Full text of "Zoologica : scientific contributions of the New York Zoological Society"

See other formats

ZOOLOGICA

SCIENTIFIC CONTRIBUTIONS
OF THE

NEW YORK ZOOLOGICAL SOCIETY

1938

Numbers 1-21

PUBLISHED BY THE SOCIETY
THE ZOOLOGICAL PARK, NEW YORK

J^eto Hork Zoological Society

General Office: 90 Broad Street, New York City

Officers;

President, W. Redmond Cross
Vice-Presidents, Kermit Roosevelt and Alfred Ely
Chairman, Executive Committee, W. Redmond Cross
Treasurer, Cornelius R. Agnew
Secretary, Fairfield Osborn

Scientific Staff

Zoological iparfe
W. Reid Blair, Director

Raymond L. Ditmars, Curator of Mammals and Reptiles
Lee S. Crandall, Curator of Birds
Charles R. Schroeder, Veterinarian
O.aude W. Leister, Ass’t to the Director and Curator, Educational Activities
H. C. Raven, Prosector
Edward R. Osterndorff, Photographer
William Bridges, Editor and Curator of Publications

Aquarium

Charles M. Breder, Jr., Acting Director
Christopher W. Coates, Aquarist
Ross F. Nigrelli, Pathologist
G. M. Smith, Research Associate in Pathology
Homer W. Smith, Research Associate in Physiology

department of C Topical 3&esearct)

William Beebe, Director and Honorary Curator of Birds
John Tee-Van, General Associate
Gloria Hollister, Research Associate
Jocelyn Crane, Technical Associate

Cbitorial Committee

Fairfield Osborn, Chairman

Charles M. Breder, Jr.
William Bridges

W. Reid Blair
WJlliam Beebe

CONTENTS

Part 1. March 31, 1938. page

1. The Significance of Differential Locomotor Activity as an

Index to the Mass Physiology of Fishes. By C. M. Breder,
Jr., & R. F. Nigrelli. (Text-figures 1-8) 1

2. A Study of the Anoplocephaline Cestodes of North American

Rabbits. By John G. Arnold, Jr. (Text-figures 1-23) 31

3. The Templeton Crocker Expedition. XIII. Penaeidae from

the Region of Lower California and Clarion Island, with
Descriptions of Four New Species. By Martin D. Burk-
enroad. (Text- figures 1-34) 55

4. Fibro-epithelial Growths of the Skin in Large Marine Turtles,

Chelonia mydas (Linnaeus). By G. M. Smith & C. W.
Coates. (Plates I-XVIII) 93

Part 2. July 14, 1938.

5. Plankton of the Bermuda Oceanographic Expeditions. VIII.

Medusae Taken During the Years 1929 and 1930. By
Henry B. Bigelow. (Text-figures 1-23) 99

6. Classification of Carp-like Fishes. By J. T. Nichols 191

7. Tissue Responses of Cyprinodon variegatus to the Myxo-

sporidian Parasite, Myxobolus lintont Gurley. By Ross F.
Nigrelli & G. M. Smith. (Plates I-VII) 195

8. Electrical Characteristics of the Electric Tissue of the Elec-

tric Eel, Electrophorus electricus (Linnaeus). By R. T.
Cox & C. W. Coates. (Plate I; Text- figures 1 & 2) 203

DEC 3 0 IBM

PAGE

9. Colonial Rattlesnake Lore, 1714. By James R. Masterson 213

Notes : 217

Part 3. September 28, 1938.

10. Morphology of the Hypophysis of the Common Goldfish

(Carassius auratus L). By W. Randal Bell. (Plates
1 & 2; Text-figures 1 & 2) 219

11. Pathology of Dirofilaria Infestation. By Joel Hartley.

(Plates I-V) 235

12. Papilloma of the Skin Occurring in an Electric Eel, Elec-

trophorus electricus (Linnaeus). By C. W. Coates,

R. T. Cox & G. M. Smith. (Plates I-VI) 247

13. Arithmetical Definition of the Species, Subspecies and

Race Concept, with a Proposal for a Modified Nomen-
clature. By Isaac Ginsburg. (Text-figures 1-4) 253

14. Eastern Pacific Expeditions of the New York Zoological

Society. XIV. Introduction, Itinerary, List of Stations,

Nets and Dredges of the Eastern Pacific Zaca Expedi-
tion, 1937-1938. By William Beebe. (Text-figures
1 & 2) 287

15. Eastern Pacific Expeditions of the New York Zoological

Society. XV. Seven New Marine Fishes from Lower
California. By William Beebe & John Tee-Van.
Plates I-III; Text-figures 1-5) 299

16. Deep-sea Fishes of the Bermuda Oceanographic Expedi-

tions. Family Anguillidae. By A. Vedel Taning.
(Text-figure 1) 313

PAGE

Part 4. December 31, 1938.

17. The Phylogeny of the Characin Fishes. By William K.

Gregory & G. Miles Conrad. (Text-figures 1-37 ) 319

18. Eastern Pacific Expeditions of the New York Zoological

Society. XVI. Holothurians from the Western Coasts
of Lower California and Central America, and from
the Galapagos Islands. By Elisabeth Deichmann.
(Text-figures 1-15) 361

19. A Note on Eggs and Young of Leioheterodon madagas-

cariensis (Dumeril & Bibron). By Roger Conant.
(Plate I) 389

20. An Unusual Aberrantly Colored Pleuronectid. By C. M.

Breder, Jr. (Plate I) 393

21. A Diagnostic Test for the Recognition of Tuberculosis in

Primates. By Charles R. Schroeder. (Plates I & II) 397

Index to Volume XXIII 401

vii

ZOOLOGICA

SCIENTIFIC CONTRIBUTIONS

OF THE

NEW YORK ZOOLOGICAL SOCIETY

VOLUME XXIII
Part 1

Numbers 1-4

PUBLISHED BY THE
THE ZOOLOGICAL PARK,

March 31, 1938

SOCIETY
NEW YORK

CONTENTS

PAGE

1. The Significance of Differential Locomotor Activity as an

Index to the Mass Physiology of Fishes. By C. M. Breder,
Jr., & R. F. Nigrelli. (Text-figures 1-8) 1

2. A Study of the Anoplocephaline Cestodes of North American

Rabbits. By John G. Arnold, Jr. (Text-figures 1-23) 31

3. The Templeton Crocker Expedition. XIII. Penaeidae from

the Region of Lower California and Clarion Island, with
Descriptions of Four New Species. By Martin D. Burk-
enroad. (Text-figures 1-34) 55

4. Fibro-epithelial Growths of the Skin in Large Marine Turtles,

Chelonia mydas (Linnaeus). By G. M. Smith & C. W.
Coates. (Plates I-XVIII) 93

ZOOLOGICA

SCIENTIFIC CONTRIBUTIONS
OF THE

NEW YORK ZOOLOGICAL SOCIETY

The Significance of Differential Locomotor Activity as an Index
to the Mass Physiology of Fishes.

