natural conditions, but occurs only after oxygen in the external en-
vironment and in the tissues has undergone decrease to a critical concentration.
The indicator then becomes a hydrogen acceptor and is reduced. This reduction
is considered to be catalyzed by one or more dehydrogenases. Obviously this
reduction of the indicator represents oxidation of some intracellular substrate by
loss of hydrogen to the indicator and becomes the characteristic reaction with
sufficient oxygen decrease.

Decrease of available oxygen can be brought about in various ways. In some
of the earliest studies of intracellular reduction of indicators highly toxic reducing
agents were used, e.g., sodium hyposulphite and hydrochloric acid, also another,
rongalite, containing formalin. These required extreme caution in use and in
•certain cases their use led to errors as regards regional differentials in rates of
reduction. Regions most susceptible to the toxic effects were injured, so that
reduction in them was delayed or did not occur, though when they were not
greatly injured reduction was more rapid in them than in any other parts (Child,
1941a, pp. 90-92 and footnote). In the earliest studies of indicator reduction in
echinoderms these toxic reducing agents were not used. After staining by oxidized
dyes a number of embryos or larvae were sealed in a small volume of water or, in
some cases, dilute dye solution and oxygen decrease resulted from oxygen uptake
of the living material. With this procedure the length of time before reduction
varied with number of individuals and volume of fluid. Intracellular reoxidation
occurred rapidly on opening the sealed preparation and reduction and reoxidation
could often be repeated several times before the intracellular concentration of
oxidized indicator became toxic. Some years later this method was used for
intracellular indophenol reduction in Dendraster (Child, 1941b).

In recent indicator studies, however, a much more satisfactory reducing agent,
sodium hydrosulphite (NaHSCX or Na2H2S2O4), has been used. A fraction of
a milligram is sufficient to bring about reduction in echinoderm developmental
stages in one ml. of water. The chief difficulty is to keep the quantity used small
enough so that reduction will not be too rapid for observation of regional dif-
ferentials. This agent is not appreciably toxic in concentrations much higher
than those required for reduction, though in sealed preparations animals finally
die from lack of oxygen. Hydrosulphite has been used for reduction in all cases
considered in this paper and in thousands of other individuals.

Figures are essentially optical sections along the polar axes. They do not
indicate actual differences in size in the different species. Arrows, drawn only
for the left side, though the two sides are similar, point in the direction of decrease
in rate of reduction. Mesenchyme and dissociated entoderm cells in the blastocoels
are indicated in dotted outlines or areas.

REDUCTION PATTERNS OF DENDRASTER AND STRONGYLOCENTROTUS EXOGASTRULAE

Exogastrular reduction patterns and their variations are similar in these two
echinoids. A highly effective method for producing exogastrulation in these
forms is exposure to low temperature during early development with later de-

REDOX PATTERNS AND EXOGASTRULATION

27

velopment at a much higher temperature. Figures 4-8 are from a lot kept at 10° C.
for 29 hours from the 2-4 cell stage and later at 22°-24° C. Neither change of
temperature was sudden. At the low temperature development did not usually

FIGURES 4-15. Reduction patterns in Dcndrastcr exogastrulae : Figures 4-8, exogastrulae re-
sulting from temperature change ; Figure 9, 2 days in sodium azide M/600 ; Figures 10-15, 2 days
in water after 28 hours in LiCl M/50. Further data in text.

28 C. M. CHILD

progress beyond blastula stages and if these were left too long at the low tempera-
ture death occurred. Figures 4—8 represent exogastrulae with high degrees of
differential recovery at the higher temperature. Ectoderm and mesenchyme attain
complete pluteus development and in Figures 4—6 there is more or less development
of three entodermal segments. More extreme degrees of exogastrulation occur
with slightly longer exposure to low temperature. In Figures 4—6 entodermal
reduction progresses from the tip, though in Figure 6 the thin-walled region
adjoining the ectoderm reduces from the ectoderm. In Figure 7, with less
entodermal development, reduction progresses from the tip, but in Figure 8, with
very similar degree of entodermal development, entodermal reduction progresses
from the ectoderm. Reduction in the entodermal cell-wall progresses from the
blastocoel outward. Ectodermal reduction pattern is like that of the normal
pluteus; tips of the oral lobe and of the arms are the high ends of reduction
gradients. Figure 9 with ectodermal development stopped in a prepluteus stage,
and entodermal reduction progressing from the tip was exposed to sodium azide
only after it attained the blastula stage (azide M/600, 2 days). With this rela-
tively later exposure to azide only a small number of exogastrulae appeared; in-
hibited entogastrulae developed in an estimated 95 per cent of the lot, and the few
exogastrulae were not extremely inhibited by azide. Figures 10-15 represent
exogastrulae two days in water after 28 hours from the 2-4 cell stage in LiCl M/50.
All are from a single container and serve as examples of the variations in form
and development under more or less similar conditions. In Figures 10-12
entodermal polar reduction progresses from the tip and in Figures 10 and 11 a
slight ectodermal polar gradient, decreasing basipetally, still persists. Figure 12
is particularly interesting; evidently it was greatly inhibited by LiCl, apparently
with entodermization of much of the ectoderm, but reduction progressing from
the entodermal tip suggests that after return to water the entoderm underwent a
high degree of differential recovery, with the tip becoming the region of most rapid
reduction. Figure 13 is also of interest, as suggesting some degree of recovery
of the entoderm at and near the tip, though not sufficient to prevent occurrence of
reduction in both directions at the ect-entodermal junction. In Figures 14 and 15
reduction progresses in both or only in one direction from the ect-entodermal
junction, the usual course of reduction in the early study of exogastrulation.
Similar varieties of form and course of reduction occur very generally in a single
container, except with extreme degrees of inhibition by the exogastrulating agent
or other conditions. In all cases in which the entodermal cell- wall is thick enough
to show the cell-wall gradient clearly, reduction progresses from the blastocoelar
surface outward.

Exogastrulae of Strongyloccntrotus differ so little from those of Dendraster
that they require only brief attention. Figures 16 and 17 are forms two days in
water after two days in azide M/800 from 2-4 cell stages. There was evidently
considerable differential recovery after return to water, with further development
of entoderm. Reduction progresses from the entodermal tip and in Figure 16 a
slight polar gradient is present in the ectoderm. Figure 18, from a lot two days
in LiCl M/50 from the 2-cell stage with dilution of the solution to approximately
half water after that period, is much like Figure 12 of Dendraster. Here also
there is probably entodermization of ectoderm and differential recovery of entoderm

REDOX PATTERNS AND EXOGASTRULATION

29

with reduction progressing from the entodermal tip. Many other exogastrulae in
this lot were very similar. In Figure 19, after three days in LiCl M/40 without
return to water, reduction is from the ect-entodermal junction. Figure 20, three
days in LiCl M/60 without return to water, also showed reduction from the
ect-entodermal junction. Figure 21, from a lot one day in LiCl M/30, followed
by two days in water, shows reduction progressing from the ect-entodermal junc-
tion, like most others in the lot, but in a few individuals reduction progressed from
the entodermal tip. Perhaps the Strongylocentrotus exogastrulae, not merely

19

FIGURES 16-21. Reduction patterns in Strongylocentrotus exogastrulae: Figures 16 and
17, 2 days in water after 2 days in sodium azide M/800 ; Figure 18, 2 days in LiCl M/50 followed
by dilution to approximately M/100; Figure 19, 2 days in LiCl M /40 ; Figure 20, 3 days in
LiCl M/60; Figure 21, 2 days in water after one day in LiCl M/30. Further data in text.

those of the figures, but many hundreds of exogastrulae in many lots, suggest that
recovery of the evaginated entoderm, so that reduction progresses from its tip
occurs less frequently in this echinoid than in Dendrastcr. This difference is
perhaps to be expected, as Strongylocentrotus is in general somewhat more sus-
ceptible to inhibiting conditions than Dendrastcr. In both forms reduction in the
cell-wall progressed from the blastocoelar surface outward in all cases in which
the cell-wall was not so thin that the gradient was uncertain.

It is a point of incidental interest that in these echinoid exogastrulae there are
usually free cells in the blastocoel ; immigration of mesenchyme or of some part of

30 C. M. CHILD

it often occurs before actual evagination of entoderm. Not infrequently most of
the mesenchyme cells reach the ectodermal region, but, except in the lesser degrees
of exogastrulation such as Figures 4-8 of Dendr aster, the ectodermal factors
localizing mesenchyme are obliterated, and, in Figure 9, almost obliterated. In
those exogastrulae skeleton does not develop, or at most a few irregularly localized
spicules or rods appear. In addition to mesenchyme cells in the blastocoel, cells
may dissociate internally from the entoderm; these are usually still capable of
reduction and oxidation of indicators. These dissociated cells in the blastocoel
reduce earlier than other parts of the exogastrulae, and, as pointed out above,
doubtless play some part in decreasing oxygen content in the blastocoel by their
own oxygen uptake, i.e., as external oxygen decrease occurs these cells probably
determine their own rapid reduction and may also be factors in determining the
cell-wall gradient decreasing from the blastocoel outward. All figures of echinoid
exogastrulae were drawn from individuals in which direction of the cell-wall
gradient was clearly distinguishable.' In some exogastrulae the entoderms may
become so thin that direction of the cell-wall gradient becomes difficult or impossible
to determine. Occasionally it was noted at the time of observation that reduction
seemed to progress from the external entodermal surface inward. With rapid
and extreme oxygen decrease, this is of course possible, but it seems quite beyond
question that the characteristic cell-wall reduction gradient decreases from the
blastocoel outward in the evaginated entoderm and also in the ectoderm, if that is
not too thin to show a distinct cell-wall gradient.

REDUCTION PATTERNS IN EXOGASTRULAE OF PATIRIA

With exposure to exogastrulating agents, beginning in the earlier blastula
stages, development of differentially inhibited entogastrulae may precede exogastrula-
tion. For example, Figures 22 and 23 are from a lot in which every individual
of several samples, including large numbers, was an inhibited entogastrula after
15 hours in LiCl M/30 from early blastula stages. In many of these, entodermal
dissociation was already occurring. After 48 hours in LiCl every individual
of numerous samples was an exogastrula and in most of them the invaginated
part of the entoderm was dissociating or dissociated. Figure 24 is an exogastrula
with its development stopped in early stages during three days in a high concentra-
tion of LiCl (M/7.5). Only the cell-wall gradient, decreasing from the blastocoel
outward, is distinguishable. Figure 25 is also stopped in an early stage of
exogastrulation by three days in azide M/250. Entodermal dissociation is begin-
ning internally and here also only the cell-wall gradients are distinguishable.

Figures 26 and 27, two days in LiCl M/30 from early blastula stages, are
cases of polar entodermal reduction progressing from the tip and without entodermal
dissociation. This pattern of reduction has not been observed as frequently in
Patiria as in the echinoids ; in those cases in which it has been observed, no evidence
of entodermal dissociation has appeared. In Figures 28-32 polar entodermal
reduction progresses in both directions from an entodermal region near the ect-
entodermal junction and all show more or less entodermal dissociation. Figure
28, 2 days in LiCl M/30 from the early blastula, may have been originally an
entexogastrula ; if it was, the invaginated part of the entoderm has dissociated
internally and the remaining entoderm cells have come together and are intact.

REDOX PATTERNS AND EXOGASTRULATION

31

Figure 29, with similar LiCl exposure, is an entexogastrula with the invaginated
part of the entoderm in process of dissociation. In Figures 30 and 31, also with
the same LiCl exposure, entodermal dissociation occurred earlier, perhaps during
an entogastrula stage, and the remaining entoderm has healed. Figure 32, three
days in LiCl M/30, is an example of the great length sometimes attained by the

30

31

FIGURES 22-32. Reduction patterns in Patiria exogastrulae : Figures 22 and 23, inhibited
entogastrulae after 15 hours in LiCl .17/30, after 48 hours in LiCl xvere exogastrulae ; Figure 24,
exogastrula stopped in early stage after 3 days in LiCl Af/7.5 ; Figure 25. stopped in early
exogastrulation after 3 days in sodium azide M/25Q ; Figures 26-31, 2 days in LiCl .17/30 from
early blastula stages ; Figure 32, 3 days in LiCl .17/30. Further data in text.

32 C. M. CHILD

•evaginated entoderm when little or no dissociation occurs. These figures suggest
that the most active and therefore the most susceptible entodermal region often
undergoes dissociation; other less susceptible regions remain intact and may
apparently develop some degree of tolerance to this concentration of LiCl and
undergo further elongation (Figs. 31 and 32).

Figures 30 and 31 are intended to indicate another interesting characteristic of
starfish exogastrulae. In the course of observations of reduction of oxidized
diazine green it was found, first by accident and later confirmed by many cases,
that when staining by this dye was continued somewhat longer than the usual 5—15
minutes with 1/100,000 or 1/50,000, reduction occurred as the arrows in these
figures and in Figures 28 and 29 indicate. Reduction to the red diethyl safranine
occurred first ; reduction to colorless followed in the dissociated cells but a variable
region of the ectoderm and of the entodermal tip remained red and did not reduce
to colorless at any time. In Figures 30 and 31 this is indicated by the deep
shading of the apical region of the ectoderm and of the entodermal tip. In some
cases in which lots were left in oxidized diazine green for half an hour or somewhat
more, these two regions remained blue-green, i.e., did not reduce at all. These cases
are regarded as indicating differential injury of these regions, the most susceptible
of the individual, while the less susceptible are still able to reduce the dye, even to
colorless. In use of diazine green on other organisms and even with other less
toxic dyes it has been found that with over-staining by oxidized dyes reduction is
retarded or may not occur at all. Indophenol reduction is retarded similarly if
the intracellular concentration of indophenol becomes sufficiently high.

Although the length of the evaginated entoderm in the starfish exogastrulae
varies greatly, further differentiation of the entoderm with development of two or
three segments has been observed only in nine individuals among the thousands
of exogastrulae observed. In the starfish, as in other echinoderms, degrees of
exogastrulation and indicator patterns are dependent on experimental procedures,
temperatures and susceptibilities of individuals and different lots of eggs. It is of
course possible that with different exposure periods to exogastrulating agents
or other differences in experimental conditions frequencies of the different reduction
patterns may differ. However, the present paper is primarily concerned with
occurrence, rather than with frequencies of the different patterns.

There is no primary mesenchyme in the starfish. The dissociated cells in the
blastocoels of most starfish exogastrulae are cells dissociated from the entoderm, but
usually still capable of reducing and reoxidizing the indicators. They unques-
tionably contribute to the low oxygen content in the blastocoel and, like the
mesenchyme of the echinoids, usually reduce before other parts. Apparently the
entoderm cells which dissociate into the blastocoel represent the most susceptible
entodermal region; when there is no dissociation, reduction progressing from
the entodermal tip appears to be more frequent than in cases of dissociation of
this region. The progress of reduction in the entodermal cell-wall from the
blastocoel outward is even more distinct in the starfish than in the echinoids; the
larger size of starfish stages and the greater thickness of the evaginated entoderm
in most of the exogastrulae account for this difference.

REDOX PATTERNS AND EXOGASTRULATION

ENTODERMAL REDUCTION IN NORMAL DEVELOPMENT WITHOUT
EXTERNAL OXYGEN DECREASE

Although this section concerns certain observations on normal development,
rather than on exogastrulae, it is included here as an example of determination by
metabolic activity of decrease in oxygen content in internal cavities below that in
external environment, and therefore as bearing on 'certain questions of reduction
pattern.

It was recently observed that in normal plutei and somewhat earlier stages of
Dendrastcr after 10-15 minutes in oxidized diazine green the mesenchyme cells
and the archenteron became distinctly red in well-aerated water, though the
ectoderm remained completely oxidized. Indophenol reduction of the archenteron
also occurs wjthout external oxygen decrease, though this reduction consists
merely in loss of color and is less striking, and must also be observed through the
more or less deep blue ectoderm. In the midgut with thicker cell-wall than other
parts of the entoderm reduction appears to progress from the internal cavity toward
the blastocoelar surface. As noted above, oxygen diffusing inward must now
pass through two cell-walls to reach this cavity and muscular contraction of the
gut begins to occur at about this time. To what extent oxygen reaches the
archenteric cavity from the mouth is not known at present ; evidently it is not
sufficient in amount to prevent low oxygen content and early reduction in the
cavity. Apparently the hindgut also undergoes early reduction from the inside
outward.

In the later stages of normal larval development of Patiria entodermal reduc-
tion of diazine green to the red diethyl safranine occurs more rapidly and to a
greater degree than in Dendrastcr after a few minutes in the oxidized dye and in
well-aerated water, while ectoderm remains completely oxidized. Indophenol re-
duction also occurs under the same conditions, though if the ectodermal reaction
becomes deep in color it may be difficult to observe. Here, even more distinctly
than in Dendr aster, reduction progresses from the entodermal cavity toward the
blastocoel, at least in the midgut and apparently in the hindgut. The foregut be-
comes thin- walled as it enlarges, and direction of reduction is less clearly dis-
tinguishable there. Oxygen entering through the mouth is evidently not suf-
ficient in amount to prevent this early entodermal reduction. Probably the sea
urchin will also show early entodermal reduction in later stages of normal larval
development without external oxygen decrease, but this question must await another
breeding season.

DISCUSSION AND CONCLUSIONS

As regards the pattern of indicator reduction, it must again be emphasized
that intracellular reduction of the indicators represents an intracellular oxidation
in the living tissues. With oxygen decrease to a critical point the indicator be-
comes a hydrogen acceptor with catalysis of the reaction by dehydrogenase. Some
substrate in the cells loses a hydrogen to the indicator. The substrate concerned
in this dehydrogenase oxidation is undoubtedly different from that involved in
the oxidase or oxidation patterns. In other words, the pattern of indicator reduc-
tion is actually the pattern of certain intracellular oxidations, a component of the
metabolism of the living cells.

34 C. M. CHILD

In the echinoderms thus far investigated by means of redox indicators the
patterns of intracellular oxidation, catalyzed by an oxidase or by oxidases, and the
patterns of indicator reduction, catalyzed by one or more dehydrogenases, are the
same, as regards regional differentials, in early development under natural condi-
tions to stages just preceding gastrulation and in early stages of entodermal invagina-
tion. As those stages are attained, the primary mesenchyme cells of the echinoids
and the prospective entoderm, previously the least active region, evidently undergo
a considerable activation, apparently involving rapid growth of the entoderm.
With this activation a new reduction gradient decreasing from the basal region,
i.e., opposed in direction to the reduction gradient of earlier stages and also to the
oxidase or oxidation gradient, appears. The oxidase gradient remains unaltered
and still decreases from the apical region basipetally over the entire individual.
The new acropetal reduction gradient varies in length according to degree and
duration of oxygen decrease. It may extend into the ectoderm. Attention was
repeatedly called to the appearance of this new reduction pattern, and it was
suggested that it indicated a change in physiological condition, decreasing from
the basal region acropetally, and perhaps representing a step in differentiation.5 In
later gastrula and larval stages oxidation and reduction patterns again become
similar as regards regional differentials and show the same relation to morphogenesis
in normal development. In exogastrulae differences in relations of oxidase and
reduction patterns appear. The polar patterns of both indicator oxidation and
reduction may be decreased or entirely obliterated. The polar oxidation pattern
in the evaginated entoderm is still present with decrease from the entodermal tip,
though its differential may become slight when entodermal elongation is inhibited
at an early stage (Child, 1953c). The reduction pattern of the evaginated
entoderm may decrease from the tip or from the ect-entodermal junction or an
entodermal region near it. In general, reduction from the entodermal tip, like
that in entogastrulae, seems to occur more frequently, as might be expected, in the
less extreme degrees of exogastrulation, though differences in susceptibility to
inhibiting conditions differ so greatly in individuals and in different lots of eggs
that it may also appear in more extreme forms. These variations in the entodermal
reduction pattern indicate that the inhibiting conditions, noted above as necessary
for exogastrulation, may obliterate, or perhaps reverse, the polar entodermal
gradient. Under these conditions, the region at or near the ect-entodermal junc-
tion becomes the most active and most rapidly reducing region, not through increase
in its own activity, but in consequence of inhibition of other regions.

In the starfish exogastrulae, and to a lesser degree in the echinoids, the dis-
sociation into the blastocoel of cells from the entodermal tip apparently constitutes
loss of the most active and most susceptible cells from the entodermal cell-wall.
Even after dissociation, these cells, free in the blastocoel, may, and almost always
do, reduce more rapidly than other cells, but they are no longer a part of the
entodermal gradient pattern. After such dissociation, together with other condi-

5 Child, 1936a; 1941a, pp. 133-143 and figures on these pages: 1941b; 1944. The writer
is indebted to Dr. E. L. Tatum for the suggestion that the new reduction pattern might be
associated with an increase in synthetic activity chiefly in the prospective entoderm, and that
the two opposed gradient patterns may perhaps be regarded as indicating a competition of
different intracellular substrates as regards oxidation, one catalyzed by oxidase with relatively
high oxygen tension, the other catalyzed by dehydrogenase and favored by low oxygen tension.

REDOX PATTERNS AND EXOGASTRULATION

tions tending to obliterate the entodermal polar gradient, reduction from the
ect-entodermal region becomes increasingly probable. In the starfish reduction
from the tip of the evaginated entoderm has usually been observed only when
dissociation of entodermal cells did not occur. In the exogastrulae of Dendraster
determined by change from low to high temperature reduction progressed from
the entodermal tip with few exceptions in large numbers observed. Differential
recovery after return to water following exposure to the inhibiting agent may also
permit reactivation of the entodermal gradient and reduction from the tip.

The general polar and ventrodorsal gradients of the echinoderm embryo and
larva are obviously different in character from the gradients between the surfaces
of the cells which constitute the cell-walls of the embryo and larval stages. Polar
and ventrodorsal gradients are differentials from cell to cell, involving the entire
individual or extensive regions of it, though slight differentials may be present
within the limits of single cells, e.g., in cleavage stages (Child, 1953c). These
general body-gradients are apparently determined in the ovary ; at least this ap-
pears clearly to be the case as regards the polar gradient. However, this gradient
can be decreased or even obliterated by differentially inhibiting conditions, and
with differential recovery or with certain degrees of differential tolerance, the
differential may become greater than in normal development, not only in echino-
derms, but also in various other organisms. In a coelenterate early developmental
stage it has been possible to determine experimentally a new polarity, and in
reconstitution new polarities have been determined experimentally in many
forms (Child, 1941a, Chapters X and XI). When and how the ventrodorsal
gradient is determined still remains uncertain, though it may perhaps also be in
the ovary in association with the growth of the gonad. It can be modified ex-
perimentally by the same conditions as the polar gradient and it can be experi-
mentally reversed in direction, apparently by a differential inhibition (Pease,
1941, 1942a, 1942b). The cell- wall gradients are usually differentials between
the two exposed surfaces of single cells, though in the thick entodermal masses of
some exogastrulae and certain other modified forms the cell-wall is not a single
cell-layer, and the cell-wall gradient becomes a multicellular differential. In the
preceding paper it was pointed out that the presence in early normal development
of the cell-wall gradient, decreasing from the blastocoel outward, did not support
the earlier conclusion, based on reduction alone, that oxygen content in the
blastocoel was normally less than externally (Child, 1936a). If this were the
case, it seemed improbable that the oxidation gradient of the cell-wall could decrease
from a region of lower, to one of higher, oxygen content. Moreover, the cell-wall
oxidation gradient is not present from the beginning of development. Its appear-
ance as a visible gradient, decreasing from the blastocoel outward, is associated with
the appearance of the blastocoel. No visible indication of presence of this gradient
has been observed in the 16-cell stage, but at 32 cells a slight cell-wall oxidation
differential has been observed and later it becomes increasingly visible. It still
remains a question as regards the conditions which determine the origin of this
gradient with its high end toward the blastocoel in normal development. Further
investigation is necessary to determine whether it is in any way correlated with
position of the nucleus in the cell. Most figures of echinoderm larval develop-
ment do not show the nuclei. There are, however, a few figures with nuclei in the

36 C. M. CHILD

earliest studies of exogastrulae by Herbst,6 but even these are not satisfactory. In
some of them, at least some of the nuclei seem to be slightly nearer the blastocoelar,
than the outer surface of the wall ; in others no distinguishable difference with
respect to the two surfaces of the wall appears. As regards the presence of this
oxidation gradient there is no question. It has been observed in many hundreds
of blastulae and early gastrulae, and its presence has been confirmed by others.
It is possible to advance various hypotheses as regards conditions determining this
gradient but at present they are little more than guesses.

The presence of this gradient in normal development by no means excludes
the possibility that with external oxygen decrease oxygen content in the blastocoel
may become much lower than outside in consequence of oxygen uptake by cells
of the wall and cells which happen to be in the blastocoel, that the indicator reduction
gradient may also decrease in the cell- wall from the blastocoel, and that immigrated
mesenchyme and dissociated entoderm cells in the blastocoel may reduce most
rapidly of all. Reduction from the internal cavity of the gut in later larval stages,
entirely without external oxygen decrease, is further evidence along this line.

The cell-wall oxidation gradient of the entoderm undergoes reversal in direction
with entodermal evagination ; it was suggested in the preceding paper that this
reversal is an essential factor in exogastrulation (Child, 1953c) . In the exogastrular
ectoderm there is no reversal of the oxidation gradient.

The indicator reduction gradient in the cell-wall of the evaginating or
evaginated entoderm of exogastrulae undergoes no reversal -and no definite changes.
It becomes visible only after marked external oxygen decrease, and evidently in
exogastrulae, as in normal development, results from more rapid oxygen decrease
in the blastocoel than externally. The only difference from normal development
is that in the evaginated entoderm the inner surface, in normal development, the
outer surface, is the blastocoelar surface of the cells.

Whether an escape or partial escape from physiological dominance of a
metabolically more active region of the developing echinoderm larva or an integrating
factor or factors of some sort which may be concerned in what we call normal
development, is concerned in exogastrulation, as suggested by J. W. MacArthur
(1924), remains a question. Obviously the normal larva does not develop as an
aggregation of independent parts. It is also obvious that the controlling or integrat-
ing factor or factors concerned in normal development are altered or, at least in part,
absent in exogastrulae and other experimental developmental modifications. In
the exogastrula the relation of ectoderm and entoderm has become different from
that in the normal larva. In the elongated exogastrulae with great over-develop-
ment of entoderm, and often more or less entodermization of ectoderm and dis-
sociation of cells, the entoderm seems, to a greater or less extent, to have gained
the upper hand. To that extent, apparently extreme in some exogastrulae. there
appears to be a partial, or in some cases almost complete, entodermal independence.
It develops as far as available material permits. Regional differential susceptibility
to experimental differentially inhibiting factors is involved in this more or less
extreme alteration or breakdown of physiological dominance or integrating and
ordering factors concerned in determining a normal individual.

6 Herbst, 1895, various figures, Plate IX ; Figures 42 and 43, Plate X. 1896, Plate XXVI.

REDOX PATTERNS AND EXOGASTRULATION

SUMMARY

1 . The paper consists primarily of a new investigation of intracellular reduction
patterns of certain redox indicators in relation to exogastrulation, with the
echinoids, Strongyloccntrotus purpuratus, and Dcndrastcr excentricus, and the
asteroid, Patiria miniata, as material. Its purpose is : first, to record results of
recent studies of these patterns, made with more adequate conditions for reduction
than in earlier work; and second, to attempt somewhat further physiological
analysis of the patterns and of their relation to oxidation patterns, than was under-
taken in the earlier study.

2. Conditions which make the reduction patterns visible involve, not only the
differentially inhibiting action of the exogastrulating agent, but also differentially
inhibiting effects of oxygen decrease externally, and in some cases, of intracellular
concentrations of oxidized dye or indophenol, and perhaps also the physiological age
of the exogastrula. Usually the significance of these different factors for individual
exogastrulae is not certainly distinguishable, but the differentially inhibiting effect
of the exogastrulating agent is probably in general the most important.

3. In the less inhibited and less extreme forms of echinoid exogastrulae, in
which ectoderm attains or approaches fully developed pluteus differentiation, the
evaginated entoderm almost always reduces progressively from the tip toward
the ectoderm, though occasional alterations of this pattern appear. In Patiria
exogastrulae, dissociation of entoderm cells from the most susceptible and most
inhibited entodermal tip and adjoining regions into the blastocoel occurs very
frequently. Entodermal reduction progressing from the tip has been observed
less frequently in Patiria than in the echinoids and thus far only when little or no
dissociation from the entodermal tip occurs. In the echinoids entodermal dissocia-
tion may also increase the cells in the blastocoel far beyond the usual number of
mesenchyme cells. In general, it appears evident that the larger the number of
dissociated cells in the blastocoel, the less frequently does reduction progress from
the entodermal tip.

4. When entodermal reduction does not progress from the tip, it begins at or
near the ect-entodermal junction, or in Patiria in the entoderm near this junction,
and progresses toward the entodermal tip and often acropetally for a short distance
in the adjoining ectoderm. Under natural conditions, these regions are the
least rapidly reducing regions of the entoderm after its activation preceding
gastrulation, and of the adjoining ectoderm. In these exogastrulae they have
become the regions of most rapid reduction, probably not by change in their own
conditions, but by more extreme inhibition of other parts and obliteration or
perhaps reversal in direction of their polar gradients.

5. In completely radial exogastrulae with rounded ectoderm lacking differentia-
tion a slight polar reduction gradient may still be visible in the apical region,
usually in cases of some degree of differential recovery after return to water,
perhaps sometimes with development of differential tolerance to the exogastrulating
agent, or a polar ectodermal reduction gradient may be completely absent. The
ventrodorsal ectodermal gradient is completely obliterated by less extreme inhibition
than the polar gradient.

6. Even if oxygen content in the blastocoel differs little or not at all from that
in the external water, as the oxidase gradient of normal development seems to

38 C. M. CHILD

indicate, it may become much lower in the blast ocoel than outside, in consequence
of oxygen uptake of cells of the cell-wall and of dissociated cells in the blastocoel
with sufficient decrease of external oxygen tension. Under these conditions re-
duction must occur most rapidly in dissociated cells in the blastocoel, which are
evidently not dead in most cases, and in the cell-wall reduction will progress from
the blastocoelar surface outward in exogastrulae, as well as in normal develop-
ment, and not only in entoderm, but also in ectoderm unless this has become so
thin that a cell-wall gradient is not distinguishable. As might be expected, the
cell-wall reduction gradient does not undergo reversal in exogastrulation, as does
the oxidase gradient. The reduction gradient in the cell-wall is not directly related
to exogastrulation.

7. In intracellular indicator reduction the indicator becomes a hydrogen acceptor
and an intracellular substrate undergoes oxidation catalyzed by one or more
dehydrogenases. Both intracellular oxidation of reduced redox indicators, catalyzed
by oxidase, and intracellular reduction, catalyzed by dehydrogenase, are directly
visible evidences of certain characteristics of oxidative metabolism, though differ-
ent enzyme systems and undoubtedly different intracellular substrates are con-
cerned in the twro oxidative reaction systems.

LITERATURE CITED

CHILD, C. M., 1936a. Differential reduction of vital dyes in the early development of echinoderms.

Arch.j.Entiv.. 135: 426-456.
CHILD, C. M., 1936b. A contribution to the physiology of exogastrulation in echinoderms.

Arch. f. Entu>., 135 : 457-493.
CHILD, C. M., 1941a. Patterns and problems of development. University of Chicago Press,

Chicago, Illinois.
CHILD, C. M., 1941b. Formation and reduction of indophenol blue in development of an

echinoderm. Proc. Nat. Acad. Sci., 27 : 523-528.
CHILD, C. M., 1944. Developmental pattern in the starfish Patiria nriniata, as indicated by

indophenol. Physiol. Zool, 17: 129-151.
CHILD, C. M., 1948. Exogastrulation by sodium azide and other inhibiting conditions in

Strongyloccntroius purpuratits. J. E.rp. Zool.. 107: 1-38.
CHILD, C. M., 1953a. Exogastrulation and differential cell dissociation by sodium azide in

Dcndrastcr c.vccntricus and Patiria miniata. Physiol. Zool.. 26: 28-58.
CHILD, C. M., 1953b. Indicator gradient pattern in oocytes and early developmental stages of

echinoderms: a reexamination. Biol. Bull., 104: 12-27.
CHILD, C. M., 1953c. Redox indicator patterns in relation to echinoderm exogastrulation.

I. Oxidation patterns. Biol. Bull.. 105 : 62-79.
HERBST, C., 1895. Experimented Untersuchungen iiber den Einfluss der veranderten chemischen

Zusammensetzung des umgebenden Mediums auf die Entwicklung der Tiere. II.

Mitteil. Zool. Stat. Ncapc!., 11 : 136-220.

HERBST, C., 1896. Experimented Untersuchungen etc. III-VI. Arch. f. Entw., 2: 455-516.
MACARTHUR, J. W., 1924. An experimental study and a physiological interpretation of

exogastrulation and related modifications in echinoderm embryos. Biol. Bull., 46: 60-87.
PEASE, D. C., 1941. Echinoderm bilateral determination in chemical concentration gradients.

I. The effects of cyanide, ferricyanide, iodoacetate, picrate, dinitrophenol, urethane,
iodine, malonate etc. J. E.rp. Zool., 86 : 381-404.

PEASE, D. C., 1942a. Echinoderm bilateral determination in chemical concentration gradients.

II. The effects of azide, pilocarpine, pyocyanine, diamine, cysteine, glutathione and
lithium. /. E.r/>. Zool., 89: 329-345.

PEASK. D. C., 1942b. Echinoderm bilateral determination in chemical concentration gradients.

III. The effects of carbon monoxide and other gases. /. Exp. Zool., 89: 347-386.

THE UTILIZATION OF SOME CARBOHYDRATES BY IN VITRO
CULTURED CHICK BLASTODERMS IN WOUND HEALING1-

RONALD C. ERASER3
Department of Zoolot/y, University of Minnesota, Minneapolis, Minnesota

Studies by Spratt (1949a, 19491), 1950a. 10501)) have shown us that there are
rather specific requirements for differentiation and form building in the early chick
embryo. By the use of chemically defined media he has been able to show the
ability of the blastoderms to utilize different carbohydrate energy sources in de-
velopment. One of the general conclusions drawn from this work is that there
are specific nutritional requirements for many of the normal processes of develop-
ment.

Former work, to be published elsewhere, devoted to the mechanical aspect of
wound healing in the chick, has shown that this process is essentially of a morpho-
logical nature. When blood carbon is applied around holes of variable sizes
produced in the extra-embryonic tissues of 24-hour blastoderms, it is observed
that those particles placed immediately at the periphery of the wounds converge
during closure of the holes, and come to lie within a very small area upon comple-
tion of healing. The distance that the carbon-marked cells move toward the
center of the wound is inversely proportional to their initial distance from the
margin, until a point is reached, beyond which the cells move outward instead of
inward. The implication from such behavior is that the closure process is effected
by a mass movement of cells, rather than by an unusually high cell proliferation at
the borders of the wounds. This conclusion has been verified by observing in
prepared slides that there is no difference in mitotic counts at any region around
the sites of the injuries.

The present study represents an extension of the general problem of wound
healing in the early chick blastoderm to include certain nutritional considerations.
It is the purpose here to determine the ability of these organisms to utilize various
media of known chemical composition for (1) the closure of wounds (for gross
tissue movements in the extra-embryonic region) and (2) the closure of wounds
in comparison to development in the embryo proper. It might be assumed that
there would be greater requirements by the rapidly developing embryonic region.

MATERIALS AND METHODS

All of the blastoderms used in the present study were of twenty-two hours
incubation (head process to head fold stages). They were removed from the
yolk, freed from the vitelline membrane and trimmed in a manner described in
detail by Spratt (1947). By the use of sharp steel needles small, approximately

1 A portion of a thesis submitted to the Graduate School of the University of Minnesota in
partial fulfillment of the requirements for the degree of Doctor of Philosophy.

2 I am deeply indebted to Dr. Nelson T. Spratt, Jr. for his guidance during this study, and
to the University of Minnesota for the excellent facilities provided.

3 Present address : Department of Biology, Reed College, Portland, Oregon.

39

40 ROXALD C. FRASER

square holes measuring 300 p. to 400 p. on a side were cut through all tissues in the
pellucid area (Fig. 2). Following injury the embryos were transferred to the
medium (see below) contained in watch glasses placed on moist cotton rings
within petri dishes. Camera lucida diagrams were made at the time of injury
and after intervals during subsequent incubation as indicated in the experiments,
to note the degree of wound closure and the development of the embryo. Two
hundred and thirty-four blastoderms were used in the study.

The preparation of the various sugar media has been described by Spratt
(1949a). By volume the constituents were: chick Ringer solution (77.5%),
penicillin-streptomycin solution (10%), phenol red (5%), phosphate buffer (5%),
bicarbonate buffer (2.5%), sugar (quantity in mg% varying in the experiments) ;
425 mg. agar per 100 ml. total volume of medium were used.

The solutions utilized were prepared as follows :

Chick Ringer solution. 0.9 per cent NaCl, 0.042 per cent KC1, 0.024 per cent
CaCl, in distilled water.

Penicillin-streptomycin solution. To 100,000 units penicillin-G-potassium and
20 mg. dihydro streptomycin sulfate (Squibb) add 20 ml. sterile Ringer saline.
Refrigerated (frozen) this is claimed to be bacteriologically effective for one week.

Phenol red. 0.001 per cent solution in Ringer.

Phosphate buffer. Add 100 ml. distilled water to 0.290 gm. Na,HPO, - - 12HX>
+ 0.052 gm. KHPO4. Autoclave to sterilize.

Bicarbonate buffer. Saturate a solution of 1.100 gm. NaHCO,, in 100 ml.
distilled water with CCX. This is accomplished by blowing expired air into the
solution through a tube for an extended period of time (ca. one hr.). Sterilize In-
filtration.

Sugar. Stock solution consists of 400 mg. monosaccharide per 100 ml. Ringer
solution (400 mg% = 0.022 M). This was diluted to the desired concentration in
the total volume of other medium ingredients. As an example, in preparing 40 ml.
of 100 mg% glucose medium the following would be incorporated : Ringer (21 ml.),
glucose stock solution (10 ml.), penicillin-streptomycin (4 ml.), phenol red (2 ml.),
phosphate buffer (2 ml.), bicarbonate buffer (T ml.), agar (170 mg.).

In making the media the agar. Ringer solution, phenol red and sugar prepara-
tion were combined and autoclaved. On cooling to approximately 45° C. the
penicillin-streptomycin and sterile phosphate and bicarbonate buffers were added.
After mixing by swirling this was poured into the watch glasses, where gelation
occurred.

Aseptic technique was used throughout the experiments. The antibiotics
were used as a precautionary measure. It was essential to guard against bacterial
contamination, a factor not prominent when using an albumen medium. All of
the equipment was dry-sterilized at 350° C. for 1.5 hours.

For the purpose of determining the nutritional requirements for wound healing
in the blastoderms the following sugar concentrations were used : glucose, 100 mg%,
50 mg%, 10 mg% and 5 mg% ; fructose, 100 mg% and 50 mg% ; galactose,
50 mg%. Embryos explanted on a non-nutrient medium (all constituents except
the carbohydrate) served as controls.

CHICK NUTRITION" IX \VOU\D HF.AI.IXG
RESULTS

41

The results of the experiments may he seen in Tahle I.

It will he observed that there was no healing except in one case in the controls
on a medium lacking a carbohydrate substrate. This would imply that there
was insufficient endogenous utilizable material for the necessary cell movements
in wound healing. It was apparent that the embryos could succeed very well in

TABLE I

Nutritional requirements for wound healing

Number healed

Medium

No. explains

8db Hrs

20 ± Hrs

Total

Saline

27

0

1

1

Glucose 100 mg%

15

12

3

15

Glucose 50 mg%

12

12

—

12

Glucose 10 mg%

15

/

7

14

Glucose 5 mg%

20

5

7

12

Fructose 100 mg%

6

0

2

2

Fructose 50 mg%

17

1

8

9

Galactose 50 mg%

12

0

6

6

this process on media containing 100 mg% and 50 mg'/e glucose. These media
have been found by Spratt (1949a) to be adequate for the development of normal-
appearing embryos. The value of the additional carbohydrate source was obvious
in these explants ; wounds healed completely on those media containing glucose
and essentially never on those without a sugar.

On the medium containing 10 mg'/c (5.5 X 10~4 AI ) glucose the blastoderms
began to show the effects of substrate dilution. While essentially all (14 out of 15)
explants healed by 20 hours, only one half had done so at 8 hours. Compare this
with those on 100 mg% and 50 mg% glucose, where practically all healing had

B

FIGURE 1. Camera lucida diagrams of an explant on agar-saline (control) medium to
illustrate progress in wound closing and embryonic development. A = initial ; B = after 8
hours ; C := after 20 hours.

42

RONALD C. FRASKK

FIGURE 2. A young chick blastoderm showing the location of a wound in the pellucid
area. X 20.

FIGURE 3. The appearance of a wounded chick blastoderm which had been incubated on

a non-nutrient (agar-saline) medium for twenty hours. The arrow indicates an unhealed
wound. X 15.

FIGURE 4. Photograph of a chick blastoderm which had been wounded and explanted on
a medium containing glucose in concentration of 5 mg% for twenty hours. Note that in this

CHICK NUTRITION IN WOUND HEALING 43

occurred by 8 hours. Of the twenty blastoderms explanted on the 5 mg%
medium only twelve had healed by 20 hours, and of these only five had done so
at the end of 8 hours.

The volume of the medium on which the embryos lay was very large (ca. 2 ml. )
compared to the size of the blastoderms. It is safe to assume that at the end of
twenty hours there was, for all practical purposes, approximately the same amount
of sugar in the medium as there was initially. It is thus inconceivable that there
was a significant depletion of the exogenous substrate. The argument that there
may have been a localized depletion at the site of the embryo also cannot be valid,
because there was a film of solution around them through which nutritional ma-
terial could pass. There was thus a constant supply of substrate available to the
explants. It is only logical to assume that the reason for the delay in closure or
the failure to heal was that, while the total amount of sugar was ample, the con-
centration of it was insufficient to meet the requirements of the embryo for this
purpose.

Both fructose and galactose were not as adequate as glucose as carbohydrate
sources. Similarly it was found by Spratt (1949a) that they were inferior to
glucose in terms of their utility in general developmental processes. This pre-
sumably reflected an inadequacy on the part of the embryos of this stage to bring
these monosaccharides into the general glycolytic scheme.

Turning to the development of the embryos it was found that the controls
continued to develop on the sub-minimal medium. By 8 hours the head was well
undercut, neurulation was evident and somites were present. At this time the
holes had closed somewhat in most cases. At 20 hours, however, there was
evidence of degenerative changes. The anterior region had developed somewhat
more than at 8 hours, but the node had become quite opaque, and had failed to
regress (Fig. 3). When a drop of saline was placed gently on the embryos at this
time, there was marked cell dispersal at the node region. This area has been
generally recognized as one of high metabolic activity (Moog, 1943 ; Hyman, 1927),
showing great sensitivity to metabolic inhibitors (Spratt. 1950b). The effects of
starvation were therefore in accordance with these previous observations. While
there was limited development in the embryos between 8 and 20 hours, there were
no significant changes in wound sizes during this period. This point will be con-
sidered in more detail shortly. Figure 1, showing camera lucida illustrations
made of an explant at time 0, 8 and 20 hours, reveals these points.

Development on glucose media in concentrations of 100 mg^/r and 50 nig9£
was fairly normal over the twenty-hour period with brain, heart and somites form-
ing. Such media, however, were not as adequate as an albumen medium for
typical embryonic development. On 10 \v\g% glucose the explants still continued
to develop over the twenty-hour period, there being no indication of nutritional
deficiency except as indicated ; there was a longer period of time required for
wound healing. If there was any damage to the node region, it was not readily
apparent in the explants.

embryo the wound has healed, that development is limited to the anterior end, and that the
node shows signs of marked deterioration. X 15.

FIGURE 5. A chick blastoderm after twenty hours of incubation on a fructose medium (50
mg%) following injury. The same features as indicated in Figure 4 may be seen. X 15.

44

RONALD C. ERASER

On the 5 mg% glucose medium there was no perceivable influence of nutritional
deficiency at 8 hours, but by 20 hours these blastoderms resembled the controls
(Fig. 4). While there was slight development between these time intervals,
again this was limited to the head region, with lower (axial) levels showing de-
generation. The somites which had formed became indistinct, the node region
failed to regress and took on an opaque appearance. Illustrating details by the
camera lucida became very difficult, due to the lack of translucency and the dispersal
of cells. In these embryos the wounds were in all stages of closure at 8 hours.
After this time about one half of the unhealed ones completed this process, as shown
in Table I.

Blastoderms on fructose and galactose media showed the effects of nutritional
deficiency in terms of development as well as in wound healing. These embryos
behaved very similarly to those on 5 mg% glucose, again showing marked node
deterioration (Fig. 5).

Pretreatment study

There was one further factor to consider in attempting to determine the minimum
concentration of exogenous substrate required for wound healing. This was

TABLE II
Wound healing on minimal media after pretreatment

Medium

No. explanted

Xo. hrs. pretreated

Xo. healed after -1 2 ± hrs.

Glucose 10 mg%

12

5

12

Glucose 10 mg%

12

10

12

Glucose 5 mg%

18

10

12

Glucose 5 mg%

18

20

18*

* Hole filled in by loose cells during embryo degeneration.

the endogenous material present in the embryo itself. It will be noticed that even
on a medium lacking an}- carbohydrate there was a certain amount of development.
Wounds in explants on this medium also started to heal, and in one case closed
completely. If, then, the effect of the carbohydrate alone was to be determined,
it was necessary to minimize the endogenous substrate factor.

This was done by "pretreating" the embryos on a non-nutrient (agar-saline)
medium prior to explanting them on another containing the sugar. The blasto-
derms were placed on the non-nutrient medium for a specified period of time,
removed and wounded, and then transferred to the sugar-bearing medium. It
was desirable to have the blastoderms use up most of their own available energy
sources without damaging them beyond recovery.

After 8 hours of incubation there had been no apparent closure in the controls.
The pretreatment period was therefore set around this interval of time. Some
embryos were starved for five hours, some for ten, and some for twenty hours.
Glucose concentrations of 10 mg^ and 5 mg^: were used, because it was around
these values that the effect of deficiency became apparent.

CHICK NUTRITION IN WOUND HEALING

Table II indicates the results obtained. From the data it seems that glucose
in a concentration of 10 mg% was sufficient for the embryos to use in healing. This
is evident even after 10 hours of pretreatment. Blastoderms explanted on a
medium containing 5 mg% glucose were found, after 12 hours incubation, to
have healed in approximately the same proportion as those not treated. Not
included in the table were twelve embryos that were pretreated, wounded, and
then transferred again to a saline-agar medium. These served as controls. None
of these had healed within the twelve-hour period. From these data it may be
concluded that 5 mg% glucose is near the minimal concentration that meets the
requirements of the blastoderms in wound closing. Differences in ability to heal
on this medium may be interpreted as an indication of variability in the embryos
themselves, when explanted on synthetic media.

After twenty hours of pretreatment the node region of the explants was under-
going degeneration. When these embryos were wounded and explanted on the
5 mg% medium, it was noticed after 12 hours that the wound had apparently healed.
This is indicated in Table II. When a drop of saline was gently placed over the
blastoderms by means of a wide-bored pipette, however, the cells in the wound
area and those at the node dispersed leaving holes in the explants. This healing
was therefore not taken as normal in the sense that it was brought about by the
general movements of normal tissues, but was the result of the association of cells
dispersed from other areas during deterioration of the embryos.

Elsewhere (Fraser, 1953) it was mentioned that attempts to incubate the
blastoderms under the medium against a cover glass failed. It was not known
whether the inability of the embryos to develop was the result of insufficient oxygen
or of some other agent, such as pressure of the medium on the embryo. In the
course of this study the effect of anaerobiosis on wound closure was explored.
Twenty embryos of 20 hours incubation were used for this purpose.

Following wounding these embryos were explanted on an agar-albumen
medium in the usual manner (see Spratt, 1947). The lids of the petri dishes in
which the embryos were incubated were kept elevated slightly with pieces of
aluminum foil to permit the removal of oxygen. Moist cotton rings were again
used in the petri dishes to maintain a moist atmosphere. The petri dishes were
placed in a large desiccator into which was poured 50 ml. of 40 per cent pyrogallic
acid and 100 ml. of 20 per cent KOH for the removal of oxygen. The lid was
sealed immediately and the desiccator placed in the incubator.

After 12 hours the explants were removed for observation. At this time it
was found that there was no embryonic development, the wounds were as large
as initially made, and the blastoderms showed extensive deterioration. They had
an opaque appearance masking all internal structure. When a drop of Ringer
solution was gently placed on them, they tore apart with much cell dispersal, quite
unlike blastoderms of a comparable age cultured on this medium under normal
conditions. This indicated that normal cell-cell adhesiveness was lacking. It was
apparent from this that oxygen is required for wound healing, and, in general,
for all normal development. Since carbon dioxide was also removed by the
alkaline pyrogallol, the effects may be in part due to its deficiency as well (Spratt,
1949b). The problem of oxygen requirement for early chick development has been
considered elsewhere (Philips, 1941, 1942; Spratt, 1950a).

46 RONALD C. FRASER

DISCUSSION

There exists, as would be expected, a good correlation between the ability of
the embryos to undergo development (including all of its component processes) and
extra-embryonic tissue movements on various media. This correlation is not
perfect, however, as evidenced by the fact that on a non-nutrient medium the
blastoderms continued to develop (even though limited to the anterior region)
after 8 hours of incubation, while the wounds failed to undergo any appreciable
change in size beyond this time. This is a strange situation in view of the fact
that one might assume that more carbohydrate (-- potential energy or carbon
skeleton source) would be utilized in morphogenesis, histogenesis of tissues,
maintenance etc. occurring in the embryo proper than in more peripheral areas of
the pellucid area, where presumably little such activity is taking place. If this
assumption is correct, and we have no basis for not believing it to be so, then an
answer must be sought for the observations.

It is not likely that more carbohydrate is required in tissue movements in-
volved in wound healing, because there is more of such activity taking place in the
development of the embryo itself. In this regard, however, it is interesting to
note that the trunk level of the embryos failed to develop, but this is associated with,
and probably a consequence of, the degeneration of the whole node region.

Since embryonic development ensues on a non-nutrient medium after general
cell movements cease in the extra-embryonic region, and since it is assumed that
more exogenous nutrient is required in the axial area, the suggestion is made
that there may be a greater concentration of endogenous substrate localized in the
embryo proper than in the outlying pellucid tissues. Although we have no. direct
evidence for this, such a localization is not hard to conceive in view of the general
consideration that substrate and corresponding enzyme are found together. The
presence of indophenol oxidase and dehydrogenase activities restricted mainly to
axial tissues has been demonstrated by Moog (1943) and Spratt (1952), re-
spectively. It is stressed, however, that the demonstration of localized enzyme
activity cannot be offered as proof that there is an accumulation of its correspond-
ing substrate.

SUMMARY

1. Wounds produced in the pellucid area of chick embryos cultured on a
non-nutrient medium failed to heal within 20 hours. Although there was little
or no change in the dimensions of the holes after 8 hours, differentiation of the
head region continued beyond this time.

2. On media of 100 mg%, 50 mg% and 10 mg% glucose the blastoderms
healed, for the most part, within 8 hours, while development continued in the
embryo proper. On a medium containing glucose in concentration of 5 mg%,
however, about one half of the wounds did not heal, correlated with degenerative
changes in the embryos, principally at the node, occurring between 8 and 20 hours
after injury.

3. Fructose and galactose were found to be quite ineffective as carbohydrate
sources for the closure of wounds and for development in general. Results using
these media were comparable to those when 5 mg^- glucose was utilized.

CHICK NUTRITION IN WOUND HEALING 47

4. By pretreating the blastoderms for 5 and 10 hours on saline-agar prior to
wounding, with subsequent transfer to media containing glucose, it was determined
that 5 mg% glucose was approximately the minimal concentration required by
the embryo for wound closure.

5. In view of the observation that embryonic development continued beyond
the time when healing stopped, and because it was assumed that more carbohydrate
must be required for the former to take place, it was postulated that there is a
greater concentration of endogenous substrate localized in the axial tissues than
in the outlying pellucid region.

LITERATURE CITED

ERASER, R. C., 1953. Studies on morphogenesis in the young chick embryo cultured in vitro.
Unpublished thesis.

HYMAN, L. H., 1927. The metabolic gradient of vertebrate embryos. III. The chick. Biol
Bull., 22: 1-39.

MOOG, F., 1943. Cytochrome oxidase in early chick embryos. /. Cell. Comp. Physiot., 22:
223-231.

PHILIPS, F., 1941. The oxygen consumption of the early chick embryo at various stages of
development. /. E.vp. Zool, 86 : 257-287.

PHILIPS, F., 1942. Comparison of the respiratory rates of different regions of the chick
blastoderm during early stages of development. /. E.rp. Zool., 90 : 83-100.

SPRATT, N. T., JR., 1947. A simple method for explanting and cultivating early chick embryos
in vitro. Science, 106: 452.

SPRATT, N. T., JR., 1949a. Nutrition requirements of the early chick embryo. I. The utiliza-
tion of carbohydrate substrates. /. E.vp. Zool., 110: 273-298.

SPRATT, N. T., JR., 1949b. Carbon dioxide requirements of the early chick embryo. Anat. Rec.,
105: 118.

SPRATT, N. T., JR., 1950a. Nutritional requirements of the early chick embryo. II. Differ-
ential nutrient requirements for morphogenesis and differentiation of the heart and
brain. /. Exp. Zool, 114 : 375-402.

SPRATT, N. T., JR., 1950b. Nutritional requirements of the early chick embryo. III. The
metabolic basis of morphogenesis and differentiation as revealed by the use of in-
hibitors. Biol Bull, 99: 120-135.

SPRATT, N. T., JR., 1952. Differentiation in reducing enzyme systems in the early chick
blastoderm. Anat. Rec., 113 : 602.

EFFECT OF LOWERED INCUBATION TEMPERATURE ON THE
GROWTH AND DIFFERENTIATION OF THE
CHICK EMBRYO l

JOHN R. HARRISON 2 AND IRVING KLEIN 3
Department of Zoology, Miami University, Oxjord, Ohio

Development of an animal embryo under normal conditions consists of three
fundamental processes: 1) growth, meaning an increase in mass; 2} differentiation,
which includes visible changes in form (morphogenesis) and invisible changes on
the molecular level ; and 3 ) maintenance, which permits the embryo to maintain its
structure intact. Many investigators have studied the effect of temperature upon
development with respect to the commercial implications and to gaining knowledge
conc^-ning the developmental processes. Broca (1862) held chicken eggs at room
temperature (20°-30° C.) for 27 days before incubating them and obtained blasto-
derms without primitive streaks. This anomaly had been described by Panum
(1860) and was later given the name of "anidian" by Dareste (1877). The anidian
chick blastoderm is often cited as an example of growth without differentiation. It
is a blastoderm in which growth, due to cell proliferation, has occurred without evi-
dence of embryonic axiation.

Dareste (1877) divided the anidians into two principal types: those blastoderms
which possess normal ectoderm and endoderm but which lack mesoderm and a
primitive streak ; and those blastoderms which possess embryonic areas more, or less
degenerated. Other types have been subsequently described : 1 ) blastoderms with-
out morphological traces of the embryo and which show no development of the
area vasculosa; 2) blastoderms without embryonic axiation but which show develop-
ment of the area vasculosa; and 3) blastoderms with a hole in the center of the area
pellucida and without trace of an embryonic structure (Grodzinski, 1933; Tur.
1907).

The present investigation was undertaken to determine if the definition of an
anidian as given by Needham, a blastoderm in which "an active proliferation of cells
goes on but no trace of axiation appears" (Needham, 1950, p. 223), is valid. A
more general definition, as given above, is accepted by many and does not exclude
the formation and later disintegration of a streak in these blastoderms. Both view-
points include the final morphological structure of a blastoderm which has increased
in size but which does not possess a primitive streak. In this respect both agree
with Dareste's (1877) original conception when he called these anomalies anidians,
meaning "without form." In addition to the re-evaluation of the anidian, this in-
vestigation was carried out in the hope that the results obtained would provide a
basis for further investigation of growth and differentiation.

1 The authors wish to express their appreciation to Dr. C. G. Grosscup for aid in the statistical
analysis, to Mr. Daniel Reynolds for help with the photography, and to Dr. H. Burr Steinbach
for critical suggestions in the preparation of the manuscript.

- A portion of this investigation was made possible by a grant from the National Science
Foundation.

3 Material in this paper was used in partial fulfillment of the requirements for the M.S.
degree from Miami University, Oxford, Ohio.

48

CHICK EMBRYO AT LOWERED TEMPERATURE 49

The present experimental procedure was followed because it is felt that no one
has followed the development of blastoderms at one experimental temperature for
prolonged periods. The principal paper on the anidian is usually stated as being
that of Edwards (1902) in which he analyzed the index of development for incuba-
tion temperatures from 21° to 31° C. for five to eight days. Within this tempera-
ture range he obtained an overall yield of about 63% anidian blastoderms. A
temperature of 25° C. was chosen as being above Edwards's (1902) physiological
zero (20°-21° C.) but sufficiently below normal temperature to produce an effect.

The blastoderms incubated in this investigation at 25° C. for two to fifteen days
show a slowed development, followed by a disintegration with continued incubation.
The two processes, growth and differentiation, are affected differentially.

MATERIALS AND METHODS

Fertile eggs were obtained locally from a flock of two year old New Hampshire
Reds during the winter of 1951-52. A total of 806 eggs was used. All eggs were
less than three days old and had been stored at temperatures of 5°-18° C. previous
to incubation. Two identical forced-draft, thermostatically controlled incubators
were employed with a constant humidity of 65 ± 5.0%. The experimental incu-
bator was held at 25 ± 0.5° C.. while the control was kept at 37.5 ± 0.5° C. All
eggs were placed on their sides in the incubators. Incubation at the normal tem-
perature produced normal embryos for all the incubation periods used.

After incubation the eggs were candled to determine the position of the blasto-
derm. Windows, about one centimeter square, were cut in the egg shells over the
blastoderm region. Blastoderms were stained by placing small squares of neutral-
red-impregnated-Bacto Agar on top of the vitelline membrane immediately over the
blastoderm (Hamburger, 1942). Each blastoderm was then measured in vivo with
respect to the total blastoderm width, the area pellucida width, and the area opaca
width on one side, using a Spencer binocular microscope equipped with an ocular
micrometer. After blastoderm measurement, the egg \vas broken into a dish con-
taining Ringer solution.4 The blastoderm was cut from the yolk, floated free, and
transferred to a Syracuse dish containing Ringer solution. In this dish the vitelline
membrane and excess yolk were removed by means of fine steel needles. After
noting the gross morphology, the blastoderm was fixed, stained with Delafield's
haematoxylin and mounted in toto for further gross examination. Representative
mounts were embedded, serially sectioned at 10 p. and remounted for an analysis of
their cellular morphology.

RESULTS
The blastoderm measurements

The experimental blastoderm measurements, with their standard deviations
and coefficients of variability, are presented in Table I. The diameter increases
from 4.07 mm. in the unincubated eggs to 8.29 mm. in eggs incubated for twelve
days. These data show clearly that the growth of the total blastoderm is de-
pendent mainly on the increase of the area opaca, as the area pellucida remains
relatively constant in size throughout this incubation period.

4 0.9% NaCl. 0.042% KC1, and 0.024% CaCl,.

50

JOHN R. HARRISON AND IRVING KLEIN

TABLE I

Average measurements of experimental blastoderms resulting from incubation

at 25° C. for 0-15 days

Width of blastoderm

Width of area pellucida

One side of area opaca

No.
days at
25° C.

No.
anidians

Size

<r*

v**

Size

a

V

Size

a

V

in mm.

in mm.

as ' ,

in mm.

in mm.

as ' ,

m mm.

in mm.

as %

0***

68

4.07

0.54

13.31

2.23

0.38

17.20

0.94

0.27

29.49

2

15

4.68

0.68

14.48

2.30

0.26

11.45

1.28

0.31

24.12

3

13

4.61

0.35

7.64

2.22

0.41

18.45

1.20

0.27

22.43

4

16

4.64

0.64

13.77

2.38

0.40

16.93

1.29

0.34

25.99

5

6

5.08

0.76

14.88

2.51

0.46

18.18

1.31

0.29

22.00

6

7

5.62

0.46

8.09

2.62

0.23

9.05

1.53

0.20

12.78

7

6

5.05

0.66

13.12

2.57

0.30

11.82

1.24

0.32

25.54

8

2

5.00

0.14

2.80

2.57

0.29

11.13

1.29

0.14

11.13

9

2

5.80

0.37

6.40

2.91

0.06

2.06

1.43

0.28

20.03

10

3

8.19

0.48

5.81

3.77

0.54

14.01

1.54

0.12

7.84

12

9

8.29

0.72

8.70

3.11

0.95

34.89

2.60

0.43

16.68

14

5

8.22

1.79

21.74

2.91

0.28

9.62

2.76

0.81

29.23

15

7

7.97

1.56

18.32

2.19

0.33

14.86

2.57

0.55

21.44

* a designates the standard deviation.

** V designates the coefficient of variability.

*** These measurements were made to obtain a pre-incubation blastoderm size to be used in
growth comparisons.

A series of eggs was incubated at normal temperature (37.5° C.) for two to
eighteen hours to determine the normal blastoderm size changes over a comparable
period of growth. In each of these runs a two hour warm-up period was used, but
this was not included as part of the incubation time. These blastoderm measure-
ments, with their standard deviations and coefficients of variability, are presented in

TABLE II

Average measurements of normal blastoderms of eggs incubated at 37.5° C.

Width of blastoderm

Width of area pellucida

One side of area opaca

incubated

of eggs

Size

<T*

v**

Size

a

v

Size

(T

V

in mm.

in mm.

as%

in mm.

in mm.

as%

in mm.

in mm.

as %

0

68

4.07

0.54

13.31

2.23

0.38

17.20

0.94

0.27

29.49

2

12

4.55

0.20

4.45

2.21

0.25

11.22

1.17

0.27

23.27

4

12

5.15

0.93

18.12

2.51

0.32

12.61

1.39

0.44

31.33

6

12

5.61

1.80

32.02

1.99

0.43

21.63

1.68

0.64

38.14

8

12

6.11

0.89

14.51

2.36

0.28

11.68

1.89

0.40

20.98

10

12

6.35

0.69

10.91

2.45

0.30

12.26

1.93

0.44

22.88

12

11

8.37

1.56

18.68

2.57

0.80

31.22

3.01

0.56

18.67

14

12

9.81

0.63

6.45

2.55

0.33

12.02

3.58

1.08

30.19

16

12

9.83

0.84

8.57

2.62

0.27

10.36

3.74

0.94

25.27

18

11

10.28

3.73

36.29

2.36

0.41

17.16

4.46

0.80

17.04

* o- designates the standard deviation.

** V designates the coefficient of variability.

CHICK EMBRYO AT LOWERED TEMPERATURE

51

Table II. Normal blastoderms increase throughout the incubation period to 10.28
mm. for eggs incubated eighteen hours. As with the experimental blastoderms,
growth of the area opaca is responsible for normal growth of the blastoderm, since
the width of the area pellucida remains relatively constant.

In Figure 1 the growth curve for blastoderms incubated at 37.5° C. has been
superimposed on the growth curve for blastoderms incubated at 25° C. The scale

COMPARISON OF BLASTODERM MEASUREMENTS
OF EGGS INCUBATED AT 25° C. & 37.5'C,

USING SIGHT FITTED CURVES

10.0

9.0

8.0

7.0

U

N 6-0

CO

5.0

4.0

INCUBATION AT:

*• x 25° C.

. 37. 5° C

DAYS

HOURS

8

10

12

14

16

16

6 8 10 12 14

LENGTH OF INCUBATION

FIGURE 1

for the hours incubation at the higher temperature has been adjusted for comparison
of the growth to about 5.75 mm. The graph shows that blastoderms must be in-
cubated at the low temperature for about thirty hours to attain the size of normal
blastoderms incubated at the higher temperature for one hour. Analysis of the
average growth rates of blastoderms incubated at the two temperatures shows that
the high temperature blastoderms grow about thirty-four times faster than those
at the low temperature. Sixty-seven blastoderms incubated for 2 to 9 days at the
low temperature had an average hourly increase of O.OOS8 mm., while one hundred

52

JOHN K. HARRISON AND IRYIXC KLEIN

and six blastoderms incubated for 2 to 18 hours at the high temperature had an
average hourly increase of 0.3016 mm. In this comparison only growth of the low
temperature blastoderms during the first nine days was considered, since it is felt
that the increase from the ninth to tenth day is not true growth, rather an increase
in diameter resulting from degenerative spreading.

ANIDIAN AND DEGENERATE STREAK
PRODUCTION AT 25° C.

100

14 15

ANIDIANS

U NORMALS
FIGURE 2

DEGENERATE

There is a corresponding slowed differentiation of primitive streaks in blasto-
derms incubated at the low temperature. Primitive streaks, as distinguished from
the more diffuse primitive shields, first appear in blastoderms incubated at the low
temperature after forty-eight hours of incubation. At the high temperature, primi-
tive streaks first appear after four hours of incubation. The ratio of development,
as measured by streak appearance, between blastoderms incubated at the high tern-

CHICK EMBRYO AT LOWERED TEMPERATURE

perature and those incubated at the Imv temperature is 1 : 12. If this ratio were to
hold true during continued development, one would expect 50% of the low tempera-
ture blastoderms to show axiation after about seventy-two hours of incubation,
since at the high temperature 50% of the blastoderms possess streaks after six hours
of incubation. The data show that it required twenty times as long ( 120 hours)
for 50% of the low temperature blastoderms to develop streaks. The extrapolation
cannot be made for 100% streak formation, since at the low temperature this per-
centage is never attained. The streaks which are formed degenerate, producing
blastoderms without axiation. The closest approach to this developmental state
is on the ninth clay when 91% of the blastoderms show axiation, either normal or
degenerate. Since 100% streak formation occurs after ten hours of incubation at
the high temperature, it takes the low temperature blastoderms about twenty-two
times as long (216 hours) to reach the same stage of development. An analysis of
variance was made to determine whether this apparent increased effect of low tem-
perature with continued incubation is significant or merely a chance phenomenon.
Comparison of percentages of blastoderms exhibiting streaks for incubation periods
of from two to nine days with the percentages occurring for individual experiments
of the same day showed that this effect of low temperature incubation is very sig-
nificant (F- ..,). Disregarding the individual variations in blastoderms, i.e., their
inherent capabilities to respond to the lowered temperature in differing ways, the
developmental picture of the average blastoderm is an enhanced effect of the low
temperature with continued incubation. The slowing effect of temperature is less
during the early period of incubation, increasing with continued exposure to the
low temperature.

The t/ross morphology

The general morphological changes occurring during prolonged incubation at
the low temperature are presented in Figure 2.5 The chart shows that anidians are
present early in incubation, comprising 71.5% of the cases. The remainder is made
up of blastoderms showing axiation. With continued incubation the percentage of
axiate embryos increases at the expense of the number of anidians. On the fifth
day of incubation a third form appears, the degenerating streak. It can be seen that
with continued incubation the percentage of anidians decreases, due presumably to
development of axiation in blastoderms which were anidian. However, at the same
time the streaks are commencing to degenerate, thus increasing the number of de-
generate forms. By the tenth day of incubation all of the streaks have disappeared,
leaving only anidians or degenerate streaks. The percentage of anidians at this day
is increased over that of the ninth day. This increase in the number of anidians
results from complete degeneration of streaks.

Examination of the various blastoderms resulting from prolonged incubation
at low temperature indicates that there are two types of anidians : 1 ) early anidians,
present from the onset of incubation to about the eighth day; and 2) later anidians,
resulting from degeneration of primitive streaks. Figure 3 is an early anidian,
corresponding to the first type described by Dareste. There is no embryonic axia-
tion present. The mid-streak blastoderm derived from the early anidian is shown

5 The data present in this chart were tested for analysis of variance with the following F
ratios: Anidian F3.0o; Normal F4.0o; and Degenerate F7.29. All of these trends are significant
at the 2% level or less.

54

JOHN R. HARRISON AND IRVING KLEIN

FIGURE 3. Early anidian ; 3 days incubation. 20 X.
FIGURE 4. Mid-streak blastoderm; 5 days incubation. 23 X.

FIGURE 5. Degenerating streak, showing tendency of node cells to disperse more slowly ;
5 days incubation. 13 X.

FIGURE 6. Later anidian, resulting from vacuolation of the streak; 9 days incubation. 13 X.

in Figure 4. These experimentally produced streaks never attain the maximum
streak length exhibited by blastoderms incubated at normal temperature. The
maximum streak length of a low temperature blastoderm was 1.999 mm. as com-
pared with 3.141 mm. maximum length of a normal temperature blastoderm.
Furthermore, as Edwards (1902) has pointed out, incubation at this low tempera-
ture never results in the formation of a notochord, neural plate or groove, or meso-

CHICK EMBRYO AT LOWERED TEMPERATURE

55

dermal somites. Rather the primitive streak, after appearing on the experimental
blastoderm, goes on to degenerate in one of the following ways : 1 ) by dispersal of
the streak cells peripherally to the margin of the area pellucida; 2) by the vacuola-
tion of the streak region ; or 3) by a combination of these forms of degeneration. In
Figure 5 degeneration has begun with the peripheral dispersion of the posterior cells.
A central clump of cells marks the node region which tends to disperse more slowly.
Degeneration of the second type is illustrated in Figure 6 in which the former streak
is vacuolated. Figure 7 illustrates the third type. Here the streak is degenerating
with a general cell dispersal and vacuolation of the posterior streak region. Figure
8 shows a secondary anidian formed by cell dispersal and some vacuolation.

The onset and particular forms of degeneration, resulting in blastoderms of the
later anidian type, appear to depend on the inherent qualities of the individual eggs,
as the same experimental conditions were used throughout the investigation. How-

a

FIGURE 7. Degenerating streak with general cell dispersal and vacuolation of the posterior
streak region; 9 days incubation. 14 X.

FIGURE 8. Later anidian, resulting from cell dispersal and some vacuolation of the streak ;
7 days incubation. 15 X.

ever, all the experimentally incubated blastoderms appear to go through the same
stages, i.e., early anidian with subsequent streak formation followed by degeneration
of the streak to produce later anidians. It is difficult to explain the morphological
changes in any other way. Final proof for this explanation, of course, would be to
follow single blastoderms throughout the prolonged period of incubation. This was
attempted but found impractical, since a very high mortality rate of blastoderms
occurs when windows are placed in unincubated eggs. However, it is felt that the
explanation for the morphological changes is supported by : 1 ) the significance of
the morphological trends . shown in Figure 2 which is great enough to eliminate
chance; and 2) the strikingly different morphological appearance of early and later
anidians.

56

JOHN R. HARRISON AXD IRVING KLEIN

The cellular morphology

Examination of serial sections of various blastoderms incubated at low tempera-
ture showed that the cellular picture was essentially normal. In those blastoderms
which had developed a streak the epiblast layer was three or four cells thick, forming
a more or less regular layer when compared with the hypoblast layer. In the latter
the cells were larger and frequently contained yolk granules. In the region of the
streak proper the cells were compact next to the epiblast layer, but loosely arranged
ventrally next to the hypoblast. In no section was mesoderm observed.

In older blastoderms in which the streak had degenerated, the picture was much
the same. In these the streak was gone, leaving the epiblast and hypoblast com-
pletely separated in the region of the area pellucida. The cells showed no evidence
of being under any stretching influence or of having changed shape. There was no
indication of basophilic granules, resulting from cytolysis. The only deviation from
normality was the presence of vacuoles. These occurred for the most part between
the epiblast and hypoblast layers, causing the later to bulge prominently. In many
of the vacuoles there was a neutral staining material, contracted in appearance as if
resulting from dehydration during staining. In the area opaca the epiblast was a
definite layer of one or two cells in close contact with the underlying yolk-laden cells.

TABLE III

Morphology of eggs incubated at 25° C. followed by two days incubation at 37.5° C.

Development in somites

Distinctly abnormal by-

TNJn

No

(

( -

days at
25° C.

of eggs

7-12

12-17

18-22

22

Mor-
phology

Degen-
eration

Hole

anidians*

Devel-
opment

Abnorma

2

12

0

6

6

0

0

0

0

100.0

0.0

4

11

0

0

6

5

0

0

0

100.0

0.0

6

12

0

3

6

3

0

0

0

100.0

0.0

8

11

1

3

7

0

0

0

0

100.0

0.0

10

10

2

6

1

0

1

0

0

90.0

10.0

12

12

1

0

0

0

6

5

0

8.3

91.7

14

12

0

1

0

0

5

5

1

8.3

91.7

20

12

0

0

0

0

0

6

6

0.0

100.0

* Tur (1907) and Grodzinski (1933) described blastoderms with holes in the center of the
area pellucidas and without traces of embryonic structure. They considered these to be non-
teratogenic formations. Such anomalies were encountered in eggs incubated at low temperature
for long periods and seemed to be the result of stress sufficient to cause a rupture in the blastoderm.

The effect of loiv temperature incubation on potency

The ability of eggs incubated at the experimental temperature to resume de-
velopment when shifted to normal incubation temperature was tested. The results
of this phase of the investigation are shown in Table III. In this work eggs were
kept for two to twenty-one days at 25° C. and then transferred to 37.5° C. for two
days. It was found that all the eggs kept for less than ten days at the low tem-
perature exhibited various degrees of normal, although slowed, development when
incubated at the normal temperature. But in eggs held for ten days or longer at
the low temperature there occurred a marked increase in the percentage of distinct

CHICK EMBRYO AT LOWERED TEMPERATURE

abnormalities. This is particularly evident in the drop in development from 90.0%
to 8.3% for the eggs held for ten and twelve days, respectively, at 25° C.

GENERAL DISCUSSION

Edwards (1902) noted that blastoderms incubated at low temperatures failed to
form notochords, neural plates and grooves, and mesodermal somites. Blastoderms
formed, at best, short primitive streaks. The failure in the present experiments to
obtain development beyond the primitive streak agrees with Edwards' work. Obvi-
ously the low temperature incubation affects one of the developmental factors con-
cerned with formation of the head process. Grodzinski (1933) believes that changes
in the primary germ wall cause peripheral dispersion of cells, resulting in failure of
head process formation. Needham (1950, p. 223) considers the anidian to be a
case in which there has been a "failure either of the formation, or more probably the
liberation, of the primary evocator." In either event a differential susceptibility to
temperature exists between the stages of development. The formation of a streak
is less sensitive than the formation of a head process, although both are dependent
upon cell movements. However, the streaks formed at the low temperature never
reach the length of streaks formed at the normal temperature.

The increased degeneration of low temperature axiate blastoderms between the
ninth and tenth days of incubation can be correlated with the nearly two-fold in-
crease in blastoderm size shown in Figure 1. Table I shows that this increase is
primarily due to an increase in the area pellucida. Prior to and following this
period, the area pellucida contributes very little to the blastoderm growth. The
degeneration of the streak region by vacuolation and cell dispersal could account
for this size increase. The tendency for the blastoderm size to decrease in Figure
1 after the twelth day is not valid, since the fourteenth and sixteenth day mean
measurements have a large standard deviation.

Another consequence of degeneration is the effect on the potency of the experi-
mentally incubated eggs. Where the degenerate streaks become the predominate
form of blastoderm, i.e., after the tenth day of incubation, the potency of eggs to
resume development when shifted to normal incubation temperature falls off strik-
ingly. If the eggs are kept at the low temperature for less than ten days, only
slowed development results. This fact was also brought out by Romanoff ct al.
(1933), who showed that exposure of 0-1 day old embryos to 29° C. for twenty-
four hours resulted in extremely retarded development, but had no significant effect
on embryo mortality.

The results obtained, showing a susceptibility of growth and differentiation to
low temperature, do not support the concept of a separation of the two processes
with low temperature incubation. Both growth and differentiation are affected
during the same period of incubation, i.e., at about the tenth day. It was pointed out
that the effect of low temperature upon streak formation is enhanced with continued
incubation. The same effect is seen in analyzing the rate of growth of the primi-
tive streaks. In such an analysis it is necessary to consider both normal and de-
generate streak lengths which introduces a question of inaccuracy due to either
shrinkage or expansion in the degenerate streaks. The growth rates of streaks in
the blastoderms incubated at the low temperature show a consistent decline from
0.0202 mm./hr. with two days of incubation to 0.0054 mm./hr. after nine days of in-

58 JOHX R. HARRISON AND IRVING KLEIN

cubation. The only exception to this consistent decline is during the eighth day when
the growth rate is higher than that of the seventh day. In comparison, the rate of
growth of streaks in the normal temperature blastoderms shows an initial high rate in
the streaks first appearing after four hours of incubation. This rate of 0.3033 mm./
hr. declines to 0.01185 mm./hr. in blastoderms incubated for ten hours and remains
more or less constant through fourteen hours of incubation. The sixteen and
eighteen hour streaks show a slight decline in rate to 0.09 mm./hr. The initially
high growth rate in the normal temperature blastoderms could be explained on the
variations in individual blastoderms, i.e., that these first streaks appear in blasto-
derms which show a faster initial development and represent a small percentage,
rather than the average.

The same phenomenon is not true for growth rates of the blastoderms incubated
at the two temperatures. The blastoderms incubated at the low temperature show
a consistent rate of growth through the tenth day of incubation at which time they
cease growing. The normal blastoderms, however, show a fairly consistent growth
rate which is broken only during the tenth to fourteenth hours of incubation when
there is an abrupt increase in rate. Although low temperature incubation does not
result in growth without differentiation, it does affect the two processes differentially.
Differentiation, as seen in axiation and head process formation, is more susceptible
than growth.

Interest in the anidian has centered on its being an example of growth without
differentiation. A better understanding of development would be possible if the
component processes could be separated and studied individually. The present in-
vestigation has shown that chick blastoderms incubated at low temperature develop
slowly. The anidians present early during incubation correspond to blastoderms
which show no axiation during the early hours of incubation at normal temperature.
However, the anidians which appear later in development are not comparable.
These correspond to the second type of anidian described by Dareste (1877) : those
which possess embryonic areas more or less degenerated. This type, however,
would not fall under Needham's definition of anidian which states that the anidian
is a blastoderm which has grown but which has failed to form a streak. Blasto-
derms which once possessed a primitive streak would not exemplify the instance
of growth without differentiation unless they continued to grow after the streaks
disintegrated. The results obtained show that growth and differentiation are
stopped at about the same time. The data obtained in the present investigation lead
to the conclusion that Needham's interpretation of the anidian anomaly is not valid.
However, the term anidian, meaning "without form," is aptly applied to either the
early blastoderms or the later blastoderms which show no axiation.

SUMMARY AND CONCLUSIONS

1. The present investigation was concerned with the effects of a prolonged in-
cubation on the size and morphology of chick blastoderms incubated at 25° C. for
two to fifteen days.

2. Comparison of the measurements of blastoderms incubated at the low tem-
perature with those incubated at the normal temperature showed that growth for
about thirty hours at the low temperature is equal to growth for one hour at the
normal temperature.

CHICK EMBRYO AT LOWERED TEMPERATURE 59

3. Comparison of the appearance and growth of primitive streaks in blastoderms
incubated at the low temperature with those incubated at the normal temperature
also showed a slowed development at the low temperature. This effect of tempera-
ture is enhanced with continued incubation.

4. The morphology of the blastoderms incubated for two to fifteen days at 25°
C. changes during the course of incubation. At the onset anidians predominate,
but give rise to axiate blastoderms after further incubation. Degeneration of the
streak by dispersal of the streak cells peripherally, by a vacuolation of the streak
region, or by a combination of both patterns, follows. Further degeneration of the
axiate blastoderms results in the formation of anidians once more.

5. A possible relationship between differentiation and growth is seen in the
cessation of growth concurrent with degeneration of the primitive streak.

6. Cytologically the blastoderms show7 normal epiblast and hypoblast forma-
tion, with the exception of numerous vacuoles appearing between the epiblast and
hypoblast in those blastoderms incubated for long periods.

7 . Degeneration at the lowered temperature is correlated with a two-fold in-
crease in the blastoderm size between the ninth and tenth days of incubation. It
also causes a marked decline in the potency of blastoderms to resume development
when shifted to normal incubation temperature.

8. Neither the early anidian, on which a primitive streak will differentiate, nor
the later anidian, resulting from degeneration of the axiate blastoderm, is a valid
example of growth without differentiation.

LITERATURE CITED

BROCA, P., 1862. Experiences sur les oeufs a deux jaunes. Annales des Sciences Naturelles, 4e

Serie. Zool., t. XVII, p. 81.
DARESTE, C. R., 1877. Recherches sur la production artificielle des monstruosites ou essais de

teratogenie experimentale. Reinwald, Paris.
EDWARDS, C. L., 1902. The physiological zero and the index of development of the eggs of the

domestic fowl, Callus domcsticus. Amer. J. Physio!., 6: 351-397.
GRODZIXSKI, Z., 1933. liber die Entwicklung von unterkuhlten Hiihnereiern. Arch. f. Entw.,

129: 502-521.

HAMBURGER, V., 1942. A manual of experimental embryology. Univ. of Chicago Press. Chicago.
NEEDHAM, J., 1950. Biochemistry and morphogenesis. Cambridge Univ. Press, Cambridge.

785 pp. Reprinted.
PANUM, P. L., 1860. Untersuchung iiber die Entstehung der Missbildungen, zunachst in den

Eiern der Vogel. Kiel.
ROMANOFF, A., LAURA L. SMITH AND R. SULLIVAN, 1938. Biochemistry and biophysics of the

developing hen's egg. III. Influence of temperature. Cornell Univ. Agr. E.vpt. Sta.

Memoir, 216 : 1-42.
TUR, J., 1907. Une forme nouvelle de 1'evolution anidienne. C. R. Acad. Sci. Paris., 144:

515-518.

NEUROSECRETION IX THE THORACIC GANGLION OF THE CRAB,

ERIOCHEIR JAPONICUS

KUNIO MATSUMOTO

Department of Bioloyy, faculty of Science,
Okayatna University, Okayaina, Japan

In a study of sacculinization in Charyhdis japonica (Matsumoto, 1952) the
author reported that the thoracic ganglia of the host crabs are damaged by the root
system of the parasites, Heterosaccus papillosus. The same kind of damage is also
observed in other species of sacculinized crabs. More recently, in a study of epi-
caridization in the fresh water crab, Eriochcir japonicus (Matsumoto, 1953), it was
observed that the thoracic ganglia of male hosts become greatly deformed by the
pressure of the parasites, Entionella flnviatilis. In order to elucidate the nature of
the abnormalities found in the ganglia of crabs subjected to so-called parasitic cas-
tration, a detailed histological study of the thoracic ganglion of normal animals
became necessary. In the course of this work the presence of neurosecretory cells
was observed in the thoracic ganglion of Eriochcir japonicus. While Enami ( 1951b)
found only one type of neurosecretory cells (a cells) in the thoracic ganglion of the
crab Sesarma, that of Eriochcir seems to contain three cytologically different types
of these cells. Their description, in the present paper, appears of interest, since
differences in cytological appearance may indicate different functions of the cellular
products. Furthermore, these cells exhibit signs of a mode of discharge of the
neurosecretory material which has so far been observed only in vertebrates.

MATERIALS AND METHODS

Eriochcir japonicus is a grapsoid crab, commonly found in the fresh waters of
Japan ; animals collected at the Asahi River in Okayama City were used for this
study. The observations are based on sectioned tissue of thoracic ganglia. The
crabs used (80 males and 72 females) were in various stages of development. They
ranged from 5 to 45 mm. carapace length. The thoracic ganglia were fixed in
Benin's solution or Zenker-formol and cut into serial sections oi 8 p thickness by the
usual paraffin method. They were stained \vith Gomori's chrome-alum hema-
toxylin and phloxine or Delafield's hematoxylin and eosin. Besides these methods
some materials were fixed in Susa or trichloracetic acid and were stained with
Mallory's triple stain or Masson's trichrome stain in order to compare them with
those described in Enami's (19511)) study.

OBSERVATIONS
1. Types of nerve cells and their location in the thoracic ganglion

There are four kinds of nerve cells in the thoracic ganglion of Eriocheir each of
which shows a definite localization. For the time being these cells are designated as
types A, B, C, and D. Their distribution is shown diagrammatically in Figure 1.
A-type nerve cells are giant elements with diameters of 80-100 /x in the adult and

60

NEUROSECRETION IN ERIOCHEIR

61

are mainly found in the medial and posterior parts of the ganglion. B-type nerve
cells are small with diameters of 15-20/x and are distributed all over the ganglion.
They are mingled with the A-cells in the medial and posterior parts. In the anterior
part, many B-cells are found on the ventral side. C-type nerve cells are also small
with diameters of 10-20 ju,; they are located in paired groups at the anterior end of

O A-cell
• B-cell

dorsal side

ventral side

FIGURE 1. Diagrammatic illustration of the distribution of four types of nerve cells in the
thoracic ganglion of Eriochcir japonicus. The dorsal and ventral sides of the ganglion are
shown separately. Each diagram is subdivided into three parts, anterior (ap), median (mp),
and posterior (pp), for convenience in description.

the dorsal side of the ganglion. D-type nerve cells are minute and are arranged in
densely packed, paired masses ventral of the C-cells.

2. Neurosecretory activity of different cell types

A-cell: The giant A-cells are considered to he neurosecretory on account of
cytological features such as modifications of their nuclei, the appearance of minute
granules, and the occurrence of small vacuoles. Some giant cells in the thoracic
ganglion contain nuclei which differ considerably from the round nuclei of ordinary
cells. These modified nuclei may be flat or crescent-shaped ; as a rule, their contents

62

KUXIO MATSUMOTO

ua*

FIGURES 2-11.

NEUROSECRETION IN KRIOCHEIR 63

show no affinity for nuclear stains. Many minute dark blue granules are seen to
gather closely around the crescent-shaped or Mat nuclei in preparations stained with
Gomori's method (Fig. 2). In the cytoplasm, the occurrence of small vacuoles is
noticeable. Some giant cells have a coarse cytoplasm as shown in Figure 3. This
appearance is considered as typical of a stage preceding that marked by vacuolated
cytoplasm illustrated in Figure 4. The latter seems to be the most vigorous stage of
secretion. Many small vacuoles appear over the whole cytoplasm ; some vacuoles
stain pale violet, others seem quite empty. This figure resembles one depicting cer-
tain neurosecretory cells in the suboesophageal ganglion of the cockroach, Leu-
cophaea, studied by B. Scharrer (1941a, Fig. 2C). Certain other A-cells have only
small vacuoles in the periphery of the cell body but never in the central portions
(Fig. 5).

Judging from all these figures, it may be assumed that the minute granules which
appear in the vicinity of the modified nuclei spread over the entire cytoplasm, while
simultaneously many small vacuoles containing neurosecretory substance appear in
the cytoplasm ; then the vacuoles gradually migrate toward the periphery of the cell
and disappear. This seems to be the mode of discharge of the neurosecretory
material in these cells ; it will be discussed in later sections of this paper.

During these secretory cycles apparent in the cytoplasm, the nucleus shows con-
comitant changes of its shape, i.e., the round nucleus gradually becomes flat or
crescent-shaped during phases of cytoplasmic activity and then expands and rapidly
regains its round shape during the resting stage of the cytoplasm. The nucleus
may, therefore, be considered as playing an important part in the neurosecretory
activity of the A-cells. These various pictures are commonly found in the thoracic
ganglion of the normal adult crab.

B-cell: B-cells are not only much smaller than A-cells but they also show a
different kind of neurosecretory behavior. The secretory activity of B-cells is illus-
trated in Figures 6-9. Since all photomicrographs shown in this paper have the
same magnification, differences in cellular size can be readily appreciated.

FIGURE 2. Giant neurosecretory cell (A-cell) showing many minute dark granules sur-
rounding a flat nucleus. Male, 30 mm. carapace length. Bouin, paraffin, 8 /*, Gomori's chrome
alum hematoxylin phloxine. Photomicrograph, X 410. Figures 3-9 are from the same speci-
men, shown at the same magnification.

FIGURE 3. A-cell with coarse cytoplasm. The minute granules are scattered throughout the
cytoplasm and the nucleus has regained its round shape. \

FIGURE 4. A-cell with many small vacuoles in the cytoplasm. The ellipsoid nucleus is
typical of this stage of secretory activity which is interpreted as the most vigorous in the cycle.

FIGURE 5. A-cell showing vacuoles only in the periphery of the cell. This figure may
represent the last stage in the secretory cycle.

FIGURE 6. Small neurosecretory B-cell with granulated cytoplasm containing many irregu-
lar masses.

FIGURE 7. B-cell containing dark staining larger masses in the cytoplasm.

FIGURE 8. B-cell in which two dark staining large masses are concentrated in the cell
periphery.

FIGURE 9. B-cell in which the dark cell inclusions have been replaced by two large vacuoles.

FIGURE 10. A group of secreting C-cells. These seem to correspond to Enami's /3 cells.
Male, 31 mm. carapace length. Zenker-formol, paraffin, 8 /*, Gomori's chrome alum hematoxylin
phloxine. Photomicrograph, X 410.

FIGURE 11. A mass of minute D-cells which never show any secretory activity. Note
round nuclei of uniform size. The cytoplasm is scarce and the cell boundaries are difficult to
discern. Same specimen as shown in Figure 10. Photomicrograph, X 410.

64 KUNIO MATSUMOTO

The first stage in the secretory cycle of B-cells may be that in which the cyto-
plasm as a whole becomes granulated and many small irregular masses appear in it.
as shown in Figure 6. Xo changes in the nucleus are observed. In the next stage,
these irregular masses show a tendency to aggregate into larger, darker staining
masses (Fig. 7). Presumably the third stage is depicted in Figure 8; the granular
material is concentrated into one or two large dark masses at the edges of the small
cell. Figure 9 illustrates the last stage of secretion. The dark masses have dis-
appeared, and large vacuoles remain in their place. The nucleus does not show any
signs of activity during the secretory process. The structure of the cell shown in
Figure 9 closely resembles the photomicrograph of a neurosecretory cell in the sub-
oesophageal ganglion of Blabcrns craniifcr published by B. Scharrer (1941a, Fig. 6).

These signs of neurosecretory activity are not observed in all B-cells present in
the thoracic ganglion, but there are regional differences. The secretory activity of
B-cells seems most pronounced in the posterior part ; it is less frequently observed in
the anterior part, and rather rare in the median part.

C-cell: C-cells have small granules and droplets in the cytoplasm which stain
with aniline blue. The number and size of the granules and droplets vary, but
in their general appearance the C-cells resemble Enami's (3 cells. When fixed in
Zenker-formol and stained according to Gomori, the C-cells show a characteristic
lumpy cytoplasm with black minute granules and small droplets (Fig. 10). The
C-cells are much less frequent than the A- or B-cells, but they all show signs of
secretory activity as described above.

D-cell: D-cells are the smallest nerve cells in the thoracic ganglion. The
nuclei are round and, for the most part, approximately uniform in size and ap-
pearance. The cytoplasm is not abundant but stains deeply with basic dyes ; the
cell boundaries are difficult to discern. These cells form two small, densely packed
masses as illustrated in Figure 11. They show no secretory activity.

3. Capillary networks in the thoracic ganglion

It is a matter of common knowledge that the crustaceans possess an open
circulatory system. There are several main arteries originating from the heart ;
these open into the haemocoel, and no capillaries are found in the majority of the
organs. Within the thoracic ganglion, however, there are exceedingly well de-
veloped capillary networks. They surround individual giant nerve cells, and enclose
groups of two or three smaller cells. Figures 12 and 13 show examples of these
capillary networks in the thoracic ganglion of Eriocheir. In Figure 12 capillaries
are seen to enclose giant nerve cells, in Figure 13 groups of smaller cells. This
capillary network branches out from five or six small arteries which enter the
thoracic ganglion in the mid-ventral region and pass through it to the dorsal side.
The rich capillary bed forms plexus around ventrally and dorsally located neuro-
secretory cells ; it also penetrates the entire medulla of the ganglion. On the
dorsal side certain capillaries join small vessels which, on leaving the ganglion, open
into the haemocoel ; others empty directly into the haemocoel at the periphery of
the ganglion.

Such a special arrangement of the circulatory system of the thoracic ganglion
suggests special functions. Neurosecretory cells must be expected to have a high

XEUROSECRET1ON IN ERIOCHEIR

65

metabolism and, therefore, require a rich blood supply. This is also the case in
vertebrates where nuclei composed of secreting nerve cells are among the most
richly vascularized of the central nervous system (see Scharrer and Scharrer,
1954). Furthermore, it seems that in the crab studied here, neurosecretory sub-
stances are given off at the periphery of the cell into these capillaries and thus reach
the general circulation of the body.

12

FIGURE 12. Capillaries closely surrounding individual giant neurosecretory cells within the
thoracic ganglion of Eriocheir. Female, 25 mm. carapace length. Capi, capillary. Zenker-
formol, paraffin, 8 /*, Masson's trichrome stain. Photomicrograph, X 410.

FIGURE 13. Capillaries enclosing groups of two or three small cells in the thoracic ganglion
of the same specimen as shown in Figure 12. Capi, capillary. Photomicrograph, X 410.

DISCUSSION

The histological examination of the thoracic ganglion of Eriocheir japonicus
showed the presence of three types of neurosecretory cells. In a recent publication,
Enami (1951b) studied neurosecretion in the central nervous system of Sesarma
and reported that only one type of neurosecretory cells (a cells) is found in the
thoracic ganglion of this crab. On account of their size, the giant A-cells in the
thoracic ganglion of Eriocheir would seem to correspond to the a cells of Sesarma.
But the morphological characteristics of these two cell types are quite different :

66 KUNIO MATSUMOTO

in particular, the central body of the a cell is never found in A-cells of Eriocheir, even
when the same fixation and the same staining methods are used.

The C-cells in the anterior region of the thoracic ganglion of Eriocheir have
the same histological structure as Enami's ft cells. But in Sesarma, these cells ac-
cording to Enami (1951b) occur in the optic ganglia, the brain, and the commissural
ganglia, but never in the thoracic ganglion. These differences in cellular distribution
may be considered as genus differences.

Some cellular details in the A-cells and B-cells of Eriocheir are quite similar
to those in the suboesophageal ganglion of cockroaches studied by B. Scharrer
(1941a). Thus neurosecretory processes in insects and crustaceans seem to have
certain features in common.

Concerning the discharge of neurosecretory substances, two possible ways were
considered in earlier studies on neurosecretion. Scharrer and Scharrer (1945)
in their review on neurosecretion described as one way the discharge of the secretory
substances into the capillaries, and as the other a transport of neurosecretory mate-
rial along nerve fibers. The latter way, i.e., the movement of neurosecretory
substances along axons, was confirmed by many recent investigators.

These observations resulted in the concept of neurosecretory systems, i.e., the
hypothalamic-hypophyseal system of vertebrates (Bargmann and Scharrer, 1951),
the intercerebralis-cardiacum-allatum system of insects (Scharrer and Scharrer,
1944; B. Scharrer, 1952), and the neurosecretory system of crustaceans (Bliss,
1951 ; Bliss and Welsh, 1952 ; Passano, 1951 }. Because of its physiological implica-
tions this concept received considerable attention in recent work on neurosecretion.
By comparison, less emphasis was placed on studies demonstrating the direct
discharge of neurosecretory substances from the cell surface into surrounding
capillaries or tissue spaces. There are many reported cases of neurosecretory
activity in which the mode of discharge of the secretory product either was not
studied in particular or could not be determined ; for example in the neurosecretory
cells of the stellate ganglion in vertebrates (Eichner, 1952) or of the central
nervous system of Limulus (B. Scharrer, 1941b).

With regard to crustaceans, Enami (1951b) suggested an axonal transport of
neurosecretory material, and Bliss and Welsh (1952) came to the conclusion that
in decapod crustaceans neurosecretory substances produced in various parts of
the central nervous system migrate along nerve fibers to the sinus gland where they
are stored and released.

The present study furnishes evidence of a different mode of discharge of
neurosecretory products in crustaceans. It is of considerable interest that, although
the circulation in crustaceans represents an open system, capillaries were found
to surround neurosecretory cells in the thoracic ganglion of Eriocheir. Moreover,
the neurosecretory substances produced in the perikaryon gradually seem to move
to the periphery of the cell and there to disappear. From these observations it may
be concluded that in the thoracic ganglion of Eriocheir, neurosecretory material
is given off directly into the surrounding capillaries and does not migrate along
nerve fibers. Thus both ways of discharge of neurosecretory products, as first
described by Scharrer and Scharrer (1945, 1954) for vertebrates, have now been
established also for invertebrates with an open circulatory system.

The physiological significance of the three types of neurosecretory cells in the
thoracic ganglion of Eriocheir is as yet unknown. Smith (1948), Enami (1951a),

XEUROSECRETION IN ERIOCHEIR 67

and Brown and his collaborators (Brown, 1949, 1950; Brown, Sandeen and Webb,
1949; Sandeen, 1950) demonstrated the existence of chromatophorotropins in the
thoracic ganglia of crustaceans. Furthermore, Brown and Cunningham (1941)
reported that neurosecretory cells in the central nervous system of Limulus furnish
chromatophorotropic principles. Hence, it seems reasonable to assume that some
of the neurosecretory cells in the thoracic ganglion of Eriocheir may be the source
of chromatophorotropic principles.

In recent studies of sacculinization and epicaridization of crabs ( Matsumoto,
1952, 1953), the thoracic ganglia of the hosts, which had changed to an intersexual
condition, were found to be damaged by the presence of the parasites. The deformed
ganglia contained fewer nerve cells than normal ones, and their distribution was
disarranged. It seems possible, therefore, that a relationship exists, either direct
or indirect, between the neurosecretory activity in the thoracic ganglion and the
development of secondary sex characters, but this possibility needs further
exploration.

The author wishes to express his gratitude to Dr. Berta Scharrer, University
of Colorado School of Medicine, Denver, for her criticisms of the manuscript and
to Dr. Frank A. Brown, Jr., Northwestern University, Evanston, 111., for his
continuous encouragement.

SUMMARY

1. The different types of nerve cells occurring in the thoracic ganglion of the
fresh w^ater crab, Eriocheir japonicus, their distribution, neurosecretory activity,
and the mode of discharge of the neurosecretory substances were studied.

2. There are four types of nerve cells which show a definite localization in the
thoracic ganglion ; three of them are considered to be neurosecretory cells. Giant
A-cells are interpreted as neurosecretory cells on account of the cyclic changes of
their nuclei and the gradual movement of many small vacuoles toward the cell
periphery. B-cells are small cells showing secretory cycles : numerous granules
appear in the cytoplasm, then concentrate into one or two masses at the edge of the
cells, and finally disappear leaving large vacuoles in their place. Their nuclei
show no changes. C-cells are also small neurosecretory cells ; they are thought to
correspond to Enami's ft cells. The minute D-cells do not posses characteristics to
suggest a secretory activity.

3. Several small arteries enter into the thoracic ganglion at the mid-ventral
region and pass through to the dorsal side, branching out into many capillaries.
These capillaries form networks and closely surround the neurosecretory cells.

4. From these observations it is concluded that the neurosecretory substances
in these cells are given off into the capillaries and thus reach the general circulation
of the body. This mode of discharge of the cellular product is of interest in view
of comparable mechanisms in vertebrates.

5. The physiological activities of these neurosecretory substances in the thoracic
ganglion of Eriocheir are as yet unknown.

LITERATURE CITED

BARGMANN, W., AND E. SCHARRER, 1951. The site of origin of the hormones of the posterior
pituitary. Amer. Sci., 39 : 255-259.

68 KUNIO MATSUMOTO

BLISS, DOROTHY E., 1951. Metabolic effects of sinus gland or eyestalk removal in the land

crab, Gccarcimis lateral is. Anat. Rec., Ill : 502-503.
BLISS, DOROTHY E., AND J. H. WELSH, 1952. The neurosecretory system of brachyuran

Crustacea. Biol. Bull., 103: 157-169.
BROWN, F. A., IK., 1949. The mechanism of color changes in Crustacea. Collecting Net,

19: 8-12.
BROWN, F. A., JR., 1950. Studies on the physiology of Uca red chromatophores. Biol. Bull.,

98: 218-226.
BROWN, F. A., JR., M. I. SANDEEN AND H. M. WEBB, 1949. Responses of the red chromatophores

of the fiddler crab. Anat. Rec., 105 : 615.
BROWN, F. A., JR., AND O. CUNNINGHAM, 1941. Upon the presence and distribution of a

chromatophorotropic principle in the central nervous system of Limulus. Biol. Bull.,

81 : 80-95.
EICHNER. D., 1952. Zur Frage der Neurosekretion in den Ganglienzellen des Grenzstranges.

Zcitschr. Zclljorsch., 37 : 274-280.
ENAMI, M., 195 la. The sources and activities of two chromatophorotropic hormones in crabs

of the genus Sesarma. I. Experimental analyses. Biol. Bull., 100: 28-43.
ENAMI, M., 195 Ib. The sources and activities of two chromatophorotropic hormones in crabs

of the genus Sesarma. II. Histology of incretory elements. Biol. Bull., 101 : 241-258.
MATSUMOTO, K., 1952. On the sacculinization of Char\bdis jafouica (A. Milne-Edwards).

Biol. J. Okayama Univ., 1 : 84-89.
MATSUMOTO, K., 1953. On the epicaridization of fresh water crab, Eriochcir japonicus (in

Japanese with English summary). Dobutsugaku Zasshi (Zoological Magazine), 62:

(in press).
PASSANO, L. M., 1951. The X organ-sinus gland neurosecretory system in crabs. Anat. Rec.,

Ill: 502.
SANDEEN, M. I., 1950. Chromatophorotropins in the central nervous system of Uca puyilator,

with special reference to their organs and actions. Physiol. ZooL, 23 : 337-352.
SCHARRER, BERTA, 1941a. Neurosecretion. II. Neurosecretory cells in the central- nervous

system of cockroaches. /. Coinp. Ncur., 74 : 93-108.
SCHARRER, BERTA, 1941b. Neurosecretion. IV. Localization of neurosecretory cells in the

central nervous system of Limulus. Biol. Bull., 81 : 96-104.
SCHARRER, BERTA, 1952. Neurosecretion. XI. The effects of nerve section on the inter-

cerebralis-cardiacum-allatum system of the insect, Lcucophaca madcrac. Biol. Bull.,

102 : 261-272.
SCHARRER, BERTA, AND E. SCHARRER, 1944. Neurosecretion. VI. A comparison between the

intercerebralis-cardiacum-allatum system of the insects and the hypothalamo-hypophyseal

system of the vertebrates. Biol. Bull., 87 : 242-251.

SCHARRER, E., AND B. SCHARRER, 1945. Neurosecretion. Physiol. Reviews, 25: 171-181.
SCHARRER, E., AND B. SCHARRER, 1954. Neurosekretion. In : v. Moellendorffs Handb. d. mikr.

Anat. d. Menschen, VI/5 (in press).
SMITH, R. I., 1948. The role of the sinus glands in retinal pigment migration in grapsoid

crabs. Biol. Bull.. 95: 169-185.

THE ANATOMY AND BEHAVIOR OF THE VASCULAR SYSTEMS
IN NEREIS VIRENS AND NEREIS LIMBATA 1

PAUL A. N I COLL
Department of Physiology, Indiana I'nii'ersity. Bloomington, Indiana -

No finer examples of contractile blood vessels may be found than those observed
in many species of annelids. Functionally considered this is rather surprising
since the annelids have few rigid structures ; as a group are highly motile ; and
rarely possess any tissue masses that are not bathed directly by their coelomic
fluid. The significance of these contractile vessels is further clouded by the fact
that within the class Chaetopoda one finds a very wide range of blood and vascular
development. This extends from species with a complete absence of any vascular
system through forms that exhibit various degrees of vascular development, coupled
with different levels of blood specialization, to specimens with completely closed
vascular systems some of which contain plasma or even cells with functional amounts
of hemoglobin (Romieu, 1923 ; Redfield, 1933).

Observations made on a variety of animals have indicated that the minute
-vessels of most vascular systems retain this power of contractility (Lutz, Fulton
and Akers, 1950). The long disputed site of this minute vascular behavior has to a
large extent been settled through the application of more refined techniques and
rigid terminology (Zweifach and Kossman, 1937; Clark and Clark, 1940; Nicoll
and \Yebb, 1946). It is now generally agreed that contractility depends on the
activity of smooth muscle-like cells that surround the endothelial tubes. Clark
and Clark (1947) have recently demonstrated that periendothelial muscle cells
develop along endothelial tubes following specific flow and pressure patterns within
the tubes and that these vascular structures then show typical contractility.

Webb and Nicoll (1944) have shown that lymphatics in the subcutaneous beds of
the bat show marked contractility which extends even to the large bulbous capillaries
(Webb, 1952). The activity of these bulbous lymphatic capillaries is of special
interest since they do not possess true muscle cells in their wall structure but have
syncytial cell mats that are the contractile elements. Their activity may reflect
a primitive mechanism and be functionally related to certain vessels observed in
the annelids.

Despite this recent revival and clarification of minute vascular contractility of
both the blood and lymphatic systems, no satisfactory explanation of its functional
significance has been advanced. It is more than likely that no single purpose is
subserved by this behavior and in the final analysis several mechanisms will be
shown to be dependent upon its existence. It was in the hope that a study of
contractility in the more primitive vascular systems of the Annelida might throw
added light on this basic activity that this investigation was undertaken.

1 This study was supported in part by grant H-676 Federal Security Agency, Public Health
Service.

- This study was carried out at the Marine Biological Laboratory, Woods Hole, under a
temporary appointment as Research Associate, Department of Physiology, Harvard University.
Boston.

69

70 PAUL A. NICOLL

The work was carried out at the Marine Biological Laboratory, Woods Hole,
during the summer of 1950. Except for preliminary observations on all species of
polychaetes collected, this study was restricted to the two species commonly
classified as Nereis virens and Nereis liinhata. The vascular anatomy and be-
havior of the entire specimen were established only for N. virens. However, at
the typical segmental level these aspects of N. liinhata were compared in detail
with the findings on N. virens.

Several deficiencies in the published descriptions of the vascular anatomy of
N . virens became evident early in attempts to analyze its vascular behavior. This,
coupled with failure to find any published descriptions of the vascular anatomy of
N. lirnbata, necessitated as a primary step a study of the vascular anatomy of
both forms.

Turnbull (1876) published the first detailed descriptions of Ar. virens, which
contains a brief discussion of its vascular system. Unfortunately he made several
mistakes that are quite significant from a functional standpoint. Because of his
excellent plates these have been handed on by most subsequent investigators.
Linville (1907) in a very brief note on the vascular system of Nereis, presumably
N. virens, actually corrected Turnbull's errors in so far as he went. His descrip-
tion of the lateral segmental vessels is essentially correct, but he failed to follow
through with sufficient detail for complete functional analysis. Federighi ( 1928)
followed Linville's description but made no significant extension or addition in
his published paper.

Carlson (1908) in a brief note describes the contractility of the vessels in
N. virens and concludes that extrinsic nerves appeared to play no part in initiating
or regulating their activity. Parker ( 1923) describes the muscle cells that are so
prominently seen after staining with methylene blue. They coil around most of
the vessels in N. virens, and he suggests that they are analogous with the Rouget
vascular cells in higher forms. Krogh (1922) also speaks of these cells as Rouget
cells. Federighi (1928) undertook an extensive analysis of the contractile elements
in the vessels of N. virens. His review of the earlier literature is quite extensive.
Following observations on living dissected specimens and detailed histological
study, he concludes that the vessels show two distinct types of motor response:
the regular rhythmical peristaltic contraction waves that pass along the vessels,
which he believes represent contraction of the endothelial cells ; and a localized
response limited to the region directly excited that is the result of the contraction of
the circular, slightly branched, smooth muscle cells on the outside of the endothelial
tubes.

MATERIALS AND METHODS

Specimens of N. virens were either obtained from the Supply Department of
the Laboratory or for the most part personally collected during low tide from the
shallow water of the cove on Nonamesset Island that lies directly across the Hole
from the Laboratory. Collections were made at least once a week since it proved
difficult to keep the worms in aquaria. Specimens of N. limbato were collected
during low tide from under the rocks along the shore between the Yacht Club
beach and the public dock adjacent to the Fisheries grounds. Fresh collections
were made every two or three days as required.

BLOOD VESSELS OF NEREIS 71

In order to evaluate the activity of a vessel in the vascular behavior of the
individual, it was necessary to work out the contribution of each vessel in the
intact living animal under conditions as near normal as possible. This required
that all information obtained by dissection studies be checked later on intact
specimens.

The manipulation of the living specimens in order that their vascular behavior
could be studied while observing them under the microscope, often at high magnifica-
tion, is a difficult task. The individuals are capable of considerable movement
of various kinds and maintain their activity for prolonged periods. Their natural
tendencies towards these various types of movement were further aggravated by
their extreme sensitivity to light stimulation. All parts of the worms seem to
possess light-sensitive receptors so that even the use of surviving segments could
not avoid this problem. Forcible restraint without anesthesia was generally
unsuccessful or impractical for microscopical studies. Of necessity, therefore, the
major portion of the analysis of vascular behavior was carried out on specimens
where some form of anesthesia was employed. The procedure of choice was to
work with fully anesthetized specimens during their recovery in sea water.

* Sodium amytal, usually applied by adding a few crystals to a petri dish of sea
water containing the worm, was the most commonly used anesthetic agent. There
is no doubt that vascular activity was reduced or even abolished during deep
anesthesia by this agent. However the worms always showed complete recovery
following their return to fresh sea water.

Narcotization was also achieved by placing the worms in dilute solutions of
ethyl alcohol and sea water. The alcohol concentrations ranged from 3.0 to 7.0%
by volume. In still other cases a type of narcotization was brought about by
placing the worms in various dilutions of sea water with distilled water. This
proved in many ways a very valuable procedure since the specimens would become
quite swollen, and thus not only sluggish or quiescent but also more transparent.
Thus internal vessels were more visible than in normal worms. Also vascular
behavior of the larger vessels was as little modified under these conditions as under
any employed.

In procedures where dissection was employed in order to expose internal
vessels or for other purposes, specimens lightly anesthetized by one of the fore-
going agents were used. The region or site of dissection was then painted with
1.0% procaine. This permitted rather extensive surgery on lightly anesthetized
worms without excessive muscular stimulation. The exposed vessels continued
to react for a limited time in what was probably a normal manner. Exposure
to sea water for long periods produced marked changes in the reactions of the
vessels. This was surprising but invariably true. Numerous attempts to use
• artificial sea water with modified salt values were not successful. For these reasons
observations following exposure were limited to an hour in most cases.

Vital staining was done with methylene blue added to sea water. The effect
was transient and the color faded, somewhat irregularly, within a few hours.
Fixing with ammonium molybdate was successful histologically but very toxic
to the tissues.

Attempts to cannulate individual vessels were not too successful. This was
surprising since no difficulty has been encountered in cannulating much smaller

72

PAUL A. NICOLL

vessels in higher forms. In the few times when cannulation was successful
attempts to inject formed material in order to visualize blood flow were dis-
appointing. India ink is quickly salted out and precipitated in the walls in the
immediate vicinity of its injection. Occasionally a few specks would remain in
solution and their progress along the vessels was very helpful in determining
flow patterns.

RESULTS
A. Vascular anatomy

Unlike some of the Annelida neither N. virens nor N. limb at a has developed a
specialized heart or hearts within their vascular system. Although Linville (1907)

N

.VIRCNS, ANTERIOR END DORSAL VIEW

1

,' s9 . C- .'
' ' /

i ..-.;

oo / <^ ; £ /

= jjf^^^

±^—±:—~.

'•"" *i

FIGURE 1. Principal vascular elements in anterior segments of Nereis virens. Dorsal view.
A, ventral longitudinal vessel ; B, dorsal longitudinal vessel ; E, recurrent branch — ventral
lateral vessel ; F, G-I plexus-lateral connective ; G, medial branch — ventral lateral vessel ;
H, stem-ventral-lateral vessel ; J, dorsal-lateral vessel.

calls the gastro-intestinal branch (vessel F) in each segment a heart, this is
hardly justified in the sense that it functions as a force pump. As explained
below these vessels appear rather as pace-makers of the segmental system, most
vessels of which are also contractile in their own right. Before considering the
flow patterns and vascular behavior it is necessary to understand the anatomical
relationships that actually exist in these forms.

In N. virens the typical reduplicated, segmental vascular pattern is extensively
modified in the anterior segments. The vascular organization in this region is an
adaptive development to fit the needs of the pharyngeal structures internally and
the reduced parapodia externally. No special function within the vascular system
itself can be ascribed to these anterior vessels. In a similar sense the segmental
pattern is modified in the posterior portion of TV. virens. Strictly speaking this
is limited to the terminal segment where the only definitive, simple and direct
connection between the dorsal and ventral vessels is to be found in the entire

BLOOD VESSELS OF NEREIS

73

animal. The zone immediately anterior to this terminal segment is the site of
formation of new segments and one can follow, more or less completely, the evolu-
tion of the definitive segmental pattern by careful comparison of the vascular struc-
tures in adjacent segments from this site forward.

Although a similar detailed study of anterior and posterior portions of N. limbata
has not yet been carried out, preliminary observations indicate this species also has
no specialized development within the vascular system at either extremity. Thus
vascular function in these species is basically concerned with segmental flow, which,
however, is modified by the contiguous union of adjacent segments.

Figures 1 and 2 represent semi-diagrammatic sketches of the principal vascular
structures in the anterior twelve segments of a typical specimen of N. virens.

'. i / .•''•

4i/ A -

TRAL VIEW

i 'VV

Sf l.j

• '- -r ,;'•*•' ,' •

V

I

.•/*"* • *'.' •', V

a- - .' - •' - •'

;.;•.•:•'/•

7 ' '•. """•*!' •.

FIGURE 2. Principal vascular elements in anterior segments of Nereis virens —
ventral view. Vessels marked as in Figure 1.

Figure 1 represents the structures as seen from a dorsal view with the dorsal
muscle mass cut away. Figure 2 is a similar view as observed from the ventral
surface.

