Color changes of animals in relation to nervous activity

=. r-q

jg -; CD

Jo

UNIVERSITY OF PENNSYLVANIA
Leidy Memorial Lectures

COLOR CHANGES IN ANIMALS
IN RELATION TO NERVOUS ACTIVITY

^

7^;

COLOR CHANGES OF ANIMALS ^
IN RELATION TO C'^

NERVOUS ACTIVITY

By
G. H. Parker

Professor of Zoology, Emeritus
Harvard University

UNIVERSITY OF PENNSYLVANIA PRESS

Philadelphia
1936

London: Humphrey Miljord: Oxford University Press

Copyright 1936
UNIVERSITY OF PENNSYLVANIA PRESS

Manufactured in the United States of America
by the Lancaster Press, Inc., Lancaster, Pa.

FOREWORD

Dr. Joseph Leidy was the first distinguished naturalist
with whom I became acquainted. As a Jessup Student
at the Academy of Natural Sciences in Philadelphia be-
tween the years 1880 and 1882 I was privileged to come
under his direct supervision. At that time Dr. Leidy
was Chairman of the Curators of the Academy and he
very generously took upon himself a kindly oversight of
the work of the Jessup beneficiaries. We spent half our
day working upon certain assigned collections in the
Academy and the other half upon the study of any sub-
ject that interested us. It was in these personal studies
that Dr. Leidy was most helpful to us. His habit was
to come about once a week to the Academy where he
would spend part of the day either in the library or in
his small private study. At such times he was always
open to approach and we were free to bring to him any
real difficulties that we had met in our work. To these
he gave kindly consideration, and after such brief inter-
views we always left with renewed inspiration and en-
couragement. He knew us well enough to call us by
our first names, a circumstance that put us in the very
appropriate relation of apprentices to master.

As I look back on these brief contacts with Dr. Leidy
I am surprised at what I unconsciously absorbed from
them. I once sat on the outskirts of a group of young
schoolchildren who had been invited by him to the Acad-
emy for a brief afternoon talk. He spoke to them on
the human skull, a subject that at first sight might seem
far from attractive to such youngsters, but before he
had finished you could see the keen natural interest that

vi COLOR CHANGES IN ANIMALS

he had awakened in them for even so dry a topic. With
his pencil he pointed out the chief apertures in the bony
brain-case and told the children of their uses. When
he came to the largest opening he called it by its tech-
nical name, the foramen magnum. He then remarked
that this sounded very learned but he would warn them
not to be overawed by such high-flown names and to
remember that to an old Roman the Latin words merely
meant a big hole. This simple experience in the use of
scientific terms was a revelation to me and gave me a
respect for the common English equivalents which I
have never lost. A lesson of this kind meant more to
me in my subsequent life as a teacher of zoology than
pages of pedagogics.

Would that in the delivery of this discourse I could
reawaken in you the childlike, lifelong enthusiasm that
Dr. Leidy had for the study of Nature in all its fasci-
nating aspects, and would that I could stir in you the
generous impulses that in a second Dr. Leidy made these
lectures possible.

G. H. Parker

Harvard Biological Laboratories
March, 1936

PREFACE

The present volume is a somewhat extended form of the
Joseph Leidy Memorial Lecture in Science delivered at
the University of Pennsylvania March 3, 1936. In ad-
dition to the historical summaries it consists in large
part of the recent studies by my students and myself
on the means of activating color-cells in the higher ani-
mals and on the significance of these processes for the
workings of the nervous system. It is believed that the
idea of the neurohumors, set forth in numerous earlier
publications and rather fully elucidated in this volume,
has a measure of truth in it for general nervous func-
tions, and it is one of the objects of this essay to point
out some of the reasons for accepting this idea and for
testing its further applications. The whole proposal is
quite obviously in a formative stage and, as every inves-
tigator knows, its outcome must await further study.

The invitation to deliver the Leidy lecture came to
me from a committee consisting of Dr. Josiah H. Penni-
man, Provost of the University of Pennsylvania, Dr.
Eliot Clark, Dr. Milton Greenman, and Dr. C. E.
McClung. To these gentlemen I wish to express my
keen appreciation of the honor of this invitation and the
great pleasure I take in accepting it. A certain personal
gratification that I feel in this acceptance I have at-
tempted to indicate in the Foreword.

I cannot conclude this brief preface without acknowl-
edging with sincere thanks the aid in preparing the
manuscript for this volume received from my wife,
Louise Merritt Parker, and from my assistant, Helen
Porter Brower. I am also greatly indebted to Dr. F. M.

viii COLOR CHANGES IN ANIMALS

Carpenter of the Museum of Comparative Zoology for
his care in the preparation of the illustrations as well
as to the editors of the following publications for the
privilege of reproducing the figures accredited to these
sources: the Journal of Experimental Zoology , the Bio-
logical Bulletin, the Journal of Experimental Biology, the
Journal of General Physiology, the Proceedings of the
National Academy of Sciences, and the Proceedings of
the American Philosophical Society.

The generous provisions for publication made by the
Committee on the Leidy Memorial Lecture are very
fully appreciated.

CONTENTS

Chapter
Foreword

Page

V

Preface

vii

I. Introduction

i

II. The Dogfish

12

III. The Killifish

11

IV. Neurohumors

40

V. The Nervous

System

and Chromatophores

58

References

66

I

INTRODUCTION

The subject of color changes in animals was a familiar
one to the ancients. Aristotle, in the second book of
his History of Animals, declared that the chameleon, an
inhabitant of the North Coast of Africa, can acquire
either a black color, like that of the crocodile, or an
ocherous one, like that of the lizard, or can be spotted
with black like the panther. These changes, according
to the Stagirite, take place over the whole body of this
animal, for the eyes change like the rest and so does the
tail. This description, together with certain other de-
tails recorded by x^ristotle for this remarkable animal,
was repeated almost verbatim by Pliny in the eighth
book of his Natural History, to which he added the pop-
ular fiction that the chameleon feeds upon air. Pliny
also recorded the color changes of the mullet, a Medi-
terranean fish much sought after as a delicacy for Ro-
man feasts. In the ninth book of his Natural History
he wrote that the masters of gastronomy inform us that
the mullet while dying assumes a variety of colors and
a succession of shades, and that the hue of its red scales
growing paler and paler, gradually changes more espe-
cially if the fish is looked at enclosed in glass. Thus
knowledge of these remarkable color changes was not
only a part of ancient lore but was passed on to pos-
terity. In Henry the Sixth Shakespeare put into the
mouth of the infamous Duke of Gloucester the boast
" I can add colors to the chameleon "; and no less a
personage than Hamlet, when asked by his uncle-king
how fares his health, replied " Excellent, i' faith; of the
chameleon's dish: I eat the air." Thus did the Bard

l

2 COLOR CHANGES IN ANIMALS

of Avon play his part in transmitting truth and fancy
about this interesting creature. It must be clear from
these few allusions that the subject of this lecture has
both the venerableness of age and the dignity of poetic
association.

From very early times till about the beginning of the
present century color changes in animals were believed
by the majority of workers to be under the exclusive
control of the nervous system. This opinion, often only
vaguely and generally expressed, steadily gained ground
in consequence of the accumulation of a large body of
favorable evidence drawn in part from purely observa-
tional work and in part from experimental investiga-
tions. One outcome of these inquiries was to show that
the eyes of animals are essential to their color changes,
for when these organs were removed or effectually cov-
ered all signs of such changes disappeared, and the given
animal so far as an alteration of its tint was concerned
was largely incapacitated. In normal animals the color
changes had long been recognized as means of harmon-
izing the creature with its surroundings. Even Aris-
totle in describing the habits of the common octopus
remarked that this cephalopod would pursue any fish
that came in its way, changing its color so as to imitate
that of the neighboring rocks. This it also did when
alarmed. In 1830 Stark, who had studied the color
changes in a number of British river-fishes such as the
perch, minnow, and the like, observed that when these
fishes were on a light background they were pale in tone
and when they were on a dark one they were of a deeper
shade. He advanced the idea that this agreement, by
which the fish was lost, so to speak, in its own back-
ground, was an advantage to it in its escape from ene-
mies. This and other instances of a like kind gave rise
to the modern theory of protective coloration, a system

INTRODUCTION 3

of animal camouflage illustrated in the colors and forms
of a great variety of organisms. From this standpoint
the importance of the eyes in color responses became at
once apparent, for an animal evidently must see that
which it tends to resemble before it can assume the like-
ness. Thus color reactions became incorporated among
the reflex activities of animals, and a wide and novel
field for investigation was thrown open.

In 1858 the celebrated British physician Joseph Lister,
then a student of medicine some thirty years of age,
published in the Transactions of the Royal Society of
London a scholarly paper on the color changes of the
common frog. Here he summarized up to his own time
the important general conclusions in this field of re-
search. According to him the eyes of any animal that
possessed the property of changing its tint were the only
channels, to use his expression, through which the rays
of light could gain access to the nervous system so as
to induce changes of color in the skin. He declared
further that the cerebro-spinal axis was chiefly, if not
exclusively, concerned in regulating the functions of the
pigment-cells. These brief statements give the physio-
logical foundations upon which has been based the ex-
perimental work on animal coloration during the last
half of the nineteenth century.

Incidentally these early investigations afforded a gen-
eral survey of the animal kingdom so far as color changes
were concerned. As an outcome of such an inspection
it was found that these changes are limited in the main
to comparatively few representatives of five important
groups of the higher animals. These are the cephalo-
pods such as the octopus, the cuttlefish, and the squid;
the crustaceans, especially the shrimps and prawns; and,
among the vertebrates, the fishes, the amphibians, and
the lizards. The highest vertebrates, the birds and

4 COLOR CHANGES IN ANIMALS

mammals, with their coverings of feathers and of hairs
almost entirely lack this capacity. Among these forms
man, so far as I am aware, is the only species which in
a feeble way keeps up this type of reaction. Our facial
blush is dependent upon a temporary enlargement of the
small blood-vessels of the skin and corresponds in all
essential respects with the reddening of the integument
seen in such fishes as the top-minnows. But even this
mild activity, once such a powerful weapon in the hands
of the female of the species, will, I fear, soon find its
place among the lost arts, for the modern generation
seems to have given up a reaction-pattern that was at
once the charm and delight of an earlier day.

Another all-important step taken by these older work-
ers was the discovery of the means by which color
changes were brought about. Over a century ago it
was found that those animals that show changes in tint
possess in their integuments a multitude of minute
bodies which by what appear to be contractions and
expansions are able to lighten, darken, or otherwise alter
the color of their possessors. These bodies were studied
in the cephalopods in 1819 by the Italian naturalist
Sangiovanni who called them cromofori, or in English
chromatophores. It is now known that chromatophores
are single integumentary cells or groups of such cells
containing pigment which by one means or another may
be concentrated and thus rendered inconspicuous, or
may be spread out and thus become exposed to view.
In the cephalopods, such as the octopus and the squid,
each chromatophore consists of a central elastic-walled
sac filled with pigment, around which is a system of
radiating muscle-fibers (Fig. 1). By means of these
fibers the spherical sac may be drawn out to a flattened
disc, thus spreading its pigment conspicuously, or it may
be allowed to contract to a minute sphere almost in-

INTRODUCTION

Fig. i. Chromatophores of the squid Loligo: a, contracted;
l>, expanded. Bozler, Zeit. vergl. Physiol., 1928, 7, 381, fig. 1.

visible. In the crustaceans, especially the shrimps, the
chromatophores (Fig. 2) are usually groups of cells which
may carry each cell for itself a distinctive color. By
internal migration these colors
become variously dispersed or
concentrated, thus adding or
subtracting their share in the
general color tone of the whole
animal. Finally in the verte-
brates the great majority of
chromatophores (Fig. 3) are single
cells of which there are several
classes: melanophores, containing
dark pigment granules; xantho-
phores and erythrophores carry-
ing yellow, orange, or red caro-
tenoid pigments; and finally,

Fig.
phore

2. Chromato-
with fully dis-
persed pigment from the
shrimp Palaemonetes .
Perkins, Jour. Exp.
Zoo!., 1928,50, 101, pi. 1 .

Inc. 3. Group of chromatophores each with fully dispersed
pigment from the tail of the killifish.

6 COLOR CHANGES IN ANIMALS

though this may not conclude the list, leucophores and
iridocytes with their semi-crystalline or crystalline gua-
nin-like contents. These various types of chromato-
phores, partly through their own innate color exposures
and withdrawals, partly through their effects in covering
and uncovering other colored cells, combined also with
such physical light changes as are induced by the thin
transparent outer layers of the skin, have united to pro-
duce that marvelous play of animal colors which once
fed the eyes of the ancient Roman gourmands and which
now drives the modern biologist almost to despair.

In 1852 Briicke published his important monograph
on the color changes of the African chameleon. In this
work he pointed out that when cutaneous nerves were
cut the denervated area of the skin made itself manifest
by darkening, that is, the dark pigment in the melano-
phores of this area became dispersed. Briicke expressed
the natural opinion that nerves severed in this way had
suffered paralysis and that the melanophores with which
these nerves were connected, having been released from
nervous control, lapsed into an inactive state. He
therefore regarded the stage of a melanophore with dis-
persed pigment as the relaxed or resting one as con-
trasted with that of concentrated pigment which he
believed to be the fully active stage. In this way he
brought chromatophores into line with ordinary muscle
fibers. This was a generalization of no small signifi-
cance and has been accepted by most later workers.
We shall see, however, that it may be open to question.

In the early seventies of the last century the French
physiologist Pouchet (1872, 1876) carried out experi-
ments upon fishes similar to those that Briicke had per-
formed on chameleons. Pouchet cut integumentary
nerves and noted, for instance in turbots (Fig. 4), that
the denervated areas darkened as they had done in the

INTRODUCTION

chameleon. Pouchet, however, showed still further that
if the spinal cord of a fish is cut, no such integumentary
darkening follows. This response took place only when
the sympathetic chains situated one on either side of the
vertebral column were severed.
He therefore declared that the
chromatophoral system was not
only under the control of nerves
but that these nerves were sym-
pathetic in origin. This con-
clusion has been abundantly
confirmed by a large number
of investigators, among whom
the chief is von Frisch (1910,
1911, 1912^, 1912^) whose
studies on the color changes in
fishes were published about the
beginning of the second decade
of this century and were a bril-
liant continuation of the mas-
terly work of his predecessors.