C. M. Breder, Jr., & R. F. Nigrelli

New York Aquarium
(Text-figures 1-8).

Introduction.

The existence of a measurable differential in the locomotor behavior of
fishes under various conditions of companionship has been established by
Schuett (1934) and confirmed by Breder and Nigrelli (1935b) and Escobar,
Minahan and Shaw (1936). Each report mentions the use of similar physical
equipment but somewhat different handling of the data. All three are
almost purely descriptive with little attempt to interpret the possible sig-
nificance of the data obtained or to explain the mechanism responsible for
the observed differences in behavior.

It is the purpose of the present communication, therefore, to interpret
the findings of the various investigators and to correlate these results, so
far as possible, with other known data on the social attitudes of fishes. To
this end a considerable amount of experimental work is reported, some of
which was briefly referred to by Breder and Nigrelli (1935b). It was early
found that before the observable results could be satisfactorily interpreted
it was necessary to consider in some detail the various items of both the
internal and external environment. Supplementary experiments, that may
seemingly have little bearing on the problem of social attitude or mass
physiology, were found necessary.

Briefly stated, the method of investigation, which was devised by
Dr. Paul Escobar and associates, consists of observing the successive posi-
tions of a fish in each of the three dimensions through a system of coordi-
nates ruled on the glass walls of an aquarium. From such observations the
distance traveled, the configuration of paths and their relation to the three
spatial axes may be calculated. That these differ for any one specimen with
companions and when alone has already been established in the three re-
ports mentioned. Before going into a full discussion of these data and pre-
senting more details of related matters, a brief consideration of the factors
involved may best be presented.

[i]

2 Zoologica: New York Zoological Society [XXIII :1

Factors Involved.

Of the numerous factors concerned, many are self evident, some are
not, and a few have not been previously discussed in this connection. It
does not follow, furthermore, that this list is exhaustive, but is intended
to indicate only those factors that must be considered for any adequate
understanding of social phenomena in fishes.

External Factors.

All the factors in the external environment may be considered under
this head and so include all influences extrinsic to the subject itself.

Space. The available space places strictures on the movements of the
included organisms. For example, any calculations based on physical for-
mulae that assume an infinite medium are accordingly modified by the limits
of the container.

The shape of the space as well as the volume is distinctly important.
This has been discussed in considerable detail by Escobar, Minahan and
Shaw (1936). Thus, its relative extent in reference to the vertical and
horizontal axes is an important differential. This latter element is ap-
parently due extrinsically entirely to the unidirectional nature of the gravi-
tational influence, illumination, etc.

Radiant energy. The impact of radiant energy is of extreme importance
since, for example, the presence of visible radiation is essential to vision,
the eyes being important receptors in the aggregating behavior of fishes.
Other wave lengths have not been studied in detail but Beebe (1935) showed
that a beam of ultra-violet would cause certain fishes to aggregate in a
closely compacted mass within its radius.

Temperature. The effects of thermal differences and thresholds are ob-
viously of considerable importance to the differential behavior of fishes.
Schooling in some cases is clearly temperature controlled; Townsend (1916)
and Breder and Nigrelli (1934, 1935a). In these cases the lowering of
temperature induces the fishes to aggregate.

Chemical conditions. Many substances dissolved in the water have a
clear effect on the “social” attitude of fishes, including the amount of C02
and 02 present; Allee (1931), Allee and Bowen (1932), Breder (1934),
Breder and Nigrelli (1934, 1935a). Eddy (1925) maintained that certain
stimulants caused aggregation and certain depressants caused dispersal.

Objects. What, for want of better term, may be considered as objects
may include both those that are inert and those that are active. Rocks,
plants, etc., serve in part to act as physical boundaries to the subjects but
more especially as boundaries to their vision which, as will develop, are of
considerable importance in the behavior pattern. Food items present a
vigorous stimulation to hungry subjects.

The companions may be here classified as homotypic and heterotypic,
following the usage of Escobar, Minahan and Shaw (1936). These may be
attractive or repellant according to the particular physiological state of
the subject at the time.

Internal Factors.

Under this head may be considered all the effects of the internal en-
vironment. These then include the physical, chemical, and psychological
make-up of the organism and cover all its inherent tendencies and char-
acteristics.

The Receptors. Since all knowledge of the external world comes by way
of the sensory receptors, considered in the broadest sense, they in their
entirety may be taken as a starting point. It so happens that in all fishes

1938]

Breder & Nigrelli: Mass Physiology of Fishes

which have been used in experiments involving their attitude to their
fellows, vision alone has been shown to be significant; Newman (1876)
Clupea harengus Linnaeus; Parr (1927) Pneumatophorus grex (Mitchill) ;
Spooner (1931) Morone labrax Linnaeus; Breder (1929) Jenkinsia stolifera
(Jordan and Gilbert) ; Bowen (1931, 1932) Ameiurus melas (Rafinesque) ;
Breder and Nigrelli (1934, 1935a) Lepomis auritus (Linnaeus). This has
been in part demonstrated by the breaking up of schools on dark nights or by
covering the eyes of individuals with vaseline and lamp black. While it has
not been previously noted, the same holds for Carassius auratus (Linnaeus),
the species on which Schuett (1934), Breder and Nigrelli (1935b) and
Escobar, Minahan and Shaw (1936) chiefly worked. Consequently there is
no need under the present limits of data to speculate on the possible sig-
nificance of the other senses so far as response of fish to fish is concerned.
However, the presence of chemically detectable food, mechanical jarring of
the aquarium, etc., rapidly change the pattern of behavior and for the pur-
pose of the present experimental studies have been rigorously guarded
against.

The Effectors. After stimulae are passed from the receptors and modi-
fied by the neuro-endocrine mechanism, the past conditioning of the fish,
the particular drives that it labors under at the moment, etc., the effectors
may produce appropriate movements or other responses. However, it cannot
be assumed that such will be locomotor, although they clearly are in many
cases. They may, for instance, be simply the rolling of an eye or the ex-
tension of the dorsal fin. Furthermore, undoubtedly many of the stimulae
may reach no evident effector at all, becoming “lost” in the internal en-
vironment with perhaps scarcely a quickened heartbeat. What such stimulae
may have to do with conditioning in fishes we can only speculate upon at
present.

The locomotor effectors, the only ones which produce results which can
be handled by the present technique, are distinctly axial. That is to say,
because of the streamlined shape of a goldfish and its built-in locomotor
mechanism, it is much more likely to move forward than in any other
direction. While it is capable of backing up, its rising and falling movements
are actually, under normal conditions, forward movements deflected up or
down and its sidewise movements are forward ones which are laterally de-
flected. Escobar, Minahan and Shaw (1936) expressed the same thought
as follows: “. . . the locomotor mechanism of fishes (with a few exceptions,
e.g., seahorse) are adapted to propel the fish along the axis of the body,
the latter being normally oriented in most species of fishes in a horizontal
plane.”