As shown in Figure 1 the homologues of vessel J in the six anterior segments
are either lost or must be accounted for in the terminal arborization of vessel B.
Only two branches of this arborization are prominent. One appears as a recurrent
vessel of large size that terminates in a prominent capillary net covering the
posterior and lateral surface of the pharyngeal muscle mass. This striking capillary
plexus also anastomoses with the third segmental homologue of vessel H and
finally forms numerous connections with the upper border of the esophageal
capillary plexus. The other prominent branch of the terminal arborization forms
the principal anterior connection with another extensive capillary net that hangs
free between the anterior-lateral pharyngeal mass and the body wall. This second
capillary net makes numerous small connections with body vessels and terminates

74

PAUL A. XI COLL

posteriorly in the arborization of the fourth segmental homologue of vessel H.
Another branch of the arborization, while not very large, is of special interest in
that it supplies a rich capillary net that closely surrounds the dorsal ganglia and
circumpharyngeal connectives of the nervous system. All A'essels are present
bilaterally except for A, B, C and D.

As shown in Figure 2, the branches of the ventral vessel in the anterior seg-
ments of N. vircns are all retained and the homologous branches of the typical seg-
mental vessel H may all be identified except for vessel F in the first four segments.

N.VIRENS PARAPODIA
POSTERIOR VIEW

FIGURE 3. Vascular patterns in a typical adult segment of Nereis z'ircns. Right parapodium
viewed from posterior surface. Semi-diagrammatic. A, ventral longitudinal vessel ; B, dorsal
longitudinal vessel ; C, G-I plexus to ventral vessel connectives ; D, G-I plexus to dorsal vessel
connectives ; E, recurrent branch — ventral lateral vessel ; F, G-I plexus-lateral connective ; G,
medial branch — -ventral lateral vessel ; H, stem ventral-lateral vessel ; I, blind ending "capillaries" ;
J, dorsal-lateral vessel.

The homologues of vessel F appear as extended recurrent structures that pass
backward to make a characteristic anastomotic contact with the anterior margin
of the gastro-intestinal capillary plexus. The connections of vessel H in segments
3 and 4 with the pharyngeal capillary plexuses have been mentioned above. The
remaining branches and terminal arborization of vessel A in the pharyngeal and
adjacent tissue are easily observed. It is important to note that no direct anterior
connections between vessels A and B occur.

Since the vascular organization throughout the entire worm is basically a
series of joined segmental patterns, a typical segment from each species has been
diagrammatically sketched in considerable detail. Despite the variations in seg-

BLOOD VESSELS OF NEREIS

75

mental pattern that are present in segments at certain locations along the animal,
the important relationships from a functional standpoint remain identical in all
mature segments. Figures 3 and 4 show the segmental vascular distribution in
N. vircns and N. limbata, respectively. Each sketch represents the vessels in the
right half of a segment as viewed from the posterior surface of the parapodium.
The body wall is cut away to expose the internal vessels and their connections with
the various lateral branches.

A previously unrecognized phenomenon in the type and distribution of capillary
vessels found in these species and their relationship with the rest of the segmental
vessels underlies the principal differences between the segmental patterns of
N. virens and N. limbata. The terminal arborizations or "capillary" vessels of

N.LlMBATAPAftAPODlA FftOM

ANTEFUOft TH\RD

FIGURE 4. Vascular pattern in a typical adult segment of Nereis limbata. Right para-
podium viewed from the posterior surface. Semi-diagrammatic. Vessels marked as in Figure 3,
except for K = direct deep lateral branch of ventral lateral vessel.

the vascular system in both forms are divided into two distinct types. The first
type clearly satisfies the classical concept of capillaries as being simple endothelial
tubes that are the terminal anastomotic arborizations of larger vessels. Blood
passes through them, on the average, in a given direction and some sort of exchange
occurs here between the blood and surrounding environment. In both species the
vascular plexus of the stomach-intestine and most of the superficial sub-epidermal
plexus of the body wall and parapodia are examples that satisfy this concept of
capillaries. The former, no doubt, serves as the site of absorption of the digestive
products while the latter is the site for gaseous exchange between blood and en-
vironment. These capillaries are non-contractile and except for differences in
their extent and distribution are essentially the same in both forms.

The second type terminal vessel found in both species is a highly contractile
blind-ending structure which supplies the internal structures and tissues of the

76 PAUL A. NICOLL

body. They are apparently quite similar to the blind-ending vessels described for
Sabella and Spirographis by Ewer (1941), who also cites several reports of such
vessels in other species. Their origin, ramification and distribution in each
species show marked differences. They have been termed vessel I in both Figures
3 and 4 where their distribution in the body wall and coelomic space is diagram-
matically indicated by solid lines. The outstanding difference between N. virens
and N. limbata in regard to these type II capillaries concerns their origin and
interrelationships with the other vessels. In N. virens they arise at numerous sites
throughout the segmental vascular system and display considerable differences in
their internal ramification and complexity of branching between individual vessels.
In N. limbata, on the other hand, all type II capillaries apparently are the arboriza-
tions of branch K of vessel H which is not present in N. virens. Thus in N. virens
both types of capillaries must be served by the same major vascular supply, vessels
G and E, while in N. limbata almost complete separation between the supplying
trunks of the two capillary types has been achieved. The functional significance of
this fact will be discussed below.

With the exception of branch K, the segmental vessels of the two species are
essentially the same. In both forms blood is supplied to all peripheral structures
through either the ventral-lateral vessel H and its branches and/or through the
gastro-intestinal vessel F and its lateral extension, vessel G. Drainage essentially
occurs through the dorsal-lateral vessel J into the dorsal-longitudinal vessel B.
In both species direct connections also occur between the dorsal vessel B ; the
ventral vessel A and the gastro-intestinal capillary plexus through the short,
variable numbered, vessels D and C, respectively. In addition some small con-
nections between B and A with the immediate musculature are present. Occa-
sionally, as is shown in Figure 4, such a vessel may arise from vessel H. No doubt
other connections between these major vessels and the various tissues and organs
will be found with further study but the major, and functionally important, branches
and their various connections are adequately diagrammed in these Figures 3 and 4.

B. Flow patterns and vascular behavior

Although the final details of flow through the vascular systems of these species
may be lacking, the fundamental character and pattern can now be described.
The total circulation of a normal intact adult specimen represents a partial
interdigitation of two flow patterns. One is the primitive segmental system upon
which is superimposed a dorsal-anterior to ventral-posterior sluggish flow involving
the entire worm. This anterior-posterior circulation no doubt represents an
adaptation to supply the modified cephalad segments of the individual where the
primitive segmental systems have been essentially lost.

The actual volume displacement, or even the direction of flow, through any
particular portion of the vascular system is dependent on the nature of the worms'
activities and may change considerably from time to time. Furthermore, flow
evaluation is made rather difficult by the lack of any reference point such as the
heart serves for most other forms. In these species the motivating force for blood
flow is supplied by the local contractions of muscle cells that surround the vessels.
This activity spreads along each vessel in a peristaltic wave form, and volume
displacement from one portion to another of the vascular system depends in part

BLOOD VESSELS OF NEREIS 77

on the relative timing of the contractile waves. They frequently conflict with
each other and flow may stop or he reversed within a given vessel when this occurs.

The anterior-posterior flow in both forms is essentially the same. The
peristaltic waves in the dorsal longitudinal vessel originate in the immature
segments just anterior to the terminal anal segmental ring and sweep cephalad.
Usually three distinct waves may be observed at one time in a large intact adult
N. virens. Not infrequently the pace-maker originates two distinct waves which
fuse in the posterior segments of the specimen and continue on cephalad as a
single peristaltic contraction. Injury or marked irritation at any point along the
vessel may set up a new pace-maker at this point. The contractile waves then
originate at the new site and sweep in both directions over the dorsal vessel. If the
injury is neither too severe nor too far forward this disturbance seems to produce
little effect on the worms' behavior. The volume of blood displaced by these
peristaltic waves is much less than would appear to be the case from simple observa-
tion. Frequently the vascular constriction is far from complete and much of the
blood within the vessel is not affected by the peristaltic contraction wave.

Blood flows caudad through either the ventral vessel or the gastro-intestinal
capillary plexus. From approximately the tenth anterior segment, where a definite
segmental flow exists, a posteriorly directed flow in the ventral vessel is readily
demonstrated despite the absence of contractile waves. In front of this region,
however, the ventral vessel is probably not an important pathway for caudal flow
of blood reaching the region via the dorsal vessel. The anterior connections of
the dorsal vessel and the medial to lateral flow through vessels H suggest rather
that blood from this area returns caudad through the gastro-intestinal plexus.

The segmental circulation would logically appear to be the more primitive
circulation but may in its present form represent a more or less degenerate flow
pattern. The general direction of blood movement is ventral-lateral to dorsal
and medial in both halves of each segment, the return occuring from the dorsal
vessel through its connectives D into the gastro-intestinal plexus. Some blood
must eventually reach the ventral vessel through vessels C but the greater portion
leaves the G-I capillaries by way of the lateral vessel F in a typical body segment.
That such a flow is possible can be readily demonstrated by placing two tight
ligatures around the entire worm and so isolating a few segments. Flow in the
various vessels can be observed to continue in the normal manner for a reasonable
length of time. Ultimately the various peristaltic waves lose their necessary time
relationships and .flow ceases. The fewer the number of segments included be-
tween the ligatures the sooner will the circulation become disorganized.

This failure of the circulation in isolated segments indicates the interdependence
that now exists between the segmental and longitudinal flow patterns. Interrup-
tion of flow in either the dorsal or ventral vessel alone produces little disturbance
in segmental flow. However, if both vessels of the longitudinal system are ligated
a few segments apart, the intervening segments are soon drained of their blood
despite the patency of the gastro-intestinal pathways and the continued normal
activity of the segmental vessels.

This was borne out in experiment on an adult N. virens where the dorsal vessel
alone was ligated about one-third of the way forward. Little disturbance in the
specimen's circulation resulted except for some over-distension of the dorsal
vessel posterior to the ligature which, however, was soon taken care of by the

78

PAUL A. NICOLL

development of antidromic waves in the posterior third of the worm. Two hours
later the ventral vessel was also ligated about 12 segments more anteriorly than
the site of ligation of the dorsal vessel. No immediate change resulted from this
second ligation but after several hours the segments lying between the two ligatures
were essentially bloodless and the peristaltic waves in the dorsal vessel, anterior to
the ligature of that vessel, were quite irregular showing both normal and anti-
dromic peristaltic waves.

Peristaltic contraction waves passing medial-laterally in vessel H and its
branches and vessel F-G, coupled with similar waves that pass lateral-medially in
vessel J, seem to provide the basic means for segmental flow. However an addi-
tional factor is also present which would seem to insure this ventral-lateral-dorsal-
medial flow pattern. It concerns the relative frequency of the peristaltic waves in the
various vessels. In both forms vessel F-G always shows a higher frequency of
the peristaltic contractile wave than occurs in vessel J or B, which is usually the
.same. Table I gives some typical values for the frequency ratios of the activity in
vessels F-G and J of N. virens. With the exception of the last value, which was
determined from a specimen in deep anesthesia, the average ratio is 1.64. One

TABLE I
Frequencies of peristaltic contractile waves in cycles / minute in vessels of N. virens

Specimen number

1

1

3

4

5

6

7

8

F-G

35

27

21

24

31

26

23

11.7

J

17

14

12

18

19

19

17

1.6

Ratio F-G/J

2.08

1.93

1.75

1.33

1.63

1.37

1.35

7.3

Average 1-7 = 1.64.

would expect, therefore, a higher mean pressure in vessel F-G than in vessel J.
This would tend, of course, to keep blood flowing in a ventral-dorsal direction
through the parapodial capillaries. Similar data are not yet available for N. limbata
but a few counts indicate the F-G/J frequency ratio in this species to be about
the same even though the absolute frequencies are much higher.

The activity in vessel H and its branches seems to contribute little towards
determination of flow by this means. The waves all pass medial-lateral-dorsal as
expected but their frequency shows considerable variation. In N. virens they
are usually about as frequent as the waves in F-G, while in N. limbata they are
much slower. It should be noted here that in both species the contractile waves
do not pass along the short connection between vessels F-G and H. Therefore
these systems do not interfere with each other's activity, and little blood is ever
seen to flow from either vessel into the other.

The much greater frequency of the peristaltic contractile waves seen in com-
parable vessels of N. limbata but not exhibited by vessel K, apparently is dependent
on the anatomical relationship of the two types of terminal arborizations (capillaries)
• exhibited by these species. The simplest explanation for the difference is that N.

BLOOD VESSELS OF NEREIS 79

limbata has, in the evolution of its vascular system, advanced to the point of nearly
complete separation of its mechanism for gaseous exchange between the blood and
environment and its mechanism for supplying the needs of the tissues and organs
of the body. Functionally this is certainly true despite the few open connections be-
tween vessels F-G and H or their branches. Blood is forced through the expansive
capillary field of the superior ligula at rapid flow rates with a resultant far greater
potential supply of oxygen being available to N. limbata than to N. vircns. This
may, in part at least, account for the marked differences in the capacities of the two
forms to develop and sustain rapid swimming movements.

In N. virens, on the other hand, no anatomical separation of the two types of
terminal arborization exists. Flow through the limited sub-epidermal capillary
plexus is continually being disrupted by the in-and-out wash of blood of the
numerous, highly contractile, blind-ending type II capillaries. Even the superficial
sub-epidermal capillaries themselves are frequently contractile in N. vircns while
their homologues in N. limbata do not exhibit contraction. Flow from vessel F-G
through these terminal vessels is therefore slow and often scanty, with minimal
forward volume displacement. Thus the potentially available blood-borne oxygen
supply to the tissues of N. vircns must be much less than is the case with N. limbata;
certainly N. vircns is quite incapable of rapid or extended swimming movements
and frequently resorts to a slow undulatory type behavior that has been ascribed
to respiratory activity.

The activity of the type II, blind-ending terminal vessels that supply the
internal tissues and organs of the body, while unique in vascular systems as well
organized as these species exhibit, probably represents a vestigial mechanism for
movement of blood in primitive systems where true circulation has not developed.
The underlying stimulus of their contraction is, without doubt, the stretch or
distension produced by the blood forced into each branch by the peristaltic wave
in the supplying vessel. If it were not for this reaction flow would soon cease
in such systems.

In N. virens the numerous branches of these type II capillaries show considerable
differences in the extent of their ramification after they branch off from the thorough-
fare channels. However each subdivision of the branch finally ends in a finger-like
tube that at rest is bent more or less back upon itself. When the incoming
peristaltic wave reaches that structure it is extended both outward and in diameter.
The recoil is sudden and precise, resulting in greater shortening and reduction of
lumen size, than the resting state exhibits.

Several attempts were made to stain with methylene blue the more superficial
of the type II capillaries in isolated parapodia of N. virens. If a ligature was
tightened around the parapodium just at the time a parapodial flush was developing
and the parapodium then cut away from its segment, the vessels were isolated
while more or less filled with blood. The parapodia could be mounted on a slide
and examined at high powers with a compound microscope and transmitted light.
In most cases such a preparation continued to exhibit marked activity in the type
II capillaries for several hours. After study of the characteristic behavior of the
type II capillaries the entire parapodium was covered with methylene blue-sea
water solutions for varying lengths of time. After the dye solution was washed
away, the type II capillaries were observed for possible staining of cells or other
contractile elements.

80 PAUL A. XICOLL

In no case were any contractile elements stained that appeared similar to the
smooth muscle cells along the larger vessels. Small signet-shaped structures
partially surrounding the vessel along the basic branches were observed. These
may have been the nuclei of small but typical smooth muscle cells which were
wrapped around the vessels, similar to those seen in arterioles of mammalian
forms. Frequently there appeared to be a thickened granular region, not clearly
marked off from the wall of the blind fingers, that lay parallel with the finger sacks
and seemed to be on their outer surface. These may be contractile elements since
frequently protoplasmic filaments of smooth muscle cells will stain in this manner
with methylene blue.

DISCUSSION

Although this paper is concerned mainly with the anatomy and behavior of
the blood vessels, together with the resultant flow patterns, some comments on
contractility per se will not be out of place. This quality of a vessel can be con-
sidered from three points of view. First, the structural elements whose activity
is responsible for contractility ; second, the immediate stimulus and control
mechanisms that determine the quantitative aspects of contractility ; and third,
the value or purpose of contractility for the organism.

Federighi (1928) comes to the rather astonishing conclusion that the typical
peristaltic contractile waves represent the activity of the endothelial tubes and not
the muscle cells in the walls of the vessel. This is certainly not in agreement with
the majority of work on the minute vessels of higher forms and is based on very
sketchy evidence where the vessels were observed at low magnification.

Several incidental observations made during this study fail to confirm Federighi's
conclusions but point instead to the circular, slightly branched muscle cells as
the structural elements responsible for contractility. Thus contractility was never
observed along any vessel where staining with methylene blue failed to show
either typical muscle cells or granular thickenings outside the endothelial tubes.
Also, direct study under high magnification always showed actual contraction of
these muscle cells to be directly correlated with the peristaltic contractile waves.
Finally the failure of the contractile waves to spread over certain branches or
short connectives free of muscle cells, such as the connection between vessels F-G
and H, but rather to spread only along the vessels with the muscular elements in
their walls argues for these muscle cells, rather than the endothelial tubes, as the
structures involved.

The stimulus for contractility, here as in other similar cases, may be divided
into two categories. One is the initial origin of the activity while the second is
its propagation. No attempt is made here to deal with the first condition. It is
sufficient to say that there are pace-maker foci at various locations along the
vascular system, and that these show the same type of behavior, and are affected
by the same type of conditions, as act on muscular pace-makers in other contractile
systems.

The spread of the contractile wave probably could be separated into an
excitatory wave and a contractile wave but this has not yet been attempted. There
is ample overlap of adjacent circular fibers to account for direct muscular transmis-
sion. In those instances where such transmission would be impossible for
anatomical or other reasons, the contractile wave could, by forcing fluid ahead of

BLOOD VESSELS OF NEREIS 81

itself, produce distension in the vessel beyond, and so lead to excitation at the
new site. Here the action would be similar to the type III deglutition waves
described for the posterior portion of the mammalian esophagus. Three examples
of this type of behavior in N. virens can be cited. One is the antidromic wave ob-
served to pass over vessel D from the gastro-intestinal margin to the dorsal vessel
immediately following the sudden distension of vessel D by the passage of a
peristaltic contraction wave along the dorsal vessel. Another example is the occa-
sional anti-dromic wave that may be observed to pass over vessel F-G when the
higher frequency of the peristaltic waves in this vessel leads to the condition that
a normal wave reaches the parapodia over vessel F-G and finds all of the small
vessels contracted. The most outstanding case, however, is the sudden and force-
ful contraction of the type II capillaries immediately after each incoming peristaltic
rush causes their sudden distension. These are, no doubt, all examples of the
behavior for simple tubular hearts discussed by Haywood and Moon (1950).

As to the ultimate value or purpose of such contractility in blood vessels no
conclusions can be made at this time. Two possibilities, not mutually exclusive,
suggest themselves. One is the simple development of a propulsive force that
produces movement of the fluid along the vascular systems. There is no doubt
that such a purpose is in fact fulfilled by the activity in these primitive vascular
systems.

A second possible function of such contractile behavior would be the production
of pressure waves within the fluid contained in these vessels. This might play
a significant role in the exchange between the blood and tissue fluid across the
capillary beds. Nicoll and Webb (1947) have suggested such a role for the
active vasomotion of the spiral smooth muscle cells found along the arterioles
of mammals.

SUMMARY AND CONCLUSIONS

1. The anatomical organization of the vascular system of N. virens has been
studied in considerable detail. Certain insufficiencies and errors in past studies
have been supplied or corrected. The detailed anatomy of N. liinbata for a typical
segment has been worked out and compared to that of N. virens.

2. The most striking observation was the discovery of many minute blind-
ending capillary-like vessels. These differ in the two species in their mode of
origin from the vascular system which permits, in the case of N. linibata, a partial
separation of a lesser or respiratory circulation from the general or systemic
circulation.

3. The flow patterns in both forms are shown to consist of an interdigitation
of two circulatory systems, one a primitive segmental type and the second a
posterior-anterior flow involving the entire worm. Neither system is independent
of the other, with the result that actual flow is rather sluggish and inefficient through-
out the entire worm. In N. liinbata the separation of the type II capillaries from
the superficial respiratory (lesser) circulation has partially overcome this deficiency.

4. No true hearts are present in the systems of either species, but rather the
majority of the vessels show contractility. Certain foci appear to function as
pace-makers for the contractile waves, and flow is determined in part by the relative
frequencies of the peristaltic waves in the different vessels.

82 PAUL A. NICOLL

5. Some discussion is given as to the nature of contractility per se. It is con-
cluded that the structural elements involved in all vascular contractility are the
slightly branched smooth muscle cells, or a primitive type granular area, of the

vessel wall.

6. The primary stimulus for the contractile waves is assigned to the inherent
activity of pace-maker foci, while their propagation along the vessels is thought to
spread by way of the circular muscle fibers. Sudden distension of a vessel, with
the resultant stretch of these muscle cells, is shown to be capable of exciting the
vessel at that site.

7. Two possible functions are suggested for the contractile waves in these
vessels: one, the propagation of blood within the system, and second, a pressure
wave that aids in exchange between the blood and interstitial fluid.

LITERATURE CITED

CARLSON, A. J., 1908. Comparative physiology of the invertebrate heart. X. A note on the

physiology of the pulsating blood vessels in the worms. Amcr. J. Physiol., 22 : 353-356.
CLARK, ELIOT R., AND ELEANOR L. CLARK, 1940. Microscopic observations on the extra

endothelial cells of living mammalian blood vessels. Amcr. J. Anat., 66: 1^19.
CLARK, ELIOT R., AND ELEANOR L. CLARK, 1947. Arterial anastomoses, some factors concerned

in their formation. Anat. Rcc., 97 : Abstr. Am. Assoc. Anat. 10.
EWER, D. W., 1941. The blood systems of Sabella and Spirographis. Quart. J. Mic. Set., 82:

587-620.

FEDERIGHI, HENRY, 1928. The blood vessels of annelids. /. Exp. Zool., 50 : 257-294.
HAYWOOD, C. A., AND H. P. MOON, 1950. The mechanics of the blood vascular system of

Ascidiella aspcrsa. J. E.rp. Bio!., 27 : 14-28.
KROGH, A., 1922. The anatomy and physiology of capillaries. Yale University Press. New

Haven, Conn.

LINVILLE, H. R., 1907. The circulatory system in Nereis. Science, 25 : 727-728.
LUTZ, BRENTON R., GEORGE P. FULTON AND ROBERT P. AKERS, 1950. The neuromotor

mechanism of the small blood vessels in membranes of the frog (Rana pipicns) and

the hamster (Mesocricetus auratus) with reference to the normal and pathological

conditions of blood flow. E.vp. Mcd. Sunjcry. 8 : 258-287.
NICOLL, PAUL A., AND RICHARD L. WEBB, 1946. Blood circulation in the subcutaneous tissue

of the living bat's wing. Ann. N. Y. Acad. Sci., 46, Art. 8 : 697-711.
NICOLL, PAUL A., AND R. L. WEBB, 1947. Studies on the behavior of the vascular network in an

intact mammal. Memorandum Report No. TSEAA-696-112. Army Air Forces

HDQS Air Material Command Engineering Division.
PARKER, G. H., 1923. Are there Rouget cells on the blood vessels of invertebrates? Anat.

Rec., 26 : 303-305.
REDFIELD, ALFRED C., 1933. The evolution of the respiratory function of the blood. Quart.

Rev. Biol.,8: 31-57.
ROMIEU, M., 1923. Recherches histophysiologiques sur le sang et sur le corps cardioque des

annelides polychetis. Arch. Morph. Gen. Exper., 17 : 339.
TURNBULL, FREDERICK M., 1876. On the anatomy and habits of Nereis vircns. Trans. Conn.

Acad. Arts Sci., 3 : 265-280.
WEBB, RICHARD L., AND PAUL A. NICOLL, 1944. Behavior of lymphatic vessels in the living

bat. Anat. Rec., 88: 351-367.

WEBB, RICHARD L., 1952. The lymphatic system. Ann. Rev. Physiol., 14: 351-367.
ZWEIFACH, B. W., AND C. E. KossMAN, 1937. Micromanipulation of small blood vessels in

the mouse. Amer. J. Physiol., 120: 23-35.

THE METAMORPHOSIS OF PARTIAL LARVAE OF PERONELLA
JAPONICA MORTENSEN,1 A SAND DOLLAR2

KAYO OKAZAKI AND KATSUMA DAN

Department of Biology, Tokyo Metropolitan University, Meguro-ku, Tokyo and
Misaki Marine Biological Station, Misaki, Kanagawa-ken, Japan

Following the discovery by Hans Driesch (1891), that in sea urchins an isolated
blastomere of the two-cell stage can develop into an harmonic larva, much work
has been done on this subject. However, since the majority of experiments in the
past have been concerned with the developmental capacity only as far as the pluteus
stage, emphasis in the present investigation was placed on the aspect of meta-
morphosing capacity. The inadequacy of past experiments is due to the difficulty
of rearing plutei to metamorphosis in the majority of sea urchin species, which,
fortunately, can be obviated in the sand dollar Peronella japonica Mortensen. In
this species metamorphosis is completed in the course of three or four days.

Peronella is found in great numbers on the sandy bottom of a shallow lagoon
in the neighborhood of the Misaki Marine Biological Station, and the breeding
season extends from the latter part of June to September.

The main developmental features of this animal have been worked out by Mor-
tensen (1921), but some new points were noticed by the authors. In the present
paper, the results obtained by operative experiments, and points regarding normal
development which have a direct concern with the description of the experiments
will be reported.

I. DEVELOPMENT OF THE WHOLE LARVA

Egg. The egg, about 300 ^ in diameter, is heavily laden with yolk and opaquely
pink. The color is imparted by a rosy pigment which is associated with doubly
refractive crystals.

Elevation and hardening of the fertilization membrane are very slow, so that
the membrane can be removed until 10 minutes after insemination. Since the
hyaline layer is extremely delicate and remains sticky, the denuded eggs adhere to
the bottom of the container, and after cleavage, the blastomeres arrange themselves
in a plane, suggesting sea urchin blastomeres in calcium-low sea water. Later, how-
ever, the larva rounds up and gives rise to a typically spherical blastula.

Micromcrcs. Frequently, in this species, at the fourth cleavage all sixteen cells
are of equal size. In such cases, the micromeres are formed in the succeeding
cleavage. The micromeres, are, as a rule, relatively large in comparison with those
of other echinoderms. Toward the end of the breeding season, however, there is
an increasing tendency to produce smaller micromeres. In eggs which are con-
sidered to be more or less overripe, as early as the eight-cell stage, the vegetal four

1 Dr. F. Uchinomi called the authors' attention to the fact that Peronella lesueuri has
recently been changed to Peronella japonica Mortensen by Mortensen (1948), for which the
authors' thanks are due.

2 This research was supported in part by the Ministry of Education Research Expenditure.

83

84

KAYO OKAZAKI AND KATSUMA DAN

blastomeres are smaller than the animal blastomeres ; and in the sixteen-cell stage,
still smaller micromeres are formed.

Blastula. At about five hours after fertilization, one or two grooves are notice-
able in the blastula, coinciding with the cleavage-furrows of the two- or four-cell
stages. However, these grooves gradually flatten out as the blastocoel expands,
and the embryo again becomes spherical. This is somewhat similar to the condi-
tion which is found in the development of Astropectcn aranciacus (Horstadius,
1939a).

At about seven hours after fertilization, the blastula acquires cilia and begins to
rotate within the fertilization membrane. From this time on, primary mesenchyme
cells migrate inward and disperse in the blastocoel, so that the embryo takes on an
opaque appearance. At nine hours, the embryo breaks through the membrane ; it is
now elongated, with a truncated posterior end having an accumulation of primary
mesenchyme cells along the vegetal wall, so that the larva looks somewhat like the

FIGURE 1. Diagram of section of successive stages showing the position of amniotic cavity
and hydrocoel. A, B, longitudinal sagittal section. A, late gastrula (18 hours) ; B, early
pluteus (26 hours), amniotic opening has shifted toward the dorsal side and blastopore has
closed. C, transverse section, early pluteus (26 hours), am, amniotic cavity, hy, hydrocoel.
c, coelom. g, gut. bl, blastopore. d, dorsal side, v, ventral side.

gastrula of regular sea urchins. The apical tuft does not usually differentiate, and
even when it can be recognized, the cilia are fewer and less conspicuous than in other
forms.

Gastrula with the amniotic imagination. Gastrular invagination begins at about
the twelfth hour. Within a few hours, another ingrowth of the ectoderm appears
in the center of the flattened oral field (Fig. 7 D, W) and develops into a stomo-
daeum-like invagination, but that it is not a true stomodaeum is shown by its sub-
sequent failure to unite with the archenteron. Instead, its sac-shaped prolongation
extends along the dorsal side of the entoderm and forms a cavity (Fig. 1, A).
Mortensen correctly identified this as an amniotic cavity. No mouth opens, and the
blastopore closes sooner or later, so that the archenteron remains as a free blind sac
within the body, without forming any functional digestive tract. This obviates the
necessity for feeding the larvae and greatly facilitates their culture.

Pluteus. At about twenty hours after fertilization, the fully formed pluteus
typically has only two post-oral arms, equal in length to the rest of the body. The
number of arms may vary, however, from none to four, without apparently causing
any essential difference in the later development, since all such larvae are able to

METAMORPHOSIS OF PARTIAL LARVAE

85

complete metamorphosis. At this stage, the hydrocoel begins to differentiate from
the coelomic sacs, which lie close to the ventral side of the body. The hydrocoel is
usually derived from the left, but occasionally from the right, coelom ; in either case,
it is formed in a nearly median position ( Fig. 1, B, C). This location of the hydro-
coel, together with the unusual median position of the amniotic imagination, forms a

FIGURE 2. Control whole imago of Peronella japonica Mortensen, 4 days after metamorpho-
sis. Aboral view. X 160.

striking contrast to the location of the corresponding structures in other echinoderms.
After the enlargement of the hydrocoel, five lobes are pushed out, arranged in a
bilaterally symmetrical fashion with regard to the median plane of the pluteus. At
about fifty hours after fertilization, the amniotic cavity constitutes a large part of
the pluteus and its inner wall is covered with well-developed adult spines. The
pluteus sinks to the bottom at this stage.

86 KAYO OKAZAKI AND KATSUMA DAN

Pigment. Simultaneously with the formation of adult spines and plates, clype-
astroid pigment (green) is deposited in the echinus rudiment and is gradually con-
centrated as development advances. The pluteus, as a whole, is greenish at this
stage, but if it dies, a splendid green color pervades the entire body. Although this
pigment is green in an alkaline fluid, it gradually loses its color completely as the
medium is acidified. The green pigment is also found in the mesenchyme cells of
the late pluteus of Astriclypeus nianni and Clypeaster japonicus, especially clearly
in the echinus rudiment. If the test of an adult Peronella, Astriclypeus or Clypeaster
is injured, the same green color appears on the test integument.

Metamorphosis. Among relatively fast-growing larvae of Peronella, meta-
morphosis sets in at about sixty hours after fertilization with the protrusion from the
amniotic cavity of rudimentary spines and tube-feet. Since the dorsal wall of the
body and the amnion are very thin at this time, they are occasionally broken through
by the protruding spines. On the contrary, slowly growing larvae, in spite of the
fact that they have the typical pluteus form, may fail to metamorphose, or finally
succeed after a delay of a week or more. When metamorphosis fails completely in
such retarded larvae, they eventually become edematous. In such a case, the al-
ready differentiated spines degenerate and disperse within the body as fine spicules.
It is interesting that in such degenerating larvae, the above-mentioned clypeastroid
pigment cannot be recognized at all.

Usually, metamorphosis is completed within seventy or eighty hours. Con-
cerning the external features of the adult, Onoda's report is available (1937). The
mouth opening, masticatory apparatus, typical spines and six-rayed spines and tube-
feet can be recognized a few days after metamorphosis (Fig. 2). The young sand
dollars survive for about ten days without being fed, and can even increase the
number of spines and tube-feet.

The most unusual feature in the development of this animal is certainly the rapid
rate of development as compared with other related forms. It can be said that the
pluteus stage is only a phantom, so to speak, and the larva is preparing for meta-
morphosis from the start.

II. DEVELOPMENT OF PARTIAL LARVAE

The fertilization membranes were removed immediately after fertilization by
squirting the eggs through a slender pipette, and the blastomeres were separated in
sea water by a fine glass needle. In this form the hyaline layer is so delicate that
it can easily be cut, even in a calcium-containing medium. Pairs of half-larvae
or quartettes of quarter-larvae were kept in separate glass containers in 5 cc. of
sea water. The plane of section at each stage is shown diagrammatically in Figure
3, and the results of the experiments are summarized in Table I and indicated in
Figure 4.

(1 ) Isolated blastomeres of the two-cell stage ; equi- and toti- potent regarding meta-
morphosing capacity.

Blastomeres isolated in this stage show partial cleavage as regards the number of
meso-, macro- and micromeres, and form more or less open half-blastulae, which,
however, soon round up and close, and gastrular invagination ocurs at the vegetal

METAMORPHOSIS OF PARTIAL LARVAE

87

(D

(3)

(4)

(4)

(6)

(5)

b

(6)

(6)

c

FIGURE 3. Diagram indicating plane of section at each stage. Dotted line, animal elements ;
fine continuous line, vegetal elements. Thick straight line, plane of sections ; broken line, original
egg-axis. Stages of operation numbered as follows: (1), 2-cell. (3), 16-cell. (4)a, grooved
blastula. (4)b, blastula with primary mesenchyme cells. (5) a, gastrula with archenteron and
two small triradiate spicules. (5)b, late gastrula with amniotic invagination. (6) a, early
pluteus with two fully developed arms but without adult skeleton. (6)b, pluteus with some adult
skeleton. (6)c, late pleuteus with well developed adult skeleton.

100

50

11

(3)

(4b>

(5)

(6)
c

FIGURE 4. Percentage of metamorphosis of separated halves olPeroneUa japonica Mortensen,
Dotted line, meridional halves. Thick continuous line, meridional pairs. Thick broken line,
vegetal (posterior) halves. Fine broken line, animal (anterior) halves. Fine continuous line,
equatorial pairs.

88

KAYO OKAZAKI AND KATSUMA DAN

^

s
1

v OJ

2 ^

§ "3

^^ *«^ ^
^>

W K -™

« i I

«.< O O

I

71 "^ —

25 §"5

&5

^\

^^

0

o

—*• ~ "* >-^

ON

00

* ^ M-O li

1

•y 2^*5

O

O

NO

^4.

NO

-• "I'S

*"

es

^^

rC

t u

6?

fe?

0 ^ 1 *• §

If)

o

^.

c S*^.i

>o

'O

— • 5— a f

I

3'> M

10

t^

00

IO

^H

E ""

"^

cs

^H

f.

09

«.2'js!|2

10

=

-t

"•=5 si

1

00

t*:

^

ON

w "

*H

es

f™^

^^

_rt C

^

^

.Q x'S 0 ^

^

o

f5

j-- si •- c

^H

c*l

-— - M Si S 'M

«J Cfl C M

1

'O

<o

^H

OO

O

J ' •=

_ra

^

65

ra "? M

to

0

^

10 "v? 2

^^ r3 x

1

O

-

r^i

0

10

•o

,iiL

w

g

OO

•? Vj u e ^

^H

•~-J% M fc o

1

CO 2 £

t-.

t*3

ro

C*3

ON

"E

IO

^H

^

"O 03

85

J£

^ ¥*g 4)

f—

<"O

-t

•^ o *^ ^

1—1

PO

2|^

1

OJ=

>o

cs

^

-

S

o

g

OO

«5 £ ^

'-'

cs

"" 2^

Hid

1/5

NO

PO

10

K

r^i

-t

"

es

= ll

8

s

H "'

-

•0

?!

-t

c

cs

Ts

a

i_

S

'o

1
w

after fertilization

er of operated larvae

er and percentage of
amorphosed pairs

er of metamorphosed
le halves

er and percentage of
amorphosed meridion
res*

'•SI

o

J2 *3

,-G b/D

o *-* ^

1«
O

S

11

S -|

S | _g

DC

£

Z

z

Z

METAMORPHOSIS OF PARTIAL LARVAE

89

U)

<U

o

I 8

.s *o

"S 03

^3 s

(U

I— I ^_)

c

o3

a

to

01

rt
3
cr

iflj

o

o

g

*C " U ^ W

oo

o>

Si-si's

o

VO

CN

00

01

00

*•' > nj

*-H

^— H

o

2s c

65

65

65

^ trj Q j^ O

CN

VO

CO

' > Cfi X 13 Qy

10

to

00

11 "

CN

to

-H

2

Ov

-t

rsi

Hi

Ipi

tO

s

<*

r--

_

f

— c

65

65

65

J2 '•/)'> "X ^3

c

00

T-H

-— - rt •= c

!2. M Si, c -s.

*— ' M 1- M
o nj t rt

^

o

LO

JO

03

65

65

65

nJ = &

o

0

T^H

X--N i- oa

^—1

~ Sra

O

^

o

O

o

IO

10

5I

65

65

65

X! 3 •£ ^' V)

o

Tt*

0

^? "MO ^ T"

^f

^"JS w C °

^ 4-» OJ

CN

o

*H

1—1

2

0

"w nj

4J •— •

T* 2 vj rt
' 2^2^

O X5

"flj

65

65

65

_ Vi,

o

0

^ o rt

^— ' OJ *J

•*-> 75

—

o

o

o

^+<

>*

iZ

5

- ll

H ^

4-1

CD

c?

'3

1
w

umber of operated larvae

umber and percentage of
metamorphosed pairs

umber of metamorphosed
animal (anterior) single
halves

umber and percentage of
metamorphosed animal
(anterior) halvesf

umber of metamorphosed
vegetal (posterior)
single halves

umber and percentage of
metamorphosed vegetal
(posterior) halves!

Z

Z

z

Z

Z

Z

j= i: ~

' - >•«

to -;— *

03
J=

2 -2

to

01 -

> E o
"03 '£ oJ

-C 03 >

bo bO b«

C C C

c/5 c/5 c/:

* -I— -H-

90

KAYO OKAZAKI AND KATSUMA DAN

side. Such operation at the two-cell stage does not destroy the basic arm-forming
capacity of the larvae, since the resultant plutei usually have two arms, and very
rarely three- or four-armed half-plutei are encountered. However, the general
tendency is toward a reduction in the number of arms as the result of the operation ;

B88§988 8888 86998889

FIGURE 5. Pair of half-imagos derived from single blastomeres of the two-cell stage, 4 days
after metamorphosis. A, aboral view. B, oral view. They do not show any difference from
whole images except for their small size. X 160.

METAMORPHOSIS OF PARTIAL LARVAE 91

i.e., the percentage of one-armed plutei is greater among the half-larvae than among
the unoperated controls. Even in this case, however, it does not mean that each of
the pairs has a right and left arm, respectively. At any rate, development of the
arms, as in the case of the whole larva, shows no connection with the capacity of the
half-larva for metamorphosis.

The young half sand dollar does not show any difference from the whole sand
dollar except that it is dwarf, the two members of the pair being completely equal
in all points (Figs. 2, 5). There is not even much delay in development as a
result of the operation.

(2) Isolated blastomeres of the four-cell stage : equi- and toti-potcnt regarding meta-
morphosing capacity.

The results of the operation are as follows :

Operated 4-cell stages 207

All members of the quartettes metamorphosed 2 sets

Three members of the quartettes metamorphosed 6 sets

Two members of the quartettes metamorphosed 24 sets

One member of the quartettes metamorphosed 65 singles

In contrast to the relative frequency with which both members of a pair of half-
larvae completed metamorphosis, all four quarter-larvae derived from isolation in
the four-cell stage were rarely able to metamorphose. The probable reasons for
this are presented below.

The development of quarter-larvae is much like that of the half-larvae described
above, except for their smaller size. However, quarter-plutei usually have only one
arm or none at all. The number of arms has, again, no essential meaning for meta-
morphosis.

The physiological condition of the quarter-images seems to be inferior to that
of the half-imagos, since the metamorphosis of the former is delayed as compared
with that of the latter, and many of the former die soon after metamorphosis. Even
when they survive for a while, they do not usually show any sign of growth and
eventually become edematous. But a point to be stressed is that even in such
quarter-imagos, no part of the body is missing.

There were, altogether, 8 instances in which three or four members of a quar-
tette metamorphosed. In these cases, not only was the general developmental con-
dition very poor, but the respective rates of growth and degrees of differentiation
were extremely variable (Fig. 6). On the contrary, there were 89 cases in which
one or two members of a quartette succeeded in metamorphosing. These images
appeared to be much better developed than those of the 8 cases in which three or
four survived, occasionally even reaching a state comparable to that of half-imagos.
This may probably mean that % of the protoplasm of a single egg is about the mini-
mum of material sufficient to permit metamorphosis ; and further, that when blasto-
meres of the 4-cell stage do not share the egg protoplasm strictly equally, the smaller
ones fall below the viable level. Harvey (1940) reported the similar fact that
although isolated quartettes from a single egg of Arbacia punctulata may all develop
into perfect dwarf plutei, there is considerable variation in size among them.

92

KAYO OKAZAKI AND KATSUMA DAM

(3) Half-larvae of the 16-cell stage: Meridional halves: equi- and toti-potcnt with
respect to metamorphosis.

Of 45 pairs of half-larvae obtained by meridional section of the 16-cell stage,
6 pairs and 13 singles completed metamorphosis. Early half-imagos resulting from
this operation are indistinguishable from those derived from the 2-cell stage.

Equatorial halves: animal half fails to metamorphose; vegetal half slwivs zvcak
capacity for metamorphosis, always followed by immediate death.

FIGURE 6. Quartettes derived from single blastomeres of the four-cell stage, 4 days after
metamorphosis. A, C, oral view. B, D, aboral view. Their developmental condition is very
poor as compared with half-imagos, and further, there is a lack of uniformity in their respective
rates of growth. X 160.

METAMORPHOSIS OF PARTIAL LARVAE 93

It is rather difficult to find differences in the mode of development of the animal
and vegetal halves before the appearance of the primary mesenchyme cells, al-
though the vegetal half tends to round up somewhat earlier than the animal half.
Simultaneously with the migration of the primary mesenchyme cells, however, a
distinction between the two halves suddenly appears. The vegetal half becomes
opaque because of the presence of the mesenchyme cells in the blastocoel, while the
animal half remains transparent (Fig. 7, A). Another striking difference between
the two regions is that on dying, the animal half becomes a brilliant pink, while the
vegetal half is turned green by the clypeastroid pigment. A detailed account of
their respective courses of development follows.

Animal half. The blastula of the animal half gradually becomes flattened—
usually in the dorso-ventral direction.3 At about 12 hours after fertilization — in
the control whole-larvae gastrular invagination begins at this time — a depression
appears in the ventral field (Fig. 7, C, An). This is exactly the position at which
amniotic invagination occurs in control whole-larvae a few hours later.4 At this
time, the animal half-larva has remarkably long cilia on almost the whole or a
part of the surface, the wall of which is more or less thickened. A few hours later,
larvae are frequently found with two, and occasionally with three depressions at
once (Fig. 7, D, An II). Some larvae have separate cells in the blastocoel, but
probably these are not mesenchyme cells, since they neither form spicules nor con-
tain clypeastroid pigment. In other words, they do not show any differentiation.
The situation is the same for amphibians (Ruud, 1925 ; Vintemberger, 1934, 1935).

Several hours later, the wall of the animal region rapidly increases in area and
begins to wrinkle (Fig. 7, E, An). The wrinkled area spreads from the animal
toward the vegetal side, so that these blastulae finally have the appearance of a mass
of many small ciliated vesicles (Fig. 7, E, An), closely resembling the isolated and
cultured Triton epidermis as described by Holtfreter (1933). Occasionally, sepa-
rate small vesicles fall off from the main mass, and some of these swim around as
small blastulae without showing any further development even after several days.
As a rule, animal half-larvae have a tinge of pink, in contradistinction to the green-
ness of the mesenchyme cells, but upon death the pink color increases its brilliance
and often pervades the entire body. Although a pink pigment exists originally
in the unfertilized egg of Peronella, and, moreover, the epidermis of normal larvae
has a slight tinge of the same color, their tones are not so deep as that of the dead
animal half. Occasionally, such a pink color is also recognized in a part of a whole
embryo which has died at the morula or the early blastula stage.

3 The animal pole region of the gastrulae is indicated by a tuft of long cilia. On animal
half-larvae after flattening, these long cilia are usually found at one pole of the long axis. Oc-
casionally, however, cases are found in which one of the flattened sides bears somewhat longer
cilia than the other surfaces (Fig. 7 B, An I).

4 At the stage of formation of the amniotic invagination, not only the animal half- but also
the control whole-larva becomes flat in the dorso-ventral direction (Fig. 7 D). It may be possi-
ble that the animal half which is released from the effect of vegetal elements flattens, and amniotic
invagination occurs earlier than normally. Moreover, Horstadius showed (1935, 1939b) that
animal halves of Paracentrotus often develop into blastulae with stomodaea, and, as before men-
tioned, the amniotic invagination of Peronella bears a striking resemblance, morphologically, to
the stomodaeum of the sea urchin larva. However, it is very difficult to judge whether the
depression of the animal half-larva of Peronella corresponds to the amniotic invagination.

94

KAYO OKAZAKI AND KATSUMA DAN

Vegetal half. The swimming vegetal blastula takes on a very dark appearance
earlier than does the whole larva, since the blastular wall is extremely thick and the
small blastocoel is filled with mesenchyme cells (Fig. 7, A, Ve). In a short time,
the cells in the vegetal region begin to dissociate and fall out of the blastular wall
(Fig. 7, B, Ve). The dissociated cells adhere to each other, forming a green mass
on the outside of the vegetal wall (Fig. 7, C, Ve).

Some larvae soon cast off this mass of dissociated cells and develop into small
gastrulae or plutei (Fig. 7, E, Ve), although a few exogastrulae with long, pro-

Ve

FIGURE 7. Comparison of development of equatorial half-larvae of 16-cell stage and control
whole larvae of Peronella japonlca Mortensen. Numbers of the upper row indicate hours after
fertilization. W, control whole larvae. An, animal halves. Ve, vegetal halves. I and II, two
types of animal half-larvae at each stage.

truding archenterons are also formed. Of 431 vegetal halves, 71 developed into
plutei with or without arms. Four of these plutei succeeded in metamorphosing.
However, such young imagos had only two or three spines on an extremely edema-
tous body, and died immediately after metamorphosis.

On the other hand, when the mass of dissociated cells becomes larger than the
larval body, which is rather frequently the case, the larvae adhere to the substratum
with a part of the sticky cell mass and consequently are unable to swim in spite of
their ciliary activity. Even if they succeed in freeing themselves from the cell mass,
they can develop only into small blastulae filled with mesenchyme cells. This is no
doubt because such larvae have lost too many cells to continue further development.

METAMORPHOSIS OF PARTIAL LARVAE

In either case, dead vegetal larvae exhibit a green color, although whole larvae
of such a young stage never do so. It is quite an impressive contrast that the animal
half turns a brilliant pink on death, while the vegetal half assumes a green color as
it dies.

(4) Half-blastulae.

Larvae were bisected meridionally at the following two stages and equatorially
at the second stage (see Fig. 3 (4) a, b) :

a. Grooved blastula (5 hours after fertilization).

b. Late blastula with primary mesenchyme cells (7-9 hours after fertilization).

Meridional halves: equi- and toti-potent with respect to metamorphosis.

Grooved blastulae were bisected along the groove. The development proceeds
in much the same way as in isolated blastomeres of the 2-cell stage. Since the
grooves of' the blastula coincide with the cleavage furrows of the two- or four-cell
stage, the distribution of material in half-larvae of this kind is identical with that
in half-larvae of the two-cell stage. Consequently, such operative results as were
obtained are quite according to expectation. This experiment furthermore indi-
cates that the regulative capacity in the meridional half has not at all decreased by
the grooved blastula stage.

Halves separated at the late blastula stage go through metamorphosis only half
as frequently as the previous stage and the physiological condition of the half-imagos
is much poorer.

Equatorial halves: metamorphosing capacity of the animal halves negligible,
of vegetal halves strong.

Among the equatorial halves, there was no instance in which both members of
a pair succeeded in metamorphosing. Commonly, the animal half became a blastula
with mesenchyme cells, and soon died without further development, except in one
instance. However, this animal became extremely edematous and did not show
typical differentiation and, moreover, its sister half (vegetal) developed only as
far as the pluteus stage. It is likely that the plane of cutting might have been
further toward the vegetal pole than usual.

In general the vegetal halves developed into plutei, usually having two arms,
and many of them metamorphosed. The young images mostly showed typical
differentiation and survived for several days after their metamorphosis.

(5) Half-gastrulae.

Larvae were operated on in the following two stages :

a. Gastrula with archenteron and two small triradiate spicules (11-15 hours after

fertilization).

b. Late gastrula with amniotic imagination (15-17 hours after fertilization).

Meridional halves: percentage of metamorphosis is minimal at the stage a,
physiological condition poor, progressive decrease in regulative capacity.

Results of operation through these stages show a general downward trend of
regulative power. As for the degree of bodily differentiation, by the time the late
gastrula stage is reached, many half-imagos show a reduced number of plates or
tooth-rudiments or spines on the side of the operation.

96 KAYO OKAZAKI AND KATSUMA DAN

The same is true with the number of arms : the later the stage of operation,
the more larvae with one arm are produced until finally all of them become one-
armed if operated on at the late gastrula stage.

However, from the standpoint of metamorphosing capacity, a rather unexpected
result was obtained. It was found that the frequency of successful metamorphosis
improves toward the end of the gastrula stage in spite of a continual loss of regu-
lating power for bodily organization. Although it is recognized that operation
during the invagination process seems to act more deleteriously than in later stages,
the situation does not seem to be so simple, since in still later stages, such as the
pluteus or imago stages, the percentage of metamorphosis keeps on improving
despite more defective organization of the larvae.

Equatorial halves: metamorphosing capacity of animal halves negligible, of
vegetal halves rather strong.

The development of the equatorial halves of the gastrulae is very much the same
as that of operated blastulae. Animal halves can only reach the stage of blastulae
with mesenchyme cells except one instance of metamorphosis in an extremely
edematous larva. On the contrary, vegetal halves do metamorphose in much better
condition and even survive several days after metamorphosis.

(6) Half-plutei.

Larvae were bisected at the following stages :

a. Early pluteus with two fully developed arms, but without adult skeleton (19-25

hours after fertilization).

b. Pluteus with some adult skeleton (45-55 hours after fertilization).

c. Late pluteus with well-developed adult spines and plates (70-90 hours after

fertilization).

The results are included in Table I and Figure 4.

Meridional halves: cqui- and toti-potent; capacity for metamorphosis stronger
than that of half-blastulae and -gastrulae.

Meridional half-larvae operated on at these stages had one arm and were not able
to form another, but many of them metamorphosed, although many half-imagos be-
came more or less edematous.

Anterior and posterior halves: equi- and toti-potent with respect to meta-
morphosis.

Anterior half-larvae apparently were not able to regenerate the posterior por-
tion, and posterior half-larvae re-formed no arms, yet both halves metamorphosed.
It is particularly worth mentioning that both anterior and posterior halves of a single
larva are able to metamorphose.

In young imagos derived from anterior halves, the physiological condition seems
to be inferior to that of those derived from meridional or posterior halves, since all
the former became edematous and died without developing, while the latter survived
longer and grewr to some extent, although eventually they also became more or less
edematous.

Usually, in any of these half-imagos, some lack of plates, spines or teeth is found
on the side of the operation. This absence of parts in the half-imagos becomes more
and more conspicuous as the stage of operation advances. However, the total num-
ber of the plates, spines and teeth found in the two halves from a bisected larva is
always greater than that of the control whole imagos.

METAMORPHOSIS OF PARTIAL LARVAE 97

Animal and vegetal halves: equi- and toti-potent in metamorphosing capacity.

As indicated in Figure 3, (5)b and (6)b, the original egg-axis is bent at the
pluteus stage, so that the anterior and posterior halves of the pluteus do not coincide
with the animal and vegetal halves, respectively, of the blastula or gastrula, regard-
ing the egg-axis. Therefore, in order to obtain half-larvae corresponding as nearly
as possible in this respect to the animal and vegetal halves of the earlier stages,
plutei were cut by a frontal section into front (animal) and rear (vegetal) parts
(see Fig. 3, (6)b).

Of 15 plutei operated on, 7 pairs and 5 vegetal halves metamorphosed. These
young sand dollars are much like those derived from meridional halves of plutei of
the same age.

(7) Bisected imagos : both halves survive for several days and continue growth.
Young imagos were bisected immediately after metamorphosis. Of 10 imagos
so treated, 7 pairs and 3 singles survived for several days after the operation and
were able to increase the number of spines and tube-feet to some extent, although
they could not regulate them to the typical numbers. It appears that physiological
recovery in the bisected animal is relatively easy even after metamorphosis.

DISCUSSION

As is indicated in Figure 4, the metamorphosing capacities of both meridional
and equatorial halves of Peronella larvae show a similar tendency to drop to a mini-
mum at stage (5)a (gastrula stage). The regulative capacity, on the other hand,
steadily decreases as the stage of operation advances. Such a relation is precisely
that which would be expected from the experiments of previous workers (Jenkin-
son, 1911 ; Horstadius, 1936, 1939b), so far as they go within the limit of the pluteus
stage. However, the fact that all the curves of half-larval metamorphosing capacity
rise steadily from the late gastrula stage on, seems to be of considerable significance.
Especially animal halves, which were found to be almost lacking in metamorphosing
capacity before the gastrula stage, acquire the capacity in the pluteus stage.

The interpretation of these curves is very difficult. However, several supposi-
tions with regard to each point will be presented, although no final conclusion can
be reached at this time.

The percentage of half-larvae operated on at the two-cell stage which are able
to metamorphose is unexpectedly low. This is probably due to the fact that there
is no way for the investigator to reject individuals with low viability at such an
early stage, although this elimination is automatically realized when older larvae
are used. If the selection of larvae with high viability were possible, a higher meta-
morphosing capacity than that which appears in the present results would un-
doubtedly be demonstrated.

When the four blastomeres of the four-cell stage are separated, all four quarter-
larvae metamorphosed. This is a further amplification of Horstadius' well known
work. Corresponding experiments on Amphibia by Ruud (1925) indicate that
the situation differs slightly in these forms, and only blastomeres carrying the future
site of the organizer are totipotent. For mammals the data are lacking, except that
Seidel (1952) succeeded in obtaining a perfect rabbit from one blastomere of a
two-cell stage which was implanted in another female in the right physiological
condition.

98 KAYO OKAZAKI AND KATSUMA DAN

Of equatorial halves separated at the sixteen-cell stage, the animal halves de-
veloped only to the blastula stage, while the vegetal halves, generally speaking, were
able to metamorphose. However, for some reason not yet understood, only a few
of such vegetal halves were able to metamorphose. On the other hand, relatively
many vegetal halves of blastulae or gastrulae were able to metamorphose, as com-
pared with vegetal halves of the sixteen-cell stage. A possible explanation might
be that since the ectoderm overgrows toward the vegetal pole as the larvae develop
(see Fig. 3), the vegetal halves of the swimming stages contain a larger amount of
ectoderm than similar halves of early stages, so that the balance between animal and
vegetal elements in later stages will approach more closely to the normal than in
earlier stages.

Although the direction of the egg-axis shows a clear correlation with the presence
and absence of metamorphosing capacity through the gastrula stage, this is lacking
in the pluteus stage, as evidenced by the fact that animal and vegetal half-larvae are
equally able to metamorphose, and produce images resembling those developing from
meridional halves of the same stage. This result indicates that some other factor
has superseded in importance the original animal-vegetal relation by the time the
pluteus stage is reached.

When larvae were operated on after the formation of the amniotic invagination,
both meridional and transverse halves metamorphosed, and the percentage of meta-
morphosis increased as the stage of operation advanced. This result is probably
due to the specific developmental mode of Peronella. Since the echinus rudiment
of this form is found in the center of the body and develops to a large size, bisection
in any direction will give each half approximately half of the rudiment, and the size
of the half-echinus rudiment which goes to each half-larva becomes larger and larger
with advance of the stage of operation. Consequently, regulation for metamorpho-
sis of half-larvae will become increasingly easier as the operational stage advances.
It seems quite possible that this central position and marked development of the
echinus rudiment in the pluteus stage constitute the factor which takes primary im-
portance, over that of the animal-vegetal axis, in determining the metamorphosing
capacity of larval regions.

The wrriters wish to express their sincere thanks to Dr. J. C. Dan for her
assistance in the preparation of the manuscript.

SUMMARY

1. In the larva of the sand dollar Peronella, the mouth does not open, and no
functional digestive tract is formed so that no feeding is necessary before meta-
morphosis.

2. Both the amniotic cavity and the hydrocoel take a median position in the
larval body.

3. Cutting experiments show that single blastomeres of the two- or four-cell stage
are totipotent regarding metamorphosing capacity.

4. Any meridional half of the larval stages has the capacity for metamorphosis.

5. The vegetal half of the sixteen-cell stage metamorphoses, but the animal half
develops only to the blastula stage.

METAMORPHOSIS OF PARTIAL LARVAE

6. The vegetal halves, but not the animal halves, of the blastula and gastrula
stages are able to metamorphose.

7. Both anterior and posterior halves of the pluteus stage are able to meta-
morphose.

8. The percentages of metamorphosis of partial larvae fall to a minimum at
the gastrula-stages after which they rise while the regulative capacity falls as a
course of a steady decrease.

LITERATURE CITED

DRIESCH, H., 1891. Entwicklungsmechanische Studien. I-II. Zeitschr. Wiss. ZooL, 53: 160-

184.
HARVEY, E. B., 1940. A new method of producing twins, triplets and quadruplets in Arbacia

punctulata, and their development. Biol. Bull., 78 : 202-216.
HOLTFRETER, J., 1933. Nachweis der Induktionsfahigkeit abgetoteter Keimtelle. Arch. f.

Entw., 128: 584-633.
HORSTADIUS, SVEN, 1935. Uber die Determination im Verlaufe der Eiachse bei Seeigeln. Pub.

Stas. Zool. Napoli, 14: 251-429.
HORSTADIUS, SVEN, 1936. tJber die zeitliche Determination im Keim von Paracentrotus Hindus

Lk. Arch. f. Entw., 135: 1-39.
HORSTADIUS, SVEN, 1939a. tlber die Entwicklung von Astropecten aranciacits. Pub. Stas. Zool.

Napoli, 17: 221-312.
HORSTADIUS, SVEN, 1939b. The mechanism of sea urchin development, studied by operative

methods. Biol. Rev., 14 : 132-179.
JENKINSON, J. W., 1911. On the development of isolated pieces of gastrulae of the sea urchin,

Strongylocentrotus lividus. Arch. f. Entiu., 32: 269-297.
MORTENSEN, TH., 1921. Studies of the development and larval form of echinoderms. Published

at the Expense of the Carlsberg Fund. Copenhagen.

MORTENSEN, TH., 1948. A monograph of the Echinodea. Copenhagen IV, 2.
ONADA, K., 1937. Note on development of some Japanese echinoids with special reference to the

structure of the larval body. II. Japanese J. Zool., 8 : 1-13.
RUUD, G., 1925. Die Entwicklung isolierter Keimfragmente friihester Stadien von Triton

taeniatus. Arch. f. Entw., 105 : 209-293.
SEIDEL, F., 1952. Die Entwicklungspotenzen einer isolierten Blastomere des Zweizellenstadiums

im Sangetierei. Naturwiss., 39 Ja. Ht. 15 : 355-356.
VINTEMBERGER, P., 1934. Resultats de 1'anto-differenciation des quarte macromeres isoles au

stade de huit blastomeres dans 1'oeuf d'un amphibian anoure. C. R. Soc. Biol., 117: 693~

696.
VINTEMBERGER, P., 1935. Sur les resultats du developpement des quatre micromeres isoles au

stade de huit blastomeres, dans 1'oeuf d'un amphibien anoure. C. R. Soc. Biol., 118:

52-53.

AN X-RAY DOSE-ACTION CURVE FOR EYE-COLOR MUTATIONS

IN MORMONIELLA

DAVID T. RAY AND P. W. WHITING 1

Howard University, Washington, D. C., University of Pennsylvania, Philadelphia, Pa.,
and the Marine Biological Laboratory, Woods Hole, Mass.

Visible mutations have been induced in many widely diverse species of organisms
by the use of x-rays and other ionizing radiations. The literature has been reviewed
by Lea (1947) and by Catcheside (1948). Rate of such mutations is low, only
one-tenth to one-fifteenth that of lethals, and hence conclusions in regard to effect
of intensity differences, fractionation of dose, wave-length and combination with
other factors such as oxygen pressure, infra-red rays and temperature have been
drawn largely from work with lethals. Because, insofar as data have been ac-
cumulated, the proportion of visibles to lethals appears to be the same under different
conditions of irradiation, these conclusions have seemed justified.

DOSE-ACTION CURVES

Dose-action curves for visibles have been shown in several organisms and have,
within the limits of error of the experiments, proved similar to those for lethals.
The curves formed are of the straight-line type of direct proportionality, meaning
that, for a given increment of dose at any interval within the range, a similar propor-
tion of mutations is added.

Drosophila. The most satisfactory information regarding dose-action curves for
visibles is available from work with the fruit-fly Drosophila inelanogaster (Timo-
feeff-Ressovsky and Delbriick, 1936). Two methods were used for identifying sex-
linked visibles — the attached-X and the C1B. The former is the more convenient.
Wild-type males are x-rayed and mated to females with their X-chromosomes at-
tached so that the offspring are one hundred per cent non-disjunctional. Since the
sons receive only paternal and therefore treated X-chromosomes, any visibles in-
duced may be observed without further breeding. The C1B method is much more
laborious. The treated wild-type males are mated to C1B females (females having
in one X-chromosome an inversion preventing crossing-over, a recessive lethal fac-
tor and the factor for bar eyes) and the bar (C1B) daughters from this cross are
set in individual cultures. If a visible is induced in a given X-chromosome, all the
males in a culture from a bar female receiving that chromosome will show the mutant
trait.

Table I shows data re-arranged from Timofeeff-Ressovsky and Delbriick ( 1936).
Linearity is indicated by both methods, but percentages for total visibles recognized
by the C1B are considerably higher for each of the three doses given. As pointed
out by the authors this is not caused by any errors in dose, because the males from

1 The research herewith reported was conducted under a contract with the U. S. Atomic
Energy Commission to the Marine Biological Laboratory and was aided by grants to P. W.
Whiting from the Johnson Fund of the American Philosophical Society and from the Board of
Graduate Education and Research of the University of Pennsylvania. The author is indebted to
Dr. John R. Freer and Dr. Sewall Wright for suggestions regarding the statistical aspects.

100

DOSE-ACTION CURVE IN MORMONIELLA

101

TABLE I

X-ray dose-action data re-arranged from Timofeeff-Ressovsky and Delbrtick (1936) on sex-linked

mutations in Drosophila melanogaster as identified by ClB and attached-X methods. The data

are presented as — mutations /total cultures (or total males) = percentage.

(lower — upper 0.95 confidence limits)

Dose
r units

ClB method.
Total visibles

Attached-X method.
Total visibles

Attached-X method.
Seven selected

ft 79378 ft ( Oft 1 ^

ft/7Q3^ ft ( ftft fK^

1500
3000
6000

10/4583=0.22 (.10- .40)
16/3396 = 0.47 (.27- .77)
15/1809 = 0.83 (.46-1.37)

u/ /yjo — U ^.UU .\jj)
13/9317=0.14 (.07-.24)
21/8442=0.25 (.15-38)
26/5183=0.50 (.33-.73)

9/18,000 = .05 (.02-.10)
28/22,500 = .12 (.08-.18)
32/14,500 = .22 (.15-31)

Cultures — 12,166

Males— 30,878

Males— 55,000

the same lot were rayed at the same time for both methods. The disparity is due
rather to the fact that the mutant type appears as a single individual in the attached-X
method but as all the males of a culture produced by a ClB female. A single mutant
may be missed by the observer, especially if the trait is not very distinct, or it may
fail to develop if the mutant type is of lowered viability. It is obvious that the
labor of dealing with the 12,166 ClB cultures in search of total visibles must have
been far greater than that of rearing and observing the 30,878 sons from the
attached-X mothers.

A method of avoiding both the viability disadvantage and the subjective factor
causing failure to observe some of the less distinct visibles is to select a limited num-
ber of easily recognized traits of high viability which may then be used in an
attached-X experiment. This was utilized by the authors who selected seven sex-
linked mutant traits (we, w, y, v, m, g, f ) in a test in which 55,000 sons of treated
wild-type males were counted. Percentage of mutations noted is here less than half
that for total visibles but undoubtedly a high degree of accuracy was attained. Dose-
action linearity is again indicated.

In Table I lower and upper 0.95 confidence limits are presented as calculated
from Kicker's (1937) table for Poisson frequency distributions. The authors give

TABLE II

X-ray dose-action data on eye-color mutations from Woods Hole wild-type stock
of Mormionella mtripennis (Walker)

Mutants

Dose
r units

Mothers

Total
sons

Sons per
mother

Bright-eyed

Dark

red

Scarlet

Orange

Oyster

Total

% (0.95 confidence limits)

0

632

18,039

28.54

0

0

1

1

.0055 (.00055-.03104)

0

1340

729

16,011

21.96

17

2

3

22

.14 (.08-.21)

8

2680

874

10,058

11.51

17

3

5

25

.25 (.16-.37)

5

4020

853

5,268

6.18

20

2

4

26

.49 (J2-.72)

1

5360

855

2,708

3.17

17

3

3

23

.85 (.54-1.27)

0

Totals

3942

52,084

71

10

16

97

14

102

DAVID T. RAY AND P. W. WHITING

standard errors following percentages of mutations. Standard errors are not ap-
plicable to very low percentages because of asymmetry in the distributions. Ricker's
method has likewise been followed in Table II and in Figure 1 for our Mormoniella
data.

A dose-action curve for production of visibles in the X-chromosomes of the egg
might be obtained by x-raying the females of Drosophila and examining their sons.
Their daughters also might be set individually and the F2 males examined. The
former procedure would be less laborious but would, like the attached-X method,
be less satisfactory because of the subjective factor in failure to recognize less dis-

I 20

1340

2680
Dose in r

4020

5360

FIGURE 1. Average numbers of sons per mother and percentages of eye-color mutations
produced by Mormoniella females, control and x-rayed with different doses. Lower and upper
€.95 confidence limits calculated from Ricker's (1937) table are given for the mutation percentages.

tinct types and because of reduced viability of some of the mutants. Testing of
daughters, although laborious, would have the advantage, as in the C1B method, of
many mutants resulting from the same mutation, but only half, instead of all, of the
males in the fraternity would be expected to show the mutant trait. This last point
might be advantageous in the case of less distinct mutant types because wild-type
sibs with the same residual heredity as their mutant brothers would be present for
comparison. Relatively little has been published with regard to radiation of females
in Drosophila, aside from a few abstracts which have recently appeared. Now, how-
ever, these investigations are actively under way. Differences in dose-action rela-
tionships are to be expected, not only between sperm and egg, but between different
meiotic stages of the latter.

DOSE-ACTION CURVE IN MORMONIELLA 103

Habrobracon. Many x-ray visibles have been produced in the parasitic wasp
Habrobracon juglandis (Ashmead) by radiation of both males and females (Whit-
ing, 1932, 1934, 1935). Treatment of unmated females results in mutant sons.
Treatment of males gives rise to F2 fraternities in some of which half the males are
mutant. The same occurs following the mating of treated females to untreated
males. When mated females are treated both eggs and sperm are exposed and
mutants may appear singly as Fx males and as many males in some F2 fraternities.
A few dominants have been produced as in Drosophila which may show directly
in the heterozygous females and some recessive mutations have been recognized
because of their dominance over a third allele in a compound female. While Habro-
bracon has the theoretical advantage of male haploidy, there are practical disad-
vantages (web- and cocoon-spinning of both parasite and host) making isolation of
virgin females and collection of food more laborious. For these reasons dose-action
curves have not as yet been obtained.

Mormoniella. In the chalcidoid wasp Mormoniella vitripennis (Walker) nu-
merous x-ray-induced eye-color mutations from wild type (dark brown) have
already been reported (Whiting, 1951). These range from dark red through
tomato, vermilion, scarlet and peach to "oyster," the last named being devoid of
pigment and transparent so that the black color of the underlying integument
shows through. Thus the appearance of the eye suggests an oyster. When oyster
wasps are placed in alcohol, the eyes become opaque white. The eye colors ranging
from tomato to oyster may be classed as "bright." All of these that have been tested
have proved to be hereditary, dependent upon recessive gene differences from wild
type. The dark reds are for the most part also hereditary but occasionally one is
found which has bred as a somatic overlap from wild type. Some very dark red-
dish-brown types occurring in wild stock have proved to be hereditary. These are
readily separated as "red-eyed pupae" from their sibs which have dull reddish-brown
eyes in the late pupal stage.

Spontaneous eye-color mutations are rare or at most very infrequent in pure
stocks, either wild-type or mutant-type. Spontaneous "mutants" are relatively fre-
quent among the offspring of heterozygous females. However, some at least of
these "mutants" are recombinants. The problem of spontaneous mutation is being
further investigated.

All of the induced eye-color mutations have been obtained by radiation of females.
The majority have appeared as single sons of unmated females but in a few cases,
when treated females were crossed to untreated males, the mutants from any one
mutation constitute about one-half of the sons of a single daughter from the cross.
Because testing daughters individually is time-consuming, rates for visible viable
mutations are determined from inspection of haploid sons of treated females. By
exclusion of dark reds and reddish-browns from the final calculations, the subjective
factor is reduced to a minimum.

Genetic evidence indicates that all males are haploid (Whiting, 1951).
Biparental diploid males such as occur as sex homozygotes in the ichneumonoid
wasp Habrobracon have not been found in Mormoniella. Females carry genes
derived from both parents, but sons of mated females as of unmated are gynogenetic
except for rare instances interpreted as possible androgenesis. Sex determination
must then be different from that in Habrobracon as it has been shown to be in the

104 DAVID T. RAY AXD P. W. WHITING

related chalcidoid Melittobia (Schmieder and Whiting, 1947). The mechanism of
sex determination has yet to be discovered for this group.

During the summer of 1952 tests were made at the Marine Biological Laboratory
by David T. Ray of wild-type and of various mutant stocks of Mormoniella to
determine which might be more suitable for a dose-action curve. Wild-type
(WH+) was selected, a stock inbred from wasps that were infesting fly pupae at
the Supply Department dock on the Eel Pond. Female pupae were isolated and
freshly eclosed virgins or very dark female pupae were placed in gelatine capsules
for raying. Treatments were given with the x-ray apparatus having two tubes
in alternate parallel, cross-firing through the specimens which were placed 13
centimeters from the targets. KVP was 182, MA 25, equivalent filtration .152 mm.
Cu, intensity approximately 2680 r/min. Ravings were made on 26 different
days. Females from each host puparium were divided among the capsules given
the different doses as a precaution to equalize distribution in case a mutant trait
were running in the stock. Among the 52,084 sons of 3,942 mothers there were
97 mutants classed as having "bright" eyes and 14 with eyes dark red (Table II).
While none of the latter occurred among the 18,039 controls, their distribution was
irregular among the treated. Rate is higher but not significantly so among those
from females given 1340-2680 r (13/26069), .050% (.026-.086), than among
those from females given 4020-5360 r (1/7976), .013% (.001-.070). Rate for
total treated (14/34045), .041% (.023-.069), is significantly higher than the
zero rate for the controls (0/18039), 0.0% (0.000-0.021). Test of one dark
red from the 2680 r treatment showed that he bred as wild type. The others
were not tested.

The 97 bright-eyed mutants were classed as scarlet, orange and oyster. One
oyster appeared among the controls. Scarlet, represented by 71 mutants among
the total 97, is by far the most frequent. Only in one instance were two eye
mutants found in a single vial and these were different, a scarlet and an oyster
from the 5360 r treatment. Each bright-eyed mutant may then be considered
to result from a separate mutation.

Stage of meiosis at time of treatment is of interest. With the controls and
lower treatments mutations might appear from first meiotic metaphase as well
as from prophases. With higher treatments, 2000 r and above, few if any
metaphase-treated eggs would be expected to produce offspring, if inferences may
be drawn from lethal rates in Habrobracon. It is very unlikely that any progeny
have been included from treated gonial cells because transfers were not made to new
vials. Death of offspring developing from eggs treated as young oocytes, if
indeed the parents survived to lay such eggs, would be expected to result from
exhaustion of food supply. Replication could occur from mutations in early
gonial divisions only, since a single egg and its accompanying nurse cells comprise
the products of the last four gonial divisions. Failure of similar mutant types to
appear in the same vial is in agreement with this expected lack of replication.
However, the presence of similar mutant types in one vial would not prove replica-
tion. They should occur rarely by chance from two separate mutations, especially
in the case of scarlet, the most frequent mutant type.

Figure 1 shows percentage of bright-eyed mutants increasing with increasing
dose. The curve dips at 2680 r but this dip is not a significant departure from
the straight line expected on the basis of single hits producing the mutations.

DOSE-ACTION CURVE IN MORMONIELLA

105

A method of calculating goodness of fit of these data to a straight line has
been suggested by Dr. Sewall Wright (Table III). Since doses given were
simple multiples of the minimum dose, 1340 r, this may be taken as the unit dose.
"Wasp-doses" are then the number of surviving wasps multiplied by the number
of 1340 r units to which the eggs were subjected. Ninety-six bright-eyed mutants
resulted from 62,763 "wasp-doses." Distribution of mutants calculated (c) on the
basis of 1340 r having the same chance (96/62,763 -- .00153) of producing a
mutation, regardless of amount of dose, would then be in proportion to the distribu-
tion of "wasp-doses" among the survivors. From the differences between the
observed number of mutants (o) and the calculated (c), chi square, 3.97, was
obtained. This deviation is insignificant.

Average numbers of sons per mother (Table II and Figure 1) show decrease
with increasing dose. Females subjected to x-radiation become sterile or die

TABLE III

Calculation of goodness of fit to linearity of bright-eyed Mormionella mutants from x-rayed mothers.
Data from Table II. (Method suggested by Dr. Sewall Wright)

Mutants

Uosc in

..

Co -c)2

r units

c

(o)

(c)

i

16,011

16,011

22

24.49

.25

2

10,058

20,116

25

30.77

1.08

3

5,268

15,804

26

24.17

.14

4

2,708

10,832

23

16.57

2.49

Totals

34,045

62,763

96

96.00

3.97

n=3

= .26

after four or five days. Many of the untreated also die at this time but a minority
may be transferred to fresh host pupae and will produce further offspring. In the
present experiment transfers were not made so that average potential fecundity of
controls is higher than that indicated.

DISCUSSION

Dose-action curves for visibles in Drosophila are essentially linear, indicating
single hits producing the mutations. There is no dip at the mid-point (3000 r),
and in two of the three experiments a slight rise occurs. The dose-action curve
for bright-eye-color mutations in Mormoniella is likewise consistent with linearity
but there is a dip at the mid-point (2680 r). However, the confidence limits are
wide enough both in Drosophila and in Mormoniella to permit an hypothesis either of
uninterrupted linearity or of a dip (at 26SO-3000 r). Neutron experiments
(unpublished) indicate a dip in the dose-action curve for eye colors in Mormoniella.
More extensive x-ray experiments are in progress with Mormoniella which should
narrow the confidence limits sufficiently to establish definitely whether or not a
dip is present. Comparison may then be made with the neutron tests and the
significance of the dip may be considered.

106 DAVID T. RAY AND P. W. WHITING

SUMMARY

1. X-ray dose-action curves for visible mutations in Drosophila are discussed.
An x-ray dose-action curve for eye-color mutations in Mormoniella is presented.
Within the limits of error of the experiments the curves may be of the straight-line
type indicating that single hits produce the mutations. However, in the Mormoniella
curve an insignificant dip occurs at the mid-point, 2680 r, suggesting the possibility
of a second factor.

2. A shortened chi square method of testing goodness of fit to a straight line
is presented. With reference to the present Mormoniella data, the deviation
is shown to be insignificant.

LITERATURE CITED

CATCHESIDE, D. G., 1948. Genetic effects of radiations. Advances in Genetics, 2 : 271-358.

Academic Press, New York.

LEA, D. E., 1947. Actions of radiations on living cells. The Macmillan Co., New York.
RICKER, WM. E., 1937. The concept of confidence or fiducial limits applied to the Poisson

frequency distribution. /. Amer. Stat. Assoc., 32 : 349-356.
SCHMIEDER, R. G., AND P. W. WHITING, 1947. Reproductive economy in the chalcidoid wasp

Melittobia. Genetics, 32 : 29-37.
TIMOFEEFF-RESSOVSKY, N. W., AND M. DELBR'UCK, 1936. Strahlengenetische Versuchc liber

sichtbare Mutationen und die Mutabilitat einzelner Gene bei Drosophila inclanoyaster.

Zeitschr. indukt. Abstramm.-u. Vererb., 71 : 322-334.
WHITING, P. W., 1932. Mutants in Habrobracon. Genetics, 17 : 1-30.
WHITING, P. W., 1934. Mutants in Habrobracon, II. Genetics, 19: 268-291.
WHITING, P. W., 1935. Recent x-ray mutations in Habrobracon. Proc. Pa. Acad. Sci.,

9 : 60-63.
WHITING, P. W., 1951. Multiple complementary alleles in Habrobracon and Mormoniella.

/. Genetics, 50: 206-214.

STUDIES ON THE HELMINTH FAUNA OF ALASKA. XVII.

NOTES ON THE INTERMEDIATE STAGES OF SOME

HELMINTH PARASITES OF THE SEA OTTER

EVERETT L. SCHILLER 1
Arctic Health Research Center, Anchorage, Alaska

According to the work of Rausch (1953), two species of helminth parasites,
Porro caecum decipiens (Krabbe, 1878) and Microp hallus pirum (Afanas'ev, 1941 ),
are pathogenic for the sea otter, Enhydra lutris (L.), on the Aleutian Island of
Amchitka. In continuation of investigations of sea otter mortality on Amchitka
during the latter part of May and early June, 1952, the writer made an attempt to
obtain information on the life cycles and developmental characteristics of these
parasites. It is the purpose of this paper to report the results of these observations.

MATERIALS AND METHODS

Collections of marine invertebrates were made with special effort to obtain
those which are known, from previous studies (Murie, 1940), to be included
in the diet of the sea otter. For the most part, these collections were restricted to
the intertidal area. Attempts to procure samples of bottom forms in the deeper
waters of Constantine Harbor by means of dragging a triangular dredge from
the stern of a small collapsible canvas boat were relatively unsuccessful.

Fishes, Lebius superciliosus (Pallas) and Hemilepidotus hemilepidotus
(Tilesius), were obtained from the waters of Constantine Harbor by means of
funnel-type fish traps baited with carcasses of birds previously autopsied in con-
nection with related parasite studies. Efforts to obtain the larger species of crabs
by the use of crab traps at a depth of about 15 fathoms were unsuccessful.
Amphipods, Anipithoc rubricate! (Montagu) and Anony.v nuga.v (Phipps), were
usually abundant on the baits when the traps were pulled for inspection.

With the exception of representative specimens preserved for purposes of
identification, most of the marine animals were autopsied in the field.

Trematode metacercariae were placed on a glass slide and excysted by gentle
pressure of a cover glass. The excysted larvae were stained in vivo with orcein
dissolved in acetic acid, by allowing the staining fluid to flow slowly under the
cover glass until the desired differentiation was attained. Additional metacercariae
which had been removed from the host tissue in the field were counted and
preserved in alcohol-formalin-acetic acid (AFA) solution. Supplementary speci-
mens were preserved in AFA, with the metacercariae intact in the host tissue.

Nematode larvae were fixed in AFA and cleared for study in liquefied phenol.

RESULTS

Two species of fishes, a greenling, L. superciliosus, and a sculpin, H. licnii-
lepidotus, abundant in the waters around Amchitka, were commonly found to con-
tain the larvae of a nematode, Porrocaecum decipiens. Observations indicate that

1 Parasitologist, Animal-borne Disease Branch, Arctic Health Research Center, Public
Health Service, Department of Health, Education and Welfare, Anchorage, Alaska.

107

Porro caecum decipiens (Krabbe, 1878)

56

500 u 1

500 t, 2

100 i)

FIGURES 1-9.
108

375

HELMINTH FAUNA OF ALASKA. XVII 109

the greenling occurs frequently in the diet of the sea otter and probably constitutes
the most important source of severe nematode infections acquired by these mammals.

The greenling and sculpin also harbored larval acanthocephalans (Corynosoma
sp.) which may represent the second intermediate stage of a species parasitic in
the sea otter. A brief discussion of this form is included.

The metacercarial stage of the trematode Microphallus pintiu was found in a
hermit crab, Pagnnts liirsiitiusculus (Dana). The finding of the intermediate host
for the last larval stage of M. pinun permits an understanding of the probable
source of the heavy infections with this trematode in the Amchitka sea otter, and
will make it possible to obtain additional information on the life cycle and
pathogenicity of this species through experimental infections.

The larval stages of Porrocaecum dccipiens and the metacercaria of M. pinun
are considered separately in some detail from the standpoint of host occurrence,
prevalence of infection, and morphology.

Stiles and Hassall (1899) published a description of Ascaris decipicns, to which
Baylis (1916) contributed further details. The species was later referred by
Baylis (1920) to the genus Porrocaecum Railliet and Henry, 1912. Although
there appears to be some disagreement concerning the validity of the name
Porrocaecum decipicns, as indicated by the discussion of the taxonomic status of
this nematode given by Johnston and Mawson (1945), the writer has preferred
to retain this name for the purposes of this paper.

Several species of marine mammals harbor the adult stage of P. dccipiens.
Around Amchitka, the harbor seal, Phoca ritnlina L., Steller's sea lion, Eumetopias
jubata (Schreber), and the sea otter, serve as definitive hosts for this nematode.
A list of species from which P. dccipiens has been recorded in the northern
hemisphere was given by Baylis (1937). Several investigators have reported
the occurrence of the larval stages in various species of fishes, and it is apparent
from these published records that this nematode has an extremely wride geographical
distribution.

Although the complete life cycle of P. decipiens has never been demonstrated
experimentally, Stiles and Hassall (1899) recognized no difference between
encysted larvae found in fishes and the youngest worms occurring in the fur seal.
Callorhinus nrsinus (L.). The seal harbored all intermediate stages between
the youngest forms and the adults of P. dccipiens. These authors concluded
that such close relationships between the definitive host and the probable inter-
mediate hosts suggested the source of infection nearly to the point of certainty— a
view commonly accepted by helminthologists.

FIGURE 1. Porrocaecum dccipiens; head and esophageal portion of second stage larva
from stomach of sea otter. (Earliest larval stage found in this animal.)

FIGURE 2. P. dccipiens ; head and esophageal portion of third stage larva from musculature
of Lcbius superciliosus.

FIGURE 3. P. decipiens; head of third stage larva from stomach of sea otter. (In process
of shedding cuticular sheath.)

FIGURE 4. P. dccipiens; tail of third stage larva from musculature of L. superciliosus.

FIGURE 5. P. dccipiens; tail of second stage larva from stomach of sea otter.

FIGURE 6. P. dccipiens; head of third stage larva from musculature of /.. superciliosus.

FIGURE 7. P. dccipiens; head of fourth stage larva from small intestine of sea otter.

FIGURE 8. P. dccipiens; head of adult male from small intestine of sea otter.

FIGURE 9. P. dccipiens; tail of adult male from small intestine of sea otter.

110

EVERETT L. SCHILLER

According to Stiles and Hassall, both the Alaskan pollock, Theragra chalco-
gramma (Pallas), and the Pacific cod, Gadus inacroccphalus Tilesius, collected in
the Bering Sea, harbored encysted larvae of P. dccipiens. They considered the
former species to be probably the chief source of infection of the fur seal. Scheffer
and Slipp (1944) reported P. decipiens abundant in the harbor seal from the
Pacific coast of the United States. They found the larval stage encysted in the
mesentery of Gadus inacroccphalus in the Aleutian Islands, where the same
species of seal was found to be parasitized by the adult worm.

FIGURE 10. Section of dorsal musculature of L. supcrciliosns showing larva of

P. dccipiens in wound cavity.

Rausch (1953) reviewed previous reports of P. dccipiens in the Aleutian
Island sea otter and recorded his observations on the prevalence of the species
in the sea otter of Amchitka. His report included an account of the pathological
changes in this animal associated with certain developmental stages of P. decipiens.

Data: A total of 106 fishes was examined during the present study on
Amchitka. These consisted of 75 greenlings, L. supcrciliosns, 15 sculpins, H.
hcmilcpidotus, 11 blennies, (Stichaeidae) Anoplarchns pnrpiiresccns Gill, and
5 tide pool sculpins, Myo.vocephalus niger niger (Bean). Of these, both L. supcr-
ciliosus and H. hemilepidotus were found to contain the larvae of P. decipiens.
Data concerning these infections are summarized in Table I. Prevalence of
infection does not appear to be correlated with either sex or size of the fishes
examined.

HELMINTH FAUNA OF ALASKA. XVII

111

Morphology: The larval stages of P. decipiens from the musculature of L.
superciiiosus ranged in length from 30 to 50 mm. All of these immature worms
possessed a boring tooth in the position of the left ventral lip (Fig. 3) and a
small conical projection at the tip of the tail (Fig. 4). The exposed part of the
boring tooth measured about 15 /A in length by 30 /JL in maximum diameter. The
terminal projection attained a length of about 17 /j.. The dorsal and ventral
lips, although visible beneath the cuticular sheath characteristic of this larval stage,
were compressed together in such a manner as to obscure the morphological
details. Examination of identical specimens from the sea otter revealed that
several were in the process of shedding the cuticular sheath (Fig. 3). The
boring tooth and terminal projection are lost during this molt. The three char-
acteristic ascarid lips are incompletely developed but appear to be functional as

TABLE I
Prevalence of larval Porrocaecum decipiens according to sex and size of fishes collected at Amchitka

Fish species: L. superciiiosus
Total number examined: 75
Total number infected: 30 (40%)

H. hemilepidotus
15

3 (20',

Sex

Male

Female

Male

Female

Number of specimens

27

48

10

5

Range in length

210-375 mm.

210-395 mm.

154-345 mm.

170-285 mm.

Average length

292 mm.

319 mm.

251 mm.

248 mm.

Range in weight

101-609 g.

105-960 g.

1 14-520 g.

57-342 g.

Average weight

336 g.

444 g.

252 g.

230 g.

Numbers infected

10 (37%)

20 (41.5%)

1 (10%)

2 (40%)

Range in numbers of larvae

1-9

1-14

j_

1-3

per infected fish

Average number of larvae per

2.7

2.9

1.0

2.0

infected fish

soon as the cuticle is shed. The dorsal lip bears two large papillae and each of
the ventral lips is provided with one. The excretory organ consists of a single,
flattened, band-like cell which extends ventrally, with several lateral branches,
through the anterior region of the worm. The excretory pore is located ventrally
between the two ventro-lateral lips. The esophagus is divided transversely into
an anterior muscular portion and a posterior glandular organ or ventriculus. The
digestive tract is well developed and an intestinal diverticulum, directed anteriad,
is present in all specimens from the musculature of the greenling. The esophageal
region of a typical larva from the musculature of this fish is illustrated in Figure 2.
Measurements of pertinent structures are included in Table II.

In comparing larvae from the fish musculature, no differences, except for those
of size, were noted. A grouping of these larvae according to length, and their
location within the fish musculature, indicates that these differences in size are
correlated with growth and development.

There was no evidence of the beginning development of reproductive organs
in larvae from the fish musculature.

112

EVERETT L. SCHILLER

A study of the immature stages of P. decipiens harbored by the sea otter revealed
that they are morphologically identical with those found in the musculature of
L. superciliosus. Some of these still retained their cuticular sheaths, a few wrere
in the process of shedding the cuticle, and others had completed this molt.

Smaller specimens (18 to 30 mm. in length), representing a stage of develop-
ment earlier than any found in the fish, were also present in the sea otter (Fig. 1).
These larvae were usually seen in dense clusters with their anterior ends deeply

TABLE II

Data on morphological details of larval Porrocaecum decipiens grouped according to location of larvae,

showing relationship between stages of development and migration through the fish musculature.

{Measurements of youngest larvae from stomach of sea otter included for purposes of comparison)

Source of larvae

Youngest lar-
vae from

In musculature
next to ab-

In thick dorsal
musculature of

In tail muscu-
lature of fish.

stomach of

dominal wall

fish.

sea otter.

of fish.

Total length

18-30 mm.

30-37 mm.

38-44 mm.

45-50 mm.

Boring tooth and
tail spike

Present

Present

Present

Present

Total length of
esophagus

Range
Average

2.66 4.16 mm.
3.29 mm.

3.29-3.64 mm.
3.41 mm.

3.38-3.85 mm.
3.61 mm.

3.45-4.35 mm.
3.84 mm.

Length of
muscular part

Range
Average

1.82-3.15 mm.
.235 mm.

1.89-2.10 mm.
1.99 mm.

1.93-2.38 mm.
2.15 mm.

2.01-2.80 mm.
2.30 mm.

Length of
ventriculus

Range
Average

770-1260 n
946 M

1.28-1.54 mm.
1.39 mm.

1.43-1.47 mm.
1.45 mm.

1.33-1.86 mm.
1.56 mm.

Length of
diverticulum

Absent in lar-
vae 18-28

mm. Begin-
ning in larvae
30 mm.

Range
Average

11 2-490 M
340 M

700-980 M
864 M

756-1048 M
870 M

700-1050 M
885 M

Distance to cervi-
cal papillae

Range
Average

392-700 n
523 M

588-728 M
679 M

700-770 M

737 M

700-7 70 M
750 M

imbedded in the mucosa of the stomach, or associated with intestinal perforations
(see Rausch, 1953). These larvae closely resembled those from the greenling,
although the ventriculus had a shrunken appearance and the diverticulum was
absent in all specimens up to 28 mm. in length. An anteriorly-directed structure
about 112 /x, in length was present in the position of the ventriculus in most larvae
29 mm. long (Fig. 1), however, and its length had increased more than four times
(490 /A) by the time the larvae had reached 30 mm. (Fig. 2). This seems to
indicate that development of the diverticulum was very rapid during this stage.
Although larvae of a comparable size were not seen in the greenling, it is

HELMINTH FAUNA OF ALASKA. XVII 113

possible that these smallest forms represent an earlier developmental stage — one
occurring in the intestine or abdominal cavity of the fish, prior to invasion of the
musculature. Measurements of the pertinent morphological characters in the
youngest larvae from the sea otter are included in Table II.

The nematode infections in the sea otter usually comprised all of the immature
stages. The pattern of development of these worms, as interpreted from the study
of morphological characteristics and location in the intermediate host (fish) and/or
in the definitive host (sea otter), is presented below:

First stage larvae: Motile larva in the egg.

Second stage larvae: Larvae up to 28 mm. in length ; boring tooth and tail
projection present; ventriculus smaller in diameter than posterior third of
muscular part of the esophagus ; lips incompletely developed ; diverticulum
absent or only slightly developed ; beginning development of reproductive
organs not evident (Figs. 1,5).

Location in intermediate host: Probably in gastrointestinal tract and/or
abdominal cavity of fish.

Location in definitive host: Attached in dense clusters to mucosa of stomach
in pinnipeds and sea otter ; associated with intestinal perforations in the
case of the sea otter.

Third stage larvae: Larvae 30 to 50 mm. in length ; boring tooth and tail projec-
tion may or may not be present ; lips incompletely developed but functional
when cuticular sheath is shed ; ventriculus well developed ; diverticulum
present, attaining a length of about 4/7 the length of the ventriculus ; beginning
development of reproductive organs not evident (Figs. 2, 3, 4, 6).

Location in intermediate host: In the musculature of fish (greenling).
Larvae possessing cuticular sheath with boring tooth and tail projection.
Location in definitive host: Attached in clusters to mucosa of stomach in
pinnipeds and sea otter. Larvae possessing cuticular sheath with boring
tooth and tail projection, cuticle being shed, or molt completed.

Fourth stage larvae: Larvae over 50 mm. in length; boring tooth and tail
projection absent ; lips completely developed, with dentigerous ridges con-
spicuous ; diverticulum well developed and equal or nearly equal to length
of ventriculus ; beginning development of reproductive organs evident, but
worms sexually immature (Fig. 7).

Located in definitive host only: In stomach and intestine of pinnipeds and
sea otter.

Adult stage: Males about 78 mm. in length; females about 110 mm. in length.
Characteristics of male: total length of esophagus, 4.5 mm., muscular portion,
3.2 mm., ventriculus, 1.2 mm.; diverticulum extends to anterior extremity of
ventriculus; spicules equal, 2.2 mm. in length; post-anal tail length, 271 /*,; six
lateral pairs of post-anal papillae, three pairs near anus and three pairs terminal
(Figs. 8, 9). All morphological characters of the adult P. decipicns examined
in this study conform to descriptions given by Stiles and Hassall (1899) and
Baylis (1916).

114 EVERETT L. SCHILLER

Located in definitive host only: In small intestine of pinnipeds and sea otter.

Larval migration, development and encapsulation: In considering the life cycle
of P. decipicns, it is assumed here that the eggs released by the adult worms in
the definitive host are passed into the sea and are ingested by the intermediate
host (fishes). The larvae apparently are released from the egg, either in the
stomach or intestine of the fish, then penetrate the walls of these organs and
migrate through the abdominal cavity into the musculature. Kahl (1938) dis-
cussed the occurrence of P. decipicns larvae in different parts of the body of the
fish (stomach, body cavity, and musculature) and presented a detailed account of
the process of encapsulation of these larvae within the muscle tissue.

The pattern of migration, development, and encapsulation of the larvae of
P. decipiens as observed in L. superciliosus appears to be essentially the same
as that in smelt, Osmerus cpurlanus, and red perch, Sebastcs norvegicus, as
described by Kahl. Living larvae were recovered from various places throughout
the fish musculature, and it was possible to correlate the stage of larval develop-
ment with the amount of host-tissue reaction and extent to which migration had
progressed. The smallest larvae (30 to 37 mm. long) were found lying in an
extended position in the muscle tissue adjacent to the abdominal wall. Macro-
scopically there was no visible evidence of tissue reaction at this location. Slightly
larger forms (38 to 44 mm. long) were found deeper in the muscle tissue dorso-
lateral to the abdominal cavity. Here, also, the larvae were lying in a more or
less extended position and there was little, if any, change in the tissue in which
they were imbedded. The largest larvae (45 to 50 mm.) were usually found in
the dorso-caudal region of the fish. These larvae were usually more or less coiled,
and apparently had ceased migrating. The worms imbedded in the form of
a loose coil appeared to have evoked moderate cellular changes, visible as cellular
infiltration contrasting in color with the adjacent tissue. More tightly coiled
individuals were seen within cavities apparently produced by them (Fig. 10). The
reaction of the surrounding tissue was more pronounced here and the cavity con-
tained a reddish-brown amorphous substance along with the worm. The degree
of tissue reaction probably is correlated with the duration of larval localization.
Other larvae occurred in compact coils within thin-walled capsules. Sections
through these capsules demonstrated that the capsule membrane is composed
of connective tissue ; however, the connective tissue formation is not nearly as
extensive as that described by Kahl (1938). This last condition represents the
most advanced stage in the process of host tissue reaction observed in the green-
ling, although on two occasions during examination of the sculpins, the charac-
teristic opaque, lenticular capsule containing a dead worm, similar to those
described by Martin (cited by Kahl, 1938), was found. The late "wound-cavity
stage" or early "encapsulation stage" in the greenling occurred most frequently
in the muscle tissue on either side of the pterygiophores of the ventral fin.

It is of interest to note that exposure of any part of the worm during dissection
of the fish usually resulted in its becoming very active, freeing itself completely
from the surrounding tissue within a few minutes. When a living larva, soon
after removal, was placed free upon the musculature and covered with another
sizeable piece of the same tissue, it re-entered and completely imbedded itself
in the muscle in less than ten minutes. These observations suggest that the

HELMINTH FAUNA OF ALASKA. XVII 115

method of penetration is mechanical. The connective tissue of the flesh apparently
offers little resistance to penetration by these worms, since they are not restricted
by the connective tissue septa as observed by Kahl ( 1<)38) to be the case in other
large species of fishes.

There is disagreement among investigators concerning the role of the fish in
the infection of marine mammals by these nematodes. Joyeux and Baer (1934)
expressed the opinion that the life cycle could be accomplished perfectly well
without this intermediate host, but that it served to accumulate and distribute
the larvae. Other authors (Pinter, 1922; Giovannola, 1936; and Fulleborn, 1923;
cited by Punt, 1941) regarded the passage of the larvae through the fish as a
physiological requisite for completion of larval growth and development. Kahl
(1939) was of the opinion that encapsulation of the larva is a method of defense
on the part of the intermediate host but is by no means indispensable for the
development of the larva. He concluded that larvae of P. dccipiens in the digestive
tract of the intermediate host had already completed the development necessary to
permit establishment within the definitive host, following ingestion. This seems
to be the case in the sea otter-greenling cycle as well, although establishment of
larvae of this developmental stage is not without adverse effect upon the sea otter.
Rausch (1953; p. 594) stated that "The earliest stage found in the sea otter
(i.e., worms having a cephalic spike) appears to be the most pathogenic. This
larval stage was always associated with intestinal perforation and seemed directly
responsible for all sea otter deaths known to have resulted from nematode infec-
tion." Inasmuch as the development ordinarily attained during migration and
localization in the fish has not been completed, these larvae may have a tendency
to continue their vigorous migration following ingestion by the definitive host.
This might explain, in part, the pathogenicity of such early stage larvae of P.
decipiens in the sea otter.

Twice during this study, several immature specimens of P. dccipiens, identical
with those found in the flesh of the greenling, were taken from the stomach of the
bald eagle, Haliactus Icucocephalus (L.), and on one occasion from the stomach
of Baird's cormorant, Phalacrocorax pclagicus (Pallas). These worms were
intermixed with the stomach contents and were probably ingested with infected
fishes. This species is not considered to be parasitic in these birds. Murie et al.
(unpublished data) reported finding L. superciliosits in the nests of the bald eagle
on several occasions. Krog (1953) has discussed the occurrence of greenling and
other species of fishes in the nests of the bald eagle on Amchitka.

Corynosoma sp.

Most of the greenling and sculpin examined in this work harbored late-stage
acanthocephalan larvae of the genus Corynosoma. These were found attached to the
mesenteries. Rausch (1953) recorded C. strumosiun (Rudolphi, 1802) in addi-
tion to an undescribed species of this genus - from both the sea otter and Steller's
sea lion at Amchitka. Afanas'ev (1941) described C. cnhydris from the sea otter
of the Komandorskii Islands. It is quite probable that the immature form in
the fish is an intermediate stage of one of these species. Immature specimens of

2 This species has been recently described by Dr. H. T. Van Cleave as Cor\nosoma viHosion
(J. Parasit., 39: 1-13. 1953).

116

EVERETT L. SCHILLER

this genus, similar to those taken from the fishes, were also found in the small
intestine of the bald eagle. It is doubtful that these worms reach maturity in this
avian host. A discussion of the status of these worms in the bald eagle has been
presented in a previous publication (Schiller, 1952). All acanthocephalan material
was studied by the late Dr. H. J. Van Cleave, Department of Zoology, University
of Illinois.

Rausch (1953) reviewed the taxonomic status of this species and presented
a discussion of the pathological changes in the intestine of the sea otter associated
with the presence of this parasite.

Microphallus pirum (Afanas'ev, 1941)

FIGURE 11. Paiittnis hirsutiusculus (about 31/f>X). Arrows indicate
metacercariae of MicropliaUns pintui.

Studies of the life cycle of trematodes of the genus Microphallus have been
mainly concerned with fresh-water species. A notable exception is the work
of Stunkard (1951) with M. linmU, whose metacercariae were found in the horse-
shoe crab, Limuhts polyphemus. His work included a critical consideration of
the systematic position of the genus Microphallus.

During the present study the metacercarial stage of M. piruui was found at-
tached rather insecurely to the inner lining of the abdominal wall and to the tissue
supporting the viscera of a hermit crab, P. hirsittinsciihis (Fig. 11). They were
usually most numerous at the juncture of the cephalothorax and abdomen, but in
heavy infections these cysts occurred throughout the abdomen and occasionally

HELMINTH FAUNA OF ALASKA. XVII

117

o
m

200 jj I 0

FIGURE 12. Metacercariae of M. f>intm from P. hirsittiuscitlits following mechanical
excystment.

FIGURE 13. Metacercaria of M. pirum from P. hirsutiusculus prior to excystment.

FIGURE 14. Metacercaria of M. pirum encysted in the hypodermis lining the
carapace of Tclmcssus sp. (Average diameter about 3COM.)

118 EVERETT L. SCHILLER

in the thoracic region, where they were attached to the hypodermis lining the
carapace.

According to Dr. Fenner A. Chace, Jr., Curator, Division of Marine In-
vertebrates, U. S. National Museum (personal communication), P. hirsutiuscitlns
is one of the most common hermit crabs on the west coast of North America, rang-
ing from the Pribilofs and Aleutians to San Diego, California, and vertically from
low tide to a depth of 17 fathoms. It also occurs in Kamchatka and Japan.

The hermit crabs collected at Amchitka were housed in shells of the gastropods
B-uccinnm baeri Middendorff and Thais emarginata (Deshayes).

Data: Forty-six (90%) of 51 hermit crabs examined in this study were found
to contain encysted metacercariae of M. pirum. The numbers of cysts in the
infected crabs ranged from 11 to 382, with an average of 87. There appeared to
be no correlation in the prevalence of infection with either sex or size of these
hermit crabs.

Description of the metacercariae: Cyst spherical; varying from 392 to 490 /A in
diameter. Cyst wall double : external wall striated and opaque, about 33 /JL in
thickness. Metacercaria occupies almost all of the space within the cyst. Body
of larva curled ventrad upon itself with anterior end innermost and lateral margins
of posterior extremity bent over ventral surface as shown in Figure 13. Excysted
larvae variable in length, but average about 580 p. Cuticular spination conspicuous.
Digestive tract well developed. Subterminal oral sucker measures 48 X 42 p.
Prepharynx 6 to 14 ^ in length ; pharynx about 42 X 19 /x. Length of esophagus
about 240 /*. Ceca 160 ^ in diameter. Genital pore located to left and adjacent
to posterior margin of acetabulum. Male copulatory papilla about 20 /x, in
diameter. Seminal vesicle about 77 /x. in length. Testes ovoid, about 80 p. long,
located near lateral margins just posterior to ends of ceca. Subspherical ovary
about 46 /JL in diameter, situated between but somewhat anterior to right testis
and acetabulum. Deeply lobed vitelline glands, incompletely developed, occur
just posterior to testes. Vitelline ducts were not observed. Uterine loops fill
body area posterior to acetabulum. The uterus is devoid of eggs.

The reproductive organs are well developed in the metacercaria of M. pint in
(Fig. 12). Except for the extent of the vitelline glands and the absence of eggs
in the uterus, the metacercaria appears to be identical with the adult worm.

The work of several authors (Strandine, 1943; Rausch, 1947; Stunkard, 1951)
suggests a considerable degree of morphological variation and a remarkable lack
of host specificity in members of the genus Microphallns. This may well be the
case with M. pirum, since a rather wride range in cyst dimensions is seen in this
species and since the adult is known to occur in the arctic fox as well as in the sea
otter — two hosts phylogenetically not closely related.

The complete life cycle of M. pirum is unknown, but because this species is a
digenetic trematode, it can be assumed that the first intermediate host is a snail.
In view of this, together with the present knowledge of the second intermediate
host, the life cycle of this species, in general, is thought to be as follows: The
eggs are released by the adult worms in the small intestine of the sea otter and/or
arctic fox and are eliminated in the feces. The miracidia gain entrance to the
body of a suitable snail in which the subsequent generations of sporocysts, rediae,
and cercariae are produced. The cercariae leave the snail and penetrate the body
of the second intermediate host, the hermit crab, in which they encyst and develop

HELMINTH FAUNA OF ALASKA. XVII 119

to the metacercarial stage. Upon ingestion of the infected hermit crab by the sea
otter and/or arctic fox, the metacercariae are liberated and attain sexual maturity
in the small intestine of the final host.

Stunkard (1953) found the herring gull, Lanis anjcntatus, to be the final host
of M. limuli and considered it very probable, in view of the lack of host specificity
among microphallid trematodes, that shore-birds may also serve as natural definitive
hosts for this species. Rausch (1953) suggested that M. pintin might infect
birds. It therefore seems appropriate to note that no infections of M. pintin were
found in any birds collected at Amchitka by the writer during the present study.
These included the following species with the number examined : red-throated
loon, Gavia stellata (Pontoppidan) (2) ; Baird's cormorant, Phalacrocorax pelagicus
resplendens Audubon (1); lesser Canada goose, Branta canadcnsis leucopareia
(Brandt) (1); Aleutian teal, Anas crccca (L.) 3 (20); lesser scaup, Aythya
affinis (Eyton) (3) ; Pacific eider, Somateria mollissima v-nigra Gray (4) ; bald
eagle, Haliactus leucoccphaliis (L.) (3) ; black oyster-catcher, Haetnatopiiis
bachmani Audubon (5) ; lesser yellow-legs, Totanus flavipes (Gmelin) (2) ;
Aleutian sandpiper, Erolia ptilocnemis (Ridgway) (5) ; Pacific godwit, Limosa
lapponica baueri Naumann (1); northern phalarope, Lobipes lobatus (L.) (1) ;
parasitic jaeger, Stercorarius parasiticus (L.) (3) ; glaucous-winged gull, Larus
glaucescens Naumann (2) ; arctic tern, Sterna paradisaca Pontoppidan (1) ;
pigeon guillemot, Cepphus columba coluinba Pallas (1); Aleutian rosy finch,
Leucostictc tephiocatis griseonucha (Brandt) (2) ; Pribilof snow bunting, Plectro-
phcna.\- nivalis tozvnsendi Ridgway (2).

The following marine invertebrates were examined, in addition to the hermit
crabs, and were found to be negative for larval stages of parasites infecting the
Amchitka sea otter: limpet, Acmaea digitalis Eschscholtz (15); mussel, Mytilus
ednlis L. (18) ; anemone, Actinea sp. (16) ; sea urchin, Strongylocentrotus
drobachiensis (Miiller) (22) ; snails, Buccinum picturatnm Ball (19), B. bacri
Middendorff and Thais emarginata (Deshayes) (37) ; amphipods, Ainpithoe
rubricata (Montagu) (13) and Anony.v nuga.v (Phipps) (33) ; isopods, Idothca
(Pentidotea) wosnesenskii (Brandt) and Ligia pallasii Brandt (29) ; barnacles,
Balanus spp. (24) ; octopus, Octopus fapollyon Berry (2).

DISCUSSION

Though the percentage of greenlings infected with the larvae of P. decipiens at
Amchitka is quite high, the number of larvae per infected fish is relatively low-
consequently a large number of fishes would have to be consumed by the sea
otter to produce the massive infections frequently found in them. This indicates
that fishes may be much more important food species for this animal than formerly
supposed. Practically all available information concerning feeding habits of the
sea otter has been derived from a study of their feces. A young otter, kept in
captivity for a short time during this investigation, was fed living greenlings. .The
flesh and viscera of the fish were consumed, but the more substantial parts of the
skeleton were usually discarded. Such feeding habits, if characteristic of sea otter
under natural conditions, would explain the scarcity of recognizable fish remains

3 The cestode parasites of this bird have been reported separately by Schiller (Proc. Helm.
Soc. Wash., 20 : 7-12. 1953).

120 EVERETT L. SCHILLER

in their feces. According to Murie et al. (unpublished data, referring to the report
on fishes collected on the 1937 Biological Survey Expedition to the Aleutian
Islands) 42 species of fishes were taken in the Aleutian Islands proper. Probably
a number of these may also serve as intermediate hosts for P. decipiens.

The relatively large number of metacercariae of M. pirum occurring in an
individual hermit crab, combined with a high prevalence of infection, would seem
to assure parasitism in any suitable final host feeding on these crabs. In considera-
tion of the tremendous numbers of worms occurring in some of the infected sea
otter at Amchitka (see Rausch, 1953), it is apparent that a great quantity of
hermit crabs must be consumed by these animals. This leads to the conclusion
that under present conditions at Amchitka, the hermit crab may also be an im-
portant species in the diet of the sea otter.

Other species of crabs, remains of which frequently occur in the feces of the
sea otter at Amchitka, may afford additional sources of infection with M. pirum.
Recent examinations of marine crabs collected on Kodiak 4 disclosed that, in
addition to Pagurns Irirsntiuscitlits, a crab of the genus l^eliiiessits, also harbored
the metacercariae of M. pi nun. The metacercariae in the latter were found
attached to the hypodermis lining the carapace (Fig. 14).

In addition to ecological relationships favoring a high degree of parasite
survival at Amchitka, crowding of the sea otter and their continual occupation of
a rather restricted home-range have resulted in a heavy concentration of parasites
here. As a consequence, any mortality due to this parasitism in the sea otter
may be expected to increase in proportion to the population density and it is con-
ceivable that disease may continue in epizootic proportions until the sea otter
population here is greatly reduced. In view of these circumstances, artificial
reduction of the population through redistribution and/or harvest of the sea
otter as recommended by Rausch (1953) may be the only practical solution.

The writer wishes to take this opportunity to express his appreciation to the
individuals whose assistance and cooperation contributed much to this work.
Identification of material was made by the following : Dr. H. Friedmann, Curator,
Division of Birds, U. S. National Museum (birds) ; Mr. Vladimir Walters. De-
partment of Biology, New York University (fishes) ; Dr. S. S. Berry, U. S.
National Museum (octopus) ; Dr. Fenner A. Chace, Jr., Curator, Division of
Marine Invertebrates, U. S. National Museum (hermit crab) ; Dr. Harald A.
Rehder, Curator, Division of Mollusks, U. S. National Museum (mollusks) ; Dr.
Robert Menzies, Scripps Institution of Oceanography, La Jolla, California
(isopods) ; and Mr. C. R. Shoemaker, U. S. National Museum (amphipods).

The U. S. Fish and Wildlife Service supported the cost of the field work on
Amchitka and the Military Air Transport Service provided transportation to
and from Amchitka.

SUMMARY

1. Two species of fishes, Lcbius snpcrciliosHS and H anile pidotus hemtiepidotus,
serve as the intermediate host for Porrocacciini dccipiens. Observations indicate
that L. supci-ciliosns is the most important source of the nematode infections
acquired by the sea otter on the Aleutian Island of Amchitka.

4 The field work on Kodiak was undertaken by Dr. R. Rausch and Miss R. V. Sacressen of
this laboratory.

HELMINTH FAUNA OF ALASKA. XVII 121

2. The morphological characteristics of the developmental stages of P. decipiens
from fish and the sea otter are described.

3. A hermit crab, Pagunis hirsntiitscitlns, has been found to harbor the
metacercariae of Microphallus pinim, an important parasite of the sea otter at
Amchitka, and this larval stage is described.

4. Some ecological relationships which favor a high degree of parasite survival
at Amchitka are discussed.

LITERATURE CITED

AFANAS'EV, V. P.. 1941. Parazitofauna promyslovykh mlekopitaiushchikh Komandorskikh

Ostrovov. Uchcnic Zapiski, Scriia biologischcskikh Nauk., 18: 93-117.
BAYLIS, H. A., 1916. Some ascarids in the British Museum. Parasit., 8 : 360-378.
BAYLIS, H. A., 1920. On the classification of the Ascaridae. I. The systematic value of

certain characters of the alimentary canal. Parasit., 12: 253-264.
BAYLIS, H. A., 1937. On the ascarids parasitic in seals, with special reference to the genus

Contracaecum. Parasit., 29: 121-130.
JOHNSTON, T. H., AND PATRICIA M. MAWSON, 1945. Parasitic nematodes. In B.A.N.Z.

Antarctic Research Expedition 1929-1931. Reports— Ser. B. Pt. 2, 5: 77-159.
JOYEUX, C. H., AND J. G. BAER, 1934. Les hotes d'attente dans le cycle evolutif des Helminthes.

Biol. Med. Paris, 24 : 482-506.
KAHL, W., 1938. Nematoden im Seefischen. I. Erhebungen iiber die durch Larven von

Porrocaecum decipiens Krabbe in Fischwirten hervorgerufenen geweblichen Verander-

ungen und Kapselbildungen. Zcitschr. f. Parasitenk.. 10: 415-431.
KAHL, \V., 1939. Nematoden im Seefischen. III. Statistiche Erhebungen iiber den

Nematodenbefall von Seefischen. Zeitschr. f. Parasitcnk., 11 : 16-41.
KROG, J., 1953. Notes and observations of birds on Amchitka Island, Alaska. Condor

55 : 299-304.

MURIE, O. J., 1940. Notes on the sea otter. J. Mammal.. 21 : 119-131.
PUNT, A., 1941. Recherches sur quelques nematodes parasites de poissons de la mer du Nord.

Ncm. du Mas. D'Hist. Nat. dc Belgique, No. 98, p. 1-110.
RAUSCH, R., 1947. Some observations on the host relationships of Microphallus opacus (Ward,

1894) ( Trematoda : Microphallidae) . Trans. Amcr. Micr. Sac., 66: 59-63.
RAUSCH, R., 1953. Studies on the helminth fauna of Alaska. XIII. Disease in the sea otter,

with special reference to helminth parasites. Ecology, 34 : 584-604.
RAUSCH, R., AND BETTY LOCKER, 1951. Studies on the helminth fauna of Alaska. II. On

some helminths parasitic in the sea otter, Enh\dra Intris ( L. ). Proc. Helm. Soc.

Wash., 18: 77-81.
SCHEFFER, V. B., AND J. W. SLIP?, 1944. The harbour seal in Washington State. Amcr. Mid.

Nat., 34 : 373-416.
SCHILLER, E. L., 1952. Studies on the helminth fauna of Alaska. V. Notes on Adak rats

(Rattus iwrrct/icits Berkenhout) with special reference to helminth parasites. /.

Mammal.. 33: 38-49.
SCHILLER, E. L., 1953. Studies on the helminth fauna of Alaska. XIV. Some cestode parasites

of the Aleutian teal (Anas crccca L.) with the description of Diorchis lomiiorum n. sp.

Proc. Helm. Soc. Wash., 20: 7-12.
STILES, CH. W., AND A. HASSALL, 1899. Internal parasites of the fur seal. In D. S. Jordan.