But even before this time,
as von Frisch himself recoo--

o

nized, a new current of ideas
had set in. This resulted from
a series of incidental obser-
vations the significance of which

was not at first fully appreciated. In 1898 Corona
and Moroni showed that when adrenalin, the secre-
tion of certain cells in the medulla of the adrenal
gland, was introduced into the circulation of a frog the
pigment in its melanophores became strongly concen-
trated. This unique observation was confirmed by
Lieben in 1906 who made an extended investigation of
the subject. Comments by Fuchs in 19 14 on these two

Fig.

Turbc

which particular nerves
have been cut wherebv
the melanophores of the
denervated regions have
been induced to disperse
their pigment thus ren-
dering the fishes dark in
those regions. Pouchet,
Jour. Anat. Physiol. ,
1876, 12, pi. 4.

8 COLOR CHANGES IN ANIMALS

pieces of work led Redfield in 191 S to investigate the
effects of this hormone on the chromatophores of the
lizard Phrynosoma with the result that adrenalin was
found also in this instance to be a potent agent in con-
centrating chromatophoral pigment. Redfield, after an
exhaustive study of the color changes in Phrynosoma,
expressed the opinion that the melanophores in this
lizard were under the control of two types of agents,
nervous and hormonal, and that both these agents in
this particular instance were concerned with the concen-
tration of the melanophore pigment, that is, with the
blanching of the animal. Here then was evidence of a
novel form of chromatophoral control, one in which hor-
mones or, as these particular hormones are now called,
neurohumors, are concerned.

In amphibians and crustaceans the process of nerve
cutting as carried out by the older investigators had
never yielded, even in the hands of the most skilful,
conclusive and satisfactory results such as had been ob-
tained from fishes and reptiles. It is therefore not sur-
prising that skeptical investigators of this subject should
turn their attention to amphibians and crustaceans with
the view of ascertaining what can be learned from them
as to the control of chromatophores. As Hogben (1924)
remarks in his volume on the Pigmentary Effector Sys-
tem, this line of attack was especially suggested by the
researches of Adler (1914), P. E. Smith (1916), and Allen
(191 7), who developed the technique of hypophysec-
tomy in anuran larvae and called attention to the ex-
treme pallor which comes over these young animals after
the removal of the germs of their pituitary glands.
These workers, however, did not appreciate clearly the
full significance of such pigmentary changes; this was
first pointed out by Atwell (191 9). In 1921 Swingle
noticed that tadpoles darkened when the intermediate

INTRODUCTION 9

part of the pituitary gland was implanted in them. The
blanching of the common frog after the removal of its
pituitary gland was incidentally recorded by Krogh in
1922. Meanwhile Hogben and Winton had been ac-
tively engaged in experiments on the color changes of
frogs in relation to nerves and the pituitary secretion.
Their papers, which appeared in 1922 and 1923 and were
based chiefly on a study of the European frogs, led to
rather startling conclusions. In these it was pointed
out that nerves played a wholly insignificant part in the
control of color changes in amphibians, if, in fact, they
had any part at all. The dark phase of the frog was
shown to depend upon a secretion derived from the
pituitary gland, probably from its intermediate part.
This secretion was carried from the gland by the blood
to the melanophores, whose pigment was thereby in-
duced to disperse. The pale phase of the animal was
not so consistently worked out. In the early steps in
his work Hogben appears to have entertained the view
that this single neurohumor, the intermediate-pituitary
secretion, was all that was involved in the frog's change
of color. This secretion when present in the fluids about
the melanophores called forth the dark reaction, and its
absence from these fluids allowed the pale phase to inter-
vene. Subsequently Hogben and his associates worked
upon other species of frogs and particularly upon the
South African toad, Xenopus. As a result of these later
investigations he and Slome became persuaded that at
least in the amphibians just named there was evidence
for a second humor which, though derived from the
pituitary complex, nevertheless induced a concentration
of melanophore pigment (Slome and Hogben, 1928,
1929; Hogben and Slome, 1931). This humor, though
a product of the pituitary gland, was thus the counter-
part of the first one. These investigations were pub-

10 COLOR CHANGES IN ANIMALS

lished a number of years after Hogben's earlier work,
and though they add somewhat to the complexity of the
picture of chromatophoral control in amphibians, they
leave that picture essentially as it was originally out-
lined by Hogben himself: nerves relegated to the back-
ground, perhaps entirely excluded, and hormones the
all-important factors. The contrast between this new
way of conceiving the adjustment of amphibian chro-
matophores to environmental changes and that envis-
aged by the older workers is enormous. As was pointed
out by Hogben in 1924, this view of color-cell activation
sets off the amphibians in strong contrast with the fishes
and the lizards, in both of which there appeared to be
ample and complete evidence for the nervous control of
their color-cells.

Quite independent of the work on amphibians, but
emerging eventually in much the same way, is that done
by recent investigators on crustaceans. Like the frogs
and toads, shrimps and other crustaceans had never
yielded in nerve-cutting experiments evidence favorable
for a nervous interpretation of the control of their chro-
matophores. In 1925 Koller noticed that when the
blood of a dark-tinted Crangon, a common Atlantic
shrimp, was drawn and injected into a pale one the
latter quickly grew dark. This at once suggested that
in these animals neurohumors may be in the blood and
may be the means of controlling the color-cells. This
idea was followed up by Perkins who in 1928 published
an account of the color changes in another /Atlantic
shrimp, Palaemonetes. Perkins was unable to repeat
with success in this form the experiment on the trans-
ference of blood carried out by Koller on Crangon, but
he nevertheless sought in the body of Palaemonetes for
an organ that might produce a humor controlling the
color-cells. This he finally found in the eye-stalks of

INTRODUCTION 1 1

the shrimp. When a number of the eye-stalks of pale
Palaemonetes were crushed and extracted with sea-
water a solution was obtained which on being injected
with proper precautions into a dark shrimp of the same
species caused the pigment in its chromatophores to
concentrate and the shrimp to blanch. Perkins showed
further that the cutting of nerves in Palaemonetes had
no effect whatever upon its color changes, but that a
temporary obstruction to the flow of blood in certain
of its blood-vessels was followed by a very profound
color change. These various observations have been
repeated and confirmed by subsequent workers and with
such a wealth of detail that we are now fully justified
in concluding that among crustaceans, as among am-
phibians, nerves play no direct part in the control of
chromatophores which at least in these groups of ani-
mals are under the exclusive influence of special hor-
mones, the neurohumors. Such secretions are produced
in some distant part of the body under excitation re-
ceived from the eye and transported from their region
of origin by the blood to the responding color-cells.
Thus it appears that color changes may be controlled
in animals through one or other of two radically different
physiological systems: a direct nervous control as seen
particularly in fishes and in reptiles, and a secretory or
neurohumoral one as exemplified in crustaceans and am-
phibians. It is now proposed to examine these two
types of control and to ask the question, are they as
different as at first sight they appear to be, or have they
elements enough in common to allow them to be brought
under one general plan of action (Parker, 1932)? In
attempting this analysis I shall discuss the color changes
of two fishes, the common smooth dogfish, Mustelus
canis, and the killifish, Fundulus heteroclitus, a small
top-minnow of the Atlantic coast.

II

THE DOGFISH

The first fish whose chromatophore system I wish to
consider is the common smooth dogfish of the New Eng-
land coast, Mustelus canis. This fish was shown first by
Lundstrom and Bard (1932) to have a decided though

limited color chanj

§J

Fig.

5-

fwo smooth

dogfish

es,

Mustelus,

orig-

inally

of

the same

tint,

twenty

-four

hours

after

On a white background it grad-
ually and slowly becomes pearly
white, often with a pinkish tint
due to the color of the blood
showing through its translucent
skin. On a black background
it darkens more quickly to a
deep slate color (Fig. 5). As
a microscopic inspection of its
skin shows, the pale phase is
due to a concentration of its
melanophore pigment (Fig. 6)
and the dark one to a disper-
sion of this coloring matter

(Fig. 7).

The dark phase was very
fully studied by Lundstrom and

Bard. They showed that after
the pituitary gland had been
extirpated this fish began to
blanch in about thirty minutes
and reached full pallor in some
twelve hours. This pallor was
maintained even when the fish
was kept on a black background where under ordinary
circumstances it would have turned dark. The part of

12

the removal of the hy-
pophysis from the fish on
the right. They now
show extreme differences
in tint. Lundstrom and
Bard, Biol. Bull., 1932,
62, pi. 1.

THE DOGFISH

the pituitary complex that was concerned with this color
change was shown by Lundstrom and Bard to be the
so-called posterior lobe, which in the dogfish probably
includes the portion designated in the higher vertebrates
as the pars intermedia. When an extract made from
this lobe was injected into a pale hypophysectomized
Mustelus a distinct general darkening of the fish oc-
curred within three minutes, and after an

?*\l*!&'*

Fig. 6. Dermal melano-
phores of a smooth dogfish,
Mustelus, showing their pig-
ment in extreme concentration
causing the fish to appear pale.
Lundstrom and Bard, Biol.
Bull., 1932, 62, pi. 4.

Fig. 7. Dermal melano-
phores of a smooth dogfish,
Mustelus, showing their pig-
ment in extreme dispersion
causing the fish to appear dark.
Lundstrom and Bard, Biol.
Bull., 1932, 62, pi. 3.

animal was found to be as deep in tint as were fully dark
normal individuals. In from five to six hours thereafter
the dogfish had returned to its original pale shade.
Tests of the effects of fractions of the posterior lobe
showed that an amount of extract that represented one
twenty-fifth of this lobe would induce a noticeable local
deepening of color, while that equal to one-tenth to one-
seventh of a lobe would excite full darkening. Com-
mercial preparations of the posterior pituitary lobe such

14 COLOR CHANGES IN ANIMALS

as " pituitrin " (Parke, Davis & Co.) and " infundin "
(Burroughs, Wellcome & Co.) also brought on darkening
in pale pituitaryless dogfishes. The melanophores in
isolated pieces of dogfish skin reacted to these various
solutions in the same way as did the color-cells in the
whole fish. From these and other results Lundstrom
and Bard concluded that the darkening of Mustelus was
due to the action of a substance from the posterior lobe
of the pituitary gland in dispersing the melanophore
pigment which under unstimulated conditions was con-
centrated in the color cells. Thus these authors gave
a very adequate account of the darkening of Mustelus,
but passed over its pale phase with only slight con-
sideration.

Two years later Parker and Porter (1934) repeated
the essential parts of Lundstrom and Bard's work and
obtained confirmatory results. They observed further
that when the defibrinated blood from a dark dogfish
was injected into a pale one, a dark area resulted show-
ing that the blood, as might have been expected, carried
a neurohumor that induced a dispersion of the melano-
phore pigment. (Incidentally it may be remarked that
the defibrinated blood from a pale dogfish when injected
into a dark one had no effect upon the tint of the re-
cipient.) These observations all support the conclusion
that the dark phase of Mustelus is due to a dispersing
neurohumor produced in the pituitary gland and carried
from that gland by the blood to the responding color-
cells.

Parker and Porter, however, went further than to
confirm the results of Lundstrom and Bard. They at-
tempted to test for blanching in Mustelus by cutting
its nerves. Whenever an integumentary nerve in a dark
dogfish was cut the area thus denervated soon blanched.
This was best seen in the fins. If a cut about one

THE DOGFISH

L5

Fig. 8. Dorsal view of a
pectoral fin from a dark dogfish,
Mustelus, showing a light band
about an hour after the initiating
transverse cut had been made
in the fin. Parker and Porter,
Biol. Bull., 1934, 66, pi. 1, fig. 1.

centimeter long is made

through a pectoral fin of

a dark dogfish at right

angles to the fin-rays and

about a centimeter and

a half from the edge of

the fin, there will be pro-
duced a light band which,

starting from the cut,

will extend over the dark

part of the fin to its pale

edge (Fig. 8). The band

will begin to appear in

from ten to fifteen min-
utes after the cut has

been made. It will reach its maximum in about a day,

after which it will gradually disappear in from two to
three days. Light bands of this
kind can be excited even in dog-
fishes in the pale phase, show-
ing that the concentration of the
melanophore pigment in such a
band is more extreme than it is
in normal pallor (Fig. 9). The
pale bands produced by cutting
are not the result of circulatory
disturbances, for, as can be seen
under the microscope, their areas
even directly after the operation
show an active and apparently
normal circulation of blood.
Moreover in many parts of the
dogfish the nerves take very

different directions from the blood-vessels and when

cuts are made in these regions the blanched bands

Fig. 9. Dorsal view
of a pectoral fin from a
pale dogfish, Mustelus,
showing a light band
several days after the
initiating transverse cut
had been made. Parker
and Porter, Biol. Bull.,
1934, 66, pi. 1, fig. 4.

16

COLOR CHANGES IN ANIMALS

Fig. io. Dorsal view of a
pectoral fin from a dark dogfish,
Mustelus, showing a light band
in process of gradual disappear-
ance two days after the initiating
cut had been made. Parker
and Porter, Biol. Bull., 1934,
66, pi. 1, fig. 2.

follow the courses of the
nerves and not those of
the vascular supply.

When a pale band on
a dark dogfish begins
to disappear, it does so
by lateral invasion. The
dark area of the fin in
general creeps in on the
pale band from the two
sides till the band is ob-
literated (Fig. 10). If,
before this invasion of
the pale band has set in, a
longitudinal cut is made
along one edge of the
band, no invasion will occur on that side but the band
will gradually disappear
by invasion from the op-
posite side and from that
side only (Fig. 11). This
condition suggests that
the disappearance of pale
bands results from the
lateral infiltration of
some darkening agent.
At least there is no evi-
dence of any other fac-
tor being involved, for a
cut made in a pale fish
is never under any cir-
cumstances followed by
the formation of a dark
band or any other such
change.