The Drives. The various internal urges that impel fishes to respond to
stimulae according to certain patterns obviously enter into the locomotor
behavior and to a large extent control it. These stimulae may arrive directly
from either the external or internal environment. The seeking of comfort,
suitable temperature, light, depth of water, current, contact with solids,
which are ordinarily understood as the various tropisms, can be shown
to be induced directly by external influences. For the purpose of the experi-
ments many of these may be eliminated by having the factors uniform
throughout the experimental tank. Others, such as gravity or light, which
necessarily must have some axis, must be allowed for in the calculations.
While thigmatropism in its full sense is not a characteristic of the goldfish,
its desire to move close to solids is of considerable importance as will
develop in some of the experiments.

Lest it might be thought that the agitation of water caused by the
passage of another fish tends to keep schools intact, it need only be pointed
out that, if fishes are at all aware of such movements, they are entirely
inadequate to produce results obtained. As previously mentioned, the ob-
scuration of vision results in the disintegration of schools.

The internal drives of hunger and sex are not easily controlled. By

Zoologica: New York Zoological Society

[XXIII:!

adequate feeding apparently the first may be neutralized, and by using gold-
fish out of the breeding season the second may not be important, but as
Escobar, Minahan and Shaw (1936) pointed out, “Even in immature in-
dividuals sex cannot be ignored, since there may exist dimorphisms and dif-
ferential physiological abilities.” Studies of both these items invite investi-
gation. The above investigators further write: “In the case of animals
with highly developed visual sense organs the contour and coloration of
their fellows in homotypic or heterotypic groups may conceivably alter
their movements resulting in various degrees of aggregation or dispersal.”
The known attitude of fishes to attempt mating with a variety of quite
unfish-like objects does not support this view, however; Lissmann (1932)
Betta splendens Regan; Breder and Coates (1934, 1935) Lebistes reticulatus
Peters; Noble (1934) Eupomotis gibbosus (Linnaeus); Breder (1936)
Lepomis auritus (Linnaeus). On the other hand, the abstract of studies by
Noble and Curtis (1935) on Hemichromis bimaculatus is suggestive of some
such differentiation.

Heterotypic schools of fishes are not uncommon and those of young
fish are frequently of an ecological nature rather than a specific one, as
discussed by Reighard (1915). Parr (1931) considered the temporary sexual
dimorphism of Gasterosteus as a factor in the breaking up of the earlier
non-sexual schools. Breder and Nigrelli (1935a) discussed other cases of
heterotypic schools. Perhaps many of the vague controversial matters re-
garding fish schools and sex recognition could be cleared up by some
modification of the technique here used for examining the factors in the
simple aggregation of essentially non-schooling species.

The territorial attitude of fishes is well marked in many species, and is
associated with tank dominance as has often been mentioned, Breder (1934,
1936). This seems to be strongest in nesting and thigmatropic species and
weakest, if present at all, in schooling species. It is at least present to the
extent of some aggregations being partial to some small area. Nothing can
be said about homing instincts or migratory movements at this time but
such would surely only be an influence evident outside of laboratory tanks.
The behavior of animals on “home” territory and in unknown country is
demonstrably different without the necessity of recourse to the plotting of
paths or trajectories.

Spiral Movements. There is a well known tendency for animals when
blindfolded or in an environment with inadequate points of reference to
move in more or less circular or spiral paths. This seems to be an inherent
element of all forms of life examined. Schaeffer (1931) expresses it as fol-
lows: “All motile organisms move spirally when -guiding senses are not
functioning, so far as known. By experiment and observation this has been
found to hold true from bacteria to blindfolded aviators.” That it is equally
true of fishes is subsequently shown. The bearing that this feature has on
mass physiology will be discussed in detail at that place. Schaeffer (1928)
discusses the general problem of spiral movement and gives a good bibli-
ography. Gordan (1936) detected such movement in the massed flight of
insects.

Random Movements. It has been stated by Lotka (1925) that the
movements of organisms in a uniform environment should be of a random
character. They then follow the Einstein (1905) equation for Brownian
movement which is usually given as follows:

A.= iLi

LSx 3t tN 5?oc'

Expressed in non-mathematical terms, the law states that the mean
squares of the displacement in any direction, in equal intervals of time, is
proportional to the elapsed time. This may be expressed graphically by
plotting time against the mean squares of the displacements. Then random
movement is expressed by a straight line with its origin at the zero point
in each axis.

1938]

Breder & Nigrelli: Mass Physiology of Fishes

This has been shown to be in agreement with the movements of protozoa
by Przibram (1913). Applied in a larger sense to a non-uniform natural
environment (in which the irregularities are essentially cancelling) Ross
(1923), Pearson and Blakeson (1906), Brownlee (1911), Edgeworth (1920),
Cole (1922), Hardy (1922) and others have shown similar phenomena in
the migratory or other movements of non-captive organisms. Thompson
(1933) has attempted to apply the equation of Einstein to the movements
of tagged fishes. Certainly there would be expected to be a large degree of
randomness under the conditions of the experiments to be discussed. Con-
sequently by the application of the formula it should be expected to hold,
except for the fact that the formula applies theoretically to a space of infinite
volume. In the case of small containers and large objects, their early “reflec-
tion” back from the sides makes its application unsuitable.

The arbitrary limits of space available made it impossible in our cir-
cumstances to use a tank sufficiently large or a fish sufficiently small to
apply the Einstein equation directly. However, by certain sub-experiments
a fairly clear indication of the relationship of random to spiral movement
was obtained which is adequate for the present purpose, at least.

Light, temperature, chemical conditions, objects, and gravitational dif-
ferential were either controlled or, as in the case of the last, pi'operly
allowed for.

The food and sex drives were minimized by adequate feeding and per-
forming the experiments out of the reproductive season. These were clearly
sufficiently overcome to give comparable results. Small differential details
are explained in connection with the items they modify.

Experimental Studies.

The data obtained by various experimental contrivances already re-
ferred to in the previous sections are given here in full descriptive detail.
The interpretation based upon these results will be found in the section
following.

Experiments in Differential Locomotion. For the purposes of this series
of experiments a rectangular aquarium 20" x 16" x 14" high was used.
This then represented a rectangular enclosure of 4,480 cubic inches. One
side and an end were ruled faintly with a glass marking crayon in two inch
intervals. The side was thus divided into 70 squares and the end into 56
squares. By viewing simultaneously from the end and side, any fish in the
aquarium could be located in any one of the 560 two-inch cubes. Such posi-
tions were then noted at ten second intervals for uniform periods. The
resulting data, a series of three reference numbers, one for each plane of
projection, in a consecutive series (from “0” to “140”) could then be
calculated in various manners, both as to an estimation of the distance
traveled and as to the actual position of the subjects in space. The data,
then, which were obtained by all these experiments, are represented by a total
of 5,499 points in space, each represented by three index figures or 16,497
projections. Exp. 35 and 36 are omitted because of the partial data taken.