The fur seals and fur-seal islands of the North Pacific Ocean. Rcpt. of Fur-Seal

Invest. 1896-1897. Pt. 3, 1899. U. S. Govt. Ptg. Off., Washington, D. C., p. 99-177.
STRANDINE, E. J., 1943. Variation in Microphallus, a genus of trematodes, from fishes of Lake

Lelanau, Michigan. Trans. Amcr. Micr. Soc., 62: 293-300.
STUNKARD, H. W., 1951. Observations on the morphology and life-history of Microphallus

limuli n. sp. (Trematoda: Microphallidae). Biol. Bull.. 101 : 307-318.
STUNKARD, H. W., 1953. Natural hosts of Micmphalliis limuli Stunkard, 1951. /. Parasit.,

39 : 225.

THE PERMEABILITY OF THE SENSORY PEGS ON THE ANTENNAE
OF THE GRASSHOPPER (ORTHOPTERA, ACRIDIDAE)

ELEANOR H. SLIFER

Department of Zooloc/y, State University of loiva, /oic'O City, Iowa,
and the Marine Biological Laboratory, Woods Hole, Mass.

In 1906 Rdhler described three types of sense organs — pegs (Keg el, sensilla
basiconica), pit pegs (Grubenkegel, sensilla coeloconica) and bristles (Sinncsborsten,
sensilla chaetica) — which are present on the antennae of a grasshopper, Acrida
turrita (Linnaeus).1 Jannone (1940) states that the antenna of another species,
Dociostaurus maroccanus (Thunberg), is provided with the same kinds of sense
organs. He counted 125 pit pegs on the antenna of a first instar female and
found between 470 and 490 on the antenna of an adult female. Eiben (1949)
showed that similar structures occur on the antennae of Melanoplus differentialis
(Thomas). He recorded the number of each type present on the antennae of
each of the six nymphal instars and of the adults of this species and found that
there is a five-fold increase in the total number of these sense organs during post-
embryonic development. In addition to the sensory structures found by Rohler
others have been described on the surface and inside the antennae of grasshoppers
(Mclndoo, 1920: Eggers. 1924; Slifer, 1936; Jannone, 1940; McFarlane, 1953)
but these need not be considered here.

Of the three types of sense organs which were found by Rohler on the antenna
of the grasshopper the basiconic pegs are most numerous. Sensilla of this kind are
known to be present in many species of insects and are generally considered to be
chemoreceptors (Snodgrass, 1926, 1935 ; Frings and Frings, 1949; Roth and Willis,
1951 ; Hodgson, 1953 and others) but, as Dethier (1953) says of these and related
structures (p. 546) : "Nothing is known concerning the chemical or physical prop-
erties of the cuticle surmounting these receptors." Richards' (1952) studies on the
antennae of the honeybee have recently supplied some information on the properties
of the cuticle of fixed and sectioned sense organs. It is the purpose of the present
paper to show that in the living grasshopper the tip of certain of the sensory pegs is
permeable to aqueous solutions of a dye.

MATERIALS AND METHODS

The species of grasshoppers which were examined in the living condition repre-
sent the three major North American subfamilies (Acridinae, Oedipodinae and
Cyrtacanthacridinae) and included male and female Orphulclla pclidna (Bur-
meister), Dissosteira Carolina (Linnaeus), Psinidia fenestralis fenestralis (Ser-
ville), Melanoplus differentialis differentialis (Thomas), Melanoplus femur-rubrum
(DeGeer) and Melanoplus me.ricanus mexicanus (Saussure). Some of these were
raised in the laboratory and others were caught in the field. Newly-hatched nymphs
of Melanoplus mexicanus nie.vicanus were used in certain experiments and newly-
molted adults of several species in others. Preserved specimens of adult Acrida
bicolor (Thunberg),2 Dissosteira Carolina, Locusta rnigratoria migratorioides

1 Known to Rohler as Try.valis nasuta L.

2 The preserved specimens of Acrida bicolor and Locusta migratoria migratorioides were
kindly sent to the writer by Dr. B. P. Uvarov of the Anti-Locust Research Centre in London.

122

PERMEABILITY OF SENSORY PEGS 123

(Reiche and Fairmaire), Melanoplus femur-rubrum femur-rubrum and Melanoplus
uic.ricanus mexicanus were also studied.

Of several dyes tried a 0.5% aqueous solution of acid fuchsin was found to be es-
pecially useful. It is a vivid stain and can be detected when present in minute quan-
tities. When used as described below it has no toxic effects and nymphs immersed
in it for an hour recovered completely after removal from it. The method used to
demonstrate the penetration of the dye was simple. If the individual to be tested
was small its head was removed, wrapped in a bit of absorbent cotton and the whole
placed in the dye. For larger insects the antennae were severed at the base,
wrapped in cotton and immersed in the stain. The cotton prevents the specimen
from rising to the surface of the solution where it would, otherwise, float. Care
must be taken that no air bubbles are trapped in the cotton for they may prevent
the stain from reaching all parts of the antennae. After a suitable interval — a few
minutes to several hours — the head or antenna was removed and dipped very
rapidly, and in turn, into distilled water, 70% alcohol and absolute alcohol to wash
off stain which was clinging to the surface. The specimen was then placed in
n-butyl alcohol or dioxan where it was left for five minutes or longer depending
upon its size. Here the antennae were removed from the head if this had not been
done earlier. Toluol was used as a final clearing agent and the antennae were
mounted in a synthetic resin (Harleco H. S. R.) which was dissolved in toluol.
In using this method it is of the first importance that passage from the stain to
n-butyl alcohol or dioxan be very rapid for the fuchsin is lost quickly if there is any
delay. Dehydrating agents in which acid fuchsin is soluble must be avoided and
the same applies to clearing and mounting media. Before any other reagent is
substituted for one of those used here a sample should first be tested by adding a
small amount of the dry, powdered dye to it. Finally, it should be noted that special
difficulties will be encountered when this method is used for studying large struc-
tures which have much soft tissue or body fluid associated with them. The water
in this tissue or fluid may dilute and carry the stain away with it while the speci-
men is being dehydrated.

RESULTS AND DISCUSSION

The basiconic or peg-like sensilla on the antennae of the grasshoppers studied
may be subdivided into at least three kinds : 3 ( 1 ) long, slender pegs with a narrowly-
rounded tip (Figs. 1 to 7, a), (2) short, stout pegs with a broadly-rounded tip
(Figs. 1 to 7, b), and (3) short, slender pegs with a pointed tip (Figs. 1 to 7, c).
Of these only the first are permeable to acid fuchsin. The other two are unaffected
by the stain. No clue as to their function has been obtained, and they will not be
considered further here.

3 Snodgrass (1935, p. 519) discusses variations in basiconic sensory structures as follows:
"Sensory pegs and cones are innervated hairs reduced in size, and there is no sharply dividing
line between sensilla trichodea and sensilla basiconica, either in the character of the external
parts or in the structure of the internal parts. In a typical sensillum basiconicum the external
process is a small peglike or conical structure (Fig. 269 A, Pg}. The walls of the process are
thick or strongly sclerotic in some cases, while in others they are thin and transparent, or the
process may terminate in a delicate membranous cap."

124

ELEANOR H. SLIFER

The extent to which the stain penetrates the long, slender pegs which were
described in the preceding paragraph, depends largely, although not entirely, upon
the time of exposure. If the antenna is left for a short time in the dye only the
extreme tip of each peg is colored and examination with an oil immersion lens may
be necessary to detect the minute red spots. This indicates that the whole outer

B

I

FIGURE 1. Sensory structures from surface of the antenna of an adult female Mclanoplus
differcntialis diffcrentialis which, nineteen hours after the final molt, was treated for 30 minutes
with an aqueous solution of acid fuchsin. A, long, slender basiconic pegs which are permeable
to the dye at their tips ; stippled area shows extent of penetration of the stain during 30 minutes ;
B, short, stout basiconic pegs which are unaffected by the dye ; C, short, slender basiconic pegs
which are unaffected by the dye ; D, surface view of coeloconic peg ; small, brownish, oval mass
of unknown origin and identity which is commonly present in such pits shown at lower right ;
E, sensory bristle which is unaffected by stain. X 1100.

surface of the peg is waterproof except at the tip. Here the usual waxy or lipoid
layers of the cuticle must be missing. The inner, permeable layers of the cuticle
extend across the tip and there is no actual opening or pore. In antennae which
have been left longer in the stain the dye will be found to have traversed the permea-
ble cuticle at the tip and to have entered the central cavity or core of the peg. The
extent of this cavity can be seen in antennae which have been allowed to dry before

PERMEABILITY OF SENSORY PEGS

125

being mounted in resin, for the air-filled core of the peg then appears black under
the microscope. Since the cavity of the peg in the living insect contains either
fluid or cytoplasm, which extends up into it from the cellular layer below, the
passage of the dye is more rapid down the central core. At the same time, but a
little more slowly, diffusion occurs laterally from the core through the inner, cuticu-
lar layers of the peg. After very long exposures the entire peg is colored and the
dye may reach the interior of the antenna itself.

A B

2

A B

4

A

J

B

B

FIGURES 2 TO 5. Basiconic pegs from the surface of the flagellum of the antennae of adults
of four species of grasshoppers. Figure 2, male Acrida bicolor; Figure 3, male Orphulella
pelidna; Figure 4, female Dissosteira Carolina; Figure 5, male Locusta migratoria nrigratorioides.
A, long, slender peg which is permeable at the tip ; B, short, stout peg which is unaffected by
dye; C, short, slender peg which is unaffected by dye. X 1100.

The tips of the long pegs stain with great regularity and the rest of the surface
of the antenna shows no trace of the dye except in those regions where an obvious
injury has occurred. To eliminate all possibility that the tips of the pegs stained
because they had been worn or abraded tests were made with the antennae of adults
which had just molted. The results with these were the same as had been obtained
with older animals. On the antennae of a freshly-molted individual, where the
cuticular surface is still perfect, only the tips of the long, basiconic pegs are colored
by the dye.

These pegs are permeable to acid fuchsin in the newly-hatched grasshopper. To
establish this point nymphs of Melanoplus nic.vicanus me.vicanus which had left the
egg a few moments before and had just shed the cuticle with which they hatch were
treated with fuchsin. The tips of their long basiconic pegs stained brilliantly but

126

ELEANOR H. SLIFER

the entrance of the dye into the cavity of the peg was slower than it is in the adult.
It was interesting to find that these sensilla in the newly-hatched nymph are of ap-
proximately the same size as are those of the adult (Figs. 6 and 7) although, as
Jannone (1940) and Eiben (1949) have shown, antennal sense organs are much
fewer in number in the former. Why penetration of the dye should be slower in the
young insect is not known but the size of the colored area at the tip suggests that the
permeable surface is even smaller than it is in the older animals.

The bristles or sensilla chaetica described by Rohler for Acrida turrita are also
present on the antennae of the species studied here (Fig. l,e) but they are few in
number and are located only on the proximal segments. Such bristles are usually
believed to have a tactile function. They are entirely unaffected by the dye in
newly-molted individuals, but these long bristles are often found to be damaged
in older animals and the stain then enters rapidly through the broken end.

B

B

FIGURES 6 AND 7. Basiconic pegs from surface of flagellum of antenna of newly-hatched and
of adult Melanoplus mexicanus mexicanus. Figure 6, newly-hatched; Figure 7, adult male.
A, long, slender peg which is permeable to dye ; B, short, stout peg which is unaffected by dye ;
C, short, slender peg which is unaffected by dye. X 1100.

As Rohler reported in 1906 another type of sense organ present on the grass-
hopper antenna is the pit peg or sensillum coeloconicum. There is considerable
evidence that coeloconic sensilla in other insects are chemoreceptors. In these sense
organs, in the grasshopper, the peg is located at the bottom of a small, globular pit
which opens to the surface through a still smaller hole (Fig. 1, d). The cavity,
in life, is filled with air and because of this it has not been possible to demonstrate,
with the method outlined above, that the tip of the peg is permeable. When the
antenna is immersed in water or in an aqueous solution of acid fuchsin the pits re-
main filled with air and this prevents the fluid from coming into contact with the
tip of the peg. The air bubbles in the pits are easily seen under the microscope.
No method for removing these bubbles, which may not be suspected of damaging
or altering the tip, has yet been devised. Attempts to remove the bubbles with a
vacuum pump and with a detergent solution were made but the results were not
satisfactory. It is possible that dye might be placed in the pit with the aid of a
microdissection syringe but this was not tried. Antennae from animals which have
been fixed in Bouin's solution and preserved in 70% alcohol provided some informa-
tion concerning the permeability of the pegs. In these the pits have filled with

PERMEABILITY OF SENSORY PEGS 127

alcohol and when such antennae are placed in fuchsin, as were the living antennae,
the dye replaces the alcohol in the pits and the tips of the pegs then take up the stain
just as do the tips of the long, basiconic pegs of the same specimen. It will be
noticed in such preparations that the stain penetrates the latter more rapidly than
it does in fresh material but the manner of entry is the same. From these observa-
tions we may conclude that the tips of the coeloconic pegs also differ from the gen-
eral cuticular surface in respect to permeability and that it is probable, although, of
course, not yet proved, that they, too, would be permeable to an aqueous solution
of fuchsin in the living condition were it possible to bring the dye into contact with
them.

Since the relatively large molecule of acid fuchsin penetrates the tip of the living
basiconic peg so readily there can be no doubt that this region is also permeable to
water and it is highly probable that a great variety of other substances could also
be shown to pass through it if methods suitable for their detection were applied.
The results reported here, then, strongly support the conclusions of many previous
investigators who, on other grounds, and with other species of insects, have believed
certain basiconic pegs to be chemoreceptors, hygroreceptors or both. The results
obtained with fixed material suggest that the coeloconic pegs may have a similar
function or functions although the evidence is less reliable than it is for the long,
basiconic pegs.

Preliminary examinations of other parts of the body of the grasshopper have
shown that long, slender pegs of the type present on the antenna occur also in other
regions, although more sparsely, and that they, too, have a permeable tip.

Whether insects other than grasshoppers also possess basiconic pegs which are
permeable to water and dyes is not known with certainty at present. A few adults
belonging to other orders (Collembola, Thysanura, Dermaptera, Isoptera, Neu-
roptera, Coleoptera, Hymenoptera and Diptera) were tested with interesting but
inconclusive results. In some specimens definite and regular staining occurred but
since only a few tests were made and since the individuals used were of unknown
age and past history it is possible that the tips of the sensilla which stained may
have previously been damaged. For critical work the animals should be freshly-
molted or, at least, known never to have been in contact with a surface or object
which might injure the tips of the pegs. These cursory tests, however, brought
out several of the difficulties which may be encountered when insects other than
grasshoppers are studied. In some the covering of long, close-set hairs retains a
film of air which prevents the stain from reaching the pegs even though the speci-
men is wrapped in cotton which is covered by the dye solution. In others the
cuticle is heavily pigmented and it is impossible to decide whether or not any stain-
ing has occurred. In still other individuals faint staining was apparent only after
many hours exposure to the dye which suggests that the rate of penetration must
be extremely slow in these forms. In certain species large, thin-walled pegs were
found barely tinged with pink. Here, seemingly, small amounts of the dye had
penetrated and then diffused through the fluid contents of the peg. Finally, it
should be emphasized that failure to stain with acid fuchsin does not mean that the
structure tested is impermeable to all materials. It may still be permeable to water
and to substances other than the dye used here. Clear-cut, positive results, such
as are given by the long, slender pegs of the grasshopper antenna, lend very strong
support to the idea that these structures, in this insect, serve as chemoreceptors, as

128 ELEANOR H. SLIFER

hygroreceptors, or, perhaps, as both. Negative results, on the other hand, prove
only that, under the conditions of a particular experiment, the structure tested is
either impermeable to acid fuchsin in detectable amounts or that any dye which did
penetrate was later lost.

SUMMARY

1. When an aqueous solution of acid fuchsin is applied to the surface of the
living antenna of a grasshopper the dye enters the tips of the largest of three types
of basiconic sense organs while the other two types are unaffected.

2. The permeability to water and to dye of these long basiconic pegs on the
antenna of the living grasshopper strongly supports the conclusions of earlier
workers with other insects that such structures may serve as chemoreceptors, as
hygroreceptors, or as both.

3. The permeability to water and to dye of the pegs of the coeloconic sense
organs on the surface of the antennae of preserved grasshoppers suggests that these,
too, may function as chemoreceptors, as hygroreceptors, or as both but the evidence
is less satisfactory than it is for the long basiconic pegs.

4. Positive results with the staining method described in the present paper indi-
cate that the structure tested is permeable to acid fuchsin and to water and, probably,
to many other substances as well. Negative results mean either (1) that the struc-
ure is completely or nearly impermeable to acid fuchsin or (2) that any of the stain
which did enter was lost in later handling.

LITERATURE CITED

DETHIER, V. G., 1953. Chapter on Chemoreception. Insect physiology, edited by K. D. Roeder,
John Wiley & Sons, New York.

EGGERS, F., 1924. Zur Kenntnis der antennalen stiftfiihrenden Sinnesorgane der Insekten.
Zcitschr, Morph. und Okol. Tierc, 2 : 259-349.

EIBEN, C. H., 1949. A study of the sense organs on the surface of the antennae of the grass-
hopper, Mclanoplus differ cntialis, during post-embryonic development. Unpublished
Master's Thesis, State University of Iowa.

FRINGS, H., AND M. FRINGS, 1949. The loci of contact chemoreceptors in insects — A review
with new evidence. - Amer. Mid. Nat., 41 : 602-658.

HODGSON, E. S., 1953. A study of chemoreception in aqueous and gas phases. Biol. Bull., 105 :
115-127.

JANNONE, G., 1940. Studio morfologico, anatomico e istologico del Dociostaurus m-aroccanus
(Thumb.) nelle sue fasi transiens congregans, gregaria e solitaria. Boll. R. Lab. Entom.
Agraria di Portici, 4 : 1-443.

MCFARLANE, J. E., 1953. The morphology of the chordotonal organs of the antennae, mouth-
parts and legs of the lesser migratory grasshopper, Melanoplus mcxicanus mexicanus
(Saussure). Can. Ent., 85: 81-102.

MclNDOO, N. E., 1920. The olfactory sense of Orthoptera. /. Comp. Ncuro., 31 : 405-427.

RICHARDS, A. G., 1952. Studies on arthropod cuticle. VIII. The antennal cuticle of honeybees,
with particular reference to the sense plates. Biol. Bull., 103 : 201-225.

ROHLER, E., 1906. Beitrage zur Kenntnis der Sinnesorgane der Insekten. Zoo/. Jahrb., Abt.
Anat. u. Onto., 22 : 225-288.

ROTH, L. M., AND E. R. WILLIS, 1951. Hygroreceptors in Coleoptera. /. Exp. Zool, 117: 451-
487.

SLIFER, E. H., 1936. The scoloparia of Mclanoplus differ entialis (Orthoptera, Acrididae).
Ent. News, 47 : 174-180.

SNODGRASS, R. E., 1926. The morphology of insect sense organs and the sensory nervous sys-
tem. Smithsonian Misc. Coll, 77 (8) : 1-80.

SNODGRASS, R. E., 1935. Principles of insect morphology. McGraw-Hill Book Co., Inc., New-
York.

THE LOCALIZATION OF HEPARIN-LIKE BLOOD ANTICOAGULANT
SUBSTANCES IN THE TISSUES OF SPISULA SOLIDISSIMA

LYELL J. THOMAS, JR.1

Marine Biological Laboratory, Woods Hole, Mass., and Department of Zoology,
University of Pennsylvania, Philadelphia, Pa.

As previously reported (Thomas, 1951) a potent blood anticoagulant resembling'
heparin can be extracted from the common surf clam Spisula (Mactra) solidissima.
This has recently been confirmed by Frommhagen et al. (1953). They attempted
to develop the Spisula anticoagulant for clinical use.

The discovery of the Spisula anticoagulant stemmed from previous studies by
Heilbrunn and his students concerning the biological significance of heparin and
related substances. There is an increasing amount of evidence that such substances
may be of rather general importance. Thus in the monograph by Jorpes (1946)
there are numerous references indicating that in addition to acting as a blood anti-
coagulant heparin may serve other functions. It is known, for example, that heparin
will inhibit growth in tissue culture and that it can inhibit the action of various
enzymes. Heparin is chemically related to the sulfated polysaccharides found inter-
cellularly in the connective tissues and in mucus secretions so that functions served
by these latter substances are also of interest when considering the significance of
heparin-like substances. As will be discussed later there is evidence that chondroitin
sulfate may be of importance in calcification.

As compared with the number of investigations concerning heparin and related
substances in mammals the number of similar investigations on adult invertebrate
animals have been relatively few. However there is evidence that these substances
may be of major importance to the eggs of invertebrate animals. Thus it is known
that the jelly coat of sea urchin eggs contains a highly sulfated polysaccharide capable
of preventing the clotting of blood. The significance of this fact and the general
importance of the jelly coat to fertilization have been discussed by Runnstrom
(1952). Polysaccharide sulfate esters seem to be of significance for reactions in
the protoplasm of egg cells as well as for reactions at the surface. Thus Heilbrunn
and Wilson (1949) found that heparin seems to inhibit the protoplasmic gelations
which normally occur during division of the Chaetopterus egg. Also, as shown
by Kelly (1953), not only the jelly coat but ako certain elements in the protoplasm
of some marine eggs show the metachromatic staining reaction for polysaccharide
sulfate esters. The metachromatic reaction is a shift in color caused by the poly-
merization of certain basic dyes such as toluidine blue (Michaelis, 1947) and is
often produced when the dye combines with large negatively charged molecules.
Heparin and other highly sulfated polysaccharides produce a very intense red
metachromatic color with toluidine blue.

From the histological investigations of Kelly it was known that the eggs and
ovaries of Spisula exhibit strong metachromatic staining and, as will be shown in
this paper, a metachromatic blood anticoagulant substance can be isolated from

1 Present address : Department of Pharmacology, Woman's Medical College, Philadelphia
29, Pa.

129

130 LYELL J. THOMAS, JR.

Spisula eggs. However it was observed that other tissues of this clam also had
a very high affinity for toluidine blue and that breis of the tissues showed an im-
pressive metachromasia in vitro. Thus it seemed desirable to isolate the meta-
chromatic substance in order to determine some of its properties. Also an attempt
has been made to discover the origin of the substance within the tissues. From the
work of Soda and Egami (1938) it was known that a heparin-like anticoagulant
can be obtained from the mucous secretions of Charonia lampas, a marine gastropod.
Thus it was suspected that the anticoagulant from Spisula might also be of mucous
origin. In part this seems to be true. However, blood clotting assays made on
extracts of various portions of the clam have revealed that in addition to an anti-
coagulant of mucous origin another anticoagulant substance is present in the tissues
of Spisula. From histological examination it appears that this latter substance is
an intercellular material possibly analogous to the chondroitin sulfate of mammalian
connective tissues. As pointed out in the discussion there is a possibility that such
substances may be of importance in calcification processes.

ISOLATION OF THE SPISULA ANTICOAGULANT BY MEANS OF A HEPARIN

EXTRACTION PROCEDURE

The clams (Spisula) were obtained from commercial fishermen along the New
Jersey coast and at Woods Hole. At first, anticoagulant preparations were made
from thoroughly washed clam meat containing all the organs except the viscera and
shell. The visceral mass was removed because of complications caused by the
gonadal material. The presence of sperm rendered it difficult to extract the anti-
coagulant. This was probably due to the basic proteins of the sperm which are
known to precipitate heparin-like substances. On the other hand the viscera from
female clams did not seem to contain any anticoagulant other than that which could
be ascribed to the ovaries and eggs.

After grinding the clam meat several times in a food chopper, extraction and
purification of the anticoagulant were carried out according to the procedure for
preparing beef lung heparin described by Homan and Lens (1948). Thus the
chopped tissue was allowed to autolyze for 24 hours and then was extracted for an
hour with a warm alkaline buffer containing half normal sodium hydroxide and
enough ammonium sulfate to maintain a pH of 9 or 10. The supernate was then
heated to 70° C.-800 C. to denature protein, filtered, and the crude anticoagulant
was precipitated together with protein by acidification to pH 2-3. The precipitate
after extraction with alcohol was digested with trypsin. Some impurities could be
removed from the digest by adding ammonium carbonate, centrifuging, and then
boiling at pH 7 followed by another centrifugation. The active material was then
precipitated by two volumes of alcohol and subjected to a partition between water
and neutralized phenol. This left the active material in the aqueous phase whereas
most of the remaining proteinaceous impurities entered the phenol phase. The
product was obtained from the aqueous phase by adding NaCl and two volumes of
alcohol. Further impurities could be removed with lead acetate at pH 5 and the
lead removed with an excess of sodium carbonate. The anticoagulant was then
obtained as the sodium salt by precipitation with a large excess of acetic acid fol-
lowed by neutralization in an alcohol-ether mixture.

HEPARIN-LIKE SUBSTAXCES FROM SPISULA 131

Anticoagulant tests were performed by the thrombin method of Jaques and
Charles (1941), with the modification that instead of beef blood citrated sheep
plasma was used. The anticoagulant activities of all clam preparations were com-
pared to commercial samples of sodium heparin (generously supplied by the Upjohn
Co., Kalamazoo, Mich.) rated in U.S. P. units by the manufacturer. These values
ranged from 120 to 156 U.S. P. units/mg.

In all, three anticoagulant preparations were made from eviscerated clams as
described above. Following the stage of phenol partition, the average anticoagulant
activity of these preparations was 40 heparin units per milligram. By further
purifying one of these preparations with lead acetate and then forming the sodium
salt, a product with an activity of 50 heparin units per milligram was obtained.
The yield obtained from these initial preparations was about 11,000 units per kilo-
gram of starting material. The preparations that were carried only through the
stage of phenol partition gave weak positive biuret and ninhydrin reactions in-
dicating traces of protein material, but the lead-purified product seemed to be
protein-free. All samples gave a positive color reaction for hexosamine (Palmer
et al., 1937) and a barium precipitate after hydrolysis with HC1 indicated the pres-
ence of ester sulfate. In view of later findings it would appear that both the yield
and activity from these initial preparations were rather low. Some of the probable
reasons for this are discussed below.

The most active product obtained thus far was a lead acetate-purified sodium
salt of the anticoagulant from Spisula mantle tissue. It had an activity of 130
heparin units per milligram which is equivalent to the activity of mammalian sodium
heparinate. The yield of this final product from one kilogram of mantle was about
19,000 units which is equal to the best yields reported for beef lung heparin
(Kuizenga and Spaulding, 1943). However, as will be discussed below, even this
yield apparently represents only a fraction of the total anticoagulant substance in
the Spisula mantle tissue. The final product was a white powder readily soluble
in water, which seemed to be free of lead and proteins. After acid hydrolysis it
gave positive tests for reducing sugar, hexosamine and ester sulfate. But it was
interesting to note that the hexosamine color and barium sulfate precipitate ap-
peared from qualitative examination to be only one half to two thirds as great as
the same reactions given by beef lung heparin of the same anticoagulant strength.

The mantle tissue anticoagulant discussed above was purified from an alkaline
extract in the same manner as the previous preparations. But in an effort to ex-
tract as much anticoagulant as possible, the mantle tissue was homogenized in a
Waring Blendor and allowed to autolyze for a relatively long period (48 hours).
It was then subjected to prolonged extraction (6 hours) with the alkaline sodium
hydroxide-ammonium sulfate mixture. However, even after this more thorough
extraction it was found that a large amount of the anticoagulant substance remained
in the tissue. After extraction, the residue was washed with several changes of
water and then with alcohol and ether. When stained with toluidine blue the
residue still showed a strong metachromatic color and, after digestion of the residue
with trypsin, the metachromatic material appeared in solution. The digest was
boiled and centrifuged, then tested for anticoagulant activity, whereupon it was
discovered that the digested residue yielded as much anticoagulant activity as the
crude product from the alkaline extract. The alkaline extract and the digested
residue each yielded about 30,000 units per kilogram of original tissue, thus in-

132 LYELL J. THOMAS, JR.

dicating a total potential yield of 60,000 units or more per kilogram of mantle tissue.
It would seem, then, that alkaline extraction is not a very efficient method of re-
moving the anticoagulant from Spisula tissues, and in subsequent work tryptic
digestion was used as the method of extraction.

Since alkaline extraction did not remove all of the anticoagulant from the tissue
it seemed wise to investigate other aspects of the procedure. By assaying tryptic
digests of fresh and autolyzed mantle tissue it was found that no change in the total
yield of anticoagulant was produced by autolysis. However more anticoagulant
could be extracted with alkali after autolysis and the crude product obtained in this
way was more active than when autolysis was omitted. This increased activity
was probably due to the preferential destruction of contaminating substances during
autolysis. Another factor of importance in obtaining a high final yield is that the
active material precipitates rather slowly out of alcoholic solution after the removal
of proteins. Even after the addition of salt to the alcoholic solution it was usually
necessary to let the mixture stand for 24 hours or more. Centrifugation prior to
this time often left some of the active material dispersed in the supernate. Partition
of the crude product between neutralized phenol and water was a very effective step
in purification if repeated two or three times. Provided that there was a clean
separation of the two phases, none of the active material was found in the phenol
phase. Most of the protein impurities following tryptic digestion could be removed
in this manner.

LOCALIZATION OF ANTICOAGULANT SUBSTANCES WITHIN THE TISSUES OF SPISULA

Extracts obtained from various organs and tissues were assayed for anticoagulant
activity and these data were compared with results obtained from metachromatic
staining of tissue sections and from in vitro observations on metachromasia.

For the histological localization of acid polysaccharides in the tissues, paraffin
sections of formalin-fixed material were stained according to the method of Sylven
as described by Click (1949). Staining was done with toluidine blue in 30 per cent
alcohol followed by destaining in 95 per cent alcohol. For the assay of total anti-
coagulant content of various organs and tissues the following method was adopted.
A small quantity of tissue was extracted with alcohol and ether to remove lipids,
and the extracted tissue was then digested with trypsin (Difco 1:250). Usually
an amount of trypsin equivalent to one tenth the extracted dry weight of the tissue
was used. After 24 hours digestion (35° C., pH 8) the digest was boiled and in-
soluble material centrifuged down. The supernate, made to a known volume and
representing a known weight of tissue, was then assayed for anticoagulant activity.
Table I represents the average anticoagulant activities per wet weight of tissue for
two series of digests made in the above manner. It seems reasonably certain that
these values represent activity due only to the heparin-like substances in the tissue.
Preliminary trials indicated that the anticoagulant action of digests of this type
could be abolished entirely by adding toluidine blue. Furthermore, no loss of
activity resulted from dialysis, phenol partition, or from precipitation with two
volumes of alcohol if these steps were carried out carefully. It was also found that
such digests had no effect on the clotting time of a purified fibrinogen-thrombin
clotting system. This indicates that a co-factor is probably necessary for the action
of the anticoagulant from Spisula as is known to be the case for heparin (Chargaff,
Ziff and Moore, 1941).

HEPARIN-LIKE SUBSTANCES FROM SPISULA 133

As can be seen from Table I the eggs of Spisula appear to contain a relatively
large amount of anticoagulant substance. A partially purified sample of this ma-
terial was obtained from a tryptic digest of eggs previously extracted with alcohol and
ether to remove lipids. After removal of some of the impurities by phenol parti-
tion, the active material was precipitated from the aqueous phase with alcohol.
The material so obtained had an anticoagulant activity of about 20 heparin units
per milligram. A large part of this anticoagulant from the egg no doubt was derived
from the jelly coat, but some may also have been derived from elements in the
protoplasm. As discussed below, not only does the jelly coat stain metachroma-
tically but also there are regions in the interior of the egg that give a metachromatic
color with toluidine blue. At the time the assays listed in Table I were performed,
attempts to obtain anticoagulant jelly coat solutions by the acid sea water treatment
of Vasseur (1947) proved unsuccessful. However this problem is being re-investi-
gated and it now seems probable that the eggs used at that time, although fertilizable,
were immature. Recent preliminary results indicate that although a thin jelly
coat is present on immature Spisula eggs it is not easily removed by acid sea water,

TABLE I

Anticoagulant activity of various Spisula tissues

Heparin units

per gram tissue

wet weight

Mantle edge inner fold 180

Mantle edge outer fold 130

Gills 140

Palps 160

Eggs 100

"Skin" of foot 45

Foot devoid of "skin" 30

Adductor muscle 20

but can be removed by treating the eggs with 3% NaCl containing 0.1 M Versene
(generously supplied by the Bersworth Chemical Co., Framingham, Mass.) at pH 8.
On the other hand, some of the jelly coat from ripe eggs seems to be rather easily
removed by acidified (pH 3.5-4) sea water.

Recently, Spisula eggs - fixed in Zenker-formal fluid and sectioned at three
microns have been stained with toluidine blue (0.1% toluidine blue in 30% alcohol
with destaining in 95% alcohol) after extraction with hot 4 per cent trichloroacetic
acid (TCA) according to the method of Monne and Harde (1951). Presumably
this extraction removes the nucleic acids. The jelly coat of both the extracted and
unextracted sectioned eggs exhibited brilliant red metachromatic staining. Un-
extracted eggs showed an intense blue to purple color in the cytoplasm and a purple
metachromasia in the nucleolus. The main bulk of the germinal vesicle was prac-
tically unstained. After TCA extraction the cytoplasmic staining was reduced to
a pale blue except for a diffuse red metachromasia in the outermost region of the
cortex. The cortical granules which had previously been obscured by the strong
cytoplasmic staining were very prominent after TCA extraction. These were seen
to stain with an intense blue color. The staining of the nucleolus was not changed

2 I wish to thank Dr. W. S. Vincent, Department of Anatomy, Syracuse Medical Center,
Syracuse, New York, for the sectioned eggs.

134 LYELL J. THOMAS, JR.

appreciably by TCA extraction except that the metachromatic red color was more
prominent. A more thorough study of the Spisula egg is planned and will be re-
ported at a later date. It would be particularly interesting to determine more
precisely whether the metachromatic staining of the nucleolus and cortical region is
due to the presence of sulfated polysaccharides. In a preliminary report Allen
(1951) mentions that mucopolysaccharide (as determined by the method of Monne
and Slautterbach, 1950) appears to be transferred from the nucleolinus to the
spindle during cleavage of the Spisula egg.

Returning now to a discussion of the adult clam, it will be observed (Table I)
that digests of the mantle, gills and palps had five or six times more anticoagulant
activity than digests of the foot and muscles. Histological sections of the mantle
edge revealed several regions of metachromasia. The mucous cells of the inner and
outer folds (adjacent to the shell) of the mantle edge are both metachromatic. How-
ever it was observed that the mucous cells of these two folds are distinctly different
with respect to intensity of staining reaction and with respect to size and shape.
Thus the mucous cells of the inner mantle surface are smaller and take the meta-
chromatic color much more intensely than the mucous cells adjacent to the shell.
No mucous cells were observed on the middle fold of the mantle edge. The mucosal
basement membranes were stained a brilliant red as were certain areas of connective
tissue in the interior of the mantle. Distally in the mantle folds this connective tissue
is a dense compact material resembling cartilage. The histological picture for the
gills and palps resembles that of the mantle. Thus the basement membranes and
interior connective tissue of these structures also are metachromatic. The mucous
cells lining the palps and gill filaments resemble the mucous cells lining the inner
surface of the mantle. Probably these mucous cells, peculiar to the mantle cavity,
provide the sticky secretion which aids in the collection of food particles. The
secretion of the large mucous cells of the outer mantle fold possibly enters into the
composition of the shell matrix.

Very little metachromasia was observed in cross sections of the foot and none
could be detected in' the muscle tissues. The foot was seen to contain large mucous
cells in the perimeter. These were stained a pronounced blue color with toluidine
blue but little, if any, of the metachromatic red color was evident. The mucosal
basement membrane in the foot, however, stained metachromatically.

The known presence of a heparin-like substance in gastropod mucus suggested
that the anticoagulant from Spisula might be of mucous origin. Also the intense
metachromasia exhibited by the mucous cells lining the gills, palps and inner mantle
surface suggested that the mucus from these cells might contain a highly sulfated
heparin-like polysaccharide. However the intensity of the metachromatic staining
observed in the connective tissues of the mantle and elsewhere suggested that there
might be another heparin-like substance in the connective tissues. In order to de-
termine this, mucosal scrapings were taken from both sides of the mantle edge and
the three folds of the mantle edge were separated from one another. During these
operations special precautions were taken to insure that no mucous contamination was
transferred from one part of the mantle edge to another. After extraction of the
tissue and mucus fractions with alcohol and ether, tryptic digests were made of these
fractions, as described previously, and the digests were assayed for anticoagulant
activity. In Table II are given the results of these assays together with the dry
weights of the starting material after extraction of lipids. The numbers at the far

HEPARIN-LIKE SUBSTANCES FROM SPISULA

135

left of the table identify the different types of data obtained from a given series of
digests. To obtain the metachromatic ratios given in Table II a series of tubes,
each containing the same amount of toluidine blue with phosphate buffer (pH 6.6)
plus increasing concentrations of digest, was matched in a comparator block with
an identical series of tubes containing buffered toluidine blue and heparin. The
ratios are the number of anticoagulant units of heparin required to produce a given
color, divided by the number of anticoagulant units of digest required to produce the
same color. Thus, for example, if a digest had a metachromatic ratio of three, only
one-third of an anticoagulant unit of digest would be required to produce the same
mixture of red and blue color with a given amount of toluidine blue as was produced
by one anticoagulant unit of heparin.

TABLE II

•s

Anticoagulant activities and metachromatic ratios for various portions of the mantle edge

A. Heparin units per mg. dry weight

Remainder of mantle edge after scraping

edge mucosa

edge mucosa

Whole mantle

Inside mantle

Outside mantle

Middle mantle

edge

fold

fold

fold

1) 1.44

0.28

0.80

2) 1.30

0.07

0.68

3) 0.98

0.08

0.84

0.64

4)

0.80

0.72

0.93

B. Total weight of mantle edge and mucosal scrapings in milligrams

1) 44
2) 32

60
48

2,480
1,700

C. Metachromatic ratio of extracts (see text)

1) 5.6

not meta-

0.8

2) 5.2

chromatic

1.0

3) 4.4

1.6

1.0

1.2

It will be noted that the tryptic digests of mucus from the outside mantle fold
had little if any anticoagulant activity. Although this mucus was metachromatic
in tissue sections, tryptic digests of this mucus were not metachromatic. On the
other hand, the digested mucus from the inner mantle fold had anticoagulant ac-
tivity and was highly metachromatic. In fact, it was about five times as meta-
chromatic per anticoagulant unit as the heparin standard. Apparently this mucous
substance has the ability to bind toluidine blue very strongly in comparison with its
ability to prevent blood clotting. In digests of mantle edge tissue after scraping off
the mucosas the metachromasia was about equal to that of heparin.

The data in Table II clearly indicate that another anticoagulant factor is present
in the mantle edge tissue besides that originating from the mucous secretions. In the
first place, digests of the outer mantle fold had nearly as much anticoagulant activity
as those of the inner mantle fold, in spite of the fact that the mucus of the outer fold

136

LYELL J. THOMAS, JR.

was nearly inactive. Quite possibly the slight activity that was associated with this
mucus can be ascribed to the small amount of tissue scraped off with the mucus.
It will be observed that the middle fold had a high anticoagulant activity but, as
stated previously, the middle fold of the Spisula mantle edge is apparently devoid
of mucous cells. It is true that the highest anticoagulant activity per weight of
starting material was given by the mucosal scrapings from the inside mantle fold.
However, these scrapings had less than twice the potency of the scraped mantle
edge and weighed only about one fiftieth as much. Furthermore, histological
examination of scraped mantle edges showed that more than half of the mucosa
had been removed. This means then that only about one twenty-fifth of the total
anticoagulant activity of the mantle edge can be accounted for by the inner mucosa.
Thus the only other obvious source of heparin-like anticoagulant in the Spisula
mantle edge is the strongly metachromatic substance in the connective tissue.

Since the middle fold of the mantle is very rich in metachromatic connective
tissue substance but has few if any mucous cells, it seemed desirable to isolate a
sample of anticoagulant exclusively from this portion of the mantle. The middle
fold was cut from mantle edge of several clams and thoroughly cleaned of debris.

TABLE III
Anticoagulant from middle fold of Spisula mantle edge

Tissue wgt., milligrams

Product wgt.,
milligrams

Heparin units per milligram

Wet

Dry

Product

Dry .tissue

4,950

830

14

45

0.75

After alcohol-ether extraction and tryptic digestion of the tissue, the digest was
boiled and centrifuged to remove impurities. Following this the active material
was precipitated from solution with two volumes of alcohol and redissolved in
water. This solution was then shaken out with two changes of phenol neutralized
with ammonium hydroxide. After acidifying the aqueous supernate to pH 4 or 5,
NaCl was added to a concentration of 1 % and two volumes of alcohol were added.
The solution was heated to 50° C. and allowed to stand for 24 hours, after which
time the active precipitate adhered firmly to the vessel. This material was washed
with an alcohol-ether mixture, then taken up in a small amount of water and allowed
to dry in a weighing bottle over CaCL. After weighing, this material was made to a
known concentration and assayed for anticoagulant activity. The results are given
in Table III. Wet weight of the starting material refers to the fresh tissue and dry
weight refers to this tissue after extraction with alcohol and ether. The heparin
units per milligram dry tissue refer to the amount of anticoagulant recovered in
the product. The final product (Table III) was quite metachromatic and as can
be seen its anticoagulant activity was almost half that of genuine heparin. Un-
doubtedly further purification would increase the activity. The amount of anti-
coagulant recovered per milligram of dry starting material is nearly as much as
was indicated from the assay of crude digests in Table II.

HEPARIX-LIKE SUBSTANCES FROM SPISULA 137

DISCUSSION

Probably the anticoagulant substance from the middle fold of Spisula mantle
is identical with the metachromatic substance detected histologically in the connec-
tive tissues. This substance, although most abundant in the mantle and associated
structures, is apparently present in other parts of the clam. Thus the basement
membrane in the foot mucosa was quite metachromatic. No metachromasia was
detected histologically in certain other tissues such as the adductor muscle, but
digests of adductor muscle were slightly metachromatic corresponding to the low
anticoagulant activity of these digests.

It seems possible that this connective tissue substance serves much the same
function in the clam as does chondroitin sulfate in the tissues of mammals. Thus
it is interesting that chondroitin sulfate has been implicated in calcification proc-
esses. Neuman and co-workers (1952) have shown that the chondroitin sulfate
in cartilage acts as a cation exchange resin. Also Miller, Waldman and McLean
(1952) found that toluidine blue and other basic dyes which have a high affinity for
polysaccharide sulfate esters can prevent the in vitro calcification of hypertrophic
cartilage. Apparently this inhibition is reversible. As they point out, there are
various interesting correlations between metachromatic staining coincident with
calcification. Thus Rubin and Howard (1950), for example, found that the meta-
chromatic staining of growing bones is most intense in those regions about to calcify.

In view of the apparent relationship between calcification and acid polysac-
charides in mammalian tissues, the recent paper by Bevelander (1952) was read
with considerable interest. From his radioautograph showing distribution of Ca45
in the mantle edge of Anodonta it would appear that calcium is taken up in certain
regions of the mantle edge connective tissues. Similar regions in the Spisula mantle
edge appear to contain an abundance of acid polysaccharide. A further investiga-
tion is planned to determine in what way mucopolysaccharides could be involved in
the calcification of molluscs.

I wish to express my sincere gratitude to Dr. L. V. Heilbrunn for his encourage-
ment and advice during this investigation.

SUMMARY

1. A heparin-like blood anticoagulant has been isolated from the surf clam
Spisula solidissiina.

2. The most potent preparation obtained had an anticoagulant activity of 130
U.S. P. heparin units per milligram and was derived from mantle tissue.

3. Anticoagulant assays made on tryptic digests from various portions of the
clam revealed that the mantle, gills and palps had about five times more anticoagu-
lant activity per gram of tissue than the foot and adductor muscles. It was also
found that the eggs of Spisula yield a high anticoagulant activity. —'

4. Toluidine blue staining of sectioned eggs revealed metachromasia in the jelly
coat, cortical region and nucleolus.

5. In order to determine the origin of the anticoagulant from the adult clam,
the results from anticoagulant assays made on digests of isolated portions of the
mantle edge were compared with histologic observations concerning metachromasia.
It was concluded that at least two substances with heparin activity are present in
the mantle tissue. One (jt these substances is present in the mucus secretion of

138 LYELL J. THOMAS, JR.

the inner mantle fold and apparently also in the mucus secretions of the palps
and gills. The mucus secretion of the outer mantle fold was nearly devoid of
anticoagulant activity. The other anticoagulant substance seems to be an inter-
cellular material in the connective tissues, possibly analogous to chondroitin sulfate.
This substance, also seems to be most abundant in the mantle, palps and gills.

6. The possibility that polysaccharide sulfate esters may be important in calcifica-
tion processes is discussed.

LITERATURE CITED

ALLEN, R. D., 1951. The role of the nucleolus in spindle formation. Biol. Bull., 101 : 214.
BEVELANDER, G., 1952. Calcification in molluscs. III. Intake and deposition of Ca45 and P32

in relation to shell formation. Bio!. Bull, 102 : 9-15.

CHARGAFF, E., M. ZIFF AND D. H. MOORE, 1941. Studies on the chemistry of blood coagula-
tion. XII. An electrophoretic study of the effect of anticoagulants on human plasma

proteins with remarks on the separation of heparin complement. /. Biol. Chem., 139 :

383-405.
FROMMHAGEN, L. H., M. J. FAHRENBACH, J. A. BROCKMAN, JR. AND E. L. R. STOKESTAD, 1953.

Heparin-like anticoagulants from Mollusca. Proc . Soc. Exp. Biol. Med., 82 : 280-283.
CLICK, D., 1949. Techniques of histo- and cytochemistry. Interscience Publishers Inc., New

York.
HEILBRUNN, L. V., AND W. L. WILSON, 1949. The effect of heparin on cell division. Proc.

Soc. Exp. Biol. Med., 70 : 179-182.
HOMAN, J. D. H., AND J. LENS, 1948. A simple method for the purification of heparin. Biochim.

et Biophys. Ada, 2 : 333-336.
JAQUES, L. B., AND A. F. CHARLES, 1941. The assay of heparin. Quart. J. Pharm. Pharmacol.f

14: 1-15.

JORPES, E., 1946. Heparin in the treatment of thrombosis. Oxford Univ. Press, London.
KELLY, J. W., 1953. Metachromasy in the eggs of fifteen lower animals. Protoplasma (in

press).
KUIZENGA, M. H., AND L. B. SpAULDiNG, 1943. The preparation of the highly active barium

salt of heparin and its fractionation into two chemically and biologically different con-
stituents. /. Biol. Chem., 148 : 641-647.
MICHAELIS, L., 1947. The nature of interaction of nucleic acids and nuclei with basic dyestuffs.

Cold Spring Harbor Symp. Quant. Biol, 12 : 131-142.
MILLER, Z., B. J. WALDMAN AND F. C. MCLEAN, 1952. The effect of dyes on the calcification of

hypertrophic rachitic cartilage in vitro. J. Exp. Med., 95: 497-508.
MONNE, L., AND S. HARDE, 1951. On the cortical granules of the sea urchin egg. Ark. Zool.

[2] 1 : 487-497.
MONNE, L., AND D. B. SLAUTTERBACK, 1950. Differential staining of various polysaccharides in

sea urchin eggs. Exp. Cell. Res., 1 : 477-491.
NEUMAN, W. F., E. S. BOYD AND I. FELDMAN, 1952. The ion binding properties of cartilage.

Metabolic Interrelations, Fourth Conference, pp. 100-112. Jos. Macy, Jr. Foundation,

New York.
PALMER, J. W., E. M. SMYTH AND K. MEYER, 1937. On glycoproteins IV. The estimation of

hexosamine. /. Biol. Chem., 119: 491-500.
RUBIN, P. S., AND J. E. HOWARD, 1950. Histochemical studies on the role of acid mucopoly-

saccharides in calcifiability and calcification. Metabolic Interrelations, Second Confer-

ference, p. 155. Jos. Macy, Jr. Foundation, New York.
RUNNSTROM, J., 1952. The cell surface in relation to fertilization. Symposia of the Society

for Experimental Biology VI. Structural Aspects of Cell Physiology, pp. 39-88.

Academic Press, New York.
SODA, T., AND F. EGAMI, 1938. A new sulfuric acid ester in the mucus of Charonia lampas.

Bull. Chem. Soc. Japan, 13 : 652.

THOMAS, L. J., JR.. 1951. A blood anti-coagulant from surf clams. Biol. Bull., 101: 230.
VASSEUR, E., 1947. The sulfuric acid content of the egg coat of the sea urchin, Strongylocentrotus

no. 6: 1-2.

droebachicnsis Mull. Ark. Kemi, Mineral. Geol. Bd. 25B,

Vol. 106, No. 2 April, 1954

THE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

THE RELATIONSHIP BETWEEN ?H AND THE ACTIVITY OF
CHOLINESTERASE FROM FLIES

L. E. CHADWICK, J. B. LOVELL AND V. E. EGNER
Medical Laboratories, Army Chemical Center, Maryland

In order to make useful comparisons of similar enzyme systems from different
organisms, or of the response of a given system to different chemical agents, some
understanding of the possible effect of changes in the conditions of assay is essen-
tial. This requirement becomes especially conspicuous in attempts to correlate
results from laboratories whose techniques are not identical, as illustrated, for
example, by certain discrepancies that will be discussed below. In commencing
a series of studies intended to bring to light the distinctive properties of insect
cholinesterases (ChE's), we have therefore found it expedient to examine in detail
the effects of altering our experimental conditions, and have already reported the
results of tests in which the activity of fly head ChE was measured in relation to
the composition of the suspending medium (Chadwick, Lovell and Egner, 1953).

Another factor with significant influence in vitro on the rate of hydrolysis of
acetylcholine (ACh) by ChE is the hydrogen ion concentration. For most ChE's
that have been studied in this connection, enzymic activity was maximal somewhat
on the alkaline side of neutrality, fell rather sharply at still higher pH values, and
declined more gradually as hydrogen ion concentration was increased. The per-
tinent references are discussed by Augustinsson (1948) ; see also Table IV below.

Three studies of the problem with insect material have been reported. Tahmisian
(1943) found a relationship of typical form and a pH optimum of 8.5 with the
enzyme from developing grasshopper eggs. Stegwee (1951), working with central
nervous tissue of the beetle, Hydrophilus, and the roach, Periplaneta, recorded
rather sharp optima at pH 7.4. Data of Babers and Pratt (1950) with fly head
suspensions are in contrast with these and all other reports in that they indicate
a peak in activity at about pH 5.75. In their preparations, ChE activity was main-
tained near half peak level between pH 6.25 and 9.00, and decreased abruptly at
higher pH values, as at values below 5.50. They comment (p. 61) that "this
activity over such a wide pH range not only was unexpected but is also unex-
plained" ; however, the most unusual feature of their results is the position of
the optimum.

At the time their work was published, our investigations of the same problem
were already in progress, and it was apparent immediately that our observations
did not agree fully with theirs. We therefore extended the scope of the experi-
ments, first in order to establish more firmly the nature of the relationship between

139

140 CHADWICK, LOVELL AND EGNER

pH and the activity of fly head ChE, and secondly with the hope of reconciling
the differences between our data and those of Babers and Pratt. In addition,
since the reduction in activity at low and high pH was found to involve some
irreversible inactivation of the enzyme, we made measurements of this aspect of
the process.

EXPERIMENTAL

Culture of flies (Musca domcstica L.) preparation of head suspensions, and
our application of Click's (1937) titrimetric method of measuring ChE activity
have been described in an earlier report (Chadwick, Lovell and Egner, 1953).
In the present experiments, data were obtained at 25.0 degrees C. on 20-ml.
aliquots, containing the equivalent of 20 heads each, with three suspension media.
Of these, the first was buffer: NaCl, 26.30 gm. ; KH2PO4, 3.85 gm. ; NaOH, 1.00
gm. ; H.,0, to one liter. This solution was designed to promote maximal enzymic
activity, which had been found to require the presence of a salt at about 0.5 N
concentration ; and was buffered lightly, so as to minimize fluctuation of pH during
assay and yet retain sufficient sensitivity for accurate determination of the rate
of production of acid. Since our results with this medium differed considerably
from those reported by Babers and Pratt (1950), whose suspensions contained
glycerol, a second series of observations was made with head tissue ground and
assayed in 30 per cent glycerol. In a third set of experiments, the brei was sus-
pended in de-ionized water. With all the suspensions, pH was adjusted to the
desired level by addition of NaOH or HC1.

The total acid production during test periods of approximately 15 minutes was
corrected by subtraction of the acid produced under conditions that were identical
except that the enzyme had been inactivated by exposing the stock brei overnight
or for a longer time to 1 X 10~5 M diisopropyl fluorophosphate (DFP). The net,
or enzymic, activity was then converted into micromoles of ACh.Br hydrolyzed per
ml. (i.e., per head) per hour.

The rate of permanent inactivation of fly head ChE at low and high pH was
determined on aliquots that were incubated at the desired pH value for definite
periods of time, and then readjusted rapidly to pH 8.0 before addition of ACh.Br
for assay.

In all these experiments, the concentration of substrate at the beginning of
measurement was 0.015 M. Other concentrations were used in a few experiments
for special purposes, as cited in the discussion.

RESULTS

Average rates of enzymic hydrolysis of 0.015 M ACh.Br at various pH values
in the three media tested are given in Table I. The data have been plotted in
Figure 1 as percentages of the average value determined for these tissue samples
in buffer at pH 8.0.

Also shown in Table I are the corresponding corrections for non-enzymic
hydrolysis. These were evidently not identical in the several media. As pointed
out in our previous paper, such variation results in part from the fact that pH

EFFECT OF pH ON FLY HEAD CHOLINESTERASE

141

is not truly constant in our method of measurement. Each addition of NaOH
during titration pushes pH to the alkaline side of the chosen value, and, for a
given amount of alkali, such excursions are greater the more weakly buffered the
solution. Error from this cause is not overly significant at pH 8.0 and below,
but increases rapidly in more alkaline solutions, where the rate of non-enzymic
hydrolysis of ACh is rising steeply with increase in pH. Since ChE activity, as
measured in well buffered solutions, increases but slightly above pH 8.0, the
error resulting from fluctuation of pH in the experimental samples is largely in
the nonenzymic fraction and should theoretically be compensated by the nearly
equal error in the controls. In practice, however, we found it difficult to obtain
satisfactorily consistent results at pH 9.5 and 10.0.

In order to supply some indication of the range of variation encountered, we
have computed standard errors for the means in each series except for those

TABLE I

Enzymic activity of fly head ChE as a function of pH

pH

4.0

5.0

5.5

6.0

7.0

7.5

8.0

8.5

9.0

9.5

10.0

In buffer

Net rate*

nil

0.34

1.97**

2.05

4.05

4.39

4.88

5.08

5.27

4.64

1.17

±s.e.

—

0.03

—

0.13

0.18

0.11

0.10

0.24

0.22

0.36

0.83

n***

4

10

5

9

10

11

31

10

13

10

10

Correction

—

—

-0.03

-0.03

-0.07

-0.09

-0.24

-0.56

-1.52

-4.78

-13.60

In de-ionized water

Net rate*

nil

0.54

1.42

1.66

2.24

2.54

2.15

1.66

2.88

1.46

0.88

±s.e.

—

0.07

0.18

0.24

0.21

0.24

0.12

0.16

0.51

—

—

n***

6

10

10

10

10

10

17

10

10

5

5

Correction

—

—

-0.01

-0.02

-0.04

-0.28

-0.56

-1.21

-3.10

-7.63

-22.17

In glycerol, 30 per cent

Net rate*

nil

0.44

0.88

1.27

1.42

0.98

1.37

0.88

1.51

0.63

0.15

n***

5

5

5

6

5

5

13

5

5

5

5

Correction

~

~

~

-0.01

-0.02

-0.25

-0.54

-1.14

-2.63

-4.24

-7.25

* Average net rates, standard errors, and corrections in micromoles ACh.Br hydrolyzed
per ml. ( = per head) per hour.
**pH, 5.75.
*** n, number of tests.
All runs at 25.0 degrees C. ; ACh.Br, 0.015 M.

where only 5 determinations were made at each pH level. We report these
calculations with some hesitation, first because of the relatively small "n," and
secondly because the measurements at different pH values in a single series were
not wholly independent. For example, aliquots of a stock brei which showed
more than average activity at one pH value tended to give higher than average
measurements at all pH levels. Thus, the sampling was not truly random, and
on this account the standard errors listed in the table should not be relied on for
estimates by the t-test of the significance of differences between means.

The second table shows the average activity remaining in samples that had been
incubated at the indicated pH values for 30 minutes and then readjusted to pH
8.0 for assay. In Table III are recorded the results of exposing samples to pH 4.0
for different periods of time, up to two hours. Some few additional data pertinent
to these experiments are cited in the discussion.

Table VI shows ChE activity as a function of pS at two pH levels, viz., 6.0
and 8.0.

142

CHADWICK, LOVELL AND EGNER

tr
ui
o.

120 -

100

80

60

40

O

< 20

1

1 1 1

1 1 1 1 1 1

3

456

78 9 10 II IE

PH

FIGURE 1. Variation in ChE activity of fly heads as a function of pH. Curve A. Residual
activity after 30-minute exposures at indicated pH values ; all measurements at pH 8.0 in buffer.
Curve B. Specific activity at indicated pH values, computed by correcting Curve C for degree
of inactivation shown in Curve A. For fuller explanation, see text. Curve C. Activity meas-
ured during 15-minute exposures at indicated pH values in buffer. The open circles give the
mean values, and the vertical bars indicate the limits for ± 3 s.e. Curve D. Activity measured
during 15-minute exposures at indicated pH values in water. Curve E. Activity measured during
15-minute exposures at indicated pH values in 30 per cent glycerol. All data shown have been
corrected for non-enzymic hydrolysis.

TABLE II

ChE activity of fly head suspensions after 30 minutes exposure at various pH values

pH

Average enzymic
activity
(per cent)

Number of
observations

pH

Average enzymic
activity
(per cent)

Number of
observations

3.0

1.0

1

8.0

100.0

10

3.5

1.9

5

9.0

93.2

5

4.0

41.9

4

10.0

85.6

5

5.0

75.0

4

10.5

83.2

5

6.0

81.1

5

11.0

27.2

5

7.0

93.4

5

12.0

nil

2

All runs at 25.0 degrees C. ; ACh.Br, 0.015 M ; samples readjusted to pH 8.0 for measurement.

EFFECT OF pH ON FLY HEAD CHOLINESTERASE

TABLE III
Rate of inactivation of fly head ChE at pH 4.0 as a function of duration of exposure

143

Time exposed

minutes

0

15

30

60

90

120

Activity in per

cent

100

65.9

41.9

30.1

24.9

20.5

Number of

observations

20

5

4

4

4

4

All runs at 25.0 degrees C. ; ACh.Br, 0.015 M; samples readjusted to pH 8.0 for measurement

DISCUSSION

1.

ChE activity as a junction of pH

Examination of the measurements in buffer convinces us that the ChE of our
fly heads differs little, in respect to the effect of pH on activity, from most other
ChE's hitherto studied (cj. Table IV). The optimum is clearly on the alkaline
side, being at least as high as 8.0 and probably as high as 9.0.

In aqueous suspensions or in 30 per cent glycerol, activity was generally low
in comparison with observations at corresponding pH values in buffer, with the
rates in glycerol somewhat less than those in water. These data provide a further
demonstration of the activating effect of 0.5 N salt and the depressant effect of
glycerol, to which we called attention earlier (1953). In water or glycerol there
appeared to be little significant change in ChE activity over the pH range from

TABLE IV

pH optima of ChE's from various sources

Source of enzyme

pH optimum

Authority

Eggs, developing, Melanoplus

Erythrocytes, human

Erythrocytes, human

Serum, human

Serum, human

Serum, human

Serum, human

Serum, horse

Serum, horse

Serum, horse

Heart extract, frog

Electric organ, Electrophorus

Gastric mucosa, pig

Brain, rat (also rabbit, guinea pig,

cat, dog)
Brain, cat

c.n.s., Periplaneta, Hydrophilus
Heads, Musca
Heads, Musca

8.5

7.6 or above
7.5 to 8.0
8.2 or above
8.0 or above

8.4 to 8.5
8.0 to 8.5
ca. 8.5

7.2 or above
8.0 to 8.5

7.5 or above
ca. 8.5

ca. 8.5
ca. 8.4

ca. 8.5

7.4

5.75

8.0 or above

Tahmisian, 1943

Plattner et al., 1928

Alles and Hawes, 1940

Plattner et al., 1928

Easson and Stedman, 1936

Click, 1937

Werle and Uebelmann, 1938

Click, 1938

Kahane and Levy, 1936

Werle and Uebelmann, 1938

Loewi and Navratil, 1926

Wilson and Bergmann, 1950

Click, 1938

Bernheim and Bernheim, 1936

Click, 1938
Stegwee, 1951
Babers and Pratt, 1950
This paper

144

CHADWICK, LOVELL AND EGNER

6.0 to 9.0, in agreement with the findings of Babers and Pratt (1950) ; however,
variation in our measurements was considerable and the curves are quite irregular.
Neither in these media nor in buffer could we find any evidence for an activity
peak in the neighborhood of pH 5.75, as reported by Babers and Pratt. This
led us to attempt one final comparison, in which activity at pH 5.75 and 7.0 was
measured under conditions as nearly like theirs as we could make them. For this
purpose, tissue was prepared in 30 per cent glycerol and diluted 1:6 for assay,
which was carried out on 9.0-ml. aliquots that contained 150 mg. of tissue and
0.045 M ACh.Br. Babers and Pratt had used 3.0-ml. samples containing 50 mg.

TABLE V
Comparison of ChE activity of fly head suspensions in 5 per cent glycerol at pH 7.0 and 5.75

pH 7.0 pH 5.75

ml. 0.02 N NaOH per 3 ml. per 20 minutes

1.05
1.03
1.09

Average
*Correction

Net

1.00
1.05
0.98
1.03
1.01

1.01

-0.04

0.97

1.02
1.01

1.04

-0.01

1.03

* Correction values from Babers and Pratt (1950). All runs at 25.0 degrees C. ; ACh.Br,
0.045 M\ tissue, 150 mg. ; total volume, 9.0 ml. Data computed to 3.0 ml. volume for sake of
comparison with results of Babers and Pratt.

of tissue, but this volume was too small for our electrodes. As in their tests, acid
produced was titrated with 0.02 N NaOH over a 20-minute test period. Five
replications were made. The results, corrected for non-enzymic hydrolysis with
values taken from Babers and Pratt (1950), are shown in Table V.

These data suggest the following comments. First, activity was nearly equal
at both pH values; i.e., evidence for a pronounced peak at pH 5.75 was not forth-
coming. Secondly, as was to have been expected, activity per unit weight of tissue,
or per head, was intermediate at pH 7.0 between the values previously found with
suspensions in water and in 30 per cent glycerol, respectively. Finally we may

TABLE VI

Activity of fly head ChE as a function of substrate concentration at two pH levels

Molar
concentration
of ACh.Br

0.001

O.OIM

0.01

0.03

0.10

Average enzymic activity in micromoles per head per hour

pH 8.0
pH 6.0

4.83
1.22

5.11
1.77

5.29
2.37

4.51

2.44

2.83
1.71

Each datum is the mean of 5 determinations. All runs on aliquots of the same stock brei in
buffer at 25.0 degrees C.

EFFECT OF pH ON FLY HEAD CHOLINESTERASE 145

note that the activity of our preparation, at both pH levels, was more than twice
the peak value reported by Babers and Pratt (1950). Unless some undetected
difference in our methods of preparing the tissue can be held responsible, this
observation indicates a possible strain difference between their flies and ours ;
and should strain differences of this magnitude exist, they could conceivably extend
to a shift in the pH optimum from above 8.0 to 5.75. This, however, seems very
unlikely in view of the bulk of evidence (Table IV) in favor of an alkaline pH
optimum for ChE's in general. The remaining alternative is to ascribe the obser-
vation of Babers and Pratt to fortuitous variation in the activity of different breis ;
i.e., to a somewhat unlikely coincidence of sampling errors, that led them repeatedly
to exceptionally high values at pH 5.75. This solution does not appeal to us,
since it is obviously indemonstrable and because the same sort of inference could,
with equal justification, be applied to our own data ; but all our efforts to find a
more satisfactory explanation have failed.

Theoretical reasons for anticipating an increase in optimal concentration of
substrate as conditions of measurement depart from the pH optimum have been put
forward by Wilson and Bergmann (1950). The data in Table VI bear on this
question, and do in fact indicate a slight shift of pSopt. in the predicted direction
at pH 6.0 as compared with pH 8.0. Although it is of interest that this shift
should appear in our results, the presence of the effect will hardly demand correc-
tion of the pH-activity data in Table I and Figure 1, for the following reasons.
The magnitude of the shift is small, the optima are relatively flat, and the standard
concentration of 0.015 M ACh.Br used routinely in our experiments is already
somewhat above the optimum for pH 8.0.

2. hiactivation of fly head ChE

The reduction in activity of fly head ChE at hydrogen ion concentrations that
depart appreciably from pH 8.0 is not wholly reversible. This fact raises a question
as to what portion of the activity change observed at different pH levels is due to
an effect of pH on reaction rate, and what portion to permanent destruction of a
fraction of the enzyme. Obviously, data such as those in Table I must reflect a
summation of both these processes.

We have attempted in a preliminary manner to separate the two effects by
measuring the irreversible inactivation of ChE that results when the suspensions
are exposed in buffer to different pH values for a constant period of time. The
interval chosen was 30 minutes, this being somewhat longer than the average total
exposure during our routine 15-minute determinations. As indicated in Table II,
the percentage inactivation observed under these conditions remained within
moderate limits until one passed below pH 5.0 or above pH 10.5. Other observa-
tions not given in the table showed that, within this pH range, there was little if
any additional loss of activity during exposures of as much as two hours ; and
further that subsequent incubation of the samples at pH 8.0 for as long as 18 hours
caused no reversal of the loss that had already occurred.

That inactivation did not take place instantaneously was demonstrated by a
series of tests at pH 4.0, where suspensions were held for periods varying from
15 minutes to two hours, before return to pH 8.0 for assay (Table III). Here

146 CHADWICK, LOVELL AND EGNER

the process of inactivation was rapid for the first 30 minutes, and followed a
slower course thereafter. Both segments of the relationship have the characteristics
of a first order reaction, as indicated in Figure 2.

Below pH 4.0 and above pH 11.0, inactivation was rapid and extensive. As
a matter of fact, the enzyme, together with large amounts of eye pigment, was
precipitated from aqueous suspensions of head tissue at about pH 5.0 to 5.1.
Some 75 per cent of the original activity could be recovered if this precipitate

2.0

\

1.9

O

1.8

1.7

1.6

UJ

O

1.5

DC
Id
Q.

O
O

-* 1.4

1.3

\
\
\

\
\
\

\
\

\
\
\
\
\
\

\

\

>0

I I I I

0 30 6O 9O 120

EXPOSURE TIME IN MINUTES

FIGURE 2. Rate of inactivation of fly head ChE in buffer as a function of duration of exposure
at pH 4.0. All measurements made at pH 8.0. For further details, see text and Table III.

was quickly re-suspended in buffer at pH 8.0; but it was evident that the enzyme
had been altered, since it was no longer as soluble as before precipitation. It was
now easily re-separated by light centrifugation. According to Augustinsson (1948),
the isoelectric point of several other ChE's has been reported as about pH 4.5.
He also notes that precipitation at this level results in more or less permanent
solubility changes.

By combining the data of Table II, which show the fraction of enzyme still
active after half-hour exposures at the various pH values, with those of Table I,

EFFECT OF pH ON FLY HEAD CHOLINESTERASE 147

which give the rates at which similar aliquots were able to hydrolyze substrate
while at the same pH levels, it is possible to construct a corrected curve that
compensates for changes in the relative amount of active enzyme. Such a curve
is shown as B in Figure 1. The divergence from Curve C, which embodies the
uncompensated activity data, is slight. This is because 30-minute exposures at
pH values between 5.0 and 10.5 inactivated only small fractions of the enzyme,
while the effect of pH on reaction rate was already considerable, well within these
limits. The compensated curve (B) emphasizes the activity peak at pH 9.0, since
the proportion of active enzyme has already begun to fall in this region, whereas
the measured activity per unit of tissue has increased slightly above the value
determined at pH 8.0. The results also show incidentally that the drop in activity
at still higher pH levels is not due wholly to denaturation of the enzyme, since the
measured activity decreased more rapidly than the enzyme was destroyed.

These data suggest a pH of about 8.0 as a suitable compromise for experiments
where an approach to maximal activity of fly head ChE is desired. Although the
true optimum probably lies as much as a full pH unit to the right of pH 8.0, the
increase in enzymic activity over this range is slight, whereas the correction for
non-enzymic hydrolysis is rapidly becoming larger. At pH 8.0, this correction is
less than 5 per cent of the average total activity measured under our conditions ;
i.e., with tissue concentration at one head per ml. ; temperature, 25.0 degrees C. ;
substrate, 0.01 5M ; and salt present in the buffered suspension at about 0.5 N
concentration. The correction could be still further reduced by shifting to even
lower pH levels, but only with increasing sacrifice of enzymic activity.

SUMMARY

1. Variation in activity of fly head cholinesterase (ChE) was measured titri-
metrically at 25.0 degrees C. with ACh.Br 0.015 M as substrate, as a function of
the pH of the assay medium over the range from pH 4.0 to 10.0. Ground tissue
obtained from Musca domestica L. was suspended at a concentration of one head
per nil. in three media : ( 1 ) buffer of composition NaCl, 26.30 gm. ; KH2PO4,
3.85 gm. ; NaOH, 1.00 gm. ; H2O, to one liter; (2) 30 per cent glycerol; (3)
de-ionized water.

2. Enzymic activity was greater in buffer than in the other media. The pH
optimum was definitely on the alkaline side, being at least as high as pH 8.0 and
probably as high as 9.0. In glycerol or water suspensions, enzymic activity
changed little between pH 6.0 and 9.0.

3. Some permanent inactivation of the enzyme was observed in half-hour
exposures at high and low pH values. This effect was measured over the pH range
from 3.0 to 12.0. Between pH 5.0 and 10.5, the degree of inactivation was moderate
and essentially complete within 30 minutes. The time course of the process was
followed at pH 4.0 for intervals from 15 minutes to two hours, and appeared to
involve a rapid phase during the initial 30 minutes and a slower phase thereafter.
Both phases had the characteristics of a first order reaction. Inactivation of ChE
resulting from exposure to low or high pH was not reversed during subsequent
incubation of the sample at pH 8.0 for as long as 18 hours.

4. Correction of the pH-activity curve to allow for changes in the relative
amounts of enzyme that result from permanent inactivation requires only minor

148 CHADWICK, LOVELL AND EGNER

alterations, since the effect of pH on reaction rate makes itself felt within pH limits
where the degree of permanent inactivation is slight.

LITERATURE CITED

ALLES, G. A., AND R. C. HAWKES, 1940. Cholinesterases in the blood of man. /. Biol. Clicni.,

133 : 375-390.
AUGUSTINSSON, K.-B., 1948. Cholinesterases. A study in comparative enzymology. Ada

physiol. Scand., 15 (Suppl. 52) : x + 1-182.
BABERS, F. H., AND J. J. PRATT, JR., 1950. Studies on the resistance of insects to insecticides.

I. Cholinesterase in house flies (Musca domestica L.) resistant to DDT. Ph\siol. Zool.,

23 : 58-63.
BERNHEIM, F., AND M. L. C. BERNHEIM, 1936. Action of drugs on the choline esterase of the

brain. /. Pharmacol. Exp. Therap., 57: 427-436.
CHADWICK, L. E., J. B. LOVELL AND V. E. EGNER, 1953. The effect of various suspension

media ou the activity of cholinesterase from flies. Biol. Bull., 104 : 323-333.
EASSON, L. H., AND E. STEDMAN, 1936. The absolute activity of choline-esterase. Proc. Roy.

Soc. London, Ser. B, 121 : 142-164.
CLICK, D., 1937. LXXII. Properties of choline esterase in human serum, Biochcm. J., 31 :

521-525.
CLICK, D., 1938. Studies on enzymatic histochemistry. XXV. A micro method for the

determination of choline esterase and the activity-pH relationship of the enzyme.

/. Gen. Physiol., 21 : 289-295.
KAHANE, E., AND J. LEVY, 1936. Sur 1'hydrolyse diastatique de 1'acetylcholine par le serum.

C. R. Acad. Sci. Paris, 202: 781-783.
LOEWI, O., AND E. NAVRATIL, 1926. Uber humorale Ubertragbarkeit der Herznervenwirkung.

X. Mitteilung. Uber das Schicksal des Vagusstoffs. Pfliigers Arch., 214: 678-688.
PLATTNER, F., O. GALEHR AND Y. KODERA, 1928. Uber das Schicksal des Acetylcholins im

Blute. IV. Mitteilung. Die Abhangigkeit der Acetylcholinzerstorung von der Wasser-

stofHonenkonzentration. Pfliigers Arch., 219: 678-685.

STEGWEE, D., 1951. Studies on cholinesterase in insects. Physiol. Comp. et Oecol., 2: 241-247.
TAHMISIAN, T. N., 1943. Enzymes in ontogenesis: choline-esterase in developing Melanoplus

differentiate eggs. /. Exp. Zool., 92: 199-213.
WERLE, E., AND H. UEBELMANN, 1938. Zur Kenntnis des Blutegeltestes. Arch. f. exp. Pathol.

u. Pharmakol., 189: 421-432.
WILSON, I. B., AND F. BERGMANN, 1950. Acetylcholinesterase. VIII. Dissociation constants

of the active groups. /. Biol. Chem., 186: 683-692.

THE AMINO ACID REQUIREMENTS OF THE CONFUSED FLOUR
BEETLE, TRIBOLIUM CONFUSUM, DUVAL.

G. FRAENKEL AND GLENN E. PRINTY 1
Department of Entomology, University of Illinois, Urbana, Illinois

The rapidly accumulating literature on the nutrition of insects contains com-
paratively few data on amino acid requirements. However, all present evidence
seems to indicate that insects require the ten amino acids which are essential
for the rat.

Work on the nutrition of Triboliurn confusitin in a chemically well-defined
medium, consisting of casein, glucose or starch, cholesterol, a salt mixture and
eight to 10 vitamins of the B-complex, has been previously published by several
authors (Fraenkel and Blewett, 1943, 1947; Fraenkel and Stern, 1951; Offhaus,
1952). An entirely successful "synthetic" diet for Triboliwm, on which growth
is as good as on the best natural diets, has not yet been reported. It has only very
recently been found that carnitine is required for adult development (French and
Fraenkel, 1954). The addition of \% brewers yeast to a synthetic diet invariably
leads to an improvement of growth. However, even in the absence of yeast, Tri-
boliiini grows sufficiently well to determine the effect of amino acid deficiencies. The
results of the present study largely confirm and extend work on similar lines by
Lemonde and Bernard (1951).

METHODS

The basic diets used in this investigation were derived from diets which had
been previously used in work with Tribolium. However, the fact that amino acid
mixtures were used in the place of casein necessitated certain modifications in the
diet. It was desirable to reduce the proportion of amino acids to a relatively low
level which would still allow for adequate growth. Tribolium grows well on a
wide range of carbohydrates ranging in concentration from 5 to 80% of the diet.
In most of our previous work the protein level used was 50%. In the present
study this was reduced to a total of 15% casein or mixtures of amino acids. In
almost all our previous work the carbohydrate in the diets had been glucose.
However, in the present study corn starch was used as the carbohydrate in all tests.
Glucose could not be used because of the Maillard reaction between sugars and
amino acids described by Friedman and Kline (1950a, 1950b). All the starch used
in the experiments to be described was from the same batch.

The diets consisted of 15 parts casein or amino acid mixture, 85 parts corn
starch, one part cholesterol, 2 parts McCollum's salt mixture no. 185 and the
following vitamins of the B-complex (expressed as p.g. per gram of the dry diet) :
thiamin 25, riboflavin 12.5, nicotinic acid 50, pyridoxin 12.5, pantothenic acid 25,
choline chloride 500, inositol 250, pteroylglutamic acid 2.5 and biotin 0.25. All
the ingredients, except the vitamins, were mixed in the dry state. To ensure a
good distribution of those ingredients which were present in very small amounts,

1 Present address : Department of Entomology, Citrus Research Station, Riverside,
California.

149

150 G. FRAENKEL AND GLENN E. PRINTY

the mixture of amino acids was first ground with an equal amount of starch, the
cholesterol and salts were then added and ground together, and the balance of
starch finally added and mixed in. The diets also contained NaHCO3 in amounts of
approximately ten per cent of the amino acid mixture (to neutralize free acids).
The vitamins were then added in solution in the amounts stated above to add 10%
water to the dry diet. After mixing the vitamin solution into the diet with a
spatula, the diets were left standing for two days in a constant temperature
chamber at about 30° C. and 60-70% relative humidity, and then ground by
hand in a mortar.

The tests were performed in shell vials, 1x2 inches, with one gm. of dry-
diet per vial, two vials to each diet. Ten first stage larvae were placed in each
vial. All the tests were performed in a constant temperature chamber at 29-30° C.
and 60-70 per cent relative humidity.

To assess the efficiency of a diet, two criteria were used. The number of
surviving larvae and their average weight were determined after a period long
enough to allow larvae on the optimal diet in a particular experiment to reach
their maximum weight before pupation had started. This period varied somewhat,
according to the composition of the diets, but was usually 20 days. On optimal
diets, with glucose as the carbohydrate and the addition of yeast, pupation may
occur after 15 days; however, since all the diets contained starch and few contained
yeast, the period required for full growth was longer. The date of pupation was
then recorded for each individual larva. From these results the average time to
pupation was calculated for each test. In some cases the pupae were kept until the
adult beetles emerged and the newly-formed beetles were examined. In following
this procedure it was considered possible that certain amino acid deficiencies
might affect larval mortality, growth rate, pupation or emergence in a different
way than others. The most significant data were usually derived from the weights
of the larvae. Since slow growth always leads to a delay in the onset of pupation,
a positive correlation should exist between weights, growth rate and days to
pupation. However, in some experiments, the number of pupae was unexpectedly
small. The data concerning adult emergence finally proved to be without signifi-
cance, since after most of the work was completed, it was discovered that the adults
of Tribolium, which were grown on artificial diets, were not viable or failed to
emerge, unless carnitine was added to the larval diets. There was, however, no
indication that carnitine was necessary for larval growth and successful pupation
(French and Fraenkel, 1954).

Growth and survival of Tribolium vary somewhat in diets run at different
times. This may be due to slight changes in temperature and humidity, a difference
in the viability of different batches of larvae and possibly other factors which are
not too well understood. It makes it necessary to include in each experiment the
appropriate positive and negative controls, and to make strict comparisons only
between diets run at the same time.

EXPERIMENTS
A. The amino acids mixtures used, and their effect on three species of insects

In the absence of data about the amino acid requirements of Tribolium when
this study was initiated, it was considered advisable to start the work with mixtures

AMINO ACID REQUIREMENTS OF TRIBOLIUM

151

which had proved successful with higher animals. Three were used altogether,
two of which were amino acid mixtures used by Rose, Oesterling and Womack
(1948) with the white rat. The third was one devised by Almquist and Grau
(1944) for chicks. Table I gives the percentage composition of these amino acid
mixtures. They were at first tested for their effect on the larvae of three beetles,
Tribolium confusum, Tenebrio molitor and Dermestes vulpinus. The diets for
Tribolium and Tenebrio were identical, except for the addition of 1.5 yu.g. carnitine
per gram of the diet for Tenebrio. Tenebrio and Tribolium received 15% amino
acids whereas Dermestes, which is a typical protein feeder, received 30% amino

TABLE I

Composition of the amino acid mixtures used in studies of the amino acid requirements of
Tribolium confusum, Tenebrio molitor, and Dermestes vulpinus

Amino acid

Per cent of the total amino acid mixture

Rose el al., 1948
19 amino acids

Rose el al., 1948
10 amino acids

Almquist and Grau. 1944
20 amino acids

DL-alanine

2.54

3.33

L-arginine hydrochloride

3.18

4.20

4.67

DL-aspartic acid

2.54

6.67

L-cystine

1.27

1.33

L-glutamic acid

12.69

16.65

Glycine

.64

6.00

L-histidine hydrochloride

6.03

8.15

2.67

L-hydroxyproline

.63

.67

DL-isoleucine

10.16

13.95

6.67

L-leucine

7.62

10.40

6.67

L-lysine

9.53

13.01

4.67

DL-methionine

5.07

6.97

3.33

DL-norleucine

.67

DL-phenylalanine

7.62

11.40

3.33

L-proline

1.27

6.67

DL-serine

1.27

1.33

DL-threonine

8.89

12.10

10.00

L-tryptophane

2.54

3.49

1.33

L-tyrosine

3.81

6.67

DL-valine

12.70

17.35

6.67

NaHCO.,

8.07

11.00

5.00

Total

108.07

111.02

105.00

acids, and no carnitine. The results of these tests are given in Table II . Dermestes
failed to grow on these diets, and Tenebrio grew very poorly. Tribolium, however,
grew on Rose's 19 amino acid mixture as well as it did on casein. With only the
10 essential amino acids in the diet, growth was somewhat delayed. The Almquist
mixture proved very much inferior. The experiments with Tenebrio and Dermestes
were first started with first stage larvae. When these larvae failed to develop
on the diets, the tests were repeated with larvae of larger size (Tenebrio larvae
of about 20 mg. and Dermestes larvae of about 10 mg.). It was expected that
larger larvae which had originally been grown on an optimal diet might be more
robust and more able to survive and overcome any adverse effect of amino acid

152

G. FRAENKEL AND GLENN E. PRINTY

TABLE II

Response of three insects to amino acid diets

Amino acid mixture and reference

19 amino acids
Rose et al, 1948

10 amino acids
Rose et al, 1948

20 amino acids
Almquist et al, 1944

Casein control diet

Dermestes vidpinus

Tenebrio molitor

Tribolium fonfusnm

is growth equal to that on casein.
— is no growth.

diets. However, the larger larvae also failed to develop. All attempts to grow
Dermestes and Tenebrio on amino acid mixtures have so far failed. The good
results obtained with Tribolium on Rose's mixtures, however, were a starting point
for further experiments.

B. The requirements oj Tribolium for individual amino acids

Two of Rose's amino acid mixtures were used, one which contained 19 amino
acids and another which contained only the 10 "essential" acids, in the proportion
shown in Table I. A series of diets was then devised in which each of the amino
acids was left out, one at a time. The results were clear cut. In every single
case in which one of the 10 essential acids was omitted, the larvae failed to grow
(Tables III and IV). Each of the remaining "non-essential" acids could be
omitted from the diet, without noticeable effects (Table IV). However, larvae

TABLE III

Effect on Tribolium larvae of omitting each amino acid from a diet containing

the 10 "essential" amino acids*

Exp. 12 — weighed at 15 days

Exp. 14 — weighed at 20 days

Avg.

Avg.

Diet

No.

wt.

Diet

No.

wt.

(mg.)

(mg.)

Casein control

18

0.55

Casein control

16

1.80

All 10 amino acids

18

0.40

All 10 amino acids

11

0.80

Without L-arginine

10

0.10

Without L-lysine

1

0.10

Without L-histidine

6

0.10

Without DL-methionine

0

Without L-isoleucine

9

0.08

Without DL-phenylalanine

1

0.10

Without L-leucine

0

—

Without DL-threonine

0

Without L-tryptophane

13

0.09

Without DL-valine

0

* None of the larvae on deficient diets survived to pupate.

AMINO ACID REQUIREMENTS OF TRIBOLIUM

153

always grew faster in the presence of 19 amino acids than of ten, in spite of the
fact that the total level of amino acids was the same in both instances (Table IV).
Superior growth of rats on a mixture of 19 amino acids, as compared with the 10
essential acids, has previously been reported by Rose et al. (1948).

TABLE IV

Tribolium confusum. Effect of omitting each amino acid from a diet containing
19 amino acids (Rose diet XXIII)

Experiment 17

20 day larvae

Pupae

Adults

Omission or other variation

Avg.

Avg.

No.

No.

wt.

No.

time

No.

normal

(mg.)

(days)

Casein

15

0.73

14

33.0

14

1

19 amino acids

20

1.11

20

30.1

16

3

No alanine

19

1.14

16

31.4

12

1

No aspartic acid

17

1.18

17

32.2

16

3

No cystine

19

1.17

18

30.9

18

2

No glutamic acid

19

1.11

15

31.4

11

1

No glycine

18

1.31

17

30.3

10

2

No hydroxyproline

12

1.09

10

30.7

9

0

No proline

18

1.00

16

32.7

13

0

No serine

18

0.84

15

32.0

14

3

No tyrosine

19

1.17

19

33.2

16

1

The 10 essential amino acids

18

0.53

12

41.5

11

1

(Rose diet XXIV)
Experiment 16

Casein

19

.96

18

30.7

15

0

19 amino acids

Ground by hand, mortar

12

.93

9

34.2

8

0

Ground in ball mill — 22 hrs.

17

.8

13

35.5

8

3

No arginine

5

.01

0

*

0

0

No histidine

4

.01

0

0

0

No isoleucine

4

.01

0

0

0

No leucine

3

.03

0

*

0

0

No lysine

3

.01

0

*

0

0

No methionine

9

.01

0

**

0

0

No phenylalanine

4

.01

0

**

0

0

No threonine

3

.01

0

0

0

No tryptophane

2

.01

0

0

0

No valine

7

.01

0

*

0

0

* One small larva at 90 days.
** Two small larvae at 90 days.

At the end of 8 weeks, when it was apparent that growth was not possible
in the absence of any of the 10 essential amino acids and when most of the larvae
had died, \% yeast was added to each of the deficient diets and the experiment run
again with a fresh lot of first stage Tribolium larvae. Growth was very much faster
after the addition of yeast, and the effect of amino acid deficiencies was largely

154

G. FRAENKEL AND GLENN E. PRINTY

obscured. This phenomenon is difficult to understand in view of the fact that the
addition of 1 % yeast only insignificantly adds to the total amount of certain of the
essential amino acids in the diet.

Since a diet which contained 19 amino acids always proved superior to one
with only 10 essential amino acids, an attempt was made to evaluate the effects
of the non-essential amino acids in the diet. Omitting any single non-essential
acid had no effect on the diet (Table IV).

It was then considered possible that the amino acids in the mixtures used in the
tests might not have been present in optimal proportions. In fact there was no a
priori reason for such an assumption to be true. Thus those differences in growth
rate, which existed between a 10 and a 19 amino acid mixture, might possibly be due
to changes in the total amount of some of the acids present. Accordingly, further

TABLE V

Response of Tribolium to D-amino acids in a medium of 19 amino acids

Substitution in diet

D-form substituted

L-form added*

20-day larvae

No. of
pupae

Av. time to
pupation
(days)

20-day larvae

No. of
pupae

Av. time to
pupation
(days)

No.

Av. vvt.

(mg.)

No.

Av. wt.

(mg.)

Casein control diet

16

0.90

4

27.0

Amino acid control diet

19

0.60

12

38.0

D-arginine
D-histidine

6
9

0.04
0.05

0
0

11

13

0.66
0.40

10
2

35.0
38.0

D-isoleucine

0

16

0.30

2

43.0

D-leucine

0

18

0.3

3

41.0

D-lysine
D-methionine

18
18

0.67
0.80

7
2

34.5
31.0

D-phenylalanine
D-threonine

17
0

0.50

0

14

0.30

6

46.3

D-tryptophane
D-valine

1
0

0.05

0

15

17

0.40
0.30

4
1

38.5
39.0

* The L-forms of the respective amino acids were added and the diets re-infested with larvae.

tests with 19 amino acids were devised in which the amount of each amino acid was
doubled in individual tests. The hypothesis was that this procedure might produce
two kinds of effects. If the diets had been improved, there would have been an
indication that the original mixtures did not contain enough of certain acids for
optimal growth. If the diets became worse, there would have been an indication
that the basic mixtures already might have contained excessive quantities of certain
acids. This experiment did not show any clear-cut changes in the efficiency of the
diets. Poorer growth resulted with double amounts of aspartic acid and valine.
Somewhat poorer growth also resulted when these amino acids were added to a
casein diet. However, these effects were only slight.

In further series of tests the D-forms of the ten essential amino acids were
individually substituted for the L- or DL-forms, in a diet consisting of 19 amino
acids. The D-forms of arginine, histidine, isoleucine, leucine, threonine, trypto-

AMINO ACID REQUIREMENTS OF TRIBOLIUM

155

phane and valine were entirely inactive (Table V). It was considered possible
that some of them might have been not merely inactive, but actually inhibitory.
Consequently the diets with D-acids, on which the larvae had failed to grow, were
later supplemented with the respective L-form and new larvae added. The larvae
grew somewhat slowly on most of the diets, which might have been due to the
age of the diets. The results, however, did not suggest that the D-forms were
inhibitory.

In one experiment (Table V) the D-forms of lysine and methionine gave as good
larval growth as the L-forms, but pupation was fairly good with D-lysine and very
poor with D-methionine. D-phenylalanine also showed good growth, but no
pupation occurred. In a repeat of this experiment (Table VI), in which carnitine
had been added to the diets, D-lysine proved entirely inactive. D-methionine was
as active, and D-phenylalanine almost as active as the respective L-forms. All
through this test the larvae pupated well and the adults were normal in the presence

TABLE VI

Response of Tribolium to the D-forms of lysine, methionine and phenylalanine

in a medium of 19 amino acids

Substitution in diet

20-day old larvae

Pupation

No. of adults

No.

Av. weight
(mg.)

No.

Av. time
(days)

Abnormal

Normal

Casein control diet :

Carnitine absent

20

.83

20

29.4

17

3

Carnitine present
Amino acid control diet :

20

.76

20

29.6

—

20

Carnitine absent

16

1.12

12

28.6

12

—

Carnitine present
D-lysine*

19
dead

1.68

19

26.1

—

19

D-methionine*

17

1.68

17

25.9

—

17

D-phenylalanine*

20

.89

16

29.2

—

16

* Carnitine present.

of carnitine. The result shows that the small number of pupae in the first test
might have been, in part, attributable to the absence of carnitine. This, however,
would not explain why D-lysine was fairly active in one, and entirely inactive in
another test.

DISCUSSION

The results on the amino acid requirements of Tribolium, as reported in this
paper, closely follow those previously reported for other insects. Lemonde and
Bernard (1951), in their work with the same insect, Triboliwn confiisum, reached
similar conclusions. They obtained some growth, and even pupation in the absence
of either lysine, threonine, phenylalanine, methionine, isoleucine, arginine, leucine
and tryptophane, although growth in all these cases wras very slow. This may
have been due to the presence, in the diets, of one half per cent of yeast. Moore
(1946) demonstrated the necessity of the 10 essential amino acids in the nutrition of

156 G. FRAENKEL AND GLENN E. PRINT Y

a carpet beetle, Attagenus sp. ; the effect of the non-essential acids was, however, not
studied. The larva of the yellow fever mosquito, Aedes aegypti L. was shown to
require glycine for normal growth in addition to the essential amino acids, plus
tyrosine for normal pigmentation, and in addition, cystine for normal emergence
(Golberg and DeMeillon, 1948). Drosophila seems to require, in addition to the
10 essential acids (Schultz et al., 1946; Rudkin and Schultz, 1947) glycine and
cystine (Hinton, Noyes and Ellis, 1951). Contrary to some of the aforementioned
authors, we have never had any indication that Tribolium benefitted by the presence
in the diet of cystine or glycine, nor did we find evidence of a toxic effect of Dl- or
L-serine as had been reported for Drosophila (Hinton et al., 1951).

Information about the nutritional value of the D-form of an essential amino
acid has been so far lacking for insects. In the amino acid requirements of man
for maintenance of nitrogen equilibrum, D-methionine was as effective as the L-form
and D-phenylalanine showed partial activity. The D-forms of valine, leucine,
isoleucine, threonine, lysine and tryptophane were inactive (Rose, 1949). In the
nutrition of the rat it is generally agreed that the D-forms of tryptophane, phenyl-
alanine and methionine show full or partial activity, while those of the remaining
essential acids are inactive (Rose, 1938; Rose et al., 1948; Nasset and Anderson,
1951). For Tetrahymena which, in addition to the 10 essential amino acids, also
requires serine, the D-forms of methionine, lysine and arginine are active, that of
leucine is inhibitory, and those of the remaining six acids are inactive (Elliott
et al., 1952). Tribolium utilizes fully or partly the D-forms of lysine, methionine
and phenylalanine. It therefore appears that the D-methionine is utilized by
Tetrahymena, Tribolium,, the rat and man, D-phenylalanine by Tribolium, the rat
and man, D-arginine alone by Tetrahymena, D-lysine possibly by Tribolium, and
D-tryptophane alone by the rat.

The authors wish to express their sincere thanks to Swift and Co., Chicago,
for the grant of a scholarship to one of us (G. E. P.). This investigation was also
supported in part by a research grant from the National Institutes of Health,
Public Health Service. The authors are greatly indebted to Dr. C. B. Berg, State
University of Iowra and Dr. A. A. Albanese, St. Luke's Convalescent Hospital,
Greenwich, Conn., for valuable gifts of D-arginine and D-lysine, respectively, and
to Merck & Co., for gifts of some of the amino acids.

SUMMARY

1. The larvae of the flour beetle Tribolium conjusum have been successfully
grown on diets which contain 19 amino acids or the 10 amino acids which are
essential in the nutrition of the rat. The larvae of two other beetles, Tenebrio
molitor and Dermestes vulpinus, failed to gro\v on similar diets.

2. Tribolium requires the following amino acids for growth : arginine, histidine,
lysine, tryptophane, phenylalanine, methionine, threonine, leucine, isoleucine and
valine.

3. On a mixture of 19 amino acids, which in addition to the above-named acids
also contains glycine, alanine, proline, hydroxyproline, glutamic acid, aspartic acid,
serine, cystine and tyrosine, growth is somewhat faster than in the presence of 10
amino acids.

AMINO ACID REQUIREMENTS OF TRIBOLIUM 157

4. Addition of any one of the non-essential acids to the mixture of the 10
essential ones has no marked effect. None of the amino acids exerted toxic effects
when added to the diet in double amounts.

5. Triboliutn utilizes fully or partly the D-form of methionine, phenylalanine
and, possibly, lysine. The D-forms of the 7 remaining essential acids were entirely
inactive, but did not show marked toxic effects.

LITERATURE CITED

ALMQUIST, H. J., AND C. R. GRAU, 1944. The amino acid requirements of the chick. /.

Nutrition, 28: 325-331.
ELLIOTT, ALFRED M., JAMES F. HOGG AND CHUNG Wu, 1952. Utilization of D-amino acids by

Tetrahymena geleii. Fed. Proc., 11 : 207.
FRAENKEL, G., AND M. BLEWETT, 1943. The vitamin B-complex requirements of several insects.

Biochem. J., 37 : 686-692.
FRAENKEL, G., AND M. BLEWETT, 1947. The importance of folic acid and unidentified members

of the vitamin B-complex in the nutrition of certain insects. Biochem. J., 41 : 469-475.
FRAENKEL, G., AND H. R. STERN, 1951. The nicotinic acid requirements of two insect species

in relation to the protein contents of their diets. Arch. Biochem., 30: 438-444.
FRENCH, E. W., AND G. FRAENKEL, 1954. Carnitine (vitamin BT) as a nutritional requirement

for Triboliitm conjusum Duval. Nature, 173 : 173.
FRIEDMAN, L., AND O. L. KLINE, 1950a. The amino acid-sugar reaction. /. Biol. Chem., 18 :

599-606.
FRIEDMAN, L., AND O. L. KLINE, 1950b. The relation of the amino acid-sugar reaction to the

nutritive value of protein hydrolysates. /. Nutrition, 40 : 295-307.
GOLBERG, D., AND B. DEMEiLLON, 1948. The nutrition of the larva of Aedes aegypti L. 4.

Protein and amino acid requirements. Biochem. J., 43 : 379-387.
HINTON, T., D. T. NOYES AND J. ELLIS, 1951. Amino acids and growth factors in a chemically

defined medium for Drosophila. Physiol. Zool., 24 : 335-353.
LEMONDE, A., AND R. BERNARD, 1951. Nutrition des larves de Tribolium conjusum Duval. II.

Importance des acides amines. Canad. J. Zool., 29 : 80-83.
MOORE, W., 1946. Nutrition of Attagenus (?) Sp. II. Coleoptera : Dermestidae. Ann. Ent.

Soc. Amer., 39: 513-521.
NASSET, E. S., AND J. T. ANDERSON, 1951. Nitrogen balance index in the adult rat as affected

by diets low in L- and DL-methionine. /. Nutritition, 44 : 237-247.
OFFHAUS, K., 1952. Der Vitaminbedarf des Reismehlkafers Tribolium conjusum. Zeitschr.

Vitamin, Horm. Perm. Forsch., 4 : 555-563.

ROSE, W. C., 1938. The nutritive significance of the amino acids. Physiol. Rev., 18 : 109-136.
ROSE, W. C., 1949. Amino acid requirements of man. Fed. Proc., 8 : 546-552.
ROSE, W. C., M. J. OESTERLING AND M. WOMACK, 1948. Comparative growth on diets con-
taining ten and nineteen amino acids with further observations upon the role of glutamic

and aspartic acids. /. Biol. Chem., 176: 753-762.
RUDKIN, G. T., AND J. SCHULTZ, 1947. Evolution of nutritional requirements in animals :

amino-acids essential for Drosophila melanogaster. Anat. Rec., 99 : 613.
SCHULTZ, J., P. ST. LAWRENCE AND D. NEWMEYER, 1946. A chemically defined medium for

the growth of Drosophila melanogaster. Anat. Rec., 96 : 540.

ANTIMITOTIC SUBSTANCES FROM OVARIES1

L. V. HEILBRUNN, ALFRED B. CHAET, ARNOLD DUNN AND

WALTER L. WILSON2

Department of Zoology, University of Pennsylvania, and the Marine Biological Laboratory,

Woods Hole, .Massachusetts

In a paper published two years ago (Heilbrunn, Wilson and Harding, 1951),
it was shown that a powerful antimitotic substance could be extracted from the
ovaries of the common starfish (Asterias forbesii). This substance tends to exert
a liquefying influence both on the cortex and on the interior protoplasm of marine
eggs, and because it prevents the mitotic gelation in somewhat the same way that
heparin does, we were led to believe that it might possibly be a heparin or heparin-
like substance. However, we had very little direct evidence to support this belief.

Accordingly, we have been eager to find out what we could as to the nature of
the antimitotic substance. The work is part of a broad general program in which
we seek to establish that all living material contains substances which favor pro-
toplasmic clotting and those which tend to prevent such clotting. It is now clear
that the colloidal behavior of protoplasm is quite similar to the behavior of verte-
brate blood ; and if this is true, it would be logical to suppose that the anticlotting
substances of living cells include heparins and heparin-like substances. For a
detailed discussion of protoplasmic clotting and how it influences not only cell
division but other vital processes as well, see Heilbrunn (1951, 1952a, 1952b).

The fact that we can obtain a potent antimitotic substance from the starfish
ovary is perhaps not surprising, for the eggs in the ovary do not divide until they
leave it. In our earlier paper (Heilbrunn, Wilson and Harding, 1951), it was
suggested that "perhaps the ovaries of many organisms are rich in heparin-like
substances." As will be seen later, this idea is apparently a fruitful one.

MATERIALS AND METHODS

In general, extracts were prepared in much the same manner as in our previous
work, except for the fact that instead of merely cutting up the ovaries, we
homogenized them before extraction with acidified sea water. Specific details
concerning the preparation of various individual extracts are given in relation to
individual experiments. When the extracts were dialyzed, cellulose dialysis tubing
was used. This was purchased from the Arthur H. Thomas Co. of Philadelphia,
and they state that according to the manufacturer, the Visking Corporation of
Detroit, the average pore diameter of the cellulose material is 24 Angstroms. The
purity of the cellulose is said to be very high, but it contains some glycerine and
approximately 0.1 per cent sulfur. During the process of dialysis, the tubes were
agitated on a shaking apparatus.

1 This investigation was supported by a research grant from the National Cancer Insti-
tute, National Institutes of Health, Public Health Service.

2 Department of Physiology and Biophysics, Colleges of Medicine, University of Vermont,
Burlington, Vt.

158

ANTIMITOTIC SUBSTANCES FROM OVARIES 159

In testing for antimitotic action, in most cases we used the eggs of a marine
worm, Chaetopterus pcrgamcntaceus. As in earlier works, the eggs were kept in a
constant temperature bath maintained at a temperature of 21° C.

Protoplasmic viscosity tests were made with an Emerson-type centrifuge. For
information about the use of the centrifuge in viscosity determinations, see Heil-
brunn (1950), Wilson and Heilbrunn (1952).

RESULTS

The substance we extract from starfish ovaries is presumably responsible for
the inhibition of mitosis in the ovary. As is well known, as soon as the starfish
eggs leave the ovary and enter sea water, the large nucleus of the immature egg,
that is to say, the germinal vesicle, breaks down and the maturation divisions
begin. In order to prevent the breakdown of the germinal vesicle and the subse-
quent maturation divisions, it is only necessary to leave the eggs in contact with
the ovary. Thus in one experiment, the ovaries of a starfish were cut up in 25 ml.
of sea water. Left in this sea water in the presence of the minced ovaries, only 2
per cent of the eggs showed germinal vesicle breakdown. However, with pro-

TABLE I

Effect of starfish ovary extract on division of Chaetopterus eggs. Eggs
exposed two minutes after fertilization

Dilution Per cent cleavage

1/100 0

1/200 0

1/400 93

1/800 97

Control 97

gressive dilution of the sea water which had been in contact with the ovary, there
was a progressive increase in the percentage of germinal vesicle breakdown, so
that when the original fluid was diluted 64 times, the percentage of germinal vesicle
breakdown rose to 69 per cent. The inhibitor effect of the ovarian substance is
largely reversible. As a matter of fact, this effect of ovarian substance both on
maturation and cleavage divisions has long been known to students of marine eggs,
and they commonly wash eggs two or three times before experimenting with them ;
that is to say, they pour off the sea water over the eggs and replace it with fresh
sea water, and then repeat this operation several times.

In the work reported previously (Heilbrunn, Wilson and Harding, 1951),
the extracts from starfish ovaries that we studied, when diluted to more than
1 part in 10, did not have much effect on Chaetopterus eggs. But in the extracts
that were prepared from homogenized ovaries, a dilution of 1 to 200 was still
effective. This is shown in Table I, which illustrates the effect on cleavage of
one of our extracts. To prepare this extract, 100 ml. of acid sea water at pH 5.8
were added to 50 g. of starfish ovaries, and the ovaries were then homogenized in
a Waring blendor. (The acid sea water was prepared as in our previous work.)
The homogenate was centrifuged in a Sorvall centrifuge at about 15,000 g. and
the resultant supernate was neutralized with 0.1 N NaOH so that its final pH
was 7.0.

160 HEILBRUNN, CHAET, DUNN AND WILSON

Nature of the antimitotic substance in starfish ovary extract. What is the
nature of the antimitotic substance in the extract from starfish ovaries? If we
knew that, then we might go ahead to discover various other antimitotic substances
in the hope that one or another of them would be useful in the treatment of
cancer. From the beginning, our suspicion has been that the potent substance in
our extracts was a heparin-like compound. Let us summarize the old and new
evidence in support of this opinion. Some of this evidence will later be presented
in more detail by individual members of our group.

1. The extract from starfish ovaries is strongly metachromatic, just as heparin
is. That is to say, the extract gives a reddish color with dilute solutions of toluidine
blue. Tests for metachromasy are best made in calcium-free sea water or distilled
water, for the calcium ions tend to prevent the metachromatic reaction.

2. The metachromatic reaction with toluidine blue disappears in the presence
of protein, just as the metachromatic reaction of heparin disappears in the presence
of protein (Kelly, 1951). If the crude extract from starfish ovaries is salted out
with varying concentrations of ammonium sulfate, the activity appears in the globulin
fraction and not in the albumin fraction. If now the globulin fraction is re-
suspended in 0.3 molar sodium chloride solution and digested with trypsin, the
resultant solution is metachromatic. This solution, after boiling to destroy trypsin,
exerts a strong anticoagulant action on sheep plasma and it also prevents cell
division in Chaetopterus eggs. Also if the re-suspended globulin fraction is
dialyzed for 48 hours against 0.3 molar sodium chloride, a metachromatic reaction
is obtained in the dialysate, and the dialysate is likewise effective in preventing cell
division in Chaetopterus eggs. These facts will be discussed more fully by one
of us (Dunn). The fact that the potent substance is able to pass through the
dialysis sac indicates that it is not a substance of high molecular weight. Perhaps
a correlation is to be found with the fact that, as Chaet (1952) has shown, ordinary
heparin in solution can break down and yield substances capable of passing through
a dialysis membrane and nevertheless capable of powerful physiological activity.
Chaet's experiments, so far presented only as a preliminary note, will be published
in extenso before long.

3. The active substance is heat-stable. Solutions containing it can be heated to
99.5° C. for 30 minutes and still retain their activity. On the other hand, when
the substance is combined with globulin as a result of the salting out procedure
described above, its activity is lost after exposure to a temperature of 80° C. for
20 minutes. (Following such inactivation, the active substance can no longer be
separated off by dialysis.)

4. The activity of the starfish ovary extracts is destroyed by dilute solutions
of periodate. Potassium periodate was added to potent extracts. Then the excess
periodate was removed by dialysis. The control, containing extract without
periodate, was dialyzed in similar fashion. The precipitates that formed were all
removed by centrifugation and the potency of the extracts was tested on Chaetop-
terus eggs. The periodate was completely successful in destroying the antimitotic
activity of the extracts. This is shown in Table II. These experiments are con-
sistent with the idea that a polysaccharide is responsible for the activity of the
extracts, but they do not constitute absolute proof of such an idea, for substances
other than polysaccharides may also be destroyed by periodate.

ANTIMITOTIC SUBSTANCES FROM OVARIES

161

TABLE II

Effect of potassium periodate on potency of starfish ovary extract

% cleavage

07

Exp. 1

Exp. 2

Exp. 3

Exp. 4

0 (control-extract alone)

0

0

0

0

0.25

96

84

100

0.5

93

92

94

0.75

98

95

96

1

96

96

Control

(no extract)

99

96

98

100

5. The dialysis behavior of the extracts is similar to that of heparin. As
already noted, Chaet (1952) has found that when preparations of commercial
heparin are dialyzed, a potent substance keeps coming through the dialysis mem-
brane. He believes that heparin is continually breaking down to produce more
of this substance. Similarly, when starfish ovary extract is dialyzed, a substance
passes through the membrane and this substance is strong in antimitotic properties.
However, the material that remains in the dialysis sac also prevents cell division.
Table III shows the results of an experiment on Chaetopterus eggs. Both the
substance or substances remaining in the sac after prolonged dialysis and the
substance that diffused through the sac into sea water completely inhibited cell
division. The active substance is not completely removed by a single dialysis

TABLE III

The effect of dialysis on the potency of starfish ovary extract in preventing cell division. (The extract

was first dialyzed against double its volume of sea water for 7 hours,

then against running sea water for 9 hours)

Contents of sac after dialysis

Dialysate

Control (dialyzed sea water)

Exp. 1

0

0

99

% cleavage

Exp. 2

0
0

99

TABLE IV

The effect of repeated dialysis on the potency of starfish ovary extract in preventing cell division. (The

extract was dialyzed against an equal volume of sea water for 12 hours and the antimitotic effect

of the dialysate tested. Then the contents of the sac were dialyzed against running sea water

for 12 hours, following which the contents were dialyzed against an equal volume

of sea water for 12 hours and the antimitotic effect

of the second dialysate tested)

Contents of sac (after 2 dialyses)
First dialysate
Second dialysate
Sea water dialysate
Second sea water dialysate
Control (sea water)

% cleavage

0

0

0
95
90
96

162 HEILBRUNN, CHAET, DUNN AND WILSON

operation. This is shown in Table IV, which gives the data on an experiment
in which a second dialysate was still strongly antimitotic and completely prevented
the division of Chaetopterus eggs. In this experiment, control tests were made
with dialyzed sea water, for it is sometimes found that dialysis tubes give off
substances that have some slight antimitotic action. (This might well be expected
from the fact that, as previously noted, the cellulose dialysis tubes consist of poly-
saccharide containing a little sulfur.) In another experiment, an active antimitotic
substance continued to pass through the dialysis membrane after seven successive
dialyses. It is quite possible that when a living cell is exposed to a heparin-like
substance or to a combination of such a substance with protein, breakdown products
of the heparin-like substance diffuse into the cell, whereas the components of larger
molecular size remain outside.

6. When an active extract of starfish ovary is placed in a dialysis sac, carbo-
hydrate diffuses through the sac into the surrounding fluid. This is shown clearly
by chromatographic tests. The carbohydrate is a polysaccharide. Chromato-
graphic tests of the dialysate also indicate the absence of nucleic acids and amino
acids. Details of these tests will be published later (by Dunn).

7. The ultraviolet absorption spectrum of starfish ovary extract is similar
to that of heparin. This is shown in Figure 1. In this figure, the open circles
show the absorption spectrum of a 0.17 per cent solution of sodium heparinate,
kindly supplied by the Upjohn Co. of Kalamazoo, Michigan. The closed (com-
pletely black) circles show the absorption spectrum of a 2 per cent solution of the
globulin fraction of starfish ovary extract in 0.5 M NaCl. In the preparation of
this fraction, 100 g. of starfish ovary were washed in 0.5 M NaCl for one hour
to remove excess mucus ; the washed ovaries were then suspended in 200 ml. 0.3 M
NaCl and homogenized in a Waring blendor. Following centrifugation, the globulin
fraction of the supernatant solution was salted out with half-saturated ammonium
sulfate solution, and the precipitate was made salt-free by dialysis against distilled
water. The triangles show the spectrum of a highly dilute solution of starfish ovary
extract. This was prepared from an extract made by extracting 20 gm. of homo-
genized starfish ovary in 40 ml. of acid sea water (pH 5.8 ) and then neutralizing the
resultant solution. This extract was then diluted with ordinary sea water until it
was only 0.26 per cent of its original strength. The dilution was made in order to
obtain a curve at about the same position on the graph at the heparin curve. The
readings on the (Beckman) spectrophotometer were made by Dr. Lester Goldstein.

The general similarity of the three curves in Figure 1 is obvious, and although
this similarity does not provide proof that the starfish ovary extract does actually
contain a heparin-like substance, it is certainly consistent with such a view.

Indeed, no one of the seven arguments that we have presented is in itself very
cogent, but taken as a whole they do indicate rather strongly that the starfish ovary
extract contains a heparin-like substance and that this substance is responsible
for its anticlotting and antimitotic activity.

Antimitotic substances in fish ovaries. In our thinking about the starfish ovary
extract, we were bothered by two facts. In the first place, we knew of no evidence
in the biochemical literature of heparin or heparin-like compounds splitting to form
compounds of lower molecular weight capable of passing through dialysis mem-
branes. Secondly, and this from a practical clinical standpoint is more important,

ANTIMITOTIC SUBSTANCES FROM OVARIES

163

we were soon led to believe that the starfish ovary extract which had so drastic a
liquefying action and so strong an antimitotic effect on invertebrate marine eggs,
was rather powerless on vertebrate cells and tissues. The starfish ovary extract
acts on starfish eggs, Chaetopterus eggs, eggs of the sea urchin Arbacia and eggs
of the clam Spisula ; but we found no very great antimitotic activity on frog eggs,
and the antimitotic action of the extract on embryo mouse cells in tissue culture
was much less than that of ordinary commercial heparin. [The studies on tissue
culture cells were made by Carol Bocher and were presented by her as a Master's
thesis (Bocher, 1952).] Moreover, although ordinary heparin, or a breakdown

0.55

220 240 260

WAVE LENGTH,

280

300

FIGURE 1. Comparison of the absorption spectra of heparin and two samples of starfish
ovary extract. The open circles show the absorption spectrum of sodium heparinate. For
further details, see text.

164 HEILBRUNX, CHAET, DUXX AND WILSON

product of it, stops the frog heart in diastole (Kraus, Fuchs and Merlander, 1931 ;
Chaet, 1952), starfish ovary extract has no effect on frog heart, although it does
stop the clam heart in diastole. Also, in a few preliminary experiments, we found
that starfish ovary extract had no obvious toxic action when injected subcutaneously
into mice.

Fortunately, our attention was called to the fact that some fish of the family
Tetraodontidae have in their ovaries a substance which has a very potent pharmaco-
logical and toxic action. This substance stops the heart of a toad or frog in
diastole, and it thus acts like heparin, or rather like the breakdown product of
heparin studied by Chaet (1952). Partly because of the toxic nature of this
substance — it kills a few people in Japan each year — the literature concerning it is
quite voluminous and goes back several hundred years. Useful papers include
those of Tawara (1910), Ishiwara (1924), and Yudkin (1944, 1945). The sub-
stance has been called tetrodotoxin, and it is manufactured under that name in
Japan. There it is used in the treatment of neuralgia and arthritis. Chemical
study of tetrodotoxin has indicated that it is a carbohydrate of no very great
molecular weight, containing both amino nitrogen and sulfur ; also it is precipitated
by alcohol. In all these properties, except for its low molecular weight, it resembles
heparin. Like heparin also, it has an inhibitory effect on the clotting of bird and
mammalian blood.

Here, then, is the kind of substance we have been looking for. Also, it is an
ovarian substance and therefore of particular interest to us. Accordingly, we
began experiments on the antimitotic properties of substances in the ovaries of the
common puffer of the Atlantic coast, Sphcroides maculatus. This fish belongs to
the Tetraodontidae, and extracts of its ovaries stop the frog heart in diastole
(Yudkin, 1945). The work on the puffer ovary is being done by Pierre Couillard,
who first called our attention to the paper by Yudkin. Suffice to say at this point
that extracts of the puffer ovary actually are antimitotic ; they act in much the same
manner as extracts of starfish ovary.

If it is true, as we have suggested, that ovaries in general may be rich in anti-
mitotic substances, then it ought to be possible to extract such substances from the
ovaries of many fishes, rather than just those of the family Tetraodontidae.
Accordingly, we made extracts of the ovaries of some common fishes and tested
their effect on the eggs of Chaetopterus. The extracts were made in much the
same manner as the extracts of starfish ovaries. The following experiment will
serve as an example.