Fig. 11. Dorsal view of a
pectoral fin from a dark dogfish,
Mustelus, showing a light band
in process of disappearance, an
operation here locally checked
by a longitudinal cut on one
side of the band. Parker and
Porter, Biol. Bull., 1934, 66, pi.
1, fig- 3-

THE DOGFISH 17

That the pale bands in Mustelus are the result of
nerve stimulation is rendered highly probable from the
following experiment (Parker, 1935^). If two needle
holes are made in the base of the anterior dorsal fin of
Mustelus, one about five millimeters in advance of the
other, and the two platinum electrodes of an induction
apparatus are inserted one in each hole, the preparation
after having been allowed to stand for half an hour or
so will show no change. If now the electric current is
started, in the course of ten minutes a pale band will
begin to appear in the fin and extend from a little above
the holes to the edge of the fin (Fig. 12). In twenty-

Fig. 12. Anterior dorsal fin of a dark smooth dogfish, Mus-
telus, which has been stimulated to the formation of light bands by
a transverse cut near the anterior edge of the fin and by electric
stimulation posterior to this cut. The resultant light areas are
seen in the fin peripheral to the two regions of stimulation.
Parker, Biol. Bull., 1935, 68, 2, fig. 1.

five minutes this band may become as distinct as that
produced by a cut. If in a given fin electric stimulation
and stimulation by a cut are started at approximately
the same time and in adjacent positions, two pale bands
will form, one from each center of excitation, and in a

18 COLOR CHANGES IN ANIMALS

way so that they are indistinguishable one from the
other. These results confirm the view held by Parker
and Porter that the pale phase in Mustelus is due to the
action on its melanophores of concentrating nerve-fibers
which may be stimulated by an induction current as
well as by being cut.

Combining the work of Lundstrom and Bard with
that of Parker and Porter a reasonably clear picture of
the color changes in Mustelus can be outlined. This
fish darkens in consequence of a pituitary neurohumor

Fig. 13. Two newly born smooth dogfishes, Mustelus; lower
one in the pale phase, upper one in the dark phase. Parker, Biol.
Bull., 1936, 70, pi. 1, fig. 1.

carried from the pituitary gland by the blood to the
melanophores which are thereby induced to disperse
their pigment. It blanches as a result of the action of
concentrating nerve-fibers by methods that I shall dis-
cuss more fully later.

Mustelus canis is ovoviviparous, that is, its eggs are
carried in the oviducts of the female till the young are
very fully formed when they escape from the mother's
body as active young fishes. A female may release from
four to a dozen or more such young at a time, and these

THE DOGFISH

19

at birth may measure from twenty-five to thirty centi-
meters in length. If the female dogfish is killed and
the young within her are quickly removed by what may
be termed a crude Caesarean operation, the pups, as
they are called, if handled gently, remain momentarily
passive even when immersed in sea water. After they
have been for a fraction of a minute or so in their watery
environment their gill movements begin and they will
start swimming, but with somewhat unsteady equilib-

* Life *
i ft* >

f

*.«

*

Fig. 14. Dermal melano-
phores of a newly born smooth
dogfish, Mustelus, in the pale
phase, pigment concentrated.
Parker, Biol. Bull., 1936, 70,
pi. 1, fig. 3.

Fig. 15. Dermal melano-
phores of a newly born dogfish,
Musteius, in the dark phase,
pigment dispersed. Parker,
Biol. Bull., 1936, 70, pi. 1,
% 4-

rium. In a very short time, however, they are indis-
tinguishable from pups normally born.

Fully formed young dogfishes when first removed from
the female are darkish in tint, but they will respond
immediately to the tone of the environment. In one
instance three newly born pups were put at once into
a black-walled illuminated tank. In twenty minutes
they had darkened perceptibly and in a little less than
an hour they were fully dark. Others that had been
put directly after birth into a white-walled tank were

20

COLOR CHANGES IN ANIMALS

fully blanched in about an
extreme degrees of tint in

hour and ten minutes. The
these young fishes (Fig. 13)
were quite equal to those
of the adults, and the
pigment in their melano-
phores showed as much
concentration (Fig. 14)
and dispersion (Fig. 15)
as did that in the mature
fishes. When transverse
cuts were made in the fins
of these young fishes pale
bands developed (Fig. 16)
as in the adults. In all
these respects the newly
born pups of Mustelus
agree with the adults,
and they may be said to come into the world with
their chief systems of organs including their melano-
phores fully functional (Parker, 1936^).

Fig. 16. Dorsal view of a
pectoral fin from a newly born
dogfish, Mustelus, showing two
pale bands as the result of small
transverse cuts. Parker, Biol.
Bull., 1936, 70, pi. 1, fig. i.

Ill

THE KILLIFISH

The killifishj Fundulus heteroclitus, is a small fish some
three to four inches in length, found in the coastal and
estuarial waters of the eastern shores of the United
States from Maine to Texas. It can be transferred di-
rectly from fresh water to salt water and back again
with impunity and in other respects as well it is very
hardy. In consequence it is a favorite live-bait for win-
ter fishermen and a very convenient animal for the
experimentalist. Its skin is abundantly supplied with
melanophores, the larger of which are often associated
with iridocytes. It also possesses a large number of
dermal xanthophores and a few local guanophores (Odi-
orne, 1933). Under appropriate circumstances depend-
ent for the most part on the background, Fundulus may
assume a pinkish, bluish, greenish, yellowish, or steel-
black tone, or pass over into a pale gray or almost pearly
white stage (Connolly, 1925). Some of these phases,
such as the pinkish one, are reached only very slowly
and may require days or even weeks for full accomplish-
ment. Others, like the changes from pale to dark and
the reverse, may be brought about in a few minutes.
These latter changes, as might be expected, are depend-
ent upon the action of the melanophores and have been
studied much more generally than the others. The
change from pale to dark and the reverse take place with
remarkable certainty and rapidity (Fig. 17). When a
pale Fundulus in a white-walled, illuminated vessel is
transferred to one with black walls, the fish will change
from its original pale tint to a dark one in a little less
than a minute, though the full completion of this change

22 COLOR CHANGES IN ANIMALS

may require nearly two hours. When the reverse is
tried and a dark fish is transferred from an illuminated
black-walled vessel to a white-walled one the fish

Fig. 17. Dark phase (above) and pale phase (below) of the
killirlsh, Fundulus, as the results of exposure to darkand to light
backgrounds. Parker, Jour. Exp. Bio/., 1934, II, pi. 1, fig. I.

blanches in a little over two minutes, though again the
final stage may not be reached for some four hours.
These changes take place in the times just given pro-
vided the temperature of the water in the experimental
vessels is approximately 20° C. At lower temperatures
the reaction times may be much lengthened, but under
all circumstances darkening appears to be accomplished
always more rapidly than blanching, a rule that holds
for the corresponding changes of a number of other ani-
mals (Parker, Brown, and Odiorne, 1935). It is an
interesting fact that when a normal Fundulus, either
pale or dark in tint, is put in an environment of com-
plete darkness it does not assume a dark tone as might
be expected, but it blanches strikingly and remains pale
(Parker and Lanchner, 1922). This fact, which has
been already noted in other fishes by von Frisch (191 1),
makes it obvious that there must be a significant differ-
ence in these forms between being in the dark and seeing

THE KILLIFISH 23

black. Another feature that should be noted in passing
is that a Fundulus from which the eyes have been re-
moved is pale in darkness, as normal individuals are,
but moderately dark in bright illumination. This dark-
ening of blinded fishes in illuminated vessels occurs to
an equal degree whether these vessels have black walls
or white walls (Parker and Lanchner, 1922). Such en-
vironmental differences are responded to, as might be
expected, only when the eyes of the fish are functional.
The general color conditions just described for Fun-
dulus are exhibited by this fish the year round, but
during the breeding season, the height of which is toward
the end of June, all color responses are greatly intensi-
fied, especially in the males. In fact at this time of
year the male is readily distinguished by a dark eye-like
mark on its dorsal fin, a nuptial secondary sex-character,
which is formed by an aggregation of melanophores in

Fig. 18. Four enlarged views of the dorsal fin of the killifish,
Fundulus, showing the nuptial mark in the male (1, 2) and its
absence in the female (3, 4). Figures 1 and 3 represent the dark
phase and figures 2 and 4 the pale phase of this fish. Parker and
Brower, Biol. Bull., 1935, 68, 5.

24 COLOR CHANGES IN ANIMALS

the fin membrane (Fig. 18). This mark can be seen in
the males from April to November, but not at other
times of year (Parker and Brower, 1935). Such sea-
sonal markings and the accentuated color changes that
accompany them during the breeding season undoubt-
edly play an important role in the sporting of the sexes
in mating. The less pronounced year-round changes of
darkening and blanching are probably of a character
purely protective.

Some of the observations recorded in the preceding
paragraphs are not of great moment for the present
discussion and have been little studied, but I have intro-
duced them here that you may appreciate to some extent
the complexity in the reactions of the melanophores in
the killifish, and that you may have some understanding
of what we are dealing with when we attack the prob-
lem of even the simpler color changes in this fish. The
main features to be kept in mind in the following dis-
cussion are that in Fundulus under normal conditions
the animal turns quickly dark in an illuminated, black
environment, and less quickly pale in a similar white
one. It is hardly necessary to say that the dark phase
of the fish is one in which the pigment of its melano-
phores is fully dispersed, and the pale phase one in which
this coloring matter is compactly concentrated.

We may now proceed to inquire into the mechanism
of the melanophore changes in Fundulus and, since they
are dependent upon the eyes, we are naturally led to
ask first of all how much of a part nerves play in these
reactions. A host of investigators have attempted to
identify chromatophoral nerves by the very reasonable
step of cutting given nerve-strands and then looking for
possible color changes in the integumentary areas that
have been thus denervated. This method was em-
ployed by Briicke, by Pouchet, by von Frisch, and many

THE KILLIFISH

others. It can be easily and conveniently used on the
tail of Fundulus. The tail of this fish is a blunt, sym-
metrical organ like the free end of a spatula, supported
by a system of over twenty fin-rays along which
nerves pass from the root of
the tail toward its free edge.
A short cut made transverse
to these rays and near their
proximal ends severs one or two
of them with their associated
nerve-strands, thus producing
an elongated denervated area
extending from the cut to the
free edge of the tail (Fig. 19).
This cut, excepting in its im-
mediate neighborhood, produces
no disturbance in the blood
supply to the denervated part
of the tail, for collateral vessels
are so numerous in this organ
that a short distance along
the band from the cut a normal flow of blood can
always be seen under the microscope.

If a cut as described is made in a pale Fundulus, the
denervated area will begin to appear as a darkened one
within thirty seconds of the time of the operation, and
the band will continue to deepen in tint till it reaches a
maximum a few hours later. Such a band corresponds
to the darkened areas produced by nerve cutting in the
chameleons and other lizards (Briicke, 1852; Bert, 1875;
Keller, 1895; Redfield, 191 8; Hogben and Mirvish,
1928*2, 1928^; Zoond and Eyre, 1934) as well as in a
great variety of fishes (Pouchet, 1876; von Frisch, 191 1;
Wyman, 1924; Hewer, 1927; Giersberg, 1930; Fries,
1 931; Smith, 1931^; Mills, 1932*2; Parker, 1934^, 1935^).

Fig. 19. Diagram of
a caudal fin of a killifish,
Fundulus, showing a
band of dark melano-
phores produced by cut-
ting radial nerves near
the root of the tail.
Parker, Pro. Nat. Acad.
ScL, 1934, 20, 307, fig. 1.

26

COLOR CHANGES IN ANIMALS

As I have already stated, the majority of workers begin-
ning with Briicke (1852) and extending down even to
Sand (1935) have expressed the opinion that such sev-
ered chromatophoral nerves are paralyzed and that the
associated melanophores after the severance of the nerve
lapse into a state of inactivity comparable to the relaxed
condition of inactive muscle.

That this view of the relation
of the chromatophoral nerves
and their associated melano-
phores is probably erroneous is
seen from the following experi-
ments (Parker, 1934*3:). If a
dark caudal band is induced in
a pale Fundulus in the way
already described and the fish
is kept in a white illuminated
vessel, the band will gradually
fade, as was first pointed out
by Pouchet in 1876 and subse-
quently confirmed by von Frisch
(191 1). In the caudal band of
Fundulus this fading will occur
in a few days more or less
variable with different individ-
uals. If, after the caudal band
in Fundulus has faded, a new
transverse cut is made within the area of the old band
and slightly distal to the original cut, a wholly new dark
band will appear reaching from the new cut over a part
of the old band to the edge of the tail (Fig. 20). The
formation of this new band could not possibly take place
if the nerve-fibers originally cut had been paralyzed by
that step. The formation of the second band within
the limits of the first is to be interpreted, in my opinion,

Fig. 20. Diagram of
a faded band in the tail
of a killifish within which
a new short cut has been
made. This cut induced
the formation of a new
small band within the
larger one showing that
the severed nerve-fibers
of the original band were
still active. Parker, Pro.
Nat. Acad. Set., 1934,
20, 308, fig. 3.

THE KILLIFISH

27

as a temporary reactivation of the original fibers with
a corresponding response of the melanophores. When
these fibers were originally cut they were not paralyzed
by this operation, but were profoundly stimulated, and
this stimulation continued, though with diminishing
energy, for one or more days as indicated by the state
of their effector organs, the melanophores. This at least
is the interpretation that I
have been led to put on this
phenomenon. I am fully aware
that it is quite contrary to the
teachings of conventional neuro-
physiology. But this type of
physiology has been developed
on the basis of the nerve-
muscle preparation, and there
may be important differences
between such preparations
and those based on melano-
phores.

If cut melanophoral nerve-
fibers remain active for a long
time after severance, there must
be a continuous flow of activa-
ting impulses from the region of
the cut to the melanophores
that constitute the dark band.

One test of the correctness of this view would be to
attempt to block such a flow. This cannot well be done
with drugs because of the readiness with which they
diffuse, but it can be accomplished by the application of
cold. If a sharply bent, capillary glass-tube carrying
a chilled mixture of water and alcohol several degrees
below zero centigrade is applied to a region on a fully
formed caudal band about midway its length, a very

Fig. 21. Diagram of
a fully formed band in
the tail of a killiflsh to
which a cold block {A)
had been applied with
the result that the distal
half of the band faded
in about a quarter to
half an hour after the
application of the block.
Parker, Pro. Nat. J cad.
Set., 1934,20, 309, fig. 5.