By a suitable geometrical treatment of the data a trajectory of the
fish, as measured, at ten-second intervals may be calculated. This, reduced
to meters per hour, is given in Table I.1 Obviously these figures are minimal
values as the fishes do not necessarily travel in a straight line between the
fairly long intervals. The methods used by Schuett (1934) and Escobar,
Minahan and Shaw (1936), give closer approximations to the actual dis-
tance traveled. However, our readings were made basically for another
purpose and are of course comparable among themselves, giving properly
relative values. A study of the data will reveal that there is a very genuine
separation between the rate of travel by a fish alone and one with com-

1 All Tables will be found at the end of this paper.

Zoologica: Neiv York Zoological Society

[XXIII :1

panions, the details of which are given in the discussion. The first 22
experiments were made by one observer by using a mirror set at an angle
of 45° to the ruled end so that both side and end could be observed at one
time. This was found to introduce slight complications, either by irregu-
larities in illumination or perhaps by the fish seeing his image. This was
subsequently overcome by enclosing the aquarium in a light-tight box, pro-
vided with a side and an end “peep hole” and with artificial illumination.
This required the simultaneous observations of two people and even here
certain irregularities in behavior could be traced to light entering laterally
from around the viewing holes. The details of the separate experiments are
given in Table I and discussed along with their interpretations.

Another method of handling the data is to consider the fish positions in
each plane of projection separately and plot these as frequency curves. Ob-
viously, any correlative changes in the shape of such curves to be found
associated with the different conditions indicate a change in the swimming
pattern of an active organism. These data are given in Table II. In certain
cases it is more convenient simply to consider the aquarium as divided in
half of each of its three axes and note the percentage of positions to either
side of the mid-lines. These data are given in Table III. Still another way
to treat the data is to count the number of crossings of these mid-lines in
each plane. This is not the same as the previous treatment as various
degrees of activity may give the same distribution or various speeds of
travel may include more or less crossings of the mid-lines. Each of the
three treatments, however, is nevertheless closely related to the others, and
taken together they increase the emphasis on the differential behavior. Still
another treatment is to note simply the number of changes in direction and
finally to plot out the changes in position along a time axis. These items are
considered at this point to indicate the flexibility of the technique rather
than for the purpose in hand, the details of which are considered along with
the inferences based upon them.

Experiments in Random and Spiral Movements. It is thoroughly evi-
dent that the aquarium used in the preceding experiments would be inade-
quate to make even a cursory study of random and spiral movements in any
but nearly microscopic organisms. In order to obtain some estimation of
these features a pan six feet in diameter and two inches deep was con-
structed. The bottom was covered with a uniform layer of sand and the
vertical sides were enameled white to make them as little evident as pos-
sible under the circumstances. The pan was illuminated from six feet above
by four electric fixtures spaced so as to provide a uniform vertical light,
and was surrounded by a curtain of heavy white sail canvas to exclude
extraneous light and disturbances. A motion picture camera was mounted
directly over the center for the production of permanent records. This was
found, however, not of any particular value as the experiments developed.
The speed with which the fish found the edge of even this pan was so rapid
as to give records of little value. It may be noted here that the restriction
of the subjects to an essentially two-dimensional field in no way invalidates
the calculations. Even a single dimension may be used for such purposes.

Sketch paths of five fish are shown in Text-figure 1. The longest repre-
sents not quite 15 seconds, as they were only continued until the subject
responded to the side. This was always immediately evident because in
every case the subject then began to circle close to it. Table VI gives the
details of the data. There is very considerable evidence of typical spiral
movements in both the figure and the tabular data. In these five cases the
tendency was to turn to the left, resulting in a counter-clockwise traveling
about the edge of the pan. These fish were gently placed at the points
shown and had never been in the pan before. It was quickly found by a
second trial that the conditioning had been such that the subjects learned
that there was a sheltering edge not far away if they traveled in any

1938]

Breder & Nigrelli: Mass Physiology of Fishes

Paths of five fishes of different species in a shallow pan, protected from outside
influences. 1. Carassius No. 1. 2. Carassius No. 2. 3. Lebistes reticulatus,
female (fish lost sight of at the point shown, but a moment later was found
traveling counter-clockwise around edge) . 4. Brachydanio rerio. 5. Villarius
catus, 2 cm. long (heavy dotted portion of path picked from a motion picture
record, each dot representing every eighth frame or at intervals of % of a
second). All other paths are free-hand tracings.

straight line. This in itself is rather remarkable, since actual sight of the
edge through two or more feet of water is not involved, as is evident from
the fact that on second trials they seldom went to the nearest edge but
seemingly struck out in a random direction when released but maintained
practically a straight path. How this could be we do not yet understand.
The short distances at which Breder and Coates (1935) found that Lebistes
could distinguish even evident objects also bears on this.

After proceeding to this point it became evident that such an apparatus
was inadequate for these studies and that suitable equipment was not avail-
able, on the basis of space alone, in the laboratories. Because of this, slow
moving aquatic gasteropods were substituted for fishes as subjects. Their
behavior, readily followed, gave some interesting data on animal behavior,
which would seem to be in accord with what the preceding fragmentary

Zoologica: New York Zoological Society

[XXIII :1

experiments would lead us to believe. Fortunately for the work, the snails
all seemed dissatisfied with a position on a flat horizontal surface and
showed a strong disposition to come to rest on the vertical sides of the pan.
The only difference in physical equipment was the placing of a large piece
of French plate glass over the sand so as to give the creeping snails an
absolutely plane surface to move over. Under this was inserted a pale
gray piece of cardboard, pencil ruled in small, scarcely discernible, squares.