Ovarian material was gathered from females of the small fish, Fundulus
heteroclitus. To 10 gm. of this material, 10 ml. of acid sea water at a pH of 5.3
were added. (The acid sea water had had its bicarbonates removed.) The ovaries
were then homogenized in a small glass homogenizer, and the homogenate diluted
to a volume of 25 ml. and centrifuged in a high-speed Sorvall centrifuge. The pH
of the supernatant solution was 6.15; it was brought to a pH of 8.00 with 0.1 N
NaOH, and the resultant solution was used in the experiment. Chaetopterus eggs
were fertilized, and then two minutes after fertilization, some of the eggs were
placed in each of 4 dishes, A, B, C, D. The first dish, A, contained full strength
extract ; in B the extract was diluted with an equal volume of sea water, so that the
resultant mixture was half strength. C had three parts of sea water for each

AXTIMITOTIC SUBSTANCES FROM OVARIES 165

part of extract, and D had seven parts of sea water for each part of extract. Of
the control eggs in sea water, 100 per cent gave off polar-bodies, but in A, polar-
body formation was completely inhibited. In B, there was 12 per cent polar-body
formation ; in C, 27 per cent, and in D, 42 per cent. Thirty minutes after fertiliza-
tion, at a time when it is known that the gelation has developed (Heilbrunn and
Wilson, 1948), centrifuge tests were made rapidly with a hand centrifuge, as in
previous studies from this laboratory. These tests showed that whereas the proto-
plasm of normal eggs had a viscosity well above the arbitrary value of 8 (and
presumably, in accordance with earlier studies, about 14), the eggs in A exposed
to the full strength Fundulus ovary extract had a protoplasmic viscosity of about 4.
The eggs in B had a protoplasmic viscosity value of about 6. At 55% minutes after
fertilization, 50 per cent of the control eggs had cleaved, and a few minutes later,
97 per cent had cleaved. No one of the eggs in A ever cleaved, and only 25 per
cent of those in B. The C eggs showed 27 per cent cleavage, and the D eggs 31
per cent. Some of the eggs from A and B were transferred to ordinary sea water.
The eggs in A were badly injured, and following transfer to sea water they did
not divide. When the eggs in B were transferred to sea water (after a 63-minute
exposure to the extract), 53 per cent showed cleavage. Thus the effect of the
extract is to some extent reversible.

Results similar to these were obtained with extracts from the ovaries of various
other fishes. In all our studies with fish ovary extracts, we noticed that the active
antimitotic substances tended to lose their potency in a relatively short time. Thus
when the experiment with the Fundulus ovary extract was repeated a day later
with the same extract, which had been kept overnight in a refrigerator, the effect
both on the cleavage and on the protoplasmic viscosity was decidedly less. It
should be noted that the extracts we use are very crude ; no doubt in addition to
anticlotting substances, they contain thromboplastic substances which promote
clotting. When the extracts age, the effect of these thromboplastic substances may
tend to override the effect of the anticlotting substances.

Because of the fact that powerful anticlotting and antimitotic substances are
found not only in the ovaries of fishes of the family Tetraodontidae but also in
ovaries of other fishes as well, we began to wonder if there might not be some
evidence to indicate a pharmacological action or a toxicity of ovaries of fish not
belonging to the Tetraodontidae. Literature in support of this idea does indeed
exist, but it is very hard to assemble. For over four hundred years, scientific
writers have commented on poisoning due to the eating of fish, but many of the
articles that were written are in obscure journals, difficult of access. Gudger
gathered together numerous references for Dean's bibliography of fishes, published
in 1916-1923, and in the fifteen years following the collection of these references,
he was able to find 180 others (Gudger, 1930). In the West Indies, there is a
special word — ciguatcra — that means fish poisoning ; and in the East Indies and the
South Seas, there is frequent reference to fish poisoning. Useful sources of infor-
mation include papers by Taft (1945), Cohen, Emmert and Goss (1946), Von-
fraenkel and Krick (1945), Gilman (1942), and Gudger (1918, 1930). Books
by Phisalix (1922) and Pawlowsky (1927) may also be consulted. When men
are poisoned by eating fish in the tropics, there is often uncertainty as to the

166 HEILBRUNX, CHART, DUXX AND WILSON

cause. Always there is a possibility that the fish may have spoiled ; also there
is an old superstition that fish become poisonous because they have eaten poisonous
fruit. In the case of the barracuda, often found to be poisonous, it seems clear
that only larger fish contain poison and then only at certain seasons of the year
(Chisholm, 1808). This seems to indicate that the gonads may be involved.
According to Coker (1930), the roe of garpikes (genus Lepisosteus) is said to be
toxic, and in Germany it is well known that the ovaries of the barbel, a large
cyprinid fish (Barbus vulgaris) are poisonous. According to McCrudden (1921),
the ovaries of the pike are even more toxic than those of the barbel. Kohler, in
1933, writing in a magazine for practicing physicians, states: (p. 292) "Manche
an sich ungiftige Fische geben zur Laichzeit unter nicht naher bekannten Verhalt-
nissen Ursache zur Yergiftungen." There is thus clear indication that fish ovaries
may contain potent substances, substances which under certain conditions have a
serious effect when ingested.

In fishes, the ovaries are not alone in containing substances that prevent the
clotting of protoplasm. This is only to be expected if, as we believe, all types of
living material contain anticlotting as well as clotting factors (see Heilbrunn,
1952b). We were not surprised, therefore, that when we extracted the testis
of the toadfish (Opsanus tan) in the same manner that we extracted the ovary,
we were able to obtain a substance that kept the protoplasm fluid and prevented
cell division. As a matter of fact, in the Tetraodontidae, the testis is toxic as well
as the ovary (Remy, 1883), and apparently contains the same type of substance
that the ovary does. Also it will be remembered that the starfish testis contains
much antimitotic substance (Heilbrunn, Wilson and Harding, 1951). In animals
that breed only once a year, at times when little or no mitosis is occurring in the
testis, this organ may presumably be rather rich in substances which prevent cell
division. We found too that fish liver might also contain easily recognizable
amounts of anticlotting and antimitotic substances — we used the liver of the angler
or goosefish (Lophins piscatorius} . This is to be correlated with the fact that
the liver is a ready source of heparin and also with the fact that in Tetraodon
the liver may be poisonous as well as the ovary (Tani, 1940). It is possible
that the tetrodotoxin of the ovary is secreted in the liver.

The results reported in this paper provide additional evidence to show that
many diverse types of living tissue, and indeed possibly all types of living cells, con-
tain substances that prevent the clotting of protoplasm and exert a powerful anti-
mitotic action. The ovaries of many animals are especially rich in such substances.
Many of the anticlotting substances are either heparins or heparin-like substances.
There is good reason to believe that the various substances vary widely both in
molecular size and molecular composition.

At the present time the search for antimitotic substances in various organs and
tissues of various organisms is being continued. There is undoubtedly a large
and diverse group of naturally occurring heparin-like compounds which can act
as anticlotting and antimitotic agents. Out of this large group of compounds, it
should be possible to discover some which may be of real value in the treatment
of tumors. More work is urgently needed. We need to know more about the
chemistry of these heparin-like compounds, and their effect should be tested not only
on simple isolated cells, but also on tumors.

ANTIMITOTIC SUBSTANCES FROM OVARIES 167

SUMMARY

1. Starfish ovaries contain a substance which prevents maturation divisions
in the eggs contained in these ovaries.

2. By homogenizing starfish ovaries before extracting them with acid sea water,
we have been able to prepare antimitotic extracts much more powerful in their
action on Chaetopterus eggs than the extracts reported on previously.

3. There is additional many-sided evidence to indicate that the potent substance
in these extracts is a heparin-like compound. Some of this evidence comes from
chromatographic studies ; also from studies of the absorption spectrum of the extract.
Moreover, the potency of the extract disappears after treatment with periodate.

4. Ovaries of various species of fishes contain antimitotic substances which
resemble in their action the substance or substances in starfish ovary extracts. In
at least one family of fishes, the ovaries are known to contain a potent substance
of heparin-like chemical composition and heparin-like properties.

LITERATURE CITED

BOCHER, C., 1952. The effect of an extract of starfish gonads and heparin upon cells grown

in tissue culture. Master's thesis, University of Pennsylvania, Philadelphia.
CHAET, A. B., 1952. The action of heparin preparations on cells and tissues. Biol. Bull.,

103 : 281.

CHISHOLM, C., 1808. On the poison of fish. Edinburgh tiled, and Surg. Jour., 4: 393.
COHEN, S. C., J. T. EMERT AND C. C. Goss, 1946. Poisoning by barracuda-like fish in the

Marianas. U. S. Naval Bull., 46: 311-317.
COKER, R. E., 1930. Studies on common fishes of the Mississippi River at Keokuk. Bull. U. S.

Bureau Fisheries, 45 : 141-225.

DEAN, B., 1916-1923. A bibliography of fishes. 3 vols. Amer. Museum Nat. History, N. Y.
GILMAN, R. L., 1942. A review of fish poisoning in the Puerto Rico-Virgin Islands area.

U. S. Naval Bull, 40 : 19-27.
GUDGER, E. W., 1918. Sphyraena barracuda, its morphology, habits, and history. Carnegie

hist. Wash., 12 (Publ. no. 252) : 53-108.
GUDGER, E. W., 1930. Poisonous fishes and fish poisonings, with special reference to ciguatera

in the West Indies. Amer. J. Trop. Med., 10: 43-55.
HEILBRUNN, L. V., 1950. Viscosity measurements. Chap. 4 of F. M. Uber's Biophysical

research methods. Interscience Press, N. Y.
HEILBRUNN, L. V., 1951. The colloid chemistry of narcosis. In: Mecanisme de la Narcose

(Colloques Internat. du Centre. Nat. de la Recherche Sci.), pp. 163-178.
HEILBRUNN, L. V., 1952a. The physiology of cell division. In: Modern trends in physiology

and biochemistry. Academic Press, N. Y., pp. 123-134.
HEILBRUNN, L. V., 1952b. An outline of general physiology. W. B. Saunders Co., Philadelphia.

Third Ed.
HEILBRUNN, L. V., AND W. L. WILSON, 1948. Protoplasmic viscosity changes during mitosis

in the egg of Chaetopterus. Biol. Bull., 95 : 57-68.
HEILBRUNN, L. V., W. L. WILSON AND D. HARDING, 1951. The action of tissue extracts on

cell division. /. Nat. Cancer Inst., 11 : 1287-1298.

ISHIWARA, F., 1924. Studien iiber das Fugutoxin. Arch. f. c.rp. Path. u. Pliann.. 103: 209-218.
KELLY, J. W., 1951. Effects of x-ray and trypsin on the metachromasy of heparin-basic protein

combinations. Biol. Bull., 101 : 223.

KOHLER, F., 1933. Muschel- und Fischvergiftung. Fortschrittc d. Mcd., 51 : 291-292.
KRAUS, F., H. J. FUCHS AND R. MERLANDER, 1931. Uber das Koagulin des Muskels III.

Zeitschr. f. d. ges. exp. Med., 79: 59-75.
MCCRUDDEN, F. H., 1921. Pharmakologische und chemische Studien uber Barben- und Hecht-

rogen. Arch. f. exp. Path. u. Pharm., 91 : 46-80.
PAWLOWSKY, E. N., 1927. Gifttiere und ihre Giftigkeit. Fischer, Jena.

168

HEILBRUNN, CHART, DUNN AND WILSON

PHISALIX, M., 1922. Animaux venimeux et venins. 2 vols, Masson, Paris.
REMY, C., 1883. Sur les poissons toxiques du Japon. C. R. Soc. de Biol., 35 (Mem.) : 1-28.
TAFT, C. H., 1945. Poisonous marine animals. Texas Reports on Biol. and Med., 10 : 43-55.
TAXI, I., 1940. Changes and individual variations of the toxicity of "hugu" (globe fishes).

Jap. J. Med. Sci., IV. Pharm., 13: 174-177.

TAWARA, Y., 1910. Uber das Tetrodongift. Biochcm. Zcitschr.. 30: 255-275.
VONFRAENKEL, P. H., AND E. S. KRicK, 1945. Fish poisoning by barracudas in the Marianas.

U. S. Naval Bull, 40 : 19-27, 1942.
WILSON, W. L., AND L. V. HEILBRUNN, 1952. The protoplasmic cortex in relation to stimulation.

Biol. Bull, 103 : 139-144.
YUDKIN, W. H., 1944. Tetrodon poisoning. Bull. Bingham Oceanographic Collection, 9 (1) :

1-18.
YUDKIX. W. H., 1945. The occurrence of a cardio-inhibitor in the ovaries of the puffer,

Sphcroides maciilatns. J. Cell. Comp. Physiol., 25: 85-95.

FURTHER INVESTIGATIONS ON THE INTERACTION BETWEEN
SPERM AND JELLY-COAT IN THE FERTILIZATION OF

THE SEA URCHIN EGG l

A. MONROY, L. TOSI.- G. GIARDINA AND R. MAGGIO

Laboratory oj Comparative .•Inatomy, The Uniz'crsity of I'd'cmio. I'ntcruio. Italy

The role played in fertilization by fertilizin, the sperm-agglutinating substance
of egg water, was first emphasized by F. R. Lillie (1914). This substance has
been shown (Tyler and Fox, 1939, 1940; Tyler, 1940) to be derived from the
gelatinous coat of the egg and the evidence shows no other macromolecular con-
stituents to be present in the latter. The gelatinous coat slowly dissolves as the
eggs stand in sea water, providing thus the egg water of Lillie (1914), who had
erroneously supposed its agglutinating property to be due to some substance
actively secreted by the egg. Eggs deprived of their gelatinous coat are still
fertilizable but require insemination with larger amounts of sperm (Tyler, 1941).
In such eggs there is evidence (Tyler, 1941 ct scq.) that a layer of fertilizin remains
as a part of the vitelline membrane of the surface of the egg proper. It appears
likely that reaction of the sperm with fertilizin on the surface of the egg is essential
for successful fertilization to occur (see Tyler, 1948a, 1948b, 1949a, 1949b).

The well known fact that a jelly-coat solution causes sperm agglutination sug-
gests already either a binding of jelly-coat molecules to the surface of the sperm
or at least a modification of the sperm surface induced by the jelly-coat. The
observation that the organic material of a purified solution of sea urchin jelly-coat
can be practically completely adsorbed by sperm (Tyler, 1948b), speaks strongly
in favor of the former interpretation to which the results presented in this paper
give further support.

Our investigations started from the study of the mechanism by which the sperm
are able to go through the jelly-coat in fertilization. An enzymatic mechanism
was suggested by the finding of a jelly-coat splitting enzyme in extracts of sea
urchin sperm (Monroy and Ruffo, 1947; Lundblad and Monroy, 1950; Yasseur,
1951). The activity of this enzyme was, however, weak. Furthermore its exist-
ence in some species of sea urchin was questioned (Krauss, 1950). Monroy and
Tosi (1952), re-investigating the matter, interpreted their new results as being
contrary to the early assumption of an enzymatic splitting and discussed the
possibility of a quite different mechanism by which the sperm may be able to find
its way to the egg through the jelly-coat.

1 This investigation has been supported by grants of the Istituto Superiore di Sanita.
Rome and Consiglio Nazionale delle Ricerche. A generous gift of chemicals by the
Farmitalia, S. A. is also acknowledged.

2 Present address : Stazione Zoologica, Napoli.

169

170 MOXROY, TOSI, GIARDINA AND MAGGIO

Summary of the experiments suggesting a non- enzymatic splitting of the jelly-coat
by the sperm ( Monroy and Tosi, 1952)

The previously mentioned conclusion on the non-enzymatic splitting of the
jelly-coat by the sperm rested upon the following experimental evidence :

1. Living sperm added to a jelly-coat solution agglutinate and cause a sudden
decrease of viscosity of the jelly. Irrespective of the quantity of sperm added, the
viscosity drops in the course of about one minute and then remains constant.
Washing of the agglutinated sperm with sea water does not restore their ability
either to be re-agglutinated (see Lillie, 1914; Tyler, 1948b) or to "depolymerize"
a fresh viscous solution of jelly-coat. That suggests that the reactive groups at
the surface of the sperm have been blocked by the first reaction with the jelly-coat.

2. The interesting observation was recently made (Metz and Donovan, 1951)
that fixed sperm can be agglutinated by jelly-coat solutions. We have also found
that fixed sperm added to a solution of jelly-coat are able to "depolymerize" it just
as well as living sperm do (for further details see Monroy and Tosi, 1952).

From these experiments it was suggested as a working hypothesis that in the
reaction between sperm and jelly-coat, surface groups of the former react with
groups in the jelly. As a result of this reaction fragments of the jelly-coat sub-
stance would become attached to the sperm surface. That would account for the
apparent depolymerization of the jelly-coat.

THE NEW EXPERIMENTS

In order to develop further the analysis of the reaction between jelly-coat and
sperm, two kinds of experiments were undertaken :

1. Attempts to recover the agglutinating factor from the agglutinated sperm

2. Quantitative estimation of some typical component of the jelly-coat substance
before and after reaction with sperm

Material

Jelly-coat and sperm of Arbacia li.rula were used throughout these experiments.
Jelly-coat solution was prepared according to Tyler (1949b). Sperm were
obtained by spontaneous shedding following cutting of the shells and were
centrifuged once at 500 X g for 10 minutes in order to remove impurities and
excess of seminal fluid. In our experience, washing with sea water prior to
agglutination proved to be detrimental to the sperm.

1. Recovery of the agglutinating factor from agglutinated sperm

To a jelly-coat solution in sea water, sperm were added to saturation, i.e.,
to the point when the solution had lost its agglutinating ability. The sperm were
then centrifuged in the cold, washed twice with cold sea water and finally suspended
in 4% formaldehyde at room temperature for one hour, being gently stirred from
time to time. The suspension was centrifuged at high speed in the cold and the
supernatant dialyzed against several changes of sea water and then tested for

INTERACTION OF SPERM AND JELLY-COAT 171

agglutinating ability. Non-agglutinated sperm were treated similarly and the
solution used as a control.

The extract from agglutinated sperm proved to have agglutinating power,
whereas no effect whatsoever was obtained with the extract from non-agglutinated
sperm.

Attempts to elute the agglutinating factor from agglutinated sperm by possibly
a less drastic procedure (changes of pH, incubation at 37° C, salt solutions) were
unsuccessful. In most cases, indeed, nucleic acid passed into the solution in rather
large amounts and the solution was devoid of any agglutinating activity. A very-
small contamination with nucleic acid was actually found also in the formalin extract
(small absorption peak in the U. V. at 260 m/*) ; we do not know whether this fact
may be of any importance.

In the discussion following the presentation of these data by one of us (A. M.)
at the Symposium on "The biochemical basis of morphogenesis" (Utrecht, August,
1952), Prof. Runnstrom objected that on account of the great ability of jelly-coat
molecules to polymerize, especially under the influence of Ca-ions, it is possible that
after a certain number of jelly-coat molecules have reacted with the sperm surface,
other molecules may simply form a cloud around them. The recovered agglutinat-
ing ability may thus be due to the secondarily linked jelly molecules rather than to
those which have become linked directly to the sperm surface.

The possibility of the formation of a cloud of jelly-coat molecules around the
sperm, after the first ones have reacted with it, although difficult to imagine on
the basis of our immunological knowledge, cannot be ruled out with certainty.
This, however, does not invalidate the main fact that in the reaction between jelly-
coat and sperm, jelly-coat molecules become attached to the sperm surface.

2. Importance of salts for the molecular architecture of the jelly-coat

The view that Ca-ions play an important role in the molecular architecture of
the jelly-coat has been entertained especially by Yasseur ( 1949) who conceives the
polysaccharide chains of the jelly-coat substance as being held together by Ca-
bridges. Our experiments, however, in which the effect of addition of CaCl, on
the viscosity of jelly-coat solutions was tested, have demonstrated that Ca causes
a decrease of viscosity of the latter (Fig. 1). Much more dramatic is the effect
of a 3.6% NaCl solution and of sea water, the former causing a drop of viscosity of
60% and the latter a drop of 70%. Urea, on the contrary, has no effect, thus
suggesting that H-bonds play no part in holding together the molecules of the jelly-
coat.

Furthermore there is another observation to which we would like to draw
attention. The jelly-coat stains metachromatically with toluidine blue both in vivo
and in vitro: as it is known this is a reaction given by all sulfonated polysaccharides.
Now, in vitro this reaction is positive only if the jelly-coat is dissolved in distilled
water whereas it is negative, i.e., there is no metachromatic colour change, if the
jelly-coat is dissolved in sea water, probably because under these conditions the
— SO3 groups of the jelly are present in a salt form. All this would suggest that
the native jelly-coat must have a very low salt content, if any, and the — SO3 groups
are either free or linked to some anionic group by very weak bonds. Evidently

172

MONROY, TOSI, GIARDINA AND MAGGIO

when the jelly-coat is brought into solution by the ordinary acid treatment the
—SO:, groups become free to react and they do in fact react with the ions of
the sea water.

To the influence of salts is also probably due the swelling and dissolution of the
jelly-coat of unfertilized eggs when standing in sea water.

3. Quantitative estimation of a component of the jelly-coat substance before and
alter reaction with spcnu

In order to obtain further evidence for the attachment of the jelly-coat material
to the surface of the sperm, experiments were performed to determine quantitative
changes in some typical component of the jelly-coat as a result of its reaction with
sperm. Analyses of sea urchin jelly-coat have proved it to be a hexosamine-free
sulfonated glycoprotein ( Yasseur, 1949, 1950; Tyler, 1948a, 1949a, 1949b). The

DIST. WATER

6 M UBEA

0,00-1 M CACI2

0.01 M CACI2

0.4 M

O.e M

SEA -WATER

FIGURE 1. Effect of different salt solutions and urea on the viscosity of a jelly-coat solu-
tion in distilled water. The viscosity of the undiluted sample is taken as 100. To 3.0 cc. of
this, 0.5 cc. of the solutions to be tested were added.

carbohydrate components are different in the different genera thus far studied
(reviewed by Runnstrom, 1951 ). As to the jelly-coat of Arbacia li.nila, fucose is
by far its largest component, only small amounts of galactose being also present
(Minganti, personal communication). Fucose estimation in the jelly-coat solution
before and after reaction with sperm was therefore thought to be a good indication
for the attachment of jelly-coat substance to the sperm. Such analyses are com-
plicated by the fact that -Irbacia sperm in sea water and in jelly-coat solution
release a fucose-containing phosphate ester. The study of the latter phenomenon
is being carried out by one of us ( L. T.). The results of the analyses therefore
had to be corrected for a blank in which sperm were suspended in sea water.

Experimental

In a typical experiment, to 1.0 cc. of jelly-coat solution in sea water, 2.0 cc.
of sperm suspension were added and after 30 seconds the agglutinated sperm were

INTERACTION OF SPERM AND JELLY-COAT 173

centrifuged off. The supernatant was collected quantitatively, brought to 4.0 cc.
with sea water and 1.0 cc. of chilled 10' 'c perchloric acid was added. The solution
was kept in the cold for one hour and then centrifuged. The addition of perchloric
acid was found to be necessary as otherwise sperm may remain in suspension and
interfere with the colorimetric estimation of fucose. Sea water was substituted
for jelly-coat in the blank for the fucose released by the sperm. The zero point,
i.e., the estimation of fucose content of the jelly-coat before reacting with the sperm,
was done diluting 1.0 cc. of the jelly-coat solution to 4.0 cc. with sea water and
then adding 1.0 cc. of perchloric acid. Aliquots of these solutions were used for
fucose determination according to Dische and Shettles (1948). Three fucose
standards were run with each analysis.

The results of these analyses are summarized in Table I.

Quite independently, similar results have been recently communicated in a short
paper by Hultin et al. (1952).

These experiments conclusively indicate that the interaction between jelly-coat
and sperm consists of a reaction in which surface groups of the sperm react with

TABLE I
Effect of the treatment with sperm on the fucose content of Arbacia jelly-coat solution

Mg fucose/cc. in
Jelly-coat control Sperm treated jelly-coat*

234.0 44.0

9.8 4.2

120.0 49.0

19.8 2.2

42.0 25.6

* Corrected for the fucose released by the sperm in sea water.

and bind to some groups in the jelly. Xow, Metz and Donovan (1951 ) have shown
that alkylation of the - — NH, groups in the sperm results in prevention of the
agglutination by jelly-coat solution. That makes it highly probable that the reaction
occurs between -— NH., groups at the surface of the sperm and -— SO:; groups in
the jelly-coat.

Model experiments

If the conclusion as to the groups entering in the reaction between jelly-coat
and sperm is correct, it must be possible to duplicate the experiment using ion-
exchangers having reactive groups similar to those of the sperm. Anionic exchange
resins in fact have proved to be a satisfactory model. Most of our experiments were
carried out with Amberlite IR 4B which is a weakly basic phenol-formaldehyde type
resin with =N as an active group. Before use, the resin was blotted on filter
paper to avoid dilution of the sample. Jelly-coat solution in distilled water was
shaken with the resin and the viscosity was measured before and in the course
of the experiment. As seen in Table II, treatment of the jelly-coat with the resin
causes a drop of viscosity of the jelly. When the viscosity had reached that of
water, a very low agglutinating titer was found and chromatography showed that
the fucose spot had entirely disappeared, while the one of galactose was apparently

174

MONROY, TOSI, GIARDINA AND MAGGIO

unchanged (Fig. 2). This different behavior of fucose and galactose is very
peculiar and deserves further attention.

The resin used in the treatment of the jelly-coat solution was thoroughly washed
with distilled water and eluted with 0.5-2% Na-bicarbonate or 5% ammonium
sulphate. The eluate, after dialysis against distilled water and sea water, proved
to be endowed with agglutinating ability.

That by this procedure the jelly-coat fraction adsorbed on the resin can be
recovered almost quantitatively is demonstrated by the following experiment.

Two samples of 3.0 cc. of jelly-coat solution in distilled water were shaken with
about 3.0 g. of resin. The supernatant was collected quantitatively, the resin
washed three times with distilled water and the washings added to the first super-
natant. Finally the material adsorbed on the resin was eluted as previously
described and fucose estimated in the supernatant and in the eluate (Table III ).

TABLE II

Effect of treatment with Amberlite IR-4B on viscosity, fucose content
and agglutinating liter of Arbacia jelly-coat solution

% control

% treated

Fucose Mg/cc. in

Agglutinating liter

Control

Treated

Control

Treated

1.52
12.10

1.0
2.0

present*
350.0

absent*
125.5

not tested
not t<

1/50
jsted

2.48
4.23
14.5

1.01
1.04
1.92

present*
136.0
460.0

absent*
1.0
185.0

not t<
> 1/10.000
1/100.000

;sted
1/50
1/5000

* Qualitative estimate from paper chromatography.

CONCLUSIONS

The results of the present experiments suggest that in the reaction between
sperm and jelly-coat substance, "molecules" of the latter adhere to the sperm surface.
As previously mentioned, a similar point of view has been defended also by Tyler
(1948b). Coating of bacteria by mucin is a well known phenomenon (reviewed
by Olitzki, 1948). We think, however, that in this case not a simple coating but
an actual chemical reaction occurs. Very likely the reaction is between the — NH2
groups at the surface of the sperm and — SO3 groups in the jelly-coat. This reac-
tion may also account for the so-called depolymerization of the jelly-coat caused
by the sperm. In a previous communication (Monroy and Tosi, 1952) it was
suggested that as a consequence of this reaction the molecules of the jelly-coat would
undergo fragmentation. As, however, nothing is known as to the length of the
molecules of the jelly-coat before and after reaction with the sperm, it is difficult at
present to decide whether a fragmentation actually occurs or the decrease of viscosity
is due to the fact that whole jelly-coat molecules bind to the sperm surface.

Following the multivalence theory of antigens and antibody of Heidelberger
(1939), Tyler thinks (1948b) that fertilizin molecules (i.e., jelly-coat molecules)
may each bind with two or more spermatozoa and each spermatozoon may in turn

INTERACTION OF SPERM AND JELLY-COAT

175

St

c

FIGURE 2. A chromatogram showing the disappearance of the fucose spot after treatment
•of a jelly-coat solution with Amberlite IR 4B(Tr). C = untreated jelly-coat. The greater
intensity of the galactose spot in the treated sample is due to the higher concentration of the
hydrolysate. St = standards of : Ga = Galactose ; Gl = glucose ; M = Maltose ; F = Fucose.

bind to several iertilizin molecules. According to our findings, the jelly-coat in
the living state should have a very low salt content, if any, whereas when dissolved
in sea water its -— SO3 groups probably react and bind with the ions of the sea
water. However, the jelly-coat in sea water not only agglutinates sperm but
Vasseur has maintained (1949) that Ca reinforces the agglutinating power of the
jelly-coat solution. That may be an indication that the type of bond that is estab-

TABLE III

Recovery of fucose after treatment of jelly-coat solution with Amberlite IR-4B

Mg Fucose in

Control
(untreated)

1002.0
91.0

Eluate

119.0
20.4

Treated

Supernatant

850.0
65.6

176 MOXROY, TOSI, GIARDIXA AXI) MAGGIO

lished between sperm surface and jelly-coat is different when the sperm cross the
jelly-coat at fertilization and when they are agglutinated by a solution of jelly-coat.
When a spermatozoon crosses the jelly-coat surrounding the egg, it is coated
by a halo of jelly-coat molecules which will not react with any other sperm and
therefore a number of reactive groups of these molecules will remain free. Xow the
question arises whether the spermatozoon carries inside the egg its jelly-coat halo
or the whole sperm surface is left outside. Should the former prove to he the case,
then one could assume that the free groups of the jelly-coat molecules enter into
reaction with the egg cytoplasm and this reaction may be of great importance either
for the activation of the egg or for the reaction that establishes the block against
polyspermy. In fact it is known that parthenogenetically activated eggs can be
entered by a number of spermatozoa (reviewed by Monroy, 1953). Evidence on
this point may be, however, hard to obtain.

We are indebted to Prof. A. Tyler, California Institute of Technology, for
interesting discussions and for reading the manuscript, and to Prof. T. Ajello,
Dept. of Pharmaceutical Chemistry, University of Palermo, for his suggestions.
The technical assistance of Mr. A. O. Oliva is also acknowledged.

LITERATURE CITED

DISCHE, Z., AND L. B. SHETTLES, 1948. A specific color reaction of methylpentoses and a

spectrophotometric micromethod for their determination. /. Biol. Clicin., 175: 595-603.
HEIDELBERGER, M., 1939. Quoted by Tyler, 1948b.
HULTIN, E., G. KRISZAT, S. LINDVALL, G. LUNDBLAD, H. Low. J. RUNNSTROM, E. VASSEUR

AND E. WICKLUND, 1952. On the interaction between the gametes of the sea-urchin

at fertilization. Ark. f. Kcmi, 5: 83-87.
LILLIE, F. R., 1914. Studies of fertilization. 6. The mechanism of fertilization in Arbacia.

/. E.\-p. Zool, 16: 523-590.
LUNDBLAD, G., AND A. MONROY, 1950. Mucopolysaccharase activity of sea-urchin sperms.

Ark. f. Kemi, 2 : 343-347.
KRAUSS, M., 1950. On the question of hyaluronidase in sea-urchin spermatozoa. Science, 112:

759.
METZ, C. B., AND J. DONOVAN, 1951. Specific fertilizin agglutination of dead sperm. Biol. Bull..

101 : 202.'

MONROY, A., AND A. RUFFO, 1947. Hyaluronidase in sea-urchin sperm. Nature, 159: 603.
MONROY, A., AND L. TOSI, 1952. A note on the jelly-coat sperm interaction in sea-urchin.

Experientia, 8: 393.
MONROY, A., 1953. Biochemical and structural changes at fertilization. Proc. Symposium on

"The biochemical and structural basis of Morphogenesis'' in Arch. Neerl. Zool., 10:

19-25.
OLITZKI, L., 1948. Quoted by H. P. Lambert and J. Richley, 1952. The action of mucin in

promoting infections. Brit. J. E.vpcr. PathoL, 33 : 327-339.
RUNNSTROM, J., 1950/51. The problems of fertilization as elucidated by work on sea-urchins.

The Harvey Lectures, Ser. XLVI, 116-152.
TYLER, A., AND S. W. Fox, 1939. Sperm agglutination in the keyhole limpet and the sea-urchin.

Science, 90: 516-517.
TYLER, A., AND S. W. Fox, 1940. Evidence for the protein nature of the sperm agglutinins

of the keyhole limpet and the sea-urchin. Biol. Bull., 79: 153-165.
TYLER, A., 1940. Sperm agglutination in the keyhole limpet, Megathura crenulata. Biol. Bull.,

78: 159-178.
TYLER, A., 1941. The role of fertilizin in fertilization of eggs of the sea-urchin and other

animals. Biol. Bull., 81 : 190-204.

INTERACTION OF SPERM AND JELLY-COAT 177

TYLER, A., 1948a. On the chemistry of the fertilizin of the sea-urchin Strongylocentrotus

pnrpuratus. Anat. Rec., 101 : 8-9.

TYLER, A., 1948b. Fertilization and immunity. Physiol. Rev., 28: 180-219.
TYLER, A., 1949a. Serological aspects of fertilization. The Collecting Net, 19: 6-8.
TYLER, A., 1949b. Properties of fertilizin and related substances of eggs and sperm of marine

animals. Amer. Nat., 83: 195-215.
VASSEUR, E., 1949. Effect of calcium ions on the agglutination in Strongylocentrotus droe-

bachiensis Mull. Ark. j. Ketni, 1 : 105-116.
VASSEUR, E., 1948. Chemical studies on the jelly-coat of the sea-urchin egg. Acta Chem.

Scand.,2: 900-913.
VASSEUR, E., 1950. L-Galactose in the jelly-coat of Echinus csculentus eggs. Acta Chem.

Scand., 4: 1144-1145.
VASSEUR, E., 1951. Demonstration of a jelly-splitting enzyme at the surface of the sea-urchin

spermatozoon. E.\-p. Cell Res., 2 : 144-146.

THE CHARACTER AND ULTIMATE FATE OF THE LARVAL

SALIVARY SECRETION OF PHORMIA REGINA MEIG.

(DIPTERA, CALLIPHORIDAE)

HERBERT H. MOOREFIELD AND G. FRAENKEL

Department of Entomology, University of Illinois, Urbana, Illinois

The study of insect salivary glands has now attained a near classic interest,
and diverse fundamental researches involving them have been stimulated in many
fields. Fraenkel and Brookes (1953) have recently reviewed the literature con-
cerning these glands in Diptera ; especially the cytoplasmic changes occurring
during the late larval period of Phormia regina and several species of Drosophila,
together with the subsequent passage of cellular components into the lumen of the
glands. Various functions have been assigned to this accumulated salivary ma-
terial and a few postulations have been made regarding its disposition, but it is a
noticeable fact that none of the previous workers has offered any experimental
evidence as to the fate of the final larval gland contents. Fraenkel and Brookes
(1953) have also described the manner in which Phormia and certain Drosophila
orally release a fluid that flows along the underside of the insect, solidifies and
securely attaches the newly formed puparium to the surface upon which it rests.
By observing the comparative sizes of the salivary glands and investigating the
volumes of glandular contents in the larvae before and after this secretion was
released, they concluded from this indirect evidence that the secretion produced
in the salivary glands was the same material which glued the puparium to its
substrate.

The purpose of the present investigation was to collect the secretion from the
late larval salivary glands and the material on the exterior of the puparium ; to
analyze them chromatographically in order to establish the identity of these products
and to characterize it as well as the limited quantities would permit.

METHODS

Laboratory cultures of adult Phormia were maintained on sugar and water.
Eggs were collected on pork liver ; the larvae were reared in battery jars with
moist wood shavings and were furnished fresh liver daily. Prior to pupation,
this species leaves its food supply, migrates through the shavings and evacuates its
crop. This is referred to as the "empty-crop" stage in this work, and dissections
revealed that during this period, the salivary gland lumen contains the greatest
quantity of fluid attained throughout the larval life.

The salivary secretion was collected from empty-crop larvae by dissection in a
modified Ringer solution (Ephrussi and Beadle, 1936). This was accomplished
by cutting off the terminal third of the maggot with scissors, and then manipulating
the insect with two pairs of fine forceps to turn it inside out over one of the points
of the forceps. With the aid of ,a dissecting microscope, the glands which pre-
viously extended well into the abdomen were then readily discernible, floating free

178

FATE OF SALIVARY SECRETION IX PHORMIA 179

in the saline, and could be easily teased free of the fat body. They were removed
to a fresh solution of Ringer's to be freed of haemolymph, then quickly dipped into
a dish of distilled water to remove the salts, and finally transferred to a clean,
oversize slide (2 X 3"). Here the glands were punctured with a teasing needle
and the secretion permitted to flow out onto the glass. The material from many
insects was accumulated on a single slide and pooled by washing it off with distilled
water. The volume was reduced in a vacuum desiccator over solid XaOH.

Full grown, empty-crop larvae were placed in petri dishes (about ten insects
per dish) to pupate on a layer of clean sand. When the maggots pupated, the
material which ordinarily flowed along the underside of their bodies and later
anchored them now poured into the sand and merely aggregated a few sand grains
at the anterior end of the puparia. By collecting the tanned puparia, carefully
chipping off these small clumps of sand grains, pooling them and treating with
warm water, the external secretion was extracted.

Hydrolysates were prepared by placing the secretions in small Pyrex tubes,
evaporating to dryness in the desiccator, adding 2 ml. of 6 N HC1 and sealing off
the tubes. These were then steam-autoclaved for 18 hours at 15 Ibs. pressure.
After hydrolysis, the acid was removed by vacuum desiccation, distilled water added
and completely evaporated four times to free the samples of HC1.

The hydrolyzed and raw secretions were analyzed by paper partition chromatog-
raphy (Consden, Gordon and Martin, 1944), using the ascending modification of
Williams and Kirby (1948). Two-dimensional chromatograms were prepared
on Whatman No. 1 paper (9 X 11"), using phenol and water (80 gms. and 20
ml.) as the first phase solvent and a water-saturated mixture of equal parts of
collidine and lutidine in the second phase. A 0.2% ninhydrin (in water-saturated
butanol) spray was used to develop the colors. The final chromatograms were air
dried and examined in transmitted light with the aid of a light-box viewer. Re-
sultant spots were identified by, first, Rf values ; second, cochromatography (an
authentic sample of a known substance is added to the original application on the
paper, intensifying the provisionally identified chromatographic spot which then
acts as a single entity — behavior as such in different solvents presents strong evi-
dence that the known and unknown are identical materials) ; and third, specific
reactions for some of the components (to be described in a later section).

RESULTS AND DISCUSSION

Four separate hydrolysates of the secretion collected from the salivary glands
were prepared, and four from the material gathered from larvae which had pupated
in the sand. The majority of these preparations were made from insects of
different generations. In no cases was there any discrepancy in the number of
spots on well prepared chromatograms, and the similar composition of these ma-
terials was repeatedly demonstrated by identical patterns consisting of the same
components, present in constant relative intensities. The typical pattern, distri-
bution and identification of spots are portrayed in Figure 1.

To test for free amino acids, unhydrolyzed preparations were chromatographed,
and a single spot moved off the original site of application. This spot was present
in both the glandular secretion and material collected from the puparia ; but, when
portions of these preparations were first dialyzed and then chromatographed, the

180

HERBERT H. MOOREFIELD AND G. FRAEXKEL

spot was absent. Cochromatography proved the free amino acid to be lysine.
Chromatograms of dialyzed and nondialyzed materials, after hydrolysis, yielded
lysine spots in each, but the nondialyzed preparations gave a more intense lysine
spot. This indicated that, although lysine was present as a free amino acid, there
was also lysine conjugated in the protein molecule.

ARGININE

GLUTAMIC
ACID

LYSINE

CYSTINE

COLLIDINE

LUTIDINE

VALINE

GLYCINE

ALANJINE

TYROSINE

THREONINE

SERINE

ASPARTIC ACID

/ T^GLUCOSAMINE

CYSTEIC ACID

FIGURE 1. Chromatogram of hydrolyzed salivary secretion of larval Phormia rcgina.

FATE OF SALIVARY SECRETION IN PHORMIA

181

Table I presents a complete list of the identified components found in this study
together with a compilation of reported analyses made on dipterous salivary glands
and associated products.

Cystine and its oxidation product, cysteic acid, appear brown when sprayed
with ninhydrin, and both were present in the salivary material. Dent (1947)
suggested that cysteic acid may be produced by secondary decomposition during
chromatography ; however, a sample of cystine hydrolyzed alone and treated as
were the salivary products showed but a single cystine spot when chromatographed.

TABLE I

Comparison of cinematographic analyses of dipterous salivary glands and associated products

Component

Salivary
gland
chromosome
D. melaiio.

Salivary
gland
chromosome
D. virilis

Salivary
gland
(in toto)
D. virilis

Salivary
gland
protein
D. mclano.

Salivary
gland*
secretion
Phormia

Vasuzumi and
Miyao, 1950

Blumel and
Kirby, 1948

Blumel and
Kirby, 1948

Kodani,
1948

Glycine

+

+

+

4.

4.

Alanine

4.

4.

4.

4.

4-

Serine

4-

4-

4-

Threonine

4-

4-

4-

Valine

4.

4-

4-

4-

4-

Leucine

4-

4.

4-

4-

4-1

Tyrosine
Phenylalanine

Proline

+

+

~

Aspartic acid
Glutamic acid

+

. +

+

+

+

Arginine

Lysine

+

+

+

+

Cystine

4-

Cysteic acid
Methionine sulfoxide

+2

+

Glucosamine

4-

4-

Unidentified

four

two

Free amino acids

none

none3

lysine

* Analysis of material from salivary gland and exterior of puparium.

1 Leucine and/or isoleucine.

2 Probably in error (Dent, 1948).
3LaCour and Drew (1947).

Cystine oxidized with 309f H,CX was used for identification of the cysteic acid
spots. Dent (1948) pointed out that by superimposing the peroxide directly on
the hydrolyzed material on the paper, cystine would be quantitatively converted to
cysteic acid. In chromatograms so treated, the cystine spot did not appear, and
the cysteic acid spot was intensified. Further confirmation of cystine was accom-
plished by spraying the chromatograms with an iocline-azide spray reagent. As
demonstrated by Chargaff, Levine and Greene (1948), the sulfur-containing amino
acids were revealed as white spots against a brown background. After the location
of the spot was marked, and the iodine faded, the paper could be resprayed with
ninhydrin.

182

HERBERT H. MOOREFIELD AND G. FRAEXKEL

The spot labeled Xl occupies the position ascribed by Dent (1947) to methionine
sulfoxide and was later reported as such in the salivary secretion of Drosophila by
Kodani (1948). When methionine was treated with 30% H2O2, mixed oxidation
products of methionine sulfoxide and methionine sulfone resulted, and when this
mixture was cochromatographed with the salivary secretion, X, was reinforced by
methionine sulfoxide. However, neither adding the peroxide to the original ma-
terial on the paper before chromatography, nor treating the final chromatogram
with the iodine-azide reagent gave the anticipated reactions of a sulfur-containing
amino acid in this position. Beta-aminoisobutyric acid has also been reported to
have the same Rf values as methionine sulfoxide (Grumpier, Dent, Harris and
Westall, 1951), but as this compound has never been identified as a protein hy-
drolysate product, it is unlikely that this amino acid is Xv

A second unknown substance, labeled X2, was characterized by a yellow
ninhydrin reaction. Although this reaction is typical of the imino acids, X2 does
not migrate to the site occupied by any of the known substances of this nature.

TABLE II
Nitrogen composition of salivary products of Phormia regina

Product

Dry weight
of material
analyzed

Total

nitrogen

Per cent
nitrogen

Mean
nitrogen
per cent

Secretion collected from sali-
vary glands

1.4 mg.
3.6

0.111 mg.
0.289

7.9
8.(T

8.0

Secretion collected from pu-
paria (in sand)

2.9

3.4
4.8

0.231
0.261
0.382

8.0

7.7
8.0

7.9

4.9

0.388

7.9

Secretion collected from pu-
paria and dialyzed

7.3
5.9

0.732
0.597

10.0
10.0

10.0

A compound with a similar ninhydrin color and Rf values has been reported on
chromatograms of free amino acids in potatoes (Dent, Stepka and Steward, 1947),
and has also been found free in other dipterous tissues (unpublished data). In the
salivary material, this unknown substance is undoubtedly conjugated in the protein
molecule, as it appears only after hydrolysis.

In Table II, micro-Kjeldahl nitrogen determinations are compiled. These ma-
terials were collected in small aluminum-foil boats, and dried to constant weight
before analysis. The nitrogen content of the fluid collected from the glands is
in good agreement with that of the material emitted into the sand. The nitrogen
values are higher in dialyzed samples, but are still unusually low for a typical
protein. Kodani (1948), working with the glandular secretion of D. melanogaster,
reported 10.8% nitrogen in samples exhaustively extracted with ether, and further
demonstrated that a considerable quantity of crystalline, inorganic salt was present
after hydrolysis and evaporation of HC1 from the residue. He proposed that the
low nitrogen figures could be accounted for by the presence of the salt together
with a large amount of glutamic acid and glucosamine, both of which are low in

FATE OF SALIVARY SECRETION IX PHORMIA 183

nitrogen. The Phormia products, when permitted to dry on glass slides, fre-
quently crystallized in dendritic patterns which could be a result of the presence
of inorganic salts. However, if this were the principle contributing factor to the
low nitrogen figures, it would be expected that dialyzed samples would contain more
nitrogen than the determined values of W%. Lesher (1952) presented data
suggesting that the substance synthesized by the larval salivary gland of Drosophila
robusta is a conjugated protein composed of a protein bonded to a polysaccharide,
i.e., a mucopolysaccharide. This could possibly explain the low nitrogen figures
obtained by Kodani and in this study.

Biuret and ninhydrin (triketohydrindene hydrate) tests were both strongly
positive and indicative of the proteinaceous nature of the salivary gland secretion.
The protein was water soluble and could be precipitated with hot or cold 10%
trichloroacetic acid. Millon's reaction was positive, confirming the presence of
tyrosine, and the xanthoproteic test also gave a strong reaction. The Hopkins-
Cole test for tryptophane was slightly positive, but as only acid hydrolysates were
prepared, the presence of this amino acid was undetected. Although glucosamine
was demonstrated as a constituent of the protein, results of the Molisch test were
doubtful. Specific carbohydrate tests, Benedict's, Barfoed's and SelivanofFs, were
all negative.

SUMMARY

1. In a comparative chromatographic study of the fluid in late larval salivary
glands, and the substance which is responsible for adhering puparia of Phormia
rcgina to their substrate, evidence is presented indicating that these materials are
identical in nature and composition ; this constitutes convincing proof that the
"puparial cement" is the ultimate fate of the larval salivary secretion.

2. The identity of these secretions, collected from two different sites (the
salivary glands and the exterior of the puparia), has been demonstrated by like
nitrogen composition ; the presence of a single free amino acid, lysine, in each
fluid ; the same components in the protein constituent (amino acids — glycine, alanine,
serine, threonine, valine, leucine, tyrosine, proline, aspartic acid, glutamic acid,
arginine, lysine and cystine ; two unknown substances, and the carbohydrate,
glucosamine) ; as well as by the reactions to several biochemical characterization
tests.

LITERATURE CITED

BLUMEL, J., AND H. KIRBY, 1948. Amino acid constituents of tissues and isolated chromosomes

of Drosophila. Proc. Nat. Acad. Sci., 34: 561-566.
CHARGAFF, E., C. LEVINE AND C. GREEN, 1948. Techniques for the demonstration by chroma-

tography of nitrogenous lipide constituents, sulfur-containing amino acids, and reducing

sugars. /. Biol. Chem., 175: 67-71.
CONSDEN, R., A. H. GORDON AND A. J. P. MARTIN, 1944. Qualitative analysis of proteins;

a partition chromatographic method using paper. Biochcm. J ., 38 : 224-232.
CRUMPLER, H. R., C. E. DENT, H. HARRIS AND R. G. WESTALL, 1951. /3-aminoisobutyric acid

(a-methyl-/3-alanine) : A new amino acid obtained from human urine. Nature, 167 :

307-308.
DENT, C. E., 1947. The amino-aciduria in Falconi syndrome. A study making extensive use

of techniques based on paper partition chromatography. Biochetn. J., 41 : 240-253.

184

HERBERT H. MOOREFIELD AND G. FRAEXKEL

DENT, C. E., 1948. A study of the behavior of some sixty amino acids and other ninhydrin-

reacting substances on phenol-collidine filter-paper chromatograms, with notes as to

the occurrence of some of them in biological fluids. Biochcin. J .. 43: 169-180.
DENT, C. E., W. STEPKA AND F. C. STEWARD, 1947. Detection of the free amino acids of plant

cells by partition chromatography. Nature, 160 : 682-683.
EPHRUSSI, B., AND G. W. BEADLE, 1936. A technique of transplantation for Drosophila. Amer.

Nat., 70: 218-225.
FRAENKEL, G., AND V. J. BROOKES, 1953. The process by which the puparia of many species

of flies become fixed to a substrate. Biol. Bull, 105 : 442-449.
KODANI, M., 1948. The protein of the salivary gland secretion in Drosophila. Proc. Nat.

Acad. Sci, 34 : 131-135.
LACouR, L. F., AND R. DREW, 1947. Partition-chromatography and living cells. Nature,

159: 307-308.
LESHER, S. W., 1952. Studies on the larval salivary gland of Drosophila. III. The histochemi-

cal localization and possible significance of ribonucleic acid, alkaline phosphatase and

polysaccharide. Anat. Rci., 114: 633-652.
WILLIAMS, R. J., AND H. KIRBY, 1948. Paper chromatography usine capillary ascent. Science,

107 : 481^83.
YASUZUMI, G., AND G. MIYAO, 1950. A qualitative analysis of the amino acids in isolated

chromosomes. E.\-p. Cell. Res., 2: 153-157.

SIZE DISTRIBUTION, EROSIVE ACTIVITIES, AND GROSS META-
BOLIC EFFICIENCY OF THE MARINE INTERTIDAL SNAILS,
LITTORINA PLANAXIS AND L. SCUTULATA 1 •-

WHEELER J. NORTH

Division of Marine Biochemistry, Scripps Institute of Oceanography, University of California,

La Jolla, California

The two common representatives in Southern California of the nearly world-
wide genus, Littorina, are L. plana.vis and L. sciitulata. The former generally
occurs at higher levels in the intertidal, but the zones of distribution of the two
species overlap. L. plana.ris may often be found 5 to 10 feet above spring high
tide level, but L. sciitulata prefers a zone two or three feet on either side of the
high tide mark. Published information concerning these two species of snail is
scanty in spite of their great abundance and their availability. Their importance to
the high intertidal community, however, warrants an extensive study of their
ecology, and the present paper describes some of the basic biology of these interest-
ing animals.

SIZE DISTRIBUTION

Inspection of colonies of Littorina at different places along the La Jolla shore
has revealed that the majority of snails at any given locality fall within certain
rather well denned size limits. Lysaght (1941 ) noted a similar condition in L.
neritoides on the Plymouth Breakwater. In order to gain a more exact picture
of size distributions in the present study, three typical" Littorina environments were
chosen, and height measurements were made of all the periwinkles found within a
selected area, representative of the environment.

Environmental description

The three environments are shown in Figure 1. The first (Fig. la) is a group
of pools at Whale View Point in La Jolla. The area is subjected to vigorous
wave action at even moderate tides. The second environment (Fig. Ib) is the
seaward edge of the top of a broad shelf of rock that extends out from a cliff
about 1/4 mile north of the Scripps Institution of Oceanography. The top of
Shelf Rock is 1 to H meters above spring tides, but the area under discussion is
well splashed at high tide and a large wave may occasionally wash over it. The

1 This work was supported by a research grant from the Rockefeller Foundation and
represents a portion of a thesis submitted in partial fulfillment of the requirements of the Ph.D.
degree, University of California. The author would like to express his gratitude to Professor
Denis L. Fox for his continued interest and friendly supervision, and to Dr. Arthur L. Kelly
and Messrs. James S. Kittredge and John R. Maher for helpful assistance.

2 Contribution from the Scripps Institution of Oceanography, New Series No. 685.

185

186

WHEELER J. NORTH

third environment (Fig. Ic) is a protected portion of Shelf Rock some three meters
removed from the second environment and splashed only occasionally by large
waves.

All three localities are sedimentary sandstone, but the grain size is much finer
at Shelf Rock. By exerting pressure the point of a knife may be forced into

FIGURE 1. Photographs of the three environments studied. a. Whale View Point; arrow
points to typical tidepool from which snails were gathered, b. Pools of Shelf Rock ; arrow
points to region studied, c. Dry area of Shelf Rock ; Region studied enclosed by white line and
marked with arrow labeled 3 : arrow labeled 2 points to second environment, about 3 meters
away.

STUDIES ON CALIFORNIA LITTORINA

187

IOOT

50

Percent of
population
with equal or
less height

0.5

Height of snails, cm

FIGURE 2. Cumulative curves for the Littorina plana.ris populations of three environ-
ments. A. Whale View Point; B. Shelf Rock pools; C. Dry area, Shelf Rock. Measurements
made in August, 1951, shown as circles ; measurements made in July, 1953 shown as dots.

the sandstones to a depth of about half an inch. The top layers of sand particles
may easily be removed by scraping. The topography is characteristically very
irregular, great numbers of pools, depressions, basins, and small holes being
present.

Results of measurements

The size distribution of L. planaxis and L. scutulata in each of these environ-
ments is shown by means of cumulative curves in Figures 2 and 3. The greatest
dimension, "height," was measured, being the distance from the tip of the spire
to the lower lip of the aperture. It can be seen that the populations at Whale View
Point are rather small, averaging about 0.4 cm. The snails in the pools of Shelf
Rock are generally intermediate in size, approximately 0.8 cm., while those on
the sheltered dry area of Shelf Rock, although more heterogeneous than the other
groups, are generally the largest, averaging almost 1.1 cm. in height. In the latter
environment there were insufficient numbers of L. scutulota to enable the construc-
tion of a reliable curve.

188

WHEELER J. NORTH

In attempting- to explain the variation in size distribution with locality, two
possibilities suggest themselves. A favorable set of spawn in different areas at
different times, for example, would produce curves of the type that have been
obtained. Equally plausible is the contingency that some factor or factors operate
in the environment to produce selection for a particular size of periwinkle. If
the first hypothesis is correct, curves obtained from measurements made at a later

IOOT

Percent of
population
with equal or
less height

Height of snails, cm

FIGURE 3. Cumulative curves for the Littorina scutulata populations of two environments.
A. Whale View Point; B. Shelf Rock pools. Measurements made in August, 1951, shown as
circles ; measurements made in July, 1953, shown as dots.

date should be shifted towards the right along the abscissa ; that is, effects of
growth should be evident. If the latter hypothesis is the proper explanation for
the phenomenon, the position on the abscissa of a curve should remain constant
with time. Figures 2 and 3 show cumulative curves for populations in the first
and second environments, constructed from measurements made about two years
after the data discussed above were obtained. Measurements were not attempted
for the third environment as appreciable numbers of large periwinkles had been
permanently removed from this locality for other experiments. The expected

STUDIES ON CALIFORNIA LITTORINA 189

height increase of L. planaxis during this time can be calculated from data given
below, and would be of the order of 0.5 cm., but the two sets of curves occupy
approximately the same position. Environmental factors, therefore, appear to
determine the size distribution of snail populations at any particular locale.

It should be mentioned that the three environments were under constant ob-
servation over the period that intervened between the measurements of Figures
2 and 3. The populations were never noticeably different in their size distribution
during this time.

Factors affecting size

Wave action and salinity may be among environmental mechanisms capable
of causing selection for a given snail size. In many places these two factors are
inversely correlated. The higher levels of the spray zone are not wetted by the
ocean so frequently as are the lower levels, and have therefore more opportunity
for undergoing evaporation and achieving correspondingly greater salinities. The
water of the pools and the surface moisture films of the lower levels, on the other
hand, are renewed often by waves, and salinities at these localities are therefore
not elevated.

As already noted, the three environments studied are subjected to different
degrees of wave action. The snail populations composed of larger individuals
appear to occupy the drier areas of the spray zone. In order to determine whether
water currents might be more effective in removing large than small snails the
following experiments were performed.

Four small (approx. 0.6 cm.), four intermediate (0.9-1.2 cm.), and four
large (1.4-1.6 cm.) specimens of L. planaxis were placed on a small flat rock
in a tidepool. When all the snails had emerged from their shells and were ob-
served creeping, the rock was moved rapidly through the pool, creating a water
current across the shells. After several vigorous swings through the water, only
three small snails were left. A repetition of the experiment using six small, four
intermediate, and four large snails ended with only four small snails remaining
attached to the rock. When twelve medium-size L. scutulata (approx. 0.9 cm.)
were placed on the stone with three small and two large L, planaxis for comparison
.purposes, the large specimens of the latter species were washed off with relative
ease. After much effort three of the L. scutulata were eventually swept away,
leaving nine of this species and the three small L. planaxis when the experiment
was discontinued.

When a stream of sea water from a hose was allowed to play against the
shells of L. planaxis creeping on a flat rock, large individuals were washed off
more readily than small ones, and a snail of a given size was removed with greater
ease when the stream was directed against the posterior part of the shell.

In order to gain an idea of the current velocities necessary to dislodge a snail,
specimens of L. planaxis were placed in a Plexiglass tube of 3.5 cm., inside diam-
eter, and after the animals had emerged from their shells and were crawling on
the surface of the Plexiglass, the tube was gently filled with sea water. A current
of known velocity was then allowed to flow through the tube and notations were
made of the ability of the snails to remain attached for 10 seconds. The results

190 WHEELER J. NORTH

are given in Table I. A total of 10 large and 8 small snails was used in the
experiment, and 6 trials were conducted.

Several hypotheses offer plausible explanations for the ability of small L.
plana.ris to withstand currents that remove large individuals. Physical factors
might include greater water friction acting as drag on larger shells, or frictional
forces slowing the current in the vicinity of the rock-water interface 3 thus favoring
smaller snails. Biological factors might include loss of vigor and of tenacity with
increasing age, or a disproportionate growth of the various parts such that the
sole of the foot does not increase as rapidly as the surface area of the shell. Figure
4 shows that the last hypothesis is not supportable, hence one or more of the other
factors may contribute to the distribution phenomenon.

TABLE I

Effect of current velocity on two size groups of Littorina planaxis

Velocity of current Fraction of snails

flowing over snails removed by current

Size group meters/second per cent

Large snails 3.4 90

(1.3 to 1.6 cm.) 2.3 90

2.0 80

Small snails 3.4 50

(approx. 0.7 cm.) 2.0 50

The absence of large snails from areas exposed to vigorous wave action may
thus be explained, but account has not yet been taken of the absence of small
snails in the high, dry areas of the spray zone. The solution to this problem may
lie in physiological age changes rendering the animal more capable of coping
with exposure to air or to more variable salinity .conditions. Salinity has been
shown to influence shell size and shape in Littorina (Thorson, 1946; Agersborg,
1927) and further study along these lines will be necessary before a complete
explanation of the observed size distribution can be proposed.

EROSIVE ACTIVITIES

Many intertidal animals bring about erosion of rock. Littorina species often
occur in small basins above high water mark (Fischer-Piette, 1932; Clench, 1938;
Lysaght, 1941) and have been credited by some investigators (Brunelli, 1928;
Welch, 1929) with the production of depressions in rock. The snails feed by
applying a file-like ribbon, the radula, to the substratum and transfer bits of it
to the mouth by a scraping action. On the cliffs around La Jolla the animals
scrape algae and fine detritus from the rocks and at the same time remove particles
of the rock itself. Because of their great abundance, erosion resulting from their
feeding is believed to be appreciable, and a quantitative estimation of the magni-
tude of the erosion will be useful in determining its importance with respect to
other erosive processes.

Since the animals may not behave as they do in nature when kept in the)
laboratory over long periods of time, it seemed wisest to obtain as much data

3 An idea for which I am indebted to Professor Roger R. Revelle.

STUDIES ON CALIFORNIA LITTORINA

191

as possible from snails in their normal environment. The simplest means for
accomplishing this appeared to be to determine the number of times daily the gut
contents are completely renewed, and also the proportion of the gut contents
that is inorganic matter. The product of the two quantities would yield the daily
rate of erosion.

0.6 1.2-f

0,4 0,8--

0.2 0.4 --

L. plonoxis *
L. scutulqto »•

(
1.0

0.5

— t—
2.0

1.0

FIGURE 4. Relationship of shell surface to surface of the sole of the foot for Littorina
plana.vis and L. scuttilata, indicating that the increase in area of each with age is proportionately
the same. The sole area (At) was measured while the animals were crawling up the sides of a
glass vessel. Shell surface (As) was arbitrarily taken as I/T of T V r + H2, the formula for
the curved surface of a right cone.

Time required to renew material in the gut

The number of times daily the gut contents are renewed will depend on the
rate at which material is passed along the gut, which in turn may vary with the
amount of time the snail is able to feed. Dissection of specimens of L. plana.ris
taken both from pools (46 specimens ) and from areas that become dry at low tides
(30 specimens ) revealed that throughout the day the intestines of the former group
were always full of unconsolidated material, whereas the intestines of the latter

192

WHEELER J. NORTH

TABLE II

Summary of an experiment to determine the time required for Littorina to cycle food completely
through the gut; May 18, 1953; water temperature range 17° to 30° C.;

20 snails per size group

Species

I [fight cm.

Time elapsed until

First blue
fecal pellet
observed, hr.

Half of group
defecated blue
feces. hr.

Entire group
defecated blue
feces, hr.

L. f>lanaxis
L. planaxis
L. planaxis

0.4-0.5
0.7-0.95
1.3-1.6

If

2
2|

2
3!

^

3

5

_*

L. sciitulata
L. sciitulata

0.3-0.45
0.8-0.95

1!

2|

1
4

3
5

* Thirteen had defecated blue feces after 1\ hours.

group were generally at least partially full hut sometimes almost empty. It was
concluded that animals in the pools graze sufficiently to keep the gut full at all
times. Snails feeding on areas that become dry at low tide appear to graze only
when the surface is moist, and the amount of material in the gut consequently is
variable.

In order to determine the rate at which material is passed through the gut,
groups of snails were allowed to graze on a section of Shelf Rock stained. with the

2 +

Weight of

sand in the
gut, mg

o.'s i.o

Height of snail, cm

FIGURE 5. Relationship of sand contained in the gut to height of Littorina plana.ris.

STUDIES ON CALIFORNIA LITTORINA 193

harmless dye methylene blue. After a period of an hour and a half had elapsed,
the animals were removed to an unstained area and carefully observed for the first
appearance of blue fecal pellets. While they were on the stained and unstained rock,
the periwinkles were washed with fresh sea water every 10 minutes, and their
behavior during the experiment seemed normal. Five such experiments were
conducted on two spring and on three summer days. For snails of height 0.8
cm., cycling times of 2^ to 6 hours were obtained. The results were always similar
and are described in Table II for one of the experiments.

A typical 0.8-cm. snail in a Shelf Rock pool, therefore, feeds sufficiently to keep
the gut full at all times and probably renews the material completely four to eight
times daily.

Inorganic matter in the gut

Sand in the gut was determined by extracting snails from their shells, in-
cinerating the bodies in a tared crucible, cooling, adding a little concentrated
hydrochloric acid to dissolve body ash, carefully decanting, rinsing, re-incinerating
and weighing. Four size groups of 50 L. planaxis each were thus analyzed and
the results are presented in the curve of Figure 5. An 0.8-cm. snail therefore
contains on the average about 1.6 mg. of sand in the gut.

Calculated rate of erosion

Taking the Shelf Rock region, and considering the simplest case of a snail
feeding in a pool, an estimation of erosion may now be. made. It has been shown
(Fig. 2) that the population in this environment averages 0.8 cm. in height. The
average amount of inorganic material contained in the gut of an 0.8-cm. snail was
found to be 1.6 mg., and, if we take the most conservative cycling period of 4
times daily, we obtain the value of 1.6 X 4 = 6.4 mg. of inorganic material passing
through the gut of the snail per day. This presumably is equivalent to the amount
of rock eroded daily. The density of Shelf Rock is approximately 2.5 g./cc. and
calculations show that 100 snails, 0.8 cm. in height, would be capable of excavating
a basin of 86 cc. yearly, almost a liter in a decade.

The concentration of L. planaxis in the environment under discussion is of
the order of one snail per 30 cm.2 Erosion by this species alone, therefore, may
be calculated to deepen the pools one cm. every 40 years. If L. scutulata is con-
sidered, the snail concentration increases to one snail per 12 cm.,2 and assuming that
0.8-cm. individuals of both species have similar feeding rates and weight of gut
contents, erosion by both Littorina species combined will deepen the pools one
cm. every 16 years. The snail concentration has remained fairly constant over
the two-year period that is covered by these observations.

Comparison with erosion from other sources

Other erosive processes acting on the sandstone rocks of this area have been
studied by Emery (1941, 1946). For exposed surfaces a general erosion rate
of one cm. every 20 years was estimated, while for pools at Whale View Point
nocturnal pH changes at low tides were calculated to cause a deepening of the

194

WHEELER J. NORTH

pools by one cm. every 33 years. The different processes causing rock erosion are
probably additive in some cases and in other instances facilitate each other and are
not additive.

GROSS EFFICIENCY

The average, gross metabolic efficiency, or the ratio of ingested food to organic
matter incorporated as living tissue, may now be estimated for L. planaxis with the
aid of additional data that have been obtained.

Growth

Growth rates were determined by measuring the increases in height of marked
specimens of L. planaxis on Shelf Rock at various intervals for a period of a

u.ub-

0.04-
Growth
increment,
cm/mo

0

• o
o o co o o o

00 0 - ° 0 0 °

o° °0 o

0.5 1.0 1.5
Height of snail, cm

FIGURE 6. Twenty-six monthly growth increments observed in specimens of Littorina plana.ris

at Shelf Rock.

year and a quarter. The methods of converting an increase in spiral length to a
height increase, used by Moore (1937) and Lenderking (1951) with other lit-
torines, seemed to be complicated in the present case, since the apex angle of L.
plana-xis varies considerably, and it therefore appeared simplest to measure height
increments over a long period of time. Figure 6 illustrates the results of the
experiment, showing the growths of positively identified individuals from the snails
that were recovered, out of a total of 300 originally marked. It may be noted
that an 0.8-cm. snail has an average height increment of about 0.02 cm. per month.
To convert this value to an increase in tissue weight the relationship between
snail height and dry tissue weight was obtained and is shown in Figure 7. Speci-
mens of L. plana.vis were dried in a vacuum oven for 24 hours at 80° C. and 460
mm. pressure. After cooling in a desiccator, the shell and dry tissue were weighed
and the shell weight was obtained after removing the dry tissue by a half hour's
immersion in boiling 20% KOH. Dry tissue weight was then readily calculated

STUDIES ON CALIFORNIA LITTORINA

195

0.06T

0.04--

0.02 +
Weight of

dehydrated
tissue, g

0.5

Height of snail, cm

FIGURE 7. Relationship between dry tissue weight of Littorina planaxis and height.

by subtraction. Computations using Figure 7 indicate that an 0.8-cm. snail grow-
ing 0.02 cm. per month would have an increase in dry weight of 0.6 mg. for a
like period.

Ingcstion of organic matter

Determination of the amount of food ingested monthly could be readily ac-
complished if the organic matter per cent of the material swallowed by the snail
were known. In scraping the surface of rocks, however, the snail has an excellent

TABLK III

Organic content of various materials

Material analyzed
Snail feces

Light manual knife scrapings of
rock surface

Loose sand from tidepool
bottom

Organic matter
per cent

2.4
2.0
3.0

3.2

2.4
2.6

0.8
1.0

Average

2.7
2.5
0.9

196 WHEELER J. NORTH

mechanism for removing only a thin, organic-rich layer of material. Even the
lightest of scrapings made by a knife might remove much underlying rock along with
the surface "skin" of organic matter. To estimate the organic matter per cent of the
material swallowed by the snail, therefore, it seemed best to assume that only
a small fraction of the organic molecules will be assimilated through the gut wall,
and that the feces would therefore have approximately (and conservatively) the
same organic content as the ingested material. Corrections for this assumption
can then be made and the efficiency recalculated.

Fresh snail feces may easily be obtained in abundance, and the results of
organic analyses on these and other materials by Walkley and Black's rapid
titration method (chromic acid digestion) are given in Table III.

It was found in calculating erosion that an 0.8-cm. snail voids 6.4 mg. oi
inorganic material daily and this amounts to 192 mg. per 30-day month. If the
feces are 2.7% organic matter the amount of organic material voided per month is
5.3 mg. Making use of the assumption explained above, this is approximately
equal to the monthly ingested organic matter.

Calculation of gross efficiency

For an 0.8-cm. snail in the pools of Shelf Rock the gross efficiency may now be

Organic matter added as tissue 0.6

estimated as ^ —:- -r- -X 100 which equals ^r X 100 = 11%.

Organic matter ingested 5.3

In order to refine the calculation let us consider how much of the ingested
organic matter might reasonably be assimilated across the gut wall from the food,
of which some is built into new tissue. There must also have been assimilated a
sufficient quantity of material to account for organic matter lost in respiration,
and since the animals are poikilothermic and slow in movement, it would seem that
this loss should be less than 4 times the amount added as new tissue. Taking
the latter, then, as 20% of the total assimilated, and all other losses combined as
80%, and knowing that the animal adds 0.6 mg. per month as new tissue, we have
3 mg. per month as an estimate of the amount of organic matter assimilated across

the gut wall. Recalculating the efficiency gives a value of -p-^r =-=- X 100 = 7%.

0 .0 ~t O . \J

It will be recalled that in calculating erosion the time for cycling food through
the gut was taken conservatively as 6 hours. The average value might be less,
which would increase the erosion rate and depress the efficiency.

SUMMARY

1. Size distribution curves for populations of the marine interticlal snails
Littorina planaxis and L. scutnlata in three environments are presented.

2. It is concluded that environmental factors cause the observed size distri-
butions, and the amount of wave action at a given locale appears to be one of the
influencing factors. L. planaxis was observed to cling to a smooth surface in
current velocities of two to three meters per second for 10 seconds.

3. The time for littorines to cycle food completely through the gut varied
with size. Snails 0.8 cm. in height required from 2\ to 6 hours.

STUDIES ON CALIFORNIA LITTORINA 197

4. Erosion resulting from the snails' feeding activities was estimated for certain
tidepools and found to deepen tidepools one cm. every 16 years. This is of the
same order of magnitude as other erosive processes which have been studied in the
same region.

5. Growth of L. planaxis was found to average 0.02 cm. per month in height
increment, and 0.6 mg. per month in dry weight of organic matter.

6. The gross metabolic efficiency was computed and is estimated to be in the
neighborhood of 7%.

LITERATURE CITED

AGERSBORG, H. P. K., 1927. The distribution, variation, and evolution of certain prosobranchiate

mollusca from the littoral zone of the coasts of New England and Norway. Anat. Rec.,

37: 149.
BRUNELLI, G., 1928. Sulla natura biofisica della erosione foveolare della arenaria nella costa

Tirrena. Atti. R. Acad. Nas. Lincei Rend. Cl. Sci. Fis. Mat. e Nat., 8: 423-424; Biol.

Abstr., 3, 1929.
CLENCH, W. J., 1938. A new species of Olivia from Santo Domingo with notes on other marine

forms. Nautilus, 51: 109-114.
EMERY, K. O., 1941. Rate of surface retreat of sea cliffs based on dated inscriptions. Science,

93: 617-618.

EMERY, K. O., 1946. Marine solution basins. /. Geol, 54 : 209-228.
FisCHER-PiETTE, E., 1932. A propos du charactere euryhaline des littorines. Bull. Lab. Mar.

Saint-Servan, 9 : 16-17.
LENDERKING, R. E., 1951. Observations on Littorina anguilijera, Lam., from Biscayne Key,

Florida. Quart. J. Fla. Acad. Sci., 14: 247-250.
LYSAGHT, A. M., 1941. The biology and trematode parasites of the gastropod Littorina neri-

toides, L., on the Plymouth breakwater. /. Mar. Biol. Assoc., 25 : 41-67.
MOORE, H. B., 1937. The biology of Littorina littorea. Part I. Growth of the shell and tissues,

spawning, length of life, and mortality. /. Mar. Biol. Assoc., 21 : 721-742.
THORSON, G., 1946. Reproduction and larval development of Danish marine bottom invertebrates,

with special reference to the planktonic larvae in the sound (0resund). Meddelelser

Fra Kommissionen For Danmarks Fiskeri-Og Havundersjzfgeleser Serie : Plankton.

Bind 4. Nr. 1.
WELCH, R. J., 1929. Littorina perforations in indurated chalk. Irish Nat. J. Belfast, 2: 131.

THE ECOLOGY OF PHYTOPLANKTON BLOOMS IN

MORICHES BAY AND GREAT SOUTH BAY,

LONG ISLAND, NEW YORK ^

JOHN H. RYTHER
Woods Hole Oceanographic Institution, Woods Hole, Massachusetts

In recent years Great South Bay and Moriches Bay have supported an
extremely heavy growth of phytoplankton which characteristically appears early
in the spring and persists throughout the summer and fall. At the peak of their
growth, the contributing organisms have been found to exceed concentrations
of ten million cells per ml., and the resulting turbidity may reduce Secchi disc
transparency to less than one foot in places.

Records maintained since 1922 by Mr. J. B. Glancy show that the recurrent
annual populations of algae in Great South Bay have consisted almost without
exception of small, unicellular, green organisms 2-4 /t in diameter, which have
been referred to locally as "small forms." This type of population differs greatly
from the mixed communities of diatoms, green flagellates, and dinoflagellates
which are typical of the plankton flora in other bays and estuaries of the same
general region. Another striking feature is the persistence of the "small form"
populations throughout the spring, summer, and fall, in marked contrast to the
seasonal succession of dominant species which is commonly observed elsewhere.
It is the purpose of this report to analyze the physiological factors which may
account for the dominance of "small forms" in the recurrent plankton blooms.

These dense growths of algae have greatly reduced the value of the surround-
ing region as a recreational area, and are also considered to be the principle cause
of the failure of what was formerly a prosperous oyster industry in Great South
Bay. Correlated with and suspected as a cause of the algal blooms is the existence
of a large duck industry which now consists of over 40 individual farms centered
along the tributary streams and coves of Moriches Bay. These farms are so
situated that their waste products eventually enter the bays, greatly enriching the
water and presumably creating conditions conducive to the development of the
plankton blooms.

The results of a study of various aspects of the hydrography, chemistry, and
biology of Great South Bay and Moriches Bay, will appear elsewhere (Ryther,
unpublished data). This investigation has revealed that the embayments under
consideration represent a unique ecological environment which results partly from
pollution contributed by the surrounding duck farms, and partly from the topo-
graphic and hydrographic features of the area. The pollution not only provides
an extensive fertilization of the bay waters with nutrients essential to phytoplankton
growth, but, in addition, the presence of organic nitrogen compounds and the low
ratio of nitrogen to phosphorus in the pollutants create conditions which differ

1 Contribution No. 685 from the Woods Hole Oceanographic Institution.

198

ECOLOGY OF PHYTOPLANKTON BLOOMS 199

considerably from the chemical composition of natural sea water. The shallow
nature of the embayments contributes to the development of abnormally high
water temperatures which may reach 30° C. during the summer months. Their
extremely low flushing rate results in the retention for long periods of time of
both the pollutants and their entraining fresh water, which further builds up the
concentrations of nutrients and the resulting phytoplankton crop and also reduces
salinities to approximately 50% of normal sea water in the open embayments
and less than 10% in the estuaries.

This report includes the results of experiments performed with the two dominant
species of "small forms," and, for contrast, with the common neritic diatom,
Nitsschia closterium. Growth rates of these algae were studied in relation to the
temperature, salinity, and nutrient conditions peculiar to Great South Bay and
Moriches Bay as described above. The results of these experiments were then
related to the distribution of the "small form" population and associated physical
and chemical conditions existing in the embayments.

The author expresses his sincere thanks to Mr. J. B. Glancy, the Messrs.
G. H. Vanderborgh, Sr., and Jr., and the New York State Department of Con-
servation for their generous assistance in providing laboratory and boat facilities
in the field. Appreciation is also tendered to Dr. Win. Butcher and Dr. R. A.
Lewin for their help in the identification of the phytoplankton. The investigation
was carried out, in part, with the assistance of a grant from the National Science
Foundation.

CULTURE METHODS

The "small form" population of 1952 and 1953 consisted principally of two
species of algae. That which was the more numerous was identified by Dr. Wm.
Butcher as his recently described Nannochloris atomus (Butcher, 1952). The
other species was tentatively placed by Dr. Ralph Lewin in the genus Stichococcus,
possibly S. cylindriciis Butcher.

Because of their similarity of appearance, it was impossible to distinguish
between the two species in routine examinations of the natural population. Con-
sequently they will be considered together as a single community and referred to
collectively as the "small form" population.

The Nannochloris, Stichococcus, and Nitsschia used in the experiments were
isolated from water samples collected from Great South Bay. Pure cultures were
obtained by the agar streak method incorporating the use of penicillin (200
units/ml.) and streptomycin (10-20 units/ml.) in the enriched agar. Colonies
isolated from streaks were transferred several times on the antibiotic agar and then
inoculated into liquid media.

Pure cultures were used in all experiments involving nutrient studies, while
cultures of Nannochloris and Stichococcus used in the temperature and salinity
experiments were unialgal but not bacteria-free.

Growth studies were made with algal cultures grown in 300 ml. of media in
500-ml. Erlenmeyer flasks. Lighting was provided by a bank of mixed daylight
and white fluorescent lamps which produced 500 foot candles of illumination.

200

JOHN H. RYTHER

Temperature was controlled by keeping cultures immersed in running tap
water, which provided a range of 20-30° C. over a six-months period by varying
the rate of flow of the water. This method permitted control to ± 2° C. for
periods of one to two weeks, the normal duration of the experiments. Cultures
grown at temperatures of 5-15° C. were maintained to a variability of ±1° C.
in an illuminated, constant temperature box. Except where the effect of tempera-
ture upon growth was being studied, cultures were routinely grown at 18-22° C.

The same basic medium was employed for the culture of all three species of algae,
and consisted of a modification of the artificial sea water of McClendon et al.
described in Sverdrup, Johnson and Fleming (1942), enriched with N, P, Si,
and Fe. At full strength (34.62°/00) it consists of the following, in parts per
thousand :

NaCl

MgCl2-6H2O

MgSO4-7H2O

CaCl2

KC1

NaHCO3

KBr

26.726
2.260
3.248
1.153
0.731
0.198
0.058

H3BO,

Na2SiO3-9H2O
NH4C1

Na2HPO4-12H2O
Fe Citrate

0.058
0.020
0.053
0.020
0.001

Since the growth of Nitzschia is relatively poor in ammonia-nitrogen, as will
be demonstrated, NH4C1 was replaced with 0.10°/oo KNO, in media used for
growing the diatom. Several of the other constituents have been altered in con-
centration or replaced with other ingredients in the various experiments, as will
be discussed below.

Growth in all experimental cultures was determined by cell counts with a Levy
hemacytometer. Cultures were grown for periods of 10 days to two weeks, and
cell counts were made at intervals of two to three days.

The growth rate, expressed as divisions per day (d), was calculated for the
entire period of growth from the expression :

where Ct and C0 are cell concentrations at times t and o, respectively.

EXPERIMENTS AND THEIR APPLICATION TO FIELD OBSERVATIONS

1. Nutrients

A. Experiments

A series of laboratory experiments was conducted to determine the relative
growth rates of Nannochloris, Stichococcus, and Nitsschia in each of several
different forms of nitrogen. Pure cultures of the three species were used, and
each was grown at its optimum salinity, as will be discussed in a later section.
All three algae were grown in media containing one mg. atom per liter of nitrogen
in each of the following forms : nitrate, nitrite, ammonia, urea, uric acid, 1-cystine,
asparagin, and glycocoll. The resulting growth rates, in divisions per day, are
given in Table I.

ECOLOGY OF PHYTOPLANKTON BLOOMS

201

TABLE I

The effect of the nitrogen source upon the growth rales
of Nannochloris, Stichococcus, and Nitzschia

Nitrogen source
(1 mg. A N/L)

Growth rate: Divisions per day

Nannochloris

Stichococcus

Nitzschia

N03

0.54

0.57

0.37

N02

0.65

0.57

0.40

NH,

0.72

0.61

0.04

Urea

0.62

0.56

0.21

Uric acid

0.68

0.59

0.23

1-cystine
Asparagin
Glycocoll

0.77
0.68
0.66

0.62
0.62
0.61

0.00
0.15
0.07

It may be seen that the diatom grew about equally well in nitrate and nitrite,
poorly in ammonia, and slowly or not at all in the organic N compounds. Both
Nannochloris and Stichococcus showed good growth in all of the forms of nitrogen
tested. However, the growth rate of both species in nitrite and nitrate was slightly
lower than that in ammonia, and, in general, was less than growth rates in the
organic compounds. The best growth of Nannochloris was observed in cultures
containing 1-cystine.

TABLE II

The effect of the N:P ratio upon the growth rates of Nannochloris,
Stichococcus, and Nitzschia

N:P ratio
(by atoms)

Growth rate: Divisions per day

Nannochloris

Stichococcus

Nitzschia

15:1
5:1

0.68
1.32

0.53
0.88

0.39
0.40

In another series of experiments, growth rates of the three species of algae were
determined in media containing the same concentrations of nitrate-nitrogen (one
mg. atom per liter) but two different concentrations of phosphate-phosphorus
(0.066 and 0.200 mg. atoms per liter). By varying the concentrations of phos-
phorus in this manner, the resulting N : P ratios by atoms in the two media
were 15:1 and 5:1, respectively.

In this experiment, increasing the P concentration, or lowering the N : P ratio,
had no effect upon the growth of the diatom, Nitzschia, but approximately doubled
the growth rates of Nannochloris and Stichococcus (Table II).

B. Application to field observations

Richards (unpublished data) measured the nitrogen present as uric acid,
ammonia, nitrite, and nitrate in the duck farm effluents, the tributaries to Moriches

202 JOHN H. RYTHER

Bay receiving these effluents, and in Moriches and Great South Bays on August
21, 1952. His analyses showed that uric acid could be detected only in the duck
farm effluents, ammonia was present in both the effluents and the tributary streams,
while nitrite and nitrate were found nowhere except as traces. No appreciable
concentrations of inorganic N in any of the forms tested for was found in either
Moriches Bay or Great South Bay.

Phosphorus, on the other hand, appears to have been present in excess of the
requirement of the phytoplankton during most of the year, not only in the polluted
estuaries, but also throughout the bay waters. Filtered water samples collected
from several locations in Great South Bay and Moriches Bay failed to support the
growth of Stichococcns, if untreated or enriched with phosphate. However, dense
growth of the alga occurred in all samples if enriched with ammonia-N indicating
that the latter w-as the principal limiting factor to the growth of the algae population.

The population maxima normally occurred in Moriches Bay and its polluted
tributaries. The distribution pattern of the organisms in Great South Bay and
Shinnecock Bay is strongly suggestive that their presence in these waters was
largely the result of the seaward flushing of Moriches Bay water, and that growth
was principally confined to an area close to the source of the nutrient rich duck
farm pollutants.

In its original state in the duck wastes, nitrogen occurs as excreted uric acid
and amino compounds contained in the undigested food residues. Investigations
in this laboratory by Vaccaro, Norton and Plunkett (unpublished) have disclosed
that bacteria present in the duck farm effluents are capable of decomposing uric
acid with great rapidity. In water samples collected from these effluents, the
contained uric acid was found to decrease to 10% of its original concentration in
an average of 15 hours at 2° C. The nitrogenous end products of this decom-
position were not investigated, but Copeman and Dillman (1937) observed that the
decomposition of the uric acid of guano in water was accompanied by an increase
in ammonia from 32.2% to 85.0% of the total nitrogen in four days.

According to the classical concept of the nitrogen cycle of the sea, the decom-
position of organic nitrogen to ammonia is followed by the nitrification of the
ammonia to nitrite and nitrate. This phase of the cycle appears never to occur
to any appreciable extent in the Moriches Bay area. Since nitrogen is the limiting
factor to the growth of the phytoplankton, it is utilized as quickly as it becomes
available and before decomposition to nitrite and nitrate can occur.

It follows that those organisms will have a distinct advantage which are able
to utilize the nitrogen in the earliest stages of its decomposition. The laboratory
experiments have demonstrated that Nannochloris and Stichococcns are particularly
well adapted to growth in organic nitrogen and its early decomposition products.
In contrast, the diatom, Nitsschia closterium, grew poorly or indifferently in these
N forms. The advantage of the "small forms" over the latter in the Moriches
Bay environment is therefore obvious.

Nitrogen and phosphorus are contained in duck faeces at a ratio of approxi-
mately 3.3 atoms of nitrogen to one atom of phosphorus. Total N and total P
data from the tributaries of Moriches Bay and from the bay itself show N : P
ratios ranging from 2.3 : 1 to 4.4: 1 (Richards, unpublished data). Approximately

ECOLOGY OF PHYTOPLAXKTON BLOOMS 203

one-half of the total phosphorus consisted of inorganic phosphate in these analyses.
On the other hand, no appreciable concentrations of available nitrogen were found
anywhere in the bay waters, as previously discussed. If it is assumed that half
of the total phosphorus and all of the total nitrogen were incorporated in particulate
matter, which consisted principally of algal cells, then the N : P ratio of the phyto-
plankton would range from 4.6 : 1 to 8.8 : 1 by atoms.

Various authors have pointed out that the ratio of nitrogen to phosphorus in
open ocean water is rather constant at approximately 15:1 by atoms, which
is closely reflected in the ratio of these elements in marine phytoplankton (Red-
field, 1934; Cooper, 1937, 1938; Fleming, 1940). Harvey (1940) found that
natural populations of diatoms utilized about 20 atoms of nitrogen to one atom
of 'phosphorus.

Ketchum and Redfield (1949) showed that laboratory cultures of Nitzschia
closteriuin contained N and P at a ratio of 11.6: 1, but cultures of six species of
Chlorophyta, including Stichococcus bacillaris, had N : P ratios ranging from
3.5: 1 to 6.6: 1, or two to three times as much phosphorus per atom of nitrogen
as the diatoms. This compares favorably with the estimated N : P ratio of the
phytoplankton of Moriches Bay which consisted predominantly of the Chlorophyta,
Stichococcus and Nannochloris.

These data indicate that there is a basic difference between the chemical com-
position of the green algae, typical of fresh and brackish water, and oceanic diatoms,
at least with respect to the N : P ratio in the cells. The growth studies bear out
this contention by demonstrating the fact that the Chlorophyta grow much more
rapidly in water containing three times as much phosphorus per atom of nitrogen
as normal sea water, while this increase in phosphorus has no effect upon the growth
of the diatom, Nitzschia closterium.

The N : P ratio of the polluted water of Moriches Bay and its environs thus
appears to be another factor favoring the growth of the "small forms" and their
competition with organisms similar to Nitzschia in their nutritional requirements.

2. Salinity

A. Experiments

To determine the effect of salinity upon the growth rates of Nannochloris,
Stichococcus, and Nitzschia, each species was grown in a series of dilutions of the
artificial medium described above. This series consisted of 100, 75, 50, 25, and 1
per cent solutions of the indicated concentrations of NaCl, MgCL, MgSO4, CaCl2,
and KC1. The concentrations of the other ingredients, which together account
for less than 0.5°/o<>, were not altered to avoid possible deficiencies of those nutrients.

The salinities of the resulting solutions, in °/oo» totaled, respectively, 34.51, 25.91,
17.44, 8.91, and 3.80, with an additional 0.048°/0o in each of the Nitzschia media as
a result of replacing NH4C1 with KNO3.

These experiments indicate a salinity optimum at or near that of full sea water
for Nitzschia, a typical marine species, while the "small forms" grow well within
a wide range of salinities, with optima at about 50% sea water, and may be
considered as brackish water species (Table III).

204

JOHN H. RYTHER

TABLE III

The effect of salinity upon the growth rates of Nannochloris,
Stichococcus, and Nitzschia

Growth rate : Divisions per day

Salinity: °/m

Nannochloris

Stichococcus

Nitzschia

34.51

0.45

0.44

0.40

25.91

0.61

0.55

0.34

17.44

0.69

0.61

0.28

8.91

0.56

0.58

0.09

3.80

0.37

0.52

0.00

B. Application to field observations

The highest concentrations of "small forms" normally occurred in Moriches
Bay in water of approximately 15°/0o salinity, or close to the physiological optima
of the organisms. While this does not imply that salinity was the factor con-
trolling the distribution of the phytoplankton, particularly in view of the preceding
discussion concerning the distribution of nutrients, it nevertheless represents
another environmental condition favoring the growth of the "small forms" over
that of the more typical marine forms.

Of still greater importance, however, is the fact that Nannochloris and Sticho-
coccus were able to grow remarkably well within the entire range of salinities
tested in the experimental cultures, in contrast to the diatom, which was unable
to grow at all in the low salinity cultures. This is particularly significant in view
of the wide range of salinities which were observed in Great South Bay and
Moriches Bay and their tributaries. Over a million "small forms" per ml. were
present throughout the late spring and summer in the upper Forge River, directly
opposite a group of duck farms, in water of less than l°/oo salinity. In addition,
evidence was obtained in the experiment described in Section IB that nitrogen-
enriched water from Fire Island Inlet (26°/oo salinity) was able to support a prolific
growth of Stichococcus.

The population which thus becomes established in the highly enriched rivers
and estuaries is able to continue growth as it is borne out to sea until its source
of nutrients becomes depleted. At no time does salinity limit its growth and
thereby permit the succession of other forms.

3. Temperature
A. Experiments

Growth rates of the three algae were determined for cultures grown at tem-
peratures ranging from 5° to 30° C. at intervals of 5° (Table IV). Nitzschia
was grown in the full strength artificial medium (34.51°/0o) while the "small forms"
were grown in the half-strength media (17.44°/0o) found to give optimum growth
in the preceding section.

ECOLOGY OF PHYTOPLANKTON BLOOMS

205

The diatom multiplied within a temperature range of 5-25° with optimum
growth at 15°. Its division rate was relatively high at the lower limit of 5°, but
decreased rapidly above 20°.

No appreciable growth of Nannochloris or Stichococcus occurred at temperatures
of 10° or lower, but both species grew rapidly at temperatures of 15-30°. No
clear-cut optimum could be detected between 15° and 25°, but growth rates of
both "small forms" decreased considerably at 30°.

B. Application to field observations

When field studies were begun on April 1, a mixed diatom bloom occurred in
the bays, which consisted of 890,000 Leptocylindrus minimus, 35,000 Thallasiosira
nana, and 12,000 Skelctonema costatum per ml. at a station in central Moriches
Bay. There were also present at that time 211,000 "small forms" per ml. On
May 13 the situation was reversed, with "small forms" dominating the plankton
and diatoms reduced to a total of 40,000 cells per ml. The latter subsequently
disappeared from the plankton and did not reappear until the following February.

The principle cause of this succession of dominants appears to be temperature.
Experimental evidence has shown that Nannochloris and Stichococcus are unable
to grow appreciably at the 10° temperature which prevailed throughout the embay-
ments on April 1, while the diatom, Nitsschia, maintained a relatively high growth
rate at temperatures as low as 5°.

During the period of May-September, temperatures in Moriches Bay ranged
from 13° to 30°. The growth experiments demonstrated that both Nannochloris
and Stichococcus divide rapidly within that temperature range. The seasonal
distribution of inorganic phosphorus in the Forge River at a station close to the
source of pollution indicates that the heaviest enrichment of the bay waters also
occurred during the same period of May-September. The situation therefore
exists that during that part of the year when the bay is most heavily enriched with
nutrients, temperatures may be expected which will fall approximately within the
range for optimum growth of the "small forms."

The month of July, 1952, was characterized by abnormally high temperatures
throughout the Long Island area. Water temperatures of the three embayments
under consideration ranged from 28-30° during the July 22-23 survey. A

TABLE IV

The effect of temperature upon the growth rates of Nannochloris,
Stichococcus, and Nitzschia

Growth rate: Divisions per day

T .or1

Nannochloris

Stichococcus

Nitzschia

5

0.00

0.00

0.29

10

0.14

0.12

0.42

15

0.65

0.48

0.48

20

0.80

0.48

0.40

25

0.71

0.43

0.19

30

0.32

0.26

0.00

206 JOHN H. RYTHER

decrease in the "small form" population at that time was generally observed through-
out the area. This may be explained by the fact that the July temperatures
exceeded the optima of the "small forms." The division rates of both Nannochloris
and Stichococcus at 30° were observed to be approximately one half of that at 25°
(Table IV). This population drop was obviously not associated with a nutrient
depletion, as indicated by unusually high inorganic phosphorus concentrations on
that date.

The presence of relatively large numbers of "small forms" between October and
April, when temperatures were presumably too low to permit their growth, may be
explained on the basis of the slow flushing time of Moriches Bay and the low
death rate of the organisms. This subject will be discussed at length elsewhere.

It is significant, however, that the slow rate of disappearance of the static
population during the winter months not only accounts for the presence of the
organisms during that part of the year when they are unable to grow, but it also
provides for a substantial seed population by the following spring, when conditions
again become favorable for growth. The latter may be one of the chief reasons for
the annual recurrence of the "small form" populations.

DISCUSSION

The "small forms," Nannochloris and Stichococcus, have been found to be
particularly well adapted for growth under the peculiar physical and chemical con-
ditions which occur in Great South Bay and Moriches Bay. In addition, it has
been brought out that the slow flushing time of the bay waters allows for the
retention of a considerable fraction of the summer bloom during the unproductive
winter months, thereby providing for a substantial seed population on the following
spring. This combination of circumstances alone is perhaps sufficient to account for
the presence of "small form" blooms in these waters every year.

It is significant, however, that ecological conditions not only approach an
optimum for the growth of the "small forms" but are also quite unsatisfactory for
the development of the diatom, Nitsschia clostcrium. It is perhaps suggestive that
the physiological characteristics of Nitsschia considered here are representative of
diatoms, dinoflagellates, and other plankton flora normally found in unpolluted
estuaries. While this provides a convenient hypothesis, such an assumption is
unwarranted on the basis of the existing evidence.

The utilization of ammonia and amino acids by the green algae has been
described by many workers (Schreiber, 1927; Braarud and F0yne, 1930; Algeus,
1946, 1949, 1950, and other papers). If diatoms in general are similar to
Nitsschia in the matter of their nitrogen utilization, their absence from the summer
blooms of Moriches Bay could be explained on that basis alone. This, however,
does not appear to be the case.

Harvey (1940) found that natural populations of mixed diatoms could utilize
nitrogen as nitrate, ammonia, urea, uric acid, and several amino acids, and that the
growth of some species appeared to be better in ammonia than in nitrate. Chu
(1943) observed that several diatoms (Nitzschia palca, Fragilaria crotonensis,
Asterionella gracillima') grew equally well in ammonia or nitrate.

A recent paper by Harvey (1953) describes the exponential growth of Nitsschia
closterium cultures in media containing ammonia as a source of nitrogen. Since

ECOLOGY OF PHYTOPLAXKTOX BLOOMS 207

this is contradictory to the results of the experiments described in this paper, in
which Nitsschia was found to grow very poorly in ammonia, these experiments were
repeated, using both the artificial medium and natural sea water containing one
mg. atom per liter of ammonia as a nitrogen source. The growth of Nitsschia
in both of these media was equally as poor as that reported in the earlier experi-
ments. Since the concentration of ammonia used here was probably higher than
that employed by Harvey, there is the possibility that this was the factor which
was toxic to the diatom (see Algeus, 1946) although it obviously was not so for
either Nannochloris or Stichococcus. It is equally plausible that the Nitsschia used
by Harvey and the present author represented different physiological varieties.

From the existing evidence of both the laboratory and the field observations,
it appears that the only time of the year when organisms other than the "small
forms" are able to dominate the phytoplankton is that period during which tem-
peratures are too low to permit the growth of the Chlorophyta. In this respect,
however, the natural situation in the bay waters differs somewhat from that which
might be expected from the culture work. The experiments have shown that the
"small forms" are able to grow very slowly if at all at the temperatures observed
in the bays after October, while diatoms did not begin to flourish there until
February. Although the bays were heavily enriched only during the duck growing
season, relatively high concentrations of nutrients appear to be present throughout
the year, presumably from the decomposition of the rich sediments in the tributaries
receiving the duck farm effluents.

A remaining possibility exists that the absence of diatoms and other forms from
the water between October and February may have been due, in part, to the
production of inhibitory substances by the "small forms" and the accumulation of
these products in the bay waters. This is suggested by the work of Pratt and his
group (1944, and earlier papers) who found that Chlorclla produces an antibiotic
which inhibits its own growth, and by experiments of Rice (1949) who demon-
strated that Chlorclla and Nitsschia fntstruluni produce substances which are
mutually inhibitory.

Lefevre and his co-workers (1951) have further shown that filtrates of both
laboratory cultures and of pond water containing blooms of different species of
algae produce inhibitory effects upon a wide variety of phytoplankton organisms.

It is perhaps significant that Pratt's group (Pratt, Oneto and Pratt, 1945)
found that the maximum inhibitory effect of Chlorella was produced by senescent,
non-dividing cultures, typical of the late fall and winter population of "small forms"
in Great South Bay and Moriches Bay.

SUMMARY

1. The phytoplankton bloom in Great South Bay and Moriches Bay during the
spring, summer, and early fall of 1952 consisted of the Chlorophyta, Nannochloris
atomus and Stichococcus sp., to the virtual exclusion of other species. These or-
ganisms persisted throughout the year, but were accompanied by minor diatom
blooms during the winter and early spring.

2. Growth rates of Nannochloris, Stichococcus, and the diatom, Nitsschia clo-
sterium were determined from laboratory cultures grown under various conditions

208 JOHN H. RYTHER

I

of temperature, salinity, and nutrients which are peculiar to the Great South Bay-
Moriches Bay area.

3. Nannochloris and Stichococcus grew well in nitrogen present as nitrate, ni-
trite, ammonia, urea, uric acid, cystine, asparagin, and glycocoll. Nitzschia grew
equally well in nitrate and nitrite, but showed poor growth in ammonia and the
organic N compounds.

4. Nannochloris and Stichococcus grew approximately twice as fast in media
containing an N : P ratio of 5 : 1 as they did in media with a 15 : 1 ratio of these ele-
ments. The growth rate of Nitzschia was the same in both media.

5. Nannochloris and Stichococcus appear to be brackish water forms with sa-
linity optima of about 170/00, but both species grew well within a salinity range of
3-34°/oo- Nitzschia, a typical marine form, was unable to grow in low salinity
water.

6. Nannochloris and Stichococcus grew at temperatures of 10-30° C., with very
slight growth at 10° and an optimum range of 15-25°. Nitzschia grew within a
temperature range of 5-25° with its best growth at 15°.

7. Pollution from duck farms bordering Moriches Bay heavily enriches the bay
waters with plant nutrients. The presence of organic nitrogen compounds and the
low ratio of nitrogen to phosphorus in the pollutants favor the growth of Nan-
nochloris and Stichococcus over that of the more typical estuarine phytoplankton.

8. The peculiar nature of the pollutants together writh low salinities and high
water temperatures occurring at the time and place of maximum enrichment of the
bay waters are factors which may explain the persistent dominance of the Nan-
nochloris-Stichococciis community in the annually recurring plankton blooms in
Great South Bay and Moriches Bay.

LITERATURE CITED

ALGEUS, S., 1946. Untersuchungen iiber die Ernahrungsphysiologie der Chlorophyceen. Bot.

Notiscr, 1946 : 129-278.

ALGEUS, S., 1949. Alanine as a source of nitrogen for green algae. Physiol. Plant., 2: 266-271.
ALGEUS, S., 1950. Further studies on the utilization of aspartic acid, succinamide, and asparagin

by green algae. Physiol. Plant., 3 : 370-375.
BRAARUD, T., AND B. FJ^YNE, 1930. Beitrage zur Kenntnis des Stoffwechsels im Meere. Au-

hand, Norske Videnskaps-Akademi Mat. Nat. Klasse, 1930, no. 14 : 1-24. Oslo.
BUTCHER, R. W., 1952. Contributions to our knowledge of the smaller marine algae. /. Mar.

Biol. Assoc., 31: 175-191.
CHU, S. P., 1943. The influence of the mineral composition of the medium on the growth of

planktonic algae. Part II. The influence of the concentration of inorganic nitrogen

and phosphate phosphorus. J. Ecol., 31 : 109-148.
COOPER, L. H. N., 1937. On the ratio of nitrogen to phosphorus in sea water. /. Mar. Biol.

Assoc., 22 : 177-182.
COOPER, L. H. N., 1938. Redefinition of the anomaly of the nitrate-phosphate ratio. /. Mar.

Biol. Assoc., 23 : 179.
COPEMAN, P. R. R., AND F. J. DILLMAN, 1937. Changes in the composition of guano during

storage. /. Agr. Sci., 27 : 178-187.
FLEMING, R. H., 1940. The composition of plankton and units for reporting populations and

production. Proc. Sixth Pac. Sci. Cong., Calif., 1939, 3 : 535-540.
HARVEY, H. W., 1940. Nitrogen and phosphorus required for the growth of phytoplankton.

/. Mar. Biol. Assoc., 24: 115-123.
HARVEY, H. W., 1953. Synthesis of organic nitrogen and chlorophyll by Nitzschia closterium.

J. Mar. Biol. Assoc., 31 : 477-487.

ECOLOGY OF PHYTOPLANKTON BLOOMS 209

KETCHUM, B. H., AND A. C. REDFIELD, 1949. Some physical and chemical characteristics of
algae grown in mass culture. /. Cell. Comp. Physiol., 33 : 281-299.

LEFEVRE, M., H. JAKOB AND M. NISBET, 1951. Compatibilites et antagonismes entre Algaes
d'eau douce dans les collections d'eau naturelles. Trav. Assoc. Internal. Limn. Theor.
et Appl, 11: 224-229.

PRATT, R., 1944. Studies on Chlorclla vulgaris. IX. Influence on growth of Chlorella of con-
tinuous removal of chlorellin from the culture solution. Amcr. J. Bot., 31 : 418-421.

PRATT, R., J. F. ONETO AXD J. PRATT, 1945. Studies on Chlorclla z'ltlgaris. X. Influence of
the age of the culture on the accumulation of chlorellin. Amer. J. Bot., 32 : 405^408.

REDFIELD, A. C., 1934. On the proportions of organic derivatives in sea water and their relation
to the composition of plankton. James Johnstone Memorial Volume. The University
Press. Liverpool.

RICE, T. R., 1949. The effects of nutrients and metabolites on population of planktonic algae.
Ph.D. Thesis, Harvard University, Department of Biology.

SCHREIBER, E., 1927. Die Reinkultur von marinen Phytoplankton und deren Bedeutung fur
die Erforschung der Produktionsfahigkeit des Meerwassers. ll'iss. Meersuntersuch.
Abt. Helgoland N.F., Bd. 16, No. 10:1-34.

SVERDRUP, H. U., M. W. JOHNSON AND R. H. FLEMING, 1942. The oceans, their physics, chem-
istry, and general biology. Xew York, Prentice-Hall, Inc.

THE PHYSIOLOGY OF INSECT DIAPAUSE. VIII.

QUALITATIVE CHANGES IN THE METABOLISM

OF THE CECROPIA SILKWORM DURING

DIAPAUSE AND DEVELOPMENT1

HOWARD A. SCHNEIDERMAN 2 AND CARROLL M. WILLIAMS

The Bioltxjical Laboratories, Harvard University, Cambridge 38, Massachusetts

In the Cecropia silkworm the termination of pupal diapause and the progress of
adult development are accompanied by large and predictable changes in respiratory
metabolism. Thus, as described in the preceding paper of this series, the respira-
tion of Cecropia midway its adult development is approximately seven times that of
the diapausing pupa. In the present study efforts were made to ascertain the
enzymatic basis of the quantitative changes in respiration. Considerable evidence
was already at hand pointing to pronounced alterations in the cytochrome system
in synchrony with the termination of diapause in the eggs of the grasshopper Mela-
noplus (Bodine and Boell, 1934a. 1934b), the eggs of the commercial silkworm
Bomby.r (Wolsky, 1943), and the pupa of the Cecropia silkworm (Sanborn and
Williams, 1950 ; Pappenheimer and Williams, 1952 ; Schneiderman and Williams,
1952). For this reason attention focussed on the role of the terminal oxidases in
relation to diapause and development.

The principal terminal oxidases which have been demonstrated in animals and
higher plants are cytochrome oxidase, flavoproteins, and copper-containing proteins
such as ascorbic acid oxidase and tyrosinase (Lardy, 1949; Goddard and Meeuse,
1950). Among these, all save ascorbic acid oxidase are thought to function as
terminal oxidases in certain animal cells, though the precise role which tyrosinase
may play has never been satisfactorily denned (Sussman, 1949).

In animals such as insects, when hemoglobin and other erythrocruorins are
absent, carbon monoxide inhibits cytochrome oxidase and tyrosinase (Warburg.
1949), but apparently fails to inhibit flavoproteins or any other enzymes or sub-
strates. It is true that carbon monoxide forms spectroscopically identifiable com-
plexes with certain peroxidases, but peroxidase activity remains uninhibited (The-
orell, 1953). Carbon monoxide's inhibition of cytochrome oxidase and tyrosinase
can be distinguished in that the former is reversed by light while the latter is not
(Warburg and Negelein, 1928; Kubowitz, 1937, 1938). Carbon monoxide there-
fore affords a remarkably specific tool for tracking the participation of the cyto-
chrome oxidase system in biological reactions. In the present study we have ex-
ploited this specificity in an effort to characterize the terminal oxidases of the Ce-
cropia silkworm during diapause and development.

1 This study was aided by the Lalor Foundation, by a grant from the U. S. Public Health
Service, and by an Institutional Grant to Harvard University from the American Cancer
Society.

2 Former Atomic Energy Commission Fellow. Present address : Department of Zoology,
Cornell University, Ithaca, New York.

210

METABOLISM OF SILKWORM 211

For reasons considered elsewhere ( Schneiderman and Feder, 1954), the effects
of high concentrations of carbon monoxide were studied by positive pressure tech-
niques. Animals were placed in transparent, air-filled, polymethyl methacrylate
(Lucite) chambers, compressed with carbon monoxide, and measurements of res-
piration performed by means of respirometers developed for use at high pressures.
Under these conditions the oxygen tension remained unchanged at its normal value
of one-fifth of an atmosphere, while the carbon monoxide pressure could be increased
to as high as seven atmospheres. The positive pressure techniques were supple-
mented by experiments performed at atmospheric pressure and testing the effects
of carbon monoxide, oxygen tension, and cyanide.

MATERIALS AND METHODS

Diapausing pupae and developing adults of the giant silkworm, Platysamia
cecropia, were used as experimental animals. In the case of the pupal material the
brains were commonly removed to stabilize the animals in permanent diapause
(Williams, 1946). In order to avoid the complication of "post-injury metabolism"
(Schneiderman and Williams, 1953a), one month or longer was allowed for the
recovery of animals subjected to surgical manipulation. All experiments were
performed at 25° C.

1. Metabolic studies

In studies performed at atmospheric pressure the insects were placed in indi-
vidual 45-cc. vessels of the type described previously (Schneiderman and Williams,
1953a), and the oxygen consumption determined by the Warburg method. Ex-
periments at positive pressures were carried out in high pressure respirometers
(Schneiderman and Feder, 1954). Measurements were begun approximately I1/-*
hours after compression and continued for 20 to 30 hours in order to compensate
for the discontinuous release of carbon dioxide by diapausing pupae (Punt, 1950;
Schneiderman and Williams, 1953a, 1953b). The gas volume of the individual
respirometers was sufficiently large to preclude any important decrease in oxygen
tension during the experimental period.

At the end of positive pressure experiments, acid was added to the alkali and the
displaced carbon dioxide measured volumetrically (Schneiderman and Williams,
1953a). The total output of carbon dioxide was measured from the moment the
respirometers were sealed to the moment the animals were removed, and the average
carbon dioxide production estimated during this period. The over-all respiratory
quotient for the duration of the experimental period was calculated from the average
carbon dioxide production divided by the average oxygen consumption.

2. Experimental gases

The gases were obtained in commercial cylinders and assayed as follows :

Nitrogen (Airco), 99.5% N, plus less than 0.5% oxygen. Oxygen (Airco),

99.5% oxygen plus less than 0.5% nitrogen. Carbon monoxide (Matheson Co.),

96.8% carbon monoxide, 0.36% carbon dioxide, 0.97% hydrogen, 1% nitrogen,

0.8% saturated hydrocarbons, 1.19 mg. iron per liter, 0.32 ing. sulfur per liter.

212 H. A. SCHNEIDERMAX AXD C. M. WILLIAMS

Prior to its use, the carbon monoxide was bubbled through a solution of 10% sodium
hydroxide to remove carbon dioxide and iron carbonyl compounds.

In one series of experiments, 200 liters of extremely pure carbon monoxide were
prepared by the action of hot concentrated sulfuric acid (C. P. reagent) on formic
acid (analytical reagent). The carbon monoxide was passed, in turn, through an
aqueous solution containing 5% pyrogallic acid and 25% KOH, a dry ice-acetone
trap, CaCL, Mg(ClO4)2 ("Anhydrone"), a liquid nitrogen trap, and then com-
pressed to 100 psi in small steel cylinders. Since the effect of this pure carbon
monoxide on respiration could not be distinguished from that of the alkali-treated
commercial carbon monoxide, the less expensive commercial gas was used in subse-
quent experiments.

Mixtures of carbon monoxide, oxygen, and nitrogen were prepared under pres-
sure in steel or Lucite tanks and their compositions checked by gas analysis (Scho-
lander and Roughton, 1953).

3. Cyanide experiments

To appraise the metabolic effects of cyanide, the insects were first weighed and
their water content assumed to equal 75 per cent of the live weight. Then, by means
of an extremely small (30) gauge hypodermic needle, each pupa was injected just
lateral to the midline of the thoracic tergum with 0.05 to 0.09 ml. of freshly prepared
neutralized KCN. The latter's concentration was regulated to establish a specific
final concentration after dilution with the fluid volume of the insect. At the pH
of the insect, KCN exists almost wholly as HCN ; the molar concentration of HCN
within the insect was calculated on this basis.

Immediately after injection each pupa was enclosed in a Warburg vessel. As
recommended by Robbie ( 1946 ) , mixtures of KCN and KOH were placed in the
vessel for the absorption of carbon dioxide ; in this manner the HCN concentration
of the chamber was balanced against the internal concentration established within
the insect by the injection. At internal HCN concentrations of 10~3 M or greater,
a constant external HCN tension of 5 X 10~4 was employed — the highest concentra-
tion that one can establish by means of KCN-KOH mixtures. In one series of
experiments the experimental animals were equilibrated via the tracheal system with
a specific tension of HCN for 60 hours and then studied without the actual injection
of cyanide.

EXPERIMENTAL RESULTS

1. Effects of carbon mono.vidc of atmospheric pressure on the respiration of dia paus-
ing pupae

In experiments performed on two diapausing pupae and five brainless diapausing
pupae the rate of oxygen consumption was first measured in air, then in 6 per cent
oxygen plus 94 per cent nitrogen, and, finally, in 6 per cent oxygen plus 94 per cent
carbon monoxide (carbon monoxide/oxygen = 16:1). During the course of the
experiment the respirometers were flushed periodically with the experimental gases
to prevent the uptake of oxygen from appreciably diminishing the oxygen tension in
the vessels. As recorded in Table I, it is evident that between the first and eighth
hours of exposure to carbon monoxide only one pupa showed any appreciable inhi-

METABOLISM OF SILKWORM

213

TABLE I

The effects of carbon monoxide on the oxygen consumption of two
diapausing and five brainless diapausing pupae*

Rate of oxygen consumption

Relative rate of oxygen

(mm.Vgm. live wt./hr.)

consumption in 16:1 CO/O2 (%)

Type of pupa

Between

Between

Air

16:1 N2/O:

After 1

1st and

8th and

8th hour

28th hour

Diapausing

9.0 (76%)

11.8 (100%)

126

87

—

Diapausing

11.5 (91%)

12.6 (100%)

132

132

—

Brainless diapausing

8.8 (106%)

8.3 (100%)

116

94

144

Brainless diapausing

9.7 (99%)

9.8 (100%)

85

94

85

Brainless diapausing

9.9 (89%)

11.1 (100%)

95

93

84

Brainless diapausing

11.8 (91%)

13.0 (100%)

106

98

89

Brainless diapausing

13.4 (106%)

12.7 (100%)

73

65

79

* All measurements performed at a total pressure of one atmosphere.

%

bition of respiration. Between the eighth and twenty-eighth hours four of the brain-
less diapausing pupae showed about 15 per cent inhibition.

These results demonstrate that only a small fraction of the metabolism of Ce-
cropia pupae is carbon monoxide-sensitive when the carbon monoxide/oxygen ratio
is 16: 1. To test the effects of still higher ratios, a considerable number of experi-
ments were performed making use of the positive pressure respirometers.

2. Effects of high pressures of nitrogen on the respiration of brainless diapausing
pupae

Figure 1 illustrates results typical of a number of control experiments in which
the respiration of five brainless diapausing pupae was determined in one atmosphere
of air and then in air compressed with five atmospheres of nitrogen. It is evident
that positive pressures of five atmospheres of an inert gas such as nitrogen were
without notable effects on either the rate of oxygen consumption or carbon dioxide
output. The slight increase in carbon dioxide output is most probably an artifact
attributable to the flushing of stored carbon dioxide from the animal during the
period of decompression (Schneiderman and Williams, 1953b).

3. Effects of high pressures of carbon monoxide on the respiration of brainless
diapausing pupae

Brainless diapausing pupae were placed in air-filled respirometers and their
respiration measured after compression with five atmospheres of nitrogen ; they
were then decompressed to air and the measurements repeated after recompression
with five atmospheres of carbon monoxide (carbon monoxide/oxygen ratio of
25 : 1 ) . Figure 2 records the results obtained in an experiment utilizing five
brainless diapausing pupae. A comparison of the respiration in nitrogen and in
carbon monoxide reveals that about one-third of the oxygen consumption was
inhibited by the high pressure of carbon monoxide. Carbon dioxide production
was affected to a lesser degree to yield an apparent increase in the respiratory
quotient.

214

H. A. SCHXEIDERMAN AND C. M. WILLIAMS

MM3>

/ A Nil

cn

02 CONSUMPTION

^__^— 5ATM H2 m

ffllaa

C02 PRODUCTION

[ftl±l

MM^/ HR

/ANIMAL

I ATM AIR | |

02 CONSUMPTION

C02 PRODUCTION

DIAPAUSE

I ATM AiR f I

DEVELOPMENT

FIGURE 1. The average respiration of five brainless diapausing pupae in air at one atmos-
phere compared with the average respiration of the same animals in air compressed with
5 atmospheres of nitrogen.