COLOR CHANGES IN ANIMALS

remarkable occurrence will be noticed. The band prox-
imal to the region of the application of the cold tube
will remain unchanged, but that distal to it will in the
course of half an hour or less gradually fade out (Fig.
21). This is what would be expected if the band was
maintained by a continuous flow
of impulses from the cut toward
the free edge of the tail.

A cold block may be used in
still another way to test this
question. If the cold tube is
applied to a spot near the center
of the tail of Fundulus and a
denervating cut is made some
distance proximal to the region
of application, a dark band will
begin to form and will extend
from the cut to the region of
the cold block but, as might
be expected, will not pass be-
yond this block. If now a cut
is made immediately distal to
the block and in line with
the first one, an additional
band will form from the new
cut to the edge of the fin (Fig.
2.2). From this experiment two
important conclusions may be
drawn: first, that cold in the neighborhood of o° C.
serves as a real block to nervous impulses over chro-
matophoral fibers; and, second, that what is transmitted
from the central organs over these fibers is not an inhibi-
tory influence that is checked when the fibers are cut,
but a true activating influence that may be excited
locally by severing the nerve.

Fig. 22. Diagram of
a caudal fin of a killifish
across a part of which a
capillary tube {A) carry-
ing a chilling mixture
was placed. The cut (B)
was followed by the for-
mation of a dark band
which reached from the
cut to the chilled area
but did not enter it. The
cut (C) gave rise to a
dark band which reached
to the edge of the tail.
Parker, Pro. Nat. Acad.
Set., 1934, 20, 307, fig. 2.

THE KILLIFISH

29

Precisely what induces the excitation at the cut and
maintains it cannot now be stated definitely. Since the
tail of the fish is in almost continuous lateral movement
during the life of the animal it might be supposed that
after the cut is made the continual rubbing of the sev-
ered ends of the nerve-fibers in the wound is the occasion
of the prolonged activity of these fibers. That such,
however, is not the case can be shown by the simple
experiment of cutting a window
in the tail instead of making
merely a transverse slit (Fig.
23). Under such circumstances
the nerve-fibers are no longer
rubbed on the rough surfaces of
the wound, and yet from a
window a caudal band is formed
and maintained precisely as it
is from a simple transverse cut.
Mechanical stimulation prob-
ably has nothing to do with
the maintenance of the band.
There are, however, enough
other disturbances in the cut to
account for continuous stimula-
tion, but precisely what these disturbances are is at
present unknown.

Nerves that are cut as these caudal nerves have been
would naturally be expected to undergo degeneration,
and it is reasonable to ask whether this process plays
any part in the activities under consideration. After a
denervated darkened band in the tail of a Fundulushas
blanched, the activity of its severed nerve-fibers can be
tested by recutting them. If the band darkens a second
time, the fibers must be regarded as still functionally
active; if not, they may reasonably be suspected of de-

Fig. 23. Diagram of
a caudal fin of a killifish
in which a small window
was cut resulting in the
formation of a dark
band. Parker, Pro. Nat.
Acad. Set., 1934, 20, 308,
% 4-

30

COLOR CHANGES IN ANIMALS

%

Fig. 24. Tail of a killifish.,
Fundulus, eighteen days after a
primary cut and immediately
after a more proximal secondary
cut had been made resulting in
the formation of a dark block
between the two cuts. There
is no evidence of the regenera-
tion of dispersing nerve-fibers.
Parker and Porter, Jour. Exp.
Zoo/., 1933, 66, pi. i, fig. 3.

of dark melanophores
will appear between the
two cuts showing that
the central ends of the
chromatophoral nerves
are still normally active
(Fig. 24). This darken-
ing, however, will not
extend as a rule distal
to the first cut, a con-
dition which shows that
at this stage no new
nerve-fibers have grown
out from the old nerve-
stump into the dener-
vated area. At about
the eighteenth day after

generation. By this
means it can be shown
that the dispersing chro-
matophoral nerve-fibers
of Fundulus remain ac-
tive for some four to five
days after they have been
cut. Functional inactiv-
ity begins to appear in
about five days, thereby
showing that the early
stages of degeneration
have set in. If at any
time up to the eighteenth
day after the initial cut
was made, a second cut
is made slightly proximal
to the first one, a block

at *^ —

Fig. 25. Tail of a killifish,
eighteen days after a more prox-
imal secondary cut had been
made from which a dark band
extends partly over the old pri-
mary band and thus shows
evidence of some regeneration
in the dispersing nerve-fibers.
Parker and Porter, Jour. Exp.
Zoo/., i933>66^ P1- T> fig- 4-

THE KILL! FISH 31

the initial cutting or shortly thereafter, a second cut
in preparations of the kind described will be followed
by short dark streaks that can be traced into the region
of denervation (Fig. 25). As can be seen in fishes pre-
pared for this purpose and experimentally tested from
day to day, these dark bands extend farther and farther
each day till eventually they reach the edge of the fin
(Fig. 26). They are of
course due to the pres-
ence of melanophores

" :v

-

with dispersed pigment
and reflect the progres-
sive regeneration of the
dispersing nerve-fibers.
Tn Fundulus, under the
circumstances described,
this regeneration begins
about the eighteenth day
after the initial cut and
is completed on about
the twenty-fifth day.
The approximate dis-
tance covered by the
growing nerve in these
seven days is some six
millimeters, and the rate
of regenerative growth consequently is about 0.8 milli-
meter per day (Parker and Porter, 1933).

By an ingenious method that is an improvement over
the one just described, Abramowitz (1935) has made a
second determination of the rate of regeneration not
only for the dispersing fibers but also for the concen-
trating ones in Fundulus, for there is evidence, as we
shall see later, for two kinds of chromatophoral nerve-
fibers in this fish. In the method employed by Abramo-

Fig. 26. Tail of a killifish
twenty-five days after a primary
cut had been made. The prox-
imal secondary cut induced the
formation of a complete new-
caudal band showing that the
regeneration of the dispersing
nerve-fibers had been fully estab-
lished. Parker and Porter, Jour.
Exp. Zool., 1933, 66, pi. 1, fig. 6.

32 COLOR CHANGES IN ANIMALS

witz a Funduhis with a blanched caudal band, on which
the rate of regeneration in the dispersing nerve-fibers is
to be measured, is kept in a white-walled vessel. When
a measurement is to be made the fish is placed for five
minutes or more in a black-walled receptacle. Here it
naturally darkens, including such part of its caudal band
as has regenerated its dispersing fibers. The extent of
this part is measured and the fish returned to the white-
walled vessel where it remains till another measurement
is desired. The same method can be applied to the
concentrating nerve-fibers except that in this instance
the experimental fish is retained in a black-walled recep-
tacle and transferred temporarily for testing to a white-
walled one. In this instance the extent of the pale band
beyond the initial cut is of course what is measured.
By these methods Abramowitz calculated the rate of
regeneration in both sets of nerve-fibers and found them
in both instances to be about 0.5 millimeter a day. The
earlier determination by Parker and Porter, 0.8 milli-
meter a day, is not far from this more accurate figure
by Abramowitz, and both fall within the range of the
rates already published for the regeneration of nerves
in the frog (1.34 mm. per day, Harrison, 19 10; 0.24 mm.
per day, Williams, 1930; 1.44 to 0.20 mm. per day,
Speidel, 1933). As can be seen from these studies, the
degeneration of the chromatophoral nerves in Fundulus
begins some five days after they have been cut, and re-
generation is a matter of weeks later. Obviously these
degenerative and regenerative processes have nothing
to do with the formation of the caudal bands and their
blanching, all of which may occur in a day or so after
the formation of the band. Over the early period of
four or five days the chromatophoral nerves are func-
tionally active, and this activity quite reverses the older
conception concerning severed nerves. After they are

THE KILLIFISH 33

cut they are not at once paralyzed, as was formerly
thought, but they are by the very act of cutting thrown
into a state of superactivity which may last in Fundulus
for two, three, or more days. The resulting dispersion
of pigment in the associated melanophores then is not
indicative of a stage of inactivity comparable with that
of relaxed muscle, but is rather one of unusual excita-
tion. From these several lines of inquiry it appears to
be reasonably well established that the melanophores of
Fundulus are provided with nerve-fibers whose action
incites a dispersion of pigment in these color-cells. The
dark phase in this fish results then from the stimulation
of its melanophores by nerves, and the fibers concerned
may be designated, in consequence of this action, as
dispersing fibers. It must be evident at once that this
opinion is quite contrary to that which for the last half-
century has been generally espoused by the majority of
workers in this field. Nevertheless the facts presented
in the preceding pages warrant, in my opinion, the
acceptance of this newer conception.

If from the standpoint of the most recent work the
dark phase of Fundulus is due to the action on its
melanophores of a system of dispersing nerve-fibers,
what can be said about the pale phase? In the color
changes of fishes and of reptiles the pale phase has long
been regarded as occasioned by the direct action of
chromatophoral nerves. This view has been held in
consequence of the ease with which blanching can be
excited through electrical stimulation of integumentary
and other nerves. Such responses were observed as
early as 1852 in the chameleon by Briicke, whose obser-
vations on this point have been confirmed and extended
on the same animal by Hogben and Mirvish (19280,
1928^) and on Phrynosoma by Redfield (191 8). The
same appears to hold for fishes, as was demonstrated on

34 COLOR CHANGES IN ANIMALS

Phoxinus by von Frisch (191 1) and on numerous other
fishes by Spaeth (19 13), Schaefer (1921), Wyman (1924),
and others. As Sand declares (1935), electrical stimu-
lation of nerves in fishes has always been found to cause
the melanophores to contract.

This generalization is certainly supported by what is
known of Fundulus. If one or two rays in the tail of
a dark Fundulus are included between the electrodes of
an inductorium, on making the currents a pale band
develops between the region stimulated and the edge of
the tail. If, following the procedure of von Frisch, the
electrodes are applied to the occipital region of Fundulus
so as to stimulate its medulla, the whole fish becomes
pale with almost incredible quickness. These observa-
tions, which have been many times repeated and con-
firmed, point indubitably to the conclusion that the
chromatophoral nerves of Fundulus contain fibers that
are concerned with the concentration of pigment and
therefore with the consequent blanching of the fish.
These fibers may, therefore, be designated concentrating
fibers in contrast with the dispersing fibers already dis-
cussed. In Fundulus then it would appear that its
melanophores are provided with a double equipment of
nerve fibers, one set to excite dispersion and the other
concentration of pigment.

The conclusion thus arrived at raises the much-dis-
puted question concerning the double innervation of
effectors. So far as chromatophores are concerned, this
idea was apparently first suggested in 1875 by the
French physiologist Paul Bert, who in a very incidental
way and without proof of any serious kind advanced it
in explanation of ordinary chromatophoral responses.
Bert's declaration never received a thoroughgoing con-
sideration and has been allowed to drift on in a rather
indeterminate way. The number of investigators who

THE KILLIFISH 35

believe that their researches on the whole are favorable
to Bert's view (Carnot, 1896; Sollaud, 1908; Redfield,
1 91 8; Kahn, 1922; Giersberg, 1930; Smith, 1931*2) are
about equal to those who take the opposite stand (Spaeth
and Barbour, 191 7; Hogben, 1924; Gilson, 1926; Sand,
1935). The inconclusiveness of much of the work on
this question probably resulted from the fact that it
involved experiments with drugs mainly from the stand-
point of mammalian physiology, a somewhat uncertain
procedure when transferred to the lower vertebrates,
and further that no small part of it was done on am-
phibians where as we now know, thanks principally to
Hogben and his associates, chromatophoral nerves are
mostly, if not entirely, absent.

It is not my purpose here to attempt to deal with
this question in a general way, for, as will be shown
toward the end of this essay, the truth is that double
innervation is not a general question. Melanophores
may have double innervation, as I believe to be the case
in Fundulus, or single innervation as has been shown to
be true of the dogfish, or no innervation at all as appears
to be the case in lampreys (Young, 1935) and in the
frog. The point that I wish to discuss here is whether
there is reason to assume double innervation for the
melanophores in Fundulus.

Some of the evidence leading up to this view has
already been given, but there remain certain important
aspects of this question still to be considered. In 1932
Mills pointed out that a close scrutiny of the melano-
phores at the edge of a caudal band in Fundulus showed
very important differences depending upon the state of
the color-cells. The differences here considered are best
seen when the band itself is somewhat blanched. If
under such circumstances a given fish is darkened by
being kept in an illuminated black-walled vessel, the

36 COLOR CHANGES IN ANIMALS

pigment in certain melanophores at the edge of its band
will be found to be as fully dispersed as those on the tail
in general. When this fish is made to blanch by a short
retention in an illuminated white-walled vessel these
same melanophores will be found not to have concen-
trated their pigment fully but to have come to rest with
their coloring matter in a position between the extremes
of concentration and of dispersion; in other words, these
melanophores can disperse their pigment fully but can-
not concentrate it fully. In a corresponding way other
melanophores can be found also at the edge of the band
that can concentrate their pigment fully but cannot
disperse it fully. These conditions, so far as the band
as a whole is concerned, may cause its edge to appear
to shift slightly, depending upon whether the tail as a
whole is dark or pale. An understanding of this pecul-
iar situation is possible on the assumption of double
innervation, but not on that of single innervation.
When nerve-strands are cut, as in the excitation of a
caudal band, both kinds of fibers, assuming both kinds
to be present, must of course be severed. Nerve-fibers
in the tail do not pass out into that structure on strictly
radial lines, but scatter somewhat irregularly. Conse-
quently near the edge of a caudal band it would not be
surprising to find certain melanophores whose concen-
trating fibers had been eliminated by the cut, but whose
dispersing fibers were still intact, and others in which
the reverse was true. Under these circumstances some
melanophores would be open to excitation for dispersion
but not for concentration and vice versa, a condition of
affairs that would result in exactly what has been ob-
served. Double innervation then will explain this pe-
culiar state; single innervation will not.

Another aspect of the problem of double innervation
is found in the regeneration of chromatophoral nerves.