The trial paths of certain snails are given in Text-figures 2 and 3.
Here, too, there are undoubted spiral movements, but the bias is slightly to
the right. The behavior of these snails yielded further data bearing on the
subject of random movement which is not entirely clear at this writing.
It was found, for example, that the first trial of Physa No. 1 was much more
nearly random than the second. The actual paths of these two trials are

1938]

Breder & Nigrelli: Mass Physiology of Fishes

given to Text-figure 2A. The data of these, plotted according to the Ein-
stein equation (i.e., square of the distance from the origin against time),
are also shown here and indicate that the second more closely approximates
that of a straight line while the first more closely resembles that of a
spiral path (Text-figure 2B). See Text-figure 4 for illustrations of
typical lines of translation and their representation by this type of plotting.
If the means of 5 consecutive points be plotted, curves are obtained which
are given in Text-figure 2C. Of the second little can be said because of the
slight amount of data, although it is obvious that this line must become more
and more nearly a straight one as the quantities taken to obtain mean values
are increased. The first path in this graph also more nearly approaches a
straight line, for similar reasons. The angle of these lines to the horizontal
gives a measure of the speed of recession from the origin and insofar gives
a measure of the behavior of the organism. At this point a difficulty inherent
in the method becomes apparent, since by an appropriate selection of mean
values a given path such as the first can be either represented as indicating
a spiral movement or a random movement. This is not so serious as it might

Zoologica: New York Zoological Society

[XXIII :1

be first thought, since what it actually indicates is that elements of both
are present and either one may be emphasized and studied according to
the scale of operations under consideration. This seeming paradox becomes
clear when it is considered that an animal or other object may be cutting
close spirals, turning cart-wheels or showing other vagaries of locomotion
and still be moving in a random fashion, if considered in a larger sense.
On the other hand an animal may be moving in very large spirals on which
are overlaid a measure of randomness considered in a finer sense. This only
becomes confusing when the size of the spirals and the random paths are of
similar magnitudes. However, it is just in such cases that the value of this
approach is greatest.

1938]

Breder & Nigrelli: Mass Physiology of Fishes

•O'*-! 93 93

2 °'I5 %

IS u

“ CQ Qh

3 S bOC
-K £ c c
w S.3 o

3 ft'O
”3 “ 3 0*
>

•3 £t-

•5 £*3 «

— « J?

aS O

•H O0 «W
0-0-0
o 00 ^
ea .g

O) H cj

s ftw!5

e,«tn'0
e o 2 41

§ 5 2

s§a-

® -g § 8

•§§~Q
W5 ^--5-3

.-an:

<u -O
3 3
bfi 3

V eo

.. C

c3^*r
_ o _ £

O ^ CO <D

J^d 5
e . «

l^or

-si aj

•h bfl ai v

0«C Sr1

03 ■+■>

j- X be o r i

+3 9><C ■*jU

S‘SH^ rt

P 03 „

a ^.Std .

03^ M.s^

§S 13 “ 2

tr3 o *o ^
-o o 4_» jjjo
<c

.2'a^ x

S,eS^

2-"

> w

The underlying biological causes of these differences between the snail
paths No. 1 and No. 2 are not so evident but would seem to have their roots
in some conditioning, orientation, or more broadly some learning process. By
plotting the data still another way (i.e., trial number against distance from
the origin at the end of fixed periods), what may tentatively be called a
learning curve is obtained. This is given in Text-figure 2D. Further data
of other individuals of Physa are given in Text-figure 5. The learning curves

Zoologica: Neiv York Zoological Society

[XXIII :1

of three specimens show considerable differences, including one individual
which apparently failed to learn, performing more poorly on each successive
trials (Text-figure 5B). However, it is clearly evident that the mean of
these three snails nevertheless shows some shift to an increasing speed of
recession from the origin on succesive trials. These observed differences
may be due to individual background or other factors in the life of the
animals. Text-figure 5C shows the positions of Physa No. 3 at the end of
four different periods in each of its three trials, again indicating its lack
of learning. The essential random and spiral nature of the paths are
also indicated in Text-figure 5D in which the mean trial of the Physa No. 3
shows a marked resemblance to the first path of No. 1 given in Text-figure
2B. The means of all trials of these three Physa are likewise given, which
emphasizes the characteristics common to all of them (Text-figure 5A).

Data on another species, Planorbis, are given in Text-figure 3 and show
similar characteristics. The actual paths are given and the resulting plot-
tings. The mean of four trials shows again the spiral and random ten-
dencies inherent in these paths. The learning curve is well marked and,
as in the other species, is shown both with the data as picked from the
curve and the square roots of those values.

Similar studies of two specimens of V. vivipara, the data of which are
not given in full, appear in the mean values for all in Text-figure 5A. This
species was added to the previous two, since it is not a pulmonate. The
essential similarity of its behavior is evident. The difference in shapes of
the curves of the three species is referable directly to their activity and
speed. It is to be noted how basically similar are all patterns.

Significance of Observed Behavior.

The data obtained by means of the experiments previously described
and presented in Tables I and V in most cases may be better understood if
presented in graphic form. The index of the rate of travel under various
conditions as expressed by the data of Experiments 1 to 22 of Table I is
analyzed in Table IV. It may be seen at once from this table that the two
fishes tested move much faster when alone than when with companions. This
holds not only for the mean values obtained but also for both the maximum
and minimum values. It is also to be noted that there is but one case of
overlapping of extreme figures and that there is virtually complete separa-
tion between the two sets of values. Actually this is strictly true for all
of the mean values, the only ones of full statistical significance. Maximum
and minimum values show a separation of over 50% and the minimum con-
siderably more. The close agreement between the two test fishes is to be
especially noted. These data support the experiments of Schuett (1934) and
Escobar, Minahan and Shaw (1936). The data of these investigators are
given in Table III. This has been equated as well as is possible from their
data to the present experiments. Since, however, their procedure differed
somewhat from the present it cannot be thoroughly comparable, although
it is fully adequate to establish the validity of the increase in speed of
travel in Carassius with solitude. These data are presented in graphic form
in Text-figure 6. Some of the irregularities are doubtless due to differences
in technique and methods of calculation. This is largely obviated in the top-
most line which represents the travel of a fish with companions as a per-
centage of the fish alone. The apparent differences smooth out to a con-
siderable extent under such treatment, well indicating the basic agreement
between the three sets of studies. Lest it may be thought that the dif-
ferent 02 and C02 concentrations associated with various numbers of fish
are responsible for these differences it is but necessary to point out that
Schuett (1934) successfully ruled out this factor for a much smaller
aquarium than the one here used.

1938]

Brecler & Nigrelli: Mass Physiology of Fishes

Text-figure 6.

Comparison of the data of Schuett (1934), Escobar, Minahan and Shaw (1936),
with the present. The topmost line was computed by reducing the amount
of travel of the fish alone to unity. Then the indicated figures represent the
extent of travel reduction with three companions.

The new experiments, each representing a special feature that in some
cases shows more clearly just what transpired by other treatment of the
data, may be now considered in reference to rate of travel. Another fish,
“C”, in the shadow box (Exp. 23) previously described and illuminated by
a hundred watt light, showed a speed of 162.15 M/h when alone. The addi-
tion of other fish (Exp. 24) one-half hour later resulted in even a higher
rate of travel (166.98 M/h). These fish changed abruptly from daylight
were clearly worried and it would seem that their rapid movements were
responsible for the failure of the accommodated test fish to reduce its speed.
Further data on this appear in the consideration of the effect of light.
Later, on the removal of the box (Exp. 25 and 26), the travel reduced to
values comparable to those of “A” and “B” with companions; 61.70 and
33.39, mean 47.54.