FIGURE 2. The average respiration of five brainless diapausing pupae in air compressed
with 5 atmospheres of nitrogen compared with the average respiration of the same animals in air
compressed with 5 atmospheres of carbon monoxide.

FIGURE 3. The average respiration of three brainless diapausing pupae, lacking abdominal
ganglia, in air at one atmosphere pressure, in air compressed with 5 atmospheres of nitrogen,
and in air compressed with 5 atmospheres of carbon monoxide.

FIGURE 4. The average oyxgen consumption of three developing animals on the sixth day
of adult development, in air at one atmosphere pressure, and in air compressed with 5 atmospheres
of carbon monoxide. The normal average oxygen consumption of diapausing pupae in air at
one atmosphere pressure is also recorded.

METABOLISM OF SILKWORM 215

4. Effects of liigh pressures of carbon monoxide on the respiration of brainless
diapausing pupae lacking abdominal ganglia

Taken at face value, experiments of the type just considered suggest that about
one-third of the metabolism of diapausing pupae is mediated via the cytochrome
oxidase system. However, it was noted that diapausing pupae showed a con-
spicuous depression in the frequency and amplitude of spontaneous muscular move-
ments of the abdominal segments in the presence of high pressures of carbon
monoxide. It seemed possible that the observed inhibition by carbon monoxide
might arise from a suppression of these movements rather than from inhibition of
the pupa as a whole.

This possibility was tested by a repetition of the preceding experiment on a
series of brainless diapausing pupae in which the intersegmental muscles of the
abdomen had previously been denervated by removal of the chain of eight abdominal
ganglia. A total of four such animals were studied in detail. Figure 3 records
the respiratory exchange of three of these individuals in air, in air compressed
with five atmospheres of nitrogen, and in air compressed with five atmospheres
of carbon monoxide. The rates of oxygen consumption under all three conditions
were indistinguishable — a result which indicates that in the absence of muscular
movements of the abdomen the metabolism of diapausing pupae is insensitive to
carbon monoxide.

5. Effects of carbon monoxide on pilocar pine- stimulated muscular activity and
respiration

It was known from studies to be considered elsewhere that the injection of
suitable concentrations of pilocarpine causes diapausing pupae to move their
abdomens continuously for up to a year thereafter. Consequently, animals stimu-
lated in this manner afforded ideal material for testing the sensitivity of the
abdominal motion and the accompanying respiration to inhibition by carbon
monoxide. To this end, each of a series of nine diapausing pupae was injected
with 0.1 ml. of 0.1 M pilocarpine hydrochloride that had previously been neutralized
to pH 6.6 with sodium hydroxide. Two days later, the pupae were enclosed in an
air-filled Lucite chamber, compressed with specific pressures of carbon monoxide
or oxygen, and the effects on abdominal motion noted.

Carbon monoxide inhibited the abdominal motion to a degree dictated by the
carbon monoxide/oxygen ratio. Thus, when a ratio of 10 : 1 was established by the
addition of 30 psi carbon monoxide to the initial atmosphere of air, abdominal
motion was markedly inhibited. When the ratio was then decreased to 3:1 by
the addition of 7 psi of oxygen, vigorous movements reappeared. Further com-
pression with carbon monoxide once again restored the inhibition. In virtually all
cases abdominal motion ceased when the ratio was as high as 15 : 1 , but was resumed
within 10 minutes after the decompression and return to air.

Ten days after the experiment just considered the oxygen consumption of five
of the continuously wriggling pupae was measured at atmospheric pressure in air,
and in specific mixtures of oxygen, nitrogen, and carbon monoxide. The results
summarized in Table II reveal that 16: 1 carbon monoxide/oxygen caused a prompt
cessation of the abdominal motion and inhibited the oxygen consumption by approxi-
mately 30 per cent.

216

H. A. SCHNEIDERMAN AND C. M. WILLIAMS

TABLE II

The effects of carbon monoxide on the oxygen consumption and abdominal

motion of five diapausing pupae injected with 0.1 ml. of 0.1 M

pilocarpine hydrochloride*

Rate of oxygen consumption
(mm.'/animal/hour)

Relative rate of oxygen consumption
in 16:1 CO/O2 (%)

Air

16:1 Ni/O2

During 1st hour

During 3rd hour

81 (89%) (+)
86 (77%) ( + )
113 (94%) (+)
133 (116%) ( + )
179 (80%) (+)

91 (100%) ( + )

112 (100%) (+)
120 (100%) (+)
115 (100%) (+)
223 (100%) (+)

52 (-)
90 ( -)
58 (-)
81 (-)
75 (-)

52 (-)
87 (-)
58(-)
81 (-)
75 (-)

Average: 118 (89%)

132 (100%) (+)

72 (-)

71 (-)

* All measurements performed at a total pressure of one atmosphere.
(+) Records the presence of abdominal motion; ( — ) the absence of same.

Taken along with the previously mentioned experiments, these findings provide
a consistent body of evidence that the contraction of the intersegmental muscles of
the diapausing pupa is inhibited by carbon monoxide, whereas the other tissues of
the dormant insect are not inhibited by carbon monoxide.

6. Effects of carbon monoxide on the increased respiration accompanying adult
development

After the termination of pupal diapause the onset and progress of adult develop-
ment are accompanied by a rapid increase in respiration. Thus on the sixth day of
adult development the average respiration is approximately five times that during
diapause. In order to ascertain the carbon monoxide-sensitivity of this additional
metabolism accompanying development, the respiration of four animals on the sixth
day of adult development was first measured in air and then in air compressed with
five atmospheres of carbon monoxide. The results, illustrated in the case of the
three individuals in Figure 4, demonstrate a striking effect of this 25 : 1 carbon
monoxide/oxygen ratio on the respiration of developing animals. About two thirds

TABLE III

The effects of carbon monoxide on the oxygen consumption of developing adults*

Rate of oxygen consumption

Relative rate of oxygen consumption

Days after

(mm.Vgm. live wt./hr.)

in 16:1 CO/O2 (%)

initiation

of adult

development

Air

16:1 N2/O2

During 1st
hour

During 4th

hour

During 8th
hour

2

54 (102%)

53 (100%)

68

56

55

51

84 (129%)

65 (100%)

64

52

39

6^

99 (121%)

82 (100%)

68

56

54

* All experiments performed at a total pressure of one atmosphere.

METABOLISM OF SILKWORM

217

of the oxygen consumption was inhibited and the metabolism dropped to a level
almost as low as that of diapausing pupae. Consequently, it appears that the
increased oxygen consumption accompanying adult development is completely
or almost completely inhibited by carbon monoxide.

Table III records analogous findings in an experiment in which three developing
adults were exposed to a mixture of carbon monoxide and oxygen at a total pres-
sure of one atmosphere. To compensate for the utilization of oxygen, the Warburg
vessels were reflushed with the experimental gas every 2l/2 hours. It will be noted
that the 16 : 1 carbon monoxide/oxygen inhibited the oxygen consumption of the
developing insects by approximately 50 per cent.

7. The effects of carbon monoxide on the post-injury metabolism of diapausing
pupae

In addition to the increased metabolism which accompanies the onset of adult
development, the pupa during diapause can undergo a substantial increase in its
metabolism under certain experimental conditions. Thus, after small localized
injury to the pupal integument, the oxygen consumption and carbon dioxide pro-
duction are considerably enhanced for one to several weeks thereafter (Sussman,
1952; Schneiderman and Williams, 1953a). This result has been regularly
observed in both normal pupae and in pupae immobilized by prior removal of the
abdominal ganglia.

Experiments were performed to test the sensitivity of the injury metabolism
to carbon monoxide. To this end, the brains and abdominal ganglia were removed
from six diapausing pupae. Two months later the rate of oxygen consumption
was determined for each animal. A V-shaped 4-mm. incision was then made in the
thoracic tergum of each animal, and the rate of oxygen consumption measured one
day later. Three of the pupae were then placed in air-filled respirometers, com-
pressed with five atmospheres of carbon monoxide, and the measurements repeated.

TABLE IV

The effects of carbon monoxide on the injury-stimulated respiration of brainless
diapausing pupae lacking abdominal ganglia and connectives

Rate of oxygen consumption* (%)

No. 1

No. 2

No. 3

No. 4

No. 5

No. 6

Prior to injury

In air

100

100

100

100

100

100

24 hours after injury

In air

158

208

108

198

146

146

30 hours after injury

In air

153

183

118

—

—

—

In CO

—

—

—

183

152

128

Percentage difference between 1st
and 2nd post-injury measure-
ments

-3

-12

+9

-8

+4

-12

* Initial pre-injury oxygen consumption varied from 91 to 143 mm.3/animal/hour.

218

H. A. SCHNEIDERMAN AND C. M.. WILLIAMS

The respiration of the other three pupae was again measured in air at one atmos-
phere. The results recorded in Table IV show that the extra respiration stimu-
lated by injury is uninhibited by carbon monoxide.

8. Effects of cyanide on the respiration of diapausing pupae

Diapausing pupae were injected with specific concentrations of cyanide and
their oxygen uptake then ascertained. A typical set of measurements is plotted
in Figure 5. In control experiments in which distilled water was injected, the
oxygen uptake began to increase about five hours after the injection, and the
typical pattern of injury metabolism became apparent. When cyanide was injected

O

I

-I
<f

5

DISTILLED M.O

I0"3 M KCN

5. IO'3 M KCN

-• 1

30

HOURS

FIGURE 5. The effects of cyanide injection on the oxygen consumption of diapausing pupae.
Concentrations of cyanide refer to calculated final internal concentrations.

METABOLISM OF SILKWORM

219

to attain a final internal concentration of 10~3 M, a slight inhibition was usually
observed, though in some cases the effect did not differ from that of distilled water.
With further increase in cyanide to a final concentration of 2 X 10~3 M, an
immediate inhibition of 50 per cent always occurred which persisted at a steady
level for about ten hours and then gradually returned to normal. Still higher
concentrations of cyanide (5 X 10~3 iM and above) caused a prompt inhibition of
70 to 90 per cent, followed by the death of the animal several days later.

In the interpretation of experiments of this type it is necessary to separate the
effects of injury from those of cyanide. This can most easily be accomplished by

TABLE V

The effects of cyanide on the respiration of diapausing pupae

Prior to injection

After injection

Calculated internal
concentration
of HCN

Average
rate of oxygen
consumption
(mm.3 /gin. live
wt./hr.)

Lowest hourly
rate of oxygen
consumption as
per cent of
average initial
rate*

Lowest hourly
rate of oxygen
consumption in
48-hour period as
per cent of
average initial
rate

Behavior

0

10.7

97

98

Normal

11.4

87

96

15.8

79

104

17.4

92

108

Average

89

102

5X10~4 M

9.9

92

80

Normal

10.3

96

102

10.5

88

97

12.1

91

89

12.5

94

98

Average

92

93

5X10~4 M

8.1

95

93

Normal

(Exposed to HCN

9.1

95

91

gas for 60 hours

11.0

91

98

but not injected)

13.0

97

87

13.5

90

96

14.7

93

76

Average

93

90

1X10-3 M

8.5

87

110

Normal, or slight decrease in

9.7

95

107

abdominal motility

1 1 4

eo

QJ.

11.4

94

51

Conspicuous decrease in ab-

10.5

83

85

dominal motility for 12 to

12.2

100

72

24 hours

28

93

67

Average

91

84

* Measured during a four- to eight-hour period prior to cyanide injection

220

H. A. SCHNEIDERMAN AND C. M. WILLIAMS
TABLE V (Continued)

Calculated internal
concentration
of HCN

Prior to injection

After injection

Average
rate of oxygen
consumption
(mm.Vgm. live
wt./hr.)

Lowest hourly
rate of oxygen
consumption as
per cent of
average initial
rate*

Lowest hourly
rate of oxygen
consumption in
48 hour period as
per cent of
average initial
rate

Behavior

2X10-3 M

8.1
11.0
13.0

95
91
97

53
50

52

Electrically inexcitable for
one day. Spontaneous ab-
dominal movements reap-
peared after 2 days

Average

94

52

5X10-3 M

9.1
13.5
14.7

96

90
93

39
13
36

Died

Average

93

29

1X10-2 M

11.9

13.7
20.2

90
97
93

21
19
26

Died

Average

93

22

focusing attention on the oxygen uptake in the first few hours after injection and
prior to the onset of the injury metabolism. As a measure of cyanide inhibition
we chose the lowest oxygen uptake measured over a one hour interval during this
period. To prevent any normal hour-to-hour variations in oxygen uptake from
being interpreted as cyanide inhibition, we also recorded the lowest hourly rote of
oxygen consumption during a four- to eight-hour period prior to cyanide injection.
Table V summarizes such calculations on a series of 31 diapausing pupae. Detect-
able inhibition of abdominal motion and of respiration appeared only when the
internal cyanide concentration, as calculated, was increased to or above 10~3 M.
In the evaluation of these findings it is worth recalling a technical difficulty
mentioned in the section on Methods ; namely, that it was impossible by the use
of KCN-KOH mixtures to establish HCN concentrations higher than 5 X 1O4 M
in the air surrounding the insect. Consequently, at internal concentrations higher
than 5 X 10~4 M, detoxification and unspecific combinations of the injected HCN,
along with the loss of HCN by diffusion from the tracheal system, necessarily
reduced the internal concentrations below the calculated values. This fact com-
plicated a decision as to the cyanide-sensitivity of the abdominal muscles. How-
ever, in experiments of short duration performed on a total of 18 diapausing pupae,
we found that 10~3 M cyanide uniformly impaired the contractility of the abdominal
muscles of normal pupae and totally eliminated the contractility of pilocarpine-
stimulated muscles. In the latter case the muscles no longer responded to electrical
stimulation. Consequently, we conclude that the carbon monoxide-sensitive abdom-
inal muscles of diapausing pupae are likewise cyanide-sensitive.

'METABOLISM OF SILKWORM

221

Warburg (1949) emphasizes the fact that cyanide is a specific inhibitor of heavy
metal enzymes only at concentrations up to about 10~3 M ; above this level cyanide
undergoes significant combinations with a large number of substrates and metabolic
intermediates. In the case of Cecropia we attribute the lethal effects of cyanide
concentration of 5 X 10~3 M and above to these unspecific side-reactions of cyanide.

9. Effects of cyanide on tJic respiration of developing adults

After the termination of diapause and the initiation of adult development, the
effect of cyanide was easier to ascertain by virtue of the absence of injury metab-
olism, the latter being peculiar to diapausing pupae (Schneiderman and Williams,
1953c). Table VI summarizes the inhibition by cyanide of the respiration of 19

TABLE VI
The effects of cyanide on the respiration of developing adults

Calculated

internal
concentration
of HCN

Days after
initiation
of adult
development

Average rate
of oxygen
consumption
prior to
injection
(mm.Vgm.

Oxygen consumption as per cent of average
initial oxygen consumption at intervals
after injection

Behavior
after
injection

live wt./hr.)

1.5 hrs.

8.5 hrs.

20 hrs.

26 hrs.

40 hrs.

10~4 M

3

69

(124)

(66)

(110)

Emerged

9

119

79

52

122

Emerged

12

136

83

41

124

Emerged

15

166

70

39

54

Died

17

279

(92)

(20)

(30)

Died

Average

77

44

100

2X10-"M

10

112

90

5

5

Died

12

114

68

7

7

Died

12

140

67

6

6

Died

13

186

63

5

5

Died

14

206

69

12

12

Died

Average

71

7

7

5X10-4 M

9

95

57

4

6

Died

10

100

52

6

7

Died

10

118

43

25

6

Died

11

128

41

10

7

Died

13

160

42

12

7

Died

Average

47

11

7

10-3 M

8

110

27

6

7

Died

11

150

21

7

6

Died

12

131

30

9

7

Died

13

162

23

9

7

Died

Average

25

8

7

Figures in parenthesis () were not included in the average since these animals were not at
comparable stages of adult development.

222

H. A. SCHNEIDERMAN AXD C. M. WILLIAMS

developing adults. The results have been averaged for animals between the 8th
to 15th day of adult development. In contrast to the findings on diapausing pupae,
cyanide at final concentrations of 2 X 1O4 M or higher was lethal ; moreover 10~4 M
cyanide now inhibited the oxygen uptake 35 to 80 per cent. In Table VI it also
appears that the proportion of total metabolism which was cyanide-sensitive under-
goes definite increase during the course of the twenty-two day period of adult
development.

10. Effects oj oxygen tension on the respiration of brainless diapausing pupae
lacking abdominal ganglia

The experimental results, up to this point, demonstrate that systematic changes
occur in the insect's dependency on metabolism sensitive to carbon monoxide and
to cyanide. Aside from the intersegmental muscles of the abdomen, the metabolism
of the diapausing pupa and the extra metabolism which it exhibits in response to
injury are substantially insensitive to carbon monoxide and cyanide ; by contrast,
the metabolism of the developing insect is markedly inhibited by these agents.

100

90

80

<r

30

10

21

100

270

02 TENSION AS PERCENT OF I ATMOSPHERE

FIGURE 6. The effects of oxygen tension on the oxygen consumption of three brainless
diapausing pupae lacking abdominal ganglia and connectives. The abscissa is marked off in
arbitrary units.

METABOLISM OF SILKWORM

Carbon monoxide-stable respiration has generally been found to require oxygen
tensions far higher than does carbon monoxide-sensitive respiration (see Discus-
sion). For this reason the effect of oxygen tension was studied in relation to the
metabolism of diapause and development.

The oxygen consumption of three brainless diapausing pupae, previously
immobilized by the removal of their abdominal ganglia, was determined at a series
of oxygen tensions ranging from 1 to 270 per cent of an atmosphere of oxygen.
The results of measurements at six different oxygen tensions are recorded in
Figure 6. Each individual determination represents a steady-state value obtained
after exposure to the gas mixtures for several hours. After each determination
the pupae were returned to air for three days. The respiration in air was then
re-determined before the pupae were exposed to a new oxygen tension.

As Figure 6 indicates, the oxygen consumption was independent of oxygen
pressure at tensions from 5 per cent to 100 per cent of an atmosphere. In 3 per
cent oxygen a conspicuous decrease was evident. But even in one per cent oxygen
the average oxygen consumption was still 40 per cent of that in air. At the
extremely high oxygen tension of 2.7 atmospheres, there appeared to be a slight
depression, presumably attributable to the toxic effect of oxygen (Williams and
Beecher, 1944).

Four pupae were stored in one per cent oxygen for ten hours. When returned
to air, the oxygen consumption increased approximately 25 per cent above the
normal rate in air and persisted at this level for several hours. This observation
suggests the accumulation of a small but definite oxygen debt at the low oxygen
tension. It is clear, however, that diapausing pupae possess limited ability to
accumulate an oxygen debt for, as demonstrated in numerous experiments, several
days of exposure to 0.5 per cent oxygen is lethal (L. D. 50c/'c = 3 days at 25° C.).

DISCUSSION
1. Insensitvuity of insects to compression

Interpretation of the experimental results obviously requires a decision as to
whether the experiments at high gaseous pressures were complicated by unspecific
or narcotic effects of pressure per sc (Behnke, 1940; Lawrence ct a!., 1946).

As far as nitrogen is concerned, narcotic effects have not been demonstrated
in any experiments on insects. Chadwick and Williams (1949) found that
Drosophila could fly in 4.5 atmospheres of nitrogen plus one atmosphere of air,
although the wingbeat frequency was decreased because of the increased gas density.
Case and Haldane (1941) observed that Drosophila was active at 10 atmospheres
of nitrogen plus one atmosphere of air, and Williams (unpublished) found that
seven hours of exposure to 24 atmospheres of nitrogen plus one atmosphere of air
failed to affect the vitality of Drosophila upon subsequent return to air. More-
over, experiments on the Cecropia silkworm at all stages of development, from egg
to adult, have demonstrated that prolonged exposure to 6.7 atmospheres of nitrogen
plus one atmosphere of air fails to retard embryonic or adult development, heart
beat, movement, or the spinning of the cocoon (unpublished observations). And
as demonstrated in the present study, the oxygen consumption was the same in air
and in air compressed with five atmospheres of nitrogen. These results give
assurance that nitrogen, at the pressures utilized, was not a narcotic.

224 H. A. SCHXEIDERMAX AXD C. M. WILLIAMS

\Ye feel that the same conclusion is valid in the case of carbon monoxide. Thus,
the oxygen consumption of immobilized diapausing pupae was the same in air
and in air compressed with five atmospheres of carbon monoxide. Moreover,
we shall subsequently show that many of the effects of high pressures of carbon
monoxide on the post-diapausing insect are reversed by light (Schneiderman
and Williams, 1954). Consequently, high pressure techniques appear to be useful
and uncomplicated tools for experiments on insects.

2. Significance of carbon monoxide-sensitive respiration

As mentioned in the Introduction, suitable pressures of carbon monoxide are
known specifically to inhibit the function of two enzymes, cytochrome oxidase
and tyrosinase. Inhibition of cytochrome oxidase is reversed by light ; inhibition
of tyrosinase is not. Though the light-reversibility of carbon monoxide's action
on post-diapausing Cecropia has already been described in the case of the male sex
cells of Cecropia (Schneiderman, Ketchel and Williams, 1953) and will be con-
sidered in further detail at a later occasion, for our present purposes the phenomenon
is doubly significant since it excludes tyrosinase as the critical target of carbon
monoxide. This conclusion is further substantiated by the failure of phenylthiourea
or any of a number of other potent anti-tyrosinases to duplicate the effects of
carbon monoxide or of cyanide (Schneiderman and Williams, 1953a). We are
therefore persuaded that the actions of carbon monoxide on Cecropia are due to
its combination with cytochrome oxidase.

The factors which condition the quantitative effects of carbon monoxide on res-
piration mediated by the cytochrome oxidase system are four in number : ( 1 ) the
relative affinity of the insect's cytochrome oxidase for carbon monoxide and for
oxygen; (2) the carbon monoxide/oxygen ratio established at the site of enzyme
action; (3) the degree to which cytochrome oxidase limits the transfer of electrons
from substrate to oxygen; and (4) the oxidation of carbon monoxide to carbon
dioxide. We shall briefly consider each of these points as it pertains to the present
study.

Detailed measurements of the relative affinity of cytochrome oxidase for carbon
monoxide and oxygen are available only in the case of yeast (Warburg, 1949) and
mammalian heart muscle (Ball et al., 1951). A 17:1 carbon monoxide/oxygen
ratio inhibits the cytochrome-catalyzed respiration of yeast 75 per cent and the cyto-
chrome oxidase activity of heart muscle 64 per cent. Since the relative affinities
are so similar for such dissimilar cell types, it is a reasonable presumption that the
insect cytochrome oxidase does not differ greatly. This probability has been con-
firmed by the finding that a carbon monoxide/oxygen ratio of 16: 1 causes a light-
reversible inhibition of the cytochrome oxidase activity of homogenates of the tho-
racic muscles of Cecropia moths by approximately 50 per cent (Pappenheimer and
Schneiderman, unpublished observations).

In the positive pressure experiments on Cecropia a ratio of 25 : 1 was routinely
established in the air surrounding the insect. The utilization of oxygen in the res-
pirometer gradually lowered the oxygen tension over a period of 30 hours from 21
per cent to as low as 14 per cent, and thus increased the carbon monoxide/oxygen
ratio. And, in each instance, the utilization of oxygen in the tissues decreased the
internal oxygen tension still further. Consequently, the recorded ratio of 25 : 1 is

METABOLISM OF SILKWORM

a minimal estimate of the effective carbon monoxide/oxygen ratio that was actually
established in the insect's tissues at the site of enzyme action. For these several
reasons we conclude that carbon monoxide in the positive pressure experiments ef-
fectively inhibited a large proportion of the insect's cytochrome oxidase — probably
not far short of 100 per cent.

As mentioned above, one might anticipate that an excess of cytochrome oxidase
relative to cytochrome c and other electron donors would tend to camouflage the
participation of cytochrome oxidase in the metabolism of diapause. However, it is
worth recalling that carbon monoxide combines exclusively with reduced cytochrome
oxidase ; that is, with functional oxidase receiving electrons from cytochrome c. An
excess of the oxidase would necessarily be present in the oxidized form and there-
fore incapable of combining with carbon monoxide.

Finally, there is circumstantial evidence that the oxidation of carbon monoxide
to carbon dioxide was not a complicating factor in the present study. If such an
oxidation occurred, as described in the case of frog muscle by Fenn and Cobb (1932)
and Stannard (1940), the oxidation of each molecule of carbon monoxide would be
recorded manometrically as ly* molecules of oxygen consumed and one molecule
of carbon dioxide produced. The theoretical R.Q. of this process is 0.66, and in
the case of the Cecropia pupa would thus tend to decrease the normal R.Q. of 0.78.
However, since as recorded in Figures 2 and 3, a slight increase in R.Q. was actu-
ally observed in the presence of carbon monoxide, the oxidation of carbon monoxide
was not a serious complication in the present experiments.

Thus, in summary, the conclusion seems acceptable that metabolism insensitive
to high ratios of carbon monoxide /oxygen signals the utilization of terminal oxidases
other than cytochrome oxidase.

3. Significance of cyanide-insensitive respiration

Cyanide is a far less specific inhibitor of cytochrome oxidase than is carbon
monoxide. It inhibits not only cytochrome oxidase, but also catalase, peroxidase,
and tyrosinase, and, as previously mentioned, at concentrations higher than about
10~3 M, cyanide also combines with various substrates, metabolic intermediates, and
enzymes possessing carbonyl groups. For this reason cyanide-sensitivity is far less
significant than cyanide-insensitivity. Cyanide-insensitivity strongly suggests that
neither cytochrome oxidase nor any of a number of other enzymes are prerequisite
for the reaction in question.

4. Respiration during diapause and development

As judged by its insensitivity to cyanide and carbon monoxide, virtually all of
the metabolism of the diapausing pupa appears to proceed via pathways independent
of cytochrome oxidase, save for the metabolic events responsible for the contraction
of the abdominal muscles. It is therefore of particular interest and importance that
the intersegmental muscle of the abdomen is the only major tissue within the dia-
pausing pupa containing a high concentration of the classical cytochrome system
(Williams, 1951 ; Pappenheimer and Williams, 1952).

The termination of diapause and the onset of adult development, however, usher
in a new situation in which a progressively larger fraction of metabolism becomes
dependent on the presence of a functional cytochrome oxidase system. Evidently,

226 H. A. SCHXKIDKRMAX AX1) C. M. WILLIAMS

at this time, a carbon monoxide- and cyanide-sensitive respiration mediated by cyto-
chrome oxidase is superimposed on the carbon monoxide- and cyanide-stable me-
tabolism of diapause.

Analogous changes in the relative activities of carbon monoxide-sensitive and
carbon monoxide-stable respiratory systems have been demonstrated in a variety of
plants, animals, and micro-organisms (Bodine and Boell, 1934a, 19341); Wolsky,
1943, 1949; Paul, 1951 ). In the case of Cecropia the shift to cytochronie oxidase-
mediated respiration is synchronized with the action of the insect's "growth and
differentiation hormone" in terminating the pupal diapause — a correlation which
suggests that the change in metabolism, in itself, is a part of the biochemical action
of the hormone.

5. Effects of oxygen tension on the respiration of brainless diapausing pupae

The experimental results demonstrate that the metabolism of diapausing Ce-
cropia pupae becomes independent of oxygen tension when the latter is five per cent
of an atmosphere or higher. The tension of oxygen is usually considered to limit
respiration at the cellular level only when it approximates zero (Krogh, 1916; Op-
penheimer, 1925 ) ; this consideration is valid in most organisms since cytochrome
oxidase, the usual terminal oxidase, is saturated by oxygen at tensions ranging from
0.25 to 2.5 mm. Hg (Winzler, 1941). However, flavoproteins when functioning
as terminal oxidases are ordinarily thought to be unsaturated at low oxygen tensions.
Thus, the "old yellow enzyme" which Warburg and Christian (1932) isolated from
yeast was markedly influenced by variations in oxygen tension : in 100 per cent of
an atmosphere of oxygen the respiration which it mediated was nearly five times
that in 5 per cent oxygen. A corresponding dependency on oxygen tension has also
been observed for flavoprotein-mediated respiration in vivo. Thus, in thin sections
of the Arum spadix (James and Beevers, 1950), flavin-catalyzed oxygen uptake
increased progressively as the oxygen tension was raised to one atmosphere.

The fact that the respiration of diapausing pupae is independent of oxygen at
tensions greater than 5 per cent of an atmosphere, and the further fact that one per
cent oxygen sustains 40 per cent of the normal respiration suggest that the carbon
monoxide- and cyanide-stable oxidase of Cecropia differs from the flavoproteins
reported in plants and bacteria and studied in vitro. However, we cannot exclude
the possibility that such an oxidase may be present in relative excess in Cecropia
and that it may fail to limit electron transmission even when driven slowly at low
oxygen tensions.

6. Identification of the terminal o.vidascs mediating respiration in diapausing pupae

The results of the present study permit the following characterization of the ter-
minal oxidases in diapausing Cecropia pupae. The principal terminal oxidase in
the intersegmental muscles of the abdomen is cytochrome oxidase ; in other major
tissues of the diapausing insect it is not cytochrome oxidase. The latter unknown
oxidase is uninhibited by high concentrations of carbon monoxide (carbon monoxide
/oxygen ratios of 25: 1), or by cyanide concentrations up to 10'* M, or by phenyl-
thiourea. Moreover, the oxidase in question is saturated by oxygen tensions of
5 per cent of an atmosphere or less.

METABOLISM OF SILKWORM

On the basis of these several lines of evidence, the most probable candidates
appear to be either an autoxidizable flavoprotein transferring electrons from reduced
pyridine nucleotides to molecular oxygen, or an autoxidizable heme-containing
enzyme which fails to combine with either cyanide or carbon monoxide.

SUMMARY

1. The respiration of the Cecropia silkworm was studied after the injection
of cyanide or in the presence of specific mixtures of oxygen, nitrogen, and carbon
monoxide. Positive pressure techniques were utilized to test the effects of carbon
monoxide/oxygen ratios as high as 25 : 1 .

2. It was found that the respiration of the diapausing pupa is only slightly
affected by high concentrations of carbon monoxide or cyanide. This minor effect
was accounted for in terms of the cyanide- and carbon monoxide-sensitivity of the
contraction of the intersegmental muscles of the pupal abdomen. The other tissues
in the dormant insect showed no detectable inhibition by high concentrations of
cyanide or carbon monoxide.

3. The termination of the pupal diapause and the progress of adult development
are accompanied by a marked increase in the insect's sensitivity to cyanide and
carbon monoxide. The effects of these agents are then no longer limited to
muscular tissue but extend to the insect as a whole. Cyanide or carbon monoxide
appear to act exclusively on the extra metabolism accompanying development and,
thereby, to reduce the overall metabolism to the old diapausing level.

4. The modes of action of cyanide and carbon monoxide within the diapaus-
ing and non-diapausing insects are considered in detail. Insensitivity to these
agents, as in most tissues of the diapausing pupa, argues in favor of the presence
and utilization of a terminal oxidase other than cytochrome oxidase.

5. It is concluded that cytochrome oxidase is the principal terminal oxidase
of only the somatic musculature of the diapausing pupa. Months later, with the
termination of the pupal diapause, cytochrome oxidase becomes the principal
terminal oxidase of the growing, post-diapausing insect as a whole.

6. These qualitative changes in the insect's metabolism are synchronized with
the secretion of the hormone responsible for the termination of diapause and the
development which follows, and appear to be a more or less immediate result
of the hormonal action.

LITERATURE CITED

BALL, E., C. F. STRITTMATTER AND O. COOPER, 1951. The reaction of cytochrome oxidase with

carbon monoxide. /. Biol. Chcm., 193 : 635-647.
BEHNKE, A. B., 1940. High atmospheric pressures ; physiological effects of increased and

decreased pressure; application of these findings to clinical medicine. Ann. Int. Mcd.,

13: 2217-2228.
BODINE, J. H., AND E. J. BOELL, 1934a. Carbon monoxide and respiration : action of carbon

monoxide on the respiration of normal and blocked embryonic cells (Orthoptera).

/. Cell. Comp. Physiol., 4: 475-482.
BODINE, J. H., AND E. J. BOELL, 1934b. Respiratory mechanisms of normally developing and

blocked embryonic cells (Orthoptera). /. Cell. Coinp. Physiol., 5: 97-113.
CASE, E. M., AND J. B. S. HALDANE, 1941. Human physiology under high pressure. I. Effects

of nitrogen, carbon dioxide and cold. /. Hygiene, 41 : 225-249.

CHADWICK, L., AND C. M. WILLIAMS, 1949. The effects of atmospheric pressure and com-
position on the flight of Drosophila. Biol. Bull., 97: 115-137.

228 H. A. SCHNEIDERMAN AND C. M. WILLIAMS

FENN, W. O., AND D. M. COBB, 1932. The burning of carbon monoxide by heart and skeletal

muscle. Amer. J. Physiol., 102 : 393-401.
GODDARD, D. R., AND J. D. MEEusE, 1950. Respiration of higher plants. Ann. Rev. Plant

Physio/., 1 : 207-232.
JAMES, W. <)., AND H. BEEVERS, 1950. The respiration of Arum spadix. A rapid respiration

resistant to cyanide. New Phytologist, 49: 353-374.

KROGH, A., 1916. The respiratory exchange of animals and man. Monogr. Biochem., Long-
mans, Green and Co., London.
KUBOWITZ, F., 1937. liber die chemische Zusammensetzung der Kartoffeloxydase. Biochem.

Zeitschr., 292 : 221-229.
KUBOWITZ, F., 1938. Spaltung und Resynthese der Polyphenoloxydase und des Hamocyanins.

Biochem. Zeitschr., 299: 32-57.

LARDY, H. A., 1949. Respiratory enzymes. Burgess Publ. Co., Minneapolis.
LAWRENCE, J. H., W. F. LOOMIS, C. A. TOBIAS AND E. H. TURPIN, 1946. Preliminary observa-
tions on the narcotic effect of xenon with a review of values for solubilities of gases in

water and oils. /. Physiol., 105 : 197-204.

OPPENHEIMER, C., 1925. Die Fermente, 2 : 1402. Georg Thieme, Leipzig.
PAPPENHEIMER, A. M., AND C. M. WILLIAMS, 1952. The effects of diphtheria toxin on the

Cecropia silkworm. /. Gen. Physiol., 35 : 727-740.
PAUL, K. G., 1951. Iron-containing enzymes — A. Cytochromes. The enzymes. Vol. II, part 1.

Academic Press, New York.

PUNT, A., 1950. The respiration of insects. Physiol. Comp. et Occol., 2 : 59-74.
ROBBIE, W. O., 1946. The quantitative control of cyanide in manometric experimentation.

/. Cell. Comp. Physiol., 27: 181-209.
SANBORN, R. C., AND C. M. WILLIAMS, 1950. The cytochrome system in the Cecropia silkworm

with special reference to the properties of a new component. /. Gen. Physiol., 33:

579-588.
SCHNEIDERMAN, H. A., AND N. FEDER, 1954. A respirometer for metabolic studies at high

gaseous pressures. Biol. Bull., 106: 230-237.
SCHNEIDERMAN, H. A., M. KETCHEL AND C. M. WILLIAMS, 1953. Physiology of insect diapause.

VI. Effects of temperature, oxygen tension, and metabolic inhibitors on in vitro sperma-

togenesis in the Cecropia silkworm. Biol. Bull., 105: 188-199.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1952. The terminal oxidases in diapausing and

. non-diapausing insects. Anal. Rec., 113: 55-56.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1953a. The physiology of insect diapause. VII.

The respiratory metabolism of the Cecropia silkworm during diapause and development.

Biol. Bull., 105 : 320-334.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1953b. Discontinuous carbon dioxide output by

diapausing pupae of the giant silkworm, Platysamia cccropia. Biol. Bull., 105: 382.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1953c. Metabolic effects of localized injury to

the integument of the Cecropia silkworm. Anat. Rec., 117 : 640-641.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1954. The 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.
SCHOLANDER, P. F., AND F. J. W. RouGHTON, 1943. Microgasometric estimation of the blood

gases. I. Oxygen. /. Biol. Cheat.. 148: 541-550.
STANNARD, J. N., 1940. An analysis of the effect of carbon monoxide on frog skeletal muscle.

Amcr. J. Physiol., 129: 195-213.

SUSSMAN, A. S., 1949. The functions of tyrosinase in insects. Quart. Rev. Biol., 24 : 328-341.
SUSSMAN, A. S., 1952. Tyrosinase and the respiration of pupae of Plat \samia cccropia L.

Biol. Bull., 102: 39-47.
THEORELL, H., 1951. The iron containing enzymes. B. Catalases and peroxidases. "Hydro-

peroxidases." The enzymes. Vol. II, part 1. Academic Press, New York.
WARBURG, O., 1949. Heavy metal prosthetic groups and enzyme action. Oxford at the

Clarendon Press.
WARBURG, O., AND W. CHRISTIAN, 1932. Uber ein neues Oxydationsferment und sein Absorp-

tionsspektrum. Biochem. Zeitschr., 254 : 438-458.
WARBURG, O., AND E. NEGELEIN, 1928. Uber den Einfluss der Wellenlange auf die Verteilung

des Atmungsferments. Biochem. Zeitschr., 193 : 339-346.

METABOLISM OF SILKWORM 229

WILLIAMS, C. M., 1946. Physiology of insect diapause: the role of the brain in the production

and termination of pupal dormancy in the giant silkworm, Platysamia cccropia. Biol.

Bull., 90: 234-243.
WILLIAMS, C. M., 1948. Extrinsic control of morphogenesis as illustrated in the metamorphosis

of insects. Grozvth, 12 : 61-74.
WILLIAM, C. M., 1951. Biochemical mechanisms in insect growth and metamorphosis. Fed.

Proc.. 10: 546-552.
W'ILLIAMS, C. M., AND H. K. BEECHER, 1944. Sensitivity of Drosophila to poisoning by oxygen.

A mcr. J. Physio/.. 140: 566-573.
WIXZLER, R. J., 1941. The respiration of hakers yeast at low oxygen tension. /. Cell. Comp.

Physiol., 17: 263-276.
WOLSKY, A.. 1943. Respiratory mechanism of eggs of the silkworm moth. III. Effect of

treatment with HC1 on respiration. Maqyar Biol. Kutato intczct Munkai, 14: 445-464.

(Chcm. Abstracts, 1947.)
WOLSKY, A., 1949. The effect of carbon monoxide on the respiration of artificially bivoltinized

silkworm eggs. Current Science (India), 18: 323-325.

A RESPIROMETER FOR METABOLIC STUDIES
AT HIGH GASEOUS PRESSURES l

HOWARD A. SCHNEIDERMAN 2 AND NED FEDER *
The Biological Laboratories, Harvard University, Cambridge 38, Massachusetts

The use of carbon monoxide as a specific inhibitor of cytochrome oxidase is
accompanied by one serious complication — the inhibition is reversed by oxygen
(Warburg, 1949). The degree to which one can inhibit the enzyme is therefore
dependent, not alone on the carbon monoxide pressure, but also on the pressure
of oxygen that is simultaneously present. In short, the inhibition of cytochrome
oxidase is a function of the carbon monoxide/oxygen ratio. To achieve an effec-
tive inhibition of the enzyme, this ratio must be high. Fifty per cent inhibition
requires approximately a 10 to 1 ratio ; 75 per cent inhibition, a 20 to 1 ratio. For
greater degrees of inhibition, still higher ratios are necessary (Warburg, 1949;
Ball eiaL, 1951).

To achieve this goal it has been customary to use an atmosphere containing 95
per cent carbon monoxide and 5 per cent oxygen ; indeed, an atmosphere of 98 per
cent carbon monoxide and 2 per cent oxygen has occasionally been employed.
However, oxygen at these low pressures fails to satisfy the normal respiratory
requirements of most plants and animals in vivo or in vitro (Tang, 1933). Experi-
mental results are thereby complicated by anoxia, and there is uncertainty as to
whether an observed effect is due to the presence of carbon monoxide or a!
deficiency in oxygen. The earth's atmospheric pressure (760 mm. Hg) is too
low to permit one to inhibit cytochrome oxidase effectively by substituting carbon
monoxide for the nitrogen in air. If the oxygen tension is to be maintained at its
normal value (/f,th atmosphere ), then several atmospheres of carbon monoxide must
be superimposed.

For this reason there has long been a need for a simple and practical method for
measuring oxygen consumption and carbon dioxide production at high gaseous
pressures. This objective seems doubly attractive in view of the insensitivity of
most biological preparations to pressure f>cr sc, as long as the latter is not extremely
high.

Two methods have been described for this purpose ; namely, that of Libbrecht
and Massart (1937) and that of Stadie and Riggs (1944). Both of these meth-
ods utilized a pressure chamber containing a manometric apparatus of the War-
burg type. The chamber designed by Stadie and Riggs had a capacity of 60
liters and enclosed 6 Warburg manometers and vessels. The apparatus was con-

1 This study was aided by grants to Professor C. M. Williams from the Lalor Foundation
and the U. S. Public Health Service, and by an Institutional Grant to Harvard University from
the American Cancer Society.

2 Former Atomic Energy Commission Fellow. Present Address : Department of Zoology,
Cornell University, Ithaca, N. Y.

3 Present address : Philadelphia General Hospital, Philadelphia, Pa.

230

HIGH PRESSURE RESPIROMETER 231

structed so that the necessary manipulations and readings were made from the out-
side, while high pressure and constant temperature were maintained inside. As
might lie anticipated, such an apparatus was extremely costly, presented an explo-
sive hazard, and required careful checks to police a dozen separate gaskets and
fittings.

The present paper describes a simple and practical technique for measuring gas
exchange at positive pressures up to seven atmospheres. The respirometer itself
is inexpensive, safe, leak-proof, and yields results of the same degree of accuracy as
conventional manometric and volumetric techniques. In its present form it is suit-
able for studies of intact animals, plants, and tissues where agitation is not required.
However, the principle of the technique is readily adaptable to studies of solutions,
slices, and homogenates. and also to studies at pressures below atmospheric.

PRINCIPLE OF METHOD

A small glass respiration chamber, containing the experimental animal plus a
carbon dioxide absorbant, is joined to a graduated capillary tube. Provisions are
made so that the capillary can subsequently be sealed with a fluid index drop. After
assembly, the respiration chamber is enclosed in a large polymethyl methacrylate
( Lucite ) chamber capable of withstanding high internal gas pressure. At the start
of the experiment the capillary tube is patent ; the lumen of the capillary therefore
affords a direct connection between the gas space of the respiration chamber and
the Lucite compensation chamber. Consequently, when the latter is filled with gas
to a desired pressure, gas passes through the capillary and fills the respiration cham-
ber at the same pressure. When a desired pressure is attained, an index drop is
tipped into the graduated capillary, thus sealing the respiration chamber. The
measurements are then performed in the same manner as in an ordinary volumeter
of the Fenn type (Fenn, 1935 ). Since the plastic compensation chamber is closed
off from the outside air, excursions of the index drop are independent of changes
in atmospheric pressure (cf. Gerard and Hartline, 1934).

APPARATUS (see Fig. 1)

a. Compensation chamber. The plastic compensation chamber is a transparent
Lucite cylinder fitted with brass endplates, gaskets, and needle valves. Figure 2
shows the chamber and its component parts. The Lucite cylinder is 4" I. D.
X 4.50" O. D. >: 18" long. The endplates are 5" X 5" X 0.5"" brass plates with a
0.25" deep circular channel milled on the inner surface to receive the Lucite cylinder.
A rubber gasket is inserted into this channel. Half-inch Hoke needle valves are
threaded and silver brazed in the center of each endplate. The two endplates are
held together by four brass rods, 0.5" in diameter. Endplate A is bolted to the
rods, while endplate B is removable. On endplate A the rods extend 0.5" beyond
the reducing valve so that the tank may be placed on end. The internal volume of
the assembled compensation chamber was 3460 cc.

b. Capillary volumeter. The size of the respiration chambers and capillaries is
dictated by the dimensions of the experimental animal, the rate of oxygen consump-
tion, and the desired sensitivity. The size most frequently used in this laboratory

232

HOWARD A. SCHNEIDERMAN AND NED FEDER

B

FIGURE 1. High pressure respirometer : an animal-containing capillary •volumeter, a reference
volumeter, and a capillary barometer are mounted on the capillary volumeter frame.

FIGURE 2. A 3460-cc. Lucite compensation chamber and its component parts.

HIGH PRESSURE RESPIROMETER

is shown in Figure 3.4 It consists of a 20-cni. length of 2 mm. bore capillary tubing,
calibrated in 0.005 cc. units, and fitted by a 20/40 standard taper joint to a 30-cc.
Pyrex shell vial, 85 mm. long. To the tip of the capillary is fused a 3.5 cm. length
of 7-mm. bore Pyrex tubing. The latter is slightly constricted at its distal open
end and serves as a reservoir for the index drop solution. The calibrated volume
of each capillary is 550 mm.3 The total volume of each volumeter to the tip of the
capillary is 45 cc.

FIGURE 3. A 45-cc. capillary volume'.er.

The respiration chamber can be attached to a standard Warburg manometer by
means of an adapter previously described (Schneiderman and Williams, 1953).
Thus the respiration of the animal in air before and after exposure to high pressure
may be conveniently measured by conventional methods without removal of the ani-
mal from the respiration chamber.

c. Capillary barometer. Measurements of absolute pressure, accurate to within
one per cent, are required for the proper determination of gas exchange in the
present apparatus (see discussion of calculations below). Since the standard Bour-
don type pressure gauges are subject to errors in excess of 5 per cent, a capillary
barometer is utilized. Each such barometer consists of a one-cc. pipette of approxi-
mately 2 mm. internal diameter, graduated in 0.01 cc., and sealed at one end. By
means of a long hypodermic needle a drop of colored detergent solution is placed
in the closed end of the capillary and another drop at the beginning of the gradu-
ations. In each experiment three capillary barometers of this type are enclosed in
the Lucite compensation chamber. The pressure in the closed system is calculated
by application of the gas law from readings of the capillary barometer taken before
and after compression and from a reading of the local barometric pressure.

d. Water bath. Glass aquaria make the most satisfactory water baths since the
glass sides facilitate reading the capillaries.

e. Reagents. ( 1 ) The index drop solution has the following formula : 1 part
"Aquet" (detergent of Emil Greiner Co.), 500 parts distilled water, a few drops of
dilute H2SO4 to prevent carbon dioxide absorption by the index drop, and a few
crystals of acid fuchsin to give the solution a red color. This fluid flows easily,
keeps almost indefinitely at room temperature, and forms an index drop which re-
sponds regularly to slight pressure changes. (2) The carbon dioxide absorbant
is carbonate-free 1 N KOH. (3 ) The grease used on the ground-glass joints con-
necting the respiration chambers to the capillaries and on the gaskets of the brass
endplates is Dow-Corning silicone stopcock grease. Conventional organic greases

4 The assistance of Dr. Conrad Yocum in the design of the final capillary volumeter is
gratefully acknowledged.

234 HOWARD A. SCHNEIDERMAN AND NED FEDER

have a tendency to oxidize or react in other ways with oxygen and carbon monoxide
under pressure.

PROCEDURE

A roll of filter paper is deposited in the bottom of each respiration chamber and
moistened with 0.5 cc. of 1 N KOH. A small paraffin-coated tube is placed in the
chamber to support the experimental animal. The latter is inserted and the ground-
glass joint in the respiration chamber plugged with the capillary tube.

Four animal-containing respiration chambers, two reference volumeters not
containing animals, and three capillary barometers are mounted in a plywood frame
( Fig. 1 ) and held in place with rubber bands. The assembly is then placed in a
horizontal position in the compensation chamber with the base of the frame flush
against the brass endplate A. By means of a hypodermic syringe 0.05 cc. of the
index drop solution is introduced into each of the index drop reservoirs, care being
taken not to occlude the capillaries themselves. With valves A and B open, brass
endplate B is now bolted on. One then records the temperature of the room, the
position of the index drop in the capillary barometers, and the atmospheric baro-
metric pressure.

The experimental gas is supplied from standard cylinders through a manually
controlled reduction valve connected by a flexible 0.25" bore copper tubing to valve
A. About 10 liters of experimental gas are flushed slowly through the compensa-
tion chamber.5 Valve B is closed and the experimental gas introduced under pres-
sure through valve A to approximately the desired pressure, as indicated by the
capillary barometers. Valve A is then closed.

The compensation chamber is now tilted to a vertical position so that the index
drops flow into the lumina of the calibrated capillaries. Valve A is opened slightly
and sufficient gas introduced under pressure to force a drop of index fluid a few
centimeters into each of the six capillary tubes. Valve A is then closed and the
compensation chamber returned to a horizontal position. Valve A is reopened and
gas under pressure is slowly admitted until the drops have traversed the lengths
of the capillary tubes to the proximal end of the calibrations. Valve A is then
closed, the flexible coupling disconnected, and the compensation chamber immersed
in a water bath controlled to ± 0.01° C.

Valve A is now opened carefully until a barely detectable outward movement of
the index drops is observed. The rate of gas escape is adjusted so that 10 to 20
minutes are required for the drops to migrate to the distal end of the capillaries.
Valve A is finally closed when the distal ends of the index drops are about 2 centi-
meters from the distal ends of the calibrations. By running the drops up and down
in this manner, one wets the walls of the measuring capillaries and thereby assures
both a sensitive response and a constant size in the index drops.

Temperature equilibration requires 80 to 100 minutes. After equilibration, the

•' If this flushing procedure is not carried out, then, upon compression of the compensation
chamber with the experimental gas mixture, some of the air in the compensation chamber will
be forced into the respiration chambers along with the experimental gas. This introduces con-
siderable error ; for example, if the air-filled compensation chamber is compressed with 5
atmospheres of carbon monoxide without prior flushing, the carbon monoxide/oxygen ratio in
the compensation chamber will be approximately 25 : 1 while the carbon monoxide/oxygen ratio
in the vt.lnmcters will be less than 6: 1.

HIGH PRESSURE RESPIROMETER

bath's temperature and the position of the drops in the capillary barometers are
recorded to determine the absolute pressure in the compensation chamber. The
two barometric pressures that agree most closely are averaged. Readings of the
position of the index drop in each calibrated capillary also begin at this time. The
positions of the drops are recorded at intervals ranging from 30 minutes to 6 hours,
as dictated by the rate of oxygen consumption and the duration of the experiment.
Thermobarometric corrections are applied to each reading, taking into account the
fact that the actual volume of gas in the reference volumeters is slightly larger than
that in the animal-containing chambers. Calculation of thermobarometric correc-
tions may be simplified by enclosing in each thermobarometer a glass rod of approxi-
mately the same volume as the animal in the experimental chamber.

To calculate the oxygen consumption in mm.3 at S. T. P. from the excursion of
the index drop, the following calculations are employed :

Let: v - volume in mm.:i of capillary that the index drop traversed.

P - absolute pressure in atmospheres after compression.

V r — gas volume of respiration chamber in cc. (I.e., 45 cc. minus volume
of organism and reagents).

Vc — gas volume of compensation chamber in cc. (i.e., 3460 cc. minus vol-
ume of 6 respiration chambers and frame — ca. 3000 cc. i.

T = Temperature of bath.

A calibration factor F is calculated at each pressure and temperature to convert
the measurements of v to mm.3 oxygen. Thus :

Fv = mm.3 oxygen consumed at S. T. P.
Fenn (1935) has shown that the value of F is provided by the formula

273

T '

Under ordinary experimental conditions the quantity in brackets is very nearly 1.01.
Thus

276
mm." oxygen consumed " rv -- r X ^^ X r.

Corrections for the vapor pressure of water and for the solubility of oxygen in
the insect and the reagents were not applied since the combined errors thereby intro-
duced were less than one per cent.

At the conclusion of the experiment, the compensation chamber is slowly decom-
pressed and unbolted and the animals removed from the respiration chambers. By
the addition of acid the total carbon dioxide produced in each respiration chamber
during the experiment is displaced from the alkali and measured volumetrically in
the gas analyzer described by Bliss ( 1953 ) or manometrically by coupling the res-
piration chamber to a standard Warburg manometer. The average carbon dioxide
output may then be calculated.

236

HOWARD A. SCHNEIDERMAN AND NED FEDER

ACCURACY OF METHOD AND RANGE OF APPLICATION

The calibrated capillaries can be read to ±0.1 division. This and the total
capillary excursion ( 110 divisions) establish the theoretical limits of accuracy. The
actual limits are, of course, determined in large measure by the degree of agreement
between the thermobarometers. Table I records the results of a series of readings
on two sets of thermobarometers in two typical sets of experimental conditions.
The maximum standard deviation of ± 0.16 corresponds to an error of about 0.1
division in reading the positions of the index drops, in close agreement with the
theoretical limits. Therefore any individual reading corrected for thermobarometric
change is accurate to within ± 0.2 division. A capillary excursion of 30 divisions
would thus be accurate to ± one per cent.

TABLE I

Typical series of thermobarometric readings

Time reading
taken (hours)

Difference between initial and subsequent thermobarometric readings

Ti

T2

T3

*

T5

Standard
deviation

Series 1

Five thermobarometers each containing 1 ml. 1 N KOH compressed
with 5 atmospheres of nitrogen

2.00

.8

.8

.8

.9

.7

.07

4.00

1.4

1.4

1.4

1.6

1.4

.08

4.70

1.7

1.8

1.6

1.7

1.6

.07

16.90

2.6

2.8

2.6

2.7

2.4

.16

18.65

2.9

3.0

2.9

3.0

2.8

.07

Series 2

Two thermobarometers each containing 1 ml. 1 N KOH compressed
with 5 atmospheres of carbon monoxide

1.02

.7

.7

0

2.12

1.1

1.3

.15

7.50

2.2

2.3

.1

20.75

1.7

1.5

.15

The experimental method has been applied without difficulty to studies of or-
ganisms and tissues having oxygen uptakes between 20 and 1000 mm.3/ hour, and by
the use of high pressures of carbon monoxide the role of cytochrome oxidase has
been studied in both animals (Schneiderman and Williams, 1954), and plants
(Hackett, Yocum and Thimann, personal communication).

We wish to express our sincere appreciation to Professor Carroll M. Williams
in whose laboratory these experiments were performed.

HIGH PRESSURE RESPIROMETER 237

SUMMARY

A simple and practical apparatus is described for the measurement of oxygen
consumption and carbon dioxide production at positive pressures up to seven atmos-
pheres. It consists, essentially, of a series of capillary respirometers enclosed in a
large Lucite compensation chamber capable of withstanding a positive pressure.
The details of the construction and operation of the apparatus and the accuracy and
range of application are considered.

LITERATURE CITED

BALL, E. F., C. F. STRITTMATTER AND O. COOPER, 1951. The reaction of cytochrome oxidase

with carbon monoxide. /. Biol. Chcm., 193 : 635-647.
BLISS, D. E., 1953. Endocrine control of metabolism in the land crab, Gccarcimis lateralis.

I. Differences in the respiratory metabolism of sinusglandless and eyestalkless crabs.

Biol. Bull., 104: 275-296.
FENN, W. O., 1935. The differential volumeter for the measurement of cell respiration and

other processes. American Instrument Company, Inc. publication.
GERARD, R. W., AND H. M. HARTLINE, 1934. Respiration due to natural nerve impulses. A

method for measuring respiration. /. Cell. Comp. Physiol., 4: 141-160.
LIBBRECHT, W., AND L. MASSART, 1937. Compt. rend. Soc. Biol., 194: 299. (Cited in Stadie

and Riggs, 1944.)
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1953. The physiology of insect diapause. VII.

The respiratory metabolism of the Cecropia silkworm during diapause and development.

Biol. Bull., 105: 320-334.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1954. The physiology of insect diapause. VIII.

Qualitative changes in the metabolism of the Cecropia silkworm during diapause and

development. Biol Bull., 106: 210-229.
STADIE, W. C., AND B. C. RIGGS, 1944. An apparatus for the determination of the gaseous

metabolism of surviving tissues in vitro at high pressures of oxygen. /. Biol. Chcm.,

154: 669-686.
TANG, P. S., 1933. On the rate of oxygen consumption by tissues and lower organisms as a

function of oxygen tension. Quart. Rev. Biol., 8 : 260-274.

WARBURG, O., 1949. Heavy metal prosthetic groups and enzyme action. Oxford at the Claren-
don Press.

THE PHYSIOLOGY OF INSECT DIAPAUSE. IX. THE CYTO-

CHROME OXIDASE SYSTEM IN RELATION TO THE

DIAPAUSE AND DEVELOPMENT OF THE

CECROPIA SILKWORM 1

HOWARD A. SCHNEIDERMAN - AND CARROLL M. WILLIAMS
1'hc Biological Laboratories, Harvard University, Cambridge, Massachusetts

During the pupal diapause the respiratory metabolism of the Cecropia silkworm
proceeds at a low and relatively constant rate which, except in the case of the inter-
segmental muscles of the abdomen, is insensitive to carbon monoxide and cyanide.
However, with the termination of diapause and the initiation of adult development,
a carbon monoxide- and cyanide-sensitive respiration appears and increases pro-
gressively, being superimposed on the carbon monoxide-stable respiration of dia-
pause. It was concluded from these and other observations that the metabolism of
the developing insect is largely mediated by the cytochrome oxidase system while
that of the diapausing pupa is not ( Schneiderman and Williams, 1954).

But respiratory measurements in themselves can provide only circumstantial evi-
dence that the coupling of metabolism to cytochrome function is causally related to
the termination of diapause and the development which follows. The problem is
basically morphogenetic in character and therefore demands solution in morpho-
logical terms. Is the change in terminal oxidase coincidental, or is there an obliga-
tory coupling between the function of the cytochrome oxidase system and the actual
development of the insect? The present study was designed to answer this ques-
tion by direct observations of the effects of carbon monoxide on the growth of the
Cecropia silkworm during successive stages of metamorphosis.

MATERIALS AND METHODS
1. Experimental animals

Experiments were performed on embryos, mature larvae, pupae, developing
adults, and adults of the giant silkworm Platysainia cecropia. The pupae were of
three types: (a) Normal diapausing pupae removed from their cocoons and stored
continuously at 25° C. ("unchilled diapausing pupae"), (b) Diapausing pupae
such as the preceding, except that the brains had been removed and plastic windows
established in the facial region and at the tip of the abdomen ("brainless diapausing
pupae" ). (c) "Previously chilled diapausing pupae" — animals that had been stored
at 5° C. for approximately six months and provided with plastic terminal abdominal
windows. As previously reported (Williams, 1946), prolonged exposure to low

1 This study was aided by the Lalor Foundation, by a grant from the U. S. Public Health
Service, and by an Institutional Grant to Harvard University from the American Cancer Society.

2 Former Atomic Energy Commission Fellow. Present address : Department of Zoology,
Cornell University, Ithaca, New York.

238

CYTOCHROME AND SILKWORM DEVELOPMKXT 239

temperature renders the brain competent to secrete its hormone and results in the
initiation of adult development approximately two weeks after such pupae are re-
turned to 25° C.

In previously chilled pupae provided with plastic windows, the heart-beat and
the initiation and day-to-day progress of adult development could be observed di-
rectly under the dissecting microscope. As has been emphasized in the previous
papers of this series, the visible initiation of adult development is an event of special
significance since it signals the end of the months of pupal diapause. Table 1 re-
cords the time sequence of adult development as observed beneath facial and terminal
abdominal windows at 25° C., from the first visible signs of hypodermal retraction
to the emergence of the adult moth approximately 22 days later. The table records
the average tempo of development of a large series and permits one to estimate the
stage of development to within ± 12 hours in the vast majority of individuals.

TABLE I

Time-table for the development of male chilled Cecropia at 25° C. as witnessed in
pupae equipped zvith facial and abdominal ivindozvs *

Day Characters

0 Initiation of hypodermal retraction just ventral to imaginal disc of genitalia; no retraction
elsewhere.

1 Hypodermal retraction under terminal window extends half way up each side ; the aedeagus
and harpal lobes have tripled in size and migrated slightly toward center of window ; hypo-
dermal retraction under facial window has occurred only along posterior margin and is
restricted to the midline and the lateral angles ; no retraction of leg hypodcrmis.

2 Initiation of retraction of leg hypodermis, harpes show considerable enlargement and
sharply denned outer edges ; beginning of midventral fold between harpes ; the aedeagus
has migrated about half way to center of window.

3 Facial retraction nearly complete ; eye lobes partially visible ; terminal retraction com-
plete except dorsally ; mid-ventral fold of genitalia extends dorsally to aedeagus ; harpes
show considerable molding and beginning of subdivision into upper and lower lobes ; tips
of dorsal harpal lobe slightly forked.

4 Facial and terminal retraction complete ; eye lobes well developed but unpigmented ; fur-
ther subdivision of harpes into dorsal and ventral lobes ; aedeagus has a cone-shaped, trans-
parent, undivided, membranous tip.

5 Palps and "stalks" of antennae visible for first time. Harpes considerably enlarged and
show well developed upper and lower fleshy, semi-transparent lobes ; no pubescence ; no
eye pigment.

6 Membranous tip of aedeagus subdivided into two or three semi-transparent processes ;
harpal lobes with sharp edges ; extremely delicate transparent pubescence along outer edge
of upper harpal lobes ; no pubescence of lower lobes ; no eye pigment.

7 Initiation of pink eye pigment ; transparent pubescence now extends along outer edge of
lower harpal lobes ; genitalia deeply telescoped into preceding segment.

<5 Generalized reddish brown eye pigment ; genitialia fully formed but fleshy and unpig-
mented ; pubescence generally distributed over outer side of all harpal lobes, but longer and
"silky" along edge of upper lobes.

9 Dark reddish brown eye pigment ; long silky hairs on upper harpal lobes and shorter silky-
hairs on lower lobes.

10 Dark brown eye pigment ; long silky hairs on all harpal lobes ; membranous tip of aedeagus
with fleshy spine.

11 No further change. (Continued on next page)

* The same time-table may also be used for the female insect, save for the characteristics
pertaining to the male genitalia. Characters printed in italics are visible without windows and
can be seen by moistening the overlying cuticle with 70 per cent alcohol. The adult genitalia of
Cecropia have been described and figured by Michener (1952).

240 H. A. SCHNEIDERMAN AND C. M. WILLIAMS

TABLE I — Continued

rjay Characters

12 Tan .streak of pigment present on each side of mouth opening; white hairs on upper harpal
lobes and on face; earliest tan pigment on genitalia along surface of gnathos and on ridge
connecting upper and lower harpal lobes on each side.

13 Tarsal claws black; facial cuticle with pale diffuse tan pigmentation; coarse white hairs
on harpes ; tannish pigmentation of triangular plate (annulus) below base of aedeagus, the
pigment extending bilaterally to lower tip of lower harpal lobes ; the latter, in turn, show-
minute black punctate spots ; tip of aedeagus dark brown ; tan pigmentation of upper harpal
lobes ; spine on membranous tip of aedeagus still transparent.

14 Spine on tip of aedeagus black; black, fully-formed antenna! barbs.

Persistency of coarse white hairs.

17 Three black spots along posterior edge of each foreiving; the coarse white hairs on geni-
talia show initiation of pale pink pigmentation.

18 Generalized but incomplete wing pigmentation; red, pink, and white hairs on genitalia;
cuticle "soft" only in region of forewings.

19 Complete wing pigmentation; softening of cuticle extends to dorsum of abdomen.

20 Cuticle "soft" throughout but not crisp; moulting fluid partially absorbed under facial and
abdominal windows.

21 Cuticle crisp throughout; moulting fluid fully resorbed except under abdominal ivindow;
cuticle semi-transparent.

22 Animal distended; adult emergence.

2. Experimental methods

All experiments were performed at 25° C. Three techniques were utilized in
the management of the various gas mixtures :

a. In the flow method one or more insects were enclosed in a glass tube through
which an approximately streamlined and steady flow of a specific gas mixture was
maintained. The mixtures were prepared in pressure cylinders and analyzed prior
to use.

b. In the static pressure method each animal was placed in a shell vial and the
latter loosely plugged with cotton. The vial was then sealed in an individual 2.5-liter
air-filled steel chamber and compressed with a specific gas, the pressure being read
on a gauge calibrated in pounds per square inch. Alternatively, one or more ani-
mals were enclosed in a 3.5-liter air-filled polymethyl methacrylate (Lucite) cham-
ber and compressed with a specific gas. The oxygen tension in the chambers was
that of air (20.9 per cent of an atmosphere), while the pressure of the added gas
was the gauge pressure. After storage at 25° C. for specific periods the chambers
were slowly decompressed, the animals returned to air, and observations continued
over a period of several weeks.

c. In the constant composition pressure method, a series of insects was placed
in a Lucite holder so that their terminal abdominal windows faced uppermost ; the
holder was then enclosed in a 3.5-liter air-filled Lucite chamber (Fig. 1). The
animals were therefore visible through the transparent wall of the chamber and
could be studied under the dissecting microscope. A glass trough containing 10 per
cent NaOH was placed in the chamber for the purpose of absorbing carbon dioxide.
Control experiments revealed that the reaction of carbon monoxide with the concen-
trated alkali to produce formate occurred so slowly that it did not detectably dimin-

CYTOCHROME AND SILKWORM DF.YKI.OPM KXT

241

ish the total carbon monoxide pressure. The chamber also contained a calibrated
capillary barometer for the measurement of absolute pressure (Schneiderman and
Feder, 1954). The air-filled tank was compressed with carbon monoxide, the final
pressure being recorded on the tank gauge and the capillary barometer. The latter
was read, at three-day intervals and the oxygen consumed by the animals replaced
by the addition of a corresponding amount of oxygen. On each such occasion a
sample of gas was removed and analyzed ( Scholander and Roughton, 1943), thus

FIGURE 1. Transparent pressure chamber for studying the effects of high pressures of
carbon monoxide on the day-to-day progress of development. Thirty animals have been equipped
with plastic windows and sealed within the chamber in the presence of five atmospheres of car-
bon monoxide.

242 H. A. SCHXEIDERMAX AND C M. WILLIAMS

giving double assurance that the oxygen tension in the chamber remained within
the desired limits. At the termination of the exposure period the chamber was
slowly decompressed, the animals returned to air, and observations continued.

3. Experimental gases

The compressed gases (oxygen, nitrogen, and carbon monoxide) were handled
as previously described ( Schneiderman and Williams, 1954). In one series of ex-
periments extremely pure carbon monoxide was prepared ( ibid. ) . Since the latter
was indistinguishable from alkali-washed carbon monoxide in its effects on growth,
the less expensive commercially available carbon monoxide was utilized in subse-
quent experiments.

RESULTS

1. Effects of high pressures of nitrogen

Chilled and unchilled diapausing Cecropia pupae and post-diapausing animals at
several stages in adult development were placed in individual air-filled 2.5-liter steel
or Lucite chambers and compressed with from 4 to 7 atmospheres of nitrogen. The
static pressure technique was utilized and each experiment continued for 21 days.
Under this treatment the animals behaved as in air at atmospheric pressure. Spon-
taneous movements of the abdomen and the beating of the heart continued without
interruption. Moreover, the rate of adult development was the same as in air at
one atmosphere, and the resulting adults were normal in all respects.

From these control experiments we learn that pressures up to seven atmospheres
of an inert gas such as nitrogen are without detectable effects on the adult develop-
ment of Cecropia. It is also clear that the 525 cc. of oxygen initially present in each
air-filled chamber was sufficient to permit a pupa to undergo normal adult develop-
ment without interference from oxygen lack or from the accumulation of metabolic
carbon dioxide during the 21 -day period of confinement.

2. Effects of carbon monoxide on diapausing pupae

Diapausing pupae and brainless diapausing pupae were exposed to carbon mo-
noxide by all three of the above-mentioned experimental methods. When the car-
bon monoxide/oxygen ratio was increased above 10: 1, spontaneous movements of
the abdomen showed considerable reduction in both amplitude and frequency. Re-
sidual extremely feeble movements, occasionally detectable even at 15:1 carbon
monoxide/oxygen, completely disappeared after further increase in the ratio. When
decompressed and returned to air, normal abdominal motion reappeared within a
few hours. In contrast to the paralysis of the intersegmental muscles, the heart
continued to beat normally throughout the 21 days of exposure to carbon monoxide
even when the carbon monoxide/oxygen ratio was 25: 1.

It will be recalled that diapausing pupae initiate adult development after continu-
ous storage at 25° C. for five months or longer (Williams, 1946). This behavior
was unimpaired by three weeks of prior exposure to high pressures of carbon monox-
ide. Evidently, within the diapausing insect the viability of neither the pupal tissues,
nor the anlagen of the adult tissues, nor the endocrine organs themselves is dependent
on enzymes inhibited by carbon monoxide.

CYTOCHROME AND SILKWORM DEVELOPMENT 243

^. Inhibition of wound hcalnuj in diapausing pnpac b\ carbon monoxide

Although l>rainless diapausing pupae are incapable of initiating adult develop-
ment (Williams, 1946), they retain the ability to repair integumentary wounds.
One can study this process to good advantage by removing a disc of hypodermis
plus overlying cuticle and covering the wound with a plastic window, the latter being
sealed in place with melted paraffin. Spindle-shaped blood cells promptly adhere
to the window and begin to string out tenuous cytoplasmic processes. The latter
interlace and form a fenestrated tissue which, after 4 or 5 days, is transformed into
a transparent, shiny membrane. Meanwhile, the hypodermis begins to close in
around the margins of the wound, accompanied by minute tracheae and tracheoles.
A continuation of this centripetal growth leads to a central closure of the wound
after about 10 to 14 days.

In order to ascertain the effects of carbon monoxide on wound healing, the fol-
lowing experiment was performed. From a series of six previously chilled diapaus-
ing pupae the brains were removed and facial and abdominal windows established
in each individual. Two days after the operation, three animals were placed in a
transparent 3.5-liter air-filled Lucite tank and compressed via the static pressure
method with five atmospheres of carbon monoxide (carbon monoxide/oxygen ratio
of 25 : 1 ) . The other three animals served as controls and were maintained in air.
Each individual was examined daily under the dissecting microscope for signs of
regeneration. After a total of 13 days the control group in air had completely re-
paired the wounds under both facial and abdominal windows. By contrast, the
experimental group in carbon monoxide showed no evidence of repair. But when
decompressed and returned to air, repair began at once and was completed within
13 days.

Thus, it is clear that even in the diapausing pupa the localized morphogenesis
inherent in the repair of a wound is completely inhibited by carbon monoxide.

4. Inhibition of adult development by carbon monoxide

All three of the techniques for the administration of carbon monoxide were uti-
lized in a study of the adult development of previously chilled pupae and of animals
that had already initiated adult development. The progress of development in each
individual was judged by observations of its genitalia, the day-to-day changes being
compared wdth the normal tempo already defined (Table I). Each experiment was
continued for 21 days.

As recorded in Table II it is of special interest that during exposure to carbon
monoxide the termination of pupal diapause, as signalled by the onset of adult de-
velopment, was blocked or greatly delayed. Moreover, individuals which already
showed early adult development at the outset of the experiment remained alive in
most cases, but further development was either prevented or greatly inhibited.

It is also clear from Table II that the degree of inhibition was a function, not
of the carbon monoxide concentration alone, but of the carbon monoxide/oxygen
ratio. When the latter was higher than 20: 1, development was completely or al-
most completely blocked. Such animals, when returned to air, promptly resumed
normal development where they had left off and produced normal adult moths.
However, when development was incompletely blocked in carbon monoxide/oxygen
ratios less than 20: 1, the insects, upon return to air, continued in a pattern of ab-

244

H. A. SCHNEIDERMAX AND C. M. WILLIAMS

TABLE II

Effects of twenty-one days exposure to various carbon monoxide /oxygen ratios on previously
chilled Cecropia pupae and on animals at specific stages of adult development

CO/0,
ratio

Gas content
of chamber
(atmospheres)

Stage of development
at outset

Num-
ber
of
animals

Num-
ber

of
sur-
vivors

Average
rate of
development
in CO as %
of rate in air

Development after return
to air

33:1

6.7 CO + 1 air

Previously chilled pupae

6

6

(0.5)

6 normal adults

<5-25% development

9

8

1

5 normal adults; 3 died

25:1

5 CO + 1 air

Previously chilled pupae

8

5

(2)

2 normal adults; 3 died

<5-25% development

6

5

2

3 normal adults; 2 died

20:1

4 CO + 1 air

Previously chilled pupae

8

5

2

No data

19:1

0.95 CO +0.05

<5-25% development

6

6

11

3 abnormal adults; 1 normal

02

adult; 2 died

26-50% development

3

3

12

3 slightly abnormal adults

70% development

1

1

18

1 slightly abnormal adult

15:1

3 CO + 1 air

Previously chilled pupae

2

2

(3)

1 abnormal adult; 1 normal

adult

<5-25% development

2

2

15

1 abnormal adult; 1 normal

adult

10:1

4 CO +0.2 O2

Previously chilled pupae

8

8

(7)

3 abnormal adults; 3 normal

+ 1 air

adults; 2 died

<5-25% development

10

10

14

6 abnormal adults; 4 died

25-50% development

4

2

25

2 died at 70% stage of devel-

opment

5:1

1 CO + 1 air

Previously chilled pupae

2

2

(15)

No data

1:1

1 CO +0.8 Oz

<5-25% development

3

0

70

3 died at 70% stage of devel-

+ 1 air

opment

Total

78

Parentheses () indicate that one or more individuals initiated development in the presence
of carbon monoxide.

normal development and produced adult moths with various abnormalities. The
latter included defective scales, hairs, and pigmentation, along with incomplete or
malformed eyes, legs, antennae, and genitalia.

The endocrine competency of the brain itself was found to be unaffected by expo-
sure to carbon monoxide. Thus, brains removed from previously chilled pupae
after 21 days of exposure to 25 : 1 carbon monoxide/oxygen retained their activity
and evoked adult development when implanted into brainless diapausing pupae.

5. Effects of carbon monoxide on mature larvae

Mature larvae at the outset of spinning were exposed to carbon monoxide by the
static pressure method for one to six days. In control experiments four atmos-
pheres of nitrogen was substituted for the carbon monoxide.

The results recorded in Table III show that neither the behavior nor the viability
of the caterpillars was affected by four atmospheres of nitrogen. By contrast, no
individual was able to survive exposure to 33 : 1 carbon monoxide/oxygen for as
long as five days. Moreover, in the presence of carbon monoxide/oxygen ratios
as low as 1 : 1, the spinning of a normal cocoon was inhibited, the insect either failing
to spin or spinning only a flat sheet of silk.

CYTOCHROMK AND SILKWORM DEVELOPMENT

245

TABLK III

Effects of various carbon monoxide /oxygen ratios on mature fifth instar Cecropia larvae

CO ( ).,
ratio

Gas content of chamber

(atmospheres)

Duration
of
exposure
(days)

Type of spinning
behavior in
chamber

Type of spinning
behavior after removal
from chamber

Pupated

—

1 air

2

Normal cocoon

Normal cocoon

+

—

4 N2 + l air

2

Normal cocoon

Normal cocoon

+

33 : 1

6.7 CO + 1 air

5

None

Dead on removal

0

2

None

Normal cocoon

+

25:1

5 CO + 1 air

6

None

Dead on removal

0

20:1

4 CO + 1 air

1

None

Normal cocoon

+

15:1

3 CO + 1 air

5

Flat sheet

None

+

4

Flat sheet

Normal cocoon

+

2

Flat sheet

None

+

1

Flat sheet

None

+

5:1

1 CO + 1 air

3

Flat sheet

Normal cocoon

+

3:1

0.67 CO + 1 air

5

Flat sheet

None

+

2:1

0.4 CO + 1 air

3

None

Normal cocoon

+

1:1

0.2 CO + 1 air

3

None

Normal cocoon

+

3

Flat sheet

Normal cocoon

+

3

Flat sheet at first,

None

+

then continued with

a normal cocoon

6. Effects of carbon monoxide on fertile eggs and embryos

Embryonic development of Cecropia, from oviposition to hatching, requires about
ten days at 25° C. From the sixth to the tenth day, one can easily track the prog-
ress of embryonic development under the dissecting microscope and thereby estimate
the stage of embryonic development.

By means of the static pressure method, freshly oviposited fertile eggs were
exposed to a carbon monoxide/oxygen ratio of 20: 1 for 1, 3, and 5 days, respec-
tively. Similar experiments were performed on developing embryos which had
already completed 10, 30, 70, and 90 per cent of embryological development.

In control experiments compression with four atmospheres of nitrogen was with-
out major effects on viability, and approximately 90 per cent of eggs and embryos
hatched. However, even one day of exposure to 20 : 1 carbon monoxide/oxygen
considerably decreased the viability of the embryos. When returned to air, only
10 per cent of the eggs eventually hatched and only 50 per cent showed any detect-
able progress in embryonic development. Three days of exposure to the 20 : 1 mix-
ture was lethal in nearly all cases ; when returned to air, no eggs hatched and almost
all of the embryos were already dead. It is clear that both the development and
the viability of eggs and embryos are extremely sensitive to brief exposure to carbon
monoxide.

7. Effects of carbon monoxide on tJie adult moth

By the use of the static pressure method, adult Cecropia moths. 12 to 36 hours
after emergence, were exposed to various carbon monoxide oxygen ratios for pe-
riods up to five davs. The results summarized in Table IV demonstrate that the

246

H. A. SCHNEIDERMAN AX I) C. M. WILLIAMS

TABLE IV

Effects of various carbon monoxide / oxygen ratios on tin- viability of adult Cecropia moths

CO/O2

C,.is content
of chamber
(atmospheres)

X umber
of
animals

Duration
of
exposure
(days)

Behavior in chamber

Behavior after removal from chamber

—

1 air

2

2

Fluttering

Flying

—

4 Ns+1 air

2

2

Fluttering

Flying

33:1

6.7 CO + 1 air

2

2

Slight tremors
which ceased

Flaccid upon removal. Recovery
after 1 and 10 minutes. Feeble

after 2 hours

coordinated motion within 3 hours,
but no flight. Died within 4 days

20:1

4 CO + 1 air

2

5

Slight tremors
which ceased

Flaccid and dead upon removal

after 2 hours

4

3

Slight tremors
which ceased
after 2 hours

Flaccid upon removal. Recovery
after 10, 10, 20, and 60 minutes.
Extremely feeble uncoordinated
activity within 3 hours. Died
within 3 days without regaining
coordination

2

2

Slight tremors
which ceased

Flaccid upon removal. Recovery
after 1 and 10 minutes. Consid-

after 2 hours

erable coordinated activity within

3 hours, but no flight. Both lived
for 6 days after removal, one fe-
male laid eggs

8

1

Slight tremors
which ceased
after 2 hours

Flaccid upon removal. Recovery
after 30 seconds. Flying within
3 hours

moth is definitely sensitive to carbon monoxide. After three days of exposure to
a carbon monoxide/oxygen ratio of 20:1, the insects showed considerable de-
crease in vitality when returned to air ; exposure for five days was lethal. Equiva-
lent compression with nitrogen had no effect.

8. Photoreversibility of the carbon monoxide inhibition

Six pupae showing early adult development were placed head-down in an air-
filled Lucite tank such as illustrated in Figure 1, and compressed with carbon mo-
noxide to a 'final carbon monoxide/oxyen ratio of 20: 1. Three individuals were
illuminated continuously with a 250- watt mercury vapor lamp (General Electric
AH-5 ) via their terminal abdominal windows. The light was collected with a re-
flector and passed through a solution of sodium nitrite to cut off the ultraviolet and
through 5 cms. of water to eliminate the infra-red (Bowen, 1949). Three control
animals were loosely wrapped in aluminum foil to maintain them in darkness, and

CYTOCHROME AND SILKWORM DEVELOPMENT

247

placed in the same chamber. The latter was immersed in a water bath at 25° C..
the distance from the light source to the animals being approximately 25 cms.

Exposure to carbon monoxide and simultaneous illumination were continued for
5 days. The chamber was then decompressed and the experimental animals com-
pared with the controls. The genitalia of the illuminated animals had progressed
an average of 3.5 days ; that is, at 70 per cent of the rate in air. By contrast, the
genitalia of the unilluminated individuals showed no detectable progress. This dif-
ference was particularly striking at the anterior and posterior ends of the illuminated
animals in that the illuminated genitalia showed considerable progress in develop-
ment whereas the unilluminated facial region showed no morphological advance.
Since light-reversibility is a distinguishing property of carbon monoxide's inhibition
of cytochrome oxidase, the demonstration of light-reversibility is especially critical,
confirming for the insect as a whole the phenomenon as previously encountered in
cultures of isolated Cecropia spermatocytes ( Schneiderman, Ketchel and Williams,
1953).

9. Effects of o.vygen tension on animals at the initiation of adult development

Pupae showing the first day of adult development were exposed to continuously
flowing mixtures of oxygen and nitrogen for specific periods, usually 21 days. The
results recorded in Table V show that development was retarded by 13 per cent in

TABLE V

Effects of oxygen tension on the adult development of Cecropia
(animals on first day of development at outset)

Average rate

Oxygen tension
(per cent of
an atmosphere)

Number
of
animals

Days
in gas
mixture

of develop-
ment as per
cent of rate

Average rate of development
after return to air

Final state

in air

100

3

21

90

2 at 100% normal rate;

1 normal adult;

1 at 60% normal rate

2 slightly abnormal

21 (air)

3

21

100

100%

Normal adults

5

3

21

87

100%

Normal adults

3

3

21

52

100%

Abnormal adults

1

3

15

0

After 10 days in air, de-

1 adult with minimal de-

velopment began again

fects in antennal struc-

and continued at 100%

ture; 2 dead after 45%

normal rate

and 70% development

Less than 0.5

3

7

0

0

Dead when removed from

gas

5 per cent of an atmosphere of oxygen, and by 10 per cent in an atmosphere of pure
oxygen. Between these limits the rate of development was independent of oxygen
tension. Evidently, a gradient in oxygen pressure slightly in excess of 5 per cent
of an atmosphere is sufficient to meet the oxygen requirements of the developing
tissues. Those individuals which underwent development in the presence of oxygen
pressures less than 5 per cent showed abnormalities similar to those encountered
after exposure to carbon monoxide (cf. section 4).

248 H. A. SCHXEIDERM AN AM) C. M. WILLIAMS

10. Effects of o.vvgen tension on mature larvae

Eleven mature larvae were subjected for one to four days to specific low oxygen
tensions established by the flow method. The effects were judged in terms of the
insect's spinning behavior and subsequent pupation.

Essentially normal cocoons were spun until the oxygen tension was reduced
below 3 per cent of an atmosphere. At 2.5 per cent oxygen the animal usually spun
silk in a flat sheet ( rf. section 5 ) . At tensions lower than 2 per cent, spinning
ceased ; however, animals that had been exposed to this low tension for three days
spun normal cocoons when returned to air.

11. Effects of anoxia on larvae, diapausing pupae, and adults

Mature larvae and adult moths were killed by one day of exposure to tank nitro-
gen containing less than 0.5 per cent oxygen. When diapausing pupae were treated
in like manner, the heart ceased to beat after 4 to 7 hours. Half the animals were
dead after 72 hours ; the survivors, when returned to air, showed resumption of
heart beat and abdominal motion after one to two days.

DISCUSSION
1. Systematic changes in sensitivity to carbon monoxide

In the preceding paper of this series, evidence derived from respiratory studies
on the Cecropia silkworm demonstrated that marked changes occur in the sensitivity
of respiration to carbon monoxide during embryonic and post-embryonic develop-
ment. The results of the present study reaffirm these changes by demonstrating
that diverse physiological activities of the insect show parallel variations in sensi-
tivity to carbon monoxide. In the analysis of these findings it is convenient to sub-
divide the physiological activities of the insect into processes concerned with "main-
tenance" and with "growth and activity." The first of these include the minimal
metabolic events which sustain the viability and status quo of the organism. The
second category includes physiological processes responsible for morphogenesis and
similar highly involved and specialized activities.

Prolonged survival in the presence of high pressures of carbon monoxide signi-
fies that the gas fails to block the function of any tissue or organ required for the
maintenance of life. Death signifies that the function of at least one such tissue or
organ is blocked by carbon monoxide. In these terms it is clear that both the main-
tenance and the growth-and-activity processes are blocked by carbon monoxide in
the egg, embryo, and larva. After pupation, however, the maintenance and survival
of the diapausing pupa in the dormant state are insensitive to carbon monoxide.

The carbon monoxide-stable mechanism apparently remains intact during the
early stages of adult development. But, here also, carbon monoxide continues to
block development and to inhibit the contraction of all muscles except the heart.
Finally, in the late stages of adult development and in the adult moth, carbon mo-
noxide once again interferes with maintenance as well as with growth and activity.

Evidence has heretofore been presented that the target of carbon monoxide in
the insect is cytochrome oxidase (Schneiderman and Williams, 1954). The light-
reversibility of carbon monoxide's inhibition of growth is strong confirmation of this

CYTOCHROME AND SILKWORM DEVELOPMENT

249

view. Moreover, as was inferred in the previous study, the ability of the diapausing
pupa to survive in the presence of high concentrations of carbon monoxide signifies
that the loss or inactivation of the carbon monoxide-sensitive cytochrome oxidase
system at the time of pupation is compensated by the development of activation of
a carbon monoxide-stable respiratory system capable of underwriting the maintenance
requirements and the heart-beat of the diapausing insect. This finding affords a
remarkably clear illustration in biochemical terms of the dissociability of "mainte-
nance" and "growth" (Needham, 1942, p. 505 ff.).

2. 1 lie cytochrowie-cytochrowie (i.viduse svstcm and the energetics of development

The dependency of the growth and activity processes of Cecropia at all stages of
development upon respiration mediated by cytochrome oxidase finds many parallels.
From a study of the literature we have assembled in Table VI a number of processes

TABLK VI

Vital processes in which the inhibitory action of carbon monoxide
IKIS been found to be reversed by light

Material

1 . Arbacia eggs

2. Cecropia

spermatocytes

3. Drosophila

4. Avcna (oat)

5. Pisitni (pea)

6. Solatium (white

potato)

7. Dancus (carrot)

8. Rat

9. Ptcridium (bracken

sperm)
10. Fundulus (fish) heart Heart-beat

Reference

Cell division (mitosis) Clowes and Krahl (1940)

In vitro spermatogenesis (meiosis and Schneiderman, et al. (1951,

spermiogenesis) 1953)

Adult development \Yolsky (1937)

Growth of isolated coleoptile sections Hackett and Schneiderman

(cell elongation) (1953)

Growth or isolated stem sections (cell Hackett and Schneiderman

elongation) (1953)

Water uptake by tissue slices Hackett et al. (1953)

Salt accumulation by tissue slices Weeks and Robertson

(1950)

Incorporation of radioiodine in surviv- Schachner ct al. (1943)

ing thyroid tissue

Movement of bracken spermatozoids Rothschild (1951)

I 1 . Frog nerve

Action potential

Fisher and Cameron (1936,

1938)
Schmitt (1930)

where a light-reversible carbon monoxide inhibition has been reported. These in-
clude meiosis, mitosis, differentiation, cell elongation, water uptake, salt accumula-
tion, flagellar movement, and nerve conduction. As Lemberg and Legge ( 1949)
have reasoned (p. 383 ) : "Whether the respiration of the resting cell is always cata-
lyzed by the cytochrome system or not, it has become increasingly clear that the
functional activity of the cell depends on this system." See also Drabkin (1948).
For our present purposes it is of special interest that the inhibition of cytochrome
oxidase within the post-diapausing Cecropia establishes and enforces an artificial
diapause during the period of exposure of carbon monoxide. It is also noteworthy
that even in the diapausing pupa the inhibition of this enzyme prevents wound-
healing. From these several lines of evidence we learn that carbon monoxide-
sensitive metabolism plays an obligatory role in the energetics of development.

250 H. A. SCHNEIDERMAX AND C. M. WILLIAMS

Tlie absc'iice of all but a trace of a complete cytochrome oxidase system in the
diapausing pupa therefore assumes special significance (Williams, 1951). Since the
presence and function of this system appear to he prerequisite for adult development,
its virtual absence in the dormant pupa can, in itself, account for the developmental
stand-still of diapause.

In diapausing embryos of the grasshopper, Melanoplits, and of the commercial
silkworm, Bouiby.v, the absence of a cytochrome-mediated respiration has been
attributed to an inactivation of the cytochrome oxidase that is already present ; the
oxidase is thought to be re-coupled to metabolism in synchrony with the termination
of diapause (Bodine and Boell, 1938; Wolsky, 1949). But, in the case of the Ce-
cropia silkworm, the termination of diapause and the onset of development are ac-
companied by an actual synthesis of a new cytochrome system — not a mere re-
coupling of enzymes already present ( Sanborn and Williams, 1950 ) . The results
of the present investigation therefore link the respiratory and enzymatic studies and
demonstrate that cytochrome oxidase is the terminal oxidase in processes energizing
the insect's development.

The present study confirms the fact that qualitative as well as quantitative
changes occur in the energy metabolism of the Cecropia silkworm during the course
of metamorphosis. It also contributes to a coherent body of evidence that the cyto-
chrome oxidase system plays an obligatory role in the energetics of morphogenesis.
We are therefore persuaded that the recruitment and resynthesis of the cytochrome
oxidase system are among the biochemical changes set in motion by the growth and
differentiation hormone — changes which couple the endocrine action to the termina-
tion of the pupal diapause.

The experiments reported in Sections 5 and 10 were performed in collaboration
with Dr. William Van der Kloot and those in Section 9 in collaboration with Mr.
Roger Milkman. The photograph in Figure 1 was made by Dr. Roman Vishniac
and is used with the permission of Time, Inc.

SUMMARY

1. The effects of mixtures of carbon monoxide and oxygen on the growth and
metamorphosis of the Cecropia silkworm were examined at successive stages of
embryonic and post-embryonic development.

2. Embryos, mature larvae, and adults are killed by five days of exposure to
carbon monoxide/oxygen ratios of 20: 1 or 25: 1. Diapausing pupae, by contrast,
survive at least 21 days of exposure to carbon monoxide/oxygen ratios as high as
33:1.

3. While failing to interfere with the viability of diapausing pupae, carbon mo-
noxide blocks or greatly retards the termination of the pupal diapause ; it also in-
hibits the healing of experimental wounds in the pupal integument.

4. The ability to survive in the presence of high pressures of carbon monoxide
persists throughout the early stages of adult development. Exposure of the devel-
oping, post-diapausing insect to suitable pressures of carbon monoxide establishes
and enforces an artificial diapause which is reversed upon return to air.

5. The inhibition of adult development by carbon monoxide is light-reversible;
the degree of inhibition is a function of the carbon monoxide/oxygen ratio. These

CYTOCHROMK AND SILKWORM DEVELOPMENT 251

findings indicate that the effects of carbon monoxide are due to the poisoning of
cytochrome oxidase.

6. Resistance to carbon monoxide, as in the diapausing pupa, signals the pres-
ence and utilization of an oxidase other than cytochrome oxidase.

7. On the basis of these several lines of evidence, it is concluded that growth
and metamorphosis, at all stages in the life history, are dependent on metabolism
catalyzed by cytochrome oxidase. The function of cytochrome oxidase is likewise
prerequisite for the maintenance of life of the embryo, larva, and adult.

8. Only the diapausing pupa survives without regard to the presence or func-
tion of cytochrome oxidase, the maintenance metabolism of the pupae being served
by an unidentified oxidase which is insensitive to carbon monoxide.

9. With the termination of pupal diapause the growth and differentiation of the
adult moth again requires the function of the cytochrome oxidase system. This fact
is considered in relation to the endocrine control of the pupal diapause.

LITERATURE CITED

BODINE, J. H., AND E. J. BOELL, 1938. The influence of some dinitrophenols on respiratory

metabolism during certain phases of embryonic development. /. Cell. Cotnp. PliysioL.

11: 41-63.
BOWEN, E. J., 1949. The chemical aspects of light. 2nd edition. Oxford at the Clarendon

Press.
CLOWES, G. H. A., AND M. E. KRAHL, 1940. Oxygen consumption and cell division of fertilized

sea urchin eggs in the presence of respiratory inhibitors. /. Gen. PliysioL. 23 : 401-411.
DRABKIN, D. C., 1948. Distribution and metabolic aspects of derivatives of iron protoporphyrin

(hemin). Fed. Proc., 7: 483-492.
FISHER, D. C., AND J. A. CAMERON, 1936. Effect of light on the CO-poisoned embryonic

Fundulus heart. Biol. Bull., 71 : 404.
FISHER, D. C., AND J. A. CAMERON, 1938. The frequency of the CO-poisoned heart at different

mean light intensities. /. Cell. Comp. PliysioL, 11 : 433-454.
HACKETT, D. P., AND H. A. SCHNEIDERMAN, 1953. Terminal oxidases and growth in plant

tissues. I. The terminal oxidase mediating growth of Arena coleoptile and Pisitui

stem sections. Arch. Biochcin. Biophysics, 47 : 190-204.
HACKETT, D. P., H. A. SCHNEIDERMAN AND K. V. THIMANN, 1953. Terminal oxidases and

growth in plant tissues. II. The terminal oxidase mediating water uptake by potato

tissue. Arch. Biochcin. Biophysics, 47 : 205-214.
LEMBERG, R., AND J. W. LEGGE, 1949. Hematin compounds and bile pigments. Interscience

Publ. Inc., New York.
MICHENER, C. D., 1952. The Saturniidae ( Lepidoptera ) of the western hemisphere. Bull.

Am. Mns. Nat. Hist.. 98: 341-501.

NEEDHAM, J., 1942. Biochemistry and morphogenesis. Cambridge University Press.
ROTHSCHILD, LORD, 1951. Cytochrome-catalysis of the movement of bracken spermatozoids

(Ptcridium aqmlinum). Proc. Roy. Soc., London, Ser. B, 138: 272-277.
SANBORN, R. C., AND C. M. WILLIAMS, 1950. Oxidative enzymes in relation to pupal diapause

and adult development in the Cecropia silkworm. Atmt. Rcc., 108: 70.
SCHACHNER, H., A. L. FRANKLIN AND I. L. CHAiKOFF, 1943. The effect of cytochrome oxidase

inhibitors on the formation in vitro of thyroxine and diiodotyrosine by thyroid tissue

with radioactive iodine indicator. /. Biol. Chein., 151 : 191-199.
SCHMITT, F. O., 1930. On the nature of the nerve impulse. I. The effect of carbon monoxide

on medullated nerve. Amcr. J. PliysioL. 95: 650-661.
SCHNEIDERMAN, H. A., AND N. FEDER, 1954. A respirometer for metabolic studies at high

gaseous pressures. Biol. Bull.. 106: 230-237.
SCHNEIDERMAN, H. A., M. KETCIIEL AND N. FEDER, 1951. The cytochrome system in relation

to in vitro spermatogenesis in the Cecropia silkworm. Anat. Rcc.. Ill : 102.

252 H. A. SCHXEIDERMAX AND C. M. WILLIAMS

SCHNEIDERMAN, H. A., M. KETCHEL AND C. M. WILLIAMS, 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: 188-199.
SCHNEIIIKKMAX, H. A., AND C. M. WILLIAMS, 1953. The physiology of insect diapause. VII.

The respiratory metabolism of the Cecropia silkworm during diapause and develop-
ment. Biol. Bull., 105: 320-334.
SCHNEIDERMAN, H. A., AND C. M. WILLIAMS, 1954. The physiology of insect diapause. VIII.

Qualitative changes in the metabolism of the Cecropia silkworm during diapause and

development. Biol. Bull.. 106: 210-229.
SCHOLANDER, P. F., AND F. J. W. RouGHTON, 1943. Microgasometric estimation of the blood

gases. I. Oxygen. /. Biol. Chan.. 148: 541-550.
\\ I-KKS, D. C., AND R. N. ROBERTSON, 1950. Studies in the metabolism of plant cells, VIII.

Dependence of salt accumulation and salt respiration upon the cytochrome system.

Australian J. of Sci. Res., B.. 3 : 487-500.
WILLIAMS, C. M., 1946. Physiology of insect diapause: the role of the brain in the production

and termination of pupal dormancy in the giant silkworm, Platysamia cccropia. Biol.

Bull.. 90: 234-243.
WILLIAMS, C. M., 1951. Biochemical mechanisms in insect growth and metamorphosis. Fed.

1'ruc.. 10: 546-552.
WOLSKY, A., 1937. Production of local depressions in the development of Drosophila pupae.

Nature, 139: 1069-1070.
WOLSKY, A., 1949. The effect of carbon monoxide on the respiration of artificially bivoltinized

silkworm eggs. Current Science (India), 18: 323-325.

LETHALITY AND THE BIOLOGICAL EFFECTS OF X-RAYS

IN PARAMECIUM: RADIATION RESISTANCE

AND ITS VARIABILITY

RALPH \\TCHTERMAN' AXI) FRANK H. J. FIGGE -

l>epartmcnt of Biolof/y. Temple University; Department of Anatomy, University of Maryland
Medical School; Marine Biological Laboratory, ll'nods Hole, Massachusetts

It has been known for a long time that Paramecium and certain other Protozoa
are able to survive exceedingly high dosages of x-rays (see review, Wichterman,
1953). \Yith low, sub-lethal dosages, paramecia become perceptibly accelerated.
In normal bacterized culture media, dosages of 200,000 roentgen (r) and above
usually retard motility in Paramecium, and there are generally no survivors above
510,000 r. Occasionally, survivors of this high dosage produce clones which, after
overcoming irradiation effects, reproduce and flourish in a manner comparable to
controls (Wichterman, 1948). X-ray survival curves for microorganisms as re-
ported in the literature vary considerably, apparently depending upon the conditions
employed for irradiation. We find, for instance, that with certain methods and un-
der certain conditions in the irradiation of Paramecium caudatum, the LD 50 — that
dosage which results in the death of 50 per cent of irradiated organisms — may vary
from 75.000 r to 350,000 r.

The purpose of the present investigation was to establish a standard, repeatable
method of irradiation and to analyze the causes of radiation resistance and variability
in Paramecium.

To fully appreciate the insensitivity of paramecia to x-radiation. we need only
examine the LD 50 dosages of other organisms. According to Lea (1947), the
50 per cent survival dosage for yeast is 30,000 r ; for the bacterium B. coli, 5600 r,
and for spores of B. mcsentericus, 150,000 r. For the algae Chlorella, -Inkistro-
desuuts, and Chroococcns, the LD 50 is 22,000 r, 11,000 r, and 9.000 r. respectively
(Bonham and Palumbo, 1951 ). In this connection, it is to be noted that bacteria
in culture fluid, as well as those in the body of Paraincciitin and the symbiotic Chlo-
rclla in Paramecium hursaria, can be destroyed by x-rays without killing the para-
mecia (Wichterman. 1948). It is thus possible to sterilize such cultures to yield
species-pure clones of Paramecium as well as colorless races of the normally green
species, Paramecium hursaria. The recent accounts given by Curtis (1951) and
Nickson (1952) for some vertebrate animals commonly used in the laboratory are
seen to vary, but relatively low dosages of x-rays are required to produce 50 per
cent lethality. For instance the LD 50 for "baby" rats is given as 510 r but 590-
1280 r for adults. The LD 50 for other animals follows: mice, 400-840 r; guinea

1 Supported by the Committee on Research, Temple University and aided by a contract be-
tween the Office of Naval Research, Department of the Navy and Temple University (NR
135-233 ) .

'- Supported by grants from the Anna Fuller Fund and the American Cancer Society, Mary-
land Division, Inc.

253

254 R. WICHTERMAX AND F. H. J. FIGGE

pigs, 200-310 r ; rabbits, 790-1500 r ; dogs, 300-335 r ; monkeys, 500 r. According
to Sparrow and Rubin (1952), it has been estimated that the LD 50 for man would
be approximately 400 r when the x-radiation is received over the whole body in a
fairly short period of time. It is therefore worthy of note that Parainccium can-
datum, with an LD 50 of approximately 340,000 r when irradiated in Nylon syr-
inges, has a radiation resistance 850 times as great as that of man and some common
vertebrate laboratory animals.

As a test animal for the evaluation of irradiation effects and associated phe-
nomena, Paraiiicchiin has many useful features. Beginning with a single specimen,
it is possible to obtain for experimentation a genetically uniform, pedigreed strain
of enormous numbers of paramecia. This allows for speed and precision of obser-
vation generally impossible with other test animals. In addition to being a com-
pletely isolated cell, Paraineciiiui is a structurally complex organism; hence mor-
phologic changes as a result of irradiation can be determined readily. Irradiation
effects are manifested in loss of motility, which may include a change in ciliary action
or its complete cessation, dysfunction of contractile vacuoles, change in rate of cyclo-
sis, vacuolization, blistering of the pellicle, changes in body shape, and finally dis-
integration of the body. Also the division rate, which is an index of vitality, can
be compared with the control specimens and expressed in quantitative terms. Addi-
tional advantages in x-radiation experiments with paramecia may lie in the field of
biochemistry, especially in regard to the effects on respiratory mechanisms which
appear to be greatly involved.

MATERIALS AND METHODS

In the present study, all irradiation work was done at the Marine Biological
Laboratory, Woods Hole, Massachusetts. The x-ray generator operates simul-
taneously two water-cooled Coolidge tubes in alternate parallel. One tube was
mounted rigidly on a platform on the floor, and the other tube was supported on a
counter-balanced arm which allowed it to be moved vertically and in line directly
over the fixed tube. Paramecia in irradiation chambers were thus cross-fired from
above and below. The x-ray tubes operated at 182 kv. pk., and 25 ma., with an
equivalent filtration of 0.2 mm. of copper. When the tubes were brought very close
together (position A), which was the position used for all experiments, intensity
was 6300 r per minute. Not only were the tubes water-cooled, but an electric fan
was directed upon them, and the irradiated materal \vas surrounded by an ice cham-
ber. Temperature determinations were made by the use of a thermo- junction and
galvanometer. The junction was placed directly into the control irradiation cham-
ber ; thus it was possible to determine the small temperature changes — which proved
to be insignificant — during the entire time specimens were irradiated. Most of the
irradiation work was done at a temperature of 15° C.

Although different species of Paramecium were irradiated and results indicated
species differences in regard to x-ray susceptibility, the results reported here are
based upon the use of P. caudatinn." Cultures were begun with a single specimen
and cultivated in covered flasks containing either lettuce or hay infusions which
were inoculated with the bacterium Acrobactcr acrogcncs as the food source.

3 The original strain of Paramecium caudahtm (57-14) was kindly supplied by Dr. Lauren
C. Oilman, University of Miami.

EFFECTS OF X-RAYS IN PARAMECIUM

A

255

V

V

DROPS OF CULTURE FLUID
+ 5 — 20 PARAMECIA

ICC OF CULTURE FLUID
+ 100 PARAMECIA

FIGURE 1. Drawings illustrating how conventional plastic boxes were used to irradiate
Paramccium caudatum in drops and larger volumes of fluid.

256 R. WICHTERMAX AND F. H. J. FIGGE

Usually vegetative specimens to be irradiated were removed with a micropipette
from rich clonal cultures of pH 7.1 following the logarithmic growth phase. Such
active and vigorous animals were commonly uniform in size and shape. The
environmental culture fluid to he irradiated with the paramecia contained fewer
bacteria than during the active growth phase.

For most of the investigations, two types of irradiation chambers were em-
ployed. At first, the chambers used consisted of rigid, transparent, plastic boxes
with tightly fitting lids and of a type -commonly used in such experiments with
microorganisms. The boxes measured approximately 24 X 24 y. 18 mm. with a
volume of about 6 cc. (Fig. 1, A). It was possible to irradiate four boxes contain-
ing paramecia at one time. To study the influence of the ratio of the numbers of
animals to volume of fluid, drops of uniform size were suspended as hanging drops
from the lids of the boxes. The drops, each containing 10, 25, 50 and 100 para-
mecia, were then irradiated. Additional variations were made utilizing the plastic
boxes as shown in Figure 1 and described later. Subsequent experiments indi-
cated that the number of paramecia per unit of volume was not as important in
determining the lethal effects of x-radiation as the depth of the exposed culture
medium, volume of the moist air-space, and the amount of surface of the culture
medium exposed to the air in the radiation chamber.

A new type of radiation chamber was therefore employed to avoid the compli-
cating factor of the air-space which appeared to diffuse from the moist air and
which appeared to be extremely lethal to paramecia ( Fig. 2 ) . This new chamber
consists of a Nylon hypodermic syringe of 2 cc. capacity and graduated in units
of one-tenth of a cc. (0.1 cc. ). A tightly fitting Lucite cap is applied over the
tapering tip of each syringe. The syringe absorbs very little irradiation, elimi-
nates air from the irradiation chamber, and permits the introduction of various sub-
stances to be tested during irradiation. The syringes may be sterilized in an auto-
clave. Accurate sampling of specimens after intervals of irradiation without
changing the depth of the medium is also a desirable feature. A Plexiglas holder 4
measuring 11.5 X 8.5 X 2.5 cm. was designed to hold four syringes, all of which
could be irradiated at the same time. The syringe-chamber method is thus ideal
for the study of lethality of x-rays in Paraincciuui and should prove to be useful
for similar studies with other microorganisms. Before sampling and immediately
after irradiation, the syringe was quickly rotated between the fingers of both hands
in order to distribute the paramecia uniformly. Usually 100 specimens in two cc.
of fluid were placed in each syringe and irradiated in steps of 20,000-50,000 r. By
expressing 0.2 cc. of irradiated fluid after a given dosage, it was possible to de-
liver into sterile Pyrex spot plates a precisely countable number of specimens—
commonly ten — for the establishment of survival curves. Animals were examined
immediately after irradiation, then placed in moist chambers for subsequent
observation.

«

RESULTS AND DISCUSSION

Irradiation with x-rays markedly increases the viscosity of the protoplasm of
Poranicchini caudatuni ; greater dosages lead to irreversible coagulation. Prior to

4 The Plexiglas syringe holder with self-contained ice chambers was constructed by Mr.
Michael Troisi, Instrument Maker, Temple University.

EFFECTS OF X-RAYS IX PARAMFCIUM

257

FIGURE 2. Photograph showing four 2-cc. Xylon syringes (with Lucite caps in place)
being used as irradiation chambers. An ice well is present on each side of the syringe holder.
(Slightly less than actual size.)

258

R. WICHTERMAX AXD F. H. J. FIGGE

death, paramecia become immobilized, change shape to become broadly ellipsoidal
and settle on the bottom of the irradiation chamber. Contractile vacuoles function
more slowly and sometimes become abnormally large. Active cyclosis ceases as
the protoplasm becomes conspicuously darker and vacuolated. Clear, transparent,
structureless, blister-like swellings appear on the pellicle prior to death. Near
death, waves of trichocysts are extruded, suggesting that these structures — com-
monly thought of as organelles of defense — represent a response to an injury re-
action. Specimens frequently become sub-spherical before their disintegration
(Fig. 3).

B

C

D

E

FIGURE 3. Effects of high dosage x-radiation on Paraiiicchiin candatiiin (X 190). A: Un-
irradiated control specimen. B : Irradiated with 255,000 r resulting in slight change of body
shape; animals generally recover from this dosage. C: Irradiated with 340,000 r (approxi-
mately the LD 50 dosage) in which locomotion and cyclosis are retarded. D : Irradiated with
425,000 r in which body shape becomes broadly ellipsoidal ; greatly decreased locomotion ; vacu-
olization. E and F : Irradiated with 510,000 r resulting in cessation of locomotion and cyclosis,
increased vacuolization, blistering of pellicle, darkening (coagulation) of protoplasm followed
by disintegration and death. (Photographs taken of specimens irradiated in Nylon syringes
immediately after removal from x-ray generator. )

Our data are based on specimens observed for at least 24 hours after irradi-
ation, commonly longer. The survival curves based upon this method are sigmoid
as is the case with most irradiated biological material. Occasionally the slope of
the curve is so steep approaching lethality as to be almost vertical. For a 24-hour
period, the LD 50 for Paramecium candatitiiif\s approximately 340,000 r (Fig. 4).
It was soon found that the I. L. D. (immediate lethal dose), as defined by Back
and Halberstaedter ( 1945 ) — that dosage which produced a complete cessation of
motility within 10-15 minutes after irradiation — -was not reliable as a useful end-
point. We have found that such immobilized paramecia may appear to be dead,
but if examined hours later may be seen to be not only as active as control speci-
mens but may eventually divide and produce successful clones. However, it is of
interest to note that Back and Halberstaedter report the I. L. D. to be approxi-
mately 350,000 r, a dosage close to our results when using the syringe method.

The results showing percentage survival after irradiating paramecia in drops
and larger volumes of fluid in plastic boxes (Fig. 1) and in Nylon syringes are

EFFECTS OF X-RAYS IX PARAMECIUM

259

TABLE I

Survival of Paramecium caudatum after rocntgen irradiation
in plastic boxes and nylon syringes

Influence of the degree of exposure of animals and culture medium
to air during irradiation

No. of

No. of

Per cent survival dosage in kr.

; i ,

groups

observed

85

128

170

212

255

Mm

340

383

425

510

Paramecia in hanging

4

55

0

0

•o

0

0

0

drops in 6-cc. plastic

12

260

0

0

0

0

0

0

boxes containing 1 cc.

12

140

5

0

0

of culture fluid

Paramecia in 1 cc. of

10

1000

100

culture fluid in bottom

4

400

100

0

of plastic boxes with

12

335

100

100

0

0

0

0

cover (volume 6 cc.)

Paramecia in 1-2 cc. of

culture fluid in Xylon

36

1335

100

100

95

94

81

57

44

19

2

0

syringe (noair bubbles)

given in Table I. From this tabulation, it may be seen that the paramecia in hang-
ing drops in plastic boxes were much more sensitive to roentgen radiation than
the paramecia in the one cc. of culture fluid placed in the bottoms of the plastic
boxes (Fig. 1, A). Dosages of 170 kr. killed nearly all of the paramecia in the
drops whereas such dosages failed to kill any of the paramecia in the one cc. of fluid
in the bottom of the plastic boxes. In most instances, the paramecia placed in
hanging drops in the covers of the plastic boxes and the paramecia in the culture
fluid in the bottom of the boxes were irradiated simultaneously. Variations in the
concentration of paramecia in the drops and in the culture fluid in the bottom of
the box did not alter this great difference in radiation sensitivity between drop and
culture fluid in the bottom of the box. The only essential difference between these
two conditions was the difference in the relative amount of surface exposed to air
in the chambers.

It was also quite apparent that even the paramecia in the culture fluid in the
bottom of the boxes succumb to the radiation in an almost "all or none" manner.
When a dose of 170-200 kr. was exceeded, all paramecia died; in lower dosages,
all lived. Some experiments were performed in which the influence of the depth
(volume) of the culture medium was tested, since it was thought that variations
in culture medium might have been responsible for an x-ray nitration effect. This
did not appear to be the reason, however, for the differential sensitivity in drops,
as compared with sensitivity in larger volumes of culture fluid. In some experi-
ments, drops with 5-20 paramecia were placed in plastic boxes and one cc. of cul-
ture fluid containing 100 paramecia was placed in inverted lids above and below
the plastic box chamber containing the drops (Fig. 1, B ). The two cc. of culture
fluid in the lids thus partially shielded the paramecia in the drops in the boxes.
In other similar boxes containing drops with 5 and 20 paramecia per drop, the one

260

R. WICHTERMAX AXD F. H. J. FIGGE

cc. of culture medium above and below was omitted (Fig. 1, C). Both sets of
boxes were irradiated with 170 kr. This dose killed all of the paramecia in the
drops in both boxes. All of the paramecia in the one cc. of culture fluid in the
inverted lids survived. In the case of the paramecia in the inverted lid on top of
the box (uncovered and exposed to the atmospheric oxygen at the surface of the
culture medium), survival was 100 per cent. Thus it was apparent that all the
paramecia in the drops in the plastic container were killed even though they wrere
partially shielded by two cc. of culture fluid (one cc. above and one cc. below).

Because of this differential sensitivity resulting from differences in the degree
of exposure of the culture medium to air, the plastic boxes and hanging drops were
abandoned and the Nylon syringes were utilized as x-radiation chambers for the
reasons given earlier. The irradiation of paramecia in the syringes ( which con-
tained no air bubbles ) yielded results that were much more uniform. Using the
syringe method, the results of seven experiments involving nine different dosage
groups and 36 determinations are shown in Table I and the survival curve of Fig-
ure 4. In the Xylon syringes, most of the paramecia survived a dose of 170-212
kr. (lethality := 5-6 per cent). As this dosage is exceeded, however, the per cent
of animals that survive 24 hours after irradiation takes a sharp drop. Generally
no animals survived a 510 kr. dose.

100-

90-

80-

70-

60-

50-

40-

30-

20 H

10 -

X-RAY SURVIVAL CURVE
FOR PARAMECIUM

170

255

DOSE IN KR.

340

425

510

FIGURE 4. X-ray survival curve for Paraincciiuii caudatnin irradiated in Nylon syringes.
This curve is based on seven experiments and after irradiated paramecia were observed for 24-
hour period after expulsion from Nylon syringes. Coincident points are not indicated. Each
point represents observations on 10-25 counted paramecia.

EFFECTS OF X-RAYS IN PARAMECIUM

261

LETHALITY OF ROENTGEN RADIATION IN PARAMECIA

P

R

0

B

I

T

S

98

95
90

80 -

70
60
50
40
30
20

10 -
5 -

2

I

20

40

100

I 1 1^

200 300 500

DOSE IN

FIGURE 5. Dosage-effect curve for lethality of roentgen radiation in Furnnicciiiiii cm/datum.
Results were recorded on probability paper for plotting percentages directly on a probit scale.

The data for the per cent survival after irradiation in Nylon syringes were also
plotted on log-probit paper ( Fig. 5 ) . The significance of such a curve to assist in
the analysis of data concerned with all-or-none responses is described by Bliss
(1952). From this curve, it may be seen that while there was not a straight line
relationship at the higher and lower percentages, one was present in the important
range between 10 and 90 per cent. From an examination of this curve, it may be
concluded that the LD 50, 24 hours, for Paraincchtiii caiidatuin irradiated in Nylon
syringes is approximately 340 kr.

From the experiments with plastic boxes, it was concluded that the number of
paramecia per unit of volume was not as important in determining the lethal effects
of x-radiation as the depth of the exposed culture medium, volume of the moist air-
space, and the amount of surface of the culture medium exposed to the air in the
irradiation chamber. This gave rise to the hypothesis that some toxic gaseous sub-
stance, possibly ozone (Taylor, 1935) was diffusing into the fluid from the irradi-
ated moist air-space of the chamber. However, we were unable to detect ozone
formation in the irradiated air of the chamber, even with the most sensitive tests.
The toxic factor derived in whole or in part from the moist air in the closed boxes

262 R. WICHTERMAX AXU F. H. J. FIGGE

(hiring irradiation is probably oxygen or a derivative of oxygen, hydrogen peroxide
or some other oxidation product.