THE KILLIFISH 37

This is pointed out by Abramowitz in work which is in
process of publication and from which T am permitted
to make the following excerpt. When new nerve-fibers
grow out from a proximal stump of a severed caudal
nerve they take, as already mentioned, a distal course
over the previously formed band and grow at an approx-
imate rate of half a millimeter a day. If during this
regeneration the progressing front of the growing nerve
as indicated by the melanophores under its control is
studied closely, it will often be found to vary in position,
depending upon the states of the melanophores used in
the test. If the fish is rendered dark by keeping it in a
black-walled vessel, the front of the regenerating band
of nerve as judged by the melanophores may be at one
place; if the fish is immediately thereafter blanched by
being put into a white-walled vessel the front as judged
in the same way may be measurably elsewhere. In cer-
tain fishes the developing front as shown by dispersed
melanophores may be in advance of that shown by the
concentrated ones; in other fishes the reverse may be
true. This disagreement in the position of the advanc-
ing front as shown by the two states of the melano-
phores, a disagreement that in a way is like that already
described on the edge of the fully formed band, is quite
consistent with the idea of double innervation, but very
difficult if not impossible to reconcile with single inner-
vation. Obviously the two sets of regenerative nerve-
fibers do not always grow forward at precisely the same
rate. In some instances the concentrating fibers are in
advance, in others the dispersing fibers. Here again the
idea of double innervation is a necessary part in the
explanation of a well-ascertained functional state, that
concerned with the regeneration of melanophore nerves.
In consequence of these two important lines of evi-
dence, one from the work of Mills and the other from

38

COLOR CHANGES IN ANIMALS

that of Abramowitz, as well as from the original con-
sideration already set forth, it seems fair to conclude
that the melanophores of Fundulus possess a double
innervation, and that the two sets of nerve-fibers, dis-
persing and concentrating, are real elements in the
neuro-melanophore organization of this fish. It is inter-
esting to observe that in the figures of melanophore
nerves in fishes published in 1893 by Ballowitz (Fig. 27),

Fig. 27. Innervation of a chromatophore from the perch.
Ballowitz, Zeit. wiss. Zool., 1893, 56, pi. 38, fig. 21.

each color-cell receives several nerve-fibers and not sim-
ply one, as muscle-cells ordinarily do. Of these nerve-
fibers, which are sometimes rather numerous, one at
least is presumably a concentrating fiber and another a
dispersing one. From this standpoint it is indeed pos-
sible that melanophores may differ somewhat in their
functional capacities depending upon a larger or a
smaller number of one or other kind of fiber. Thus
some cells may be more active in concentrating their
pigment than in dispersing it in consequence of a pre-

THE KILLIFISH 39

dominant concentrating innervation. But this point,
so far as I am aware, is purely speculative.

Evidence for double innervation such as that which
has been presented is not easily and quickly gathered.
The only other fish that has been exhaustively studied
from this standpoint is the common freshwater catfish,
Ameiurus nebulosus, whose melanophore system (Par-
ker, 1934^) appears also to involve two sets of nerves.

Perhaps the most important advance made in the
physiology of chromatophores during the last decade
and a half has been the establishment of the fact that
pituitary secretions are often of the utmost significance
in color changes. Now what part do these secretions
play in the melanophore activities of Fundulus} To
test this question Matthews (1933) removed the pi-
tuitary glands from a number of killifishes, and after
their recovery he subjected them to changes of environ-
ment to ascertain whether they had lost to any extent
their capacity to alter their tint. Having observed no
such loss Matthews concluded that this gland in Fun-
dulus was of little or no importance in controlling color
changes. Matthews, however, made the interesting ob-
servation that an extract from the pituitary gland of
this fish when applied to an isolated scale was followed
by a concentration of pigment in the melanophores of
the scale. Rather the reverse of this effect was recorded
by Kleinholz (1935) who showed that when pituitary
extract from a Fundulus was injected into another on
whose tail was a partly blanched caudal band, this band
darkened though the fish as a whole did not. These
various observations demonstrate that under normal
conditions the pituitary gland in Fundulus is probably
of no importance in the control of the color changes,
though the exceptional responses obtained from its se-
cretion as applied to isolated scales by Matthews and
to caudal bands by Kleinholz call for further elucidation.

IV
NEUROHUMORS

The preceding extended examination of the melano-
phore system in Fundulus leads to the conclusion that
the pituitary secretions in this fish in all probability play
no real part in its color changes, which seem to be con-
trolled exclusively by two sets of autonomic nerves, one
concentrating and the other dispersing in function. It
might seem that this would be the end of our quest, but
there are other phases of the subject that lead us on to
rather novel and interesting fields of inquiry. These
have to do with the way in which the nerves control the
melanophores, a process which is approachable in the
blanching of caudal bands in fishes.

Dark spots produced on the skins of fishes by cutting
integumentary nerves were long ago recognized as tem-
porary. Pouchet in 1876 noted that these darkened
regions subsequently became as pale as the rest of the
fish, and von Frisch (191 1) observed that the contrast
between the dark area and the pale general surface of
a minnow may vanish in from twelve to fourteen days.
Smith (1931^), who worked also on the minnow, con-
firmed these early observations. When the ophthalmic
branch of the trigeminal nerve on one side of a minnow's
head is cut, a dark area appears and covers the anterior
dorsal aspect of the half of the head concerned. This
area disappears in the course of several days when the
fish is kept on a white background, but it does not dis-
appear when the fish remains on a black background.
This led Smith to the conclusion that some non-nervous
agency was here involved. This phenomenon was in-
vestigated in Fundulus by Mills (1932^) who showed

40

NEUROHUMORS

41

that a caudal band in Fundulus did not disappear uni-
formly as a whole but was subject to a gradual reduction
which beginning on the two lateral edges of the band
spread slowly toward its axis, the last portion of the
band to be seen. The steps in the peripheral disappear-
ance of the bands in Fundulus was demonstrated photo-
graphically by Parker (1935c) who succeeded with the
help of Abramowitz in photographing from hour to hour
identically the same area in a fading caudal band on
this fish. At the outset all melanophores across the
whole dark band had their pigment about equally dis-
persed (Fig. 28). Some nine hours later those on the

Fig. 28. Caudal band of a living killifish a quarter of an hour
after it has been formed by the cutting of a single fin-ray. This
ray is represented at the region of the photograph by its four
branches. The pigment in all denervated melanophores is fully
dispersed. Parker, Proc. Amer. Philos. Soc, 1935, 75, pi. 3,
fig- Pl-

edge of the band showed greater pigment concentration
than those near its axis (Fig. 29). Finally after about
two and a quarter days those near the axis had as con-

11

COLOR CHANGES IN ANIMALS

-•"•!44tf M4ui4&£-' .1"

Fig. 29. The same portion of the caudal band as is shown in
Fig. 28 photographed nine and a quarter hours after the initiating
cut had been made. The pigment in the marginal denervated
melanophores is more concentrated than that of the axial color-
cells. Parker, Pro. Amer. Philos. Soc, 1935, 75, pi. 3> % l6-

*€>*

-.iff ir*>*%< lilW-t

rrMX1

?wm

Fig. 30. The same portion of the caudal band as is shown in
Fig. 28 photographed fifty-four hours after the initiating cut had
been made. The pigment in all denervated melanophores is very
fully concentrated. Parker, Pro. Amer. Philos. Soc, 1935, 75,
pi. 3, fig. 18.

NEUROHUMORS 43

centrated pigment as those on the edge, and the band
as a whole had become quite pale (Fig. 30). In other
words, the concentration of pigment begins on the edge
of the band and, as was clearly stated by Mills, proceeds
toward its axis. Subsequently Abramowitz found that
relatively large dark areas, such as may be produced on
one side of the head of Fundulus by cutting the ophthal-
mic nerve of that side, also disappeared by peripheral
reduction.

This method of blanching is of no small significance
for the topic at hand. It indicates that the influence
that induces blanching proceeds not from below, in
which case the whole dark area would become pale at
the same time, but that it invades the area from the
side. The importance of this will be more fully appre-
ciated if we turn first to the structure of the tail.

The tail of Fundulus, like that of most bony fishes,
is made up of two layers of skin supported within by
relatively stout fin-rays separated by considerable inter-
vals (Fig. 31). A few melanophores are lodged in the

<&«&&***

Fig. 31. Cross section of the caudal fin of the catfish, Amei-
uruSy showing near the middle a fin-ray cut across and on each
side above and below the integument with melanophc
Parker, Jour. Exp. Zoo/., 1934, 69, pi. 3, fig. 14.

lores.

cavities of the fin-rays, but the great majority of them
rest on the deep surface of each layer of skin (Fig. 32).
The space between the two layers of skin where it is not
occupied by the fin-rays is filled with loose connective

44

COLOR CHANGES IN ANIMALS

tissue, and through this connective tissue and the fin-
rays run the blood-vessels and nerves that supply the
tail. In this way the two layers of melanophores in the
tail are subtended by tissue rich in blood and lymph.
Yet when a dark caudal band fades, as has just been
pointed out, it does so not uniformly as though it were

Fig. 32. Section of the integument of the caudal fin of the
catfish, Ameiurus^ showing two small spherical melanophores in
the epidermis and three large ones in the derma. Parker, Jour.
Exp. Z00L, 1934, 69, pi. 3, fig. 15.

acted on from below by some constituent of the adja-
cent blood and lymph, but laterally as though it were
attacked at its edges. It can, however, be made to fade
uniformly by injecting into the circulation of a given
fish a small amount of dilute adrenalin, after which the
whole band will blanch evenly, the axis as rapidly as
the edges.

Further evidence that the blood of a Fundulus does
not aid in blanching a band is seen in the fact that if
the defibrinated blood from pale fishes is injected into
a dark one no change in tint is seen in the recipient.
And the reverse is also true, namely, blood from a dark
fish has no effect upon the tint of a pale one (Matthews,
J933)- The fading of a caudal band as ordinarily ob-

NEUROHUMORS 45

served in Fundulus, therefore, is not to be attributed to
some action of the subjacent blood and lymph. It must
be due to some influence that enters the band from
the side.

It is a noteworthy fact, as Smith (1931^) pointed out,
that bands or like areas do not fade when fishes are
kept permanently on a black background. Fading oc-
curs only when the adjacent region in the tail or other
part of the body is pale, as though some agent from the
neighboring pale area made its way into the band.
Such an agent might well be a neurohumor produced
in the pale area by the concentrating nerve terminals of
that region and transmitted laterally into the band.

Such a neurohumor may be conceived of as spreading
slowly from the region of its production into that of the
band and of inducing there the same kind of change that
it is capable of accomplishing over the rest of the body.
That the action is in all probability a slow diffusion of
this kind is not only shown in the beginning of these
changes on the periphery of the band or other dark
areas, but by the fact that bands of different widths take
different amounts of time in which to disappear (Parker,
1934^). A band of a width of one millimeter on a pale
Fundulus will require on the average a little over twenty-
six hours in which to disappear; another two millimeters
in width calls for nearly fifty-two hours in which to
blanch. These determinations support the hypothesis
that a neurohumor produced in the pale region of the
tail and responsible for the tint of that region diffuses
from the pale region into the band and thus causes it in
time to blanch.

Not only is there reason for believing that concen-
trating fibers act on melanophores through concentrat-
ing neurohumors, but also that dispersing fibers act on
melanophores by corresponding means. After a caudal

46

COLOR CHANGES IN ANIMALS

band on a pale Fundulus has become about as pale as
the rest of the fish, the animal may be placed on a black
background, whereupon it will darken after a few min-
utes in all parts except the band. This will remain pale
for some time, but will in the course of several hours
gradually darken till finally it is as dark as the rest of
the fish. Such procedure may be repeated back and
forth many times, the fish
blanching or darkening quickly,
and the band following the
general tint of the fish, but
always with a lag of some hours.
A situation of this kind is
clearly explicable on the as-
sumption of two sets of neuro-
humors, one dispersing, the
other concentrating, and each
produced by its appropriate
nerve endings.

Evidence for the lateral spread
of the dispersing neurohumor
can be seen in certain experi-
ments that are better carried out
on the tail of the catfish, Amei-
urus, than on that of Fundulus.
The tail of the catfish presents
much the same conditions as that of Fundulus. If
one of its fin-rays is cut, a dark caudal band results
much as in the killifish. This band will likewise blanch
(Fig. 33) if the catfish is kept in a white environment,
though the process is slower than it is in Fundulus. If
now in a catfish with a pale caudal band two new dark
bands are made by cutting the fin-rays adjacent to that
of the pale band, and the cuts for the new bands are
made not near the root of the tail as that for the pale

Fig. t)i). Diagram of
the caudal fin of a hy-
pophysectomized catfish,
AtJieiurus, in the dark
phase. Above is a faded
caudal band and below a
dark newly excited one.
Parker, Jour. Exp. Zool.,
1934, 69, pi. 2, fig. 6.

NEUROHUMORS

47

hand was, but about halfway out on the rays toward
the tip of the tail, the result will be a central pale band
whose distal half will be abutted laterally by dark bands.
The proximal half of the pale band will be surrounded
only by the pale portions of the tail. Under such con-
ditions the distal half of the pale central band which is
flanked by the dark half-bands
will be seen to darken slowly,
whereas the proximal half will
remain light (Fig. 34). This
experiment shows quite clearly
how adjacent dark areas may
bring about a dispersion of
pigment in a pale area, a
change easily understood from
the standpoint of an invading
dispersing neurohumor. The
darkening just described occurs
only when the invaded region
of the pale band is a denervated
one. If the dark half-bands are
excited on either side of an
innervated pale ray, no such
deepening of tint occurs (Fig.
35). Apparently the normal
concentrating fibers of such an
innervated area are too active
in maintaining the pale state
to be overcome by an invading dispersing neurohumor
(Parker, 1934^).

According to this general view, then, the two sets of
melanophore nerves in Fundulus act on its color-cells
through appropriate neurohumors which are produced
by the proper nerve terminals and which excite in one

Fig. 34. Diagram of
the caudal tin of a hy-
pophysectomized catfish
with a faded caudal band
between two newlv cut
dark half-bands. The
half of the faded band
not flanked bv the new
dark bands has remained
pale; the half flanked by
the dark bands has dark-
ened slightly. Parker,
Jour. Exp. Zoo/., 1934,
69, pi. 1, fig. 5.

48

COLOR CHANGES IN ANIMALS

case pigment concentration and in the other pigment
dispersion (Mills, 1932^).