A rotating motor-driven disc placed under the electric bulb, causing
the rapid passing of shadows, clearly caused fright and the speed increased
(Exp. 27). Some hours later this was even higher (Exp. 28) and three days
later (Exp. 29) the speed was still high although it is hard to imagine that
this effect lasted that long. This feature needs further study.

Another fish, “D”, was then studied in relation to the amount of light

Zoologica: New York Zoological Society

[XXIII :1

used. The data are given in Experiments 30 to 35. If the values of fish alone
are plotted, there seems to be an increase of speed with an increase of
illumination but the data are inadequate for a thorough understanding of
this phenomenon. The other items cannot be elucidated in terms of speed
of travel at this time but are discussed in subsequent connections.

The distribution of the experimental fishes in space is given in Table V,
expressed in terms of percentage. If again the first 22 experiments are
combined their mean distribution alone and with companions may then be
plotted and compared. These calculations are given in Table V and the
resulting graphs are shown in Text-figure 7. It will be noted at once that
the curves obtained from fishes alone are convex downwards and those from
fishes with companions are convex upwards. An examination of the data
of Table II will show that this is a consistent feature of each experiment.
The separation is not so clear in some individual cases as in the integrated
means of Text-figure 7, but in each of these there is a reasonably clear and
evident reason because of the conditions of the experiments and to that
extent forms a measure of the particular influences at work as noted in Table
I. The minor distortions of the curves of Text-figure 7 are each indicative
of tendencies in the subject and form an excellent quantitative measure of
them. A consideration of the curves in detail yields the following data,
each treated according to its separate coordinates.

The vertical curve of the fish alone shows that the fish tended in a
regular manner to keep close to the bottom of the aquarium. Actually it
occupied the bottom 2" of water 28% of the time and the surface 2" only
8%, with a quite regular grading of the intermediate strata. Considering
the 14 individual experiments upon which this curve of the fish alone is
based, there are the following extreme divergencies: Percentage of time

Text-figure 7.

Configurational behavior of goldfish alone and with companions in respect to
three dimensions. Solid line: fish alone. Dotted line: fish with three com-
panions. The short vertical pecked lines on the latter indicate the limits,
counting from the central position that covers 50% of the readings. A pro-
jection of these on the solid line shows that over 50% is there excluded by
them instead of included.

1938]

Breder & Nigrelli: Mass Physiology of Fishes

spent in the bottom layer varied from a minimum of 13% to a maximum
of 60% while time spent in the surface layer varied from a minimum of 0%
to a maximum of 31%. In only three of the experiments did the fish spend
more time in the surface layer than in the bottom one, and in only one was
a peak value reached in the middle layer. This may simply be taken as a
mathematical expression of the normal attitude of a solitary Carassius as
based on casual but extensive observations, to the effect that they are apt
to be found near the bottom, and if not there, at the surface, and least
commonly in mid-water.

If now the curve of the fish with companions be considered, it is clear
that a considerable change in distribution has taken place, the peculiar
S-shape of the curve apparently indicating two loci, or that the fish is
hesitant between two sites of approximately equal preference. The bottom
layer accounts for 19% of the time while an intermediate layer of from
6" to 8" from the surface accounts for 21% of the time. The surface layer
is visited about as frequently as before (6%), but the distribution between
surface and bottom is utterly different. Over 50% of the time is spent in
layers 2, 3 and 4 (57%) ; the small values of the layers below these, except
the bottom one, indicate rapid dashes through them to cruise along in the
latter. With the same fish alone over 50% of the time was spent in
layers 5, 6 and 7 (61%). If we divide the tank into a top and bottom half
it appears that alone the fish spent 68% below and 32% above such a mid-
line. With companions this was inverted to read 54% below and 46% above,
indicating a more nearly uniform distribution.

The “longitudinal” curve running at right angles to gravity shows a
strong tendency for the solitary fish to keep to either end of the tank, most
markedly so at the right end. This was the end at which the mirror stood
and the curve gives a measure of the attractiveness of this device. If it may
be assumed that either end was equally attractive except for the mirror,
then the difference of the readings at either end (56-12 = 44%) may be
taken as a measure of this attraction. When companions are added and a
more median position is selected, as evidenced by the shape of the curve,
the strong attraction for the mirror end reduces but does not disappear.
Thus, between the two sets of experiments the attractiveness dropped from
a value of 56% as compared with all other influences in this plane to one of
17% on the addition of three companions. The non-mirrored end dropped
from 12% to 3% and since the latter is the lowest value on the curve, pre-
sumably indicates that its attractive value fell to zero. The median positions
3, 4, 5, 6 and 7 account for 60% of the time, whereas when alone the terminal
position 10 accounted for 56% of the time.

The “transverse” curve is esentially similar to the longitudinal one and
with fish alone either side seemed to be equally attractive (33% and 36%
respectively). Since there was no differential as in the longitudinal measure,
this is what one should naturally expect. The addition of three companions
caused the sides to lose their relative attractiveness to the extent that the
four central coordinates, Nos. 3, 4, 5, and 6, accounted for 60% of the time.

Dividing the transverse measures in half, we find that alone the
distribution to left and right is 49% and 51% respectively and with com-
panions 55% and 45%. These differences are likely too small to be signifi-
cant and seem to indicate merely the degree of accuracy obtainable by this
method on a half and half distribution. If, however, we take the longitudinal
measure and handle it in the same way, the figures for the fish alone stand
left 23% and right 77%. With companions this becomes 41% and 59%. The
bias by the mirror in this is thus overcome by the introduction of three
fish to the extent above indicated. The biological significance of the basic
distortion of the spatial distribution curves of these fish with and without
companions is quite apparent. The interpretation of them would seem
to be somewhat as follows: Since goldfish are gregarious, the behavior re-

Zoologica: New York Zoological Society

[XXIII :1

fleeted in these curves is evidently a measure of this social disposition.
The species may be said to be an aggregating one but not a schooling one
in the restricted sense, as may be checked by reference to goldfish in a
fairly large pond. The exact behavior in a small aquarium, however, be-
cause of its limitations of space, can only be properly appreciated by some
such treatment as the present. With these considerations in mind it would
seem that both the increased speed of travel and the tendency to seek the
walls of the aquarium would be due to an attempt to find company. This
finding of company with the resulting slowing down of movement and the
formation of an aggregation remote from the aquarium walls would then
indicate a satisfaction of that drive. Since the aggregating tendency im-
plies an urge to seek a position close to but not in contact with some object
(another fish), it also may be that approach to some inanimate object or
surface is used as a substitute to partially satisfy that urge when no com-
panions are present. In fact, it may even be that the movements of the
other fish is the determining factor in the expression of the preference
of approach to a fish as against a surface. It must be borne in mind, how-
ever, that the glass walls of the tank are optically not very evident. If other
sensory organs enter here they are not yet evident, for as previously men-
tioned the fish school in general appears to be a purely visual affair. The
fact that aggregations form remote from the walls is not surprising on a
purely mechanical basis. If each fish is considered as having a “field” of
influence about it, the formation of the aggregation in mid-water would
be expected. Fish in a central position would exert the effect of their
presence at every point on the surface of a sphere at whose center they were
located. Fish close to a side could exert only one-half that, i.e. on the
surface of a hemisphere. Fish at the junction of two sides would exert a
quarter as much and one in a corner one-eighth as much for similar reasons.
It is not necessary to labor the point that such objects, moving either
primarily at random or in spirals and mutually affecting each other, would
aggregate at approximately the tank center equi-distant from the sides.
Divergence from this would then indicate other influences.