When sealed Xylon chambers of air are irradiated with 400 kr. and unirradiated
paramecia then drawn into such chambers without outside air being permitted to
enter, the paramecia live for as long a period of time as the controls. This shows
conclusively that the irradiated air by itself is not toxic to the animals. Also when
unirradiated paramecia are placed in irradiated fluid (400 kr. ) exposed and not
exposed to air, and in irradiated mixtures of air and culture fluid, paramecia are
not killed.

It has been known for a long time that water exposed to ionizing radiations forms
hydrogen peroxide which may be lethal to ciliates (Taylor, Thomas and Brown.
1933 J . This does not hold for oxygen-free pure water in which no hydrogen perox-
ide can be demonstrated even photocolorimetrically (sensitivity 0.1 y per ml. )
(Bonet-Maury, 1951 ). In irradiation chambers containing clear culture fluid with
bacterized paramecia, minute amounts of the enzyme catalase originate from the
microorganisms and tend to offset the toxic effect of hydrogen peroxide. Accord-
ing to Dale (1951), one molecule of catalase can decompose 5,000,000 molecules
of hydrogen peroxide per minute at 0° C. Kimball and Gaither ( 1952. 1953 ) .
using Paniiiieciiiin anrclla, report that hydrogen peroxide is of major importance
in the production of certain kinds of nongenetic effects but only under certain
circumstances.

A study of the biological effects of ionizing radiations upon Paraineciitin must
take into account the effect of these radiations on the environment in which these
organisms live. The culture fluid in which the specimens are irradiated consists
mainly of wrater with organic matter from the hay or lettuce infusions. A great
body of literature demonstrates that as a result of irradiation of water, hydrogen
peroxide, hydrogen and oxygen are formed in which the amounts and relative pro-
portions depend upon such factors as dissolved oxygen concentration, radiation ionic
density, close, temperature and pH. Water that is irradiated oxidizes reducing
agents and reduces oxidizing agents (Bonet-Maury, 1951).

In the irradiation of paramecia, another factor that plays a part besides the effect
of ionizing radiations of water on the cell is the effect of the accompanying x-rayed
bacteria present in the culture as the food source. Experiments in which the irra-
diated bacteria of paramecia cultures were plated out at intervals up to 350 kr. show
the bacteria to have a far lower LD 50 than the paramecia. Another factor to take
into account is the indirect or direct effect of radiations of the dead bacteria and
their fragmented cells upon paramecia. The experiments with bacteria also showed
the necessity of bacterizing spot plates containing irradiated paramecia and fluid
if one is to make observations over long periods of time. Failure to do this will
result in slower division rates ; perhaps ultimate starvation of the paramecia in ir-
radiated paramecia samples.

SUMMARY

1. Irradiation with x-rays markedly increases the viscosity of the protoplasm
of Paramccium caitdatuui ; greater dosages lead to irreversible coagulation. With
increased irradiation, paramecia become immobilized, become broadly ellipsoidal and
settle on the bottom of the irradiation chambers. Contractile vacuoles function

EFFECTS OF X-RAYS IX PARAM KCIUM 263

more slowly and occasionally become abnormally large. Prior to death, cyclosis
ceases and the protoplasm becomes darker and vacuolated. Clear, blister-like swell-
ings appear at the pellicle. Before death, waves of trichocysts are extruded sug-
gesting that their function may represent an injury-reaction. Finally, paramecia
frequently become sub-spherical before their disintegration.

2. It was found that one of the most important factors influencing the lethal
effects of x-radiation was the degree and extent of exposure of the fluid containing
paramecia to air. Paramecia in hanging drops were killed by dosages (170 kr. )
that exhibited no lethality for paramecia in larger volumes of culture fluid. This
difference in lethality occurred even though the numbers of paramecia per unit vol-
ume were kept uniform in both drops and larger volumes.

3. A new method using Xylon syringes was devised to minimize the variability
of x-radiation effects.

4. Survival curves were established for Paramecium candatiiin using this new
method. It was found that the LD 50, 24 hours was approximately 340 kr.

LITERATURE CITED

BACK, A., AND L. HALBERSTAEDTER, 1945. Influence of biological factors on the form of
Roentgen-ray survival curves. Amcr. J. Roentgenol., 54: 290-295.

BLISS, C. I., 1952. The statistics of bioassay. Academic Press Inc., New York. Pp. 445-628.

BONET-MAURY, P., 1951. Hydrogen peroxide formation in water exposed to ionizing radi-
ations. Brit. J. Radio!., 24 : 422-428.

BONHAM, K., AND R. F. PALUMBO, 1951. Effects of x-rays on snails, Crustacea and algae
Growth. 15: 155-188.

CURTIS, H. J., 1951. Advances in biological and medical physics. Academic Press Inc., New.
York. Volume 2, pp. 1-50.

DALE, W. M., 1951. Some aspects of the biochemical effects of ionizing radiations. Brit. J.
Radio!.. 24: 433-435.

KIMBALL, R. F., AND N. GAITHER, 1952. Role of externally produced hydrogen peroxide in
damage to Paramecium aurclia by x-rays. Proc. Soc. Exp. Biol. Ifcd., 80: 525-529.

KIMBALL, R. F., AND N. GAITHER, 1953. Influence of oxygen upon genetic and nongenetic ef-
fects of ionizing radiation on Paramecium aurclia. Proc. Soc Erf' Biol Ifcd 82'
471-477.

LEA, D. E., 1947. Actions of radiations on living cells. The Macmillan Co., New York. Pp.
1-402.

NICKSON, J. J., 1952. Symposium on radiobiology. John Wiley and Sons, Inc., New York.
Pp. 1-465.

SPARROW, A. H., AND B. A. RUBIN, 1952. Survey of biological progress : Effects of radiation
on biological systems. Academic Press Inc., New York. Volume 2, pp. 1-43.

TAYLOR, C. V., 1935. The effects of x-rayed medium on living cells. Estratto dagli Atti del I
Congresso Internazionale di Elettro-radio-biologia. Vol. II.

TAYLOR, C. V., J. O. THOMAS AND M. G. BROWN, 1933. Studies on Protozoa, IV : Lethal ef-
fects of the x-radiation of a sterile culture medium for Colpidimn canipvliiin. Phvsiol.
Zoo/., 6 : 467-492.

WICHTERMAN, R., 1948. The biological effects of x-rays on mating types and conjugation of
I\iraincciiiui hitrsaria. Bio!. Bull., 94: 113-127.

WICHTERMAN, R., 1953. The biology of Paramecium. The Blakiston Company, Inc., New
York. Pp. 1-527.

ERRATUM

In the paper by John R. Gregg and Norma Ornstein on
"Explant systems and the reactions of gastrulating amphibians
to metabolic poisons," which appeared in the December, 1953 issue
of THE BIOLOGICAL BULLETIN (Volume 105, No. 3), paragraph
(3 ) on page 476 should read as follows :

"(3) Among the inhibitors that we have studied, sodium bar-
bital is unique in suppressing all three of Em (en, en), St (m, en)
and Sp (ec, en), but in allowing Fu (m, ec, en) to occur to some
extent. The precise embryological interest of this result is not
clear."

Vol. 106, No. 3 June, 1954

THE

BIOLOGICAL BULLETIN

PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY

EFFECT OF ADENOSINETRIPHOSPHATE (ATP) ON THE

ENDOGENOUS OXYGEN UPTAKE OF DEVELOPING

GRASSHOPPER EMBRYOS x

JOSEPH HALL BODINE AND WILLIAM LIONEL WEST
Zoological Laboratory, State University of Iowa, loiva City, loiva

A phosphate transfer system has been found in practically every tissue thus
far investigated (Lardy, 1949). It functions directly or indirectly in almost
every phase of metabolism and has become so well established that its presence
and functions are very often inferred without further demonstration. Previous
studies in this laboratory have revealed that homogenates of the embryos of the
grasshopper, Melanoplus differential-is, in 0.25 M sucrose can oxidize hexose
phosphates to a greater extent than glucose (Bodine and West, 1953). Bodine
and Thompson (1938) reported that labile phosphate is found in both the embryo
and yolk while Lu and Bodine (1953) found a gross transformation of phosphorus
from yolk to embryo. However, very few direct observations seem to have been
made on the chemical or functional nature of the phosphate transfer system in
developing organisms. Recently Albaum and Kletzkin (1948) and Calaby (1951)
confirmed the presence of ATP in insects. Humphrey and Siggins (1949) pre-
sented indirect evidence that glycolysis in insect muscle involves the phosphate
transfer system while Sacktor (1953) describes a specific ATPase in flight muscle
mitochondria.

The present paper is concerned with results of a study on the effects of ATP
on the endogenous O2 uptake of grasshopper embryos at different developmental
stages. These results are discussed in the light of a functional phosphate transfer
system as exhibited by other organisms.

MATERIALS AND METHODS

Embryos of the grasshopper, Melanoplus differentialis, were dissected from
eggs in Ringer solution (buffered at pH 6.8 with M/15 phosphate) and washed free
of adhering yolk (Bodine and Boell, 1934, 1936). The washed embryos, sus-
pended in a suitable volume of the selected medium, were homogenized using a
Pyrex glass tube with a tight fitting selenite rod as a pestle. The pestle rotated at
1150 r.p.m. and the time of homogenation was two and one-half minutes at 0° C.

Two suspension media were used, Ringer solution containing 0.0035 M mag-

1 Aided by a grant from the National Institutes of Health.

265

266

J. H. BODINE AND W. L. WEST

nesium chloride (pH 6.8) and 0.25 M sucrose containing 0.0035 M magnesium and
0.0035 M calcium chlorides and 0.03 M phosphate (pH 6.8) (Bodine and West,
1953). One hundred intact embryos or homogenates containing the equivalent of
one hundred embryos per cubic centimeter were used throughout this investigation.

Oxygen uptake determinations (air as gas phase) were carried out by standard
Warburg techniques at 25° C. ; 0.5 cc. of substrate were tipped from the sidearm
to make the final volume of the reactants 1.5 cc.

Adenosinetriphosphate (ATP) (sodium salt) was obtained from the Sigma
Chemical Company, St. Louis, Missouri.

RESULTS

The effect of ATP on the endogenous O2 uptake of intact embryos (mitotically
active or blocked) was investigated over a range of concentrations from 2.5 to
10.0 //.moles per 1.5 cc. Data for a typical experiment are summarized in Table I.
From an examination of this table it is apparent that ATP has little, if any, sig-

TABLE I

Prediapause
(17 days)

Diapause
(40 days)

Postdiapause
(3 days)

E

H

E

H

E

H

Control (sucrose)
ATP

17.0

16.0

5.7
9.4

10.0
9.3

4.6
6.5

17.4
16.5

5.9

13.3

ATP+glu
ATP+glu-l-PO4
Glucose -1-PO4

15.0
19.6
18.8

10.2
13.2
8.5

9.1
14.2
15.2

8.0
9.7
6.5

16.7
18.2
18.5

14.3
13.6
9.0

Shows O2 uptake (cc.) for 100 minutes for prediapause, diapause and postdiapause embryos
(E) and their homogenates (H) in 0.25 M sucrose plus Mg++ and Ca++ after addition of ATP,
5 rnioles per flask; glucose 1.0%; glucose-1-phosphate 0.5%. Stimulation due to hexosephos-
phate has previously been pointed out (Bodine and West, 1953). Data in table are taken from
one series of experiments and represent averages from a minimum of 8 determinations. All
data from different experiments have been statistically analyzed and differences, indicated in
text, found to be significant.

nificant stimulating effect on the respiration of intact embryos either in 0.25 M
sucrose or Ringer solution. This lack of effect may be related to or conditioned
by the permeability of the intact embryo to these reagents.

The effect of ATP on the endogenous O2 uptake of homogenates of embryos
in 0.25 M sucrose is strikingly different from that of the intact embryo. ATP
augmented the endogenous respiration of homogenates in sucrose (Table I). The
concentration effect was found to be quite variable at high concentrations and this
is attributed to the formation of clumps which entangled the mitochondrial ele-
ments, thus preventing or interfering with electron transfers. This clumping
effect was more apparent in diapause and postdiapause stages at the 10 /tmole level
of ATP. Clumping is believed to be caused by an involvement of embryonic actin,
myosin, and ATP and is given support by the observation that clumping seldom
occurred in the prediapause stages before 17 days at which time the percentage

ATP AND OXYGEN UPTAKE 267

stimulations were more consistent. Maximal augmentation of respiration was
obtained at the 5 //.mole level where the clumping effects were absent. This
concentration has been selected as optimal in these experiments. ATP, when
tipped from the sidearm, produces a lag before maximum augmentation of oxygen
uptake. Homogenates made in 0.25 M sucrose plus ATP (employed only for
diapause) showed a greater oxygen uptake than when ATP was added to the
sucrose homogenate. The magnesium ion was necessary for maximal stimulation
by ATP.

Homogenates made in Ringer and Mg++ showed no stimulation of endogenous
oxygen uptake when ATP was added.

Combinations of ATP and glucose produced no marked hexokinase activity
in either the intact embryo or its homogenate. Similarly, no marked phosphoglu-
cokinase activity was apparent.

Washed nuclei in sucrose or Ringer showed no response in their endogenous
oxygen uptake to these concentrations of ATP.

DISCUSSION

The exact nature of the labile phosphorus compounds of the phosphate transfer
system in this material has not yet been satisfactorily demonstrated, due largely
to various inherent technical difficulties. However, it is known that the labile
phosphorus component of the embryo is adsorbed on activated charcoal (method
of Crane and Lipmann, 1953), which is a characteristic of the adenosine-containing
nucleotides (unpublished data). The ability of hexosephosphates to stimulate
endogenous respiration of intact embryos is quite unusual and no active mechanism
has been revealed (Bodine and West, 1953). ATP, unlike the hexosephosphates,
seems to have no stimulating effect on the endogenous respiration of the intact
embryo (mitotically active or blocked). Similarly, glucose plus ATP gave no
increased endogenous O2 uptake, indicating no marked hexokinase activity at or
near the cell membrane.

ATP markedly stimulates endogenous respiration of the homogenates in
0.25 M sucrose (Mg++, Ca++) and thus one can infer a functional phosphate transfer
mechanism. This effect may take place through "active" phosphorylation of
endogenous substrates, making them more available for oxidation, or "active"
dephosphorylation by a specific ATPase, increasing the concentration of high
energy phosphate acceptors (ADP + AMP) and permitting the oxidation of avail-
able endogenous substrates or a combination of both. (This discussion presupposes
that oxidation and phosphorylation are linked.) Studies are in progress to clarify
this point.

ATP does not stimulate the endogenous respiration of Ringer homogenates.
The mitochondria lose their morphological integrity in this medium and show a
marked functional difference to added succinate and hexosephosphates. Thus
structural integrity of the mitochondria in this material seems related to their
functions.

Combinations of glucose or hexosephosphates with ATP in sucrose homogenates
yield variable results. No effort was made to remove the endogenous substrate,
and at present it can be said that there appears to be no marked hexokinase or
phosphohexokinase activity in this material.

268 J. H. BODINE AND W. L. WEST

SUMMARY

1. A study has been made on the effects of ATP on the endogenous O2 uptake
of grasshopper embryos and homogenates at different developmental stages.

2. ATP has little, if any, effect upon the CX uptake of the intact embryo.

3. ATP augments the O2 uptake of homogenates in sucrose.

4. The magnesium ion is necessary for maximal stimulation of ATP.

•

LITERATURE CITED

ALBAUM, H., AND M. KLETZKIX, 1948. Adenosinetriphosphate from Drosopliila melanogastcr.

Arch. Biochcm., 16 : 333-337.
BODINE, J. H., AND E. J. BOELL, 1934. Respiratory mechanisms of normally developing and

blocked embryonic cells. /. Cell. Comp. Physiol., 8 : 357-366.
BODINE, J. H., AND E. J. BOELL, 1936. Respiration of embryo versus egg. /. Cell. Comp.

Physiol., 8 : 357-366.
BODINE, J. H., AND V. THOMPSON, 1938. Phosphorus distribution in the grasshopper egg.

/. Cell. Comp. Physiol., 12 : 247-254.
BODINE, J. H., AND W. L. WEST, 1953. Carbohydrate metabolism of the developing egg and

embryo. Biol. Bull., 104: 1-11.
CALABY, J. H., 1951. Adenosinetriphosphate from insect muscle. Arch. Biochem. Biophysics,

31 : 294-299.
CRANE, ROBERT K., AND FRITZ LIPMANN, 1953. The effect of arsenate on aerobic phosphoryla-

tion. /. Biol. Chcm., 201 : 235-243.

HUMPHREY, G. F., AND LORRAINE SIGGINS, 1949. Glycolysis in the wing muscle of the grass-
hopper, Locusta migratoria. Australian J. Expcr. Biol. Med. Sci., 27: 353-359.
LARDY, H. A., Ed., 1949. Respiratory enzymes. Burgess Publishing Company, Minneapolis.
Lu, KIAO-HUNG, and J. H. BODINE, 1953. Changes in the distribution of phosphorus in the

developing grasshopper (Mclauoplus differentialis) embryo. Physiol. Zool., 26: 242-

254.
SACKTOR, BERTRAM, 1953. Investigation on the mitochondria of the house fly, Musca donifstica

L. /. Gen. Physiol., 36: 371-387.

NUTRITIONAL STUDIES ON THE AMOEBO-FLAGELLATE,

TETRAMITUS ROSTRATUS *

MORGAN M. BRENT -
Department of Zoology, University of California. Berkeley 4, California

Nutritional studies on amoebae have lagged far behind those on ciliates and
flagellates. The earliest controlled work on the nutrition of members of the
Sarcodina concerned the cultivation of single species of amoebae in the presence
of bacterial mixtures. Here, most emphasis was placed upon the culture medium
itself, the living bacteria being considered as obligate factors for growth of the
phagotrophic forms. Scattered reports of axenic (Dougherty, 1953) cultures of
amoebae have been recorded in the literature, but the earliest accounts have not been
corroborated by other investigators. Included in these is the reported cultivation
of some amoebae in vitro on sterile mammalian tissues by Williams (1911). She
believed that these amoebae were parasites, but this is doubtful in the light of
present research upon known parasitic forms. Oehler (1924) claims to have
grown several unknown species of free-living amoebae under axenic conditions
upon water-agar containing coagulated serum. As far as is known this work has
not been re-investigated.

One of the first species to be grown free of living bacteria and which can be
obtained today is the free-living amoeba, Acanthamocba castellanii. Castellani
(1930) found that this amoeba was capable of growing upon a killed bacterium and
dead yeast. Cailleau (1933b) succeeded in obtaining an entirely liquid medium for
A. castellanii consisting of peptone and added salts. This has been modified some-
what by Storm, Hujr\Jer and Cowperthwaite (1951) who grew the amoeba upon a
medium containing acid-hydrolysate of gelatin and a low concentration of skim milk.
The same authors (1951 ) have also reported growth of Hartmannella rhysodcs upon
an autoclavable liquid medium containing hydrolysates of gelatin and free oleate
esters.

Reich (1935) reported axenic cultivation of the soil amoeba, Mayorella pales-
tinensis, upon a clear medium consisting of peptone, a balanced salt solution, and
added dextrose. The latter component appeared indispensable for good growth.

On the other hand, many investigators have failed to grow amoebae axenically.
Among these is Wherry (1913) who, in preliminary experiments, could not succeed
in growing an amoebo-flagellate (probably Naeglcria gruberi) free of the bacterial
flora. Rice (1935) succeeded in growing the marine amoeba, Flabcllula mira, in
monobacterial cultures, but failed to cultivate this form axenically by adding either
killed bacteria or various amino acids to the basal medium.

1 A portion of a dissertation presented to the graduate school of Northwestern University
in partial fulfillment of the requirements for the Ph.D. degree in biology. The author wishes
to express his appreciation to Dr. William Balamuth for his helpful suggestions and interest
during the course of this investigation.

-Present address: Department of Zoology, University of California, Berkeley, California.

26Q

270 MORGAN M. BRENT

Although a few of the parasitic flagellates have been cultured axenically in vitro,
the nutrition of the parasitic amoebae has provided a more difficult problem. In
the case of Entamocba histolytica, the amoebae apparently require not only certain
metabolites provided by the basal medium but also an anaerobic environment medi-
ated by the associated bacteria (Snyder and Meleney, 1943), as well as the bacterial
bodies themselves (Shaffer, 1952). As far as is known at present, E. histolytica
has not been grown in vitro free of the influence of living protoplasm (other
Protozoa, bacteria, embryonic tissue cultures).

As one approach to the nutrition of the parasitic amoebae it seemed desirable to
investigate a free-living coprophilic form, since these Protozoa pass sporadically
through the lower digestive tract of various animals. Although sharing the same
environment as intestinal parasites in this respect, they seem to be incidental guests
of their hosts and might conceivably stand at the threshold of parasitism. Little
work has been done on the amoebo-flagellate, Tctramitns rostratus, apart from
morphological and cytological studies. According to most investigators this organ-
ism is a coprophile which actively feeds and multiplies in its trophic stages upon
substrates rich in organic material. It has also been found in fecal cultures of
various animals by Bunting (1922) and Rafalko (1951). In view of these facts
this organism provided an excellent opportunity for a nutritional investigation
which might link free-living and parasitic forms.

MATERIALS AND METHODS

The strain of Tetramitits used in this investigation was obtained in December,
1950 by Dr. William Balamuth from Miss Lois Norman at the Communicable
Disease Center in Chamblee, Georgia. It had been found in a "sterile" urine
specimen in Austin, Texas in the same year. Routine cultures of the wild stock
were maintained at room temperature upon a medium consisting of 0.5% Difco
yeast extract plus 0.5% Difco peptone (YP) made up in distilled water. The
diphasic nature of this organism was first demonstrated by Bunting (1922, 1926),
who pointed out a reversible amoeba-to-flagellate transformation in its life cycle.
Although both stages of the present strain appeared in the bacterized maintenance
broth only the amoeboid form was present on solid agar-containing media or under
axenic conditions.

Handling of cultures

In order to obtain large numbers of bacteria-free amoebae for use in nutritional
studies it was thought advisable to establish the trophozoites in a two-membered
culture with a penicillin-sensitive bacterium. Washed cysts from the wild stock
of Tetramitus were first sterilized by a series of chemical procedures. These in-
cluded successive treatment at 23° C. with 1 : 50,000 HgCL for one hour and with
1 : 5000 KMnO4 for 30 minutes. The cysts were then implanted into a tube of YP
broth containing Micrococcus pyogcncs var. aiirens. Bacteria-free amoebae could be
obtained from this culture by inoculating 6- to 14-day-old trophozoites onto plates
of YP medium containing 1.5% agar plus 500 units/ml, of penicillin and incubating
them for 3 days at room temperature. Their sterility was routinely tested by

NUTRITION OF TETRAMITUS ROSTRATUS 271

inoculating them into Difco fluid thioglycollate medium and Difco stock culture
agar (SCA).

In assaying prepared media the general plan was as follows : Penicillin-treated
amoebae were gently flushed off the agar surface with sterile tap water and pooled
in a test tube. The organisms were inoculated in 0.1-ml. amounts into 150 X 18
mm. cotton-stoppered tubes containing 5 ml. of sterile test broth. The viability
of the treated amoebae used in each experiment was tested by inoculating them into
broth containing living bacteria. All cultures were incubated in a moist chamber
at 30° C. in a slanting position and observed at regular 2- to 4-day intervals for as
long as 10 to 14 days. Transplants were made in approximately 0.2-ml. amounts.
In order to eliminate carry-overs consideration was given only to the fourth sub-
culture when evaluating positive results. Positive cultures were always re-checked
for bacterial sterility in fluid thioglycollate medium and SCA.

As required in population runs, amoebae from broth cultures were counted upon
four hemocytometer fields and the results averaged.

Preparation of media

In axenic assays, initial experiments were designed to modify the concentration
of components of the original YP medium. In some cases substitution for the
yeast extract was made with Anheuser-Busch autolyzed or pepsin-digested yeast.
Protein digests including BBL trypticase, BBL phytone, Difco proteose-peptone
and Difco tryptone in concentrations from 0.1% to 3.0% were substituted for the
peptone fraction. More complex media involved the addition of Cerophyl (de-
hydrated cereal grass leaves), liver extract, cream, whole egg, blood and selected
vitamins to the basic YP medium in varying concentrations.

A review of the literature suggested types of media which have been used to
grow Protozoa axenically. Pressed-yeast juice was prepared according to the
method of Johnson and Baker (1942), in which they cultivated Paramecium multi-
micronucleatuin. Variations of their medium, sterilized by Selas filtration, con-
sisted of adding the concentrated juice to distilled water in amounts ranging from
3.0% to 50.0% in approximately two-fold concentrations. Cailleau (1933a, 1933b)
reported growth of A. castellanii upon two kinds of media. Her first medium
(1933a) and variations of her later medium (1933b) were tried. Trypticase, phy-
tone and peptone in concentrations from 1.0% to 3.0% were substituted for the
peptone fraction in her newer medium. Reich's medium (1935) for M. palestinen-
sis was also utilized in the screening procedures.

All media except Johnson's and Baker's pressed-yeast juice were sterilized by
autoclaving for 20 minutes at 15 pounds pressure. The hydrogen-ion concentra-
tions were not critically controlled but were adjusted when possible to approximate
neutrality.

Various species of dead bacteria were utilized as a food source in many of the
experiments. The organisms were grown upon YP agar in large petri dishes for
24 to 48 hours. After maximum growth was obtained the cells were scraped off
the plates, suspended in tubes of distilled water and autoclaved for 30 minutes at
15 pounds pressure.

272 MORGAN M. BRENT

Preparation of bacterial hydrolysates and extracts

It later became evident that certain substances present in the dead bacterial cells
(particularly in Bacillus cereus and B. subtilis) were required for growth of the
amoebae. The B. cereus cells were fractionated in the following manner : Thirty
grams of freshly harvested bacteria were added to an equal weight of alumina and
the aggregate ground by hand with dry ice. One hundred and sixty milliliters of dis-
tilled water were added to the crushed cells and the alumina centrifuged out. The
suspended cells were divided among three beakers in equal amounts and enough 1 N
HC1 and 1 N NaOH added separately to two of the containers to make 0.1 N
solutions of acid and base, respectively. The contents of the third beaker were
adjusted to pH 7.00. All three of the vessels were autoclaved for 30 minutes at
15 pounds pressure yielding acid-, alkaline-, and neutral-hydrolyzed fractions. The
contents from each beaker were then divided into two parts :

(a) One-half of the cell suspension from each beaker was subjected to dialysis
through a Visking casing for 28 hours at 4° C. in liter beakers with four changes of
cold distilled water. The dialysates were discarded for the purposes of these
experiments. After dialysis the pH of the three non-dialyzable 3 fractions contain-
ing the cell- residues was adjusted to neutrality. The cell-residues were collected
and washed and the non-dialyzable supernatants concentrated to 5 ml. by boiling.

(b) The pH of the remaining non-dialyzed acid, alkaline and neutral fractions
was also adjusted to neutrality; the cell-residues were collected and the non-
dialyzed supernatants concentrated in a similar manner.

All fractions were then re-autoclaved and stored at 4° C.

In later investigations it became necessary to extract the fat-soluble fractions
of the neutral-hydrolyzed non-dialyzed supernatants with various fat solvents.
This was accomplished by shaking each supernatant with an equal volume of
solvent, collecting the latter and repeating the procedure several times. The solvent
was evaporated to dryness and the sediment brought up to the original volume with
distilled water. Before extracting with alcohol or acetone it was first necessary
to evaporate the supernatants to dryness ; the insoluble materials were then centri-
fuged out.

Vitamins, purines, pyrimidines and ami-no acids

In experiments requiring special nutrients, mixtures of vitamins, purines,
pyrimidines and amino acids were added in place of certain bacterial fractions. A
stock solution of vitamins was prepared consisting of 50 mg. each of choline-HCl,
folic acid, inositol, nicotinic acid, paramino-benzoic acid, calcium pantothenate,
pyridoxine-HCl, riboflavine, thiamine-HCl and 25 /xg of biotin. The purine-
pyrimidine mixture consisted of 50 mg. each of adenine sulfate, cytidylic acid,
guanine, thymine, uracil and xanthine. Both of the above mixtures were separately
suspended in 500 ml. of distilled water, filtered through No. 03 Selas filters and
stored at 4° C. The amino acid stock mixture contained 50 mg. each of L-aspara-
gine, DL-alanine, L-cystine, L-glutamic acid, DL-methionine and DL-tryptophane.
The above acids were added to 50 ml. of distilled water and autoclaved for 10
minutes at 15 pounds pressure.

3 Non-dialyzablc refers to a retained fraction after being subjected to dialysis, as opposed
to non-dialyscd fractions which were not subjected to dialysis.

NUTRITION OF TETRAMITUS ROSTRATUS

EXPERIMENTS AND RESULTS

273

Selection of a medium

Experiments with most empirical formulae and those media already known to
support axenic growth of the free-living Protozoa described above, proved un-
successful for Tetnutiitits. Axenic growth of Tetramitus could be obtained, how-
ever, when washed autoclaved Bacillus cereus was added to a medium consisting
of 0.5% yeast extract, 0.5% peptone and 1.0% Wilson liver concentrate (N.F.). In
the yeast-peptone-liver medium (designated as YPL) sterile young amoebae were
2-3 times larger than normal (60 ^ in diameter) and usually appeared opaque and
immobile, but they became progressively more active as the cultures became older.
Through varying the pH from 5.5 to 7.9 it was found that better growth could be ob-

300

12

FIGURE 1. Comparison of growth of populations of T. restrains with living vs. autoclaved
B. cereus in 5 ml. of YPL medium at 30° C. Inoculum— 870/ml. •: Growth with living
bacteria; o: Growth with dead bacteria. Note: Each point represents a separate tube.

tained near neutrality (6.8-7.3). Altering the temperature from 20° C. to 35° C.
demonstrated that optimum growth (150 per mm.3) could be reached in approxi-
mately 12 days at 30° C.

A distinct lag phase was observed when the amoebae were grown with dead
B. cereus (Fig. 1), which was considerably longer than that obtained with living
bacteria. This would seem to indicate that although this medium is capable of
supporting growth and reproduction of the amoebae, it is far from perfect and is
incapable of supplying all the factors derived from cultures containing living bac-
teria. Preliminary experiments to shorten this lag with Selas-filtered, precon-
ditioned bacterial cultures have not proved promising.

Attempts were made to substitute other autoclaved bacteria for B. cereus in YPL
broth. Considerable growth of the amoebae could be obtained with killed Es-
cherichia coli, Neisseria catarrhalis, Sarcina lutea and Bacillus subtilis by culturing

274

MORGAN M. BRENT

at 8-day intervals. No growth could be obtained with autoclaved Pscudomonas
flu or esc ens. Substitution of autoclaved yeast Harris, Anheuser-Busch and Difco
whole yeast failed to replace this bacterial factor.

Although the liver portion of the medium could be reduced to 0.25% without
appreciably affecting growth, its complete elimination resulted in poor numbers
of amoebae. However, amoebae have been cultured in this medium (YP) for two
months with the addition of large amounts of autoclaved B. cereus or B. subtilis at
each sub-culture. The liver-deficient medium would not support growth of the

TABLE I

Growth-supporting capacity of fractions of Bacillus cereus for Tetramitus roslratus

in sterile yeast- peptone-liver medium

Neutral hydrolysis

Acid hydrolysis

Alkaline hydrolysis

Non-
dialyzed
super-
natant

Non-
dialyzable
super-
natant

Non-
dialyzed
super-
natant

Non-
dialyzable
super-
natant

Non-
dialyzed
super-
natant

Non-
dialyzable
super-
natant

-

-

-

-

-

-

Neutral
hydrolysis

Non-dialyzed
cell-residue

*

Non-dialyzable
cell-residue

—

+

—

—

—

Acid
hydrolysis

Non-dialyzed
cell-residue

+

Non-dialyzable
cell-residue

+

+

Alkaline
hydrolysis

Non-dialyzed
cell-residue

—

*

—

Non-dialyzable
cell-residue

—

—

* Present only in initial culture.
+ Growth of amoebae.
- No growth of amoebae.

Note: All supernatants were added as 1 pt. to 4 pts. of YPL medium,
excess (approx. 0.1 ml. wet mass).

Cells were added in

amoebae with added autoclaved E. coli or N . catarrhalis. Apparently B. cereus and
B. subtilis contain at least one or more factors found in liver extract which seem to
be lacking in the two non-spore-forming bacteria. These factors remain to be
elucidated.

Bacterial hydrolysates

Since something in the bacterial cell appeared necessary for sustained growth
of Tetramitus in YPL medium, partial hydrolysis of B. cereus was carried out as

NUTRITION OF TETRAMITUS ROSTRATUS 275

described, in order to separate possible essential fractions. It was found at the
outset that the substance (s) survived autoclaving for as long as 50 minutes, and a
thorough washing of the intact cells failed to remove any activity.

It can be seen from Table I that amoebic growth resulted in YPL medium with
the addition of the neutral-hydrolyzed non-dialyzable cell-residue to the neutral-
hydrolyzed non-dialyzed supernatant. Although the non-dialyzed supernatant was
routinely added as one part to four parts of YPL medium (by choice), it was
found that it would maintain trophic growth in one-half this amount as long as the
cell-residue was added. No growth could be obtained using the supernatant alone
even when raised to 50 % of the total medium. It can also be seen that the non-
dialyzable supernatant is completely inactive whether used alone or with the non-
dialyzable cell-residue. This residue still retained a small quantity of fat which
was demonstrated with the Sudan Black B stain of Burdon (1946). The results
obtained with the neutral-hydrolyzed fractions, therefore, would seem to indicate
that at least two factors from the bacterial cell are required for growth of the
amoebae : one dialyzable found in the bacterial supernatant and the other non-
dialyzable found in the cell-residue.

Macerated cells treated with 0.1 N HC1 and then subjected to dialysis retained
their activity. Microscopical examination of the cells revealed mostly disintegrated
cell husks with some fat retained in the debris. The non-dialyzed supernatant
from these cells (Table I) was inactive when utilized with the neutral-hydrolyzed
non-dialyzable cell-residue.

Alkaline hydrolysis of B. cercns seemed to destroy more than one factor re-
quired for growth of Tetramitus. Combinations of alkaline-hydrolyzed fractions
together or with neutral-hydrolyzed fractions were inactive. Examination of the
cell fragments revealed the absence of fat. When alkaline-treated fractions were
added with normal autoclaved whole cells growth ensued, demonstrating that there
is no toxicity factor involved here.

1. Preliminary attempts to replace the dialysable fraction

Utilizing the lead obtained from neutral hydrolysis, experiments were performed
to discover the nature of the factor (s) which were removed by dialysis. In these
experiments the non-dialyzable fraction was provided by the non-dialyzable cell-
residue. The possible dialyzable factors were sought separately in a mixture of
vitamins, a mixture of purities and pyrimidines and amino acids. The vitamin
mixture described above added in 0.002-ml., 0.02-ml. and 0.2-inl. amounts per 5 ml.
of YPL medium gave no indication of supporting growth. The mixture of purines
and pyrimidines added in 0.02-ml., 0.2-ml. and 2.0-ml. amounts to the vitamin-
containing medium also failed to maintain growth. Mixed amino acids added in
0.25-ml., 0.5-ml. and 1.0-ml. amounts in combination with the vitamins and the
purine-pyrimidine mixture gave no promising results.

In view of the fact that the substituted mixtures showed no activity, it was
postulated that non-dialyzable fractions in the bacterial supernatant were also re-
quired for growth. This immediately suggested lipoidal material. To test this
possibility, the non-dialyzable supernatant was utilized with the non-dialyzable cell
residue as non-dialyzable fractions. No growth of the amoebae resulted with the
addition of the vitamin mixture to these fractions in YPL medium. This was

276 MORGAN M. BRENT

re-investigated with the lipoid-extracted portions of the supernatant added as one
part to four parts of YPL medium. In no case was growth of Tetramitus observed
when eitlier benzene-, alcohol-, ether-, or acetone-extracted supernatant was added
to the vitaminized medium containing the non-dialyzable cell-residue.

2. Preliminarv at tempts to replace (lie non-dialyzable fraction

In routine investigation of the non-dialyzable fraction, neutral-hydrolyzed non-
dialyzed supernatant was added to every tube of YPL broth as one part to four
parts of medium ; and in addition the vitamin mixture was added in a concentration
of 0.1 ml. per 5 nil. of broth.

In order to eliminate the possibility that an essential metal might be tied up in
the bacterial residue, the cells were completely ashed by flaming them in a Pyrex
tube and then added to the culture medium. No growth of Tetramitus could be
obtained. The separate addition in 1.0% and 5.0% proportions of peptone,
tryptone, tryptose, proteose-peptone, yeast extract, trypticase and phytone failed to
replace this factor. Skim milk in concentrations of 0.01% to 0.5% has also failed.
Experiments are planned to replace this factor with known proteins and polysac-
charides.

DISCUSSION

Although only preliminary experiments have been carried out on the nutritional
requirements of Tetramitus, it has been shown that the dead bacterial cell provides
some essential constituent (s) for axenic growth of the amoebae in the yeast-peptone-
liver medium. The types of killed bacteria it can use in this medium are non-
specific since it has utilized gram-negative and gram-positive representatives of the
cocci group, gram-positive spore-formers and a gram-negative coliform. It ap-
pears, however, that although Tetramitus will grow in YPL medium with most
of the species of autoclaved bacteria investigated, only the spore-formers were
utilized by the amoebae in the liver-deficient medium. Since Bacillus cereus and
B. subtilis contained large quantities of fat, it is interesting to speculate whether
they provide certain lipoidal substances present in the liver extract. This hypothesis
should be tested by adding to the YP medium the fat-extracted portions of these
cells together with the autoclaved non-spore-forming bacteria.

The finding that Tetramitus can utilize certain heat-stable metabolites found in
microorganisms is not a new one when considering Protozoa in general. Johnson
(1936) was able to obtain growth of the holotrichous ciliate, Glaucoma ficaria, in
suspensions of 11 species of dead bacteria as well as 6 species of dead flagellates,
using a balanced salt solution as a basal medium. Glaser and Coria (1935) es-
tablished Paramecium caudatuni and P. multimicronucleatum as well as other
Protozoa upon a medium containing dead yeast cells as an indispensable nutrient.
Van Wagtendonk and Hackett ( 1949) secured axenic growth of Paramecium aurelia
but had to provide a 24-hour-preconditioned, autoclaved lettuce infusion culture
of A. aerogenes with autolyzed yeast. When either component was omitted growth
of the ciliates stopped, indicating essential substances other than the bacterial
fraction.

It is evident from the preliminary work on Tetramitus that this organism is not

NUTRITION OF TETRAMITUS ROSTRATUS 277

as fastidious in its growth requirements as the parasitic amoebae. The fact that
it can grow and multiply aerohically in the absence of other living protoplasm tends
to make the nutritional approach an easier one. With Entamoeba histolytica, on
the other hand, nutritional studies have been hampered by its dependency upon
associated living organisms and upon its extreme sensitivity to oxygen. Recently
a certain substance (s) in dead bacteria has also proved essential for this latter
species. Karlsson, James and Anderson (1952) have shown that when an auto-
claved culture of a streptobacillus in liver-proteose-peptone medium was used as a
substrate for E. histolytica under antibiotic treatment to suppress bacterial growth,
fair growth of the amoebae resulted. Use of filtered media resulted in the loss
of activity, suggesting the active material was present in the cells. Later Karlsson
(1952) showed that 90% of the cells' activity was destroyed during the first 20
minutes of autoclaving. In the present work on Tctramitus the bacterial fractions
are clearly heat-stable, since prolonged autoclaving for 50 minutes does not seem
to destroy their activity. The streptobacillus factor for Entamoeba was completely
destroyed by 0.1 N NaOH in 5 minutes at room temperature but could withstand
treatment with 0.1 N HC1 for one hour, suggesting similarity to those factors
found in the B. cere us cell. It is also interesting to note that Karlsson's strepto-
bacillus fraction could not be extracted with fat solvents.

Although no definite decision can as yet be made, the dialyzable factor in the
cell-extracted supernatant of B. ccreus suggests some protein fragment, for example,
a polypeptide or some other dialyzable substance of relatively low molecular weight.
There exists the possibility that several essential substances, both dialyzable and
non-dialyzable, may be present in the supernatant. Collection and analysis of the
dialysates would be of value in elucidating these fractions. The supernatant frac-
tion appears to remain with the cell-residue when subjected to acid hydrolysis
(Table I). Further evidence for this was shown by complete inactivity of the
acid-hydrolyzed non-dialyzed supernatant. Preliminary experiments have shown
that this acid fraction contained no inhibiting substance when added to acid-
hydrolyzed cells.

The non-dialyzable fraction in the cell residue would suggest substances either
proteinaceous or polysaccharide in nature. More complete analysis of this fraction
is required before arriving at any conclusions.

SUMMARY

1. Tetramitns was cultivated indefinitely under axenic conditions upon a
medium consisting of 0.5% Difco yeast extract, 0.5% Difco peptone and 1.0%
Wilson liver concentrate (N.F.) with selected types of autoclaved bacteria.

2. The liver concentrate could be eliminated with subsequent reduction in
trophic growth, only if killed B. ccreus or B. subtilis was utilized as the bacterial
fraction. Other bacteria (Ncisscria catarrhalis, Escherichia coli) could not be
substituted for these spore-formers in this medium.

3. It was found that the B. cereus cell contained at least two heat-stable frac-
tions necessary for growth ; neutral hydrolysis yielded a non-dialyzable fraction
found in the cell-residue and a dialyzable fraction found in the bacterial cell super-
natant.

4. Roth factors were stable to autoclaving with 0.1 N HC1 for 30 minutes at

278 MORGAN M. BRENT

15 pounds pressure and were retained in the acid-treated cell-residue. Alkaline
hydrolysis destroyed activity of all the fractions.

5. Preliminary attempts have failed to substitute for the factors found in the
neutral-hydrolyzed bacterial supernatant by employing selected vitamins, amino
acids, purines and pyrimidines.

LITERATURE CITED

BUNTING, M., 1922. A preliminary note on Tctramltus, a stage in the life cycle of a coprozoic

amoeba. Proc. Nat. Acad. Sci., 8: 294-300.
BUNTING, M., 1926. Studies of the life-cycle of Tctramitus restrains Perty. /. Morph., 42:

23-81.
BURDON, K. L., 1946. Fatty material in bacteria and fungi revealed by staining dried, fixed slide

preparations. /. Bact., 52 : 665-680.
CAILLEAU, R., 1933a. Culture d'Acanthamocba castcllanii en milieu liquide. C. R. Soc. Biol.,

Paris, 113: 990-992.
CAILLEAU, R., 1933b. Culture d'Acanthamoeba castcllanii sur milieu peptone. Action sur les

glucides. C. R. Soc. Biol., Paris, 114: 474-^76.
CASTELLANI, A., 1930. An amoeba found in cultures of a yeast: Third note. /. Trap. Mcd.

Hyg., 33 : 221-222.
DOUGHERTY, E., 1953. Problems of nomenclature for the growth of organisms of one species

with and without associated organisms of other species. Parasit., 42: 259-261.
GLASER, R. W., AND N. A. CORIA, 1935. The culture and reactions of purified Protozoa. Amcr.

J. Hyg. ,21: 111-120.
JOHNSON, D. F., 1936. Growth of Glaucoma ficaria Kahl in cultures with single species of other

microorganisms. Arch. f. Protistcnk., 86: 359-378.
JOHNSON, W. H., AND E. G. S. BAKER, 1942. The sterile culture of Paramecium multimicro-

nuclcata. Science, 95 : 333-334.
KARLSSON, J. L., 1952. Studies on the physical properties of a growth factor for Endamocba

histolytica. Amcr. J. Trap. Mcd. Hyg., 1: 548-551.
KARLSSON, J. L., M. B. JAMES AND H. H. ANDERSON, 1952. Studies on nutritional principles

for Endamocba histolytica in autoclaved bacterial cells. E.vp. Parasit., 1 : 347-352.
OEHLER, R., 1924. Weitere Mitteilungen iiber gereinigte Amoben- und Ciliaten-zucht. Arch.

f. Protistcnk., 49: 112-134.
RAFALKO, J. S., 1951. Mitotic division in the amoebo-flagellate, Tctramitus rostratus. J.

Morph., 89 : 71-90.
REICH, K., 1935. The cultivation of a sterile amoeba on media without solid food. /. Exp.

Zool, 69: 497-500.
RICE, N. E., 1935. The nutrition of Flabellula mira Schaeffer. Arch. f. Protistcnk., 85: 350-

368.
SHAFFER, J. G., 1952. Studies on the growth requirements of Endamocba histolytica. V.

Studies on the nature of some of the factors in the Shaffer-Frye medium that affect the

propagation of E. histolytica. Amer. J. Hyg., 56: 119-138.
SNYDER, T. L., AND H. E. MELENEY, 1943. Anaerobiosis and cholesterol as growth requirements

of Endamocba histolytica. J. Parasit., 29: 278-284.
STORM, J., S. H. HUTNER AND J. COWPERTHWAITE, 1951. Preliminary notes on the nutrition of

two small amoebae in pure culture. Proc. Amer. Soc. Protosool., 2 : 3.
VAN WAGTENDONK, W. J., AND P. L. HACKETT, 1949. The culture of Paramecium aurclia in

the absence of other living organisms. Proc. Nat. Acad. Sci., 35: 155-159.
WHERRY, W. B., 1913. Studies on the biology of an amoeba of the limax group. J'ahlkampfia

sp. no. I. Arch. f. Protistcnk., 31 : 77-94.
WILLIAMS, A., 1911. Pure cultures of amebae parasitic in mammals. /. Med. Res.. 25: 263-

283.

THE PENETRATION OF RADIOACTIVE PHOSPHATE
INTO MARINE EGGS x- ~

SUMNER C. BROOKS AND EDWARD L. CHAMBERS *
Department of Zoology, University of California, Berkeley, California

Needham and Needham (1930) showed that the gastrulae and plutei of the
echinoderm Dcndraster excentricus have a higher total phosphate content than that
of the unfertilized eggs. These authors suggested that the increased phosphate
content of the larvae was related to the formation of the skeletal spicules.

Brooks (1943a) obtained results, using radioactive phosphate, which indicated
that soon after first cleavage in the fertilized eggs of Arbacia punctulata, the intake
of radiophosphate was accelerated. During the winter of '46 to '47 these experi-
ments were repeated using more refined methods and the eggs of several different
species of sea urchins, as well as the eggs of the gephyrean worm, Urechis caupo.
The results are presented in this paper. Radiophosphate was found to enter the
fertilized eggs of sea urchins more than one hundred times faster than it entered the
unfertilized eggs (Brooks and Chambers, 1948). There was no evidence for
alternating phases of intake and loss of phosphate ions, such as have been reported
to occur during the early period of ion uptake by single Nitella cells and by uniform
populations of egg cells (Brooks, 1939a, 1939b, 1940, 1943a, 1943b). The
previously obtained results are to be ascribed to the considerable variability inherent
in the methods which had been used (Brooks, 1951).

Independently Abelson (1947), using the eggs of Arbacia punctulata, and
Lindberg (1948), using the eggs of Psaniniechinus miliaris, demonstrated the rela-
tively more rapid penetration of radioactive phosphate into the fertilized, as com-
pared to the unfertilized, sea urchin egg.

METHODS

Materials. Eggs shed from the ovaries of the freshly collected Pacific coast
sea urchins Strongylocentrotus purpuratus, S. franciscanus and Lytechinus pictus,
and eggs obtained from the "egg collectors" (MacGinitie, 1935) of the worm
Urechis caupo were used. The eggs were strained through cheese cloth and
washed four times by centrifugation. In each experiment eggs from only a single
animal were used, unless otherwise stated.

Conduct of the experiment. Egg suspensions containing 1 ml. eggs/liter sea
water were used. Impaired development occurs if the concentration exceeds 5-6

1 Presented at the 28th Annual Meeting of the AAAS, June, 1947, at San Diego, California.

2 This investigation was aided by a research grant (C-559) to the University of California
at Berkeley, from the National Cancer Institute of the National Institutes of Health, U. S. Public
Health Service.

3 Work done while Porter Fellow of the American Physiological Society, 1946-1947; now
at the University of Oregon Medical School, Portland 1, Oregon.

279

280

S. C. BROOKS AND E. L. CHAMBERS

nil. eggs/liter sea water. The pH of the sea water surrounding the eggs varied
from 8.0 to 8.2 throughout the duration of each experiment, and the temperature
of the egg suspensions was maintained at 15 ± 0.1° C., except that suspensions of
Lytechiniis pictns eggs were maintained at 20° to 21° C.

The eggs were kept suspended by using a stirrer rotated at 50 r.p.m. P32 of
high specific activity was added directly to the egg suspensions. The initial con-
centration of orthophosphate in the suspension fluid, after addition of PSJ, varied
from 50 to 434 /j.g P/liter (see Protocols). When thoroughly mixed, the homog-
eneous egg suspension was divided into two lots, one of which was inseminated by
adding one drop (0.05 ml.) of solid sperm, directly removed from the testis, to
1000 ml. of suspension. Examination of the eggs shortly after insemination showed
approximately two to three spermatozoa at the periphery of each egg. The re-
mainder of the experiment consisted of removing samples of both unfertilized and
fertilized eggs at frequent intervals. At the completion of the experiment, the

GIflSS ROD

MENISCUS

INCLUDED FLUID = V,
EGGS= ve

i'KE. 1. Diagram of Hopkin's tube containing sample of eggs.

eggs previously left unfertilized were inseminated. Development of these eggs, as
well as those inseminated earlier, was followed, without diluting the egg suspension,
through the pluteus stage for the sea urchin eggs and the trochophore stage for
Urccliis eggs. Development in every case was normal as compared to controls in
sea water. This indicated that the P3- had been used in concentrations, of the
order of 1 to 30 juc/liter, which were below the toxic level.

All experiments were rejected in which (1) the time from insemination to
50 per cent cleavage deviated appreciably from normal, (2) there was undue pro-
longation of the period between the time when the eggs first started to cleave to the
time when cleavage was completed, and (3) less than 95 per cent of the eggs
developed to normal swimming embryos.

Removal of egg samples for radioactivity measurements. Each sample of eggs
was taken by drawing up one- to two-mi, quantities of the well stirred homogeneous

PENETRATION OF PS-' INTO MARINE EGGS 281

suspension into a large lion- pipette with ;i wide mouth, and depositing the aliquots
in a Hopkin's vaccine tube (Fig. 1 j up to the 10.0-ml. mark. The tube was
then centrifuged at 86 X g for 60 seconds in a hand centrifuge. This force was
just sufficient to drive the eggs into the narrow end of the tube. Within 30 seconds
the supernatant was decanted and the fluid remaining within the narrow prolonga-
tion of the Hopkin's tube drawn off to a level just above the eggs, using a capillary
pipette. The tube was then immediately inverted, and the walls dried with filter
paper. The "end point" of penetration of isotope into the eggs was taken as being
at the end of the 60 seconds' centrifugation.

The total volume (Vt) of the eggs together with the suspension fluid con-
tained within the narrow prolongation was then determined (Fig. 1). This volume
(Vt) amounted to 0.03 to 0.05 ml. in the different experiments. The Hopkin's
tube was fixed in a holder fastened to the mechanical stage of a horizontally placed
low power microscope provided with an ocular hair line. By operating the stage,
the level of the meniscus could be read on the stage scale. Since each tube had
been previously accurately calibrated with mercury, the stage scale readings could
be converted directly into volume.

After completion of the reading, a thin glass rod with rounded ends was in-
serted into the tube in order to seal off the mouth of the narrow prolongation
(Fig. 1). By holding the rod in place with the index finger, any radioactive
solution adhering to the upper walls of the tube was washed out with distilled
water without disturbing the eggs at the bottom. After removing the rod, the
eggs, together with washings from the bottom of the tube, were transferred to a
flat nickel dish 3 cm. in diameter and 3 mm. deep, and dried. The dried material
formed a thin even layer on the bottom of the dish, amounting to no more than
1 mg. solids/cm.2 of surface. The radioactivity was measured using a Geiger-
Miiller tube, having a thin mica window 8 cm. in diameter. Samples of the decanted
supernatant fluid were dried in the identical dishes and the radioactivity measured.

The question arose as to how accurately the 10.0-ml. aliquots represent the
suspension as a whole. This was determined by taking a batch of unfertilized
eggs and removing the jelly by several washings. A dozen 10.0-ml. samples were
taken as above described in the Hopkin's tubes and centrifuged for ten minutes at
2000 X g. The top of the eggs packed in the narrow prolongation of the Hopkin's
tubes formed a perfectly straight line, and its level was measured as previously
described. The volumes thus obtained were within a maximum range of 0.2 per
cent of each other, indicating the validity of the sampling procedure used.

The advantage of the above described method for determining the quantity of
radioactive isotope in the eggs is that, by eliminating the necessity for washing the
cells, errors which might arise from injury to the eggs and from outward leaching
of ions or compounds are avoided.

Egg volume measurements. The mean diameter of fertilized eggs in the early
one-cell stage was determined by averaging 25 individual measurements made with
a filar micrometer. The unfertilized eggs of the sea urchin are never spherical
when freshly removed from the ovaries, and Urechis eggs in the unfertilized state
are indented on one side. Soon after fertilization the eggs of both species become
spherical with only slight changes in volume (Tyler, 1932). The average diameter
of 5. purpuratus eggs is 81.3 //,, S. franciscanus eggs 119 p, and Urechis caupo eggs

282 S. C. BROOKS AND E. L. CHAMBERS

110 /A. The number of eggs per ml. of suspension was determined as follows.
Using a wide-mouthed pipette a sample, approximately 0.2 ml. in volume, was
withdrawn from the homogeneous suspension, deposited on a ruled slide, cov-
ered with a coverslip, and weighed in order that the volume of the sample could
be accurately determined. The total number of eggs on the slide was then counted.
This procedure was repeated twice and the results averaged. Knowing the
number of eggs in a given mass of sea water and the average diameter of one egg,
the total volume of egg protoplasm (Ve} in a 10.0-ml. volume of suspension could
be readily calculated.

Egg volume determinations were also carried out by centrifuging the jelly-free
unfertilized eggs for 10 minutes at 2000 X g. Results obtained by this method
were not significantly different from volume determinations carried out by counting
the number of eggs in aliquots and measuring diameters.

Calculations. The concentration of P32 within the eggs was calculated as fol-
lows. Knowing the total volume of eggs with included fluid (F/) contained within
the narrow prolongation of the Hopkin's tube (Fig. 1), and the volume of eggs in
10.0 ml. of suspension (Fe), the volume of the included fluid alone is: (Fj) =
(F, ) — (F<?). The included fluid volume (Fj) represents the quantity of fluid
exterior to the protoplasmic surface of the eggs. This volume, therefore, includes
the space occupied by the egg jelly and the space enclosed within the fertilization
membrane, structures which allow free diffusion of phosphate ions. The radio-
activity of the included fluid is obtained by multiplying (Ff) X C.P.M. (counts per
minute) of one ml. suspension fluid. Since the radioactivity of the total sample
is known, the radioactivity of one ml. of eggs is :

(C.P.M. of sample) - (C.P.M. of included fluid)
C.P.M./ml. eggs = - ^— A (1)

' e

Under the conditions of the counting method used, 1 C.P.M. = 3.6 X 10~G
,u,c P32. Using this conversion factor, the results have been expressed in terms of
/AC P32/ml. eggs.

Accuracy of method. The following sources of error were taken into consider-
ation in calculating the standard deviation for each determination of the /AC P32/ml.
eggs (see Tables I and II) : the sampling error, error in determining the egg volume
(Fe) and the included fluid volume (Fi), error in determining the C.P.M. of the
sample, of the background, and of the supernatant fluid. When the radioactivity
contributed by the included fluid in each sample is more than half that of the entire
sample, the error of the method is considerable. Accuate results are obtained when
the concentration of isotope in the eggs exceeds the concentration of isotope in the
suspension fluid.

PROTOCOLS

Experiment 1. P32 was added to a suspension containing 1.00 ml. S\ purpiiratits
eggs/liter 120 minutes after the eggs had been removed from the ovaries, and the
eggs were inseminated five minutes later. The initial concentration of P32 in the
sea water was 1.22 /Ac/liter, and the orthophosphate concentration approximately
71 /Ag P/liter. First cleavage started 97 minutes after insemination, was 50 per cent
complete at 104 minutes and finished at 110 minutes.

PENETRATION OF

INTO MARINE EGGS

283

Experiment 2. P32 was added to a suspension containing 1.09 ml. 5\ purpuratus
eggs/liter 135 minutes after removal of the eggs from the ovaries, the eggs being
inseminated 12 minutes later. The initial concentration of P32 in the sea water

IJO

3
O
O

EXPT. I

2nd Cieovoge

3rd Cieovoge

"° -„ 100 ml Suspension Fluid of Fertilized Eggs

Unfertilized Eggs ° "'IT-——! «

-5*10 30 60 90 120 150 180 210 240

Time in Minutes After Inseminotion

270

30 -

28

26-

24

2.2

18

<D 1.6
CL

EXPT. 2

Cieovoge

100ml Suspension Fluid
of Fertilized Eggs

-12 0 20 40 60 80 100 120 140 160 ISO

Time in Minutes After Inseminotion

58
54
50
46

</>42
0>

IU38

E

0>

in

3
£24

5 20
1.6
1.2
S
.4

EXPT. 3

2nd Cieovoge/

100ml Suspension Fluid
of Fertilized Eggs

P"odded
I Sperm
• Jodded

330

-55 *IO 30 50 70 90 110 130 150 170 190

Time in Minutes After Insemination

.70

60

.50

o>

O-

30

o
o

EXPT. 4

'nd Cleavage

Cieovoge

Cieovoge

100 ml Suspension Fluid of Fertilized Eggs

• » „

Unfertilized Eggs

-10 0 20 40 60 80 100 120 140 160 180 200 220 240 260

Time in Minutes After Inseminotion

FIGURE 2. Uptake of P32 by unfertilized and fertilized sea urchin eggs. Expts. 1, 2 and 3 :
S. purpuratus eggs. Expt. 4 : S. franciscanus eggs. At beginning of each experiment P32
added to suspension and then divided into two lots. Sperm added to one lot at 0 minutes in
all experiments — fertilized eggs. The second lot was not inseminated — unfertilized eggs.
Ordinates : MC P32/ml. eggs. Abscissae : Time in minutes before and after insemination, 0 —

time of insemination. (O O) fertilized eggs, (•- -•) unfertilized eggs,

MC P32/100 ml. suspension fluid of fertilized eggs.

284

S. C. BROOKS AND E. L. CHAMBERS

TABLE I

Uptake of P32 by the unfertilized and fertilized eggs of S. purpuratus. Expts. 1, 2, and 3

Time after
insemination
in minutes

Unfertilized,
/iC P32/ml. eggs

Fertilized,
tic P32/ml. eggs

Suspension fluid of
fertilized eggs,
/ic P'Vml.

Expt. 1, S. purpuratus

-5.0

P32 added to suspension

0.0

One-half of suspension inseminated

0.00122

8.3

0.00184 ± .0009

0.00111 ± .0009

0.00122

59.6

—

0.175 ± .003

0.00103

92.4

0.00810 db .0009

0.359 ± .004

0.00085

173.8

- —

0.756 ± .006

0.00048

225.1

—

1.01 ± .01

0.00024

266.5

0.0188 ± .0010

—

Expt. 2, 5. purpuratus

-12.0

P32 added to suspension

-8.2

0.034 ± .002

0.0

One-half of suspension inseminated

0.00885

21.0

0.046 ± .002

0.047

± .003

0.00872

56.0

0.050 ± .002

0.494

± .009

0.00826

84.4

— .

1.27

db .09

0.00770

87.1

0.063 ± .002

—

—

127.2

—

2.09

± .14

0.00665

152.0

0.069 ± .003

—

—

171.1

—

3.02

± .16

0.00560

Expt. 3, S. purpuratus

-5.5

P32 added to suspension

-2.5

0.0809 ± .0064

0.0

One-half of suspension inseminated

0.0335

18.0

0.0924 ± .0061

0.0444

± .0044

0.0333

46.6

—

0.496

± .071

0.0328

81.5

0.102 ± .007

1.68

± .09

0.0316

122.5

0.119 ±.006

—

128.0

—

3.46

± .10

0.0294

179.2

—

5.79

± .12

0.0270

was 8.9 /uc/liter, and the orthophosphate concentration approximately 133 ^g P/
liter. First cleavage started at 113 minutes after insemination, 50 per cent had
cleaved at 118 minutes and completed at 123 minutes.

Experiment 3. P32 was added to a suspension containing 1.10 ml. S. purpuratus
eggs/liter 127 minutes after the eggs had been removed from the ovaries, and the
eggs were inseminated 5.5 minutes later. The initial concentration of P32 in the sea
water was 33.5 juc/liter and the orthophosphate concentration approximately 434
/xg P/liter. First cleavage started 107 minutes after insemination, was 50 per cent
complete at 114.5 minutes and was finished at 121 minutes.

Experiment 4. P32 was added to a suspension containing 0.72 ml. .V. jranciscanus

PENETRATION OF P3- INTO MARINE EGGS

285

eggs/liter 120 minutes after the eggs had been removed from the ovaries, and the
eggs inseminated 10 minutes later. The initial concentration of P3 in the sea
water was 1.2 ju.c/liter, and the orthophosphate concentration approximately 59 //.g
P/liter. First cleavage started 108 minutes after insemination, was 50 per cent
complete at 113 minutes and reached completion at 117 minutes.

Experiment 5. P3i was added to a suspension containing 1.29 ml. U. canpu
eggs/liter 180 minutes after removal of the eggs from the animal, and the eggs were

TABLE II

Uptake of P32 by the unfertilised and fertilized eggs of S. franciscanus and Urechis caupo.

Expts. 4 and 5

Time after
insemination
in minutes

Unfertilized,
MC P'Vml. eggs

Fertilized,
MC P»/mI. eggs

Suspension fluid of
fertilized eggs,
j»c Pw/ml.

Expt. 4, S. franciscanus

-10.0

P32 added to suspension

0.0

One-half of suspension inseminated

0.00118

6.3

0.0031 ± .0013

0.0012 ± .0012

0.00118

74.8

0.0038 ± .0014

0.127 ± .005

0.00109

108.2

0.0075 ± .0015

0.240 ± .005

0.00100

163.2

—

0.435 ± .007

0.00086

206.7

0.0094 ± .0015

0.600 d= .009

0.00076

252.0

— -

0.718 ±.011

0.00066

Expt. 5, Urechis caupo

-10.0
-5.0

P32 added to suspension
0.0007 ± .0004

0.0
33.3

One-half of suspension inseminated

0.0028 ± .0005

0.00089
0.00089

43.0

0.0039 ± .0006

—

—

91.5

—

0.0066 ± .0005

0.00088

100.8

0.0056 ± .0006

—

—

176.9

—

0.0145 ± .0006

0.00087

184.1

0.0103 ± .0007

—

—

223.2

—

0.0220 ± .0007

0.00085

266.0

—

0.0369 ± .0010

0.00083

275.0

0.0174 ± .0007

—

—

517.0

—

0.285 ± .005

0.00052

inseminated 10 minutes later. The initial concentration of P32 in the sea water was
0.89 /xc/liter, and the orthophosphate concentration about 50 /Ag P/liter. First
cleavage was 50 per cent complete at 119 minutes after insemination, second
cleavage at 160 minutes and third cleavage at 209 minutes.

RESULTS
Uptake of P3- by unfertilised and fertilised eggs

The results of five experiments are presented in Figures 2 and 3. The data
are abbreviated in Tables I and II, in which only a few of the determinations are

286

S. C. BROOKS AND E. L. CHAMBERS

presented for each experiment. The first column in each table indicates the time
when samples of unfertilized and fertilized eggs were removed for radioactivity
measurements. P3L> was added to the suspension of the unfertilized eggs at the
beginning of the experiment. Shortly thereafter the first sample was removed,
the suspension divided into two lots, and one lot inseminated. In every case the
time of insemination is set as zero time. In the second column the quantity of P3-
which has penetrated the unfertilized eggs after various time intervals is shown.
The third column presents the same data for the fertilized eggs. In the fourth
column the decrease in concentration of P32 in the suspension medium is shown.

Eggs of Strongylocentrotus purpuratus and S. franciscanus. The uptake of P32
by the unfertilized eggs is shown in Figure 2, interrupted line with solid circles.

10 ml Suspension Fluid
of Fertilized Eggs

.005 =,

-10 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280

Time in Minutes After Insemination

FIGURE 3. Uptake of P:12 by the unfertilized and fertilized eggs of Urcchis caupo, Expt. 5.

Legend as for Figure 2.

In Experiments 2 and 3 the initial uptake of P32, during the first three to four
minutes after adding the isotope, appears to be greater than the uptake in the
subsequent 200 minutes. This initial increase, however, undoubtedly does not
represent penetration of P32 into the eggs, but is an artifact arising from a small
error in determining the absolute volume of the eggs, or from the initial absorption
of a small quantity of the isotope to the extraneous coats or the surface of the eggs.
In Experiments 1 and 4, in which the concentrations of P (as orthophosphate) and
P32 in the suspension fluid were much less than in Experiments 2 and 3 (see
Protocols), the absence of an initial phase of rapid P32 uptake is evident. In all

PENETRATION OF P^ INTO MARINE EGGS 287

experiments, with the exception of the initial phase in Experiments 2 and 3, the
uptake of P32 by the unfertilized eggs occurred at a constant rate.

The uptake of P32 by fertilized eggs (Fig. 2, solid line with open circles) during
the first 10 to 15 minutes following fertilization is identical to the uptake shown
by unfertilized eggs. By 15 to 20 minutes following fertilization the rate of uptake
increases until by 50 to 60 minutes in 5". purpnratus, and 80 to 90 minutes in 6".
jranciscanus the uptake has reached a maximum rate. Thereafter, except for
minor variations, the uptake occurs at a constant rate through the third cleavage.
The minor variations in the rate of uptake which occurred were within the range
of experimental error. No change in rate of uptake during the cleavage cycles
was evident.

Results obtained using the eggs of Lytcchinus pictus are essentially identical
to those obtained using the eggs of the two species of Strongylocentrotus (Chambers
and Whiteley in Whiteley, 1949).

Eggs of Urcchis caupo. The uptake of P3'- by the unfertilized eggs (Fig. 3,
interrupted line with solid circles) was observed to occur at a slow constant rate.
The uptake of P32 by the fertilized eggs (Fig. 3, solid line with open circles) was
essentially identical to that of the unfertilized eggs throughout the period of
maturation and the first two cleavages (the eggs, laid in the germinal vesicle stage,
do not mature until after insemination occurs). Shortly after the second cleavage
the rate at which the fertilized eggs removed P32 from the medium increased. Even
after the fourth cleavage the rate continued to increase.

Loss of P32 from the eggs

Experiments wrere carried out to determine whether or not P32 contained within
the eggs is lost to the medium when the P32 in the sea water surrounding the
eggs is removed. The experiments were carried out using the eggs of Lytechinns
pictus as follows : Unfertilized and fertilized eggs were exposed to sea water con-
taining P32 and approximately 60 jug P as orthophosphate/liter for one hour.
Samples of the suspension were then taken to determine the quantity of P32 which
had entered the eggs, and immediately thereafter the remainder of the suspension
was gently centrifuged, the supernatant decanted, and replaced by fresh non-
radioactive sea water containing about 60 ju,g P as orthophosphate/liter. After
washing three times by centrifugation, a suspension of the washed eggs was pre-
pared containing 3.0 ml. eggs/liter sea water. Ten-mi, samples of this suspension
were removed at various intervals of time for radioactivity determinations. The
washing of the eggs was repeated at frequent intervals, in order to remove any P32
which may have entered the medium from the eggs. The results are shown in
Figure 4. The quantity of P32 remaining within the eggs is expressed in terms of
the per cent of the quantity of P32 within the eggs immediately preceding the first
washing. The times when the eggs were washed are indicated by the small arrows
in the figures. During careful washing of the eggs, in spite of the greatest pre-
cautions it is impossible to avoid destroying or losing some of the eggs. This is
particularly true of the fragile unfertilized eggs. Accordingly, the volume of un-
fertilized eggs in each 10-ml. sample taken was determined by the centrifugal
method, and correction made for any loss of eggs which may have occurred during
the repeated washings.

288

S. C. BROOKS AXD E. L. CHAMBERS

The results obtained on the unfertilized eggs (Fig. 4, curves in upper half) reveal
that in two experiments 15 per cent, and in one experiment 4 per cent of the P3-
initially contained within the eggs was lost to the medium during the first 100
minutes after washing was started. The P3- continues to be lost at a slow rate
over a long period of time. However, the results obtained after 300 minutes from
the time the eggs were first washed are open to question, because fertilization and
development of these eggs were impaired.

The fertilized eggs are far more resistant to the washing procedure, since they
are protected by their fertilization membranes. It was not possible, however, to
correct for such losses of eggs as may have occurred, during washing, as egg volumes

o
o

UJ

.Unfertilized Eggs

Ill

^ IUU

z
o

!HS^: — *^r

— ^

>-

®T5:::^==^_ \

£ 8°

h-

^ ——A

— - — n ;

z

UJ

' =Gb^

T^Of-

0

z

^^Ferlillzed Eggs-^

0 60

^ Ul X \

1 '00

r.y0nn i i \

^x ^^ U V

-I

}_

t ^- CO

rt|®| | |

T

i 80

-

b.

0

h-
z 60

It 1

-

tr

UJ

Q.

0 50 100 150 200 250 300 350 400

TIME IN MINUTES AFTER EGGS SUSPENDED IN NON-RADIOACTIVE SEA WATER

FIGURE 4. Effect of washing unfertilized and fertilized Lytcchinns Rictus eggs containing P32

in non-radioactive sea water.

in the samples cannot be determined when the eggs are fertilized. The results of
two experiments on fertilized eggs are shown in Figure 4 (curves in lower half).
A loss of P3- from the samples taken after the first series of washings occurred,
but subsequently, no appreciable loss of P3- from the eggs was observed. The
initial loss of P32 is undoubtedly due to the loss of some eggs, for which no correction
could be made.

In six experiments carried out using unfertilized and fertilized S. purpnratns
eggs, similar results were obtained.

The influence of extraneous coats on the uptake of P32

The uptake of P32 by the fertilized eggs of Lytechinus pictits from which the
extraneous coats had been removed was compared to that of normal eggs possessing

PENETRATION OF P32 INTO MARINE EGGS 289

all their coats intact. The experiment was performed as follows : A suspension of
unfertilized eggs was divided in two beakers. The eggs in one beaker were
inseminated. One-half of this suspension was centrifuged, the supernatant dis-
carded and two minutes after insemination, at the time when observation revealed
that the fertilization membranes were rising, the eggs were suspended in a mixture
of 95 parts 1 M urea, pH 8.0, and five parts sea water. The eggs were allowed to
settle, and five minutes after insemination the urea solution was decanted and
replaced by sea water. The decantations were repeated several times until the
eggs had been washed free of the urea solution. Observation of the eggs revealed
that the jelly coats and the fertilization membranes had been completely removed
and that the hyaline plasma layer did not form (Moore, 1930). When the urea-
treated eggs cleaved, they separated into twro spherical blastomeres, fastened to-
gether only by delicate stalks. The fertilized untreated eggs and the urea-treated
eggs cleaved 100 per cent and at the same time. At 50 minutes after insemination,
P32 was added to all three suspensions, the control unfertilized eggs, the control
fertilized eggs, and the urea-treated fertilized eggs. Samples were removed at
intervals for radioactivity determinations. The results are shown in Table III.
They reveal that in the 25-minute interval during which the uptake of P32 was
measured, 75 times as much P32 entered the fertilized eggs as the unfertilized eggs,

TABLE III

Uptake of P32 by unfertilised, normal fertilized and urea-treated fertilized Lytechinus pictus eggs

Uptake of P32,
Condition of eggs in C. P.M. /ml. eggs

Unfertilized 2,000

Fertilized, controls 150,000

Fertilized, urea treated 145,000

and that as much P32 entered the denuded urea-treated eggs as the control fertilized
eggs. Since the urea-treated eggs in sea water may still have possessed a thin but
invisible coating of proteinaceous material, at the end of the 25-minute interval
these eggs were washed for 10 minutes in three changes of an isotonic mixture of
10 parts 0.53 M KC1 and 90 parts 0.52 M NaCl at pH 6.0 and then suspended in
this mixture. Such a treatment should have dissolved away any remaining ex-
traneous material surrounding the eggs (Chambers, 1940). The washing of the
urea-treated eggs in the Na/K mixture did not remove an appreciable quantity of
P32 from the eggs, revealing that no significant quantity of P32 is absorbed to the
coats, which surround the eggs external to the protoplasmic surface.

Rate of penetration into the eggs

The quantity of P32 in /AC entering one ml. eggs in a given interval of time,
t2 — tlt may be obtained directly from the graphs. Rates of penetration were
determined only for the constant phases of uptake, i.e., from five minutes (^) to
200 minutes (t2) after addition of P32 for the unfertilized eggs, from 70 minutes
(^) to 170 minutes (f2) after insemination for the fertilized sea urchin eggs, and
from 250 minutes (#t) to 275 minutes (f2) after insemination for the fertilized
Urechis eggs.

Tn spite of a very considerable decrease in the concentration of P32 in the medium

290 S. C. BROOKS AND E. L. CHAMBERS

surrounding the fertilized sea urchin eggs, the uptake of P3'- by the eggs remained
constant throughout the duration of Experiments 1-4. In experiments on the
unfertilized eggs, and the fertilized eggs of Urechis caupo, no appreciable change in
concentration of P32 in the suspension fluid occurred, since only a small quantity of
P32 penetrated the eggs.

Fertilised eggs of S. purpuratus and S. franciscanus. The decrease in concen-
tration of P32 in the medium surrounding the fertilized eggs could be due either
to an exchange of P32 from the external medium for P inside the eggs, or to an
accumulation of P within the eggs, depleting the P in the suspension fluid. Experi-
ments described in this paper reveal that when fertilized eggs, which had been
exposed to sea water containing P32, are immersed in radioactive-free sea water con-
taining about 60 /Ag P/liter no appreciable quantity of P32 leaves the eggs. Cham-
bers and White (1949, 1954) have shown that fertilized eggs remove ortho-
phosphate from a medium containing between 10 to 100 /Ag P/liter at the same rate
as the P32.

The specific activity (/AC P32//Ag P) of the P (as orthophosphate) in the medium,
therefore, remains constant throughout the duration of the experiment. Accord-
ingly, the rate of penetration of orthophosphate into the eggs can be accurately
calculated as follows :

,, s e e

P/ml. eggs/mm. = — - T - ' <2>

where //,g P8 = initial concentration of orthophosphate in the suspension fluid in
/Ag P/ml., /AC Ps32 = initial concentration of P32 in /AC/ml. in the suspension fluid,
/AC Pe32 at t2 and ^ = concentration of P32 in //.c/ml. eggs at the time in minutes t2
and £v

The rates of penetration of orthophosphate, calculated according to equation (2)
are shown in Table IV, including three experiments from Chambers and White
(1954.)

For the fertilized eggs of 5\ purpuratus, the rate of P uptake, from 70 to 170
minutes after insemination, at 15° C., in four experiments, varied from 0.54 /Ag
P/ml. fertilized eggs/minute at an external orthophosphate concentration of 416
/Ag P/liter to 0.28 /Ag P/ml. fertilized eggs/minute at an external orthophosphate
concentration of 20 /Ag (P/liter (Table IV, columns 3 and 5). With a twenty-fold
change in concentration of P, only a 1.9-fold change in the rate of penetration of
P occurred. As long as the orthophosphate concentration exceeds about 20 /Ag
P/liter, the more dilute the orthophosphate concentration, the greater is the fraction
of P in the medium which is absorbed by the fertilized eggs in a given period of
time.

The rate of penetration of orthophosphate into fertilized Strongylocentrotus
franciscanus eggs from 70 to 170 minutes after insemination, at 15° C., in three
experiments, ranged from 0.11 to 0.17 /Ag P/ml. fertilized eggs/minute, with the
concentration of orthophosphate in the external medium varying from 63 to 20 /Ag
P/liter (Table IV, columns 3 and 5). The rate of uptake by the fertilized S.
franciscanus eggs is about half that of S. purpuratus eggs.

Unfertilized eggs of S. purpuratus and S. franciscanus. In the experiments
carried out with unfertilized eggs, no change could be detected in the concentration

PENETRATION OF P3-' INTO MARINE EGGS

291

of P3- in the suspension fluid. Although the concentration of orthophosphate in
the sea water surrounding unfertilized sea urchin eggs remains constant or slowly
increases (Chambers and White, 1954), the amount of increase is not sufficient to
appreciably alter the specific activity of the orthophosphate in the medium as long
as the egg suspension is dilute (1.0 ml. eggs/liter suspension) and the concentra-
tion of orthophosphate in the medium exceeds 50 ,ug P/liter.

Unlike fertilized eggs, unfertilized eggs containing P32, when immersed in non-
radioactive sea water, slowly lose P32 to the surrounding medium. The rate, how-
ever, at which P32 is lost from eggs which had been exposed to P32 sea water is
over a hundred times slower than the rate at which the P32 originally entered the
eggs. In view of these considerations and the linear P32 uptake curves, equation

TABLE IV

Rate of penetration of phosphate into unfertilized and fertilized S. purpuratus, S. franciscanus

and Urechis caupo eggs at 15° C.

Experiment

Condition
of eggs

Initial and
final P cone.
^ P/liter
susp. fluid

^g P/ml. eggs/min.

Fertilized,
fig P/ml. eggs/min.

Unfertilized

Fertilized

Unfertilized,
Mg P/ml. eggs/min.

S. purpuratus,

Unfert.

71

0.0035

Expt. 1

Fert.

56 to 29

0.30

86

5. purpuratus,

Unfert.

133

0.0035

Expt. 2

Fert.

122 to 84

0.37

106

5. purpuratus,

Unfert.

434

0.0041

Expt. 3

Fert.

416 to 359

0.54

132

5. purpuratus*

Unfert.

78

0.0026

Fert.

78 to 20

0.28

106

S. franciscanus,

Unfert.

59

0.0015

Expt. 4

Fert.

55 to 42

0.17

113

5. franciscanus^

Fert.

53 to 20

0.11

—

S. franciscanus\

Fert.

63 to 20

0.14 ot 0.17

—

U. caupo,

Unfert.

50

0.0033

Expt. 5

Fert. §

47 to 46

0.024

7

* From Chambers and White (1954), Expts. 5 and 6.
t From Chambers and White (1954), Expt. 7.
j From Chambers and White (1954), Expt. 8.
§ After third cleavage.

(2) may be used to calculate the rate of entry of orthophosphate into unfertilized sea
urchin eggs.

In the case of the unfertilized eggs of >5\ purpuratus the rate of penetration of
orthophosphate at 15° C. in three experiments varied between 0.0041 to 0.0026
/Ag P/ml. unfertilized eggs/minute with concentration of P (as orthophosphate) in
the medium ranging from 434 to 71 p.g P/liter (Table IV, columns 3 and 4). For
the unfertilized eggs of S. franciscanus the rate of P uptake, in one experiment,
was 0.0015 fug P/ml. unfertilized eggs/minute, at an external orthophosphate con-
centration of 59 .fig P/liter.

Fertilised and unfertilised Strongylocentrotus eggs compared. As shown in
Table IV, column 6, phosphate penetrates fertilized S. purpuratus eggs 86 to 132
times more rapidly than unfertilized eggs.

292 S. C. BROOKS AND E. L. CHAMBERS

In 6". franciscanns eggs, approximately a 113-fold increase in rate occurs after
fertilization.

Eggs of Urechis caupo. The rates of penetration of orthophosphate into the
unfertilized and fertilized eggs were arbitrarily calculated according to equation (2).
In view of the slow rate of P32 uptake, the inappreciable change in concentration of
P32 in the suspension fluid, and the dilute egg suspension, it is probable that the
use of equation (2) is justified.

Phosphate penetrates the unfertilized eggs of Urechis caupo and 5. purpuratus
at about the same rate (Table IV, columns 3 and 4). The important difference
between the Urechis and the sea urchin egg is that in the former species, after
fertilization, no increase in the rate of P uptake occurs. After the Urechis eggs
have undergone second cleavage, however, the rate starts to increase, attaining a
rate 7 times that of the unfertilized eggs by the time of fourth cleavage (Table IV,
column 6). Even by this time the rate has not reached its maximal level (Fig. 3).

DISCUSSION

When fertilized sea urchin eggs which had been exposed to sea water containing
P32 are immersed in radioactive free sea water, no appreciable quantity of P32 leaves
the eggs. Furthermore, fertilized sea urchin eggs remove orthophosphate from
sea water at the same rate as P32 (Chambers and White, 1949, 1954). These data
reveal that the entry of P32 into the fertilized sea urchin eggs measures the rate at
which phosphate accumulates within the cells.

On the other hand, when P32 is added to a suspension of unfertilized eggs, no
appreciable change in concentration of P32 occurs in the suspension fluid. The
concentration of orthophosphate in the suspension fluid surrounding unfertilized
eggs remains constant or slowly increases (Chambers and White, 1954). When
unfertilized eggs, containing P32, are immersed in non-radioactive sea water, P32
slowly washes out into the external medium. On the basis of these data, the
conclusion may be made that the uptake of P32 by unfertilized eggs measures the
rate at which phosphate enters the eggs, presumably by exchange, at the same time
that the internal concentration remains constant, or even decreases. The change
from the unfertilized to the fertilized state, therefore, involves not only a change in
magnitude but also a reversal of "driving forces." Although the two-fold increase
following fertilization in permeability to water (Lillie, 1916) and non-electrolytes
(Stewart and Jacobs, 1932) may contribute to the striking increase in uptake of
orthophosphate which follows fertilization, it is probable that changes in "driving
forces" play the dominant role.

Of interest is the relatively constant rate at which phosphate accumulates within
fertilized sea urchin eggs, irrespective of large changes in concentration in the ex-
ternal medium. This resembles the constancy in the rate of oxygen consumption
of cells, over a wide range of different oxygen tensions, as long as the tension
exceeds a certain minimal value. Apparently the primary factor which determines
the rate of phosphate entry into fertilized eggs is the rate at which phosphate is
bound or combined within the cells.

The question arises as to the importance of phosphate in sea water for the
development of the eggs. Both Loeb (1907) and Herbst (1898) reached the
conclusion that phosphate in the medium is not necessary for normal development.

PENETRATION OF P™ INTO MARINE EGGS 293

However, these investigators used artificial sea water prepared from the individual
salts, which had not been specially purified, and the only criterion for the absence of
phosphate was the lack of a positive test with a molybdate method which was far
too insensitive. Herbst used highly dilute sea urchin egg suspensions in the order
pf several drops of eggs to a beaker of sea water. In view of the data presented
in this paper, under such conditions even a trace of phosphate would have been
sufficient to adequately supply the eggs. Eggs with a very low intracellular in-
organic phosphate content, such as the eggs of S. purpuratus (Chambers and White,
1949) and 5". drobachiensis (Chambers and Mende, 1953a, 1953b), may have more
need for an external source of phosphate than eggs with a high content of inorganic
phosphate, such as Arbacia punctulata (Chambers and Mende, 1953b), and Para-
centrotus lividns (Zielinski, 1939).

On the basis of experiments in which the rate of P32 uptake by S. purpuratus
spermatozoa was measured in a suspension containing 1.0 ml. solid sperm/liter
sea water, it was found that one ml. of solid sperm takes up P32 more slowly than
a corresponding volume of unfertilized eggs (Chambers and White, unpublished
data). Accordingly, the amount of P32 which would enter the few excess sperma-
tozoa attached to the outer surface of fertilized eggs is infinitesimal, compared to the
amount actually found to enter the eggs.

Relation to oxygen consumption. The slow rate of P32 uptake by unfertilized
sea urchin eggs (two to six hours after removal from the ovaries) is observed during
the period when the eggs would show a fairly constant and low rate of oxygen
consumption (Borei, 1948, 1949). The prominent increase in the rate of uptake
of P32, which occurs after sea urchin eggs are inseminated and represents the ac-
cumulation of phosphate within the fertilized eggs, takes place during a period when
the rate of oxygen consumption increases markedly (Borei, 1948; Tyler and
Humason, 1937).

Laser and Rothschild (1939) describe a marked increase in the rate of oxygen
consumption of Psammechinus miliaris (sea urchin) eggs at 20° C. within the first
five minutes after insemination, followed by a fall to the original unfertilized level
within ten minutes. During the corresponding period in the eggs of sea urchins, no
increase in the rate of penetration of orthophosphate was observed. However,
within the first 6 to 7 minutes after insemination at 16° C., a prominent decrease in
the concentration of intracellular inorganic phosphate has been noted in the eggs of
6". purpuratus (Chambers and White, 1949) and 5*. drobachiensis (Chambers and
Mende, 1953b).

The rate of oxygen consumption of sea urchin eggs reaches a maximum some
time before first cleavage (Runnstrom, 1933) and remains fairly constant during
the next several cleavages. The rate of P32 uptake (at 15° C., for both species of
Strongylocentrotus eggs) reaches a maximum between 30 to 40 minutes before
first cleavage and remains constant thereafter through the first two or three cleavage
cycles.

Zeuthen (1949, 1950a, 1950b, 1951) has demonstrated that superimposed on
the basic oxygen consumption curves of sea urchin eggs are definite waves of
relatively small magnitude, the minima corresponding to the periods of cytoplasmic
division, the maxima to the prophases. Although no alterations could be detected

294 S. C. BROOKS AND E. L. CHAMBERS

in the rate of P:!- uptake during the first few cleavages, it should be emphasized
that the accuracy of the P32 uptake method described in this paper is such that
waves of considerably greater magnitude than those described by Zeuthen would
not have been detected. However, using a method of much greater accuracy,
cyclic variations in the rate of P32 uptake have been observed during the later
segmentation stages of sand dollar eggs (Chambers, White and Zeuthen in Zeuthen,
1951).

In the experiment carried out using the eggs of Urechis caupo, obtained from
freshly collected animals, the rate of P32 uptake did not increase following fertiliza-
tion until after the second cleavage. This may be related to the fact that an increase
in the rate of oxygen consumption does not occur in these eggs, obtained from
freshly collected animals, until during the later cleavage stages (Tyler and Humason,
1937; Horowitz, 1940).

Changes in the rate of oxygen consumption appear to parallel changes in the
rate of accumulation of phosphate in the marine eggs studied, at least before and
after fertilization and during the early cleavage stages. Following fertilization,
the sea urchin eggs accumulate phosphate, since energy from oxidative processes is
utilized in the synthesis of organic phosphorous-containing compounds (Chambers
and Mende, 1953b).

SUMMARY

1. A method is described for measuring the concentration of radioactive isotope
in cells without washing the cells free of the surrounding radioactive medium.

2. P32 as orthophosphate penetrates the unfertilized eggs of all species at a slow
and constant rate.

3. During the first 10 to 15 minutes following the insemination of Strongylocen-
trotus purpuratus and S. jranciscanus eggs, the rate of P32 uptake is essentially
identical to that of the unfertilized eggs. The rate of uptake increases by 15 to
20 minutes, and reaches a maximum by 50 to 90 minutes after insemination. There-
after, through the first three cleavages the rate remains constant, within the limits
of error of the method, as long as the concentration of P in the medium is in excess
of 20 Mg P/liter.

4. Following insemination of Urechis caupo eggs, the rate of P32 uptake does
not increase. After the second cleavage, however, the rate of P32 uptake increases,
and a maximum rate has not been attained even after the fourth cleavage.

5. When fertilized eggs containing P32 are suspended in non-radioactive sea
water, they slowly lose P32 to the external medium. On the other hand, no P32 is
lost from fertilized eggs containing radioactive phosphate when they are washed in
non-radioactive sea water.

6. P32 is not absorbed to the extraneous coats of fertilized eggs.

7. The rate of penetration of phosphate into the unfertilized and fertilized eggs
has been calculated in terms of the /xg P entering one ml. eggs/minute. The rate at
which phosphate enters fertilized Strongylocentrotus eggs is relatively independent
of the external concentration, as long as this exceeds 20 ju,g P/liter. Phosphate
enters fertilized Strongylocentrotus eggs 86 to 132 times faster that it penetrates
the unfertilized eggs.

PENETRATION OF P32 INTO MARINE EGGS 295

LITERATURE CITED

ABELSON, P. H., 1947. Permeability of eggs of Arlmcia punctulata to radioactive phosphorus.

Biol. Bull., 93: 203.
BOREI, H., 1948. Respiration of oocytes, unfertilized eggs, and fertilized eggs from Psammechi-

nus and Asterias. Biol. Bull., 95 : 124-150.
BOREI, H., 1949. Independence of post-fertilization respiration in the sea-urchin egg from the

level of respiration before fertilization. Biol. Bull., 96: 117-122.
BROOKS, S. C., 1939a. Ion exchanges in accumulation and loss of certain ions by the living

protoplasm of Nitella. J. Cell. Coin/). Physiol., 14: 383-401.
BROOKS, S. C., 1939b. Intake and loss of radioactive cations by certain marine eggs. Proc. Sue.

K.rp. Biol. Mcd., 42: 557-558.
BROOKS, S. C., 1940. The intake of radioactive isotopes by living cells. Cold Sprint/ Harbor

Symposia Quant. Biol., 8: 171-177.
BROOKS, S. C., 1943a. Intake and loss of ions by living cells. I. Eggs and larvae of Arbacia

punctulata and Asterias jorbcsi exposed to phosphate and sodium ions. Biol. Bull., 84:

213-225.
BROOKS, S. C., 1943b. Intake and loss of ions by living cells. II. Early changes of phosphate

content of Fundulus eggs. Biol. Bull., 84: 226-239.
BROOKS, S. C., 1951. Penetration of radioactive isotopes, PS2, Na34 and K42 into Nitella. /. Cell.

Comp. Physiol., 38: 83-93.
BROOKS, S. C., AND E. L. CHAMBERS, 1948. Penetration of radioactive phosphate into the eggs

of Strongyloccutrotits purpuratus, S. franciscanus. and Urechis caupo. Biol. Bull., 95:

262-263.

CHAMBERS, E. L., AND T. J. MENDE, 1953a. The adenosine triphosphate content of the un-
fertilized and fertilized eggs of Asterias forbesii and Strongylocentrotus drobachiensis.

Arch. Biochcm. Biophys., 44 : 46-56.
CHAMBERS, E. L., AND T. J. MENDE, 1953b. Alterations of the inorganic phosphate and arginine

phosphate content in the eggs of Strongylocentrotus drobachiensis following fertiliza-
tion. E.i-p. Cell Research, 5: 508-519.
CHAMBERS, E. L., AND W. E. WHITE, 1949. The accumulation of phosphate and evidence for

synthesis of adenosine triphosphate in the fertilized sea-urchin egg. Biol. Bull., 97 :

225-226.
CHAMBERS, E. L., AND W. E. WHITE, 1954. The accumulation of phosphate by fertilized sea

urchin eggs. Biol. Bull., 106: 297-307.

CHAMBERS, ROBERT, 1940. The relation of extraneous coats to the organization and permea-
bility of cellular membranes. Cold Spring Harbor Symposia Quant. Bio!., 8: 144—153.
HERBST, C., 1898. Uber z\vei Fehlerquellen beim Nachweis der Unentehrlichkeit von Phosphor

und Eisen fiir die Entwickelung des Seeigellarven. Arch. f. Entw., 7: 486-510.
HOROWITZ, N. H., 1940. The respiratory metabolism of the developing eggs of Urechis caupo.

J. Cell. Comp. Physiol, 15: 299-308.
LASER, H., AND LORD ROTHSCHILD, 1939. The metabolism of the eggs of Psanunecliinus iniliaris

during the fertilization reaction. Proc. Roy. Soc. (London), Scries B, 126: 539-557.
LILLIE, R. S., 1916. Increase of permeability to water following normal and artificial activation

in sea-urchin eggs. Amcr. J. Physiol., 40: 249-266.
LIXDBERG, O., 1948. On the turnover of adenosine triphosphate in the sea-urchin egg. Arkiv.

Kemi, Mineral. Gcol., 26B : No. 13 : 1-4.
LOEB, J., 1907. The chemical character of the process of fertilization and its bearing upon the

theory of life phenomena. Univ. Cal. (Berkeley) Pub. Physiol., 3: 61-81.
MACGINITIE, G. E., 1935. Normal functioning and experimental behavior of the egg and sperm

collectors of the echiuroid, Urechis caupo. J. R.vp. Zool., 70: 341-355.
MOORE, A. R., 1930. Fertilization and development without membrane formation in the egg

of the sea-urchin, Strongylocentrotus purpuratus. Protoplasma, 9 : 9-17.
NEEDHAM, J., AND D. M. NEEDHAM, 1930. On phosphorus metabolism in embryonic life. I.

Invertebrate eggs. Brit. J. E.rp. Biol., 7 : 317-348.
RUNNSTROM, J., 1933. Zur Kenntnis der Stoffvvechselvorgange bei der Entwicklungserregung

des Seeigeleis. Biochcm. Zcitschr., 258: 257-279.

296 S. C. BROOKS AND E. L. CHAMBERS

STEWART, D. R., AND M. H. JACOBS, 1932. The effect of fertilization on the permeability of the

eggs of Arbacia and Asterias to ethylene glycol. /. Cell. Comp. Physio!., 2 : 275-283.
TYLER, A., 1932. Change in volume and surface of Urechis eggs upon fertilization. /. Exp.

Zool, 63 : 155-173.
TYLER, A., AND W. D. HUMASON, 1937. On the energetics of differentiation, VI. Comparison

of the temperature coefficients of the respiratory rates of unfertilized and fertilized eggs.

Biol. Bull., 73 : 261-279.

WHITELEY, A. H., 1949. The phosphorous compounds of sea-urchin eggs and the uptake of radio-
phosphate upon fertilization. Aiucr. Naturalist, 83: 249-282.
ZEUTHEN, E., 1949. Oxygen consumption during mitosis ; experiments on fertilized eggs of

marine animals. Amcr. Naturalist, 83: 303-322.
ZEUTHEN, E., 1950a. Respiration during cell division in the egg of the sea-urchin Psammechimis

miliaris. Biol. Bull, 98: 144-151.
ZEUTHEN, E., 1950b. Respiration and cell division in the egg of Urechis caupo. Biol. Bull.,

98: 152-160.
ZEUTHEN, E., 1951. Segmentation, nuclear growth and cytoplasmic storage in eggs of echino-

derms and amphibia. Pubbl. stas. zool. Napoli, 23 (Suppl.) : 47-69.
ZIELINSKI, M. A., 1939. Carbohydrate metabolism and phosphorus compounds in the fertilized

eggs of the sea-urchin (Paracentrotus lividus). Ada Biol. Exptl. (Lodz), 13, No. 4:

35-48.

THE ACCUMULATION OF PHOSPHATE BY FERTILIZED

SEA URCHIN EGGS L

EDWARD L. CHAMBERS 2 AND WILLIAM E. WHITE3
Department of Zoology, University of California, Berkeley, California

Radioactive phosphate enters fertilized sea urchin eggs far more rapidly than it
enters the unfertilized eggs (Brooks and Chambers, 1948, 1954; Abelson, 1947;
Lindberg, 1948; Whiteley, 1949). Investigations described in this paper demon-
strate that the entry of P32 into the eggs represents an accumulation of phosphate
within the eggs. In addition, the concentrations of P and P32 in the inorganic and
organically bound phosphate fractions of the eggs have been measured, with the
purpose of determining in which fractions the phosphate, accumulated by the fer-
tilized eggs, is incorporated, and whether the process of accumulation is associated
with alterations in the distribution of P within the eggs.

MATERIALS AND METHODS

Eggs of the Pacific coast sea urchins Strongylocentrotus purpuratus, S. fran-
ciscanus, and Lyt echinus pictus were prepared for use, and measurements of egg
volume, and of P32 concentration in the eggs and suspension fluid carried out as
described previously (Brooks and Chambers, 1954). The jelly was removed from
the eggs by repeated washings in sea water. The sea water used in the experi-
ments was filtered through fine mesh filter paper. The experiments were performed
at 15 ±0.1° C. unless otherwise stated. The pH of the sea water in which the
eggs were suspended was measured at intervals throughout the duration of the
experiments, and varied between pH 8.0 to 8.2. Carrier-free P32, as orthophos-
phate, was added to the egg suspensions in amounts which varied from 0.2 to 2 /AC
P32/liter of suspension. The concentration of orthophosphate in the sea water
was measured using the Deniges- Atkins method (Atkins, 1923) with corrections for
reagent blank and salt error (Cooper, 1938).

Trichloroacetic acid extracts of unfertilized and fertilized eggs were prepared
as described by Chambers and Mende (1953a). Measurements of the P and P32
content of the inorganic and easily hydrolyzable phosphate fractions of the trichloro-
acetic acid-soluble extracts were carried out using the isobutyl alcohol extraction
method of Borbiro and Szent-Gyorgyi (1949). After measurement of the phos-
phomolybdate concentration in the isobutyl alcohol extracts, aliquots were pipetted
into flat dishes, evaporated, and the P32 concentration measured using a Geiger-
Miiller counter. In all experiments described in this paper samples of the egg

1 This investigation was supported by a research grant (C-559) to the University of Cali-
fornia, at Berkeley, from the National Cancer Institute of the National Institutes of Health,
U. S. Public Health Service. Preliminary reports of the research described in this paper have
been published (Chambers, Whiteley, Chambers and Brooks. 1948; Chambers and White, 1949).

2 Now at the University of Oregon Medical School, Portland 1, Oregon.

3 Now at the University of Kansas Medical School, Kansas City, Kansas.

297

298

E. L. CHAMBERS AND W. E. WHITE

suspensions, as originally prepared, were kept for observation. If the suspension
was of unfertilized eggs, these were inseminated at the completion of the experiment.
Over 95 per cent of the eggs in these samples developed to normal swimming
gastrulae.

RESULTS

Removal of P and P"- from the suspension fluid, and uptake of Pz- by the eggs in
suspensions of sea urcliin eggs

Suspensions of unfertilized and fertilized eggs were prepared in filtered sea
water containing 2.5 to 5.6 ml. eggs/liter. Small quantities of orthophosphate and
P32 were added to the suspensions. The initial concentration of orthophosphate in
the suspension fluid varied between less than 4 //,g to 78 p.g P/liter, and the initial
concentration of P32 from 0.23 to 0.28 /xc/liter.

Unfertilised eggs. The results obtained using suspensions of unfertilized 5\
purpuratus eggs are shown in Table I. Measurements of P and P3- concentrations
were begun two hours after the eggs had been removed from the ovaries. In
Experiments 1, 2 and 3 the concentration of P in the external medium increased,

TABLE I

Concentration of P in the suspension fluid of a suspension of unfertilized S. purpuratus eggs.

Experiments 1 to 5

lixpt. No.

Ml. eggs/1.
suspension

Time between
P analyses
in minutes

Initial P
cone., ng/\.
susp. fluid

Final P
cone., itg''\.
susp. fluid

fig P lost/'
ml. eggs/min.

Mg P entering/
ml. eggs/min.

1

5.6

97

38

55

0.032

0.0011

2

4.1

120

8

20

0.024

—

3

4.7

312

30

56

0.019

—

4

5.5

115

4

5

0.0

—

5

2.5

159

78

78

0.0

0.0026

while in Experiments 4 and 5 no appreciable change in the P concentration could
be detected (Table I, columns 4, 5 and 6). There was no measurable change in
concentration of P3- in the suspension fluid in any of the experiments. In Experi-
ments 1 and 5 the uptake of P32 by the eggs was measured and the quantity of P
entering one ml. eggs/minute calculated (Table I, column 7), as previously de-
scribed by Brooks and Chambers (1954). The results reveal that P enters un-
fertilized eggs whether or not the eggs simultaneously lose P to the external medium.
At the completion of the experiments the unfertilized eggs were inseminated, and
they developed normally through the gastrula stage.

Fertilised eggs. The results obtained using suspensions of fertilized S. pur-
puratus and 6". franciscanus eggs are shown in Tables II and III, and Figure 1.
The eggs were inseminated two hours after removal from the ovaries, washed free
of spermatozoa by gentle centrifugation, and suspended in sea water containing
known concentrations of orthophosphate and P32.

In Experiment 6 (Table II, Fig. 1) at 20.5 minutes after insemination 2.5 ml. ot
S. purpuratus eggs were suspended in a liter of sea water containing 78 /j.g P/liter.
A prominent decrease in concentration of P and P32 in the medium occurred (Expt.

ACCUMULATION OF PHOSPHATE BY EGGS

299

6, Table II, columns 2 and 4, and Fig. 1 ). The rate of uptake of P;J by the eggs
(Table II, column 6) was identical to the rate of disappearance of P and of P32 from
the medium (Table II, compare columns 3, 5 and 7). The initial lag in the dis-
appearance of orthophosphate and of Pn- from the medium (Fig. 1, Expt. 6, from
0 to 30 minutes) is due to the fact that the uptake of P by fertilized eggs does not
reach a maximum until about one hour after insemination (Brooks and Chambers,
1954). Subsequently, orthophosphate is removed from the medium at a constant
rate until the concentration falls to 15 to 20 p.g P/liter (Fig. 1, Expt. 6). The
rate of uptake then falls off sharply.

TABLE II

Concentration of P and P32 in the suspension fluid, and of P3- in the eggs in

suspensions of fertilized eggs.

Experiments 6 and 7

Time after
initial
measurement
in minutes

Mg P/l.
susp. fluid

Per cent
initial P
cone, in
susp. fluid

CPM Pas/I,
susp. fluid

Per cent
initial P32
cone, in
susp. fluid

CPM P32 in
eggs/h
suspension

Per cent
initial P«2
cone, in eggs

Experiment 6. S. purpuratus eggs

0.0*

78

100.0

61,000

100.0

0

0.0

17.5

75

96.0

56,600

92.6

4,780

7.8

60.5

48

61.5

36,780

60.2

24,900

40.8

91.5

22

28.2

16,960

27.8

42,920

70.5

129.5

6

7.7

6,600

10.8

54,480

90.0

213.0

<2

<3.0

3,200

5.2

58,140

95.5

387.5

<2

<3.0

2,360

3.8

—

—

Experiment 7. S. franciscanus eggs

O.Oj

53.5

100.0

76,000

100.0

0

0.0

31.0

40.0

74.8

62,000

81.6

15,200

20.0

68.0

27.5

51.4

—

—

—

—

139.0

10.0

18.7

17,160

22.6

57,380

75.4

192.0

5.0

9.3

9,200

12.1

67,020

88.0

* Eggs inseminated 20.5 minutes before initial measurment.
t Eggs inseminated 63 minutes before initial measurement.

In Experiment 7 (Table II, Fig. 1) at 63 minutes after insemination 3.5 ml. of
S. franciscanus eggs were suspended in a liter of sea water containing 53.5 p.g P/
liter. The results are similar to those obtained in Experiment 6. The eggs remove
orthophosphate from the external medium at a constant rate until the concentration
falls below 20 /xg P/liter, when the rate of uptake by the eggs falls off sharply
(Fig. 1).

In Experiment 8 (Table III) at 43 minutes after insemination 4.9 ml. of
S. jranciscanus eggs were suspended in a liter of sea water. The P and P32
concentrations in the suspension fluid were measured at the beginning and at the
end of successive 30-minute periods. Additional amounts of P and P32 were added
to replenish the external medium prior to each 30-minute period. As in the two

300

E. L. CHAMBERS AND W. E. WHITE

previous experiments, the decrease in concentration of P3'2 parallels the decrease in
concentration of P in the suspension fluid (Table III, columns 6 and 7). The
results show that as long as the concentration of orthophosphate is over 18 to 20
p.g P/liter, the rate of uptake of P and P32 remains fairly constant during the first
430 minutes after insemination (Table III, column 8).

Distribution of P3- between the trichloroacetic acid-soluble and -insoluble fractions
of the eggs

Suspensions of Lytechinus pictus eggs were prepared containing one ml. eggs/
liter of sea water maintained at a temperature of 20 to 21° C. Two hours after

O O EXPT. 6, SUSPENSION S.purpuratuS EGGS

X X EXPT. 7.SUSPENSION S. f TO nciscanus EGGS

0 20 40 60 80 100 120 140 160 180

TIME IN MINUTES AFTER INITIAL MEASUREMENT OF P CONCENTRATION

FIGURE 1.

the eggs had been removed from the ovaries, carrier-free P32 was added. At various
intervals of time duplicate 20-ml. samples of the suspension were removed, the
eggs washed three times by centrifugation at 86 X g for three minutes in non-
radioactive sea water, the supernatant sea water decanted, an equal volume of ice
cold 10 per cent trichloroacetic acid added to each of the duplicate samples, the
trichloroacetic acid-soluble and -insoluble fractions separated, and the P32 content
of the fractions measured.

In unfertilized eggs, between 95.7 to 95.9 per cent of the P32 was recovered in
the trichloroacetic acid-soluble extracts, with 4.1 to 4.3 per cent in the acid-insoluble
fractions, after the eggs had been exposed to P32 for two hours.

In the process of washing the unfertilized eggs in non-radioactive sea water

ACCUMULATION OF PHOSPHATE BY EGGS

301

prior to homogenization, from 4 to 8 per cent of the P32 initially present in the eggs
was removed. The effect of this loss of P32 is to decrease the relative proportion
of P32 contained in the trichloroacetic acid-soluble extracts of the washed eggs by
0.2 to 0.3 per cent.

In experiments carried out using fertilized eggs, the eggs were inseminated two
hours after removal from the ovaries and P32 was added at the time of insemination.
In eggs exposed to P32 for a period of 45 to 120 minutes after insemination, and
washed for a period of 20 to 30 minutes in non-radioactive sea water, 96.3 to 96.4
per cent of the P32 was found in the trichloroacetic acid-soluble extracts, and 3.6
to 3.7 per cent in the acid-insoluble residue. No appreciable quantity of P32 is lost
from fertilized eggs when washed in non-radioactive sea water (Brooks and
Chambers, 1954). The slightly lower proportion of P32 in the acid-insoluble
fraction of fertilized eggs, as compared to the unfertilized, may be entirely due
to the loss of P32 from the unfertilized eggs when they are washed prior to
homogenization.

TABLE III

Suspension of fertilized S. franciscanus eggs. Disappearance of P and P32 from the suspension
fluid during thirty-minute periods, following successive additions of P and P32.

Experiment 8

Time after

Initial concentration:

Final concentration:

Per cent

Per cent

msem., in
30 minute

decrease
P cone.

decrease
P32 cone.

/ig P/ml.
eggs/min.

intervals

«g P/l.

CPM P»2/l.

Mg P/l.

CPM PS 2/1.

50 to 80

63.5

598,000

41.0

356,000

35

37

0.15

110 to 140

50.5

473,000

25.7

248,000

49

48

0.17

160 to 190

47.0

—

23.5

—

50

—

0.16

210 to 240

57.0

578,000

35.8

350,000

37

39

0.14

290 to 320

38.5

385,000

18.0

214,000

53

45

0.14

325 to 355

(16.0)

(220,000)

(10.0)

(140,200)

(37)

(35)

(0.04)

400 to 430

57.5

550,000

30.7

286,000

47

48

0.18

In a series of experiments, after exposing the inseminated eggs to P32 for 50
minutes and washing, the eggs were allowed to develop 400 minutes in non-radioac-
tive sea water to the early blastula stage prior to homogenization. No P32 was lost
from the fertilized eggs during the long period of development in the sea water free
of P32, even though the medium surrounding the eggs was repeatedly replaced by
fresh sea water. The quantity of P32 found in the acid-soluble extract amounted
to 93.0 per cent of the total, with 7.0 per cent in the acid-insoluble fraction, as com-
pared to 96.4 and 3.6 per cent, respectively, in the corresponding experiment on
fertilized eggs homogenized immediately after washing. The experiment reveals
that a substantial portion of the phosphate, initially accumulated in the eggs, be-
comes incorporated in the acid-insoluble fraction. This conclusion is based on the
consistency with which a lower percentage of P32 was found in the acid-insoluble
residue of fertilized eggs continuously exposed to P32.

In the control experiment P32 was added only after the eggs had been suspended
in trichloroacetic acid. Even after repeated washing of the acid-insoluble residue
with trichloroacetic acid, 1.0 per cent of the P32 was retained in this fraction. This
experiment indicates that the quantity of P32 organically combined in the acid-

302

E. L. CHAMBERS AND W. E. WHITE

insoluble residue is probably less by at least one per cent than the quantities actually
found.

The distribution of P:)J between the trichloroacetic acid-soluble and -insoluble
fractions of S. purpuratus eggs, both unfertilized (four experiments) and fertilized
(six experiments) is essentially identical to that found in the eggs of Lytechinus
pictus.

Lipids and phospholipids were extracted from the acid-insoluble residue of the
fertilized L. pictus eggs. The acid-insoluble residue, after complete extraction with
a mixture of three volumes ethanol and one volume ether, retained 92.1 per cent
of the original P32 content. The ethanol-ether extract was dried, and the residue
extracted with petroleum ether. The petroleum ether fraction containing the
phospholipids accounted for 7.5 per cent of the total P32 content of the trichloroacetic
acid-insoluble fraction. The remaining 0.5 per cent was in the petroleum ether-
insoluble fraction.

TABLE IV

Distribution of P and P32 in the acid-soluble extracts of Strongylocentrotus purpuratus eggs.

Experiments 9, 10 and 11

Expt. No.

Condition
of eggs

Mg P/ml. eggs± std. dev.

Per cent total P32:

Inorg.
P

Labile
P

Inorg. -Habile
P

Inorg.

P32

Labile

P32

Acid stable

P32

9

Unfertilized

58±.6

408±4

466±4

24

66

10

Fertilized

16±.2

456±5

472±5

6

88

6

10

Unfertilized

69±.7

415±4

484 ±5

.

Fertilized

34±.4

451 ±5

485±5

8

84

8

11

Unfertilized

77±.8

460 ±5

537±5

—

Fertilized

39±.4

496±5

535±5

9

84

7

Distribution of P and P32 in the trichloroacetic acid-soluble extracts of S. purpuratus

eggs

The quantities of inorganic P and P liberated after 10 minutes' hydrolysis in
1 N HC1 at 100° C. in the trichloroacetic acid-soluble extracts of unfertilized and
fertilized eggs were determined. The results of three representative experiments
are shown in Table IV, Experiments 9, 10 and 11. Five ml. of S. purpuratus eggs
were suspended in a liter of sea water containing 20 to 50 p.g P as orthophosphate/
liter. The suspension was divided into two equal lots. Carrier-free P32, 1 /xc/100
ml. suspension, was added to one lot of unfertilized eggs one hour after removal
from the ovaries. The other lot was inseminated two hours after the eggs had been
removed from the ovaries, and at the same time duplicate 100-ml. samples were
removed from the suspension of unfertilized eggs, centrifuged, and the trichloroace-
tic acid extracts prepared. Thirty minutes after insemination 0.1 ;u,c P32/100 ml.
suspension was added to the fertilized eggs, and at 60 minutes after insemination,
duplicate 100-ml. samples were removed, the fertilized eggs washed twice in non-
radioactive sea water by centrifugation, and the trichloroacetic acid extracts pre-

ACCUMULATION OF PHOSPHATE BY EGGS 303

pared. The results show that following insemination, a prominent decrease in the
concentration of inorganic P occurs within the eggs (Table IV, column 3), and
at the same time a corresponding increase in the concentration of P liberated after
hydrolysis (Table IV, column 4). Within the errors of the measurements, the
sum of the inorganic P and P liberated after hydrolysis is the same both before
and after insemination (Table IV, column 5).

The distribution of P32 between the various P fractions in the trichloroacetic
acid extracts is shown in Table IV, columns 6, 7 and 8. The results reveal that
the major portion of the P32 is associated with easily hydrolyzable organic P com-
pounds. Following insemination, with the accompanying decrease in quantity of
inorganic P and the increase in amount of P liberated after hydrolysis, the pro-
portion of P3- in the easily hydrolyzable P fraction increases markedly. The
proportion of P32 in the acid-stable P compounds is small, in spite of the fact that
Whiteley (1949) reports the presence, in trichloroacetic acid extracts, of 511 p.g
acid-stable P/ml. 5". purpuratus eggs, which amounts to approximately one-half
of the total P content in the acid-soluble extract.

DISCUSSION

The experiments presented in this paper establish conclusively that the entry
of P32 into the fertilized eggs quantitatively measures the accumulation of ortho-
phosphate within the eggs. However, P32 probably enters unfertilized eggs by an
exchange process, since the quantity of orthophosphate in the medium surrounding
the eggs either remains constant, or slowly increases. Measurements of the distri-
bution of P3- in the trichloroacetic acid-soluble extracts of the eggs reveal that in
both unfertilized and fertilized eggs the P32 is confined primarily to the intracellular
inorganic phosphate fraction and the easily hydrolyzable organic P compounds.
In fertilized eggs, between 84 to 88 per cent of the P32 entering the eggs is found
in the easily hydrolyzable fraction, indicating that the accumulation of phosphate
by fertilized eggs involves primarily its incorporation in the easily hydrolyzable P
compounds.

In cells actively metabolizing substrate the intracellular inorganic phosphate
concentration may be markedly lower than in slowly metabolizing cells, devoid of
or with a limited supply of substrate (yeast: MacFarlane, 1936, 1939; bacteria:
Wiggert and Werkman, 1938, O'Kane and Umbreit, 1942 ; brain tissue : Schachner
et al., 1942 ; retinal tissue : Bumm and Fehrenbach, 1931 ; liver : Lundsgaard, 1938).
Furthermore, many investigators have shown that orthophosphate rapidly enters
and accumulates in actively metabolizing cells (diatoms: Ketchum, 1939a, 1939b ;
yeast: Hevesy ct a!., 1937, Mullins, 1942; bacteria: Vogler and Umbreit, 1942,
Wiggert and Werkman, 1938, O'Kane and Umbreit, 1942, Hotchkiss, 1946; brain
tissues: Schachner ct al., 1942). When the same cells are devoid of substrate,
phosphate ions enter slowly and the cells may even lose phosphate to the external
medium.