It is no easy task to ascertain the nature of these
neurohumoral substances in Fundulus. Attempts to
extract them from the skin of this fish have failed, prob-
ably because they are so much involved in the scales
that it is impossible to bring solvents easily into close
contact with them. This, how-
ever, is not true of all fishes.
In some, as will be shown pres-
ently, they have probably been
in a measure isolated. One
point concerning their nature
in Fundulus seems to be clear,
namely, that they are not
carried in the blood. As al-
ready stated, the defibrinated
blood of a pale Fundulus has
no effect on the tint of a dark
fish and vice versa. This in-
dicates that these particular
neurohumors are not soluble
in water. If they are not open
to aqueous solution the only
other probable means of dis-
solving them is oil or fat, and it
is my opinion that these neuro-
humors are oil-soluble (Parker, 1933^, 1933^) and are
transmitted, not through the lymph or other watery
fluids between cells, but from cell to cell by means of
their lipoid or oily constituents. Under such circum-
stances transmission would be slow, as in fact it is known
to be, and would be limited to tissues in which the cells
are more or less in contact with one another. In this
respect the melanophores of Fundulus are a favorable

Fig. 35. Diagram of
the caudal fin of a catfish
with a normal inner-
vated band between two
newly cut dark half-
bands. The intermedi-
ate band retains its pale
tint throughout its whole
length. Parker, Jour.
Exp. Zool., 1934, 69, pi.
2, %. 7-

NEUROHUMORS 49

group of cells, for their processes are so richly branched
and intertwined that the necessary contacts for such a
transmission must be more than abundant.

The conception to which we are finally led respecting
the control of melanophores in Fundulus is as follows:
this control is accomplished through two sets of auto-
nomic nerves, concentrating and dispersing, and, though
it is what would be termed a strictly nervous control,
it is nevertheless based upon a special type of hormone,
a neurohumor, which ordinarily passes directly from the
nerve terminal to the effector cell, the melanophore,
over an almost submicroscopic distance, but under other
circumstances may make its way over stretches of a
millimeter or two from its region of origin to distant
effectors by way of the lipoid constituents of the inter-
vening tissues.

As a fish from which to attempt the extraction of an
oil-soluble neurohumor, Mustelus is much more favor-
able than Fundulus. The phase of Mustelus that is
suspected of being associated with such a neurohumor
is the pale one, and the parts that show this phase to
best advantage are the fins. Dogfishes were therefore
put in a white-walled illuminated tank and after a few
days, when they had become fully blanched, they were
killed and their fins removed. It was a matter of good
fortune that in the preparation of the fins the cutting
of nerves intensified their paleness rather than the re-
verse. The fins immediately after their removal were
ground to a pulp in an ordinary kitchen grinder, and
the pulp from the fins of one ordinary dogfish was then
thoroughly mixed with about two cubic centimeters of
Italian olive oil. This mixture was further ground by
hand for about half an hour in a rough porcelain mortar
till it reached the consistency of a thick paste and then
it was set aside to undergo extraction. In most in-

50 COLOR CHANGES IN ANIMALS

stances it was sterilized by heat before it was extracted,
but in the beginning this step was avoided. Whether
the paste was sterilized or not, its extraction was always
carried on at the low temperatures of an ordinary ice
refrigerator. After the paste had stood some fifteen
hours or so, it was mixed with its own volume of sterile
seawater, and the thick liquid that resulted was set
aside to allow the oil to rise to the top. In this way
there was collected a water-and-oil emulsion which after
having been roughly filtered through sterile cheesecloth
was vigorously agitated and injected subcutaneously in
appropriate amount into a dark dogfish. Very soon
after the injection had been made there commonly ap-
peared on the skin of the dogfish and a little in front of
the point of insertion a few small white spots which
however soon disappeared. As these spots appeared
when small amounts of indifferent fluids were injected
as checks they were regarded as of purely operative
origin. In from one to two days after the injection
relatively large pale areas made their appearance in the
skin immediately over the region into which the fin
extract had been introduced (Fig. 36). These large
areas were very persistent and, as could be shown under
a low power of the microscope, they were produced by
the concentration of melanophore pigment. That the
pale skin included in these spots was essentially normal
was demonstrated by the injection of pituitrin into a
fish with such a spot. Shortly after an injection of this
reagent had been made, particularly if the region of
injection was close to the pale spot, it disappeared by
the darkening of its melanophores only to return after
a few hours as the effect of the pituitrin wore off.

These large pale spots were not produced by injec-
tions of seawater, oil, oil extracts of dark fins or of
muscle, seawater extracts of pale fins, or defibrinated

NEUROHUMORS

51

Fig. 36. Left side of the trunk of a smooth dogfish, Muslelus,
in the region of the anterior dorsal fin showing a secondary light
spot due to an injection of 0.5 cc. of an emulsion of olive-oil
extract of blanched fins and seawater made a little over a day
previously. Parker, Jour. Gen. Physiol., 1935, 18, 840, fig. 1 A.

Fig. 37. Right side of the same fish as is illustrated in Fig. 36
showing no change of color after the injection of 0.5 cc. of an
emulsion of olive-oil and seawater. Parker, Jour. Gen. Physiol.,
1935, 18, 840, fig. 1 B.

52 COLOR CHANGES IN ANIMALS

blood from pale or from dark fishes (Fig. 37). They
were produced from oil extracts, sterilized or not ster-
ilized, of pale fins, and from cold ether extract and
Soxhlet ether extracts from the same. These various
tests lead to the conclusion that the induced pale areas
in Mustelus are due to the action of some substance that
can be extracted from the pale fins of this fish by olive
oil or ether. The exact source of this substance cannot
be stated, for it has been taken from the whole pale fin
only. That it is not in dark fins and not soluble in
water leads to the conclusion that it is in all probability
the concentrating neurohumor concerned with the nerv-
ous blanching of Mustelus, but proof of this view is far
from complete. The few known properties of the sub-
stance are its solubility in olive oil and in ether, its
insolubility in water, and its resistance to dry heat up
to no° C. It is probable that even in the oily Soxhlet
extracts it was present at most in extremely small
amounts (Parker, 1935^).

The only other fish that has been examined for the
possible presence of oil-soluble neurohumors is the cat-
fish Ameiurus. In this fish dark and pale phases are
well marked (Fig. 38) but the dark phase is the only
one favorable for study. Ameiurus (Parker, 1934^/) has
a melanophore system almost a duplicate of that of
Fundulus except that in addition to concentrating and
dispersing fibers Ameiurus has an active pituitary neuro-
humor which supplements the function of its dispersing
nerves. Extracts of the skins and fins of dark catfishes
were prepared as in the case of pale dogfishes, and the
final extract was injected subcutaneously into light cat-
fishes. This operation was followed in a little less than
an hour by the formation of dark splotches on these
fishes (Fig. 39). Such splotches, which were caused by
the dispersion of pigment in the melanophores of the

XEUROHUMORS

53

Fig. 38. Pale phase (above) and dark phase (below) of the
common catfish, Ameiurus. Parker, Jour. Exp. Zoo!., 1934, 69,
pi. 1, fig. 1.

Fig. 39. A catfish, Ameiurus, into which an injection of olive-
oil extract of the dark fins and skins of five other catfishes had
been made anteriorly from the black dot below the adipose fin.
The resulting dark area is superficial to the region where the
injected fluid escaped from the needle point. Parker, Jour. Exp.
Biol., 1935, 12, pi. 1, fig. 2.

54 COLOR CHANGES IN ANIMALS

region concerned, disappeared spontaneously after a few
days. When they were first formed they could be tem-
porarily obliterated by an injection of adrenalin. Ex-
traction of the skin of Ameiurus by ether, hot or cold,
yielded residues that were slightly active in darkening
the skin, but they were by no means so effective as were
the ether extracts in the case of Mustelus. However,
the evidence from the catfish supports the view that in
Ameiurus a dispersing neurohumor is present which is
soluble in oil and in this respect resembles the concen-
trating neurohumor of Mustelus.

The survey that has just been made of the means by
which the melanophores of Fundulus> of Mustelus, and
of other related species of fishes are activated indicates
with reasonable certainty that the distinction between
excitation by nerves and excitation by hormones is not
a fundamental one. In what is ordinarily called direct
stimulation of melanophores by nerves, as occurs for
instance in Fundulus, there is sufficient ground to as-
sume that of the two sets of nerve-fibers present each
one when active produces a substance, a neurohumor,
that can excite in a melanophore an appropriate re-
sponse. This neurohumor is believed to be produced
by the numerous nerve-terminals that surround the
color-cell. It must pass from its region of origin, the
terminal organ, over the almost submicroscopic space
to the responding cell. In the sense that it passes from
one place to another it is a hormone, but it is a hormone
that ordinarily travels over only a very short distance.
However, as already demonstrated in Fundulus, it may
pass over as much as a millimeter or so of intervening
space. It is therefore in all essential respects as much
a hormone for the activation of melanophores, as the
pituitary secretion is. So far as transmission is con-

NEUROHUMORS 55

cerned the pituitary secretion differs from that in the
fin only in the much greater distance that the former
must cover (Parker, 1934^). As previously suggested,
all these agents, be they short-range or long-range, acti-
vate the melanophores in essentially the same way.
Hormonal excitation and nervous excitation so far as
color-cells are concerned are really one in principle; both
are carried on by special hormones, the activating neu-
ron u mors.

It would be quite impossible at present to attempt a
catalogue of neurohumors. From what has been men-
tioned in discussing the conditions in Fundulus and in
Mustelus there appear to be at least two classes of these
substances, the water-soluble and the oil-soluble or, as
they have been designated, hydrohumors and lipohu-
mors (Parker, 1935^). Hydrohumors are soluble in
water and especially in blood, lymph, or other watery
body fluids. In consequence they spread rapidly and
far and ordinarily bring about responses over the whole
animal. They are well represented by the chromato-
phoral secretions of the pituitary gland as seen in Mus-
telus^ Ameiurus and a host of other creatures. Lipo-
humors are soluble in lipoids, fats, fat solvents and the
like. Since such substances are essentially stationary
in the animal body, the lipohumors after dissolving in
them must diffuse through them and consequently move
very slowly from place to place. Lipohumors are there-
fore relatively local in their action and do not excite
responses of the body as a whole. They are appropriate
to animals that can change their color patterns and
maintain them thus changed as, for instance, certain
flatfishes. When such fishes are on a coarsely varie-
gated background they show an appropriately coarse
melanophore pattern which is strikingly readjusted to

56 COLOR CHANGES IN ANIMALS

5»

Fig. 40. A flat-fish, Paralichthys albiguttus, on checker-board
patterns of different sizes. All figures are from the same fish.
The length of this fish was 14 cm.; the sides of the checker-board
squares were 2 mm., 5 mm., 10 mm., and 20 mm. Mast, Bull.
United States Bur. Fish., 1916, 34, pi. 21.

NEUROHUMORS 57

a finely variegated background by a pattern of finer
texture (Fig. 40). Lipohumors are represented by the
concentrating and the dispersing neurohumors of Fun-
dulus and of Ameiurus, and by the concentrating neuro-
humor in Mustelus. Thus the two main groups of
neurohumors with their numerous representatives prom-
ise a rich field for functional investigation.

V

THE NERVOUS SYSTEM AND
CHROMATOPHORES

Brucke (1852) in his account of the color changes in
the African chameleon compared chromatophores to
ordinary muscle and declared that color-cells with their
pigment concentrated were in a condition comparable
to that of active, contracted muscle. This idea that the
phase of concentrated pigment is the active phase of a
chromatophore has been accepted by the majority of
workers (Keller, 1895; von Frisch, 1912^; Spaeth, 1916;
Giersberg, 1930; Sand, 1935). Carlton (1903), how-
ever, was led to reverse this view for Anolis in that he
declared that in this lizard the concentrated state was
the state of rest. Babak (1913) went still further and
expressed the view that both extremes, that of full con-
traction and of full dispersion, were conditions of activ-
ity probably in contrast with some intermediate resting
phase.

It is not my intention to discuss this question at
length. In fact it would probably be ill advised to do
so, for, in my opinion, more work should be done in this
general field before a sound conclusion can be reached.
Suffice it to say that all the views thus far expressed
are based more or less implicitly on a supposed simi-
larity between chromatophores and muscle, especially
skeletal muscle. Such a comparison, as I have else-
where intimated (Parker, 1935^/), appears to be quite
erroneous, and I believe that we should do well in
reflecting on the physiology of chromatophores not to
let it bias our thoughts.

It is probable that the active states of chromato-

58

THE NERVOUS SYSTEM 59

phores are those in which the contained pigment is
actually moving in the cells and the quiescent ones those
in which it is at rest (Redfield, 191 8). There is reason
to believe that in what may thus be called the active
state of a chromatophore its protoplasm is relatively
fluid and mobile, that is, in a sol condition, for the con-
tained melanin particles then show Brownian move-
ment, whereas in what has been called a state of rest
the protoplasm is firmer, in a gel state, in which the
melanin shows little or no Brownian motion (Gilson,
1926; Parker, 1935^/). This conception of activity and
rest in color-cells is wholly unlike that advanced by the
older workers, for it nullifies any comparison between
these cells and those of skeletal muscle.

Pouchet's discovery (1876), confirmed by von Frisch
(191 1 ), that chromatophores are controlled by what they
then called the sympathetic nervous system, but what is
now designated the autonomic system, was an important
step forward, for it put chromatophores in the category
of effectors such as glands and smooth muscle, and re-
moved them from that of ordinary muscle. Spaeth
(1916) emphasized this distinction when he declared
that chromatophores were modified smooth-muscle cells,
and it is certainly true that these two types of tissue
have many points in common. The resting and active
states of chromatophores as just described are in their
essentials very like those of smooth muscle. The active
state of this tissue is when its fibers are shortening or
elongating. Its resting state is when they are main-
taining constant lengths. Smooth muscle is primarily
a tonus tissue. Chromatophores may remain weeks in
a condition with dispersed or with concentrated pig-
ment, conditions of extreme tonus. But I do not agree
with Spaeth in declaring that in consequence of these
similarities chromatophores must be regarded as modi-

60 COLOR CHANGES IN ANIMALS

fied smooth-muscle cells. The sparse and scattered
innervation of smooth-muscle cells with only a nerve-
terminal here and there in a wealth of cells, is in strong
contrast with the multitude of nervous end-organs that
surround even a single chromatophore (Fig. 27).