If a space is enclosed that includes 50% of the positions, counting from
the center both ways, Text-figure 7 indicates that it also includes the peaks
of each of the curves of the fish with companions. Short vertical lines show
these limits in that figure. These lines projected to include the curves of
fish alone for the longitudinal and transverse readings include a very
small percentage, i.e., 10% and 16% respectively. That for the vertical com-
ponent is somewhat different because of the fact that it is at right angles
to the main axis of the fish and parallel to the gravitational field. Here the
curve of the fish alone is a gradual increase from one end to the other and
the corresponding value is intermediate between the one on either side,
i.e., 15%, 39%, and 46%.

Schuett (1934) in considering the speed of movement alone has shown
that there is some optimum of crowding above and below which the fish
move with greater rapidity. This would seem clearly to be for the reasons
above outlined. Since these fish place themselves in regard to one another
at some “standard” distance, the findings of Schuett would naturally follow
and the whole phenomenon become one of seeking an equilibrium. Fast
random swimming certainly would tend more quickly to restore a “lost”
goldfish to its group as well as tend more quickly to disperse an aggrega-
tion too closely packed for comfort.

Having considered the primary significance of the configurational dis-
tribution of fishes, the details of the further experiments may be discussed.
Experiment 23 made in the shadow box alone shows much greater equality
of distribution at either end of the horizontal components; longitudinal
32% and 22%, transverse 32% and 28%. Here the confusing outside ele-
ments were eliminated, giving another evidence of the influence of the end

1938]

Breder & Nigrelli: Mass Physiology of Fishes

mirror of the 22 primary experiments. The vertical component was in essen-
tial agreement with the previous, the surface layer .01% and the bottom
and highest figure 37%. This agreement is surprisingly close when it is
considered that this single experiment was only 1/22 as long as the mean
values previously discussed. Experiment 24 showed no change in speed, as
previously noted, and agreeing with this it also showed no change in con-
figuration. The presumed reason has already been mentioned, but it may be
pointed out that the rate of speed and configuration of pattern are clearly
not locked to each other on any mechanical basis. Consequently in the pres-
ent case we have two separate approaches, both of which indicate that in
this set of two experiments the expected response to companions that has
already been established here and by others did not take place. Experiments
25 and 26, without the shadow box, showed irregularities not unlike those
of the earlier experiments. As the speed of the fish slowed down, as pre-
viously noted, the configuration took on an approach to that of fish “A” and
“B” with Companions. All that can be said of this is that fish “C” either
responded more slowly to companions or was simply not as reactive. Per-
haps this could eventually be shown as a matter of individual difference in
the “psychic” attitude of various goldfish.

The light was then interrupted (Exp. 27) and the fishes sought either
end of the tank, spending 28% and 29% of the time in the two end divisions.
The next day (Exp. 28) they returned to a more nearly normal distribution,
the ends receiving only .08% and 19% of the positions.

Fish “D” with three companions (Exp. 30) behaved in a normal fashion.
This fish was then tested alone with various degrees of illumination (Exp.
31 to 35). The results of this can be best understood by an inspection of
Table V. Experiment 35, with a very faint light, just enough to enable the
making of the necessary observations, showed the fish to have spent much
time in mid-water. The three central coordinates accounted for 57% of
the time. Experiment 31, with a 10-watt bulb, showed a considerable resem-
blance to this one with the same coordinates accounting for 66% of the
time. These are the only two readings obtained with vertical characteristics
of this nature and they are the two most weakly illuminated ones. This
again suggests the visual nature of the movements and aggregating be-
havior in these fish. The other three experiments (32, 33 and 34) with 40,
100 and 200-watt bulbs all showed a strong predilection for the bottom,
but the data cannot decide if there is a gradual change with increased
illumination or whether it is simply a matter of passing some threshold.
Text-figure 8 gives this data graphically. It is to be noted that in case of
“no illumination” the fish spent more time in the top layer than in the
bottom. The various elements of this figure, including that of speed, would
seem to indicate that the lack of light was stimulating to locomotor activity
but without any configurational influence, that a low illumination (10 watts)
was quieting but again with no influence on the configuration, but that a
higher illumination (40 watts) drove the fish sharply to the bottom and
increased the locomotor activity. Increase of illumination above this point
seemed to have very little effect, but seemed, if the figures can be trusted to
such details, not to hold the fish so closely to the tank floor. It is doubted if
such slight differences are significant, however. If anything, it would seem
that there may be an activity and behavior threshold somewhere between
the illumination of the 10 and 40 watt lamps.

Experiments 36 and 37 represent the simultaneous paths of “C” and
“D”. For practical reasons they could be taken in two ordinates only. The
extent of agreement is close, but no closer than the agreement between suc-
cessive observations on the same fish. Consequently it is inferred that these
observed differences of such order are due to inherent variant characters
in the fish rather than unrecognized extrinsic elements.

A fifth fish, “E”, was tested in Experiments 38 to 41. The four fish,

Zoologica: New York Zoological Society

[XXIII :1

Text-figure 8.

Behavior of goldfish in respect to vertical distribution under different amounts
of illumination. See text for explanation.

A to D, which were examined at other times, were the companions of the
test fish. The speed of travel under varying light conditions was less with
companions than without, but the separation was not as great as in “A”
and “B”. The configuration, however, was in quite good agreement.

Another method of measurement of such activity is to plot the actual
transit of the mid-line or to check the reversals of direction. The amount of
labor involved in such calculations, however, is disproportionate to the re-
sults, which so far as can be seen from the present data give no further
information than that already obtained from the more readily applied
methods herein discussed.

An attempt was made to calculate the displacements according to the
formula of Einstein (1905) for random movement, but it soon became ap-
parent that the small size of the vessel in relation to the size and movement
of the fish made it impossible of application. The thesis was taken that if
such a fish moved at random, then any diversion from such randomness
could be used as a measure of mental integration on the part of the fish.
It is now clear, however, from a study of the movement patterns, that in
any case such behavior is not of a fully random nature, but is at all times
guided by a series of integrated perceptions, however slight.