Unfertilized sea urchin eggs, at least after a period of sojourn in sea water,
present the picture of cells with limited available or utilizable substrate. They
possess a characteristically low metabolic rate (Borei, 1948), have a high inorganic
phosphate content, a relatively low content of easily hydrolyzable P (see also

304 E. L. CHAMBERS AND W. E. WHITE

Chambers and Mende, 1953b), may slowly lose phosphate to the external medium,
and P penetrates the eggs at an extremely slow rate (Brooks and Chambers, 1954).
However, after the eggs are fertilized, the eggs behave as if an abundant supply
of substrate had been made available, or had become utilizable. The oxygen con-
sumption increases, the inorganic phosphate content of the eggs is strikingly lowered,
the quantity of easily hydrolyzable P increases (see also Chambers and Mende,
1953b), and the eggs now accumulate phosphate, absorbing it from the external
medium.

The fertilized eggs would appear to accumulate orthophosphate against a con-
centration gradient of a thousand-fold or more (compare column 2, Tables II and
III with column 3, Table IV). This, however, is unlikely since the analytically
determined inorganic P content of cells probably represents, in addition to the true
intracellular ionic orthophosphate, hydrolysis products of highly labile phosphate
esters and orthophosphate which, in the living cell, had been present in undissociated
salt-like complexes. The binding of orthophosphate by electrostatic forces has
been shown to occur, for example, in the protein aldolase (Velick, 1949). De-
naturation of proteins may abolish their ability to bind anions (Klotz and Urquhart,
1949). Furthermore, the anions of an extracting agent, such as trichloroacetic acid,
would tend to displace phosphate ions which, in the living cells, had been present
in undissociated salt-like complexes and in equilibrium with free orthophosphate
ions. It is proposed that in the living sea urchin eggs the actual concentration of
free ionic orthophosphate is only a fraction of the analytically determined inorganic
P. Following fertilization of the eggs, along with the demonstrated decrease in
concentration of the analytically determined inorganic P, the concentration of free
orthophosphate ions in the egg protoplasm may be reduced to such a low order of
magnitude as to favor the entry of orthophosphate from the external medium.

The hypothesis has been advanced that the penetration of orthophosphate into
cells requires esterification at the cell surface. The marked effects of changes in
temperature and of certain metabolic inhibitors (Kamen and Spiegelman, 1948,
Villee et al., 1949) on the rate of penetration of orthophosphate have been cited in
support of this hypothesis. However, Jacobs and co-workers (1935) have em-
phasized that changes in temperature may cause marked shifts in "equilibrium"
states, and such alterations would have to be taken into account before the effects
of temperature changes on the rate of penetration of orthophosphate could be
properly evaluated. Similarly, metabolic inhibitors must induce profound changes
in "equilibrium" states within cells. For example, Spiegelman, Kamen and Suss-
man (1948) have shown that azide prevents the decrease in concentration of intra-
cellular inorganic P which normally occurs when yeast ferments glucose.

The claim has also been made that orthophosphate must enter cells by a process
of esterification at the cell surface, since the specific activity of the P in certain
organic compounds may be higher than that of the intracellular inorganic P (e.g.,
Lindberg, 1950). Such an interpretation from specific activity measurements is
open to serious question, since the analytically determined inorganic P of cells
is undoubtedly derived from several different components, and does not represent
the true ionic orthophosphate content of the living cell.

The observed great differences in the rates of penetration of orthophosphate
into cells at different levels of metabolic activity may just as well be explained by

ACCUMULATION OF PHOSPHATE BY EGGS 305

changes in "driving forces" such as the rate at which orthophosphate is esterified
within the cells, and changes in the concentration gradient of free orthophosphate
ions.

The authors desire to thank Miss Nylan Jeung for technical assistance during
the course of this investigation.

SUMMARY

1. The concentration of phosphate in the external medium of a suspension
of unfertilized Strongylocentrotus eggs remains constant, or increases, while in a
^suspension of fertilized eggs, the concentration of phosphate in the external medium
decreases.

2. Fertilized Strongylocentrotus eggs absorb P32 and phosphate from sea water
at identical rates, revealing that the exchange of phosphate between the cell interior
and the external medium is inappreciable.

3. The rate at which phosphate is removed from sea water by fertilized Strongy-
locentrotus eggs is relatively independent of the external concentration as long as
this exceeds 15 to 20 micrograms P per liter.

4. When unfertilized and fertilized sea urchin eggs are continuously exposed
to sea water containing P32 and more than 20 micrograms P per liter, 95.9 to 96.4
per cent of the P32 which enters the eggs is found in the trichloroacetic acid-soluble
fraction, with 3.6 to 4.1 per cent of the P32 being recovered in the acid-insoluble
fraction. The distribution of P32 between these two fractions is not significantly
different in the unfertilized, as compared to the fertilized eggs. Although a slightly
lower proportion of P32 was found in the acid-insoluble residue of unfertilized eggs,
outward leaching of P32 during the washing of the unfertilized eggs may well ac-
count for the difference noted.

5. If fertilized Lytechinus pictus eggs containing P32 are suspended in a non-
radioactive medium shortly after insemination, the proportion of P32 in the acid-
insoluble fraction increases from 3.6 per cent at the two-celled stage to 7.0 per cent
at the blastula stage.

6. The concentration of inorganic P in the trichloroacetic acid-soluble extracts
of the eggs decreases prominently following insemination. A corresponding in-
crease occurs in the quantity of P liberated after 10 minutes' hydrolysis of the
extracts in 1 N HC1 at 100° C.

7. The major portion of the P32 which enters the eggs is found in the easily
hydrolyzable P fraction of the trichloroacetic acid-soluble extracts. After fertiliza-
tion, the proportion of P32 in the easily hydrolyzable P fraction increases.

LITERATURE CITED

ABELSON, P. H., 1947. Permeability of eggs of Arbacia punctulata to radioactive phosphorous.

Biol. Bull. 93: 203.
ATKINS, W. R. G., 1923. The phosphate content of fresh and salt waters in its relationship to

the growth of the algal plankton. /. Marine Biol. Assoc. U. K., 13: 119-150.
BORBIRO, M., AND A. SzENT-GvoRGYi, 1949. On the relation between tension and ATP in

cross striated muscle. Biol. Bull., 96 : 162-165.
BOREI, H., 1948. Respiration of oocytes, unfertilized eggs, and fertilized eggs from Psam-

mechinus and Astcrias. Biol. Bull., 95 : 124-150.

306 E. L. CHAMBERS AND W. E. WHITE

BROOKS, S. C, AND E. L. CHAMBERS, 1948. Penetration of radioactive phosphate into the eggs
of Strongyloccntrotus purpuratus, S. franciscanus, and Urechis caupo. Biol. Bull., 95:
262-263.

BROOKS, S. C., AND E. L. CHAMBERS, 1954. The penetration of radioactive phosphate into marine
eggs. Biol. Bull., 106 : 279-296.

BUMM, E., AND K. FEHRENBACH, 1931. Uber verschiedene Wege des Zuckerabbaues im
tierischen Organismus II. Hoppc-Scylcr's Zeitschr. physiol. Chcm., 195: 101-112.

CHAMBERS, E. L., AND T. J. MENDE, 1953a. The adenosine triphosphate content of the unfer-
tilized and fertilized eggs of Astcrias forbcsii and Strongylocentrotus drobachiensis.
Arch. Biochein. and Biophys., 44: 46-56.

CHAMBERS, E. L., AND T. J. MENDE, 1953b. Alterations of the inorganic phosphate and arginine
phosphate content in the eggs of Strongyloccntrotus drobachiensis following fertiliza-
tion. ILvp. Cell Research, 5: 508-519.

CHAMBERS, E. L., AND W. E. WHITE, 1949. The accumulation of phosphate and evidence for
the synthesis of adenosine triphosphate in fertilized sea urchin eggs. Biol. Bull., 97:
225-226.

CHAMBERS, E. L., A. WHITELEY, R. CHAMBERS AND S. C. BROOKS, 1948. Distribution of radio-
active phosphate in the eggs of the sea urchin Lytechinus pictus. Biol. Bull., 95: 263.

COOPER, L. H. N., 1938. Salt error in determinations of phosphate in sea water. /. Marine Biol.
Assoc. U. K., 23: 171-178.

HEVESY, G., K. LIXOERSTR^M-LAXG AND N. NIELSEN, 1937. Phosphorous exchange in yeast.
Nature, 140: 725.

HOTCHKISS, R. D., 1946. Gramicidin, tyrocidine and tyrothricin. Advances in Enzymol., 4:
153-199.

JACOBS, M. H., H. N. GLASSMAN AND A. K. PARPART, 1935. Osmotic properties of the
erythrocyte. VII. The temperature coefficients of certain hemolytic processes. /.
Cell. Comp. Physiol., 1 : 197-225.

KAMEN, M. D., AND S. SPIEGELMAN, 1948. Studies on the phosphate metabolism of some
unicellular organisms. Cold Sprin<j Harbor Symposia Quant. Biol., 13: 151-163.

KETCH UM, B. H., 1939a. The absorption of phosphate and nitrate by illuminated cultures of
Nitzschia closterium. .-inter. J. Bot., 26: 399-407.

KETCHUM, B. H., 1939b. The development and restoration of deficiencies in the phosphorus
and nitrogen composition of unicellular plants. /. Cell. Comp. Physiol., 13: 373-381.

KLOTZ, I. M., AND J. M. URQUHART, 1949. The binding of organic ions by proteins. Com-
parison of native and of modified proteins. J . Ainer. Chcm. Soc., 71 : 1597-1603.

LINDBERG, O., 1948. On the turnover of adenosine triphosphate in the sea-urchin egg. Arkiv
Kemi. Mineral. Geol, 26B : No. 13, 1-4.

LINDBERG, O., 1950. On surface reactions in the sea urchin egg. E.rp. Cell Research, 1 : 105-
114.

LUNDSGAARD, E., 1938. The phosphate exchange between blood and tissue in experiments with
artificially perfused livers and hind limb preparations. Skand. Arch. Physiol., 80: 291-
302.

MACFARLANE, M. G., 1936. CXCV. Phosphorylation in living yeast. Biochein. J. (Lon-
don), 30: 1369-1379.

MACFARLANE, M. G., 1939. LXX. The phosphorylation of carbohydrate in living cells.
Biochem. J. (London), 33: 565-578.

MULLINS, L. J., 1942. Permeability of yeast cells to radiophosphate. Biol. Bull., 83: 326-333.

O'KANE, D. J., AND W. W. UMBREIT, 1942. Transformations of phosphorous during glucose
fermentation by living cells of Streptococcus faecalis. J. Biol. Chcm., 142: 25-30.

SCHACHNER, H., B. A. FRIES AND I. L. CHAiKOFF, 1942. The effect of hexoses and pentoses on
the formation in vitro of phospholipids by brain tissue as measured with radioactive
phosphorus. /. Biol. Chem.. 146: 95-103.

SPIEGELMAN, S., M. D. KAMEN AND M. SUSSMAN, 1948. Phosphate metabolism and dissociation
of anaerobic glycolysis from synthesis in the presence of sodium azide. Arch. Biochem.,
18 : 409-436.

VELICK, S. F., 1949. The interaction of enzymes with small ions. /. Ph\s. Colloid Chem., 53:
135-149.

ACCUMULATION OF PHOSPHATE BY EGGS 307

VILLEE, C. A., M. LOWENS, M. GORDON, E. LEONARD AND A. RICH, 1949. The incorporation of

P32 into nucleoproteins and phosphoproteins of the developing sea-urchin embryo.

/. Cell. Comp. Physiol, 33: 93-112.
VOGLER, K. G., AND W. W. UMBREiT, 1942. Studies on the metabolism of autotrophic bacteria.

III. The nature of the energy storage material active in the chemosynthetic process.

/. Gen. Physiol., 26: 157-167.
WHITELEY, A. H., 1949. The phosphorus compounds of sea-urchin eggs and the uptake of

radio-phosphate upon fertilization. Amer. Naturalist, 83 : 249-267.
WIGGERT, W. P., AND C. H. WORKMAN, 1938. XVIII. Phosphorylation by the living bacterial

cell Biochem. J. (London), 32: 101-107.

A STUDY OF THE MECHANISM INVOLVED IN SHIFTING OF
THE PHASES OF THE ENDOGENOUS DAILY RHYTHM

BY LIGHT STIMULI

FRANK A. BROWN, JR.,1 MILTON FINGERMAN AND MARGARET N. HINES

Department of Biological Sciences, Northwestern University, and
The Marine Biological Laboratory, Woods Hole, Mass.

A persistent diurnal rhythm of color change in the fiddler crab, Uca, was first
reported by Abramowitz in 1937, and has been abundantly confirmed by numerous
investigators since that time. The character of the chromatophore rhythm is such
that the crabs darken by day and blanch by night owing to dispersion and concen-
tration, respectively, of the melanin within their melanophores. This rhythm has
been discussed most recently and described in considerable detail by Brown, Finger-
man, Sandeen and Webb (1953). These investigators demonstrated that the
rhythm not only persists for long periods in constant darkness, but actually in-
creases in amplitude to reach a maximum value only after ten days to two weeks.
This high value was then observed to persist without diminution for the longest
period of observation which was about a month. Furthermore, there appeared to
be no measurable drift of the rhythm away from its normal phase relations with
solar day-night, indicating the mechanism to have a remarkable precision of fre-
quency determination. T! : •- frequency was shown by Brown and Webb (1948)
to be independent of temp* ure over the twenty degree range from 6 to 26° C.

Although the mechanism appears to be a moderately stable one in the fiddler
crab, it was shown by Brown and Webb (1949), Webb (1950), and Brown,
Fingerman, Sandeen and Webb (1953) to be capable of having its phases readily
shifted by Lght-to-dark and dark-to-light changes under certain circumstances.
Examples ^f these shifts are (1) a backward shift of 4 to 5 hours in the phases
of the rhytiim by three consecutive midnight-to-6 A.M. periods of illumination of
animals otherwise in continuous darkness, (2) a forward shift of about 6 hours in
animals wb ;e rhythm had been inhibited by several days sojourn in continuous
bright illumination and then were placed in constant darkness at 7 A.M., and (3) a
shift of about twelve hours, or in other words a reversal, of phases by a few cycles
of illumination from 7 P.M. to 7 A.M. and darkness from 7 A.M. to 7 P.M. Brown
and Webb (1949) using illuminations of 150, 80, and 50 ft. c. found that the
brighter the light the sooner the reversal. With the highest illumination, reversal
occurred on the first day and with 40 ft. c. it occurred on the fourth. Once shifted,
the rhythms are as stable in their new phase relations as they were in their original
norm."l ones.

T >e experiments to be described were performed in order to gain further
insigb into the mechanism involved in inducing persistent shifts and certain other
modifications in the endogenous diurnal rhythm.

1 This investigation was supported by a research grant from the graduate school of North-
western University.

308

SHIFT IN DIURNAL RHYTHM BY LIGHT 309

MATERIALS AND METHODS

For the experiments to be described, 400 fiddler crabs were collected at
Chapoquoit, near Woods Hole, Massachusetts at about three o'clock on the after-
noon of June 24, 1952. They were kept in the natural daylight of the laboratory
until 7:00 P.M. when they were divided into sixteen groups of 25 crabs each and
placed in white enamelled pans in sea water to a depth of about a centimeter. One
group, the control one, whose normal rhythm was to be determined was placed
in darkness and left for the duration of the experiment. The remaining fifteen
groups, the experimental ones, were placed in the conditions of illumination to
which they were to be exposed during the night (7 P.M. to 7 A.M.) and the next
morning placed in the lower illumination or darkness to which they were to be
exposed by day (7 A.M. to 7 P.M.).

The illuminations for the experiments were obtained by frosted incandescent
lamps of various wattages held at different distances above the white pans con-
taining the animals. The illuminations were measured with a Weston photometer.
The nighttime-daytime illuminations for the fifteen groups of animals subjected
to the alternating illuminations were, in ft. candles, respectively: (1) 100-80, (2)
100-50, (3) 100-10, (4) 100-2, (5) 100-0, (6) 50-10, (7) 50-5, (8) 50-2, (9)
25-10, (10) 25-5, (11) 25-2, (12) 10-5, (13) 10-2, (14) 5-2, (15) 2-0.

The temperatures in the inside rooms of the Marine Biological Laboratory
in which the experiments were carried on did not show any significant diurnal
variation, and there was an irregular variation of only three or four degrees at
most during the course of the eleven days in which the experiment was carried
out. All the animals, furthermore, both experimentals and controls, were sub-
jected to essentially the same temperature conditions other than the rhythmic
differences in heat radiation during the periods of illumination.

The experimental groups were subjected to the twelve-hour alternating con-
ditions of illumination for six days; at 7 P.M. on June 30, they were placed in con-
stant darkness. Beginning at 11 P.M. and continuing at six-hour intervals (11

P.M., 5 A.M., 11 A.M., 5 P.M ) for four daily cycles the average chromatophore

stage of ten crabs randomly sampled from the fifteen experimental groups and the
controls were staged by the method of Hogben and Slome (1931). Through the
next or fifth daily cycle the chromatophores were staged at hourly intervals.

EXPERIMENTS AND RESULTS

A summary of the results is found in Table I.

1. Controls: The control group possessed a high-amplitude rhythm at the time
the staging of chromatophores commenced and showed no significant increase during
the five-day period of study. The highest value was found at 11 A.M., and the
lowest at 11 P.M. in every daily cycle.

2. 100-80 ft. c.: In these, it is evident that there was a strong initial depression
of rhythm-amplitude, which rapidly diminished during the five days. In every
cycle the maximum stage was now at 5 A.M. and the minimum at 5 P.M.

3. 100-50 ft. c.: This group, unlike the preceding, exhibited little or no initial
amplitude depression nor increase during the period of observation. It resembled
the preceding in having the maximum average stage always at 5 A.M. and the
minimum at 5 P.M.

310

BROWN, FINGERMAN AND HINES

TABLE I

The average stage of melanin dispersion at each of four times of day, under constant conditions, for

Uca which were earlier subjected to five days of higher illumination by night

and lower illumination by day

Ilium.

1 1 P.M.

5 A.M.

11 A.M.

5 P.M.

Ilium.

ff * \

11 P.M.

5 A.M.

11 A.M.

5 P.M.

(ft. c.)

(ft. c.)

Control

l.Of

2.3

5.0*

3.8

50-2

2.7*

i.ot

1.6

1.8

l.Of

1.9

4.7*

4.1

2.8*

1.4

i.ot

1.6

1.2f

1.9

4.5*

3.5

3.4*

1.3t

1.9

2.8

Lit

2.4

4.7*

4.4

3.0*

Lit

1.9

2.4

1.3f

3.1

4.5*

3.5

4.0*

1.4t

1.9

3.9

100-80

1.3

1.8*

1.3

i.ot

25-10

Lit

1.9*

1.8

1.6

1.2

1.2*

1.0

i.ot

i.ot

2.0*

1.7

1.3

1.3

3.6*

2.9

l.lt

1.2t

3.2*

3.2

2.3

1.9

4.3*

3.0

Lit

1.6t

3.2

4.2*

1.7

2.3

4.0*

2.3

i.ot

1.3t

3.7*

2.7

1.7

100-50

3.4

4.7*

2.2

1.2t

25-5

1.2t

1.3

2.2

2.4*

2.4

4.2*

1.7

1.2f

2.3

2.1t

2.1

2.5*

3.4

4.2*

3.5

Lit

2.2t

2.9

3.9*

3.2

3.0

4.5*

2.7

1.2t

2.4t

3.1

4.0

4.0*

3.4

3.8*

3.1

1.4t

1.8f

2.9

2.8

3.5*

100-10

3.5*

3.0

1.6

i.ot

25-2

1.3

i.ot

1.2

1.3*

1.8

3.0*

1.4

i.ot

1.3

i.ot

1.3

1.3*

2.8

3.6*

2.7

1.2f

1.4

i.ot

2.0

2.6*

2.5

3.1*

1.6

i.ot

2.6

i.ot

2.6

3.7*

2.5

3.2*

1.7

1-lt

2.6

i.ot

1.3

2.9*

100-2

1.8*

1.3

l.Of

1.4

10-5

1.2

Lit

1.5

2.0*

1.7*

1.0

l.Of

1.0

1.3t

1.3

2.2*

1.8

1.9

1.6f

2.0

2.7*

1.2t

2.0

3.4

3.6*

2.7

2.3

2.1t

3.0*

1.7t

1.7

3.0

3.3*

3.0

2.1

2.0f

3.2*

i.ot

2.4

3.0*

2.7

100-0

4.4*

3.8

l.Of

3.2

10-2

1.6

1.3f

1.9

2.1*

3.9*

3.0

l.Of

1.9

i.ot

1.4

1.6*

1.4

4.4*

1.7

i.ot

2.4

1.7t

2.1

2.3

2.9*

3.8*

2.1

l.Of

2.5

2.4

1.9t

3.1*

2.7

4.0*

2.7

i.ot

3.0

2.5

1.7f

1.7

3.0*

50-10

2.1

3.9*

2.4

1.2t

5-2

i.ot

1.2

2.1

2.3*

2.1

4.4*

3.2

1.2t

Lit

1.8

2.7*

2.1

1.9

4.9*

4.7

l.Sf

Lit

1.5

3.0*

2.6

1.5

4.3*

3.6

1.2f

i.ot

1.0

4.0*

2.2

2.4

4.6*

3.6

1.5t

1.2f

1.6

2.6*

2.5

50-5

1.9*

1.8f

1.7

1.6

2-0

2.4

1.2f

2.8

2.9*

1.6

1.4f

1.4

1.8*

2.0

1.3t

2.9

3.4*

1.3f

1.8

2.6*

2.2

3.1

L7t

3.7

4.0*

2.3

1.8f

2.7*

2.6

3.2

L4t

3.7

4.5*

2.6

1.9f

3.0*

2.4

2.6

1.5t

2.5

3.4*

* Signifies maximum values for a cycle.
t Signifies minimum values for a cycle.

SHIFT IN DIURNAL RHYTHM BY LIGHT 311

4. 100-10 ft. c.: The group subjected to these illuminations appeared to show
an intermediate degree of depression in amplitude from the beginning and no
systematic increase thereafter ; but still again, the maximum was nearly always at
5 A.M. and the minimum always at 5 P.M.

5. 100-2 ft. c.: This group showed initial amplitude depression with a rapid
increase during the five days in darkness. Now, the maximum evidently was
between 5 and 11 P.M. and the minimum between 5 and 11 A.M.

6. 100-0 ft. c.: This was the only one of the experimental groups which had
undergone a complete reversal of phases. The amplitude was very great from
the start and showed no increase. The maximum pigment dispersion was seen
at 11 P.M. and the minimum at 11 A.M. in every cycle.

7. 50—10 ft. c.: This group showed little or no initial depression of amplitude.
The maximum occurred at 5 A.M. and the minimum at 5 P.M. in every instance.

8. 50—5 ft. c.: This group initially showed not only great depression in rhythm
amplitude, hut initially almost an absence of a recognizable daily cycle. A clear
daily cycle did reappear in two or three days and gain in amplitude. But now,
strangely, the maximum was close to or at 11 A.M. and the minimum at 5 A.M.

9. 50—2 ft. c.: This group also showed an initial low amplitude of rhythm, but
one which increased rapidly. The maximum value was at 11 P.M. in every cycle
and the minimum at 5 A.M. in all but one.

10. 25—10 ft. c.: An initial amplitude depression was observed in this group but
it rapidly disappeared. The time of maximum dispersion appeared to lie between
5 and 11 A.M. and the minimum was always at 11 P.M.

11. 25—5 ft. c.: The rhythm of this group exhibited an initial depression, and
the maximum was between 11 A.M. and 5 P.M. with the minimum at 11 P.M. in four
out of the five cycles.

12. 25-2 ft. c.: In this group, there was an initial depression in amplitude which
rapidly vanished ; the daily cycle was reasonably symmetrical with an unequivocal
maximum at 5 P.M. and minimum at 5 A.M.

13. 10—5 ft. c.: A. great reduction in rhythm amplitude was initially seen in
this group. The time of maximum pigment dispersion appeared clearly to lie
between 11 A.M. and 5 P.M., and the minimum close to 11 P.M.

14. 10-2 ft. c.: Again, the amplitude gradually increased during the five days
in darkness. The time of the maximum was sometimes seen at 11 A.M. and some-
times at 5 P.M. The minimum, on the other hand was distributed between 11 P.M.
and 5 A.M. during the five daily cycles.

15. 5-2 ft. c.: The amplitude for this group was rather low throughout the
five days. The minimum was invariably at 11 P.M. and the maximum nearly always
at 11 A.M.

16. 2—0 ft. c.: There was only slight, rapidly transitory amplitude depression
in this group. The maximum value was clearly at 5 P.M. and the minimum at 5 A.M.

DISCUSSION AND CONCLUSIONS

Viewing the data of Table I as a whole, it is clearly evident that even though the
animals have all received in common a higher illumination during 12 hours from
7 P.M. to 7 A.M. and a lower one from 7 A.M. to 7 P.M., there is to be found among
the results a whole spectrum of apparent kinds and degrees of shifts of the phases

312

BROWN, FINGERMAN AND MINES

of the daily rhythm. Furthermore, for any given illumination combination, al-
though the amplitude of the rhythm might undergo considerable change during the
five-day period of observation after the animals were left in darkness, the phases of
the rhythm in no case showed any evidence of a drift in one direction or the other.
The forms of the daily variations of Table I were essentially confirmed on the last
day of the five-day series when staging of chromatophores was performed hourly.
In order to obtain a better estimate of the direction and amount of shift in the
times of the phases of the rhythms in the experimental crabs away from the con-

">

LJ
Ct
O

5
4

3

2

LJ 4
O

0

I

5
4
3
2

A

0-0

00-80

\

00-10

S,

00-2

00-0

50-10

50-5

50-2

25-10

25-5

25-2

10-5

10-2

5-2

2-0

A.M.

II

A.M.

II

A.M.

II

A.M.

II II

A. M. A. M.

II
A.M.

II

A.M.

FIGURE 1. The average form and phase relations of the daily melanophore rhythm of
Uca pugnax in constant darkness after five days of subjection to various higher illuminations
by night and lower ones by day. The night-day illumination combinations are indicated for
each rhythm.

SHIFT IN DIURNAL RHYTHM BY LIGHT

313

dition in the controls, it was considered a reasonable procedure to average the
values for each time of day for the five days for each group, plot these averages
against time of day, and draw smooth curves. This has been done, and two average
daily cycles are illustrated in Figure 1. Obviously, the amplitudes illustrated are
only the mean ones for the five-day periods, but the forms of the curves and the
relationship of the times of their various phases to the actual hour of the day are the
factors of chief concern in this consideration. These two factors showed no sig-
nificant alteration during the five days as is quite evident from Table I.

An examination of Figure 1 clearly shows that the control curve obtained
in continued darkness can be illustrated as a more or less sinusoidal one with a

TABLE II

The number of hours by which the phases of the persistent daily rhythm are shifted forward ( + ) or

backward ( — ) by alternating periods of brighter •illumination by

night and dimmer illumination by day

Shift of
maximum (hrs.)

Illumination
night-day (ft. c.)

Shift of
minimum (hrs.)

Illumination
night-day (ft. c.)

+ 12

100-0

+ 12

100-0

50-2

+9

100-2

+9

100-2

50-2

+6

25-2

+6

25-2

10-2

10-2

2-0

2-0

+3

25-5

+3

50-5

10-5

0

50-5

0

25-5

5-2

10-5

5-2

-3

25-10

-3

25-10

-6

100-80

-6

100-80

100-50

100-50

100-10

100-10

50-10

50-10

maximum at 11 A.M. and a minimum at 11 P.M. In sharp contrast with this control,
those crabs which had been subjected to 100 ft. c. by night and darkness by day,
though similarly capable of depiction as a relatively simple sinusoidal rhythm, were
in almost exactly opposite phase.

Although the great majority of the experimental groups appear capable of de-
scription in terms of simply a change in amplitude of the cycles, together with more
or less displacement forwards or backwards in time relative to the control, there are
a few that appear quite definitely to have undergone a modification of form capable
of approximate description in terms of the times of maximum and minimum having

314 BROWN, FINGERMAN AND HIKES

been displaced to different extents away from the controls. This is evident in the
50-2 ft. c. group in which the minimum appears to have been displaced to the
right by only about 9 hours while the maximum was being shifted by 12 hours.

Comparable differential shifts are also apparent in the 25-5 ft. c. group, where
the minimum appears unshifted while the maximum is moved about three hours
to the right, in the 10-5 ft. c. group where almost exactly the same situation obtains,
and in the 50-5 ft. c. group where the maximum is probably unshifted and the
minimum moved to the right, by about three hours.

The differential shifts just described give rise to persistently skewed daily cycles
as is evident from all of these curves.

If one considers the 100-0 ft. c. group with a completely reversed rhythm as
having both maximum and minimum displaced to the greatest extent, and this
to be 12 hours to the right, or forward, in each case, all the other experimental
groups tend to fall naturally into a series of lesser amounts of shift to the right,
through no shift, and finally to a maximum amount of shift to the left, or backward,
of 6 hours. This graded series is described in Table IT.

Study of Table II reveals that with 100 ft. c. by night and darkness by day,
both maximum and minimum points in the daily cycle are considered as shifted
12 hours forward. For the same illumination by night, an increase in the il-
lumination by day progressively decreases the amount of the shift. With 2 ft. c.
by day, the shift is only about 9 hours; with 10 ft. c. by day, the shift is 6 hours
backwards, and this value is not exceeded for 50 and 80 ft. c. by day. Similarly for
50 ft. c. by night the greatest amount of shift, 9 to 12 hours forward, occurs when
the day value is 2 ft. c. At 5 ft. c. by day the amount of shift has dropped to 0 to 3
hours; and elevating the day to 10 ft. c. produces again the maximum shift back-
wards of 6 hours.

A comparable series is seen for 25 ft. c. by night. Two ft. c. by day yields a
6-hour shift forwards, 5 ft. c. a 0- to 3-hour shift forwards, and 10 ft. c. a 3-hour
shift in the opposite direction. Similarly for 10 ft. c. at night, 2 ft. c. by day gives
a 6-hour forward shift, while 5 ft. c. by day yields only a 0-3-hour one.

Five ft. c. by night and two by day produces no change in either direction. Two
ft. c. by night and darkness by day gives a 6-hour shift forward.

It is evident that 25 ft. c. at night is not sufficiently great to produce the maxi-
mum shift of 6 hours backwards, only 3 being possible. And 10 ft. c. by night
appears to be capable of producing no backward shift whatsoever.

These results suggest that for darkness by day, there is a direct relationship be-
tween the number of hours of shift and the illumination by night. One-hundred ft.
c. gave the maximum shift of 12 hours forwards; two ft. c. gave only about 6 hours
forwards. It seems reasonable to postulate that for other values between 100 ft. c.
and darkness, other degrees of shift ranging down to no shift at all might be found.

In summary, the amount and direction of shift of the phases of the persistent
diurnal rhythm appear to be determined in these experiments by at least two
factors. One is the strength of the stimulus in the form of a light increase at 7
P.M., and the other is the intensity of the illumination during the period from 7 P.M.
to 7 A.M. A minimum intensity of 50 to 100 ft. c. during the 7 P.M. to 7 A.M. period
is necessary to produce the maximum 6-hour shift backwards, and the minimum
strength of the stimulus of illumination change at 7 P.M. necessary to produce the

SHIFT IN DIURNAL RHYTHM BY LIGHT

315

total 12-hour forward shift is produced by some light change between 0-100 or 0-
50 ft. c. on the one hand and 2-100 or 2-50 ft. c. on the other.

All of these results, and others that have been obtained in previous work with
respect to shift in phases of the diurnal rhythm by illumination changes, are capable
of being explained in terms of one hypothesis which will now be presented.

Let it be assumed that the endogenous rhythm in those crabs is one in which the
general form of some key aspect of the rhythm can be described as illustrated in
Figure 2. Instead of being composed of symmetrical cycles, it is skewed so that
one limb is of about 6 hours in duration and the other one about 18 hours. Let
the normal relationship of the phases of this endogenous rhythm to the solar day be

A. NORMAL

B. 0-100 FT. C.

C. 50-100 FT. C.

A.

B.

C.

I

1

12 P M,

12PM.

12PM.

12 PM.

FIGURE 2. Diagram describing an hypothesis for the mechanism of inducing persistent
shifts in rhythm phase by light stimuli. Solid curve — normal phase relations. Broken curve
—rhythm reversed by a strong illumination change stimulus at 7 P.M. Dot-dash curve — 6-
hour backward shift by bright illumination from 6 P.M. to 6 A.M. (See text for discussion.)

as indicated in the figure, with the minimum occurring at about midnight and the
maximum at about 6 A.M. Since in nature this maximum is normally correlated
with the rapid morning increase in illumination and the succeeding 12 hours of the
rhythm is normally correlated with the period of daytime, let it be assumed that
both of these are normally involved in the bringing of the phases of the endogenous
rhythm into their usual and characteristic relationship to the daily light cycle.

In this hypothesis, a strong stimulus in the form of a large increase in illumina-
tion at 7 P.M. (e.g., 0 to 100 ft. c. . . .) induces the endogenous state normally
correlated with the maximum in the endogenous cycle, or what would amount to a
displacement of the phases of the rhythm 12 hours to the right or forward. Smaller

316 BROWN, FINGERMAN AND HINES

increases in illumination at 7 P.M. (e.g., 2 to 100, 2 to 25, and 5 to 10 ft. c.) would
have progressively less effect and the cycle of the endogenous rhythm would be dis-
placed progressively less to the right and only to a degree that the displaced cycle
at 7 P.M. was brought into an equilibrium for the light-increase stimulus. To an in-
tensity change represented by the 5 to 50 or 2 to 5 ft. c. shifts at 7 P.M., the endogenous
rhythmic mechanism appears normally to be in equilibrium. At still lower strengths
of the "shift stimulus," e.g., 10-25 and 10-50, and 10-100 ft. c., the strength can
be considered less than the equilibrium one, but now the phases of the rhythm will
not automatically shift backwards. The backward shift, if permitted by the
intensity-change stimulus strength, is induced by the continuing illumination. A
value higher than 25 ft. c. during the 7 P.M. to 7 A.M. period is necessary to move
the phases of the rhythm backwards to the maximum extent of 6 hours, the limit
being determined possibly by the correlation of the time of minimum strength of the
light-shift stimulus at 7 P.M. with the minimum in the endogenous daily rhythm.
It seems reasonable to assume that the continuing illumination exerts its backward
shifting action at the time of the ascending limbs of the cycle, namely between about
midnight and 6 A.M., but that this cannot occur except in the absence of a threshold
light change at 7 P.M. Illuminations of 25 ft. c. or below produce less shift, back-
ward, as a direct function of intensity.

An endogenous daily rhythm curve of the kind illustrated in the hypothesis is
not entirely without experimental support. This postulated one has almost exactly
the same form and phase relations with respect to the day-night cycle as has the
rhythm of retinal-pigment movement in the shrimp, Palaemonetes (Webb and
Brown, 1953). Uca pugilator melanophores in autotomized legs also exhibit a 6-
1 8-hour daily cycle.

In addition to accounting readily for all the results in the current complex series
of experiments, it also explains readily the well-known shift of 6 hours backwards ob-
tained by Webb (1950) by three consecutive daily periods of bright illumination from
12 midnight to 6 A.M., and then a few days later still another backward shift of 6
hours to a total of 12 hours, by three consecutive daily periods of illumination from
6 P.M. to 12 midnight. It also provides an explanation for the value, 6 hours,
which in much of the initial work on the mechanism of shift of the endogenous
rhythm, appeared to come forth with an inexplicably high frequency.

Speculating further upon the actual nature of the physiological processes in-
volved in these light-induced shifts probably would not be very productive at this
time. One of numerous possibilities could be that the hypothetical curve describing
the diurnal rhythm is a curve describing the intensity of a physiological state which
may be altered by light stimuli. A change from darkness to light at any time during
the endogenous reduction of this state could elevate it in proportion to the strength
of the stimulus. Once abruptly altered in this manner, the endogenous, tempera-
ture-independent mechanism could take over with the cycle exhibiting a renewed
start at a point in the cycle normally characterized by this higher level. Increase in
level of this rhythmic state would be the equivalent, during the descending portion
of the curve, to moving the phases of the cycle to the right. On the other hand,
the ascending limb of the curve describing an increase in the intensity of the state
of the rhythm could be capable of being accelerated by light up to the degree that is
nearly instantaneous, provided the phases of the rhythm had not been rigidly de-

SHIFT IN DIURNAL RHYTHM BY LIGHT 317

termined at an earlier stage in the same cycle by threshold change from light to
darkness for that particular phase of the cycle. This would amount to a shift of
the phases of the cycle to a maximum of 6 hours to the left. This would not shift
further to the left, because the presence of a sub-threshold dark-to-light stimulus for
7 P.M. would earlier in each cycle have freed the cycle to move to a point with the
minimum at 7 P.M.

SUMMARY

1. A study was made of the mechanism of reversal of phases of the persistent
daily rhythm in the fiddler crab, Uca pugnax, by illumination by night and darkness
by day.

2. Fiddler crabs were subjected to a series of combinations of brighter il-
lumination by night and dimmer illumination by day.

3. A graded series of amount of shift was obtained which was capable of being
interpreted in terms of two operating factors: (a) the strength of the stimulus in
the form of the dark to light change, and (b) the absolute brightness of the higher
illumination.

4. An hypothesis is advanced which appears to account adequately for all
currently known characteristics of the mechanism of persistent shift in phases of
the daily rhythm by light stimuli.

LITERATURE CITED

ABRAMOWITZ, A. A., 1937. The chromatophorotropic hormone of the Crustacea ; standardiza-
tion, properties and physiology of the eyestalk glands. Biol. Bull., 72: 344—365.

BROWN, F. A., JR., M. FINGERMAN, M. I. SANDEEN AND H. M. WEBB, 1953. Persistent diurnal
and tidal rhythms of color change in the fiddler crab, Uca pugnax. J. Exp. Zoo/., 123:
29-60.

BROWN, F. A., JR., AND H. M. WEBB, 1948. Temperature relations of an endogenous daily
rhythmicity in the fiddler crab, Uca. Physiol. ZooL, 21 : 371-381.

BROWN, F. A., JR., AND H. M. WEBB, 1949. Studies of the daily rhythmicity of the fiddler crab,
Uca. Modifications by light. Physiol. Zool, 22: 136-148.

HOGBEN, L. T., AND D. SLOME, 1931. The pigmentary effector system. VI. The dual character
of endocrine coordination in amphibian colour change. Proc. Roy. Soc., London, Scr. B.,
108: 10-53.

WEBB, H. M., 1950. Diurnal variations of response to light in the fiddler crab, Uca. Physiol.
Zool, 23: 316-337.

WEBB, H. M., AND F. A. BROWN, JR., 1953. Diurnal rhythm in the regulation of distal retinal
pigment in Palaemonetes. /. Cell. Comp. Physiol., 41: 103-122.

THE DISTRIBUTION OF PHOSPHORUS (P31 AND P^) IN
DORSAL AND VENTRAL HALVES OF THE
RANA PIPIENS GASTRULA1

PHILIP GRANT 2. a
Department of Zoology, Columbia University, Neiv York 27, N. Y.

Studies of the distribution of enzymatic activity and oxygen consumption
in the amphibian gastrula have led to hypotheses concerning the role of metabolic
gradients in development (Child, 1941). Although considerable confusion exists
as to whether such gradients have been established, the appearance of the dorsal
lip does delineate a region of distinct morphogenetic activity from one presumably
less active (the ventral half), and does suggest that a comparison of their respec-
tive metabolic activities would reflect these morphogenetic differences.

Considering the important biological and metabolic role of phosphate com-
pounds, it was felt that a study of their distribution between these two morpho-
genetically distinct regions would reveal specific metabolic differences of a more
convincing nature. Furthermore, radioactive phosphorus was used to make pos-
sible an analysis of shifts of phosphorus, either from one region to another, or from
one component to another.

METHODS

Rana pipiens females were weighed and injected with pituitary glands to induce
ovulation. They were then injected intraperitoneally with approximately 0.1 me.
of P32 in the form of H:!PO4. Forty-eight hours later, the eggs were harvested
and fertilized. They were allowed to develop in large finger bowls at 15° C. until
Shumway stage 10.

The jelly and vitelline membrane were removed and the gastrulae were then
dissected into two halves as shown in Figure 1. Dorsal and ventral halves were
collected in separate stender dishes standing in an ice water bath. From twenty to
forty halves were transferred to 12-ml. graduated centrifuge tubes and washed twice
with full strength Holtfreter's solution. All operations were carried out in full
strength Holtfreter's solution in vessels kept in ice water and the homogenization
and extraction were completed in a 4° C. cold room.

The fractionation procedure was a modified Schmidt- Thannhauser extraction
(1945). The fractions isolated were the following: (1) total acid-soluble phos-
phorus, (2) "desoxyribonucleic acid phosphorus," (3) "ribonucleic acid phos-
phorus," (4) "phosphoprotein phosphorus," (5) "phospholipid phosphorus," and
(6) residue phosphorus.

1 This research was supported in part by a grant from Public Health Service, National
Institutes of Health, administered by Dr. L. G. Earth.

- Postdoctoral Public Health Research Fellow, National Cancer Institute.
3 Present address: Institute for Cancer Research, Philadelphia 11, Pa.

318

PHOSPHORUS IN THE FROG GASTRULA 319

After alkali digestion of the defatted material, the supernatant was poured off,
DNA was precipitated by the addition of an HC1-TCA mixture and DNA phos-
phorus determined according to the procedure outlined by Sze ( 1953) . The residue
remaining in the alkali digest was analyzed as "residue phosphorus."

After precipitation of DNA, the remaining supernatant was precipitated with
magnesia mixture overnight to obtain the inorganic phosphorus liberated from
phosphoprotein. The resulting filtrate was hydrolyzed in 60^ perchloric acid
and analyzed as "ribonucleic acid phosphorus."

All fractions isolated were hydrolyzed in 60c/( perchloric acid in a sand bath
until clear, and inorganic phosphorus was precipitated as the magnesium am-
monium complex with magnesia mixture. The precipitates were collected on filter
paper and mounted on brass discs for counting, which was done with a Geiger-
Muller end window tube (3.3 mg./cm.-) using a Nucleonic RC 2 sealer. All
samples were corrected for decay and the instrument was checked daily against
a standard beta source.

V i 0

FIGURE 1. Dissection of gastrula. V = ventral half; D = dorsal half.

The precipitates were eluted in 1 N sulphuric acid and the phosphorus determined
by the method of Berenblum and Chain (1938) using the vessel described by
Wiame (1947).

Six separate experiments were completed, each run in duplicate, with ap-
propriate reagent blanks.

The fractions isolated represent heterogeneous groups of phosphate compounds
with a wTide range of different origins (Grant, 1953). Furthermore, there is
considerable doubt as to the extent of purity of these isolated fractions, particularly
those fractions that may be contaminated with inorganic phosphorus (Davidson
et al., 1951). Because of the relative nature of the data, however, it was assumed
that any significant differences between dorsal and ventral halves should be evident
using this technique.

RESULTS AND DISCUSSION

In Table I, the distribution of phosphorus (P31 and P32) in dorsal and ventral
halves of the gastrula is shown, in absolute values and in percentages of total
phosphorus and total radioactivity. No significant differences are evident in any
of the fractions. In two experiments, whole gastrulae were extracted along with

320

PHILIP GRANT

the halves to determine the efficiency of recovery, which was fairly good. The
low recoveries of acid-soluble phosphorus and phospholipid phosphorus may be at-
tributed to loss of blastocoel fluid in the case of the former and loss of yolk granules
during the dissection procedure in the case of the latter. The data do illustrate that
dorsal and ventral halves have a similar distribution of phosphorus and that cleavage
produces a uniform apportionment of the egg constituents.

In a recent study of the regional chemical differences in the frog gastrula
(Barth and Sze, 1953), gradients of lipid and of total nitrogen (animal- vegetal)
were demonstrated ; however, no dorso-ventral gradient was evident, which agrees
with the absence of a phosphorus gradient shown in Table I. Although the
analyses of Barth and Sze were performed on several small regions of the gastrula,
their data, calculated on the basis of dorsal and ventral regions (to make them
approximately equivalent to half-gastrulae analyzed here), exhibit no dorso-ventral
differences.

TABLE I

Distribution of phosphorus (P31 and P32) in whole gastrulae and in dorsal and ventral halves

Fraction

Whole

Dorsal half

Ventral half

Hgms.
psi

% total
psi

% total

P32

/igms.

P31

% total
psi

% total

P32

Mgms.

P31

% total
psi

% total

P32

Acid-soluble P

77.0

5.35

90.00

32.5

5.10

86.74

30.5

4.34

87.58

Ribonucleic acid P

45.0

3.18

1.03

20.0

2.82

2.82

25.5

3.65

2.33

DNA P

5.0

0.39

1.34

2.5

0.40

1.23

2.0

0.33

1.14

Phospholipid P
Phosphoprotein P
Residue P

283.5
974.5
58.0

19.65
67.55
3.95

4.78
2.41
0.18

112.8
489.5
15.0

17.47
73.68
3.46

5.79
4.02
0.75

127.7
517.3
23.5

18.12
72.81
4.38

6.09
3.20
0.97

Values for Mgms. P31 expressed as micrograms of phosphorus per 100 embryos or per 100 half-
embryos. Values for % total P32 obtained from values expressed as counts per minute per whole
or half-embryo.

The values for all fractions, except DNA phosphorus and RNA phosphorus,
compare closely with Kutsky's (1950) results. The low values for these latter
two fractions may have been due to a failure to obtain complete precipitation with
magnesia mixture, since the amounts involved are relatively small. In addition,
loss of RNA phosphorus may have been due to adsorption onto the magnesium
ammonium precipitate of phosphoprotein phosphorus. However, the relative values
are significant and these indicate that no differences exist.

The per cent distribution of P32 also illustrates that no significant differences are
apparent when the halves are compared to each other, or to the whole embryo.

When the number of cells in the two regions is considered (Sze, unpublished
data), Table II is the result. Since the ventral half contains fewer, larger cells
(approximately 15,800 cells with an average volume of 67,500 ju,3 compared to the
dorsal half with 18,100 cells with an average volume of 52,600 p.3) the results are
to be expected. The larger ventral cells contain a greater proportion of cellular
constituents, particularly yolk granules, which represent about 70% of total egg
phosphorus (Grant, 1953). Thus, phosphoprotein phosphorus and phospholipid

PHOSPHORUS IN THE FROG GASTRULA 321

TABLE II

Distribution of phosphorus (I)3]) per cell of half gastrulae
Fraction Dors.il Ventral

Acid-soluble P 0.178 0.193

Ribonucleic acid P 0.109 0.161

Desoxyribonucleic acid P 0.014 0.013

Phospholipid P 6.155 8.075

Phosphoprotein P 26.750 32.750

Residue? 0.082 0.149

Values expressed as micrograms P31 per cell X 105.

phosphorus, the major constituents of yolk phosphorus (Panijel, 1950), exhibit the
greatest differences.

The specific activity data (Table III) suggest that differences between dorsal
and ventral halves may exist. However, these differences are insignificant when
tested by the comparison of individuals method. The high specific activity of the
acid-soluble fraction in the whole embryo may be attributable to the retention of
blastocoel fluid, possibly rich in highly active inorganic phosphorus.

It is possible that differences could be made more evident (that is, if they exist)
if smaller regions of the gastrula were compared ; regions similar to those analyzed
by Barth and Sze (1953). The dissection into half gastrulae includes large areas
of tissue of similar metabolic activity such that small differences between halves are
masked. Possibly, in later stages of gastrulation, where metabolic differences are
more pronounced (Brachet, 1950), dorsal and ventral halves would exhibit di-
vergencies in their phosphorus distribution and specific activity.

The residue phosphorus exhibited activities of the same order of magnitude as
the ribonucleic acid fraction, suggesting that it might be undigested nucleic acid.
It is also possible that the residue might be metaphosphate as described by Wiame
(1947) in yeast. The possible existence of metaphosphate is interesting in the
light of recent findings by Berg (unpublished data) that there is a strong meta-
phosphatase present in the developing embryo.

No specific activity values are reported for DNA phosphorus since those ex-
periments which yielded activities for this fraction failed to yield detectable amounts
of phosphorus.

TABLE III

Relative specific activity X 10s.
Whole and half gastrulae

Fraction

Whole

Dorsal

Ventral

Acid-soluble P
Ribonucleic acid P
Desoxyribonucleic acid P
Phospholipid P
Phosphoprotein P
Residue P

202.25
4.15

142.48
7.16

141.72
4.39

2.98
0.38
0.58

2.78
0.43
3.43

2.38
0.31
5.09

cts./min./Vgms. P
Values expressed as

act. injected/wt. gms.

322 PHILIP GRANT

Kutsky (1950) reports significant shifts in activity from gastrulation to neuru-
lation. These have also been found by the author (unpublished data). This
points to the need for continued study of more advanced stages of gastrulation to
demonstrate and localize these changes. This report represents the completed por-
tion of such experiments now in progress.

SUMMARY

1. A preliminary investigation of the distribution of phosphorus in dorsal and
ventral halves of the Rana pipiens gastrula was undertaken to demonstrate the
possible existence of a metabolic gradient of phosphate compounds correlated with
the apparent morphological gradient. Radioactive phosphorus was employed to
permit an analysis of shifts of phosphorus.

2. A modified Schmidt-Thannhauser extraction procedure was applied to
dorsal and ventral halves of stage 10 (Shumway) embryos obtained from a frog in-
jected with approximately 0.1 me. of P32 before inducing ovulation. Total acid-
soluble phosphorus, RNA phosphorus, DNA phosphorus, phosphoprotein phos-
phorus, phospholipid phosphorus and a residue phosphorus were extracted and
analyzed for specific activity.

3. The distribution of phosphorus (P31 and P32), expressed either as micro-
grams P31 per half embryo or as per cent of total P31 or P32, exhibited no significant
difference between dorsal and ventral halves. However, expressed as micrograms
P31 per cell, a ventral-dorsal gradient was apparent. Data expressed as specific
activity (counts per minute per microgram P31) exhibit no significant differences.

LITERATURE CITED

EARTH, L. G., AND L. C. SZE, 1953. Regional chemical differences in the frog gastrula. Physiol.

Zool, 26: 205-211.
BERENBLUM, L, AND E. CHAIN, 1938. An improved method for the colorimetric determination

of phosphate. Biochem. /., 32 : 286-298.

BRACKET, J., 1950. Chemical embryology. Interscience Publishers Inc., New York, N. Y.
CHILD, C. M., 1941. Problems and patterns of development. University of Chicago Press,

Chicago, Illinois.
DAVIDSON, J. N., S. C. FRAZER AND W. C. HUTCHINSON, 1951. Phosphorus compounds in the

cell. I. Protein bound phosphorus fractions studied with the aid of radioactive phos-
phorus. Biochem. J., 49: 311-321.
GRANT, P., 1953. Phosphate metabolism during oogenesis in Rana tcinporaria. J . Exp. Zool.,

124: 513-544.
KUTSKY, P. B., 1950. Phosphate metabolism in the early development of Rana pipiens. J. Exp.

Zool, 115: 429-460.
PANIJEL, J., 1950. L'organisation du vitellus dans les oeufs d'amphibiens. Biochim. ct Biophys.

Acta, 5 : 343-357.
SCHMIDT, G., AND S. J. THANNHAUSER, 1945. A method for the determination of desoxyribo-

nucleic acid, ribonucleic acid and phosphoprotein in animal tissues. /. Biol. Chem., 161 :

83-90.
SZE, L. C., 1953. Changes in the amount of desoxyribonucleic acid in the development of Rana

pipiens. J. Exp. Zool, 122: 577-602.
WIAME, J., 1947. fitude d'une substance polyphosphoree basophile et metachromatique chez les

levures. Biochim. ct Biophys. Acta. 1 : 234-255.

THE RESPIRATION OF NORMAL LARVAE OF
TEREDO BARTSCHI CLAPP J

CHARLES E. LANE, J. Q. TIERNEY 2 AND R. E. HENNACY
The Marine Laboratory, University of Miami, Coral Gables, I'lorida

The free-swimming, infective larval stage of Teredo in local waters does not sig-
nificantly exceed seventy-two hours in duration (Isham and Tierney, 1953). Dur-
ing this time the animals have not been observed to feed (Lane, Posner and Green-
field, 1952). The pre-attachment activities of the animal must be presumed to be
powered chiefly by glycogen. This is deposited in the ovum in granular form
during oogenesis. Additional glycogen may be contributed to the larva during
the time that it is actually embedded in the maternal gill. At the termination of this
transient, free-swimming stage, the larvae attach themselves permanently to a
wooden substratum within which they spend the rest of their adult life span. A
cellulase enzyme system exists in both larval and adult Teredo (Greenfield and
Lane, 1953). This enzyme complex may significantly facilitate the invasion of
wood by the larvae.

The act of penetration of wood renders the larva virtually immune to environ-
mental hazard except for substances in solution either in the wood itself or in the
water which constitutes the respiratory stream. Thus it is that preventive meas-
ures, to be effective, must be directed against the larva during the vulnerable first
seventy-two hours of its free-living life.

A sensitive index of physiological condition, or of the effectiveness of sub-lethal
concentrations of toxic substances, is provided by the rate of oxygen consumption of
living systems. Some of the parameters of normal respiration in free-living, pre-
attachment stages of Teredo were delimited preliminary to a study of the effective-
ness of some toxic agents. Details of this latter phase of the investigation will be
presented elsewhere. It is the purpose of the present communication briefly to
describe the methods and some of the results observed in the study of normal
animals.

MATERIALS AND METHODS

All larvae employed in this study were reared in the laboratory by methods
described by Lasker and Lane (1953).

Oxygen consumption was measured in a capillary microrespirometer (Fig. 1).
It consists of a pear-shaped chamber blown in one end of capillary tubing. The
volume of the chamber varied in different respirometers over the range of six
to 125 microliters. The volume should be kept as small as possible to increase the
stability of the system (Tobias, 1943). The opposite end of the capillary tubing

1 Contribution No. 120 from the Marine Laboratory, University of Miami. These studies
were aided by a contract between the Office of Naval Research and the University of Miami in
cooperation with the U. S. Navy Bureau of Yards and Docks.

2 Present address : Hydrographic Office, U. S. Navy, Washington, D. C.

323

324

LANE, TIERNEY AND HENNACY

bears an inside syringe-taper ground joint. This seats in the outer matching
ground joint of the thermobarometer or compensation chamber. This latter portion
of the apparatus should be as large as is consistent with ease of manipulation. We
have generally sought to have its volume at least 1000 times that of the respirometer
chamber. This insures maximum sensitivity of the system. The upper end of the
compensation chamber is closed by a stopcock. In use the entire assembly is
immersed in a constant temperature water bath maintained at 25.0° C.

The chamber is first loaded with a single animal confined in a droplet of medium
whose volume varied for different respirometers between one and ten microliters.
This volume provides a mass of medium from 100 to 1000 times the volume of the
organism. The isolation of the larva and the determination of the volume of the
medium can be effected most easily by making use of specially drawn and calibrated
micropipettes. These may be actuated either by a syringe device or by a mouth-
piece similar to that of a hemocytometer pipette. Calibration of pipettes and other
micro-glassware is readily accomplished with the micrometer burette described by

cC

5=

-v

1

(

y////////A. t i i

•^v

^

•
A

^j

_/-4

B

FIGURE 1. Sketch of components of the capillary microrespirometer. A. Thermobarometer
with outer syringe-taper ground joint in end opposite stopcock. B. Microrespirometer with
matching inner syringe-taper ground joint in end opposite respirometer bulb. C. Enlarged view
of respirometer bulb to show the disposition of droplets and the kerosene indicator fluid.

Scholander (1942). Various loading pipettes may be calibrated to deliver pre-
cisely known total volumes. The delivered volume, of course, will include the
volume of the organism. Separate pipettes are constructed for each respirometer,
and are then used only with that particular apparatus.

The droplet of medium and larva is delivered onto one wall of the respirometer
chamber. The chamber wall is previously rendered hydrophobic by the application
of a suitable silicone coating. Under these conditions the integrity of the droplet
of medium is retained for long periods of time. It has, for example, frequently
been possible to make continuous observations of the oxygen consumption of a
single larva during periods as long as twenty-four hours without opening the
sealed system.

After the respirometer has been charged with the animal and medium, a droplet
containing one to five microliters of alkali, either 10% NaOH or 5% Ba(OH).,,
is placed on the contralateral wall. The indicator fluid in the capillary is kerosene
which has been distilled at 250° C. after exhaustive oxidation with concentrated
sulfuric acid for several days. The open end of the respirometer is sealed with a

RESPIRATION OF LARVAL TEREDO

325

non-oxidizing wax. For best adhesion and complete sealing it is preferable to
employ a wax of comparatively low melting point. With the upper stopcock of
the compensation chamber open, the two portions of the apparatus are united,
seated and the joint is sealed with the same wax which was used to close the lower
end of the respirometer. The assembly is then placed in the water bath and per-
mitted to come to temperature equilibrium. Under our conditions a steady state
is generally reached by respirometers without respiring tissue within thirty minutes.
With the system sealed and equilibrated the mode of operation is as follows.
The larva extracts dissolved oxygen from the sea water medium. This creates a
diffusion gradient across the air-water interface, as a consequence of which ad-
ditional oxygen diffuses into the water from the air phase. The loss of oxygen from
the air phase causes a decrease in pressure in the air phase which is reflected in
displacement of the kerosene meniscus. The position of the meniscus is observed
with a compound microscope equipped with a long-focus objective. It is clear that

10 20 30 40 50

TIME IN MINUTES

FIGURE 2. Average rate of oxygen uptake by 24-hour larvae of Teredo. Each point is the
average of from three to 17 determinations on different animals. (1) in the presence of
M/IOOQ glucose in sea water, (2) normal sea water and (3) blank.

the sensitivity of the apparatus is limited by the resolving power of the optical
system. In our studies we have found it convenient to detect a displacement of ten
microns. This represents a change in volume of 0.002 mm.3 In the terminology
proposed by Scholander and Evans (1947) this is 2.0 yl. Naturally the sensitivity
of the entire system could be increased either by increasing the magnification of
the optical system or by decreasing the diameter of the capillary out of which the
respirometer is constructed. For the present study sufficient sensitivity was pro-
vided by the dimensions described.

RESULTS

In Figure 2 are shown average oxygen consumption values for over one hundred
individual determinations. The lowest curve is the sea water blank. Inasmuch as
this generally consisted simply of the water in which the Teredids were living at
the time of their capture, it usually contained microorganisms which showed a small

326

LANE, TIERNEY AND HENNACY

but significant oxygen consumption. Blank figures were always subtracted from
the oxygen consumption of experimental animals. The middle curve shows the
average rate and magnitude of oxygen consumption by normal 24-hour larvae.
The upper curve shows the oxygen consumption by normal 24-hour larvae when
the sea water medium had been rendered 0.001 M by the addition of appropriate
amounts of glucose.

Figure 3 shows the change in rate of oxygen consumption with increasing age
of the larvae. The increase during the first twenty-four hours is real and sig-
nificant. From seventy-two hours to three hundred hours the decrease in oxygen
consumption is progressive. This decline is associated with general involutional
changes in the larva which will lead to its death by three hundred hours if it is
denied access to wood.

60

40

o

,o — o-

50 100 150 200

AGE IN HOURS

250

FIGURE 3. Oxygen, consumption by normal larvae of Teredo at various times after release
from the maternal gill. Each point is the average of all determinations made for that age.
Data on 38 different larvae are presented in the curve.

DISCUSSION

These data serve to emphasize both similarities and differences between the
larval behavior of Teredo bartschi Clapp and T. navalis. Imai, Hatanaka and Sato
(1950) have described the large-scale culture of larvae of T. navalis. They em-
phasize the importance of suitable supplies of food organisms in the maintenance of
normal growth of the cultures. They also figure the larvae at various stages of its
free-living life. In T. navalis this pre-attachment period may occupy as much as
34 days. During this time the larva continues to grow. The maximum pre-
attachment size is approximately 245 microns in diameter.

T. bartschi, on the other hand, has been showrn to be 250 microns in diameter
when it is released from the maternal gill. The length of its normal free-swimming
life does not exceed four days.

In a previous publication from this laboratory (Lane, Posner and Greenfield,
1952) the statement was made that larvae of T. bartschi "do not appear to feed."
It was thought that the large supplies of glycogen which are characteristic of the

RESPIRATION OF LARVAL TEREDO 327

mature oocyte were sufficient to power the pre-attachment activities of the larvae
during a short free-living life. There can now be little doubt that this conception
is erroneous in view of the observed oxygen consumption of the larva during this
portion of its life. Elementary calorimetric considerations show that larvae of
T. bartschi, which weigh approximately 10 micrograms alive, and contain close to
60% moisture, fall far short of containing sufficient metabolic fuel materials to
justify the measured oxygen consumption during their free-living existence. The
deficit must be made up by the ingestion of microorganisms from their environment.

The large glycogen stores probably represent emergency reserves which are
used after 72 hours. At this stage the larvae cease to swim and assume a pedestrial
mode of progression. Clearly they do not come into contact with the same number
of suspended food organisms when crawling in two dimensions as when swimming
relatively rapidly through three dimensions.

The increased oxygen consumption at twenty-four hours is correlated with be-
havioral changes in the free-swimming larvae which have been described by Isham
and Tierney (1953). These investigators have shown that crawling with the aid
of the muscular foot replaces swimming with the velar cilia as the chief method of
locomotion at this stage of development. The enhanced oxygen consumption during
this portion of the life cycle may also be related to the post-natal development of
enzymatic mechanisms for complete glycolysis.

The increased oxygen consumption in the presence of M/1000 glucose can be
most reasonably attributed to the increased metabolism of microorganisms present
in the medium along with the larva. The possibility of direct absorption of dissolved
organic materials by Teredo larvae should not be overlooked, but it is certainly
not proved by this work.

SUMMARY AND CONCLUSIONS

1. A capillary microrespirometer is described, with the aid of which the normal
respiration of free-living larvae of Teredo bartschi Clapp has been studied.

2. Oxygen consumption during the total three hundred-hour pre-attachment
life averages approximately 25 microliters per hour, when measured at 25° C.
At twenty-four hours there is a significant augmentation. After seventy-two hours
the rate of oxygen consumption decreases regularly until the death of the larva.
The significance of these alterations in rate is discussed. Differences between larval
behavior of T. bartschi and T. navalis are described.

LITERATURE CITED

GREENFIELD, LEONARD J., AND CHARLES E. LANE, 1953. Cellulose digestion in Teredo. J. Biol.
Chem., 204 : 669-672.

IMAI, TAKEO, M. HATANAKA AND R. SATO, 1950. Breeding of the marine timber-borer Teredo
navalis in tanks and its use for antiboring test. Tohokn J. Agricul. Res., 1 : 199-209.

ISHAM, L. B., AND J. Q. TIERNEY, 1953. Some aspects of the larval development and metamor-
phosis of Teredo (Lyrodus) pcdlccllata De Quatrefages. Bull. Mar. Sci., 2: 574-590.

LANE, CHARLES E., G. S. POSNER AND L. J. GREENFIELD, 1952. Distribution of glycogen in the
shipworm. Bull. Mar. Sci., 2 : 385-392.

LASKER, REUBEN, AND CHARLES E. LANE, 1953. Origin and distribution of glycogen in Teredo
bartschi Clapp. Biol. Bull., 105: 316-319.

SCHOLANDER, P. F., 1942. Microburette. Science, 95: 177-178.

SCHOLANDER, P. F., AND H. J. EVANS, 1947. Microanalysis of fractions of a cubic millimeter of
gas. /. Biol. Chem., 169: 551-560.

TOBIAS, JULIAN M., 1943. Microrespiration techniques. Pliysiol. Rev., 23 : 51-76.

NUTRITION OF THE SEA URCHIN, STRONGYLOCENTROTUS

PURPURATUS l

REUBEN LASKER AND ARTHUR C. GIESE
Hopkins Marine Station of Stanford University, Pacific Grove, California

The gut of the purple sea urchin, Strongylocentrotus purpiiratus, reveals a mass
of algae in various stages of decomposition. Algae contain relatively small amounts
of nutrients which are readily handled by enzymes ordinarily present in animals,
but they possess galactans, alginic acid, agar and possibly some cellulose, none of
which are digested by man or most animals. The urchins in the course of evolution
may have developed enzymes which have enabled them to use these materials, or,
like many of the ungulate herbivores, they might harbor bacteria, or, like the ter-
mites, they might shelter protozoans which perform this role for them.

The first study was therefore concerned with the role of the digestive enzymes
present in the gut of the urchin. The second consisted of studies of the digestive
action of the flora of the urchin gut. The third was concerned with the over-all
nutritional economy of the sea urchin.

MATERIALS AND METHODS

Sea urchins were collected on the Monterey Peninsula at Yankee Point below
Carmel Highlands and at Pescadero Point. A large number of urchins were planted
just outside the Hopkins Marine Station to make them available for experiments
requiring an occasional fresh specimen. While some of those transplanted died or
disappeared, a fairly large number took hold and fed upon the prevalent coralline
algae. Monthly studies were made on the urchins from Pescadero Point, because
they could always be obtained even in rough weather, whereas those at Yankee
Point sometimes became inaccessible in stormy weather. All the locations from
which the urchins were taken were free of industrial waste and relatively free of
sewage. The sea urchins brought in fresh monthly were kept in the laboratory in
aerated rapidly running sea water. Even so, most of them aggregated near the top
of the water in a tank, except when they wandered in search of food. Unless only
a relatively small number of urchins were kept in an aquarium, they became un-
healthy in time.

For determination of the sugar and nitrogen content in the body fluid, ten
urchins were sacrificed the first day and ten the second after collection. The pH
of body fluid was determined with a Beckman pH meter directly in the field, and

1 The study was supported by a grant from the National Science Foundation. The authors
are indebted to Dr. L. R. Blinks, Director of the Hopkins Marine Station and to Dr. C. B.
van Niel for suggestions, advice and help as well as friendly interest in the work, to Mr. F.
Falconer, Head Librarian of the Biological Libraries, for help in bibliographic research, to
Dr. Bachman-Beam for suggestions in handling the bacteria, to Mr. W. K. Bowen for preparing
the sections of the intestine, to Dr. W. Z. Hassid for a sample of iridophycin and to Mr. R.
Aughtry and Dr. R. Bolin for unpublished data on water temperatures.

328

NUTRITION OF A SEA URCHIN 329

the pH of gut contents as soon as possible after opening a normal, healthy urchin in
the laboratory.

The body fluid was withdrawn from an urchin by excising Aristotle's lantern and
pouring out the contents of the coelom. The fluid was allowed to stand and the
clot, containing wandering blood cells, was removed by filtration. Reducing sugars
were determined in the filtrate by the method of Somogyi (1945. 1952). The non-
protein nitrogen (NPN) in the coagulum-free filtrate of body fluid was determined
by the standard Kjelclahl procedure, the proteins being first precipitated with
trichloracetic acid (TCA).

EXPERIMENTAL

1. Structure of the diijesth'e tract of the sea urchin

The mouth opening between the teeth of Aristotle's lantern leads through an
oesophagus to the stomach-intestine which is suspended in the coelom by a mesen-
tery. The oesophagus possesses well developed villus-like papillae which contain
glands, perhaps producing mucus. The intestine has two turns ; looking in at the
mouth of the urchin the first turn is clockwise and the second, doubling back upon
this, is counter-clockwise. The intestine is also lined by a glandular epithelium in
which secretory cells are found, some forming glands possessing a body and a neck
which opens into the digestive cavity. The connective tissue of this portion of the
gut is inconspicuous and the epithelium appears to be bounded by the visceral
peritoneum. Since the wall of the gut is so thin it is probable that absorption can
occur readily through any part of it.

2. Feeding habits of tJie sea urchin

'Over the last ten years, sea urchins kept in aquaria have been fed a great variety
of foods. If starved, they were found to ingest almost anything offered them such
as boiled eggs, boiled potatoes and vegetables, as well as fresh vegetables, but not
leaves of geranium or Pelargonium. However, meat and fruits were taken in
preference to vegetables. In nature the sea urchin feeds upon various algae (green,
red and brown) as well as upon the "surf-grass," Phyllospadi.i'. In local areas
the diet may be largely restricted to the most abundant alga. In the laboratory, the
sea urchins were usually fed the red alga, Iridophycus flaccidum, because of its avail-
ability and its acceptability to the urchin.

The ingested food apparently remains in the gut for a long period of time since
during starvation in an aquarium, feces were ejected for two weeks, suggesting a
very slow rate of digestion. When urchins brought in from the field were roughly
handled, they eliminated considerable amounts of material for a short period.
When slowly eliminated, the algae in the feces were found to be fairly completely
decomposed and were heavily laden with bacteria. If feeding was continuous, so
was defecation, and the algal pellets appeared to be less completely digested.

When an urchin which had been starved for some time in the laboratory was
given food, it quickly fed to capacity with any material, algal or otherwise, upon
which tests were desired. Starvation for two weeks was used as standard prac-
tice since in this time the gut will have become considerably, if not completely, cleared
of contents.

Considering the low temperature of the water in which the sea urchins live, the

330 REUBEN LASKER AND ARTHUR C. GIESE

rate of digestion is not surprising. The water temperature for Moss Beach varied
from 9.2 to 15.5 and at Stillwater Cove from 8.7 to 16.1° C. during the year. On
the rare occasions when the urchins were exposed to sunlight at low tide the
temperature may have risen considerably over this.

3. Digestive enzymes of the intestine

Since the normal food of the purple sea urchin consists of algae, the digestive
enzymes most likely to be found in the gut are those which can handle the nutrients
found in the algae. The protoplasm of algal cells, of course, contains protein, and
floridean starch is stored in red algae. However, much organic material is
present in red algae in the form of galactans or galactans mixed with other ma-
terials, e.g., agar and various gums. Enzymes which can handle proteins (pro-
teases), starch (amylase) and the various substances peculiar to algae are there-
fore of special interest. Tests were made for each of these.

For extraction of enzymes, the gut contents were flushed out with sterile 3%
NaCl, and the gut rinsed three times in the salt solution. The entire digestive
tracts of several animals were then ground with crushed Pyrex glass in a mortar
with a small amount of water, extracted with buffer solution and centrifuged. The
buffer used in extraction was a 0.5 M Mcllvaine buffer of pH 6.8-7.0. This pH
was used because the first measurements of the pH of a sea urchin gut gave readings
of 6.8—7.0. Subsequent measurements indicate that a pH of 7.2 to 7.3 is probably
more nearly correct for a freshly opened gut. When the urchin is kept in the
laboratory for a brief time the pH falls.

That a protease is present was easily demonstrated by mixing the gut extracts
with casein, adding toluol to inhibit bacterial growth and determining the increase
in NPN with lapse of time. Since some NPN appears in the control without casein,
the data given in Figure 1 are for the differences between the two. The data
demonstrate that the NPN rises rapidly after action of the enzymes on the protein.

That an amylase is present was shown by the appearance of reducing sugar
in a sample of boiled starch mixed with gut extract (toluol as antiseptic). As
seen in Figure 1 considerable reducing sugar appears after action of the enzymes in
the extract on the boiled starch. As a control the gut extract was incubated without
starch and tested for appearance of reducing sugar.

Agar and other gums contained in the algae are made up principally of poly-
merized galactose often combined with other substances. The ability of the sea
urchin to digest agar was tested by adding the enzymes extracted from the gut to
warm agar (about 37-40° C.) and mixing. The results, shown in Figure 1, may
be considered negative since the very small change in reducing sugar concentration
following exposure to the extracts is probably within the limits of error of the
experimental method. The results for iridophycin, a galactan isolated from
Iridophyciis flaccidinn by Hassid (1933), indicate that an iridophycase is present
(Fig. 1). In all cases a control was incubated without the substrate and tested
for reducing sugar. It would be desirable to test other substances peculiar to
algae but they were not available in pure form for the tests.

4. Fauna and flora of the sea urchin gut

Ciliate protozoans occur in small numbers in .the sea urchin gut, about ten to
a hundred being counted per ml. of gut contents of several sea urchins. Several

NUTRITION OF A SEA URCHIN

331

species of Entorhipidium and Lcchriopyla mystax have been identified (Lynch,
1929a, 1929b). Few were ever seen in division by Lynch; therefore they seem
to represent a static population. The protozoans appear to ingest bacteria and
small particles of well-disintegrated algae. While the activities of the protozoans,
some of which live for several days outside the intestine, should be studied, it is
unlikely that the protozoans contribute to the digestion of the algae because of their
small numbers and their feeding habits.

Bacteria are present in the gut of the sea urchin in sufficient numbers to be of
consequence. Almost every pellet of algal material in the second section of the
intestine was found to be surrounded by a translucent membrane which, upon

Q starch
O iridophycin
aga r

casein

8

16

24
Time (hrs.)

32

40

FIGURE 1. Rate of digestion of 1% casein by the extract of the intestine of the purple sea
urchin at pH 6.8 and 30° C, and of \% boiled starch, 0.1% agar and 0.02% iridophycin by the
extract of the intestine of the purple sea urchin at pH 6.8 and 30° C.

microscopical examination, proved to be a film lined with coccus and rod-shaped
bacteria. The films develop soon after the algal pellets enter the gut. When
defecated the enveloped pieces were largely digested and no longer recognizable
as algal pellets. An unenveloped piece of alga may become colorless but the cell
walls remain intact and while no bacteria are seen within the walls, they occur
around the cells damaged by the teeth. The envelope of mucilaginous material
develops in the gut of the sea urchin, but not in cultures of the bacteria from the sea
urchin gut growing on algae in vitro.

Since the enzymes from the sea urchin gut do not digest intact algal tissue, yet
the latter disintegrates in the gut, it would appear that either the algae autolyze or

332 RKUr.HX LASKER AND ARTHUR C. GIESE

the bacteria digest them. Experiments eliminated the first possibility : algae in sea
water under toluene did not autolyze. Attempts were therefore made to determine
whether the bacteria from a sea urchin were capable of digesting the algae. An
autoclaved sample of the red alga Iridophycus was inoculated with a loopfull of
bacteria removed aseptically from the hind gut of a sea urchin. In one week at room
temperature the algae had completely decomposed. The experiment was repeated
several times.

Next, to test whether the bacteria in the lumen of the gut digest agar, a constitu-
ent of the cell walls of many algae, a sample of gut fluid was aseptically removed and
transferred onto a 2% agar-sea water plate. The sample was poured over the
plate or spread over it with a sterile glass spreader. Many colonies appeared within
a few days. These were identified by the pits which they produced in the agar.
Other colonies which did not form deep pits were detected by staining the agar
surface with iodine. A clear area was noted around agar-decomposing colonies.
The number of agar-digesting bacteria found in various tests was of the order of
106 per ml. of gut contents. For an urchin fed the red alga, Iridophycus flaccidum,
the counts indicate 1.8, 2.1 and 1.4 X 106 bacteria and for one fed the brown alga,
Alaria marginata, 1.7, 1.8 and 0.7 X 10fi. In a similar sea urchin, the total number
of bacteria determined with a Petroff Hausser counting chamber was 2.6 X 1011.
In an urchin starved for a week after being fed Iridophycus, 1.0 X 106 agar de-
composers appeared out of a population of 3.3 X 106.

Several individual colonies transferred to fresh agar-sea water plates were found
to grow quite well on agar after a number of transfers, but some required additional
nutrients or growth factors which could be supplied in the extract of algae or yeast
extract. For culture purposes, the former was more convenient.

A loopfull of a pure culture of agar-decomposing bacteria was inoculated into
a sample of autoclaved Iridophycus flaccidum. Within a week, at room tempera-
ture, the alga had completely disintegrated, therefore some of the agar-decomposing
bacteria are capable of digesting this alga.

5. Possible role of the intestinal flora in nutrition

While some of the bacteria of the sea urchin intestine are capable of digesting
the algae ingested, a symbiotic relationship between the bacteria and the sea urchin
is not thereby proven. It is possible that the sea urchin maintains itself on the more
readily available nutrients in the algae which it is capable of digesting — i.e., the
floridean starch and the proteins of the algae. The presence of enzymes capable
of digestion of starch and protein supports this possibility.

On the other hand, the intestinal flora may render the nutrients in the cells
of the algae more readily available to the sea urchin, by digesting the cell walls.
However, it seems unlikely that the bacteria would spare the more generally
utilizable nutrients such as starch and protein while selectively digesting the gen-
erally less available cell wall materials. It seems more likely that the enzymes
present in the foregut of the sea urchin digest the more readily available materials
in the algal cells before the bacteria have multiplied sufficiently to offer competition,
and that the residue is then attacked by the bacteria which gradually decompose the
algal cell walls.

The bacteria nevertheless may contribute to the host by digesting the structural

NUTRITION OF A SEA URCHIN

333

components of the algae and releasing some of the nutrients which can then he
ahsorbed by the host into the body tlnid. This would constitute a type of symbiosis.
Sugar is mobilized (Table I) in the body fluid of a sea urchin soon after a meal
of algae but this could be explained adequately as a result of digestion of starch in
the algae by the amylases of the gut. Reducing sugars did not accumulate in
cultures of the agar-digesting bacteria tested. It is possible that they are present
only transiently in the intestine and might be absorbed to a small extent. Cultures
of bacteria tend to become acidified and fatty acids may be demonstrated in the
culture fluid and in the body fluid of the sea urchin. Identification tests indicate
lactic acid, judging from the position of the spot in chromatographic analysis (Reid
and Lederer, 1952) and the Friedemann-Graeser determination (1933).

TABLK I

Reducing sugar in the body fluid of the sea urchin after starvation for two weeks
and ref ceding or injection of glucose

Fed

Days fed

Weight of food
ingested in grams

Wet weight
of urchin

mg. % reducing
sugar

Iridophycus (red alga)

0

1

0
1.33

37.5
42.4

0

12.0

2

2.07

57.1

4.5

4

3.43

40.5

1.3

8

6.86

41.9

1.0

Boiled potato

0
1

0
2.89

36.9
47.8

0
47.0

2

4.69

48.4

62.0

4

4.55

38.5

1.7

8

10.09

43.9

24.3

Glucose injected

Hours time lapse

Mg. glucose injected

% glucose withdrawn

0

0.0

34.5

0

1

1.0

54.2

64.5

2

1.0

41.4

78.0

4

1.0

39.0

95.8

8

1.0

37.9

88.5

The nature of the relationship between the sea urchin and the bacteria could be
ascertained if the bacterial flora of the intestine could be removed and replaced at
will. Killing the flora by a meal of CuSO4, as has been done with some vertebrates,
kills the sea urchin as well. Defloration by high oxygen tensions is not practical
since the bacteria present in the gut are only facultative anaerobes, not obligate like
the termite protozoa. Raising sea urchin larvae aseptically to the adult stage is
not likely to be successful with present techniques. A mixture of streptomycin and
penicillin, 50 ppm in gelatin, was unsuccessful in killing the bacteria in the gut of
the urchins tested. Therefore a crucial direct test for digestion in the absence of
bacteria cannot be performed. Only circumstantial evidence can therefore be ad-
duced at present, to support the possibility of symbiosis between the sea urchin and
the bacteria, but it is not so overwhelming as to exclude other possibilities.

334

REUBEN LASKER AND ARTHUR C. GIESE

6. Reducing sugar in the body fluid and glycogen stores in tissues

Reducing sugar is usually present in the body fluid of the sea urchin as shown
in Figure 2. The amount of sugar in individual cases varies from 0 to 13 mg. per
cent. The sugar practically disappears after starvation (Table I). However, it
is mobilized quickly in such an animal after a meal of algae or boiled potato
(Table I). In the latter case the sugar rose from 0 to 62 mg. per cent in two days,
the highest value ever obtained with this species of sea urchin over the course of a
year of analyses. The intestinal amylases may be responsible for digestion but
since bacteria are present in the gut their possible share in the digestion of the
starch cannot be ignored.

Glucose injected into a starved sea urchin rapidly disappears from circulation
(Table I). This suggests that the tissues take up glucose and other reducing

10

a

o

L.

a>
0.6

E
24

(•> gonad index

O non- protein nitrogen

• re ducing sugar

.10

.08

o
.06 o

.04 3
a.

(D

X

.02

.00

0 N

1 952

M

A M
I 953

FIGURE 2. The gonad index (volume of gonad divided by the wet weight of the sea urchin)
correlated with the average content of non-protein nitrogen (NPN) and reducing sugar in the
body fluid of the purple sea urchin obtained monthly by analyses of twenty urchins.

sugars and store them in some insoluble form. Preliminary experiments have
demonstrated glycogen in various tissues, but most appears in the intestine (Hilts
and Giese, 1949). To determine the storage of glycogen a male sea urchin was
drained of body fluid and the gut tissues were rinsed free of contents in dis-
tilled water and dried in an oven. The pulverized material, including the test,
was extracted with alkali and the glycogen precipitated in alcohol. The glycogen
present in a sample of the sea urchin was found by the method of Meyer (1943)
to be 472 mg.% on the basis of dry weight, while 123 mg.'/o nitrogen was present.
The amount of glycogen per unit weight of protoplasm was estimated as follows.
Assuming that the protein of protoplasm contains all the nitrogen of the animal
and that protein constitutes about 15% of the protoplasm, 3.84 units of glycogen
are present per unit nitrogen or 0.62 unit of glycogen per unit protein, or almost
0.1 per cent glycogen in the protoplasm (muscle contains 1 per cent). The data

NUTRITION OF A SEA URCHIN

335

show that a considerable store of glycogen exists in the tissues and apparently the
tissues draw upon this store for their respiratory activities. Stott (1931) found
that in Echinus csculentus glycogen accumulated during growth of the gonads but
declined just before the annual spawning in spring.

The avidity with which glucose is taken up following injection after starvation
(Table I) suggests the possibility that the tissues of an urchin are starved for
sugar. If this were true one might expect that addition of sugar would increase
the respiration of excised tissues. Respiration of sea urchin intestine determined
with the standard Warburg technique shows that this is not so. Regardless of
whether glucose or yeast extract or both were added, the respiration of tissues freshly
removed from a well-fed or a starved sea urchin and suspended either in sea water

TABLE II

Non-protein nitrogen (NPN) of the body fluid of the sea urchin
after two weeks starvation and refeeding

Fed

Days fed

Weight of food
ingested in grams

Wet weight
of sea urchin

mg. % NPN

Iridophycus

n

0

48.0

1.20

1

2.73

56.6

4.30

2

2.87

51.2

0.80

4

3.28

45.9

3.75

Boiled egg albumin

0

0

28.9

3.19

0

0

31.0

3.20

1

3.03

35.3

6.67

2

4.72

45.0

8.20

3

2.54

44.3

3.75

or in the body fluid of the sea urchin did not increase significantly over the endo-
genous value. The average Qo2 of digestive tissue was found to be about 0.7 ml./
mg./hr.

7. Non-protein nitrogen in the body fluid

Non-protein nitrogen (NPN) is generally present in the body fluid of a sea
urchin to the extent of about 5 mg. per cent. Data (averages) for monthly samples
of groups of 20 urchins tested for a year are given in Figure 2. The NPN does not
decrease markedly after starvation (Table II), but increases markedly after feeding
with a high protein diet (Table II) rising in a day to twice the value for a control,
and to an even higher concentration the second day. It falls off again on the third
day.

The NPN seems to be regulated to some degree, since unlike the reducing
sugar, it never falls to zero even after prolonged starvation. Nitrogen compounds
stored in tissues must be liberated upon starvation to maintain the supply of
soluble NPN. Storage of the nitrogenous compounds in the gonad is suggested,
since the gonad of a sea urchin starved for a month is almost completely resorbed.

An attempt was made to determine the nature of the NPN by paper chromatog-
raphy. The filtered body fluid from which proteins were removed with TCA

336 REUBEN LASKER AND ARTHUR C. GIESE

was treated with Duolite C3 resin in the acid cycle. The anions were washed
out with distilled water and the amino acids were then separated from the cations
by removal with NH4OH and chromatographed in butyl alcohol (Redfield, 1953;
Slotta and Primosigh, 1951). Only spots for alanine and glutamic acid were
located. If other amino acids are present they occur in amounts too small to be
detectable by even the very sensitive method used.

8. Nutrition and the reproductive cycle

Collections of this species of sea urchins over many years clearly suggest
cyclic gonadal activity. Sperm are available practically all year but eggs are
available only during a limited portion of the year. However, the gonads of both
males and females undergo cyclic variation in volume. The cycle is probably
different for each ecological habitat since differences in maturity were observed
between the development of gonads of urchins at Yankee Point and at Pescadero
Point. To ascertain the nature of the cycles, the gonads were removed monthly
from twenty animals and their volume \vas determined by immersion in a graduate
partially filled with sea water. The volume of gonadal tissues, divided by the
weight of the urchin, designated the gonad index, was used for comparison of
data which are given in Figures 2 and 3.

The growth of the gonads and the development of large numbers of gametes
mean a synthesis involving the conversion of a considerable amount of nutrient
to protein and nucleic acid. The amount of nitrogen in a gonad is large — a ripe
testis containing 2.32% and a ripe ovary, 3.95% nitrogen per unit dry weight;
therefore the nitrogen present in an animal probably increase several-fold during
the breeding season. However the monthly determinations of the reducing sugar
and the NPN in the body fluid given in Figure 2 provide no evidence of excessive
mobilization or withdrawal of NPN or glucose during the breeding season.

DISCUSSION

Van der Heyde (1922) states that whereas the European sea urchin, Echinus
esculentus is herbivorous and feeds largely on Laminaria and other algae, Neapolitan
sea urchins such as Sphaerechinus and Toxopneustes may be carnivorous and may
even capture various crustaceans. Arbacia punctulata appears to be omnivorous
in nature since algae, brittle stars, hydroids and spicules of sponges are found in its
gut (van der Heyde, 1922). Strongylocentrotus droebachiensis has similar feed-
ing habits (Scott, 1901 ; Weese, 1926). The western purple sea urchin, Strongylo-
centrotus purpuratus, studied here was found to be quite omnivorous under labora-
tory conditions, but specimens found in the field invariably were feeding on any
algae available or on the "surf-grass," Phyllospadix. The gut contents of hundreds
of individuals examined appeared to consist entirely of algal pellets, although small
amounts of animal food might have been missed since critical examination for this
purpose was not attempted. Lytechinns anainesus also appears to be an algal
feeder, judging from its depredations on algal beds used for commercial extracts
(A. P. Steiner, personal communication).

Algae contain proteins and other constituents of protoplasm and may possess
small amounts of sugar and starch. In place of the latter may be found sugar
alcohols such as mannitol, dulcitol and sorbitol. However, the greater part of

NUTRITION OF A SEA URCHIN 337

the bulk of dry algae consists of the constituents of the cell walls and supporting
materials. These are made up mainly of polysaccharides other than cellulose
although the latter is sometimes found, e.g., in some of the brown algae. Algin,
found in the brown algae, is a condensation product of mannuronic acid, (CGHsOt!)n.
Hassid (1936) has demonstrated sulfuric acid esters of galactans which make up
to 40% of the dry weight of the red alga, Iridophycus flaccidum. When the
sulfuric acid is split off, hydrolysis gives rise to pure galactose. Red algae also
store floridean starch. Other compounds have also been demonstrated in algae
but the knowledge of many is incomplete and the characterizations quite vague
(see Blinks, 1951 for a review). Such lack of knowledge seriously hampers experi-
ments on digestion of the algae by the sea urchin since it is generally necessary
to use the entire algae rather than individual compounds.

Starch, glycogen, sucrose and maltose were found to be digested by ground-up
intestine and contents of the sea urchin. Echinus escnlentits (Roaf. 1908) and
sucrose and protein by a similar preparation of Strongylocentrotus droebachiensis
(Weese, 1926) and starch but not fats by Sphaerechinus granularis (Cohnheim,
1901; Scott, 1901). Weese (1926) failed to demonstrate lypolytic activity in
Echinus but van der Heyde (1922) found evidence for it in Arbacia, the gut
extracts of which developed acid when incubated with olive oil.

In the studies on S. pnrpuratus performed here, strong protease and amylase
activity were demonstrated in ground, washed intestinal tissue, but no evidence
was obtained for an invertase nor unequivocal evidence for an agar-decomposing
enzyme, even though agar is ingested by hungry urchins. The sugar content of
the body fluid of the starved sea urchin is not increased by a meal of agar. No
increases in reducing sugar were observed nor were the algae decomposed in two
weeks when they were mixed with the gut extract of the sea urchin and kept under
pentachlorophenol. However, in one series of experiments in which the galactan
iridophycin was mixed with the extract of the intestinal wall, an increase of reduc-
ing sugar was obtained, demonstrating the presence of an iridophycase in the gut.
This experiment probably has more validity than those with algae, since in the
latter case side reactions may occur by which the reducing sugar is bound. Other
algal polysaccharides should be tested in a like manner, but none were available
in pure form.

The possibility that bacteria might play a role in digestion has not been
previously seriously considered although Weese (1926) observed bacteria present
in a film about the algal particles in the gut of 5". droebachiensis. The present
study on S. purpuratus indicates that large numbers of bacteria are present, nf
which a considerable number are capable of digesting algae. The population
of bacteria is even greater per unit volume than the figures given because for these
determinations a 0.1 -ml. quantity of gut contents was ground up with sand and
an aliquot was counted or plated. The bacteria actually develop in films about
the algal particles ; therefore they occupy only a fraction of the volume sampled.
The bacteria decomposing agar and other algal cell-wall materials might well be
present in sufficient numbers to digest the algae in the gut of the urchin. Further-
more the food is retained in the intestine for a week to two weeks, a period of time
adequate for even slow digestion.

It was not possible to determine whether the sea urchins could survive loss of
their bacterial flora ; therefore the role of the bacteria could not be defined, Thev

338

REUBEN LASKER AND ARTHUR C. GIESE

may be commensals which use what is left by the digestive enzymes of the sea
urchin or they may be symbionts. The tentative conclusion is drawn that the sea
urchin possesses enzymes which hydrolyze protein and starches of algae and per-
haps slowly decompose some of the more resistant algal polysaccharides. In the
latter action they may be greatly aided by the bacteria present in the gut.

Under normal nutritive conditions reducing sugar was almost always found
in the body fluid of sea urchins taken in the field or feeding in the laboratory, the
amount being greatest just after active feeding and least after starvation. Lang
and MacLeod (1920) also report that little reducing sugar is present in the body
fluid of the echinoderms which they tested, although Myers (1920) reported an
unusually large content in S\ franciscanus. The possibility exists that additional
sugar is present in a combination with proteins or other nutrients as in some other
invertebrates (Morel and Bellion, 1910).

18

J6
o

a.14

Q.

E

O>
O

® gonad index

O Moss Beach

• Stillwater Cove

0 N
1952

M

A M J
1953

.10

.08

o

.06o

o
a.

.04s
a.

CD

.02

.00

FIGURE 3. The gonad index (volume of gonad divided by the wet weight of the sea urchin)
correlated with the water temperature during the year at Moss Beach, a habitat chosen because
it lies on the open ocean, and at Stillwater Cover, a habitat chosen because it is sheltered.

The non-protein nitrogen (NPN) showed about the same variations throughout
the entire year as the sugar but it was more closely regulated than reducing sugar
during starvation. On starvation the NPN in the body fluid remained almost
constant but the gonads shrank to a fraction of their former size. It would seem
that the gonadal nitrogen was mobilized to maintain the NPN in the body fluid
during starvation.

The gonads vary in size during the year and instead of a single growth period
several were found in the monthly examinations made over a period of a year.
The most striking was in December, but peaks occurred in March and June,
presumably followed by spawning although all the spawning periods were not
revealed by the monthly examinations. The gonadal cycle does not vary with
temperature (Fig. 3) but may be related to cyclic changes in the algae providing

NUTRITION OF A SEA URCHIN

more food or more nutritious food at one time than at another. For lack of
pertinent data in the literature it is impossible to test such a possible correlation,
but on completion of a current study of algal populations in this region during the
year, the necessary data may be available.

SUMMARY

1. The intestinal tract of the sea urchin, Strongyloccntrotns pnrpiiratus, consists
of two loops, the first clockwise, the second counter-clockwise (as seen looking in at
the mouth). Both loops are well supplied with glands.

2. Experiments present evidence for an amylase and a proteinase in the sea
urchin intestine but none for enzymes capable of digesting entire algae or agar.
However, the extract of ground gut was capable of digesting iridophycin, a galactan
from the red alga, Iridophycns.

3. Agar-digesting bacteria are present in the second loop of the intestine in
large numbers from 106 to 107 per ml. of gut contents. Total bacteria as determined
by a count with a Petroff-Hauser counting chamber reach the value of 2 X 1011
per ml. They are largely confined to the pellicle which surrounds each particle
of alga.

4. The bacteria inoculated from the gut of the sea urchin are capable of com-
pletely digesting the alga, Iridophycns flacciditm, in the course of a week. Many
of the intestinal bacteria are capable of digesting agar. Pure cultures of agar-
decomposing bacteria were isolated from the intestine.

5. The isolated bacteria grown on algae or agar do not liberate reducing sugars
into the culture medium.

6. While it seems likely that the sea urchin obtains some nutrient from the
activities of its intestinal flora, proof for this is lacking.

7. Some glycogen is stored in the tissue of the sea urchin and the body fluid
normally contains a small amount of reducing sugar. No striking changes occur
in the latter during the breeding season, but the sugar falls to zero or nearly zero
on starvation. Glucose injected into the body fluid is removed by the tissues.

8. The body fluid of the sea urchin always contains a small amount of non-
protein nitrogen (NPN), even after two weeks of starvation, and no striking change
occurs during the breeding season.

9. The stimulus that sets off the increased effectiveness of the economy of the
sea urchin resulting in the growth of the gonads and accumulation of nitrogenous
compounds in them is unknown. After gonads reach their maximal size spawning
occurs.

10. Several cycles of gonadal growth and spawning are suggested but the data
were not collected at sufficiently frequent intervals to make this certain.

LITERATURE CITED

BLINKS, L. R., 1951. Physiology and biochemistry of the algae. Chapter 14 in "Manual of
Phycology" ed. by G. M. Smith. Chronica Botanica, Waltham, Mass.

COHNHEIM, O., 1901. Versuche iiber Resorption, Verdauung und Stoffwechsel von Echinoder-
men. Zcltschr. f. Physiol. diem., 33: 9-54.

FRIEDEMANN, T. E., AND T. B. GRAESER, 1933. The determination of lactic acid. /. Biol. Chem.,
100 : 291-308.

340 REUBEN LASKER AND ARTHUR C. GIESE

HASSID, \Y. /., 1933. The isolation of a sodium sulfuric acid ester from Iridaca laminarioides

(Rhodophyceae). J. Amcr. Chcm. Soc., 55: 4163-4167.
HASSID, W. Z., 1936. Carbohydrates in Iridaca lawiinariodes (Rhodophyceae). Plant Physiol.,

11: 461-463.
VAN DER HEYDE, H. C., 1922. On the physiology of digestion, respiration, and excretion in

echinoderms. Academic Thesis de Boer Jr., den Helder (Pays Bas). Ill pp. 6 fig.
HILTS, S., AND A. C. GIESE, 1949. Sugar in the body fluid of a sea urchin. Anat. Rcc., 105 :

140.
LANG, R. S., AND J. J. R. MACLEOD, 1920. Observations on the reducing substance in the

circulating fluids of certain invertebrates and fishes. Quart. J. Exp. Physiol., 12 :

331-337.
LYNCH, J. E., 1929a. Studies on the ciliates from the intestine of Strongylocentrotus. I.

Entorhipidium gen. nov. Univ. Calif. Publ. Zool., 33 : 27-56.
LYNCH, J. E., 1929b. Studies on the ciliates from the intestine of Strongylocentrotus. II.

Lechriopyla mystax, Gen. nov., sp. nov. Univ. Calif. Publ. Zool., 33: 307-350.
MEYER, K. H., 1943. The chemistry of glycogen. Advances in Ensyuntl.. 3: 109-135.
MOREL, A., AND M. BELLION, 1910. Contribution a 1'etude du sucre du sang chez les invertebres.

Sucre libre et sucre combine du sang de 1'escargot. C. R. Soc. Biol., 69 : 27-28.
MYERS, R. G., 1920. A critical study of the blood of several invertebrates. /. Biol. Chcm., 41 :

119-135.
REDFIELD, R. R., 1953. Two dimensional chromatographic systems with high resolving power

for amino acids. Biochem. Biophys. Acta, 10: 344-345.
REID, R. L., AND M. LEDERER, 1952. Separation and estimation of saturated C0-C7 fatty acids

by paper partition chromatography. Biochem. J., 50: 60-67.

ROAF, H. E., 1908. The hydrolytic enzymes of invertebrates. Biochem. J., 3 : 462-472.
SCOTT, F. J., 1901. Food of the sea urchin (Strongylocentrotus droe bachiensis) . Contrib.

Canad. Biol., 1901 : 49-54.
SLOTTA, K. H., AND J. PRIMOSIGH, 1951. The amino-acid composition of crotoxin. Nature,

168: 696-697.

SOMOGYI, M., 1945. Determination of blood sugar. /. Biol. Chem., 160: 69-73.
SOMOGYI, M., 1952. Notes on sugar determination. /. Biol. Chcm., 195 : 19-23.
STOTT, F. C., 1931. The spawning of Echinus csculentus and some changes in gonad com-
position. /. Exp. Biol, 8: 133-150.
WEESE, A. O., 1926. The food and digestive processes of Strongylocentrotus droebachiensis.

Publ. Puget Sound Biol. Stat., 5 : 165-179.

OXIDATIVE ENZYMES IN THE THORACIC MUSCLES OF
THE WOODROACH LEUCOPHAEA MADERAE 1

W. H. McSHAN, SOL KRAMER AND VERA SCHLEGEL -
t of Zoolof/y, University of Wisconsin, Madison, Wisconsin

Our present knowledge of cellular metabolism is based largely on results ob-
tained from studies using mammalian tissues. The mechanisms of many of the
enzyme-catalyzed reactions of respiration and glycolysis have been elucidated during
the past twenty years. During recent years the methods developed for the study of
mammalian tissues have been used to study some of the oxidative enzymes in insect
tissues, with the result that considerable information has accumulated concerning
certain of the oxidative enzymes in various insect tissues.

In this connection Barren and Tahmisian (1948) found that the oxygen con-
sumption of muscle from male cockroaches, Periplaneta americana, is double that
from female roaches. Sacktor and Bodenstein (1952) reported on the cytochrome
oxidase activity of various tissues of the American cockroach, and Harvey and
Beck (1953) studied in considerable detail the succinoxidase and cytochrome
oxidase systems in the leg muscle of this form. They found that the succinoxidase
activity of muscle from the male cockroach is three times that of muscle from the
female. Spirtes (1951) demonstrated the presence of Krebs cycle enzymes such as
aconitase, isocitric, malic and succinic dehydrogenases, fumarase and condensing
enzyme, and also cytochrome oxidase and lactic dehydrogenase in the tissues of
Drosophila melanogaster ; and Bodenstein and Sacktor (1952) studied the cyto-
chrome oxidase during metamorphosis of Drosophila virilis; Sacktor (1951a, 1951b,
1952) reported on the cytochrome oxidase activity of normal and DDT resistant
house flies, Musca domestica; Sanborn and Williams (1950) studied the cytochrome
system in the tissues of the Cecropia silkworm; Watanabe and Williams (1951)
showed that succinic, a-glycerophosphate, malic and pyruvic dehydrogenases and
cytochrome oxidase are present in the sarcosomes of insect muscles ; and Collias,
McShan and Lilly (1952) reported results of studies on the succinoxidase and
cytochrome oxidase systems in the tissues of the large milkweed bug, Oncopeltus
jasdatus. Bodine, Lu and West (1952) found marked differences in the suc-
cinoxidase activity in mitotically active and blocked cells of the developing embryo
of the grasshopper, Mclanoplus differential's.

Investigations of this kind serve to clarify further our knowledge of the relation-
ship of the cellular metabolic reactions of insect tissues to those already known for
mammalian tissues. Furthermore, information obtained for insects and other lower
forms is of value from the comparative standpoint and may provide the basis for
an insight into the mechanism by which energy is provided for certain specialized
behavior patterns in insects such as the cockroach.

1 This investigation was supported in part by a grant from funds supplied by the Wisconsin
Alumni Research Foundation.

- Present address: Department of Biochemistry, University of Minnesota, St. Paul.

341

342

McSHAN, KRAMER AND SCHLEGEL

T2 Jcx Ttn X

cx

1

-Ttr

=-— TrT

Ba— 4

Tri

Cxs--

FIGURE 1. Mesal view of the right half of the thorax of the woodroach, Leucophaca
madcrae, showing some of the pigmented mesothoracic and metathoracic muscles (BaM, basalar
muscle; Tex, tergo-coxal muscle; Ttn, tergo-trochantinal muscle; Ttr, tergo-trochanteral
apodeme muscle) used in the preparation of "leg muscle" homogenate. a-b, line of incision along
coxae made to expose the entire muscles prior to removal ; Cx, coxa ; Epcx, episternalcoxal

ENZYMES OF WOODROACH MUSCLE 343

The present paper reports results of a study of the succinoxidase, cytochrome
oxidase and fatty acid oxidase of thoracic pigmented muscle from the woodroach,
Leucophaea maderae. The mechanism of action and optimum conditions for these
systems have been studied extensively in some mammalian tissues by Keilin and
Hartree (1949), Slater (1949a, 1949b), Chance (1952), Lehninger (1946), and
Lehninger and Kennedy (1948).

MATERIALS AND METHODS

The woodroaches used in this study were isolated soon after metamorphosis and
kept in dated containers so that muscle tissue could be obtained from roaches of
known age. In certain of the earlier experiments, however, adult roaches of un-
known age were used. Males and females were kept separately. All roaches
were fed on the same constant dog pellet diet and ample food and water were always
available to them.

The muscle tissue used was dissected from the nieso- and metathoracic seg-
ments immediately after the roaches were killed by severing the head and abdomen.
A mid-ventral incision through the thorax divided it into two halves. Remnants of
the gut, large tracheal tubes and fat body were quickly cleaned away, and the large
bundles of thoracic muscles were exposed as shown in Figure 1. Incisions along
the meso- and metathoracic coxae along dotted lines a-b made it possible to sepa-
rate these muscles in bulk with a few ventral and dorsal incisions. The tissue was
weighed and placed in ground glass homogenizing tubes contained in an ice bath,
and homogenized within 8 minutes after the roaches were killed. Sufficient wrater
was added to give a 2.5 per cent homogenate which was used for the succinoxidase
determinations. It was necessary to prepare a 0.5 per cent homogenate for the
cytochrome oxidase and a 10 per cent homogenate for the determinations of fatty
acid oxidase.

These muscles are sometimes referred to as the "leg muscles," and most of
these muscles are in fact concerned with leg function. Roaches, such as the cock-
roach, Periplaneta americana as well as the woodroach, Leucophaea mad era e and
others, although comparatively weak flyers, can and do fly. Woodroaches in par-
ticular were observed on rare occasions to fly distances of 10-12 feet in slow,
labored flight in the insectary. Further, Roth and Willis (1952) have shown that
the wings of male Periplaneta americana and male Blatta orientalis are vibrated
actively prior to copulation.

It is clear, then, that some muscles in roaches must function in flight and wing
vibration. Carbonell (1947) in a detailed study of the thoracic musculature of the
cockroach Periplaneta americana noted that these muscles bore little resemblance to

muscle ; Fe, femur ; Prn, pronotum ; T2, mesothoracic tergum ; Ts, metathoracic tergum ; Tr,
trochanter.

FIGURE 2. Mesal view of the mesothoracic flight muscles (BaM, basalar muscle; SaM,
subalar muscle) which lie among the leg muscles and also included in the homogenate prepara-
tions. Ba, basalare ; Epm, epimeron ; Me, meron ; Sa, subalare ; TrT, trochanteral tendon.
Other abbreviations as above.

FIGURE 3. Similar view of metathoracic flight muscles included in the homogenate prepara-
tions. Eps, episternum ; PIA, pleural apodeme ; Tn, trochatin. Other abbreviations as above.

344 McSHAX, KRAMER AND SCHLEGEL

those of other insects, and that in the musculature of the wings the cockroach
thorax differs widely from the normal scheme of wing-bearing segments as given
by Snodgrass (1935). The size of the basalar muscles (pronator-extensor of the
wings) and the subalar muscles (depressor-extensor of the wings), which lie in the
midst of the large leg muscles, led Carbonell to conclude that they must play an
important role in flight.

Dissection of the woodroach revealed that prominent basalar muscles (BaM)
and subalar muscles (SaM) are present in both the mesothoracic (Fig. 2) and
metathoracic segments (Fig. 3 ). In fact, the basalar muscles3 in both segments are
the largest and longest of all the individual muscles present. These large flight
muscles, together with the large leg muscles, are pigmented pink, in contrast to the
smaller ventral longitudinal and certain smaller oblique muscles which are a trans-
lucent white color — and it was these pink pigmented muscles as a group which
were used for the preparation of homogenates. Further, each homogenate repre-
sents not a mixture of muscles from several insects, but a preparation from the
muscles of one roach of known age.

The enzyme determinations were made by use of the conventional Warburg
apparatus. The homogenates were prepared by use of sharp-pointed, ground-glass
homogenizers. The homogenates used for the study of succinoxidase and cyto-
chrome oxidase were made with water and those for fatty acid oxidase with 0.154
M KC1. The proper amount of homogenate was placed in the flasks with the
required cof actors for each of the enzyme systems studied. The flasks were placed
in the bath at 38° C., and ten minutes were allowed for equilibration in the case of
succinoxidase and cytochrome oxidase, and 6 minutes for fatty acid oxidase. Read-
ings of oxygen consumption were taken at 10-minute intervals for at least 40
minutes. The average value for the number of 10-minute periods during which
the oxygen consumption was constant, which was usually four periods, was used
as a basis for calculating the Qo^ values.

The methods used for the determination of succinoxidase and cytochrome oxi-
dase were those reported by Schneider and Potter (1943). The optimum concen-
trations of required factors, and other conditions for maximum succinoxidase
activity of thoracic muscle of the woodroach were determined. The concentrations
of factors used for the cytochrome oxidase determinations were the same as those
that have been reported for mammalian tissues. The fatty acid oxidase determina-
tions were done by the method reported by Lehninger and Kennedy (1948). Final
flask concentrations for the different enzyme systems are given in the footnotes
to the tables.

The inhibitors were prepared in stock solutions which were in most cases 0.001
M. The solution of diethylstilbestrol was prepared by the procedure reported by
McShan and Meyer (1946). The dry weight determinations of flight muscle were
done by weighing the fresh tissue, placing it in a weighed tube and drying at 75° C.
for 24 hours, after which the dry tissue was weighed and the weight used for
calculating the percentage dry weight in terms of fresh weight.

Cytochrome c used for the determinations of succinoxidase and cytochrome oxi-
dase was prepared by a modification of the method of Keilin and Hartree (1937),

3 The basalar muscle actually arises from a tendon at the margin of the episternum adjacent
to the basalare in each segment, but Crampton (1927) regards this margin of the episternum as
an anterior portion of the basalare in the cockroach, Pcriplaneta amcricana.

KXZYMES OF WOODROACH MUSCI.K

345

or was obtained from the Sigma Chemical Company. Analytical reagent grade
chemicals were used.

RESULTS AND DISCUSSION

The succinoxidase activity of homogenates of thoracic muscle from female
roaches of different ages was determined with different concentrations of succinate,
phosphate buffer, calcium chloride, aluminum chloride and cytochrome c. The

TABLE I

Determination of the optimum concentrations of constituents required for maximum activity
of succinoxidase in homogenates of woodroach thoracic muscle

Constituents

Concentrations of variable constituents and Qn> values**

Absent

Present

Final
M in flask

Aids
CaCh

Phosphate pH 7.3

Sod. succ.
Cyto. c

variable

0.1
2X10-5

M*
Qo2

0.0

0.0083

0.017

0.033

0.050

0.066

0.100

0.133

37.0

40.0

69.0

83.0

132.0

115.0

115.0

98.0

AlCh

CaCb

Sod. succ.
Cyto. c
Phosphate

variable

0.1
2X10-5
0.05

M

0.0

0.0004

0.0008

0.0012

0.0016

0.0020

Qo2

107.0

135.0

158.0

193.0

208.0

199.0

CaCh

AlCb

Sod. succ.
Cyto. c
Phosphate

variable

0.1
2X10-5
0.05

M

0.0

0.0004

0.0008

0.0012

0.0016

0.0020

Qo2

M

126.0

175.0

177.0

166.0

186.0

177.0

None

Sod. succ.

Cyto. c
Phosphate
CaCh+AlCls each

variable

2X10-5
0.05
1.6X10 -»

0.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

Qo2

99.0

125.0

142.0

160.0

174.0

189.0

199.0

199.0

None

Cyto. c. ( X 10-5 M)

Sod. succ.
Phosphate
CaCU+AlCUeach

variable

0.1
0.05
4X10-*

M

0.5

1.0

1.5

2.0

2.5

3.0

Qoj

94.0

113.0

128.0

131.0

115.0

103.0

None

pH

Sod. succ.
Phosphate
CaCl2+AlCl2 each
Cyto. c

variable

0.1
0.05
4X10-'
2X10-5

pH

6.38

6.76

7.17

7.3

7.59

7.91

Qo:

94.0

112.0

128.0

131.0

115.0

103.0

* Final molarity in flask.

** Qo2 values are based on a dry weight content of 18.2 per cent and are averages of 2 to 5
runs using 0.1 ml. of 2.5 per cent homogenate except 0.2 ml. was used when cytochrome c and the
pH were varied.

results italicized in Table I indicate the concentrations which gave maximum
activity are 0.2 M succinate, 0.05 M phosphate buffer of pH 7.3, 1.6 X lO'3 M of
each calcium and aluminum chlorides, and 2 X I0":i M cytochrome c. Results of
runs made at different pH values show that maximum activity was obtained at
pH 7.3.

The data of Table II (Experiments 1 to 3) show that under the above conditions
the oxygen consumption was directly proportional to the amount of tissue reacting

346

McSHAN, KRAMER AND SCHLEGEL

for 0.05, 0.10 and 0.15 ml. of 2.5 per cent homogenate but not for 0.20 and 0.25 ml.
The results from Experiments 1, 2 and 3 are shown graphically in Figure 4. When
0.175 M succinate was used (Experiment 4, Table II) there was not quite a direct
proportionality between the oxygen consumption and the amount of tissue reacting.
These results indicate that a much higher concentration of succinate is required
for optimum activity of the succinoxidase system of woodroach muscle than for
this system of other tissues such as rat liver which requires only 0.05 M (Schneider
and Potter, 1943). Harvey and Beck (1953) found 0.11 M succinate optimum for
American cockroach muscle.

Essentially no oxygen was consumed when the succinoxidase system was run
without substrate, and without tissue. When the cytochrome r was left out of the

TABLE II

Relation of oxygen consumption to amount of woodroach thoracic muscle used in the

succinoxidase system

Amount of
tissue*
ml. 2.5%
homogenate

Experiment No.

1(3)**

2(3)**

3(1)**

4(1)**

Oxygen consumption

Cmm. per
10 min.

Qo2

Cmm. per
10 min.

Qo2

Cmm. per
10 min.

Qo2

Cmm. per
10 min.

Qo2

0.05

7.6

201.2

6.1

160.8

8.0

210.9

0.10

15.0

197.3

12.1

160.0

19.5

257.2

15.5

204.0

0.15

22.4

196.9

18.4

161.3

28.6

252.0

21.9

192.0

0.20

23.4

154.1

34.6

228.0

0.25

39.0

205.7

* Flask concentrations of the constituents used in the system were sodium succinate 0.2 M
except it was 0.175 M for experiment 4, phosphate buffer of pH 7.3 0.05 M, aluminum and calcium
chlorides each 1.6 X 10~3 M, cytochrome c 2 X 10~5 M. The muscle used in these experiments
was taken from adult females of unknown age.

** Number of runs with two flasks per each amount of tissue. The Qo2 values are based on
the average oxygen consumption for the first four 10-minute periods and a dry weight content of
18.2 per cent.

system the average oxygen consumption per 10 minutes was 2.9, 5.3 and 7.5 mm.3,
respectively, for 0.05, 0.10 and 0.15 ml. of 2.5 per cent homogenate. This was
presumably due to the presence of cytochrome c in the flight muscle since this
cytochrome has been shown to be present in cockroach muscle (Barren and
Tahmisian, 1948; Harvey and Beck, 1953). Homogenates of woodroach muscle
were therefore treated with sodium hydrosulfite to reduce the cytochromes and
examined by use of a Hartridge Reversion Spectroscope. Absorption bands at the
proper wave-lengths for cytochromes a, b and c were found, indicating the presence
of these cytochromes in the woodroach muscle. On the basis of this evidence it
appears that the activity of the system without added cytochrome c wras caused by
the presence of this component in the muscle homogenate.

The Qo2 values of the different amounts of muscle homogenate used in Ex-

ENZYMES OF WOODROACH MUSCLE

347

periments 1, 2 and 3 (Table II) were essentially constant and the average values
were, respectively, 198.5, 161.0 and 254.6. The over-all average Q02 was 205.
Adult female woodroaches of various ages were used for these experiments. These
QoL> values are based on a dry weight of 18.2 per cent. The dry weights given in
Table III, which range from 21.5 to 29.8 per cent, were done at different times
than were those on which the above 18.2 per cent is based. The reason for this
difference in the dry weights of these two series of experiments is not apparent.

40r

30

oc.

LJ

Q- 20

LlJ

g

ID

o

10

0.05 0.10 0.15
MUSCLE HOMOGENATE ML. 2.5 PER CENT

0.20

FIGURE 4. Results showing relation of different amounts of enzyme or tissue to oxygen
consumption. Tissue concentrations which resulted in directly proportional oxygen uptakes are
shown.

The roaches were kept under essentially constant conditions as to water, food and
temperature, but factors other than these and the possible relation of dry weight to
age may be involved.

The results given in Table III show the relation of age and sex to the dry
weight and succinoxidase activity of woodroach muscle. There seems to be some
tendency for the dry weight of muscle of both sexes to increase with increase in age
after 5 days of adulthood. These results also show that there was an increase in

348

McSHAN, KRAMER AND SCHLEGEL

the succinoxidase activity of the muscle with increase in age of the roaches. Al-
though most of these experiments were done during the spring of 1952, they were
repeated for certain ages in the spring of 1953 with similar results, Table III).
Further work, however, is necessary to clarify the variation in dry weight content
of roach muscle.