The relation of chromatophores to nerves is extremely
diverse. In amphibians these color-cells are probably
without direct nervous control and are adjusted entirely
through pituitary neurohumors. In the dogfish, Mus-
telus, blanching is a nervous function, and darkening
results from a pituitary secretion. In the killifish, Fun-
dulus, both blanching and darkening are nervous. If
we accept the melanophore system of the killifish with
its concentrating and dispersing nerve-fibers as the more
usual type, we must turn to other kinds of muscle than
smooth muscle for comparison. The best of these is
the vertebrate heart-muscle. This muscle, like smooth
muscle, is under the control of the autonomic nervous
system. Moreover it has a double innervation, sym-
pathetic and parasympathetic. The sympathetic fibers
of the heart accelerate its action, the parasympathetic
inhibit it. From this standpoint the concentrating
fibers of melanophores are believed to come under the
same category as the sympathetic fibers of the heart,
and the dispersing fibers under the same as the para-
sympathetic. Much can be said for this comparison,
but the heart as a muscle is not the typical tonus organ
that the melanophore is. Further the parasympathetic
or inhibitory fibers of the heart appear to act on that
muscle through acetylcholirt, a substance which in all
respects fulfills the requirements of a neurohumor and
yet appears to have only a very slight effect upon me-
lanophores (Parker, 1934c), an effect which as a matter
of fact is the reverse of what was to have been expected.
Thus the comparison between chromatophores and heart

THE NERVOUS SYSTEM 61

muscle is as inadequate as that between these cells and
smooth muscle. The truth is that chromatophores,
though they have some similarities with smooth muscle
and others with heart muscle, differ from both to such
an extent that they must be regarded as a type of tissue
sui generis. They are in no sense to be classed with
any kind of muscle.

Is the double autonomic innervation of chromato-
phores, such as is seen in Fundulus and in Ameiurus,
to be regarded as sympathetic and parasympathetic?
This question is not easily answered. It has been stud-
ied by Smith (1931a) from the standpoint of the action
of drugs. Smith found that cocaine, a stimulus for
sympathetic fibers, induced a concentration of pigment
in the melanophores of Fundulus, and that ergot, a sym-
pathetic depressant, checked this concentration, results
which thus favored the view that concentrating fibers
belong to the sympathetic division of the autonomic
system. In a corresponding way pilocarpin and physo-
stigmin, both parasympathetic stimulants, called forth
pigment dispersion and this was retarded by atropin, a
parasympathetic depressant. These observations, so
far as they go, point to an affirmative answer to the
question at the beginning of this paragraph, but experi-
ments with drugs are always precarious, and Smith in
his final declaration is cautious not to draw too definite
a conclusion.

In the dogfish Mustelus only one set of nerve-fibers
is present, and they are concentrating fibers. In con-
formity with what has been said about Fundulus these
fibers should belong to the sympathetic division of the
autonomic system. Following the general concepts of
vertebrate neurology they would be classed as post-
ganglionic efferent fibers whose cell-bodies lie in appro-
priate autonomic ganglia and whose axons, as non-

62 COLOR CHANGES IN ANIMALS

medullated elements, pass out over the gray rami com-
municantes and the spinal nerves to their peripheral
effectors, smooth muscles, glands, or chromatophores.
These connections and their functional relations to the
melanophores have been recently worked out with great
care by Young (1933) in the dogfish Scy Ilium. Here
two very significant observations have been made; first,
that this dogfish possesses no gray rami communicantes
and, second, that it shows no pale areas when its integ-
umentary nerves are cut. This second observation is
in strong contrast with what has been described for
Mustelus by Parker and Porter (1934) and might be
taken as ground for doubting the correctness of their
statements. A repetition of their work carried out by
Parker (1936*3) has, however, fully confirmed their
findings and incidentally has led to the interesting dis-
covery that the spiny dogfish of the New England coast,
Squalus acanthias, ordinarily shows no pale bands when
the nerves in its fins are cut. In this respect it ap-
proaches Scy Ilium. It therefore seems probable that
the observations of Young and of Parker and Porter
are not really in conflict, but that different species of
dogfishes vary in their means of melanophore activation;
in some, such as Mustelus, the pale phase is under
nervous control; in others, such as Scy Hi urn and Squalus,
this phase is induced by other means. Such diversity
in a group of even closely related species is not surpris-
ing, for, as the study of animal color change progresses,
just such individual differences are continually appear-
ing. The conclusion to be drawn from this diversity is
that the distinction between sympathetic and para-
sympathetic autonomic elements which is reasonably
clear in the higher vertebrates is by no means so definite
in fishes, where individual differences may be very pro-
nounced. In this conclusion I am in agreement with

THE NERVOUS SYSTEM 63

Young (193 1, 1933) whose neurological work on teleosts
and elasmobranches points to the inadvisability of
drawing these distinctions in the autonomic system of
the lowest group of vertebrates with too great sharpness.
So far as these two classes of fibers are concerned, the
autonomic system in fishes is like a mother liquor out
of which has crystallized the much more definite auto-
nomic components — sympathetic and parasympathetic
in the higher vertebrates.

The diversity of organization in chromatophoral
systems is apparent even more in the variety of neuro-
humors than in the types of innervations. As I have
already pointed out, it would be premature to attempt a
general consideration and classification of these activa-
tors. The most that can at present be done is to divide
them into the two groups of lipohumors and of hydro-
humors (Parker, 1935^). In Fundulus, Ameiurus, and
Mustelus the concentrating agents are lipohumors as
are the dispersing humors in Fundulus and in Ameiurus.
The ordinary hydrohumor from the pituitary gland dis-
perses melanophore pigment almost universally; that
from the medulla of the adrenal gland, adrenin, concen-
trates it with still greater uniformity. The parasympa-
thetic fibers to the vertebrate heart produce a hydro-
humor, acetylcholin, that inhibits the heart muscle, and
the same class of fibers in Fundulus produces a lipo-
humor that disperses melanophore pigment. When we
seek for generalizations in such an array of details we
find at present little beyond those associated with the
two types of solubilities already discussed.

That agents like neurohumors are active in such gen-
eral central functions as the transmission of nerve im-
pulses from one neurone to another has for some time
been surmised. The production of a neurohumor on one
side of a synapse and its reception on the other may

64 COLOR CHANGES IN ANIMALS

well be the explanation of synaptic polarization and of
the appreciable loss of time in the passage of an impulse
over such a junction. Sir Charles Sherrington in 1925
discussed these problems particularly in relation to
central nervous activation and inhibition, and pointed
out the possibility of two agents, which he termed A
and I, concerned with these operations. Such agents
were conceived of even as substances, and the chemical
trend thus given to the interpretation of synaptic func-
tions was favored by a number of investigators including
Ballif, Fulton, and Liddell (1925), Fulton (1926), and
Samojloff and KisselefF (1927). Others, following the
lead of Sir Charles Sherrington (1929), preferred to
adopt a less committal attitude and to designate these
synaptic conditions as central excitatory states (c.e.s.)
and central inhibitory states (c.i.s.) where the term
" state " was especially associated with neither action
nor substance. (Creed, Denny-Brown, Eccles, Liddell,
and Sherrington, 1932.) In this way the unsolved
problem of what lies behind these states was temporarily
put aside.

So far, however, as melanophores and their responses
are concerned, they appear to favor the more strictly
chemical interpretation just given to central nervous
operations. The concentrating neurohumors of the
color-cells would naturally represent the central excita-
tory substance, and the dispersing chromatophoral neu-
rohumors the central inhibitory substance. Such at
least might well be the hypothesized relationships.
Central synaptic functions are as a rule strikingly local-
ized. They would therefore be more successfully car-
ried out by lipohumors than by hydrohumors whose
tendency to diffuse widely in the watery environment
of the central nervous tissues would soon blot out local
limitations, Lipohumors have been shown in Fundulus

THE NERVOUS SYSTEM 65

and in Ameiurus to occur in opposing pairs. Such
pairs with their restricted fields of action may serve as
the prototypes of the activating and the inhibiting
agents in central nervous operations. Hydrohumors,
on the other hand, with their powers of rapid and free
spread would exert broad and general influences on the
whole nervous organization. Such influences could
make themselves felt in the general tone of the central
nervous system, the kinds and degrees of personality,
and in those abnormal states that fill our hospitals.

In this way neurohumors may play a very significant
part in nervous operations. Their occurrence as active
intermediaries in the chromatophoral system is a matter
of growing certainty. They also appear to have an im-
portant role in the excitation of smooth muscle and in
the control of the vertebrate heart-muscle. It is easy
to conceive of them as the effective agents in such cen-
tral nervous functions as activation and inhibition just
mentioned, in the relation of receptor cells to their
conducting neurones, and in that large body of nervous
interrelations where the integrity of nerve-units is de-
pendent upon the so-called trophic function. In these
numerous situations the idea of neurohumors affords
interesting hypothetical suggestions that lead at once to
experimental tests. The devising and applying of such
tests is no easy task, but it is just such exertions that
often yield the highest scientific returns. To approach
neurohumors and to understand their ways demands
the supreme effort of the investigator.

REFERENCES

Abramowitz, A. A. 1935- Regeneration of chromatophore
nerves. Pro. Nat. Acad. Sci. Washington, 21, 137-141.

Adler, L. 1914- Metamorphosestudien an Batrachierlarven.
Arch. Entw.-mech. Organ., 39, 21-45.

Allen, B. M. 1917. Effects of the extirpation of the anterior
lobe of the hypophysis of Rana pipiens. Biol. Bull., 32,
1 17-130.

Aristotle. 1883. History of animals. Trans. R. Cresswell,
London, 326 pp.

Atwell, W. J. 1919. On the nature of the pigmentation changes
following hypophysectomy in the frog larva. Science, 49,
48-50.

Babak, E. 1913- Ueber den Einfluss des Lichtes auf die
Vermehrung der Hautchromatophoren. Arch. ges. Physiol.,
149, 462-470.

Ballif, L., J. F. Fulton, and E. G. T. Liddell. 1925. Observa-
tions on spinal and decerebrate knee-jerks, with special refer-
ence to their inhibition by single break-shocks. Pro. Roy.
Soc. London, B, 98, 589-607.

Ballowitz, E. 1893. Die Nervenendigungen der Pigmentzellen,
ein Beitrag zur Kenntnis des Zusammenhanges der End-
verzweigungen der Nerven mit dem Protoplasma der Zellen.
Zeit. wiss. Zool. 56, 673-706.

Bert, P. 1875. Sur le mecanisme et les causes des changements
de couleur chez le cameleon. C. R. Acad. Sci. Paris, 81,
938-941.

Briicke, E. 1852. Untersuchungen iiber den Farbenwechsel des
afrikanischen Chamaleons. Denkschr. Akad. Wiss. Wien,
math.-nat. CI., 4, 179-210.

Carlton, F. C. 1903. The color changes in the skin of the
so-called Florida chameleon, Anolis carolinensis Cuv. Pro.
Amer. Acad. Arts Sci., 39, 259-276.

Carnot, P. 1896. Recherches sur le mecanisme de la pigmenta-
tion. Bull. Sci. France Belg., 30, 1-83.

Connolly, C. J. 1925. x^daptive changes in shade and color of
Fundulus. Biol. Bull., 48, 56-77.
66

REFERENCES 67

Corona, A. e A. Moroni. 1898. Contributo alio studio dell'
estratto di capsuli surrenali. La Reforma Medica, 14
(cited from van Rynberk, 1906).

Creed, R. S., D. Denny-Brown, J. C. Eccles, E. G. T. Liddell, and
C. S. Sherrington. 1932. Reflex activity of the spinal cord.
Oxford, 183 pp.

Fries, E. F. B. 1931. Color changes in Fundulus, with special
consideration of the xanthophores. Jour. Exp. Zool., 60,
384-426.

von Frisch, K. 1910. Ueber die Beziehungen der Pigmentzellen
in der Fischhaut zum sympathischen Nervensystem. Fest-
schrift R. Hertwig, 3, 15-28.

. 191 1. Beitrage zur Physiologie der Pigmentzellen in

der Fischhaut. Arch. ges. Physiol., 138, 3!9-387-

— . 19120. Ueber farbige Anpassung bei Fischen. Zool.

Jahrb. Abt. allg. Zool. Physiol., 32, 171-230.

1912^. Ueber die Farbenanpassung des Creyiilabrus.

Zool. Jahrb. Abt. allg. Zool. Physiol., 33, 1 51-164.

Fuchs, R. F. 1914. Der Farbenwechsel und die chroma-
tische Hautfunktion der Tiere. Winterstein, Handb. vergl.
Physiol., 3, 1, 1 189-1656.

Fulton, J. F. 1926. Muscular contraction and the reflex control
of movement. Baltimore, 644 pp.

Giersberg, H. 1930. Der Farbwechsel der Fische. Zeit. vergl.
Physiol., 13, 258-279.

Gilson, A. S. 1926. The control of melanophore activity in
Fundulus. Jour. Exp. Zool., 45, 457-468.

Harrison, R. G. 19 10. The outgrowth of the nerve fiber as
a mode of protoplasmic movement. Jour. Exp. Zool., 9,
787-846.

Hewer, H. R. 1927. Studies in colour changes of fish. II. An
analysis of the colour patterns of the dab. III. The action
of nicotine and adrenalin in the dab. IV. The action of
caffeine in the dab, and a theory of the control of colour
changes in fish. Trans. Roy. Soc. London, B, 215, 177-200.

Hogben, L. T. 1924. The pigmentary effector system. Edin-
burgh, 152 pp.

id L. Mirvish. 19280. Some observations on the pro-

am

duction of excitement pallor in reptiles. Trans. Roy. Soc.
So. Africa, 16, 45-52.
— — . 1928^. The pigmentary effector system.

68 COLOR CHANGES IN ANIMALS

V. The nervous control of excitement pallor in reptiles.

Brit. Jour. Exper. Biol., 5, 295-308.
Hogben, L. and D. Slome. 1931. The pigmentary effector

system. VI. The dual character of endocrine co-ordination

in amphibian colour change. Pro. Roy. Soc. London, B,

108, 10-53.
Hogben, L. T. and F. R. Winton. 1922^. The pigmentary

effector system. I. Reaction of frog's melanophores to

pituitary extracts. Pro. Roy. Soc. London, B, 93, 318-329.
— . 1911b. The pigmentary effector system.