Since the formula for Brownian movement, which may be used just as
well in two dimensions, is so closely controlled by physical influences, re-
course was made to the use of a large, shallow pan, virtually of two dimen-
sions, employing much smaller fishes. Here, too, we encountered a difficulty,
much different in nature, but one which finally led to the use of aquatic uni-
valves. A description of this device is given in the experimental section
together with the results obtained. The interpretation of these data involves
the further confirmation of the tendency of organisms to travel in spiral

1938]

Breder & Nigrelli: Mass Physiology of Fishes

paths when there is an absence of “landmarks” to guide them. The simulta-
neous action of the innate tendency to spiral and move in a broader sense
at random has been fully set forth in the experimental section. That this
is readily modified by some soi't of “leai'ning” is also appai’ent from the
experiments. Whether this is associated also with an obscure means of
orientation similar to that of “homing” as found in many animals or simply
that the snails wei’e impelled by some drive is not fully clear. We suspect
that it is the latter, however, for when left alone all three species came to
rest on the vertical sides of the pan. The pulmonates had evident reason,
but it is not clear in the case of the branchiate foi'm. If the snails could
detect these vei’tical surfaces at such a distance it is most surprising. That
the re-crossing of their previous paths had no bearing on the subject is
evident fi’om the actual paths shown in Text-figures 2 and 3, as some did
not ci’oss at all and whei’e they did thei'e is no evidence of significant
change of dii’ection.

The points that these considerations bring out enable us to detei'mine the
minimum l’equii’ements for an adequate study of the phenomenon involved:

1. A vessel must be used in which it is possible for the oi'ganism to do con-
sidei'able random and spiral wanderings without being able to “pick up”
any landmai’ks. This may be a “two dimensional” space if necessary. For
any animals of mici'oscopic size the physical limitations of most laboratoi'ies
pi-esent difficulties. 2. Photogi’aphic l'ecoi’ding at suitable time intei’vals
would be pi’actically essential for any extensive sui’vey of this field. Since
neither space, time nor suitable equipment are available to us under pi'esent
cii’cumstances it was deemed best to indicate the pi’ogi’ess of this work in
the hope that someone better situated would be able to further it.

The fundamental nature and l'easons for the existence of these spiral
movements have been speculated on by various investigators, chiefly Schaef-
fer (1928). Accoi’ding to our views, cei’tain featui'es of them may be
thought of in tex’ms somewhat according to the following: It has been
shown for invei’tebrate animals that those which move by means of muscular
effort as opposed to ciliary efforts, behave differentially under various
amounts of illumination (Welsh, 1933). The former move more rapidly,
according to their phototropism, as the units of illumination are increased,
while the latter show no such correlation. The present studies indicate a
greater amount of movement under an increased illumination at right angles
to the plane of activity. This suggests a similar increase in muscular
effort under such stimulation in at least the lower vertebrates. It should
follow from this that such spiral movements as may be present would tend
to become exaggerated. Since the muscular efforts are being intensified,
their unit acts come at shorter intervals and a given path is covered in a
shorter period of time. This would seem to culminate in the spiral paths
of insects about strong lights or even in the extreme conditions in some
fishes discussed by Breder and Harris (1935). These remarks are un-
doubtedly an over-simplification of the condition, a more full examination
revealing that a number of separate items enter into the complex of be-
havior, at least including the following :

1. Increasing speed of translation with increased illumination.

a. Direct positive or negative phototropism (Welsh, 1933).

b. Movements at right angles to axis of light (original).

2. Spiral movements inherent in organisms (Schaeffer, 1928, and
others) .

3. Tendency to expose dorsum (or other surface) to source of strong
illumination (Breder and Harris, 1935).

4. Disturbance of neural integration by large changes in illumination
(Breder and Harris, 1935).

5. Disturbance of neural integration by other environmental or patho-
logical causes (Hollister, 1934).

Zoologica: New York Zoological Society

[XXIII :1

It is not necessary for certain effects that all of these factors be present
in evident form and there may be still others, as yet unanalyzed. As far as
it is possible to understand these effects at present it would seem that the
increasing speed of normal reactions to light due to increased illumination
and the inherent tendency for organisms to show circling movements are
basic to the rest and represent all that is necessary to account for them. The
resulting movements from slight stimulae may be of value to the individual,
but beyond a cei'tain point may lead to destruction.

It is evident from many quarters that the social relationship of a fish
to its companions is not static but changes rather violently from time to
time. Among adults this is most evident during the reproductive cycles as
is emphasized in practically every paper on fish reproduction. Even in im-
mature fish diverse influences are at work. Most recently this has been
emphasized by Langlois (1935), who studied the relationships between
young bass in fish-rearing ponds. Not only did he find that changes pro-
ceeded with their development, but that quite different associations appeared
in ponds that were essentially similar in environment and original popula-
tion. This he treated by measuring the output at the end of the season.
From some he obtained a uniform group showing only the normal curve of
variation in which all fed on the food supplied. In others he found a group
of small fish and a group of large, the latter feeding on the smaller and
refusing other food as well as showing different attitudes toward schooling.
In all he enumerated eight different types of social organization, all derived
from similar starting points. This is mentioned to emphasize the dynamic
quality of fish aggregations and to indicate to some extent that the problem
cannot be handled as one involving only the varying aspects of a continuing
attitude on the part of the subjects involved.

Welty (1934) studied the learning of a maze by goldfish with and
without companions that had been trained in the same maze. He found that
those with trained companions learned the maze faster. The aggregating
tendencies of this species certainly account for his results, as indeed might
be expected. The results give a measure of the influence of an individual
that “knows” where it is going over one that does not.

Summary.

1. The relation of swimming speed to the number of fishes present as
indicated by Schuett (1934) and Escobar, Minahan and Shaw (1936) has
been further confirmed in larger bodies of water.

2. The mean positions of fish in relation to coordinates in three dimen-
sions is shown to be modified by the numbers present in a significant man-
ner, providing a quantitative technique for the study of environmental
influences in active animals.

3. The inherent tendency for animals to travel in spiral paths has been
shown for fishes and snails.

4. The essentially random nature of the broader movements of animals
has been indicated, as have been the experimental requirements for its full
study.

5. The equation of Einstein for the movement of Brownian particles
should be applicable to the higher organisms if proper data could be ob-
tained.

6. The basic nature of the tendency of organisms to travel in spiral
paths may be influenced to various degrees by impingement of the environ-
ment on the neui’O-mechanism and can be shown to reach from useful
activity to self-destruction.

1938]

Breder & Nigrelli: Mass Physiology of Fishes

TABLE I.

Data on extent of locomotion and conditions of experiments. Carassius auratus, 2 to
3 inches. Each experiment covers 140 observations. Aquaria 20 x 16 x 14 inches
high (50.8 x 40.6 x 35.6 cm.). Volume 4,480 cubic inches (72,427 cc.).

No.

Date.

Hour

Start.

No. of
Compan-
ions.

Meters

per

Hour.

Condition.

Remarks.

Fish. ■

Aug. 21