The inhibition of the succinoxidase system of woodroach thoracic muscle was
studied by use of diethylstilbestrol, cyanide, azide and malonate. Forty to 60 per
cent inhibition was obtained with 0.27 X 10~4 M diethylstilbestrol, 0.5 X 10~4 M
potassium cyanide, 0.5 X 10~3 M sodium azide and 0.5 X 10"- M sodium malonate
(Table IV). The malonate is known to inhibit the succinic dehydrogenase of the
succinoxidase system. The diethylstilbestrol (McShan and Meyer, 1946), cyanide

TABLE III

Dry weight content and succinoxidase activity of thoracic muscle
from woodroaches of different ages

Sex

Age

days

1952

1953

Dry wt.

%

Qo2*

Dry wt.

%

Qo2*

Female

0.5

25.0

120

23.4

106

5

23.9

143

10

24.2

148

24.5

110

20

26.4

145

30

29.8

140

40

27.1

188

82

27.3

155

Male

0.5

24.3

129

5

21.5

149

10

23.7

154

23.9

128

20

27.3

143

40

25.3

178

77

26.2

156

* Values based on 1 to 4 runs with 2 to 3 flasks per run, and on the oxygen uptake during the
first 3 or 4 ten-minute periods.

and azide inhibit the cytochrome oxidase of this system and this prevents the oxida-
tion of the cytochrome c when it is reduced by the action of the dehydrogenase.
These inhibitors appear to affect the succinoxidase system of woodroach muscle in
the same way as they affect this system in mammalian tissues.

Experiments were done to determine directly the succinic dehydrogenase activity
of woodroach muscle by using brilliant cresyl blue (BCB) in the system as the
mediator of hydrogen transport in place of the cytochrome system. When BCB
was used in the system a Qo2 of 76.5 was obtained as compared with 194 and 173,
respectively, when cytochrome c, and BCB plus cytochrome c were present in the
system. Similar results were obtained with rat liver which was run as a control.
When cyanide was used in the system with BCB there was an increase of 83 per cent
in the Oo2 of woodroach muscle (Oo.j of 76.5 for BCB alone) but under the same
conditions cyanide did not cause an increase in the activity of rat liver. This in-

ENZYMES OF WOODROACH MUSCLE 349

creased oxygen consumption when cyanide is added to the BCB system has been
reported previously for leg muscle of the American cockroach (Harvey and Beck,
1953).

A Qo2 of 1770 was obtained for the cytochrome oxidase of muscle from female
roaches 30 days of age when 0.05 ml. and 0.1 ml. of 0.5 per cent homogenate were
used per flask. Each flask also contained final concentrations of 0.033 M phosphate
buffer of pH 7.3, 0.0114 M ascorbic acid, 4 X 10~5 M aluminum chloride, and 8.7 X
10"5 M cytochrome c which are essentially the amounts of these factors used for rat
liver cytochrome oxidase by Schneider and Potter (1943). The Qo2 of 1770 ob-
tained for woodroach muscle is close to that of 1520 reported by Harvey and Beck
(1953) for cockroach muscle but is much greater than the Q02 of 377 and 387
found, respectively, for cytochrome oxidase of rat liver and corpora lutea from

TABLE IV

Effect of inhibitors on the succinoxidase system of thoracic muscle of the woodroach

Inhibitor

Concentration in flask Inhibition

M %

Diethylstilbestrol 0.27 (10-«) 59*

0.50 92

1.0 96

Potassium cyanide 0.5 (10"3) 40

1.0 93

10.0 97

Sodium azide 0.27 (10-') 44

0.50 56

1.0 61

1.3 79

2.7 80

5.8 78

Sodium malonate 0.5 (1Q-2) 41

1.0 60

* Values are based on 4 to 7 runs with two flasks per run.

pregnant rats by McShan, Meyer and Erway (1947) and McShan, Erway and
Meyer (1948).

The fatty acid oxidase activity is low as compared to the succinoxidase activity
of woodroach muscle, and there does not appear to be a significant change in activity
with increase in age of the roach (Table V).

The results of this study show that succinoxidase and cytochrome oxidase sys-
tems are present in the thoracic pigmented muscle of the woodroach, Leucophaca
maderae. In this muscle, however, the succinoxidase is more than twice as active
and the cytochrome oxidase more than four times as active as in rat liver. On the
other hand the fatty acid oxidase of rat liver is about ten times that of the woodroach
muscle. Perhaps this is to be expected since the liver is known to be the locus for
fatty acid metabolism.

Optimum conditions were determined for eliciting the maximum succinoxidase
activity of woodroach muscle and it was found that this muscle requires four times

350 McSHAN, KRAMER AND SCHLEGEL

the concentration of sviccinate as does rat liver. In this connection Harvey and
Beck (1953) found that the succinoxidase of leg muscle from the American cock-
roach requires 0.11 M succinate which is more than double that required by rat
liver (Schneider and Potter, 1943). These results suggest that tissues high in suc-
cinoxidase, such as roach muscle, require higher concentrations of succinate for
maximum activity than do tissues which contain a lower concentration of this
system.

The results obtained with the BCB system, inhibitors and the required cofactors
indicate that the mechanism of action of woodroach muscle succinoxidase is similar
to that of mammalian tissues.

The increase in the succinoxidase activity of woodroach muscle with increase
in age and the possible trend toward an increase in dry weight with increase in age
may have physiological significance which is not apparent at present. In this
connection Sacktor (1951b) showed that the cytochrome oxidase activity of normal
and DDT-resistant house flies changes during pupal development, and Watanabe
and Williams (1951) have reported differences in the cytochrome oxidase activity

TABLE V

Fatty acid oxidase activity of thoracic muscle from female woodroaches
Age in days Qr>2**

10* 3.1

20 5.8

40 3.0

60 4.5

Adult 3.7

Ave. for all ages 4.0

* When water was used for homogenizing muscle from a roach 10 days old a Qo2 of 2.3 was
obtained as compared to 3.1 for 0.154 M KC1.

** The amount of muscle tissue used per flask was 0.25 ml. of 10 per cent homogenate made
with 0.154 M KC1. The final flask concentrations of reagents were 0.033 M KH2PO4-K2HPO4
of pH 7.4, 0.002 M potassium octonoate, 0.07 M KC1, 0.013 M MgSO4> and 6.6 X 10~4 M KATP.
The Qo2 values are based on a dry weight content of 18.2 per cent.

of sarcosomes of Phormia isolated from insects of different ages. Further, Harvey
and Beck (1953) found that succinoxidase is three times as active in the thoracic
muscle of the male as in the female American cockroach, Periplaneta aincricana.
These results for Periplaneta have been confirmed in our laboratory. It is there-
fore of interest that the succinoxidase activity in the thoracic muscle of the wood-
roach, Leucophaea inaderae, is essentially the same in both sexes.

SUMMARY

1. The thoracic muscle of the woodroach, Leucophaea maderae, was shown to
contain high concentrations of succinoxidase and cytochrome oxidase and a low
concentration of fatty acid oxidase as compared to rat liver.

2. The conditions required for optimum activity of the succinoxidase system
were determined and it was found that this system requires four times the concen-
tration of succinate as does succinoxidase of rat liver.

3. Succinoxidase activity of thoracic pigmented muscle in the woodroach is
essentially the same in both sexes, whereas in the American cockroach, Periplaneta

ENZYMES OF WOODROACH MUSCLE 351

americana, the activity is three times as great in the muscle of the male as in that of
the female. These latter results with P. americana (Harvey and Beck, 1953) have
been confirmed in our laboratory.

4. Results were obtained which indicate that the succinoxidase activity of wood-
roach thoracic muscle increases with increase in the age of the roach.

5. The results of studies with cofactors, inhibitors and the brilliant cresyl blue
system indicate that the mechanism of action of the succinoxidase of woodroach
muscle is similar to that of mammalian tissues.

LITERATURE CITED

BARKON, E. S. G., AND T. N. TAHMISIAN, 1948. The metabolism of cockroach muscle, Peri-
planeta americana. J. Cell. Comp. Physiol., 32 : 57-76.

BODENSTEIN, D., AND B. SACKTOR, 1952. Cytochrome c oxidase activity during the metamor-
phosis of Drosophila ririlis. Science, 116: 299-300.

BODINE, J. H., KiAO-HtiNG Lu AND W. L. WEST, 1952. Succinic dehydrogenase in mitotically
active and blocked embryonic cells. Physiol. Zool., 25 : 109-123.

CARBONELL, C. S., 1947. The thoracic muscles of the cockroach Pcriplancta americana (L.).
Smith. Misc. Coll., 107 : 1-23.

CHANCE, B., 1952. The kinetics and inhibition of cytochrome components of the succinic oxidase
system. /. Biol. Chcm., 197 : 567-576.

COLLIAS, E. C., W. H. McSnAN AND J. H. LILLY, 1952. Oxidative enzyme systems of the large
milkweed bug, Oncopcltus fasciatus (D), and the effect of sabadilla on them. /. Cell.
Comp. Physiol., 40 : 507-528.

CRAMPTON, G. C., 1927. The thoracic sclerites and wing bases of the roach Periplaneta ameri-
cana and the basal structure of the wings of insects. Psyche, 34 : 59-72.

HARVEY, G. T., AND S. D. BECK, 1953. Muscle succinoxidase in the American cockroach. /.
Biol. Chem., 201 : 965-973.

KEILIN, D., AND E. F. HARTREE, 1937. Preparation of pure cytochrome c from heart muscle
and some of its properties. Proc. Roy. Soc. (London), 122B : 290-308.

KEILIN, D., AND E. F. HARTREE, 1949. Activity of the succinic dehydrogenase-cytochrome sys-
tem in different tissue preparations. Biochem. J., 44 : 205-218.

LEHNINGER, A. L., 1946. Quantitative study of the products of fatty acid oxidation in liver
suspensions. /. Biol. Chem., 164: 291-306.

LEHNINGER, A. L., AND E. P. KENNEDY, 1948. The requirements of the fatty acid oxidase
complex of rat liver. /. Biol. Chem., 173 : 753-771.

McSnAN, W. H., AND R. K. MEYER, 1946. The effect of estrogens on the succinoxidase system
of liver and pituitary tissues. Arch. Biochem., 9: 165-173.

McSnAN, W. H., R. K. MEYER AND W. F. ERWAY, 1947. Effects of estrogens on the suc-
cinoxidase system of rat tissues. Arch. Biochem., 15: 99-110.

McSHAN, W. H., W. F. ERWAY AND R. K. MEYER, 1948. Malic dehydrogenase and cytochrome
oxidase of lutein and other ovarian tissues during pregnancy and lactation. Arch.
Biochem., 16 : 379-387.

ROTH, L. M., AND E. R. WILLIS, 1952. A study of cockroach behavior. Amcr. Mid. Nat.. 47:
66-129.

SACKTOR, B., 195 la. A comparison of the cytochrome oxidase activity of two strains of house
flies. /. Econ. Entomol., 43: 832-838.

SACKTOR, B., 1951b. Some aspects of respiratory metabolism during metamorphosis of normal
and DDT-resistant house flies, Musca domestica L. Biol. Bull., 100: 229-243.

SACKTOR, B., 1952. The cytochrome c oxidase of the house fly, Musca domestica L. /. Gen.
Physiol., 35 : 397^07.

SACKTOR, B., AND D. BODENSTEIN, 1952. Cytochrome c oxidase activity of various tissues of
the American cockroach, Periplaneta americana. J. Cell. Comp. Physiol., 40: 157-161.

SANBORN, R. C., AND C. M. WILLIAMS, 1950. The cytochrome system in the Cecropia silkworm,
with special reference to the properties of a new component. /. Gen. Physiol., 33 : 579-
588.

352

McSHAN, KRAMER AND SCHLEGEL

SCHNEIDER, W. C., AND V. R. POTTER, 1943. The assay of animal tissues for respiratory enzymes.
/. Biol. Chew., 149: 217-227.

SLATER, E. C., 1949a. The measurement of the cytochrome oxidase activity of enzyme prepara-
tions. B'wchcm. L, 44: 305-318.

SLATER, E. C., 19491). A respiratory catalyst required for the reduction of cytochrome c by cyto-
chrome b. B'wchcm. J ., 45 : 14-30.

SNODGRASS, R. E., 1935. Principles of insect morphology. McGraw-Hill Book Co., New York
and London.

SPIRTES, M. A., 1951. Demonstration of the presence of Krebs cycle enzymes in Drosophila
melanogastcr. Fed. Proc., 10: 251.

WATANABE, M. I., AND C. M. WILLIAMS, 1951. Mitochondria in the flight muscles of insects.
/. Gen. Physiol., 34 : 675-689.

TIDAL RHYTHMICITY OF RATE OF WATER PROPULSION IN

MYTILUS, AND ITS MODIFIABILITY

BY TRANSPLANTATION ^ 2

KANDULA PAMPAPATHI RAO3

Department of Zoology, University of California, Los Angeles, California

While studying the rate of water propulsion in Mytilus californianus (Rao, 1953)
it was observed that the behavior of the mussels was not the same at different periods
of the day. A detailed study revealed that these differences were of the nature of
a tidal rhythm, with periods of greater activity, corresponding to the times of high
tide, alternating with those of lesser activity, corresponding to the times of low tide
in the area from which the animals were collected.

Since the discovery of a persistent tidal rhythm in Convoluta roscoffensis (Bohn,
1903; Gamble and Keeble, 1903), similar rhythms have been described for a num-
ber of marine organisms from nearly all groups, and these have been reviewed by
Calhoun (1944) and Brown, Fingerman, Sandeen and Webb (1953). Several
molluscs have been described as exhibiting tidal rhythmicity in their activity.
Littorina rudis, which is covered by water only during the semilunar high, high
tides, becomes active at 15-day intervals when kept in the laboratory (Bohn, 1904).
Brown, Bennett and Graves (1953) report a long-term tidal rhythm in Venus.
Compel (1937, 1938) reported the occurrence of a persisting tidal rhythm of oxygen
consumption in Patella, Mytilus, Pecten and Cythcrea while in Haliotis tnberculata
it was not so marked. Brown, Bennett and Webb (1953) found the same in the
crab Uca.

In the following studies an attempt was made to learn something of the nature
of this rhythm in Mytilus, using as an index of activity the rate of water propulsion.

It is a pleasure to acknowledge my indebtedness to Professor Theodore H.
Bullock for helping me in the procurement of the material ; offering me all the
laboratory facilities ; for his enthusiastic encouragement during the course of this
investigation and for critically reading through this paper. To the Chairman and
Secretary of the Department of Zoology, and the other members of the staff, T
am most grateful for several courtesies extended to me during my stay in the
Department. My especial thanks are due to Professor G. E. MacGinitie, Director
of the Kerckhoff Marine Laboratory, Corona del Mar, California, and to his staff
for allowing me to make use of their laboratory pier for the experiment in trans-
plantation of mussels. Finally I should like to place on record the promptness with

1 Aided by a grant to Dr. Theodore H. Bullock, from the National Institutes of Health, U. S.
Public Health Service.

2 Work done while the author was a holder of a Fulbright Travel Grant, awarded through
the Institute of International Education, New York, N. Y., and the U. S. Educational Foundation
in India.

3 Present address : Department of Zoology, Andhra University, Waltair, India.

353

354

KANDULA PAMPAPATHI RAO

in

<D

V>000'

Q>

•o

12 18 24 6 12 18
Jan. 20 Jan. 21

24 6 12 18 Hrs.
Jan. 22

FIGURE 1. Variations in the rate of water propulsion in a single specimen of M. calijornianus
collected inter-tidally from + 1.0 ft. and kept in darkness at 14 ± 1° C. Dotted line indicates the
tidal cycle in the locality of collection, in this and the following figures.

which the Supply Department of the Marine Biological Laboratory, Woods Hole,
Massachusetts, sent us the required supply of Mytilus cdulis.

MATERIALS AND METHODS

Mytilus calijornianus collected from about +1.0 ft. (tidal datum zero is mean
lower low water, tidal range here about 8 feet) on pilings at Santa Monica, Cali-
fornia, were transferred to aquaria containing sea water at 14 ± 1° C. One large
mussel was placed in each of three enamel-coated pans containing sea water at
9 ± 1°, 14 ± 1°, and 20 ± 1° C., respectively, while a duplicate series of three
pans contained ten to twelve mussels each, at the same three temperatures. All

o'~

A

.+5r

>>•*'

12 18 24
Mar. I

12 18
Mar. 2

24 6

12 18 Hrs.
Mar. 3

FIGURE 2. Rhythmicity in the rate of water propulsion in M. calijornianus collected from a
depth of about 30 ft. off Los Angeles, and kept in darkness at 14° C.

TIDAL RHYTHM IN MYTILUS

355

the containers were covered with lids, making them virtually dark chambers. The
method used for measuring the rate of water propulsion has been detailed elsewhere
(Rao, 1953). Measurements were made at hourly intervals round the clock for
72 hours in continuity and this was repeated at three-day intervals, over a period
of four to six weeks.

The same procedure as above was followed for M. californianus from about
+ 4.0 ft., on pilings and for M. edulis from pilings and from the underside of floats
a few feet away. A collection of M. californianus obtained from a depth of about
30 ft. off the shore near Los Angeles, and a consignment of M. edulis collected at

0>

!Ij

'

j

0.5

a>

'-5

12 18 24
Mar. 16

6 12 18
Mar. 17

24 6 12 Mrs.
Mar. 18

FIGURE 3. Variations in the rate of water propulsion over a period of 48 hours in M. edulis
collected from floats and pilings at Santa Monica, California, and kept in darkness at 14° C.
Upper graph for animals from floats and the lower one for those from pilings.

Barnstable Harbor on Cape Cod and flown to Los Angeles, California, were studied
at9± 1° C. and 14 ± 1° C.

Besides measurements on animals kept in continuous darkness, all the above
samples were subjected to continuous light and the natural day and night environ-
ment and measurements made.

RESULTS
Mytilus californianus

Individuals of M. californianus, when observed in the laboratory, exhibit a
pattern of activity (measured by the rate of water propulsion) which corresponds in

356

KANDULA PAMPAPATHI RAO

time and degree to the tidal levels in the locality from where they have been col-
lected (Fig. 1). The pattern holds good even when several individuals are
grouped together and their activity as a whole is measured. The rhythm is inde-
pendent of temperature over the whole range measured, from 9 to 20° C. (as has
been found by Brown, Bennett and Sandeen, 1953, in the fiddler crab) and persists
for over four weeks in the laboratory in continuous darkness or continuous light,
or the normal day and night environment. No indications of a diurnal rhythm in
the rate of water propulsion were noticed.

Similar results were obtained regardless of the height inter-tidally from which
animals were collected and even with mussels obtained from a sub-tidal population
at a depth of about 30 ft. off the shore (Fig. 2).

0>

. "*" 3

M—

1

0)

•o

1=0,1

\ /
\ xx

'^.-S

\ /
\»..^x

'"N

i

_^

B 24 6 12
Apr.

18 24
20

6 12
Apr.

18
21

24 Hrs

FIGURE 4. Record of rate of water propulsion in M. edulis from Barnstable Harbor on
Cape Cod, kept in darkness at 9° C. at Los Angeles, California. Dotted line indicates the tidal
cycle at Los Angeles.

Mytilus edulis

Samples of M. edulis collected from the same locality and treated similarly
showed a tidal rhythmicity in their rate of water propulsion. What is more remark-
able, mussels collected from the underside of floats showed a pattern of activity
which was quite parallel to that exhibited by mussels collected from the pilings
nearby (Fig. 3).

M. edulis collected at Barnstable Harbor on Cape Cod and studied at Los
Angeles, California — nearly 3000 miles west — showed a rhythm in their rate of
water propulsion which was out of phase with the local tidal cycle by about 6% hrs.
(Fig. 4), and this difference persisted for over four weeks in the laboratory.

Of the mussels obtained from Cape Cod, one dozen were kept in a small wire
cage and, during low tide, were secured at +1.0 ft. to a piling of the pier at the
Kerckhoff Marine Laboratory, Corona del Mar, California, to study the effect of
the local tidal schedule on these mussels. After a week's sojourn at this place,

TIDAL RHYTHM IN MYTILUS

357

they were brought back to the laboratory along with a sample of local M. edulis
attached to the same piling at the same inter-tidal height, which served as controls
for the experimental animals. Study of the activity pattern (at 9 and 14° C.)
revealed a prompt shift in the rhythm to synchronize with the local tidal cycle and

1.5

«- I
'3

0.5
+ 5

I 0

TJ

^-5

A .A

. O / 0-0 f V

V v v

\

24 6 12
Apr. 27

18 24 6 Mrs.
Apr. 28

FIGURE 5. Record of rate of water propulsion in M. edulis from Barnstable Harbor on
Cape Cod, after having been kept for one week at + 1.0 ft. in the inter-tidal at Corona del Mar,
California, and of the control. Upper graph for M. edulis transplanted from Barnstable Harbor,
and lower graph for mussels from pilings at Corona del Mar, California, serving as control.
Dotted line indicates the local tidal cycle.

there is found to be good agreement between the transplanted east-coast mussels and
the local controls (Fig. 5). They continued to keep in phase with the local tidal
cycle for a period of over three weeks in the laboratory.

DISCUSSION

A marked tidal rhythmicity of rate of water propulsion is exhibited by popula-
tions of Mytilus occurring under a great variety of environmental conditions and
persists in the laboratory for long periods (over four weeks) in phase with the tidal
cycle of their natural environment, independent of a wide range of temperature
(9 to 20° C.) and varying conditions of light and darkness. That it exhibits the
same frequency in populations from high and low inter-tidal levels and even in
sub-tidal populations (30 ft. deep) and that it persists in the laboratory, in phase
with the tidal cycle outside, for long periods under constant conditions, demonstrate
the intrinsic (or endogenous) nature of the rhythm.

It is most interesting that such a rhythm is evident in populations from the
underside of floats (and hence not subject to the direct physical effects of the tides),
with the same frequency and in phase with the local tidal cycle. It is equally of
interest that a persistent rhythm with the same frequency, but out of phase with

358 KAXDULA PAMPAPATHI RAO

the local tidal cycle, is exhibited by mussels removed nearly 3000 miles west from
their natural environment. Such instances as these indicate that the rhythm, once
set, is independent of external factors, such as cosmic influences, and can persist
over long periods in the laboratory.

Instances like the foregoing demonstration of a tidal rhythm in a single species
under a great variety of natural conditions lead one to suppose that organisms in
general have rhythmic properties and that the frequency of the rhythm is intrinsic
and perhaps inherited. But how such intrinsic rhythms at a given frequency
come to be in synchrony with rhythmic events in nature is difficult to answrer.
But the ease with which they can be reset to suit a new environment, without a

*

change in the frequency, though not abundantly demonstrated, is of sufficient
significance inasmuch as it helps us to understand the existence of so many instances
of tidal or other kinds of rhythmic behavior patterns. An intrinsic, inherited
rhythmic pattern of activity is set in phase with external events of a rhythmic nature,
which perhaps are of the same frequency as the organismic ones. Transplanta-
tion, as has been done for the first time in the above case, offers an ideal tool for
studying this phenomenon in greater detail. Likewise, studying laboratory-grown
individuals of species which show a rhythmic behavior in their natural environment,
might yield fruitful results.

But the degree to which the rhythm is marked, perhaps, is dependent on the
amplitude of the environmental rhythm. Thus the different findings (Bohn and
Pieron, 1906; Bohn, 1906, 1907; Pieron, 1906, 1908; Gee, 1913; Parker, 1916;
Crozier, 1921, and Hoffman, 1926) on the rhythmic behavior in sea anemones may
be due to the fact that the intrinsic rhythm becomes marked and measurable only
when the fluctuations of the environmental factors reach a certain, but unknown,
threshold value.

SUMMARY

1. The occurrence of a tidal rhythm in the rate of water propulsion in Mytilus
calijornianus, collected from high and low inter-tidal levels and from a depth of
30 ft. off the shore, and also in M. edulis collected from pilings and the underside
of floats, has been demonstrated.

2. Such a rhythm is independent of temperature (9 to 20° C.) and persists
in the laboratory, in phase with the external tidal cycle, for over four weeks, in
continuous darkness, or continuous light or the natural day and night environment.

3. No indications of a diurnal rhythm in the rate of water propulsion have
been observed.

4. A rhythm of similar frequency, but out of phase with local tidal cycle by
about 6% hrs., was observed in M. edulis collected from Barnstable Harbor on Cape
Cod and studied at Los Angeles, California, after transporting them by air.

5. Some of the east coast mussels were secured in the inter-tidal at Corona del
Mar, California, for a week. Examination of their activity pattern after this
period, revealed a prompt shift in their tidal rhythm to synchronize with the local
tidal schedule.

6. The intrinsic nature of the rhythm is discussed and the probable inheritable
nature of the rhythmic properties of organisms, coupled with the ease with which
they could be set in synchrony with natural environmental rhythms, are suggested
as likely causes for the widespread occurrence of rhythmic patterns in organisms.

TIDAL RHYTHM IN MYTILUS 359

7. It is suggested that the degree to which the intrinsic rhythm of the organism
becomes marked and measurable depends upon the amplitude of the environmental
rhythm.

LITERATURE CITED

BOHN, G., 1903. Sur les mouvements oscillatoires des Convoluta roscoffensis. C. R. A cad. Sci.,

Paris, 137 : 576-578.
BOHN, G., 1904. Periodicite vitale des animaux soumis aux oscillations dn niveau des hautes

mers. C. R. Acad. Sci., Paris, 139: 610-611.
BOHN, G., 1906. La persistance du rythme des marees chez {'Actinia ctjitina. C. R. Soc. Biol.,

Paris, 61 : 661-663.

BOHN, G., 1907. Le rythme nycthemeral chez les Actinines. C. R. Soc. Biol., Paris, 62 : 473-476.
BOHN, G., AND H. PIERON, 1906. Le rythme des marees et la phenomene de 1'anticipation

reflexe. C. R. Soc. Biol., Paris, 61 : 660-661.
BROWN, F. A., JR., M. F. BENNETT AND R. C. GRAVES, 1953. Rhythmic activity of the quahog,

Venus mcrcenaria. Anal. Rec., 117: 634-635.
BROWN, F. A., JR., M. F. BENNETT AND M. I. SANDEEN, 1953. Temperature independence of

the frequency of the endogenous tidal rhythmicity of the fiddler crab, Uca piiqnax.

Biol Bull, 105 : 371.
BROWN, F. A., JR., M. F. BENNETT AND H. M. WEBB, 1953. Endogenously regulated diurnal

and tidal rhythms in metabolic rate in Uca pugnax. Biol. Bull., 105: 371.
BROWN, F. A., JR., M. FINGERMAN, M. I. SANDEEN AND H. M. WEBB, 1953. Persistent diurnal

and tidal rhythms of color change in the fiddler crab, Uca puqnax. J. Exp. Zool., 123 :

29-60.
CALHOUN, J. B., 1944. Twenty-four hour periodicities in the animal kingdom. Part I. The

invertebrates. /. Tenn. Acad. Sci., 19: 179-200 and 252-262.
CROZIER, W. J., 1921. Notes on some problems of adaptation. 8. Concerning "Memory" in

actinians. Biol. Bull, 41 : 117-120.
GAMBLE, F. W., AND F. KEEBLE, 1903. The bionomics of Convoluta roscoffensis with special

reference to its green cells. Proc. Roy. Soc., London, 72: 93-98.
GEE, W., 1913. Modifiability in the behavior of the California shore-anemone Cribrina xantho-

grammica Brandt. /. Anitn. Bchav., 3 : 305-328.
COMPEL, M., 1937. Recherches sur la consommation d'oxygene de quelques animaux aquatiques

littoraux. C. R. Acad. Sci., Paris, 205 : 816-818.
COMPEL, M., 1938. Recherches sur la consommation d'oxygene de quelques animaux aquatiques

littoraux. Ann. dc Physiol, 14 : 914-931.
HOFFMAN, R. W., 1926. Periodische Tageswechsel und andere biologische Rhythmen bei den

poikilothermen Tieren (Reptilien, Amphibien, Fische, Wirbellose). Handbuch dcr

Normalen und Pathologischcn Physiologic, 17 : 644-658.

PARKER, G. H., 1916. The behavior of sea-anemones. Proc. Nat. Acad. Sci., 2: 450-451.
PIERON, H., 1906. La reaction aux mares par anticipation reflexes chez Actinia equina. C. R.

Soc. Biol., Paris, 61 : 658-660.

PIERON, H., 1908. La rythmicite chez Actinia equina L. C. R. Soc. Biol., Paris, 65: 726-728.
RAO, K. PAMPAPATHI, 1953. Rate of water propulsion in Mytilus californianus as a function

of latitude. Biol. Bull,, 104: 171-181.

THE RESPIRATORY METABOLISM OF TISSUES OF

MARINE TELEOSTS IN RELATION TO

ACTIVITY AND BODY SIZE l

F. JOHN VERNBERG

Department of Zoology, Duke University, Durham, North Carolina, and
Marine Biological Laboratory, Woods Hole, Massachusetts

Rates of oxygen uptake of tissues of fishes at different temperatures have been
investigated by various workers (Fuhrman et al., 1944, brain of large-mouthed
bass; Peiss and Field, 1950, brain and liver of polar cod and golden orfe; and
Freeman, 1950, brain and muscle of goldfish). In 1953 Vernberg and Gray
reported a direct correlation between general body activity and oxygen metabolic
rate of excised brain. They also noted that within the size range of animals used,
no relationship between body size and rate of oxygen uptake was evident in the
toadfish and the pinfish.

Although some workers reported a decrease in Qo2 of tissues with increasing
body size (Kayser, Le Breton and Schaeffer, 1925 ; Hawkins, 1928; Kleiber, 1941 ;
Weymouth, Field and Kleiber, 1942; and Weymouth et al., 1944), other inves-
tigators do not find this relationship to exist (Terroine and Roche, 1925 ; Grafe,
1925; Crandall and Smith, 1952; Bertalanffy and Pirozynski, 1953). Recently
Krebs (1950), following a determination of the Qo2 of five tissues of nine mammals,
reported that there is not a simple correlation between body size and Qo2 within
the same species, and that, in general tissues of larger species have lower values
than homologous values of tissues from smaller species.

The present investigation was undertaken for two specific reasons. First, to
continue the study of the relationship of activity and metabolism of various tissues
in marine fishes. Secondly, to examine the relationship of tissue metabolism and
body size in a group of poikilothermic vertebrates.

MATERIALS AND METHODS

The oxygen uptake of tissues was determined by the direct method of Warburg.
Liver, muscle, and brain tissue from three species of marine teleost fishes, toadfish
(Opsanus tau), scup (Stenotomus chrysops], and menhaden (Brevoortia tyrannus},
were studied. These three species of fishes were used because of their diverse
habits and differences in general activity levels. Menhaden is an extremely active
swimming form which normally lives and feeds at the surface of the ocean. On
the other hand, the toadfish is a relatively inactive pugnacious bottom-dweller, and
the scup is intermediate to these two in respect to activity.

All animals were killed by severing the spinal cord in the region immediately
posterior to the skull. Brain tissue was obtained by cutting off the roof of the
skull and removing all tissue anterior to the vagal lobes. The brain was blotted

1 Aided by a grant from the Duke University Research Council.

360

TISSUE METABOLISM AND ACTIVITY

361

TABLE I

Respiration of tissues of three species of marine fishes

Species

N

Mean Qo2

Standard deviation

Brain

Toadnsh

28

6.78 ± .290

1.54

Scup
Menhaden

20
21

10.51 ± .578
13.04 ± .721

2.59
3.30

Liver

Toadnsh

27

4.42 ± .323

1.68

Menhaden

11

11. OS ±1.173

3.89

Scup

16

14.87 ±1.219

4.88

Muscle

Scup
Toadnsh

10
18

.410± .064
.727± .084

.202
.356

Menhaden

12

1.024± .140

.485

quickly on filter paper to remove all blood and foreign matter, then weighed and
ground in a dry mortar. Sufficient amount of a phosphate buffer of pH 7.5 (glass
electrode) was added to bring the volume to 3.0 ml. and the brei transferred to a
Warburg flask. Muscles from the dorsal trunk region were treated in the same
manner, using samples weighing about 450 mg. The liver tissue was sliced with
a Stadie-Riggs tissue microtome ; each sample weighed about 125 mg. The center
well of the respirometer flask contained both 0.2 ml. of 10% KOH and filter
paper wicks.

Time between the death of the animal and the beginning of the 10-minute period
of thermal equilibration was kept constant at 10 minutes. Readings, taken at
10-minute intervals, carried for a minimum time of 60 minutes. Manometric
determinations were made in a bath maintained at 30° C. Results are expressed
in terms of wet weight Qo2. Thus Qo2 denotes microliters of oxygen consumed
per gram of wet weight per minute. The water content of the various tissues
studied was determined by drying to a constant weight at 105° C.

This study was conducted at the Marine Biological Laboratory, Woods Hole,
Mass., during the summer of 1953. All specimens were obtained from the Supply
Department and maintained in the laboratory in aerated tanks supplied with running

TABLE II
Significance of differences of means of QO2 of tissue from marine leleost fishes

Tissue

Species compared

Probability

Brain
Brain
Liver
Liver
Muscle
Muscle

Toadfish-scup
Menhaden-scup
Toadfish-menhaden
Menhaden-scup
Scup-toadfish
Toadfish-menhaden

.03

!o7

Highly significant
Highly significant
Highly significant
Significant
Highly significant
Not significant

362

F. JOHN VERNBERG

TABLE III

Water content of tissues of three species of marine fishes

Species

Tissue

No. of

determinations

Average %

Range

Toadfish

Brain

7

83.19

81.1-84.7

Liver

10

73.75

66.9-80.6

Muscle

8

82.18

79.6-86.1

Scup

Brain

6

80.40

78.8-82.4

Liver

8

76.40

71.4-78.8

Muscle

8

78.58

75.4-80.6

Menhaden

Brain

10

78.94

75.3-81.4

Liver

5

60.69

58.4-63.0

Muscle

5

72.88

70.1-73.9

sea water. Scup and toadtish could be kept very well in these tanks but the
menhaden would soon die of apparent oxygen lack. Thus it was necessary to use
these animals as soon as they were brought into the laboratory. Menhaden, par-
tially asphyxiated when brought from the traps, were not normal (Hall, Gray and
Lepkovsky, 1926).

In the statistical analysis of the data pertaining to the relationship of Qo2 to
body size, the following formulae were used :

or

M = aWb

log M = log a + b log W ',

(1)
(2)

where M is the Qo2, W the body weight, and a and b are constants, indicating the

intercept and the slope of the regression line in the log-log plot. Additional statistics

calculated were the standard error (S(\osy. iog JM) and p (coefficient of correlation).

Weights of animals used are as follows: toadfish, average 349 gms., range 78-

TABLE IV

Statistical analysis of relation of Qo2 to body size in tissues of two marine fishes

N

a

b

S(Iog ylog x)

p

Toadfish

Brain

28

2.055

.202

.1115

.780

Liver

27

9.998

-.1448

.0972

.742

Muscle

18

.327

-.1182

.2800

.409

Scup

Brain

20

4.795

.1504

.111

.568

TISSUE METABOLISM AND ACTIVITY

363

586 gms. ; scup, average 166 gins., range 83-462 gins.; and menhaden, average 345
gms., range 193-495 gins.

RESULTS

The Qo2 values of brain, liver and muscles are indicated in Table 1. Significance
of differences of means is shown in Table II.

In respect to brain tissue a definite correlation between animal activity and
oxygen consumption is noted. This is in accord with previous reported results
of Vernberg and Gray (1953). As shown in Table II, there is a significant differ-
ence between the mean Qo2 of all three species.

80 90 100

200
weight in gms.

400

500

Fig I. Qo2 of toadfish liver in relation to body size.

When comparing interspecifically the Qo2 values of liver, no correlation between
total animal activity and rate of oxygen uptake was noted. The liver of scup, the
intermediate form in regard to activity, had a higher metabolic rate than liver of
menhaden, the most active species. The degree of significance of difference of means
is not as great when comparing menhaden and scup as when comparing menhaden
and toadfish.

Scup muscle had the lowest Co., values, menhaden the highest. However,
there is no significant difference between means of toadfish and menhaden. As in

364

F. JOHN VERNBERG

the case of liver, no correlation between animal activity and metabolic rate of muscle
was noted.

Results of water content determination of the tissues studied are shown in
Table III. In general intraspecific values were fairly constant ; liver tissue showed
the greatest variation. Interspecific comparison showed that similar values were
obtained for brain and muscle tissues, but that the liver of menhaden had a much
lower water content than either scup or toadfish livers. It is well-known that
the liver of menhaden contains enormous quantities of oil and this probably accounts

80

100

200 300

weight ingms.

400

500 600

Fig. 2. Qop of toadfish brain in relation to body size.

for the lower water content. Only when comparing Qo2 values of these three
species would the significance between means be appreciably altered when results
were based on dry weights. In this case the average Qo2 values would be : toadfish
16.84 microliters/minute/gm. of dry weight, menhaden, 28.2, and scup 63.0. Thus,
on this basis, the difference between scup and menhaden liver would be highly
significant rather than significant.

Comparison of Qo2 values for liver, brain and muscle from different individuals
of the same species did not show any consistent tendency for one animal to have

TISSUE METABOLISM AND ACTIVITY

365

a higher metabolic rate for all three tissues than another animal. The Qo2 of
brain of one animal may be higher and the liver Qo2 lower than that of another.

The statistical analysis of the relation of Qo2 to body size in toadfish and scup
is presented in Table IV. Qo2 values of menhaden were not evaluated because of
the small size range of animals used (193-495 gms.). Figures 1-4 represent the
log-log plot of Qo2 values of various tissues against body weight. The middle line
is the regression of Qo2, and the two outer parallel lines give the standard error in
per cent, including % of the determinations.

Q02

14
13
12
II
10
9

8
7-

5 •

80

100

400 500 600

200 300

weight in gms.

Fig. 3. Q02 of toadfish muscle in relation to body size.

Toadfish liver (Fig. 1). There is a slight decrease with increasing body
weight but the correlation coefficient is low.

Toadfish brain (Fig. 2). A slight increase in Qo2 values with increasing body
weight is noted. In general these results correspond with the tendency observed
by Bertalanffy and Pirozynski (1953) and Elliott (1948) for oxygen consumption
of mammal brains.

Toadfish muscle (Fig. 3). Similar results to those of brain tissue.

Scup brain (Fig. 4). The same general tendency is noted for brain of scup
as that of toadfish brain.

A significant difference in mean Qo2 is noted between toadfish brain from
Woods Hole and toadfish brain from Beaufort, N. C. (Vernberg and Gray, 1953).

366

F. JOHN VKRNBERG

80

100

150 200 250

weight in gms.

300

400

500

Fig. 4. Qo2 of scup brain in relation to body size.

Determinations were made at the same temperature and the same method was
employed in both studies.

DISCUSSION

Tissue metabolism and activity

Many phases of the activity of marine fishes have been studied and certain
physiological indexes have been correlated with their activity. A direct relation-
ship between blood sugar concentration and activity was noted by Gray and Hall
(1930) ; menhaden 75.2 mg.%, scup 52.6 mg.%, and toadfish 15.4 mg.%. Hall
and Gray (1929) demonstrated a positive correlation between hemoglobin and
activity: menhaden 41 mg. % iron, scup 25.3% iron, and toadfish 13.5% iron.
A correlation between number of immature circulating erythrocytes and activity
was shown by Dawson (1933) : menhaden 16.5%, scup 4.7%, and toadfish less
than 1%. Root (1931) studied the respiratory function of blood of marine fishes
and found a definite adjustment on the part of the blood to the habits or character-
istics of the fishes. His results, as they pertain to activity, are in agreement with
those cited above.

Oxygen consumption determinations by Hall (1929) showed toadfish to have
a low resting metabolic rate with a higher rate for scup. Menhaden have been
found to have a high rate of oxygen consumption. In comparing gill area of

TISSUE METABOLISM AND ACTIVITY 367

menhaden and toadfish, Gray (1947) found that the former has about 10 times more
gill surface than the toadfish per gram of body weight and 15 times more gill area
per square cm. of body surface. Gray (1946) found the scup to be intermediate
to toadfish and menhaden in total number of gill lamellae. Thus, the physiological
indexes of activity would substantiate field observations and indicate that menhaden
is the most active form, toadfish the least active and scup intermediate. Vernberg
and Gray (1953.) found the brain of menhaden to have a higher Qo2 than that of
toadfish. The findings of the present paper demonstrate again the relationship
of brain OQ., and activity for menhaden and toadfish and include results of another
species, the scup.

In view of the fact that the comparative oxygen consumption rate of the entire
organism for menhaden is high and toadfish is low, one might surmise that the
tissues of the menhaden had a higher "basal" metabolic rate than tissues of the
toadfish.

Because so much of an animal's body consists of muscle tissue, one might
expect to find significant differences between Qo2 of muscle of menhaden and toad-
fish. However, no correlation between metabolic rates of liver and muscle with
either activity or total animal O2 consumption was observed. Thus, it would seem
that in the physiological organization of the entire organism, the coordinating
mechanisms of the more active species, the menhaden, are operating in such a
manner as to stimulate the tissues to an activitv level higher than indicated by

j *

in vitro determinations. Many factors are operative in organismic make-up and
would include such factors as hormonal and neural regulators. From the results
reported in this paper it might seem possible to suggest that an integral part of
the coordinating system of the body, the brain, is extremely important in maintain-
ing the "basal" metabolic rate of the entire organism. Thus an animal having brain
tissue with a high "basal" metabolic rate would have a high total organism "basal"
metabolic rate.

Other workers have reported results which would indicate the importance of the
brain to the general physiological functioning of the organism. In work with mam-
mals by Himwich et al. (1939) and Hoagland (1949), rhythmic potential changes
in brain tissue are dependent upon the metabolic rate of the tissue. A correlation
of brain metabolism, respiratory movements and total oxygen consumption to tem-
perature acclimatization was noted by Freeman (1950). He stated that the
metabolic activity of the brain is a major factor in determining the level of the
total oxygen consumption of a fish. The brain exerts this governing action through
its influence on the other tissues of the body.

An interesting question remains to be investigated further. If the brain tissue
Oo.j is correlated with total oxygen consumption, why then should the Qo2 of
brain tissue be slightly increased in older animals, whereas, the Qo2 of the whole
animal is decreased. Undoubtedly the role of the other factors, such as endocrine
relationships, must not be overlooked. Hoagland (1936) emphasized the modifica-
tion of respiratory rhythms by reflexes and humoral agents.

Tissue metabolism and body size

The results of this study indicate that in a poikilothermic animal such as the
toadfish, brain and muscle tissue Qo2 values do not decrease with size as does liver,

368 F. JOHN VERNBERG

but actually show a slight increase in "basal" metabolic rate with size. In general
it would seem that any decrease in basal metabolic rate of the entire organism with
increased size could not be accounted for on the basis of decline in muscle Qo2- Ber-
tanlanffy and Estwick (1953) reported that in the rat, although Qo2 of muscle
decreased slightly with body size, it was not of sufficient magnitude to account for
decreased whole-animal oxygen consumption. Recently Bertanlanffy and Perozyn-
ski (1953) concluded, after investigating 7 different tissues of rats of various sizes,
that any decline in basal metabolic rate depends not upon factors lying in the tissues
themselves but rather on regulative factors in the organism as a whole. The
present investigation would substantiate this view.

Geographical differences

Although the present study was not undertaken specifically to study geographic
physiological adaptation, a significant difference in brain tissue metabolism of two
populations of toadfishes was noted. The question arises as to whether this differ-
ence is due to genetic differences or to an acclimatization phenomenon.

Numerous workers have reported on the relationship of temperature acclimatiza-
tion to whole animal oxygen consumption (Wells, 1935a, 1935b; Fry and Hart,
1948; Sumner and Doudoroff, 1938; Fox, 1936; and Fox and Wingfield, 1937).
In general, animals from a northern habitat or acclimatized at lowered tempera-
tures consume more oxygen when determined at intermediate or elevated tempera-
tures than those that are from a southern area or acclimatized at a higher tempera-
ture. At the tissue level, Peiss and Field (1950) found that brain tissue from
an arctic-adapted fish, the polar cod, had a higher metabolic rate than a warm-
adapted southern species, the golden orfe, when determined at a temperature
which corresponded to the acclimatization temperature of the warm- adapted animal.
Freeman (1950), working with brain of goldfish, noted a similar relationship. The
temperature of the water in which animals were kept averaged approximately 10° C.
lower at Woods Hole than in the region of Beaufort. Thus, one would expect the
brain Qo2 of the northern population to be higher than the southern one. How-
ever, no attempt was made to study this phenomenon at different temperature levels
or to investigate the possible genetic differences.

SUMMARY

1. Determinations were made of the Qo2 of brain, muscle and liver of three
species of marine fishes representing different ecological habitats ; a very active con-
stantly swimming species, menhaden; a sluggish bottom-dweller, toadfish; and an
intermediate form, scup.

2. Although a direct correlation between Qo, of brain and activity of the whole
organism was noted, liver and muscle did not show any correlation with activity.
The possible significance of this relationship was discussed.

3. A slight increase in Qo2 of brain and muscle of toadfish and brain of scup
with increasing body size was noted. The Qo2 of toadfish liver decreased with
body size.

TISSUE METABOLISM AND ACTIVITY 369

LITERATURE CITED

BERTALANFFY, L. VON, AND R. R. ESTWICK, 1953. Tissue respiration of musculature in relation

to body size. Amcr. J. Physio!., 173 : 58-60.
BERTALANFFY, L. VON, AND W. J. PIROZYNSKI, 1953. Tissue respiration, growth, and basal

metabolism. Biol. Bull., 105: 240-256.
CRANDALL, R. R., AND A. H. SMITH, 1952. Tissue metabolism in growing birds. Proc. Soc.

Exp. Biol. Med., 79 : 345-346.
DAWSON, A. B., 1933. The relative numbers of immature erythrocytes in the circulating blood

of several species of marine fishes. Biol. Bull., 64 : 33-43.

ELLIOT, K. A. C, 1948. Metabolism of brain tissue slices and suspensions from various mam-
mals. /. Neurophysiology, 11 : 473-484.
Fox, H. M., 1936. The activity and metabolism of poikilothermal animals in different latitudes—

1. Proc. Zool. Soc. London, Part IV: 945-955.
Fox, H. M., AND C. A. WINGFIELD, 1937. The activity and metabolism of poikilothermal animals

in different latitudes. Proc. Zool. Soc. London, Part III : 275-282.
FREEMAN, J. A., 1950. Oxygen consumption, brain metabolism and respiratory movements of

goldfish during temperature acclimatization, with special reference to lowered tempera-
tures. Biol. Bull, 99 : 416-424.
FRY, F. E. J., AND J. S. HART, 1948. The relation of temperature to oxygen consumption in the

goldfish. Biol. Bull., 94 : 66-77.
FUHRMAN, F. A., NELL HOLLINGER, J. M. CRISMON, J. FIELD n AND F. W. WEYMOUTH, 1944.

The metabolism of the excised brain of the largemouthed bass (Hiiro salnwidcs} at

graded temperature levels. Physiol. Zool., 17 : 42-50.

GRAFE, E., 1925. Problems der Gewebsatmung. Deutsche Mcd. Wchnschr., 51 : 640-642.
GRAY, I. E., 1946. The relation between gill surface and activity in marine fishes. Anat.

Record, 96: 518.
GRAY, I. E., 1946. The relation between gill surface and activity in marine fishes. /. Elisha

Mitchell Sci. Soc., 63 : 106.
GRAY, I. E., AND F. G. HALL, 1930. Blood sugar and activity in fishes with notes on the action

of insulin. Biol. Bull., 58: 217-223.

HALL, F. G., 1929. The influence of varying oxygen tensions upon the rate of oxygen consump-
tion in marine fishes. Amer. J. Physiol., 88: 212-218.
HALL, F. G., AND I. E. GRAY, 1929. The hemoglobin concentration of the blood of marine

fishes. /. Biol. Chcm., 81 : 589-594.
HALL, F. G., I. E. GRAY AND S. LEPKOVSKY, 1926. The influence of asphyxiation on the blood

constituents of marine fishes. /. Bio!. Chem., 67 : 549-554.
HAWKINS, J. A., 1928. The metabolism of tissue from rats of different ages. /. Gen. Physio!.,

11 : 645-647.
HIMWICH, H. E., Z. HADIDIAN, J. F. FAZEKAS AND H. HOAGLAND, 1939. Cerebral metabolism

and electrical activity during insulin hypoglycemia in man. Amcr. J. Physiol., 125 :

578-585.
HOAGLAND, H., 1936. Some pacemaker aspects of rhythmic activity in the nervous system.

Cold Spring Harbor Symposia Quant. Biol., 4 : 267-276.

HOAGLAND, H., 1949. Rhythmic behavior of the nervous system. Science, 109: 157-164.
KAYSER, CHARLES, ELAINE LE BRETON AND G. SCHAEFFER, 1925. Grandeur de la respiration des

tissues et masse active au cours du developpement des organismes. C. R. Acad. Sci.

(Paris), 181: 255-257.
KLEIBER, M., 1941. Body size and metabolism of liver slices in vitro. Proc. Soc. Exp. Biol.

Med., 48 : 419-422.

KREBS, A. H., 1950. Body size and tissue respiration. Biochcm. et Biophys. Acta, 4: 249-269.
PEISS, C. N., AND J. FIELD, 1950. The respiratory metabolism of excised tissues of warm- and

cold-adapted fishes. Biol. Bull., 99: 213-224.
ROOT, R. W., 1931. The respiratory function of the blood of marine fishes. Biol. Bull., 61 :

427^56.
SUMNER, F. B., AND P. DouooROFF, 1938. Some experiments on temperature acclimatization

and respiratory metabolism in fishes. Biol. Bull., 74: 403-429.

370

F. JOHN VERNBERG

TERROINE, E., AND J. ROCHE, 1925. Production calorique et respiration des tissues in vitro chez

les homeothermes. C. R. Acad. Sci., 180: 225-227.
VERNBERG, F. J., AND I. E. GRAY, 1953. A comparative study of the respiratory metabolism of

excised brain tissue of marine teleosts. Biol. Bit!!., 104 : 445-449.
WELLS, N. A., 1935a. Variations in the respiratory metabolism of the Pacific killifish I'undiilus

parvipinnia, due to size, season and continued constant temperature. Ph\siol. Zoo/.,

8: 318-336.
WELLS, N. A., 1935b. Change in rate of respirator}- metabolism in a teleost fish induced by

acclimatization to high and low temperature. Biol. Bull., 69 : 361-367.
WEYMOUTH, F. W., J. M. CRISMON, V. E. HALL, H. S. BELDING AND J. FIELD n, 1944. Total

and tissue respiration in relation to body weight : A comparison of the kelp crab with

other crustaceans and with mammals. Physiol. Zoo/., 17: 50-71.
WEYMOUTH, F. W., J. FIELD n AND M. KLEIBER, 1942. Relationship between body size and

metabolism. Proc. Soc. Exp. Biol. Mod., 49 : 367-370.

A PERSISTENT DIURNAL RHYTHM OF CHROMATOPHORIC
RESPONSE IN EYESTALKLESS UCA PUGILATOR

H. MARGUERITE WEBB, MIRIAM F. BENNETT AND FRANK A. BROWN, JR.1

Department of Physiology and Bacteriology, Goucher College, Towson 4, Md.; Department oj
Biological Sciences, Northwestern University, Evanston, III.; and the ~ x jJL

Marine Biological Laboratory, Woods Hole, Mass.

On the basis of their response to eyestalk removal crustaceans have been classi-
fied into three types (Brown, 1948). The members of Group I, represented by
Palaemonetes, respond to eyestalk removal by dispersion of the dark pigment and
to injection of eyestalk extract by concentration of the pigment. Crago, the single
member of Group II, responds to eyestalk removal by assuming an intermediate
condition with the dark pigment partially dispersed. Group III contains all of the
brachyurans except Sesarma (Enami, 1951) and is characterized by the complete
concentration of the dark pigment on removal of eyestalks. Further investigations
have led to the development of a concept of dual hormonal control in members
of two of these groups. The evidence supporting such a concept for Crago has
been reviewed by Brown (1948). In 1952, Brown, Webb and Sandeen demon-
strated the presence in the central nervous system of Palaemonetes of a red-pigment-
dispersing substance and adduced arguments in favor of its normal functioning.

The study of the mechanism of control of the melanophores of Uca, a representa-
tive of Group III, has been complicated by the presence in these animals of a per-
sistent diurnal rhythm. Under the influence of the rhythmical mechanism the
black pigment is dispersed by day and concentrated by night. The extent of disper-
sion is susceptible to modification by such factors as light, background, and tempera-
ture and in at least one species, Uca pugnax, an endogenous tidal rhythm has been
shown to influence the condition of the chromatophores (Brown, Fingerman,
Sandeen and Webb, 1953). All of these factors are thought to act on the
chromatophores, at least in part, by virtue of alterations in the blood level of one
or more hormones. The eyestalks are known to produce a hormone which causes
dispersion of the black pigment. The central nervous system has been shown to
contain a substance which disperses the black pigment of eyestalkless animals
(Brown, 1948; Sandeen, 1950) but the participation of this substance in physio-
logical color change has not been conclusively demonstrated.

Although all efforts at direct demonstration of a substance acting to concen-
trate the black pigment of Uca have ended in failure, there are cases in which
investigators have been led to postulate the existence of such a substance (Brown
and Stephens, 1951; Brown and Hines, 1952). Furthermore, Brown and Scuda-
more (1940) reported observations which suggested that eyestalkless Uca do not
have their rhythm completely abolished.

1 This investigation was supported in part by contracts NONR-09703 and NONR-122803
with the Office of Naval Research, and in part by a grant from the Graduate School of
Northwestern University.

371

372 WEBB, BENNETT AND BROWN

The data to be reported here contribute to our understanding of two problems :
the mechanism of control of the black chromatophores of Uca pugilator, and the
mechanisms involved in diurnal rhythmicity.

EXPERIMENTAL PROCEDURE

All of these animals used in these experiments were specimens of Uca pugilator
collected at Chapoquoit beach, near Woods Hole, Mass, during August, 1953. In
the laboratory the animals were kept in white enamelled pans with a small amount
of water and at a constant illumination of about 2 ft. c.

Two types of experiments were performed. One type involved a study of the
changes occurring in chromatophores of legs which had been autotomized and were
then maintained in sea water for a period of one hour. The other type of experi-
ment consisted of injection of various concentrations of eyestalk extracts into
eyestalkless animals.

Study of changes in chromatophores of isolated legs. Animals were forced to
autotomize two or three legs each by applying pressure or by slightly injuring a
distal segment of a walking leg. The legs so obtained were placed in sea water
and observed at the time of isolation and again after thirty and sixty minutes. The
total number of legs removed at any one time varied from six to ten and the legs
were taken from two, three or five animals, depending upon the particular experi-
ment. On each occasion a minimum of six legs from normal animals and the
same number from eyestalkless animals were observed. Legs were isolated from
two such groups at 66 different times ; the total number of animals used was 226.
The experiments were performed on four different days. In one series legs were
removed every hour from 8 P.M. of one day until 8 P.M. of the succeeding day;
in the other three series legs were isolated from both normal and eyestalkless animals
as follows: 1) hourly from 8 A.M. till the next 1 A.M.; 2) hourly from 7 A.M.
until the next 1 A.M. ; 3) at 1, 2, 3, and 4 P.M. and at 8 and 9 P.M. of the same day.
The eyestalkless donors had been operated on not more than 48 hours and not less
than 8 hours before being used in an experiment.

Injection of eyestalk extract. A stock solution was made by grinding 10 dried
eyestalks and extracting in one cc. of sea water. This solution was boiled for one
minute and then cooled to room temperature. Five-hundredths cc. of this extract
(the amount used for a single injection) contained % of an eyestalk or V± of the
normal complement of eyestalk tissue of one animal. Such an extract is said to
have a concentration of one quarter. This stock extract was then used to make
up a series of concentrations as follows: 1/16, 1/64, 1/128, 1/512, 1/1024, and
1/2048. Seven groups of five eyestalkless animals were injected for each experi-
ment. Each of the five animals in a group received 0.05 cc. of one dilution injected
at the base of a walking leg. A control group of five animals received 0.05 cc. of
sea water. The state of the chromatophores of each animal was determined at 15
minutes after injection, again at 30 minutes and at 30-minute intervals until they
had returned to stage 1.

This experiment was performed four times, using two stock extracts. On one
occasion a stock extract was made up in the afternoon, part of it was used immedi-
ately and the remainder refrigerated and used in the evening. Again an extract

DIURNAL RHYTHM IN EYESTALKLESS UCA

373

was made up and part of it used in the evening while a second portion was refrigerated
and used the next morning.

Since the chromatophores are initially in stage 1, the observed chromatophore
stage minus 1 gives a measure of the dispersion present at any given time of observa-
tion. Summing the corrected values obtained during one experiment for any one
concentration of extract then gives a measure of the activity of that extract. The
activity of each extract tested was calculated in this manner.

RESULTS AND DISCUSSION

Figure 1 shows the average stage of the chromatophores of legs isolated from
eyestalkless animals, as determined 60 minutes after removal, plotted against time
of day at which autotomy occurred. The data used for this curve are those obtained

LJ

ti
£3

O
Q.

0 2
|

1 I

X
O

I

J I

I

I

\2

8

\2

8

8

PM AM PM

FIGURE 1. Average index of the melanophores in legs isolated from eyestalkless crabs sixty
minutes after isolation at various times during a twenty-four hour period.

in the complete 24-hour series of observations. The other experiments of this
type yielded entirely similar results. Since the initial average condition for all
chromatophores in legs from eyestalkless animals was stage 1.0 the distance of
any point from the abscissa gives a measure of the amount of dispersion occurring
at that time of day. It can be seen that during the hours from 8 P.M. to 1 A.M. only
very slight dispersion of the pigment occurs in 60 minutes. From 1 A.M. until
12 M. there is a gradual increase in the amount of dispersion observed, while from
noon until 8 P.M. a gradually decreasing amount is found.

Curve A of Figure 2 represents the average initial stages of the chromatophores
of legs isolated from intact animals plotted against the time of day of removal.
Curve B of Figure 2 is obtained by similarly plotting the average stages, at 60

374

WEBB, BENNETT AND BROWN

minutes after removal, of the chromatophores of legs from intact animals. These
data were obtained on the same day and at the same times as those shown in
Figure 1. Results obtained in other experiments of this series were similar to
those represented in Figure 2. It can he seen that Curves A and B of Figure 2
are similar in general shape to that describing the conditions for eyestalkless animals.
The values are low from 8 P.M. to 2 A.M., increase rather rapidly until about 5 A.M.,
remain fairly constant until 1 P.M. and then decrease gradually until 8 P.M. The
distance between a point of Curve A and the point for the same hour on Curve B
gives a measure of the change, concentration or dispersion, occurring in legs removed
at that particular hour of the day. It is immediately obvious that the chromatophores
in legs isolated from intact animals undergo more or less concentration throughout

LJ

O 4

LJ 3

a:

o

x

£2

0
cr

x
u

A • INITIAL STAGE
B o FINAL STAGE

J I 1 I I I I I

8

PM

12
AM

8

12

8

PM

FIGURE 2. Average index of the melanophores in legs isolated from normal crabs at the time of
isolation (A) and sixty minutes afterwards (B) as a function of the time of day.

most of the day. In only one case (10 A.M.) is there any noticeable degree of
dispersion.

Since all of the isolated legs were kept in sea water and at a constant light
intensity, any differences found among experimental groups must be accounted for
in terms of differences in the body fluids at the time of isolation. The fact that the
chromatophores of legs isolated from eyestalkless animals show different degrees
of dispersion at the end of 60 minutes, depending upon time of day of isolation,
clearly indicates a diurnal rhythm in eyestalkless animals marked by alterations
in the body fluids of the animals.

The results of the injection experiments are presented in Table I. The activity
values (calculated by the method previously described) are given for each of the
four times the experiment was performed. It is seen that for each of the four lowest

DIURNAL RHYTHM IN EYESTALKLESS UCA

375

concentrations the activity is lower at night than in the daytime. Although it is
possible that some decrease in activity of the extracts occurred between the time
of first preparing an extract and the time of injection any such decrease should
tend to obscure the differences rather than accentuate them. Thus a decrease
with time could result in relatively low daytime values when the extract was made
up at night. When the extract was made up in the daytime and used at night the
time elapsed was only six hours as compared with 12 hours when the reverse order
was followed. The reduction in activity observed is such as would be expected
by a two- to four-fold reduction in concentration.

The results obtained in both types of experiments clearly demonstrate that a
diurnal rhythm exists in eyestalkless Uca pugilator. The changes observed in

TABLE I

Difference between day and night activity for a series of concentrations of eyestalk extract

injected into eyestalkless animals

Concentration

Activity of extracts

Difference

Day

Night

1/16

18.8
16.0

20.0
18.4

1.2

2.4

1/64

12.4
12.0

13.7
9.0

1.3

-3.0

1/128

10.4
8.0

8.4
7.8

-2.0
-0.2

1/512

6.0
4.8

2.2
3.6

-3.8
-1.2

1/1024

1.8
2.6

0.2
0.8

-1.6

-1.8

1/2048

1.6
1.0

0.0
0.0

-1.6
-1.0

isolated legs show that the rhythm may be characterized as consisting of two distinct
phases whose time relationships correspond quite closely with those of the rhythm
found under the same conditions for normal animals. From the results obtained
following injection of eyestalk extract it is not possible to describe the duration
of the phases. The "night injections" were made in both cases shortly after 8 P.M.
and it is clear that the response of the eyestalkless assay animals was different from
that observed when injections were made in either morning or afternoon.

When an attempt is made to define the nature of the rhythm in terms of
substances in the body fluid it is immediately obvious that two substances must
be involved. A dispersing substance produced in the eyestalks has long been
recognized and it is reasonable to assume that the disappearance of this substance

376 WEBB, BENNETT AND BROWN

permits the concentration of pigment that occurs in legs isolated from normal
animals. The absence, or presence in smaller concentrations, of dispersing hormone
at night might be assumed but whether this is sufficient to account for the observa-
tions is questionable. If one postulates a single substance which causes dispersion
of the pigment and which is present during the day and absent or reduced in amount
at night then one is implicitly assuming that in the absence of any hormone the
pigment will be concentrated. Following eyestalk removal the black pigment
is maintained in the concentrated condition but disperses in isolated legs. If this
concentrated condition is maintained by virtue of the absence of chromatophorotropic
hormone then there is no logical explanation for the dispersion that follows isola-
tion. The conclusion is therefore inescapable that the pigment is maintained in
the concentrated condition by some factor which is present in the body fluid and
which disappears gradually from the isolated legs. The central nervous system
is a known source of dispersing hormone but at the present time no source of a
black-pigment-concentrating substance has been demonstrated.

Assuming that there are two antagonistic substances which function in the
control of the black chromatophores of Uca pngilator the rhythm of eyestalkless
animals appears to consist of an increased amount of concentrating factor at night.
The rhythm of normal animals appears to consist of the production primarily of
dispersing hormone in the day phase and primarily of concentrating substance at
night. Regardless of the site of production of the concentrating factor it seems
likely that control of secretion is nervous and it is clear that the structures of the
eyestalk are not essential for continued rhythmicity.

The results obtained on injection of eyestalk extract are consistent with the
interpretation that eyestalkless animals possess in their body fluid at night a sub-
stance antagonistic to the dispersing hormone of the eyestalk. The fact that no
difference was observed with the highest concentrations used may indicate that a
maximum response was obtained even at night and that therefore no further
response could be expected.

SUMMARY

1. The responses of the black chromatophores of Uca pugilator as observed in
legs autotomized and maintained in sea water are described.

2. The pigment in legs from normal animals in the day (dispersed) phase
becomes concentrated after isolation; that from normal animals in the night (con-
centrated) phase remains concentrated.

3. The pigment in legs isolated from eyestalkless animals disperses in the
daytime and fails to disperse when the legs are removed from 8 P.M. to 2 A.M.

4. The activity on eyestalkless animals of a series of concentrations of eyestalk
extract was determined in the daytime and at night. Four of the six concentrations
tested were found to be more effective in the daytime than at night.

5. The results clearly demonstrate the existence of a diurnal rhythm in eye-
stalkless animals and that the structures of the eyestalk are not necessary for this
rhythm.

6. The data provide strong evidence that a black-pigment-concentrating sub-
stance participates in the regulation of the chromatophore system of these animals.

DIURNAL RHYTHM IN EYESTALKLESS UCA 377

LITERATURE CITED

BROWN, F. A., JR., 1948. Hormones in crustaceans. Chapter V, The Hormones. Academic

Press, Inc. New York.
BROWN, F. A., JR., M. FINGERMAN, M. I. SANDEEN AND H. M. WEBB, 1953. Persistent diurnal

and tidal rhythms of color change in the fiddler crab, Uca pugnax. J. Exp. Zool., 123 :

29-60.
BROWN, F. A., JR., AND M. N. HIKES, 1952. Modifications in the diurnal pigmentary rhythm of

Uca effected by continuous illumination. Physiol. Zool., 25 : 56-70.
BROWN, F. A., JR., AND H. H. SCUDAMORE, 1940. Differentiation of two principles from the

crustacean sinus gland. J. Cell. Comp. Physiol., 15: 103-119.
BROWN, F. A., JR., AND G. C. STEPHENS, 1951. Studies of the daily rhythmicity of the fiddler

crab, Uca. Modifications by photoperiod. Biol. Bull., 101 : 71-83.

BROWN, F. A., JR., H. M. WEBB AND M. I. SANDEEN, 1952. The action of two hormones regulat-
ing the red chromatophores of Palaemonetes. /. Exp. Zool., 120 : 391-421.
ENAMI, M., 1951. The sources and activities of two chromatophorotropic hormones in crabs of

the genus Sesarma. I. Experimental analyses. Biol. Bull., 100 : 28-43.
SANDEEN, M. I., 1950. Chromatophorotropins in the central nervous system of Uca pugilator

with special reference to their origins and actions. Physiol. Zool., 23 : 337-352.

INDEX

A CCUMULATION of phosphate by fertilized
sea urchin eggs, 297.

Acid fuchsin permeability of grasshopper sen-
sory pegs, 122.

Activity of fly cholinesterase, relation between
pH and, 139.

Activity, relation of to metabolism of marine
teleost tissues, 360.

Air, role of in radiation resistance of Parame-
cium, 253.

Alaska, helminth fauna of, 107.

Algae as food for sea urchin, 328.

Algae, blooms of, in Long Island waters, 198.

Amino acid requirements of Tribolium, 149.

Aniino acids in Phormia larval salivary secre-
tion, 178.

Amoebo-flagellates, nutritional studies on, 269.

Anatomy and behavior of vascular system in
Nereis, 69.

Anidian chick embryos, 48.

Antennae of grasshopper, permeability of sen-
sory pegs on, 122.

Anticoagulant substances from Spisula tissues,
129.

Antimitotic substances from ovaries, 158.

Arbacia lixula, interaction of sperm and jelly
coat in fertilization of, 169.

ARVY, L., AND M. GABE. The intercerebralis-
cardiacum-allatum system of some Ple-
coptera, 1.

Asterias ovaries, extraction of antimitotic sub-
stance from, 158.

Asteroids, redox indicator patterns in exogas-
trulae of, 21.

ATP, effect of on endogenous oxygen uptake of
developing grasshopper embryos, 265.

Axenic culture conditions for Tetramitus, 269.

OARNACLE, respiration of normal larvae of,
323.

Basiconic sense organs on grasshopper an-
tennae, permeability of, 122.

Beetle (Tribolium), amino acid requirements
of, 149.

BENNETT, MIRIAM F. See H. M. WEBB, 371.

Biological effects of x-rays in Paramecium, 253.

Blastoderm, chick, effect of lowered incubation
temperature on, 48.

Blastoderms, chick, wound healing in, 39.

Blastomeres, isolated, of Peronella, develop-
ment of, 83.

Blood anticoagulant substances from Spisula,

129.

Blood vessels of Nereis, 69.
"Blooms," phytoplankton, ecology of in Long

Island waters, 198.
BODINE, J. H., AND W. L. WEST. Effect of

adenosinetriphosphate (ATP) on the endo-

genous oxygen uptake of developing grass-

hopper embryos, 265.
Body size, relation of to metabolism of marine

teleost tissues, 360.
BRENT, MORGAN M. Nutritional studies on

the amoebo-flagellate, Tetramitus ros-

tratus, 269.
BROOKS, SUMNER C., AND EDWARD L. CHAM-

BERS. The penetration of radioactive

phosphate into marine eggs, 279.
BROWN, FRANK A., JR., M. FINGERMAN AND

M. N. HINES. A study of the mechanism

involved in shifting of the phases of the

endogenous daily rhythm by light stimuli,

308.
BROWN, F. A., JR. See H. M. WEBB, 371.

, R. M. A new marine cercaria from

the Woods Hole region and its bearing on

the interpretation of larval types in the

Fellodistomatidae, 15.
Carbohydrates, utilization of in chick wound

healing, 39.
Carbon monoxide, effect of on diapausing silk-

worm larva, 210, 238.
Cecropia silkworm, metabolism of during dia-

pause and development, 210, 238.
Cell division, effects of ovary extracts on, 158.
Cercaria laevicardii, description of, 15.
Cercaria, marine, from Woods Hole region, 15.
CHADWICK, L. E., J. B. LOVELL AND V. E.

EGNER. The relationship between pH and

the activity of cholinesterase from flies,

139.

CHAET, ALFRED B. See L. V. HEILBRUNN, 158.
Chaetopterus, extraction of antimitotic sub-

stance from eggs of, 158.
CHAMBERS, EDWARD L., AND W. E. WHITE.

The accumulation of phosphate by fertilized

sea urchin eggs, 297.
CHAMBERS, EDWARD, L. See S. C. BROOKS,

279.

Character of Phormia salivary secretion, 178.
Chick blastoderms, wound healing in, 39.

378

INDEX

379

Chick embryo, effect of lowered incubation

temperature on growth and differentiation

of, 48.
CHILD, C. M. Redox indicator patterns in

relation to echinoderm exogastrulation. II.

Reduction patterns, 21.
Cholinesterase from fly heads, relationship

between activity of and pH, 139.
Chromatography of antimitotic substances

from ovaries, 158.
Chromatography of Phormia larval salivary

secretion, 178.

Chromatography of sea urchin gut NPN, 328.
Chromatography of sea urchin jelly-coat solu-
tions, 169.
Chromatophoric response in eyestalkless Uca,

371.

Corpus allatum in Plecoptera, 1.
Crab, neurosecretion in thoracic ganglion of, 60.
Culture method for amoebo-flagellate, 269.
Culture of chick blastoderms in vitro, 39.
Cyanide, effect of on metabolism of Cecropia

silkworm, 210, 238.
Cytochrome oxidase in thoracic muscle of

woodroach, 341.
Cytochrome oxidase system in diapause and

development of Cecropia silkworm, 238.

1~V\N, KATSUMA. See KAYO OKAZAKI, 83.

Dendraster excentricus, redox indicator pat-
terns in exogastrulae of, 21.

Developing grasshopper embryos, effects of
ATP on endogenous oxygen uptake of, 265.

Development of Cecropia silkworm, physiology
of, 210, 238.

Development of Peronella, 83.

Diapause, insect, physiology of, 210, 238.

Dietary requirements of Tribolium, 149.

Differentiation of chick embryo, effect of low-
ered incubation temperature on, 48.

Digenetic trematode larval types, 15.

Distribution of P-31 and P-32 in dorsal and
ventral halves of frog gastrula, 318.

Diurnal rhythm of chromatophoric response in
Uca, 371.

Dorsal gastrula half of frog embryo, distribution
of phosphorus in, 318.

Dose-action curve for x-ray-treated Mormo-
niella, 100.

DUNN, ARNOLD. See L. V. HEILBRUNN, 158.

UCHINODERM exogastrulation, redox indi-
cator patterns in relation to, 21.

Echinoderm (sea urchin), nutrition of, 328.

Echinoderm eggs, accumulation of phosphates
by, 297.

Echinoderm partial larvae, metamorphosis of,
83.

Echinoids, redox indicator patterns in exogas-
trulae of, 21.

Ecology of California Littorina, 185.

Ecology of phytoplankton blooms in Long
Island waters, 198.

Effect of ATP on endogenous oxygen uptake of
developing grasshopper embryos, 265.

Effect of lowered incubation temperature on
growth and differentiation of chick embryo,
48.

Effects of x-rays on Paramecium, 253.

Egg, sea urchin, interaction of jelly-coat and
sperm in fertilization of, 169.

Eggs, marine, penetration of P-32 into, 279.

Eggs, sea urchin, accumulation of phosphates
by, 297.

EGNER, V. E. See L. E. CHADWICK, 139.

Embryo, chick, effect of lowered incubation
temperature on growth and development
of, 48.

Embryo, chick, wound healing in, 39.

Embryo, frog, distribution of phosphorus in
dorsal and ventral gastrula halves of, 318.

Embryo, grasshopper, effect of ATP on endog-
enous oxygen uptake of, 265.

Embryo, sand dollar, metamorphosis of, 83.

Endocrine glands of Plecoptera, 1.

Endogenous daily rhythm shifted by light
stimuli, 308.

Endogenous oxygen uptake of developing grass-
hopper embryo, effect of ATP on, 265.

Enzyme kinetics of cholinesterase from fly
heads, effect of pH on, 139.

Enzymes, oxidative, in thoracic muscles of
woodroach, 341.

Eriocheir japonicus, neurosecretion in thoracic
ganglion of, 60.

Erosive activities of California Littorina, 185.

Exogastrulation in echinoderms, 21.

Eye-color mutation in Mormoniella, 100.

Eyestalkless Uca, persistent diurnal rhythm of
chromatophoric response in, 371.

pAUNA, helminth, of Alaska, 107.

FEDER, NED. See HOWARD A. SCHNEIDERMAN,

230.
Fellodistomatidae, interpretation of larval

types in, 15.
Fertilization of marine eggs, effect of on rate of

P-32 uptake, 279.
Fertilization of sea urchin egg, interaction of

sperm and jelly-coat in, 169.
Fertilized sea urchin eggs, accumulation of

phosphate by, 297.
FIGGE, FRANK H. J. See RALPH WICHTERMAN,

253.

FINGERMAN, M. See F. A. BROWN, JR., 308.
Fishes, antimitotic extracts from ovaries of, 158.

380

INDEX

Fly head cholinesterase, relation between activ-
ity of and pH, 139.

Food and feeding of Strongylocentrotus, 328.

FRAENKEL, G., AND GLENN E. PRINTY. The
amino acid requirements of the confused
flour beetle, Tribolium confusum, 149.

FRAENKEL, G. See H. H. MOOREFIELD, 178.

FRASER, RONALD C. The utilization of some
carbohydrates by in vitro cultured chick
blastoderms in wound healing, 39.

Function of vascular system in Nereis, 69.

, M. See L. ARVY, 1.

Ganglion, thoracic, of crab, neurosecretion in,

60.
Gaseous pressures, high, respirometer for meta-

bolic studies at, 230.
Gases, effects of on physiology of diapausing

silkworm, 210, 238.
Gastrula of frog, distribution of phosphorus in,

318.

GIARDINA, G. See A. MONROY, 169.
GIESE, A. C. See REUBEN LASKER, 328.
Glucose, role of in chick wound healing, 39.
Glycogen reserves in Teredo larvae, 323.
Gradient of phosphorus distribution in frog

gastrula, 318.
GRANT, PHILIP. The distribution of phos-

phorus (P-31 and P-32) in dorsal and ven-

tral halves of the Rana pipiens gastrula,

318.
Grasshopper antennae, permeability of sensory

pegs on, 122.
Grasshopper embryos, effect of ATP on endog-

enous oxygen uptake of, 265.
Great South Bay, ecology of phytoplankton

blooms in, 198.
Gross metabolic efficiency of California Lit-

torina, 185.
Growth and differentiation of chick embryo,

effect of lowered incubation temperature

on, 48.

UARRISON, JOHN R., AND IRVING KLEIN.

Effect of lowered incubation temperature

on the growth and differentiation of the

chick embryo, 48.

Healing, wound, in chick blastoderms, 39.
HEILBRUNN, L. V., ALFRED B. CHAET, ARNOLD

DUNN AND WALTER L. WILSON. Antimi-

totic substances from ovaries, 158.
Helminth fauna of Alaska, 107.
HENNACY, R. E. See CHARLES E. LANE, 323.
Heparin-like anticoagulant substances in Spi-

sula, 129.
High gaseous pressures, respirometer for meta-

bolic studies at, 230.

HINES, MARGARET N. See FRANK A. BROWN,
JR., 308.

Histology of crab neurosecretory cells, 60.

Histophysiological study of Plecoptera endo-
crine glands, 1.

Hydrogen ion concentration, relationship be-
tween activity of fly head cholinesterase
and, 139.

T N vitro culture of chick blastoderms, 39.

Incubation temperature, lowered, effect of on

growth and differentiation of chick embryo,

48.
Indicators, redox, in relation to echinoderm

exogastrulation, 21.

Insect diapause, physiology of, 210, 238.
Interaction between sperm and jelly-coat in sea

urchin egg fertilization, 169.
Intercerebralis-cardiacum-allatum system of

some Plecoptera, 1.

Intermediate stages of sea otter helminth para-
sites, 107.
Interpretation of larval types in Fellodistoma-

tidae, 15.

Intertidal snails, ecology of, 185.
Investigations on interaction between sperm

and jelly-coat in fertilization of sea urchin

egg, 169.

Irradiation of Mormoniella with x-rays, 100.
Irradiation of Paramecium with x-rays, 253.
Isolated blastomeres of Peronella, development

of, 83.

T ELLY-coat, interaction of with sperm in sea
urchin egg fertilization, 169.

L^ LEIN, IRVING. See JOHN R. HARRISON, 48.
KRAMER, SOL. See W. H. McSHAN, 341.
T AMELLIBRANCH, cercaria from, 15.

LANE, CHARLES E., J. Q. TIERNEY AND R. E.
HENNACY. The respiration of normal lar-
vae of Teredo bartschi Clapp, 323.

Larvae of helminth parasites in Alaska, 107.

Larvae of Peronella, metamorphosis of, 83.

Larvae of Teredo, respiration of, 323.

Larvae of Tribolium, amino acid requirements
of, 149.

Larval salivary secretion of Phormia, character
and ultimate fate of, 178.

Larval types of Fellodistomatidae, interpreta-
tion of, 15.

LASKER, REUBEN, AND A. C. GIESE. Nutrition
of the sea urchin, Strongylocentrotus pur-
puratus, 328.

INDEX

381

Lethality and biological effects of x-rays in

Paramecium, 253.
Leucophaea maderae, oxidative enzymes in

thoracic muscles of, 341.
Light, role of in shifting endogenous daily

rhythm, 308.
Littorina, size distribution, erosive activities

and gross metabolic efficiency of, 185.
Localization of heparin-like substances in

Spisula, 129.
Long Island waters, ecology of phytoplankton

blooms in, 198.

LOVELL, J. B. See L. E. CHADWICK, 139.
Lowered incubation temperature, effect of on

growth and differentiation of chick embryo,

48.
Lytechinus eggs, accumulation of phosphates

in, 297.
Lytechinus, penetration of P-32 into eggs of,

279.

J^jAGGIO, R. See A. MONROY, 169.

Magnesium ion, role of in maximal stimulation
of ATP, 265.

Marine cercaria from Woods Hole region, 15.

Marine eggs, penetration of P-32 into, 279.

Marine intertidal snails, ecology of, 185.

Marine teleosts, respiratory metabolism of tis-
sues of, 360.

MATSUMOTO, KUNIO. Neurosecretion in the
thoracic ganglion of the crab, Eriocheir
japonicus, 60.

McSHAN, W. H., SOL KRAMER AND VERA
SCHLEGEL. Oxidative enzymes in the
thoracic muscles of the woodroach Leu-
cophaea maderae, 341.

Measurement of metabolism, method for, 230.

Melanoplus differentialis, effect of ATP on
endogenous oxygen uptake of embryos of,
265.

Metabolic efficiency of California Littorina,
185.

Metabolic studies at high gaseous pressures,
respirometer for, 230.

Metabolism of ATP-treated grasshopper em-
bryos, 265.

Metabolism of Cecropia silkworm, 210, 238.

Metabolism of teleost tissues, 360.

Metabolism of Teredo larvae, 323.

Metacercariae in Alaskan fauna, 107.

Metachromasia in Spisula tissues, 129.

Metamorphosis of Cecropia silkworm, 210,
238.

Metamorphosis of partial larvae of Peronella,
a sand dollar, 83.

Method for studying metabolism at high gase-
ous pressures, 230.

Mitosis, effects of ovary extracts on, 158.

Modifiability of tidal rhythmicity of rate of
water propulsion in Mytilus by transplan-
tation, 353.

MONROY, A., L. Tosi, G. GIARDINA AND R.
MAGGIO. Further investigations on the
interaction between sperm and jelly-coat
in the fertilization of the sea urchin egg,
169.

MOOREFIELD, HERBERT H., AND G. FRAENKEL.

The character and ultimate fate of the
larval salivary secretion of Phormia regina
Meig., 178.

Moriches Bay, ecology of phytoplankton
blooms in, 198.

Mormoniella, x-ray dose-action curve for eye-
color mutations in, 100.

Musca domestica, relation between cholines-
terase activity of and pH, 139.

Muscles of woodroach, oxidative enzymes in,
341.

Mutation, eye-color, in Mormoniella, 100.

Mytilus, tidal rhythmicity of rate of water
propulsion of, 353.

XTANNOCHLORIS, ecology of blooms of,
198.

Nereis, anatomy and behavior of vascular sys-
tem in, 69.

Neurosecretion in the thoracic ganglion of the
crab, Eriocheir japonicus, 60.

Neurosecretory cells of Plecoptera, 1.

NICOLL, PAUL A. The anatomy and behavior
of the vascular systems in Nereis virens
and Nereis limbata, 69.

Nitrogen requirements of phytoplankton in
Long Island waters, 198.

NORTH, WHEELER J. Size distribution, erosive
activities, and gross metabolic efficiency of
the marine intertidal snails, Littorina
planaxis and L. scutulata, 185.

Nutrition of sea urchin, 328.

Nutrition of Tribolium, 149.

Nutritional studies on amoebo-flagellate, 269.

(~\KAZAKI, KAYO, AND KATSUMA DAN. The
metamorphosis of partial larvae of Pero-
nella japonica Mortensen, a sand dollar, 83.

Otter, sea, parasites of, 107.

Ova, marine, penetration of P-32 into, 279.

Ovaries, antimitotic substances from, 158.

Oxidases in diapausing Cecropia larva, 210,
238.

Oxidative enzymes in thoracic muscles of wood-
roach, 341.

Oxygen consumption of Teredo larvae, 323.

Oxygen uptake by echinoderm exogastrulae, 21.

Oxygen uptake of developing grasshopper em-
bryos, effect of ATP on, 265.

382

INDEX

pH, relationship between activity of fly head
cholinesterase and, 139.

Paramecium, effect of x-rays on, 253.

Parasites of sea otter in Alaska, 107.

Partial larvae of Peronella, metamorphosis of,
83.

Patiria miniata, redox indicator patterns in
exogastrulae of, 21.

Patterns of reduction in echinoderm exogastru-
lation, 21.

Pegs, sensory, on grasshopper antennae, perme-
ability of, 122.

Penetration of radioactive phosphate into
marine eggs, 279.

Permeability of sensory pegs on grasshopper
antennae, 122.

Peronella japonica, metamorphosis of partial
larvae of, 83.

Persistent diurnal rhythm of chromatophoric
response in eyestalkless Uca, 371.

Phases of endogenous daily rhythm shifted by
light stimuli, 308.

Phormia regina larva, character and ultimate
fate of salivary secretion of, 178.

Phosphate, accumulation of by fertilized sea
urchin eggs, 297.

Phosphate transfer system in grasshopper em-
bryos, 265.

Phosphorus, distribution of in dorsal and ven-
tral halves of frog gastrula, 318.

Phosphorus-32, distribution of in dorsal and
ventral halves of frog gastrula, 318.

Phosphorus-32, penetration of into marine eggs,
279.

Photoreversibility of carbon monoxide inhibi-
tion of metabolism in Cecropia, 238.

Phylogeny of digenetic Trematode larval types,
15.

Physiology of insect diapause, 210, 238.

Phytoplankton blooms in Long Island waters,
ecology of, 198.

Pigment responses in Uca, 371.

Platysamia cecropia, physiology of diapause of,
210, 238.

Plecoptera, intercerebralis-cardiacum-allatum
system of, 1.

Pollution as a factor in phytoplankton blooms,
198.

Pressure, respirometer for metabolic studies
under, 230.

PRINTY, GLENN E. See G. FRAENKEL, 149.

Propulsion, water, tidal rhythmicity of rate of,
in Mytilus, 353.

Protoplasmic viscosity, effect of x-rays on in
Paramecium, 253.

Protozoa, nutritional studies on, 269.

Puparia, role of Phormia larval salivary secre-
tion in cementing of, 178.

O ADIATION resistance and its variability,
253.

Radiophosphorus, accumulation of by fertilized
sea urchin eggs, 297.

Radiophosphorus, distribution of in dorsal and
ventral halves of frog gastrula, 318.

Radiophosphorus, penetration of into marine
eggs, 279.

Rana pipiens, distribution of phosphorus in
gastrula of, 318.

RAO, K. P. Tidal rhythmicity of rate of water
propulsion in Mytilus, and its modifiabil-
ity by transplantation, 353.

RAY, DAVID T., AND P. W. WHITING. An x-ray
dose-action curve for eye-color mutations
in Mormoniella, 100.

Redox indicator patterns in relation to echino-
derm exogastrulation. II, 21.

Relation of metabolism of marine teleost tissues
to activity, 360.

Relationship between pH and activity of fly
head cholinesterase, 139.

Respiratory metabolism of teleost tissues in
relation to activity and size, 360.

Respiration of normal larvae of Teredo barts-
chii, 323.

Respiration of sea urchin intestine, 328.

Respirometer for metabolic studies at high gase-
ous pressures, 230.

Rhythm of chromatophoric response in Uca,
371.

Rhythm of color change shifted by light stimuli,
308.

Rhythmicity, tidal, and rate of water propul-
sion in Mytilus, 353.

RYTHER, JOHN H. The ecology of phytoplank-
ton blooms in Moriches Bay and Great
South Bay, Long Island, New York, 198.

OALIVARY secretion of Phormia larva, char-
acter and ultimate fate of, 178.

Salinity requirements of phytoplankton in Long
Island waters, 198.

Sand dollar, metamorphosis of partial larvae of,
83.

SCHILLER, EVERETT L. Studies on the hel-
minth fauna of Alaska. XVII. Notes on
the intermediate stages of some helminth
parasites of the sea otter, 107.

SCHLEGEL, VERA. Se e W. H. McSHAN, 341.

SCHNEIDERMAN, HOWARD A., AND NED FEDER.

A respirometer for metabolic studies at
high gaseous pressures, 230.
SCHNEIDERMAN, HOWARD A., AND CARROLL
M. WILLIAMS. The physiology of insect
diapause. VIII. Qualitative changes in the
metabolism of the Cecropia silkworm dur-
ing diapause and development, 210.

INDEX

383

SCHNEIDERMAN, HOWARD A., AND CARROLL M.

WILLIAMS. The physiology of insect dia-
pause. IX. The cytochrome oxidase sys-
tem in relation to the diapause and devel-
opment of the Cecropia silkworm, 238.

Sea otter, parasites of, 107.

Sea urchin, nutrition of, 328.

Sea urchin egg, interaction of sperm and jelly-
coat in fertilization of, 169.

Sea urchin eggs, accumulation of phosphate by,
297.

Sensory pegs on grasshopper antennae, perme-
ability of, 122.

Silkworm, physiology of diapause in, 210, 238.

Size, relation of to metabolism of marine teleost
tissues, 360.

Size distribution of California Littorina, 185.

SLIFER, ELEANOR H. The permeability of the
sensory pegs on the antennae of the grass-
hopper, 122.

Snails, marine intertidal, ecology of, 185.

Sperm and egg jelly-coat interaction during
fertilization of sea urchin, 169.

Spisula solidissima, localization of anticoagu-
lant substances in tissues of, 129.

Stichococcus, ecology of blooms of, 198.

Strongylocentrotus eggs, accumulation of phos-
phates in, 297.

Strongylocentrotus, nutrition of, 328.

Strongylocentrotus, penetration of P-32 into
eggs of, 279.

Strongylocentrotus purpuratus, redox indicator
patterns in exogastrulae of, 21.

Studies on helminth fauna of Alaska, 107.

Study of mechanism involved in shifting of
phases of endogenous daily rhythm by light
stimuli, 308.

Succinoxidase in woodroach thoracic muscle,
341.

Survival curves for x-irradiated Paramecium,
253.

'"PELEOSTS, marine, respiratory metabolism

of tissues of, 360.
Temperature of incubation, effect of on growth

and development of chick embryo, 48.
Temperature requirements of phytoplankton in

Long Island waters, 198.
Teredo bartschi, respiration of normal larvae

of, 323.
Tetramitus rostratus, nutritional studies on,

269.
THOMAS, LYELL J., JR. The localization of

heparin-like blood anticoagulant sub-
stances in the tissues of Spisula solidissima,

129.

Thoracic ganglion of crab, neurosecretion in, 60.
Thoracic muscles of woodroach, oxidative

enzymes in, 341.
Tidal rhythmicity of water propulsion in

Mytilus, 353.

Tidepools, role of erosive activities of snails in

deepening of, 185.

TIERNEY, J. Q. See CHARLES E. LANE, 323.
Tissue respiration of marine teleosts, 360.
Tissues of Spisula, blood anticoagulant sub-

stances in, 129.

Tosi, L. See A. MONROY, 169.
Totipotency of sand dollar blastomeres, 83.
Transplantation, modifiability by, of Mytilus

water propulsion rate, 353.
Trematode larval types, interpretation of, 15.
Trematodes in Alaskan fauna, 107.
Tribolium confusum, amino acid requirements

of, 149.

I ] CA, shifting of phases of endogenous daily
rhythm of, by light stimuli, 308.

Uca pugilator, persistent diurnal rhythm of
chromatophoric response in, 371.

Urechis, penetration of P-32 into eggs of, 279.

Utilization of carbohydrates in chick embryo
wound healing, 39.

VARIABILITY m radiation resistance, 253.

Vascular system in Nereis, anatomy and beha-

vior of, 69.
Ventral gastrula half of frog embryo, distribu-

tion of phosphorus in, 318.
VERNBERG, F. JOHN. The respiratory metab-

olism of tissues of marine teleosts in rela-

tion to activity and body size, 360.
Viscosity of protoplasm as affected by x-rays,

253.

propulsion in Mytilus, tidal rhyth-

micity of rate of, 353.
Wave action as a factor in ecology of Littorina,

185.
WEBB, H. M., M. F. BENNETT AND F. A.

BROWN, JR. A persistent diurnal rhythm

of chromatophoric response in eyestalkless

Uca pugilator, 371.
WEST, W. L. See J. H. BODINE, 265.
WHITE, WILLIAM E. See EDWARD L. CHAM-

BERS, 297.
WHITING, P. W. See DAVID T. RAY, 100.

WlCHTERMAN, RALPH, AND FRANK H. J. FlGGE.

Lethality and the biological effects of x-

rays in Paramecium : Radiation resistance

and its variability, 253.
WILLIAMS, CARROLL M. See HOWARD A.

SCHNEIDERMAN, 210, 238.
WILSON, WALTER L. See. L. V. HEILBRUNN,

158.
Woodroach, oxidative enzymes in thoracic

muscles of, 341.
Wound healing in chick blastoderms, 39.

V-RA Y dose-action curve for eye-color muta-

tions in Mormoniella, 100.
X-rays, effects of on Paramecium, 253.

Volume 106 Number 1

•o

THE

BIOLOGICAL BULLETIN

PUBLISHED BY

THE MARINE BIOLOGICAL LABORATORY

Editorial Board

W. C. ALLEE, University of Florida A. K. PARPART, Princeton University

L. R. BUNKS, Stanford University BERTA SCHARRER, University of Colorado

K. W. COOPER, University of Rochester ALBERT TYLER, California Institute of Technology

L. V. HETJLBRUNN, University of Pennsylvania JOHN H. WELSH, Harvard University

M. E. KRAHL, University of Chicago DOUGLAS WmTAKER, Stanford University

E. T. MOUL, Rutgers University RALPH WlCHTERMAN, Temple University

DONALD P. COSTELLO, University of North Carolina

Managing Editor

Marine Biological Labora

WOODS HOLE, MASS.

FEBRUARY, 1954

Printed and Issued by

LANCASTER PRESS, Inc.

PRINCE & LEMON STS.

LANCASTER, PA.

A.H.T.CO.
SPECIFICATION

ELECTROPHORESIS APPARATUS

PAPER STRIP MODEL

ELECTROPHORESIS APPARATUS,
Paper Strip Model, A.H.T. Co. Specifi-
cation. With recessed, 9 X J^-inch plati-
num foil electrodes, 10 J^ inch migration
chamber, and separate variable power
supply. For preparing a series of hori-
zontal paper electrophorograms of micro
quantities (8 to 15 lambda for testing
human serum) of mixtures the components
of which are separable by migration upon
application of an electric potential. Offer-
ing the following features :

Compact, lightweight cabinet. The

transparent plastic cabinet is readily
portable, and easily inserted and
operated in refrigerators or incu-
bators.

Safety interlock switch. Raising of
the cover automatically shuts off
current to avoid accidents.

Polarity reversing switch. Changing
direction of current between tests
prolongs life of buffer solution.

Designed specifically for clinical estimation of
protein constituents of human blood serum, but
useful also for the study of other organic or in-
organic systems with similar electromigratory
characteristics.

Sharp separation by migrating components of
normal serum can be completed in approximately
6 hours, using a 250-volt potential, and the entire
procedure — including fixing, staining, drying and
evaluating — can be accomplished in from 7 to
10 hours.

Overall dimensions of cabinet: 16 inches long
X 9^2 inches wide X 6 inches high. Cabinet
includes removable plate glass cover, and phenolic
plastic paper strip carrier which permits use of a
single paper sheet or multiple strips up to a maxi-
mum width of 7f inches.

Direct Current Power Supply is a rectifier-
transformer type unit with voltmeter, milli-
ammeter, precise fine and coarse voltage controls
and polarity reversing switch. Maximum output
without load, 300 volts. With two polarized
receptacles for simultaneous connection, if desired,
to two Cabinets.

More detailed information sent upon request.

4937-W5. Electrophoresis Apparatus,
Paper Strip Model, A.H.T. Co. Specifi-
cation, as above described, consisting of
Cabinet with recessed platinum foil electrodes,
removable plate glass cover, plastic carrier for
paper strips, and Direct Current Power Supply,
300 volts, 50 milliamperes, with two receptacles.
With directions for use. For 115 volts, a.c.
only 375.00

NOTE — An improved Recording Densitometer for use with
above is now under development. Upon request,
information will be sent as soon as available.

SYMBOL OF QUALITY

A.H.T.CO.

| LABORATORY APPARATUS!

ARTHUR H. THOMAS COMPANY

WEST WASHINGTON SQUARE PHILADELPHIA 5, PA.

Teletype Services: Western Union WUX and Bell System PH-72

BIOLOGICAL ABSTRACTS

COVERS THE WORLD'S BIOLOGICAL LITERATURE

How do you keep abreast of the literature in your field? No individual
possibly could accumulate and read all of the biological contributions in the
original — yet some relatively obscure journal might publish a revealing paper
on the very subject in which you are most interested.

Biological Abstracts now publishes concise, informative abridgments of all
the significant contributions from more than 2,500 journals. As well as the
complete edition, it also is published in nine low-priced sectional editions which
are specially designed for individuals who are interested only in one or more
closely related fields.

Production costs have increased to such an extent that the active support
of all biologists is needed to maintain this important service. Write for full
details and a sample copy of the sectional edition covering your field.

BIOLOGICAL ABSTRACTS

UNIVERSITY OF PENNSYLVANIA

PHILADELPHIA 4, PA.

INSTRUCTIONS TO AUTHORS

The Biological Bulletin accepts papers on a variety of subjects of biological interest. 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
be accepted for publication. A paper will usually appear within three months of the date of
its acceptance.

The Editorial Board requests that manuscripts conform to the requirements set below;
those manuscripts which do not conform will be returned to authors for correction before 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, 85 by 11 inches. They
should be carefully proof-read before being submitted and all typographical errors corrected
legibly in black ink. Pages should be numbered. A left-hand margin of at least 1£ inches
should be allowed.

2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and
placed in correct sequence in the text. Because of the high cost of setting such material in type,
authors are earnestly requested to limit tabular material as much as possible. Similarly, foot-
notes to tables should be avoided wherever possible. If they are essential, they should be indi-
cated by asterisks, daggers, etc., rather than by numbers. Foot-notes 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.

3. A condensed title or running 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 set in a
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 7f inches, should be kept in mind in
preparing figures for publication. Illustrations should be large enough so that all details will be
clear after appropriate reduction. Explanatory matter should be included 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.

6. Mailing. Manuscripts should be packed flat; large illustrations may be rolled in a mailing
tube. All illustrations larger than 8£ 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. 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 15 and September 1, and to
Dr. Donald P. Costello, Department of Zoology, University of North Carolina, Chapel Hill,
North Carolina, during the remainder of the year.

Entered as second-class matter May 17, 1930, at the post office at Lancaster, Pa.

under the Act of August 24, 1912.

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

Supply Department

MARINE
BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS

ARVY, L., AND M. GABE Pase
The intercerebralis-cardiacum-allatum system of some Ple-
coptera 1

CABLE, R. M.

A new marine cercaria from the Woods Hole region and its
bearing on the interpretation of larval types in the Fellodis-
tomatidae (Trematoda : Digenea) 15

CHILD, C. M.

Redox indicator patterns in relation to echinoderm exogas-
trulation. II. Reduction patterns 21

FRASER, RONALD C.

The utilization of some carbohydrates by in vitro cultured
chick blastoderms in wound healing 39

HARRISON, JOHN R., AND IRVING KLEIN

Effect of lowered incubation temperature on the growth and
differentiation of the chick embryo 48

MATSUMOTO, KUNIO

Neurosecretion in the thoracic ganglion of the crab, Eriocheir
japonicus 60

NICOLL, PAUL A.

The anatomy and behavior of the vascular systems in Nereis
virens and Nereis limbata 69

OKAZAKI, KAYO, AND KATSUMA DAN

The metamorphosis of partial larvae of Peronella japonica
Mortensen, a sand dollar 83

RAY, DAVID T., AND P. W. WHITING

An x-ray dose-action curve for eye-color mutations in Mor-
moniella 100

SCHILLER, EVERETT L.

Studies on the helminth fauna of Alaska. XVII. Notes on
the intermediate stages of some helminth parasites on the
sea otter 107

SLIFER, ELEANOR H.

The permeability of the sensory pegs on the antenna of the
grasshopper (Orthoptera, Acrididae) 122

THOMAS, LYELL J., JR.

The localization of heparin-like blood anticoagulant sub-
stances in the tissues of Spisula solidissima 129

Volume 106

Number 2

THE

BIOLOGICAL BULLETIN

PUBLISHED BY

THE MARINE BIOLOGICAL LABORATORY

Editorial Board

Marine Biological Laboratory

LIBK. AIR. Y

APR 20 1954

WOODS HOLE, MASS.

W. C. ALLBE, University of Florida

L. R. BLINKS, Stanford University

K. W. COOPER, University of Rochester

L. V. HEILBRUNN, University of Pennsylvania

M. E. KRAHL, University of Chicago

£. T. MOUL, Rutgers University

DONALD P. COSTELLO, University of North Carolina

Managing Editor

A. K. PARPART, Princeton University
BERTA SCHARRBR, University of Colorado
ALBERT TYLER, California Institute of Technology
JOHN H. WELSH, Harvard University
DOUGLAS WHITAKER, Stanford University
RALPH WICHTERMAN, Temple University

APRIL, 1954

Printed and Issued by

LANCASTER PRESS, inc.

PRINCE £ LEMON STS.
LANCASTER, PA.

A.H.T.CO.

SPECIFICATION'

DENSIGRAPH

(RECORDING DENSITOMETER)

A manually operated Recorder
which provides a continuous ink
tracing indicating percentage
light transmission

• For convenient and rapid
photometric analysis of
light absorbing materials
on dry paper strips ob-
tained in electropho-
resis and chromatog-
raphy

4937-X.

DENSIGRAPH (Recording Densitometer),
A. H. T. Co. Specification, (Patent Applied
For), manually operated. For convenient and
rapid photometric analysis of light-absorbing ma-
terials by determining light transmission and
lineal separation of stained areas on dry strips of
paper as obtained in electrophoresis and chroma-
tography.

Simultaneously combines the usual operations
of scanning, indicating and recording to produce a
continuous inked tracing of the output of a photo-
voltaic cell on graph paper marked in millimeter
squares to indicate percentage light transmission.
Takes paper strips up to 40 mm wide and treat-
ment of the paper to make it translucent is not
required.

Consisting of a modified microammeter with
extra manually controlled pointer, photocell, 6-volt
lamp, adjustable slit, constant voltage transformer
to operate the lamp, and a pen which traces a
curve when the pointer on the microammeter is
followed closely by the manually controlled pointer
which is operated from the front of the cabinet by
means of a mechanically linked lever.

In use, stained paper strips are attached by ad-
hesive tape to the right edge of the graph paper,
below the adjustable slit, and advanced beneath
the photocell housing by the hand wheel at the
left of the cabinet. Lateral movement of the lever
with the right hand makes it possible to align the
manually controlled pointer continuously with
the indicating pointer of the meter and, as the rate
of travel of the graph paper is under the control
of the operator's left hand, the fidelity of the result-
ing curve depends upon the manipulative skill of the
operator. A continuous record of an electrophoro-
gram 6 inches long can be completed in approxi-
mately 5 minutes and portions of the curve can be
rechecked by simple roll-back of the graph paper.

4937-X. Denslgranh (Recording Densitometer),
A. H. T. Co. Specification, (Patent Applied For), as

above described, complete with constant voltage trans-
former, 50 ft. roll of graph paper, ink writing pen, 4 oz.
bottle of ink, roll of adhesive tape with dispenser, carrier
for 2 x 2-inch glass filters, 6 ft. three-wire cord with two-
prong plug with grounding lug, and directions for use.
Power consumption 460 watts. For use on 115 volts, 60
cycles, a.c. only 475.00

SYMBOL OF QUALITY

A.H.T.CO.

''SPEC.I F I C AT l6N •

LABORATORY APPARATUS |

Copy of Bulletin 117 sent upon request.

ARTHUR H. THOMAS COMPANY

WEST WASHINGTON SQUARE PHILADELPHIA 5. PA.

Teletype Services: Western Union WUX and Bell System PH-72

BIOLOGICAL ABSTRACTS

COVERS THE WORLD'S BIOLOGICAL LITERATURE

How do you keep abreast of the literature in your field? No individual
possibly could accumulate and read all of the biological contributions in the
original — yet some relatively obscure journal might publish a revealing paper
on the very subject in which you are most interested.

Biological Abstracts now publishes concise, informative abridgments of all
the significant contributions from more than 2,500 journals. As well as the
complete edition, it also is published in nine low-priced sectional editions which
are specially designed for individuals who are interested only in one or more
closely related fields.

Production costs have increased to such an extent that the active support
of all biologists is needed to maintain this important service. Write for full
details and a sample copy of the sectional edition covering your field.

BIOLOGICAL ABSTRACTS

UNIVERSITY OF PENNSYLVANIA

PHILADELPHIA 4, PA.

INSTRUCTIONS TO AUTHORS

The Biological Bulletin accepts papers on a variety of subjects of biological interest. 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
be accepted for publication. A paper will usually appear within three months of the date of
its acceptance.

The Editorial Board requests that manuscripts conform to the requirements set below;
those manuscripts which do not conform will be returned to authors for correction before 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, 85 by 11 inches. They
should be carefully proof-read before being submitted and all typographical errors corrected
legibly in black ink. Pages should be numbered. A left-hand margin of at least 1? inches
should be allowed.

2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and
placed in correct sequence in the text. Because of the high cost of setting such material in type,
authors are earnestly requested to limit tabular material as much as possible. Similarly, foot-
notes to tables should be avoided wherever possible. If they are essential, they should be indi-
cated by asterisks, daggers, etc., rather than by numbers. Foot-notes 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.

3. A condensed title or running 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 set in a
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 7f inches, should be kept in mind in
preparing figures for publication. Illustrations should be large enough so that all details will be
clear after appropriate reduction. Explanatory matter should be included 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.

6. Mailing. Manuscripts should be packed flat; large illustrations may be rolled in a mailing
tube. All illustrations larger than 8| 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. 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 15 and September 1, and to
Dr. Donald P. Costello, Department of Zoology, University of North Carolina, Chapel Hill,
North Carolina, during the remainder of the year.

Entered as second-class matter May 17, 1930, at the post office at Lancaster, Pa.

under the Act of August 24. 1912.

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

Supply Department

MARINE
BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS

Page
CHADWICK, L. E., J. B. LOVELL AND V. E. EGNER

The relationship between pH and the activity of cholin-
esterase from flies 139

FRAENKEL, G., AND GLENN E. PRINTY

The amino acid requirements of the confused flour beetle,
Tribolium confusum, Duval 149

HEILBRUNN, L. V., ALFRED B. CHAET, ARNOLD DUNN AND
WALTER L. WILSON
Antimitotic substances from ovaries 158

MONROY, A., L. TOSI, G. GlARDINA AND R. MAGGIO

Further investigations on the interaction between sperm and
jelly-coat in the fertilization of the sea urchin egg 169

MOOREFIELD, HERBERT H., AND G. FRAENKEL

The character and ultimate fate of the larval salivary secre-
tion of Phormia regina Meig. (Diptera, Calliphoridae) 178

NORTH, WHEELER J.

Size distribution, erosive activities, and gross metabolic
efficiency of the marine intertidal snails, Littorina planaxis
and L. scutulata 185

RYTHER, JOHN H.

The ecology of phytoplankton blooms in Moriches Bay and
Great South Bay, Long Island, New York 198

SCHNEIDERMAN, HOWARD A., and CARROLL M. WILLIAMS

The physiology of insect diapause. VIII. Qualitative changes
in the metabolism of the Cecropia silkworm during diapause
and development 210

SCHNEIDERMAN, HOWARD A., AND NED FEDER

A respirometer for metabolic studies at high gaseous pres-
sures 230

SCHNEIDERMAN, HOWARD A., AND CARROLL M. WILLIAMS

The physiology of insect diapause. IX. The cytochrome oxi-
dase system in relation to the diapause and development of
the Cecropia silkworm 238

WICHTERMAN, RALPH, AND FRANK H. J. FlGGE

Lethality and the biological effects of x-rays in Paramecium :
Radiation resistance and its variability 253

Volume 106 Number 3

f

THE

BIOLOGICAL BULLETIN

PUBLISHED BY

THE MARINE BIOLOGICAL LABO^RpgfflogJcal Laboratory

Editorial Board

JUN8- 1B54

WOODS HOLE, MASS.

W. C. ALLEE, University of Florida A. K. PARPART, Princeton University

L. R. BLINKS, Stanford University BERTA SCHARRER, University of Colorado

K. W. COOPER, University of Rochester ALBERT TYLER, California Institute of Technology

L. V. HEILBRUNN, University of Pennsylvania JOHN H. WELSH, Harvard University

M. E. KRAHL, University of Chicago DOUGLAS WmxAKER, Stanford University

E. T. MOUL, Rutgers University RALPH WlCHTERMAN, Temple University

DONALD P. COSTELLO, University of North Carolina

Managing Editor

JUNE, 1954

Printed and Issued by

LANCASTER PRESS, Inc.

PRINCE 8C LEMON STS.

LANCASTER, PA.

A.H.T.CO.
SPECIFICATION

MILLER, SMALL MODEL

KYMOGRAPH

SPRING MOTOR DRIVEN

B

• Drums are 10 inches high X 6 inches diameter

• Speed is adjustable from 0.32 mm per second
to 40 mm per second

• Will operate at slow speed for 10 hours and at
fast speed for six minutes without rewinding

S014-A.

8015-A.

KYMOGRAPH, MILLER, Small Model. Drums are 10 inches
high, made of brass tubing with cast aluminum ends, supported on polished
steel shafts, J^-inch diameter, which revolve in ball bearing races fitted
into heavy cast bases, and can be fastened in position on the shaft instantly
by means of a simple friction clamp located on top.

Opening A in the base is intended for insertion of 8015-A Apparatus
Support, Simple Form, on which to mount writing levers, etc. Opening B
in the base is threaded to take one of the 2J^-inch posts which support the
spacer rod of the Long Paper Attachment.

With belt attachment between the drum and spring mechanism. Will
operate at a slow speed for 10 hours and at fast speed for 6 minutes without
rewinding. The speed rotation at the drum surface is adjustable from
approximately 0.32 mm per second to approximately 40 mm per second
by means of two grooved aluminum pulleys and a series of five sizes of
interchangeable aluminum governing vanes. Starting and stopping are ac-
complished by means of a simple friction clutch operated by a lever on the
top face of the housing.

8014-A. Kymograph, Miller, Small Model, Spring Motor Drive, as above described.
With brass drum, set of five aluminum governing vanes from Yz to 6 inches long, and belt
for attaching drum to spring mechanism 124.75

8015-A. Apparatus Support, Simple Form, for mount-
ing writing levers, with fixed vertical rod and holder for
attachment to kymograph base, and with horizontal arm
for suspending a plummet. For attachment to 8014-A in
opening A 9.25

8015-L. Long Paper Attachment, with brass idler drum,
10 inches high X 23^ inches diameter, on heavy triangular
base with leveling screw, two attachment posts, Stainless
steel spacer rod, 36 inches long X J^-inch diameter, and
two Connectors for attaching spacer rod to the support and
to the spring-wound Kymograph 17.00

SYMBOL OF QUALITY

A.H.T.CO.

| LABORATORY APPARATUS

ARTHUR H. THOMAS COMPANY

WEST WASHINGTON SQUARE PHILADELPHIA 5, PA.

Teletype Services: Western Union WUX and Bell System PH-72

with

j

Converging Eyetubes

your EYES
do LESS WORK

AO also makes a

binocular body with

parallel eyetubes for

those who prefer them.

When using a binocular microscope your eyes are constantly shifting
from eyepieces to worktable and back again. If the microscope has
parallel eyetubes your eyes must work to readjust themselves each time
— now converging on the table at an angle of approximately 8°, now
pointing parallel into the eyetubes in order to fuse the two images.
To many people this is uncomfortable, tiring, and slows down the
— so AO optical scientists originated the familiar binocular body
with converging eyetubes which is standard on AO Microscopes.

This example of AO Design Perfection is also apparent in the re-
nowned AO Spencer Optics, dust-proof, dual-cone nosepiece, "pinch-
grip" mechanical stage, "autofocus", custom tension adjustment, and

many other features. Test
the many AO advantages
yourself. Ask your AO dis-
tributor for a demonstra-
tion, or write Dept. F185.

./imerican Uptical

INSTRUMENT DIVISION

BUFFALO 15. NF.W. YORK

INSTRUCTIONS TO AUTHORS

The Biological Bulletin accepts papers on a variety of subjects of biological interest. 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
be accepted for publication. A paper will usually appear within three months of the date of
its acceptance.

The Editorial Board requests that manuscripts conform to the requirements set below;
those manuscripts which do not conform will be returned to authors for correction before 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, 8^ by 11 inches. They
should be carefully proof-read before being submitted and all typographical errors corrected
legibly in black ink. Pages should be numbered. A left-hand margin of at least 1£ inches
should be allowed.

2. Tables, Foot-Notes, Figure Legends, etc. Tables should be typed on separate sheets and
placed in correct sequence in the text. Because of the high cost of setting such material in type,
authors are earnestly requested to limit tabular material as much as possible. Similarly, foot-
notes to tables should be avoided wherever possible. If they are essential, they should be indi-
cated by asterisks, daggers, etc., rather than by numbers. Foot-notes 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.

3. A condensed title or running 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 set in a
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 7f inches, should be kept in mind in
preparing figures for publication. Illustrations should be large enough so that all details will be
clear after appropriate reduction. Explanatory matter should be included 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.

6. Mailing. Manuscripts should be packed flat; large illustrations may be rolled in a mailing
tube. All illustrations larger than 8£ 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. 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 15 and September 1, and to
Dr. Donald P. Costello, Department of Zoology, University of North Carolina, Chapel Hill,
North Carolina, during the remainder of the year.

Entered as second-class matter May 17, 1930, at the post office at Lancaster, Pa.

under the Act of August 24, 1912.

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

Supply Department

MARINE

BIOLOGICAL LABORATORY

Woods Hole, Massachusetts

CONTENTS

Page

BODINE, JOSEPH HALL, AND WILLIAM LIONEL WEST

Effect of adenosinetriphosphate (ATP) on the endogenous
oxygen uptake of developing grasshopper embryos 265

BRENT, MORGAN M.

Nutritional studies on the amoebo-flagellate, Tetramitus
rostratus 269

BROOKS, SUMNER C., AND EDWARD L. CHAMBERS

The penetration of radioactive phosphate into marine eggs . . 279

CHAMBERS, EDWARD L., AND WILLIAM E. WHITE

The accumulation of phosphate by fertilized sea urchin eggs . 297

BROWN, FRANK A., JR., MILTON FINGERMAN AND MARGARET
N. HINES

A study of the mechanism involved in shifting of the phases of
the endogenous daily rhythm by light stimuli 308

GRANT, PHILIP

The distribution of phosphorus (P31 and P32) in dorsal and
ventral halves of the Rana pipiens gastrula 318

LANE, CHARLES E., J. Q. TIERNEY AND R. E. HENNACY

The respiration of normal larvae of Teredo bartschi Clapp . . 323

LASKER, REUBEN, AND ARTHUR C. GIESE

Nutrition of the sea urchin, Strongylocentrotus purpuratus . . 328

MCSHAN, W. H., SOL KRAMER AND VERA SCHLEGEL

Oxidative enzymes in the thoracic muscles of the woodroach,
Leucophaea maderae 341

RAO, K. P.

Tidal rhythmicity of rate of water propulsion in Mytilus and
its modifiability by transplantation 353

VERNBERG, F. JOHN

The respiratory metabolism of tissues of marine teleosts in
relation to activity and body size 360

WEBB, H. MARGUERITE, MIRIAM F. BENNETT AND FRANK A.
BROWN, JR.

A persistent diurnal rhythm of chromatophoric response in
eyestalkless Uca pugilator 371

M

LIBRARY

UH 1AZB K

i

•