II. Pro. Roy. Soc. London, B, 94, 1 51-162.

- — . 1923. The pigmentary effector system.

III. Colour response in the hypophysectomised frog. Pro.
Roy. Soc. London, B, 95, 15-31.

Kahn, R. H. 1922. Studien ueber die Innervation der Chro-

matophoren auf Grund gegensatzlicher Giftwirkungen.

Arch. ges. Physiol., 195, 337-360.
Keller, R. 1895. Ueber den Farbenwechsel des Chamaeleons

und einiger anderer Reptilien. Arch. ges. Physiol., 61,

123-168.
Kleinholz, L. H. 1935. The melanophore-dispersing principle

in the hypophysis of Fundulus heteroclitus. Biol. Bull., 69,

379-39°-
Koller, G. 1925. Farbwechsel bei Crangon vulgaris. Verh.

Deutsch. zool. Gesell., 30, 128-132.
Krogh, A. 1922. The anatomy and physiology of capillaries.

New Haven, 276 pp.
Lieben, S. 1906. Ueber die Wirkung von Extrakten chro-

maffinen Gewebes (Adrenalin) auf die Pigmentzellen. Zen-

tralbl. Physiol., 20, 108-117.
Lister, J. 1858. On the cutaneous pigmentary system of the

frog. Phil. Trans. Roy. Soc. London, 148, 627-643.
Lundstrom, H. M. and P. Bard. 1932. Hypophysial control of

cutaneous pigmentation in an elasmobranch fish. Biol.

Bull., 62, I-q.
Matthews, S. A. 1933. Color changes in Fundulus after hypo-

physectomy. Biol. Bull., 64, 315-320.
Mills, S. M. 1932^. The double innervation of fish melano-
phores. Jour. Exp. Zool., 64, 231-244.
. 1932^. Evidence for a neurohumoral control of fish

melanophores. Jour. Exp. Zool., 64, 245-255.

REFERENCES 69

Odiorne, J. M. 1933. The occurrence of guanophores in
Fundulus. Pro. Nat. Acad. Sci. Washington, 19, 750-754.

Parker, G. H. 1932. Humoral agents in nervous activity with
special reference to chromatophores. Cambridge, England,

79 PP-

. 19330. The cellular transmission of neurohumoral

substances in melanophore reactions. Pro. Nat. Acad. Sci.
Washington, 19, 175-177.

— . 1933^. Transmission of neurohumors in animals by
other means than blood and lymph. Pro. Soc. Exper. Biol.
Med., 30, 555-558-

— . 19340. The prolonged activity of momentarily stimu-
lated nerves. Pro. Nat. Acad. Sci. Washington, 20, 306-310.

. 1934^. Cellular transfer of substances, especially

neurohumors. Jour. Exper. Biol., 11, 8 1-88.
— . 1934*:. Acetyl choline and chromatophores. Pro. Nat.
Acad. Sci. Washington, 20, 596-599.

— . 1934^- Color changes of the catfish Ameiurus in rela-
tion to neurohumors. Jour. Exp. Zool., 69, 199-223.

• J935^- The electrical stimulation of the chromato-

phoral nerve-fibers in the dogfish. Biol. Bull., 68, 1-3.
— . 1935^. The chromatophoral neurohumors of the dog-
fish. Jour. Gen. Physiol., 18, 837-846.

— . 1935^. The disappearance of primary caudal bands in
the tail of Fundulus and its relation to the neurohumoral
hypothesis. Pro. Amer. Phil. Soc, 75, 1-10.

1935^/. What are the resting and the active states of

chromatophores? Pro. Nat. Acad. Sci. Washington, 21,
286-292.

— . 1936^. Color changes in elasmobranchs. Pro. Nat.
Acad. Sci. Washington, 22, 55-60.

— . 1936^. Integumentary color changes in the newly
born dogfish, Mustelus canis. Biol. Bull., 70, 1-7.

- and H. P. Brower. 1935. A nuptial secondary sex-
character in Fundulus heteroclitus. Biol. Bull., 68, 4-6.
— F. A. Brown, Jr., and J. M. Odiorne. 1935. The re-
lation of the eyes to chromatophoral activities. Pro. Amer.
Acad. Arts Sci., 69, 439-462.

— and A. J. Lanchner. 1922. The responses of Fundulus
to white, black and darkness. Amer. Jour. Physiol., 61,
548-550.

70 COLOR CHANGES IN ANIMALS

Parker, G. H. and H. Porter. 1933. Regeneration of chroma-

tophore nerves. Jour. Exp. Zool., 66, 303-309.

. 1934. The control of the dermal melano-

phores in elasmobranch fishes. Biol. Bull., 66, 30-37.
Perkins, E. B. 1928. Color changes in crustaceans, especially

in Palaemonetes. Jour. Exp. Zool., 50, 71-105.
Pliny, Sec, C. 1887. Natural History. Trans. J. Bostock and

H. T. Riley, London, 6 vols.
Pouchet, G. 1872. Du role des nerfs dans les changements de

coloration des poissons. Jour. Anat. Physiol., 8, 71-74.
Pouchet, G. 1876. Des changements de coloration sous l'influ-

ence des nerfs. Jour. Anat. Physiol., 12, 1-90, 1 13-165.
Redfield, A. C. 1918. The physiology of the melanophores of

the horned toad Phrynosoma. Jour. Exp. Zool., 26, 275-333.
Samojloff, A. und M. KisselefF. 1927. Zur Charakteristik der

zentralen Hemmungsprozesse. Arch. ges. Physiol., 215,

699-715.
Sand, A. 1935. The comparative physiology of colour response

in reptiles and fishes. Biol. Rev., 10, 361-382.
Sangiovanni, G. 18 19. Descrizione d'un particolare sistema di

organi cromoforo-espansivo-dermoideo a dei fenomeni ch'esso

produce, scoperto nei molluschi cefalopodi. Giornale en-

ciclopedico di Napoli, anno 13, no. 9.
Schaefer, J. G. 1921. Beitrage zur Physiologie des Farben-

wechsels der Fische. I. Untersuchungen an Pleuronectiden.

II. Weitere LIntersuchungen. Arch. ges. Physiol., 188,

25-48.
Sherrington, C. S. 1925. Remarks on some aspects of reflex

inhibition. Pro. Roy. Soc. London, B, 97, 519-545.
Sherrington, C. 1929. Ferrier Lecture. Some functional prob-
lems attached to convergence. Pro. Roy. Soc. London, B,

105, 33^-362-
Slome, D. and L. Hogben. 1928. The chromatic function in

Xenopus laevis. So. African Jour. Sci., 25, 329-335.

— . 1929. The time factor in the chromatic

responses of Xenopus laevis. Trans. Roy. Soc. So. Africa,

17, 141-150.
Smith, D. C. 1931a. The action of certain autonomic drugs

upon the pigmentary responses of Fundulus. Jour. Exp.

Zool., 58, 423-453-

— . 1931^. The influence of humoral factors upon the

REFERENCES 71

melanophores of fishes, especially Phoxinus. Zeit. vergl.

Physiol., 15, 613-636.
Smith, P. E. 1916. The effect of hypophysectomy in the early

embryo upon the growth and development of the frog.

Anat. Rec, 11, 57-64.
Sollaud, E. 1908. Role du systeme nerveux dans les change-

ments de coloration chez la grenouille. C. R. Acad. Sci.

Paris, 147, 536-538-
Spaeth, R. A. 1913. The physiology of the chromatophores of

fishes. Jour. Exp. Zool., 15, 527-583.
. 191 6. Evidence proving the melanophore to be a

disguised type of smooth muscle cell. Jour. Exp. Zool., 20,

193-215.

and H. G. Barbour. 1917. The action of epinephrin and

ergotoxin upon single physiologically isolated cells. Jour.
Exp. Pharm. Therap., 9, 431-440.

Speidel, C. C. 1933. Studies of living nerve. II. Activities of
ameboid growth cones, sheath cells, and myelin segments, as
revealed by prolonged observation of individual nerve fibers
in frog tadpoles. Amer. Jour. Anat., 52, 145-18 1.

Stark, J. 1830. On changes observed in the colour of fishes.

Edinburgh New Philos. Jour., 23, 327-3

Swingle, W. W. 1921. The relation of the pars intermedia of

the hypophysis to pigmentation changes in anuran larvae.

Jour. Exp. Zool., 34, 1 1 9-1 41.
Williams, S. C. 1930. Regeneration of peripheral nerves in

amphibia studied with the aid of a vital stain. Jour. Exp.

Zool., 57, 145-181.
Wyman, L. C. 1924. Blood and nerve as controlling agents in

the movements of melanophores. Jour. Exp. Zool., 39,

73-J32-
Young, J. Z. 1 93 1. On the autonomic nervous system of the
teleostean fish Uranoscopus scaber. Quart. Jour. Mic. Sci.,

74, 49J-535-

. 1933. The autonomic nervous system of selachians.

Quart. Jour. Mic. Sci., 75,^571-624.

1935. The photoreceptors of lampreys. II. The func-

tions of the pineal complex. Jour. Exp. Biol., 12, 254-270.
Zoond, A. and J. Eyre. 1934. Studies in reptilian colour re
sponse. I. The bionomics and physiology of the pigmentary
activity of the chameleon. Phil. Trans. Roy. Soc. London,
B, 223, 27-55.

INDEX

Abramowitz, A. A. 31, 37, 43

Activity of cut nerves, 29

Adler, L. 8

Adrenalin, 7, 44

Allen, B. M. 8

Ameiurus, 39, 46, 52

Amphibians, 8, 60

Aristotle, 1, 2

Atwell, VV. J. 8

Autonomic nerves, 59

Babak, E. ij 8

Ballif, L. 64

Ballowitz, E. 38

Bands,

by electric stimulation, 17

caudal, 25

disappearance of, 16, 26, 41, 45

pale, 15, 20

revival of, 26
Barbour, H. G. 3c,
Bard, P. 12, 13, 18
Bert, P. 25, 34
Blanching of Mustelus, 14
Blood exchange, 44
Blood supply, 1 5
Brower, H. P. 24
Brown, F. A. 22
Briicke, E. 6, 24, 25, 26, 33, 58
Carlton, F. C. 58
Carnot, P. 3;
Cephalopods, 4
Chameleon, 1, 6, 33
Chroma tophores,

activity and quiesence, 58

and muscle, 58

and nerves, 2
Cold block, 27
Color changes,

distribution, 3

Fundulus, 21

young Mustelus, 18
Concentrating nerves, 38, 61
Concentrating neurohumor, 45

Connolly, C. J. 21

Corona, A. 7

Crangon, 10

Creed, R. S. 64

Crustaceans, 5, 8

Dark phase of Fundulus, 33

Degeneration of nerves, 29

Denny-Brown, D. 64

Dispersing nerves, 33, 38

Dispersing neurohumor, 45

Dogfish, 12

Double innervation, 34

Eccles, J. C. 64

Erythrophores, 5

Eyes, 2, 3

Eye-stalk hormone, 10, 11

Eyre, J. 25

Fries, E. F. B. 25

von Frisch, K. 7, 22, 24, 25, 26, 34, 40,

58, 59
Frog, 3, 9
Fuchs, R. F. 7
Fulton, J. F. 64
Fundulus, 11, 21, 49, 60
Giersberg, H. 25, 35, 58
Gilson, A. S. 35, 59
Hamlet, 1 •
Harrison, R. G. 32
Hewer, H. R. 25"
Hogben, L. T. 8, 9, 10, 25, 33, 35
Hormones and nerves, 54
Hydrohumors, 55, 63
Infundin, 14
Iridocytes, 6, 21
Kahn, R. H. 3s
Keller, R. 25, 58
Killifish, 21
Kisseleff, M. 64
Kleinholz, L. H. 39
Roller, G. 10
Krogh, A. 9
Lanchner, A. J. 22
Leucophores, 6

73

74

INDEX

Liddell, E. G. T. 64

Lieben, S. 7

Lipohumors, 55, 63

Lister, J. 3

Lundstrom, H. M. 12, 13, 18

Matthews, S. A. 39, 44

Melanophores, 5, 21, 23, 24

Mills, S. M. 25, 35, 40, 48

Mirvish, L. 25, 23

Moroni, A. 7

Mullet, 1

Mustelus, 11, 12, 14, 49, 60

Nerve cutting, 8, 14, 24, 40

Neurohumor, 11, 40

Nuptial markings, 23

Odiorne, J. M. 21, 22

Palaemonetes, 10, 11

Pale phase of Fundulus, 23

Paralysis, 6, 26, 23

Parasympathetic nerves, 60, 61

Parker, G. H. 11, 14, 18, 22, 25, 31, 41

45, 47, 48, 52, 55, 58, 59, 60, 62
Perkins, E. B. 10
Phoxinus, 34
Phrynosoma, 8, 33
Pituitary gland, 8, 12, 13, 39
Pituitrin, 14
Pliny, 1
Porter, H. 14, 18,31,62

Pouchet, G. 6, 24, 25, 26, 40, 59

Protective coloration, 2, 24

Rate of nerve regeneration, 32

Redfield, A. C. 8, 25, 23, 35, 59

Regeneration of nerves, 30

Samojloff, A. 64

Sand, A. 26, 34, 35, 58

Sangiovanni, G. 4

Schaeter, J. G. 34

Scyllium, 62

Shakespeare, 1

Sherrington, C. S. 64

Slome, D. 9

Smith, D. C. 25, 35, 40, 45, 61

Smith, P. E. 8"

Sollaud, E. 35

Spaeth, R. A. 34, 58, 59

Speidel, C. C. 32

Squalus, 62

Stark, J. 2

Swingle, W. W. 8

Sympathetic nerves, 7, 59

Turbot, 6

Williams, S. C. 32

Winton, F. R. 9

Wyman, L. C. 25, 34

Xanthophores, 5, 21

Xenopus, 9

Young, J. Z. 25, 62» 63

Zoond, A. 25

$.»

SSsJ

M,#SS*

flB&*